modelId
stringlengths 5
122
| author
stringlengths 2
42
| last_modified
timestamp[us, tz=UTC] | downloads
int64 0
738M
| likes
int64 0
11k
| library_name
stringclasses 245
values | tags
listlengths 1
4.05k
| pipeline_tag
stringclasses 48
values | createdAt
timestamp[us, tz=UTC] | card
stringlengths 1
901k
|
|---|---|---|---|---|---|---|---|---|---|
Qwen/Qwen-1_8B-Chat-Int4 | Qwen | 2023-12-13T15:43:40Z | 455 | 33 | transformers | [
"transformers",
"safetensors",
"qwen",
"text-generation",
"custom_code",
"zh",
"en",
"arxiv:2309.16609",
"arxiv:2305.08322",
"arxiv:2009.03300",
"autotrain_compatible",
"4-bit",
"gptq",
"region:us"
]
| text-generation | 2023-11-30T02:59:02Z | ---
language:
- zh
- en
tags:
- qwen
pipeline_tag: text-generation
inference: false
---
# Qwen-1.8B-Chat-Int4
<p align="center">
<img src="https://qianwen-res.oss-cn-beijing.aliyuncs.com/logo_qwen.jpg" width="400"/>
<p>
<br>
<p align="center">
🤗 <a href="https://huggingface.co/Qwen">Hugging Face</a>   |   🤖 <a href="https://modelscope.cn/organization/qwen">ModelScope</a>   |    📑 <a href="https://arxiv.org/abs/2309.16609">Paper</a>    ｜   🖥️ <a href="https://www.modelscope.cn/studios/qwen/Qwen-1_8B-Chat-Demo/summary">Demo</a>
<br>
<a href="https://github.com/QwenLM/Qwen/blob/main/assets/wechat.png">WeChat (微信)</a>   |   <a href="https://discord.gg/z3GAxXZ9Ce">Discord</a>   ｜   <a href="https://dashscope.aliyun.com">API</a>
</p>
<br>
## 介绍（Introduction）
**通义千问-1.8B（Qwen-1.8B）**是阿里云研发的通义千问大模型系列的18亿参数规模的模型。Qwen-1.8B是基于Transformer的大语言模型, 在超大规模的预训练数据上进行训练得到。预训练数据类型多样，覆盖广泛，包括大量网络文本、专业书籍、代码等。同时，在Qwen-1.8B的基础上，我们使用对齐机制打造了基于大语言模型的AI助手Qwen-1.8B-Chat。本仓库为Qwen-1.8B-Chat的Int4量化模型的仓库。
通义千问-1.8B（Qwen-1.8B）主要有以下特点：
1. **低成本部署**：提供int8和int4量化版本，推理最低仅需不到2GB显存，生成2048 tokens仅需3GB显存占用。微调最低仅需6GB。
2. **大规模高质量训练语料**：使用超过2.2万亿tokens的数据进行预训练，包含高质量中、英、多语言、代码、数学等数据，涵盖通用及专业领域的训练语料。通过大量对比实验对预训练语料分布进行了优化。
3. **优秀的性能**：Qwen-1.8B支持8192上下文长度，在多个中英文下游评测任务上（涵盖常识推理、代码、数学、翻译等），效果显著超越现有的相近规模开源模型，具体评测结果请详见下文。
4. **覆盖更全面的词表**：相比目前以中英词表为主的开源模型，Qwen-1.8B使用了约15万大小的词表。该词表对多语言更加友好，方便用户在不扩展词表的情况下对部分语种进行能力增强和扩展。
5. **系统指令跟随**：Qwen-1.8B-Chat可以通过调整系统指令，实现**角色扮演**，**语言风格迁移**，**任务设定**，和**行为设定**等能力。
如果您想了解更多关于通义千问1.8B开源模型的细节，我们建议您参阅[GitHub代码库](https://github.com/QwenLM/Qwen)。
**Qwen-1.8B** is the 1.8B-parameter version of the large language model series, Qwen (abbr. Tongyi Qianwen), proposed by Aibaba Cloud. Qwen-1.8B is a Transformer-based large language model, which is pretrained on a large volume of data, including web texts, books, codes, etc. Additionally, based on the pretrained Qwen-1.8B, we release Qwen-1.8B-Chat, a large-model-based AI assistant, which is trained with alignment techniques. This repository is the one for Qwen-1.8B-Chat-int4.
The features of Qwen-1.8B include:
1. **Low-cost deployment**: We provide int4 and int8 quantized versions, the minimum memory requirment for inference is less than 2GB, generating 2048 tokens only 3GB of memory usage. The minimum memory requirment of finetuning is only 6GB.
2. **Large-scale high-quality training corpora**: It is pretrained on over 2.2 trillion tokens, including Chinese, English, multilingual texts, code, and mathematics, covering general and professional fields. The distribution of the pre-training corpus has been optimized through a large number of ablation experiments.
3. **Good performance**: It supports 8192 context length and significantly surpasses existing open-source models of similar scale on multiple Chinese and English downstream evaluation tasks (including commonsense, reasoning, code, mathematics, etc.), and even surpasses some larger-scale models in several benchmarks. See below for specific evaluation results.
4. **More comprehensive vocabulary coverage**: Compared with other open-source models based on Chinese and English vocabularies, Qwen-1.8B uses a vocabulary of over 150K tokens. This vocabulary is more friendly to multiple languages, enabling users to directly further enhance the capability for certain languages without expanding the vocabulary.
5. **System prompt**: Qwen-1.8B-Chat can realize roly playing, language style transfer, task setting, and behavior setting by using system prompt.
For more details about the open-source model of Qwen-1.8B-chat int4, please refer to the [GitHub](https://github.com/QwenLM/Qwen) code repository.
<br>
## 要求（Requirements）
* python 3.8及以上版本
* pytorch 2.0及以上版本
* 建议使用CUDA 11.4及以上（GPU用户、flash-attention用户等需考虑此选项）
* python 3.8 and above
* pytorch 2.0 and above
* CUDA 11.4 and above are recommended (this is for GPU users, flash-attention users, etc.)
## 依赖项（Dependency）
运行Qwen-1.8B-Chat-Int4，请确保满足上述要求，再执行以下pip命令安装依赖库。如安装`auto-gptq`遇到问题，我们建议您到官方[repo](https://github.com/PanQiWei/AutoGPTQ)搜索合适的预编译wheel。
To run Qwen-1.8B-Chat-Int4, please make sure you meet the above requirements, and then execute the following pip commands to install the dependent libraries. If you meet problems installing `auto-gptq`, we advise you to check out the official [repo](https://github.com/PanQiWei/AutoGPTQ) to find a pre-build wheel.
```bash
pip install transformers==4.32.0 accelerate tiktoken einops scipy transformers_stream_generator==0.0.4 peft deepspeed
pip install auto-gptq optimum
```
另外，推荐安装`flash-attention`库（**当前已支持flash attention 2**），以实现更高的效率和更低的显存占用。
In addition, it is recommended to install the `flash-attention` library (**we support flash attention 2 now.**) for higher efficiency and lower memory usage.
```bash
git clone https://github.com/Dao-AILab/flash-attention
cd flash-attention && pip install .
# 下方安装可选，安装可能比较缓慢。
# pip install csrc/layer_norm
# pip install csrc/rotary
```
<br>
## 快速使用（Quickstart）
下面我们展示了一个使用Qwen-1.8B-Chat-Int4模型，进行多轮对话交互的样例：
We show an example of multi-turn interaction with Qwen-1.8B-Chat-Int4 in the following code:
```python
model = AutoModelForCausalLM.from_pretrained(
"Qwen/Qwen-1_8B-Chat-Int4",
device_map="auto",
trust_remote_code=True
).eval()
response, history = model.chat(tokenizer, "你好", history=None)
print(response)
# 你好！很高兴为你提供帮助。
# Qwen-1.8B-Chat现在可以通过调整系统指令（System Prompt），实现角色扮演，语言风格迁移，任务设定，行为设定等能力。
# Qwen-1.8B-Chat can realize roly playing, language style transfer, task setting, and behavior setting by system prompt.
response, _ = model.chat(tokenizer, "你好呀", history=None, system="请用二次元可爱语气和我说话")
print(response)
# 你好啊！我是一只可爱的二次元猫咪哦，不知道你有什么问题需要我帮忙解答吗？
response, _ = model.chat(tokenizer, "My colleague works diligently", history=None, system="You will write beautiful compliments according to needs")
print(response)
# Your colleague is an outstanding worker! Their dedication and hard work are truly inspiring. They always go above and beyond to ensure that
# their tasks are completed on time and to the highest standard. I am lucky to have them as a colleague, and I know I can count on them to handle any challenge that comes their way.
```
关于更多的使用说明，请参考我们的[GitHub repo](https://github.com/QwenLM/Qwen)获取更多信息。
For more information, please refer to our [GitHub repo](https://github.com/QwenLM/Qwen) for more information.
## Tokenizer
> 注：作为术语的“tokenization”在中文中尚无共识的概念对应，本文档采用英文表达以利说明。
基于tiktoken的分词器有别于其他分词器，比如sentencepiece分词器。尤其在微调阶段，需要特别注意特殊token的使用。关于tokenizer的更多信息，以及微调时涉及的相关使用，请参阅[文档](https://github.com/QwenLM/Qwen/blob/main/tokenization_note_zh.md)。
Our tokenizer based on tiktoken is different from other tokenizers, e.g., sentencepiece tokenizer. You need to pay attention to special tokens, especially in finetuning. For more detailed information on the tokenizer and related use in fine-tuning, please refer to the [documentation](https://github.com/QwenLM/Qwen/blob/main/tokenization_note.md).
## 量化 (Quantization)
### 用法 (Usage)
**请注意：我们更新量化方案为基于[AutoGPTQ](https://github.com/PanQiWei/AutoGPTQ)的量化，提供Qwen-1.8B-Chat的Int4量化模型[点击这里](https://huggingface.co/Qwen/Qwen-1_8B-Chat-Int4)。相比此前方案，该方案在模型评测效果几乎无损，且存储需求更低，推理速度更优。**
**Note: we provide a new solution based on [AutoGPTQ](https://github.com/PanQiWei/AutoGPTQ), and release an Int4 quantized model for Qwen-1.8B-Chat [Click here](https://huggingface.co/Qwen/Qwen-1_8B-Chat-Int4), which achieves nearly lossless model effects but improved performance on both memory costs and inference speed, in comparison with the previous solution.**
以下我们提供示例说明如何使用Int4量化模型。在开始使用前，请先保证满足要求（如torch 2.0及以上，transformers版本为4.32.0及以上，等等），并安装所需安装包：
Here we demonstrate how to use our provided quantized models for inference. Before you start, make sure you meet the requirements of auto-gptq (e.g., torch 2.0 and above, transformers 4.32.0 and above, etc.) and install the required packages:
```bash
pip install auto-gptq optimum
```
如安装`auto-gptq`遇到问题，我们建议您到官方[repo](https://github.com/PanQiWei/AutoGPTQ)搜索合适的预编译wheel。
随后即可使用和上述一致的用法调用量化模型：
If you meet problems installing `auto-gptq`, we advise you to check out the official [repo](https://github.com/PanQiWei/AutoGPTQ) to find a pre-build wheel.
Then you can load the quantized model easily and run inference as same as usual:
```python
model = AutoModelForCausalLM.from_pretrained(
"Qwen/Qwen-1_8B-Chat-Int4",
device_map="auto",
trust_remote_code=True
).eval()
response, history = model.chat(tokenizer, "你好", history=None)
```
### 效果评测
我们使用原始模型的FP32和BF16精度，以及量化过的Int8和Int4模型在基准评测上做了测试，结果如下所示：
We illustrate the model performance of both FP32, BF16, Int8 and Int4 models on the benchmark. Results are shown below:
| Quantization | MMLU | CEval (val) | GSM8K | Humaneval |
|--------------|:----:|:-----------:|:-----:|:---------:|
| FP32 | 43.4 | 57.0 | 33.0 | 26.8 |
| BF16 | 43.3 | 55.6 | 33.7 | 26.2 |
| Int8 | 43.1 | 55.8 | 33.0 | 27.4 |
| Int4 | 42.9 | 52.8 | 31.2 | 25.0 |
### 推理速度 (Inference Speed)
我们测算了FP32、BF16精度和Int8、Int4量化模型生成2048和8192个token的平均推理速度。如图所示：
We measured the average inference speed of generating 2048 and 8192 tokens under FP32, BF16 precision and Int8, Int4 quantization level, respectively.
| Quantization | FlashAttn | Speed (2048 tokens) | Speed (8192 tokens) |
|--------------| :-------: |:-------------------:|:-------------------:|
| FP32 | v2 | 52.96 | 47.35 |
| BF16 | v2 | 54.09 | 54.04 |
| Int8 | v2 | 55.56 | 55.62 |
| Int4 | v2 | 71.07 | 76.45 |
| FP32 | v1 | 52.00 | 45.80 |
| BF16 | v1 | 51.70 | 55.04 |
| Int8 | v1 | 53.16 | 53.33 |
| Int4 | v1 | 69.82 | 67.44 |
| FP32 | Disabled | 52.28 | 44.95 |
| BF16 | Disabled | 48.17 | 45.01 |
| Int8 | Disabled | 52.16 | 52.99 |
| Int4 | Disabled | 68.37 | 65.94 |
具体而言，我们记录在长度为1的上下文的条件下生成8192个token的性能。评测运行于单张A100-SXM4-80G GPU，使用PyTorch 2.0.1和CUDA 11.4。推理速度是生成8192个token的速度均值。
In detail, the setting of profiling is generating 8192 new tokens with 1 context token. The profiling runs on a single A100-SXM4-80G GPU with PyTorch 2.0.1 and CUDA 11.4. The inference speed is averaged over the generated 8192 tokens.
### 显存使用 (GPU Memory Usage)
我们测算了FP32、BF16精度和Int8、Int4量化模型生成2048个及8192个token（单个token作为输入）的峰值显存占用情况。结果如下所示：
We also profile the peak GPU memory usage for generating 2048 tokens and 8192 tokens (with single token as context) under FP32, BF16 or Int8, Int4 quantization level, respectively. The results are shown below.
| Quantization Level | Peak Usage for Encoding 2048 Tokens | Peak Usage for Generating 8192 Tokens |
|--------------------|:-----------------------------------:|:-------------------------------------:|
| FP32 | 8.45GB | 13.06GB |
| BF16 | 4.23GB | 6.48GB |
| Int8 | 3.48GB | 5.34GB |
| Int4 | 2.91GB | 4.80GB |
上述性能测算使用[此脚本](https://qianwen-res.oss-cn-beijing.aliyuncs.com/profile.py)完成。
The above speed and memory profiling are conducted using [this script](https://qianwen-res.oss-cn-beijing.aliyuncs.com/profile.py).
<br>
## 模型细节（Model）
与Qwen-1.8B预训练模型相同，Qwen-1.8B-Chat模型规模基本情况如下所示
The details of the model architecture of Qwen-1.8B-Chat are listed as follows
| Hyperparameter | Value |
|:----------------|:------:|
| n_layers | 24 |
| n_heads | 16 |
| d_model | 2048 |
| vocab size | 151851 |
| sequence length | 8192 |
在位置编码、FFN激活函数和normalization的实现方式上，我们也采用了目前最流行的做法，
即RoPE相对位置编码、SwiGLU激活函数、RMSNorm（可选安装flash-attention加速）。
在分词器方面，相比目前主流开源模型以中英词表为主，Qwen-1.8B-Chat使用了约15万token大小的词表。
该词表在GPT-4使用的BPE词表`cl100k_base`基础上，对中文、多语言进行了优化，在对中、英、代码数据的高效编解码的基础上，对部分多语言更加友好，方便用户在不扩展词表的情况下对部分语种进行能力增强。
词表对数字按单个数字位切分。调用较为高效的[tiktoken分词库](https://github.com/openai/tiktoken)进行分词。
For position encoding, FFN activation function, and normalization calculation methods, we adopt the prevalent practices, i.e., RoPE relative position encoding, SwiGLU for activation function, and RMSNorm for normalization (optional installation of flash-attention for acceleration).
For tokenization, compared to the current mainstream open-source models based on Chinese and English vocabularies, Qwen-1.8B-Chat uses a vocabulary of over 150K tokens.
It first considers efficient encoding of Chinese, English, and code data, and is also more friendly to multilingual languages, enabling users to directly enhance the capability of some languages without expanding the vocabulary.
It segments numbers by single digit, and calls the [tiktoken](https://github.com/openai/tiktoken) tokenizer library for efficient tokenization.
## 评测效果（Evaluation）
对于Qwen-1.8B-Chat模型，我们同样评测了常规的中文理解（C-Eval）、英文理解（MMLU）、代码（HumanEval）和数学（GSM8K）等权威任务，同时包含了长序列任务的评测结果。由于Qwen-1.8B-Chat模型经过对齐后，激发了较强的外部系统调用能力，我们还进行了工具使用能力方面的评测。
提示：由于硬件和框架造成的舍入误差，复现结果如有波动属于正常现象。
For Qwen-1.8B-Chat, we also evaluate the model on C-Eval, MMLU, HumanEval, GSM8K, etc., as well as the benchmark evaluation for long-context understanding, and tool usage.
Note: Due to rounding errors caused by hardware and framework, differences in reproduced results are possible.
### 中文评测（Chinese Evaluation）
#### C-Eval
在[C-Eval](https://arxiv.org/abs/2305.08322)验证集上，我们评价了Qwen-1.8B-Chat模型的准确率
We demonstrate the accuracy of Qwen-1.8B-Chat on C-Eval validation set
| Model | Acc. |
|:--------------------------------:|:---------:|
| RedPajama-INCITE-Chat-3B | 18.3 |
| OpenBuddy-3B | 23.5 |
| Firefly-Bloom-1B4 | 23.6 |
| OpenLLaMA-Chinese-3B | 24.4 |
| LLaMA2-7B-Chat | 31.9 |
| ChatGLM2-6B-Chat | 52.6 |
| InternLM-7B-Chat | 53.6 |
| **Qwen-1.8B-Chat (0-shot)** | 55.6 |
| **Qwen-7B-Chat (0-shot)** | 59.7 |
| **Qwen-7B-Chat (5-shot)** | 59.3 |
C-Eval测试集上，Qwen-1.8B-Chat模型的zero-shot准确率结果如下：
The zero-shot accuracy of Qwen-1.8B-Chat on C-Eval testing set is provided below:
| Model | Avg. | STEM | Social Sciences | Humanities | Others |
| :---------------------: | :------: | :--: | :-------------: | :--------: | :----: |
| Chinese-Alpaca-Plus-13B | 41.5 | 36.6 | 49.7 | 43.1 | 41.2 |
| Chinese-Alpaca-2-7B | 40.3 | - | - | - | - |
| ChatGLM2-6B-Chat | 50.1 | 46.4 | 60.4 | 50.6 | 46.9 |
| Baichuan-13B-Chat | 51.5 | 43.7 | 64.6 | 56.2 | 49.2 |
| **Qwen-1.8B-Chat** | 53.8 | 48.4 | 68.0 | 56.5 | 48.3 |
| **Qwen-7B-Chat** | 58.6 | 53.3 | 72.1 | 62.8 | 52.0 |
### 英文评测（English Evaluation）
#### MMLU
[MMLU](https://arxiv.org/abs/2009.03300)评测集上，Qwen-1.8B-Chat模型的准确率如下，效果同样在同类对齐模型中同样表现较优。
The accuracy of Qwen-1.8B-Chat on MMLU is provided below.
The performance of Qwen-1.8B-Chat still on the top between other human-aligned models with comparable size.
| Model | Acc. |
|:--------------------------------:|:---------:|
| Firefly-Bloom-1B4 | 23.8 |
| OpenBuddy-3B | 25.5 |
| RedPajama-INCITE-Chat-3B | 25.5 |
| OpenLLaMA-Chinese-3B | 25.7 |
| ChatGLM2-6B-Chat | 46.0 |
| LLaMA2-7B-Chat | 46.2 |
| InternLM-7B-Chat | 51.1 |
| Baichuan2-7B-Chat | 52.9 |
| **Qwen-1.8B-Chat (0-shot)** | 43.3 |
| **Qwen-7B-Chat (0-shot)** | 55.8 |
| **Qwen-7B-Chat (5-shot)** | 57.0 |
### 代码评测（Coding Evaluation）
Qwen-1.8B-Chat在[HumanEval](https://github.com/openai/human-eval)的zero-shot Pass@1效果如下
The zero-shot Pass@1 of Qwen-1.8B-Chat on [HumanEval](https://github.com/openai/human-eval) is demonstrated below
| Model | Pass@1 |
|:------------------------:|:------:|
| Firefly-Bloom-1B4 | 0.6 |
| OpenLLaMA-Chinese-3B | 4.9 |
| RedPajama-INCITE-Chat-3B | 6.1 |
| OpenBuddy-3B | 10.4 |
| ChatGLM2-6B-Chat | 11.0 |
| LLaMA2-7B-Chat | 12.2 |
| Baichuan2-7B-Chat | 13.4 |
| InternLM-7B-Chat | 14.6 |
| **Qwen-1.8B-Chat** | 26.2 |
| **Qwen-7B-Chat** | 37.2 |
### 数学评测（Mathematics Evaluation）
在评测数学能力的[GSM8K](https://github.com/openai/grade-school-math)上，Qwen-1.8B-Chat的准确率结果如下
The accuracy of Qwen-1.8B-Chat on GSM8K is shown below
| Model | Acc. |
|:------------------------------------:|:--------:|
| Firefly-Bloom-1B4 | 2.4 |
| RedPajama-INCITE-Chat-3B | 2.5 |
| OpenLLaMA-Chinese-3B | 3.0 |
| OpenBuddy-3B | 12.6 |
| LLaMA2-7B-Chat | 26.3 |
| ChatGLM2-6B-Chat | 28.8 |
| Baichuan2-7B-Chat | 32.8 |
| InternLM-7B-Chat | 33.0 |
| **Qwen-1.8B-Chat (0-shot)** | 33.7 |
| **Qwen-7B-Chat (0-shot)** | 50.3 |
| **Qwen-7B-Chat (8-shot)** | 54.1 |
## 评测复现（Reproduction）
我们提供了评测脚本，方便大家复现模型效果，详见[链接](https://github.com/QwenLM/Qwen/tree/main/eval)。提示：由于硬件和框架造成的舍入误差，复现结果如有小幅波动属于正常现象。
We have provided evaluation scripts to reproduce the performance of our model, details as [link](https://github.com/QwenLM/Qwen/tree/main/eval).
<br>
## FAQ
如遇到问题，敬请查阅[FAQ](https://github.com/QwenLM/Qwen/blob/main/FAQ_zh.md)以及issue区，如仍无法解决再提交issue。
If you meet problems, please refer to [FAQ](https://github.com/QwenLM/Qwen/blob/main/FAQ.md) and the issues first to search a solution before you launch a new issue.
<br>
## 引用 (Citation)
如果你觉得我们的工作对你有帮助，欢迎引用！
If you find our work helpful, feel free to give us a cite.
```
@article{qwen,
title={Qwen Technical Report},
author={Jinze Bai and Shuai Bai and Yunfei Chu and Zeyu Cui and Kai Dang and Xiaodong Deng and Yang Fan and Wenbin Ge and Yu Han and Fei Huang and Binyuan Hui and Luo Ji and Mei Li and Junyang Lin and Runji Lin and Dayiheng Liu and Gao Liu and Chengqiang Lu and Keming Lu and Jianxin Ma and Rui Men and Xingzhang Ren and Xuancheng Ren and Chuanqi Tan and Sinan Tan and Jianhong Tu and Peng Wang and Shijie Wang and Wei Wang and Shengguang Wu and Benfeng Xu and Jin Xu and An Yang and Hao Yang and Jian Yang and Shusheng Yang and Yang Yao and Bowen Yu and Hongyi Yuan and Zheng Yuan and Jianwei Zhang and Xingxuan Zhang and Yichang Zhang and Zhenru Zhang and Chang Zhou and Jingren Zhou and Xiaohuan Zhou and Tianhang Zhu},
journal={arXiv preprint arXiv:2309.16609},
year={2023}
}
```
<br>
## 使用协议（License Agreement）
我们的代码和模型权重对学术研究完全开放。请查看[LICENSE](https://github.com/QwenLM/Qwen/blob/main/Tongyi%20Qianwen%20RESEARCH%20LICENSE%20AGREEMENT)文件了解具体的开源协议细节。如需商用，请联系我们。
Our code and checkpoints are open to research purpose. Check the [LICENSE](https://github.com/QwenLM/Qwen/blob/main/Tongyi%20Qianwen%20RESEARCH%20LICENSE%20AGREEMENT) for more details about the license. For commercial use, please contact us.
<br>
## 联系我们（Contact Us）
如果你想给我们的研发团队和产品团队留言，欢迎加入我们的微信群、钉钉群以及Discord！同时，也欢迎通过邮件（[email protected]）联系我们。
If you are interested to leave a message to either our research team or product team, join our Discord or WeChat groups! Also, feel free to send an email to [email protected].
|
TeeZee/DarkForest-20B-v1.0 | TeeZee | 2024-01-28T02:25:09Z | 455 | 1 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"merge",
"not-for-all-audiences",
"license:other",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-01-22T20:02:21Z | ---
license: other
license_name: microsoft-research-license
tags:
- merge
- not-for-all-audiences
---
# DarkForest 20B v1.0
## Model Details
- To create this model two step procedure was used. First a new 20B model was created using [microsoft/Orca-2-13b](https://huggingface.co/microsoft/Orca-2-13b)
and [KoboldAI/LLaMA2-13B-Erebus-v3](https://huggingface.co/KoboldAI/LLaMA2-13B-Erebus-v3) , deatils of the merge in [mergekit-config_step1.yml](https://huggingface.co/TeeZee/DarkForest-20B-v1.0/resolve/main/mergekit-config_step1.yml)
- then [jebcarter/psyonic-cetacean-20B](https://huggingface.co/jebcarter/psyonic-cetacean-20B) was used to produce the final model, merge config in [mergekit-config_step2.yml](https://huggingface.co/TeeZee/DarkForest-20B-v1.0/resolve/main/mergekit-config_step2.yml)
- The resulting model has approximately 20 billion parameters.
**Warning: This model can produce NSFW content!**
## Results
- produces SFW nad NSFW content without issues, switches context seamlessly.
- good at following instructions.
- good at tracking multiple characters in one scene.
- very creative, scenarios produced are mature and complicated, model doesn't shy from writing about PTSD, menatal issues or complicated relationships.
- NSFW output is more creative and suprising than typical limaRP output.
- definitely for mature audiences, not only because of vivid NSFW content but also because of overall maturity of stories it produces.
- This is NOT Harry Potter level storytelling.
## GGUF
[TeeZee/DarkForest-20B-v1.0-GGUF](https://huggingface.co/TeeZee/DarkForest-20B-v1.0-GGUF)
All comments are greatly appreciated, download, test and if you appreciate my work, consider buying me my fuel:
<a href="https://www.buymeacoffee.com/TeeZee" target="_blank"><img src="https://cdn.buymeacoffee.com/buttons/v2/default-yellow.png" alt="Buy Me A Coffee" style="height: 60px !important;width: 217px !important;" ></a> |
wolfram/miqu-1-120b | wolfram | 2024-03-05T19:49:44Z | 455 | 47 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"mergekit",
"merge",
"conversational",
"en",
"de",
"fr",
"es",
"it",
"base_model:152334H/miqu-1-70b-sf",
"license:other",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-02-03T22:48:14Z | ---
base_model:
- 152334H/miqu-1-70b-sf
language:
- en
- de
- fr
- es
- it
library_name: transformers
tags:
- mergekit
- merge
license: other
---
# miqu-1-120b
- EXL2: [2.4bpw](https://huggingface.co/LoneStriker/wolfram_miqu-1-120b-2.4bpw-h6-exl2) | [2.65bpw](https://huggingface.co/LoneStriker/wolfram_miqu-1-120b-2.65bpw-h6-exl2) | [3.0bpw](https://huggingface.co/LoneStriker/wolfram_miqu-1-120b-3.0bpw-h6-exl2) | [4.0bpw](https://huggingface.co/LoneStriker/wolfram_miqu-1-120b-4.0bpw-h6-exl2) | [5.0bpw](https://huggingface.co/LoneStriker/wolfram_miqu-1-120b-5.0bpw-h6-exl2)
- GGUF: [Q2_K-Q5_K_M](https://huggingface.co/LoneStriker/wolfram_miqu-1-120b-GGUF/) | [IQ3_XXS](https://huggingface.co/wolfram/miqu-1-120b-GGUF)
This is a 120b frankenmerge of [miqu-1-70b](https://huggingface.co/miqudev/miqu-1-70b) created by interleaving layers of [miqu-1-70b-sf](https://huggingface.co/152334H/miqu-1-70b-sf) with itself using [mergekit](https://github.com/cg123/mergekit).
Inspired by [Venus-120b-v1.2](https://huggingface.co/nsfwthrowitaway69/Venus-120b-v1.2), [MegaDolphin-120b](https://huggingface.co/cognitivecomputations/MegaDolphin-120b), and [goliath-120b](https://huggingface.co/alpindale/goliath-120b).
Thanks for the support, [CopilotKit](https://github.com/CopilotKit/CopilotKit) - the open-source platform for building in-app AI Copilots into any product, with any LLM model. Check out their GitHub.
Thanks for the EXL2 and GGUF quants, [Lone Striker](https://huggingface.co/LoneStriker)!
Also available: [miquliz-120b-v2.0](https://huggingface.co/wolfram/miquliz-120b-v2.0) – Miqu's younger, fresher sister; a new and improved Goliath-like merge of Miqu and lzlv.
## Review
u/SomeOddCodeGuy wrote on r/LocalLLaMA:
> I love this model. It's slow as Christmas but it's SO GOOD. You did great on this.
>
> But this model is close to getting me to shut down my ChatGPT 4 subscription lol. Between it, Deepseek and a couple others, I'm not sure I'll be using ChatGPT much anymore.
>
> Im using the Q8 at 16k, and I can't express how true it remains to its context. I might try to do some testing this weekend, but its great so far.
>
> I've been using your miqu-1 the past two days and its phenomenal. It understands everything I'm saying in ways only ChatGPT did. I've been purposefully getting more and more vague/relaxed in my speaking, and talking about the most inane stuff, and it just follows right along like a person would.
>
> Miqu-1 does ignore instructions a little. I tried to make a more sarcastic/insulting AI assistant to chat with, and specifically told it (multiple times after a few tries) to not apologize to me after, and it wouldn't stop. So if it made a jab like "Wow, great work spelling that word. Quite the whiz kid huh?", making fun of me for misspelling something, it would refuse to not follow up with "Seriously, though, sometimes misspellings happen" lol. But that's the only issue I've had with it.
(Note: All I did was merge this, though, so the credit mostly belongs to [Mistral AI](https://mistral.ai/) (giving proper attribution!) and the creators of [mergekit](https://github.com/arcee-ai/mergekit) as well as [Venus-120b-v1.2](https://huggingface.co/nsfwthrowitaway69/Venus-120b-v1.2) and [MegaDolphin-120b](https://huggingface.co/cognitivecomputations/MegaDolphin-120b) who inspired it.)
## Model Details
- Max Context: 32764 tokens (kept the weird number from the original/base model)
- Layers: 140
### Prompt template: Mistral
```
<s>[INST] {prompt} [/INST]
```
See also: [🐺🐦⬛ LLM Prompt Format Comparison/Test: Mixtral 8x7B Instruct with **17** different instruct templates : LocalLLaMA](https://www.reddit.com/r/LocalLLaMA/comments/18ljvxb/llm_prompt_format_comparisontest_mixtral_8x7b/)
## Merge Details
### Merge Method
This model was merged using the passthrough merge method.
### Models Merged
The following models were included in the merge:
- [152334H/miqu-1-70b-sf](https://huggingface.co/152334H/miqu-1-70b-sf)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
dtype: float16
merge_method: passthrough
slices:
- sources:
- layer_range: [0, 20]
model: 152334H/miqu-1-70b-sf
- sources:
- layer_range: [10, 30]
model: 152334H/miqu-1-70b-sf
- sources:
- layer_range: [20, 40]
model: 152334H/miqu-1-70b-sf
- sources:
- layer_range: [30, 50]
model: 152334H/miqu-1-70b-sf
- sources:
- layer_range: [40, 60]
model: 152334H/miqu-1-70b-sf
- sources:
- layer_range: [50, 70]
model: 152334H/miqu-1-70b-sf
- sources:
- layer_range: [60, 80]
model: 152334H/miqu-1-70b-sf
```
## Credits & Special Thanks
- original (unreleased) model: [mistralai (Mistral AI_)](https://huggingface.co/mistralai)
- ⭐⭐⭐ **[Use their newer, better, official models here!](https://console.mistral.ai/)** ⭐⭐⭐
- leaked model: [miqudev/miqu-1-70b](https://huggingface.co/miqudev/miqu-1-70b)
- f16 model: [152334H/miqu-1-70b-sf](https://huggingface.co/152334H/miqu-1-70b-sf)
- mergekit: [arcee-ai/mergekit: Tools for merging pretrained large language models.](https://github.com/arcee-ai/mergekit)
- mergekit_config.yml: [nsfwthrowitaway69/Venus-120b-v1.2](https://huggingface.co/nsfwthrowitaway69/Venus-120b-v1.2)
### Support
- [My Ko-fi page](https://ko-fi.com/wolframravenwolf) if you'd like to tip me to say thanks or request specific models to be tested or merged with priority. Also consider supporting your favorite model creators, quantizers, or frontend/backend devs if you can afford to do so. They deserve it!
## Disclaimer
*This model contains leaked weights and due to its content it should not be used by anyone.* 😜
But seriously:
### License
**What I *know*:** [Weights produced by a machine are not copyrightable](https://www.reddit.com/r/LocalLLaMA/comments/1amc080/psa_if_you_use_miqu_or_a_derivative_please_keep/kpmamte/) so there is no copyright owner who could grant permission or a license to use, or restrict usage, once you have acquired the files.
### Ethics
**What I *believe*:** All generative AI, including LLMs, only exists because it is trained mostly on human data (both public domain and copyright-protected, most likely acquired without express consent) and possibly synthetic data (which is ultimately derived from human data, too). It is only fair if something that is based on everyone's knowledge and data is also freely accessible to the public, the actual creators of the underlying content. Fair use, fair AI! |
Lewdiculous/Eris_PrimeV4-Vision-32k-7B-GGUF-IQ-Imatrix | Lewdiculous | 2024-03-27T11:01:37Z | 455 | 9 | null | [
"gguf",
"quantized",
"roleplay",
"multimodal",
"vision",
"llava",
"sillytavern",
"merge",
"mistral",
"conversational",
"license:other",
"region:us"
]
| null | 2024-03-27T02:53:08Z | ---
license: other
inference: false
tags:
- gguf
- quantized
- roleplay
- multimodal
- vision
- llava
- sillytavern
- merge
- mistral
- conversational
---
# #Roleplay #Multimodal #Vision
This repository hosts GGUF-IQ-Imatrix quants for [Nitral-AI/Eris_PrimeV4-Vision-32k-7B](https://huggingface.co/Nitral-AI/Eris_PrimeV4-Vision-32k-7B).
"More stable and with better long context handling."
**Recommended starting [SillyTavern presets here](https://huggingface.co/Lewdiculous/Eris_PrimeV4-Vision-32k-7B-GGUF-IQ-Imatrix/tree/main/sillytavern-presets-lewdicu-3.0.2-mistral-0.2).**
This is a **#multimodal** model that also has **#vision** capabilities. <br> Read the full card information if you also want to use that functionality.
Quants:
```python
quantization_options = [
"Q4_K_M", "Q4_K_S", "IQ4_XS", "Q5_K_M", "Q5_K_S",
"Q6_K", "Q8_0", "IQ3_M", "IQ3_S", "IQ3_XXS"
]
```
**What does "Imatrix" mean?**
<details><summary>
⇲ Click here to expand/hide more information about this topic.
</summary>
It stands for **Importance Matrix**, a technique used to improve the quality of quantized models.
The **Imatrix** is calculated based on calibration data, and it helps determine the importance of different model activations during the quantization process.
The idea is to preserve the most important information during quantization, which can help reduce the loss of model performance, especially when the calibration data is diverse.
[[1]](https://github.com/ggerganov/llama.cpp/discussions/5006) [[2]](https://github.com/ggerganov/llama.cpp/discussions/5263#discussioncomment-8395384)
For imatrix data generation, kalomaze's `groups_merged.txt` with added roleplay chats was used, you can find it [here](https://huggingface.co/Lewdiculous/Datura_7B-GGUF-Imatrix/blob/main/imatrix-with-rp-format-data.txt). This was just to add a bit more diversity to the data.
</details><br>
# Vision/multimodal capabilities:
<details><summary>
⇲ Click here to expand/hide how this would work in practice in a roleplay chat.
</summary>
</details><br>
<details><summary>
⇲ Click here to expand/hide what your SillyTavern Image Captions extension settings should look like.
</summary>
</details><br>
**If you want to use vision functionality:**
* Make sure you are using the latest version of [KoboldCpp](https://github.com/LostRuins/koboldcpp).
To use the multimodal capabilities of this model, such as **vision**, you also need to load the specified **mmproj** file, you can get it [here](https://huggingface.co/cjpais/llava-1.6-mistral-7b-gguf/blob/main/mmproj-model-f16.gguf) or as uploaded in the repository.
* You can load the **mmproj** by using the corresponding section in the interface:
* For CLI users, you can load the **mmproj file** by adding the respective flag to your usual command:
```
--mmproj your-mmproj-file.gguf
```
# Quantization information:
<details><summary>
⇲ Click here to expand/hide more information about this topic.
</summary>
**Steps performed:**
```
Base⇢ GGUF(F16)⇢ Imatrix-Data(F16)⇢ GGUF(Imatrix-Quants)
```
*Using the latest llama.cpp at the time.*
</details><br>
# Original model information:
# Eris Prime: Version 4.0 32k
After many trials and tribulations we have a winner: A more coherent, format stable version of Eris Prime v4 with better long context handling. |
mradermacher/Mermaid_PythonCoder-GGUF | mradermacher | 2024-05-06T04:40:17Z | 455 | 0 | transformers | [
"transformers",
"gguf",
"en",
"base_model:TroyDoesAI/Mermaid_PythonCoder",
"license:cc-by-4.0",
"endpoints_compatible",
"region:us"
]
| null | 2024-04-18T16:01:58Z | ---
base_model: TroyDoesAI/Mermaid_PythonCoder
language:
- en
library_name: transformers
license: cc-by-4.0
quantized_by: mradermacher
---
## About
<!-- ### quantize_version: 1 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: -->
<!-- ### vocab_type: -->
static quants of https://huggingface.co/TroyDoesAI/Mermaid_PythonCoder
<!-- provided-files -->
weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion.
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.Q2_K.gguf) | Q2_K | 4.1 | |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.IQ3_XS.gguf) | IQ3_XS | 4.5 | |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.Q3_K_S.gguf) | Q3_K_S | 4.8 | |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.IQ3_S.gguf) | IQ3_S | 4.8 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.IQ3_M.gguf) | IQ3_M | 4.9 | |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.Q3_K_M.gguf) | Q3_K_M | 5.3 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.Q3_K_L.gguf) | Q3_K_L | 5.8 | |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.IQ4_XS.gguf) | IQ4_XS | 5.9 | |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.Q4_K_S.gguf) | Q4_K_S | 6.2 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.Q4_K_M.gguf) | Q4_K_M | 6.6 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.Q5_K_S.gguf) | Q5_K_S | 7.5 | |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.Q5_K_M.gguf) | Q5_K_M | 7.7 | |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.Q6_K.gguf) | Q6_K | 8.9 | very good quality |
| [GGUF](https://huggingface.co/mradermacher/Mermaid_PythonCoder-GGUF/resolve/main/Mermaid_PythonCoder.Q8_0.gguf) | Q8_0 | 11.5 | fast, best quality |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time.
<!-- end -->
|
HuggingFaceFW/ablation-model-c4 | HuggingFaceFW | 2024-04-25T08:34:07Z | 455 | 3 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"arxiv:1910.09700",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-04-20T23:15:50Z | ---
library_name: transformers
license: apache-2.0
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
CortexLM/btlm-7b-base-v0.1 | CortexLM | 2024-06-13T22:07:41Z | 455 | 1 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"bittensor",
"decentralization",
"subnet 9",
"conversational",
"dataset:tiiuae/falcon-refinedweb",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-05-14T21:19:41Z | ---
license: mit
library_name: transformers
tags:
- bittensor
- decentralization
- subnet 9
datasets:
- tiiuae/falcon-refinedweb
---
<img src="https://cdn-uploads.huggingface.co/production/uploads/655a0bdf3ff5ba1b1b1c01b7/y1dKBZh8UhII6wtbs5boj.png" alt="drawing" width="512"/>
# 🚀 **BTLM-7B v0.1**
BTLM (Bittensor Language Model) is a collection of pretrained generative text models. This is the repository for the 7B pretrained model, optimized for dialogue use cases and converted for the Hugging Face Transformers format.
### Model Details
Bittensor's decentralized subnet 9 facilitated the development and release of the first version of the BTLM-7B model. This initial release comprises a sophisticated large language model designed for a variety of applications.In creating this model, significant effort was made to ensure its effectiveness and safety, setting a new standard in the decentralized open-source AI community.
⛔ **This is a pretrained model, which should be further finetuned for most usecases.**
**Training subnetwork :** 9
**Checkpoint :** 03-05-2024
[**Subnet 9 Network Leaderboard**](https://huggingface.co/spaces/macrocosm-os/pretraining-leaderboard)
[**Top Bittensor Model Checkpoint**](https://huggingface.co/tensorplex-labs/pretraining-sn9-7B-1)
### Inference
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch
model = "CortexLM/btlm-7b-base-v0.1"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
torch_dtype=torch.bfloat16,
)
sequences = pipeline(
"Tell me about decentralization.",
max_length=200,
do_sample=True,
top_k=10,
num_return_sequences=1,
eos_token_id=tokenizer.eos_token_id,
)
for seq in sequences:
print(f"Result: {seq['generated_text']}")
```
### Benchmark
| Average | ARC | HellaSwag | MMLU | TruthfulQA | Winogrande | GSM8K |
| --- | --- | --- | --- | --- | --- | --- |
| 43.32 | 45.65 | 58.29 | 44.26 | 30.45 | 70.88 | 10.39 |
[LM Evaluation Harness Repository](https://github.com/EleutherAI/lm-evaluation-harness)
## License
BTLM-7B is licensed under the [MIT License](https://opensource.org/license/mit), a permissive license that allows for reuse with virtually no restrictions. |
RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf | RichardErkhov | 2024-05-22T03:13:46Z | 455 | 0 | null | [
"gguf",
"region:us"
]
| null | 2024-05-22T00:32:57Z | Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
openbuddy-mistral2-7b-v20.3-32k - GGUF
- Model creator: https://huggingface.co/OpenBuddy/
- Original model: https://huggingface.co/OpenBuddy/openbuddy-mistral2-7b-v20.3-32k/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [openbuddy-mistral2-7b-v20.3-32k.Q2_K.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q2_K.gguf) | Q2_K | 2.55GB |
| [openbuddy-mistral2-7b-v20.3-32k.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.IQ3_XS.gguf) | IQ3_XS | 2.83GB |
| [openbuddy-mistral2-7b-v20.3-32k.IQ3_S.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.IQ3_S.gguf) | IQ3_S | 2.99GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q3_K_S.gguf) | Q3_K_S | 2.97GB |
| [openbuddy-mistral2-7b-v20.3-32k.IQ3_M.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.IQ3_M.gguf) | IQ3_M | 3.08GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q3_K.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q3_K.gguf) | Q3_K | 3.3GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q3_K_M.gguf) | Q3_K_M | 3.3GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q3_K_L.gguf) | Q3_K_L | 3.58GB |
| [openbuddy-mistral2-7b-v20.3-32k.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.IQ4_XS.gguf) | IQ4_XS | 3.7GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q4_0.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q4_0.gguf) | Q4_0 | 3.85GB |
| [openbuddy-mistral2-7b-v20.3-32k.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.IQ4_NL.gguf) | IQ4_NL | 3.89GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q4_K_S.gguf) | Q4_K_S | 3.88GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q4_K.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q4_K.gguf) | Q4_K | 4.09GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q4_K_M.gguf) | Q4_K_M | 4.09GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q4_1.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q4_1.gguf) | Q4_1 | 4.27GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q5_0.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q5_0.gguf) | Q5_0 | 4.68GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q5_K_S.gguf) | Q5_K_S | 4.68GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q5_K.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q5_K.gguf) | Q5_K | 4.81GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q5_K_M.gguf) | Q5_K_M | 4.81GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q5_1.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q5_1.gguf) | Q5_1 | 5.1GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q6_K.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q6_K.gguf) | Q6_K | 5.56GB |
| [openbuddy-mistral2-7b-v20.3-32k.Q8_0.gguf](https://huggingface.co/RichardErkhov/OpenBuddy_-_openbuddy-mistral2-7b-v20.3-32k-gguf/blob/main/openbuddy-mistral2-7b-v20.3-32k.Q8_0.gguf) | Q8_0 | 7.2GB |
Original model description:
---
language:
- zh
- en
- fr
- de
- ja
- ko
- it
- ru
pipeline_tag: text-generation
inference: false
library_name: transformers
license: apache-2.0
---
# OpenBuddy - Open Multilingual Chatbot
GitHub and Usage Guide: [https://github.com/OpenBuddy/OpenBuddy](https://github.com/OpenBuddy/OpenBuddy)
Website and Demo: [https://openbuddy.ai](https://openbuddy.ai)
Evaluation result of this model: [Evaluation.txt](Evaluation.txt)
# Copyright Notice
Base model: https://huggingface.co/mistralai/Mistral-7B-v0.2
License: Apache 2.0
## Disclaimer
All OpenBuddy models have inherent limitations and may potentially produce outputs that are erroneous, harmful, offensive, or otherwise undesirable. Users should not use these models in critical or high-stakes situations that may lead to personal injury, property damage, or significant losses. Examples of such scenarios include, but are not limited to, the medical field, controlling software and hardware systems that may cause harm, and making important financial or legal decisions.
OpenBuddy is provided "as-is" without any warranty of any kind, either express or implied, including, but not limited to, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement. In no event shall the authors, contributors, or copyright holders be liable for any claim, damages, or other liabilities, whether in an action of contract, tort, or otherwise, arising from, out of, or in connection with the software or the use or other dealings in the software.
By using OpenBuddy, you agree to these terms and conditions, and acknowledge that you understand the potential risks associated with its use. You also agree to indemnify and hold harmless the authors, contributors, and copyright holders from any claims, damages, or liabilities arising from your use of OpenBuddy.
## 免责声明
所有OpenBuddy模型均存在固有的局限性，可能产生错误的、有害的、冒犯性的或其他不良的输出。用户在关键或高风险场景中应谨慎行事，不要使用这些模型，以免导致人身伤害、财产损失或重大损失。此类场景的例子包括但不限于医疗领域、可能导致伤害的软硬件系统的控制以及进行重要的财务或法律决策。
OpenBuddy按“原样”提供，不附带任何种类的明示或暗示的保证，包括但不限于适销性、特定目的的适用性和非侵权的暗示保证。在任何情况下，作者、贡献者或版权所有者均不对因软件或使用或其他软件交易而产生的任何索赔、损害赔偿或其他责任（无论是合同、侵权还是其他原因）承担责任。
使用OpenBuddy即表示您同意这些条款和条件，并承认您了解其使用可能带来的潜在风险。您还同意赔偿并使作者、贡献者和版权所有者免受因您使用OpenBuddy而产生的任何索赔、损害赔偿或责任的影响。
|
kimdeokgi/ko-pt-model-test1 | kimdeokgi | 2024-06-04T05:14:01Z | 455 | 0 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"conversational",
"kor",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-06-04T04:39:48Z | ---
license: apache-2.0
language:
- kor
---
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
NikolayKozloff/Morfoz-LLM-8b-v1.0-Q4_0-GGUF | NikolayKozloff | 2024-06-24T14:32:52Z | 455 | 1 | null | [
"gguf",
"llama-cpp",
"gguf-my-repo",
"tr",
"base_model:Morfoz-Aigap/Morfoz-LLM-8b-v1.0",
"license:apache-2.0",
"region:us"
]
| null | 2024-06-24T14:32:31Z | ---
base_model: Morfoz-Aigap/Morfoz-LLM-8b-v1.0
language:
- tr
license: apache-2.0
tags:
- llama-cpp
- gguf-my-repo
---
# NikolayKozloff/Morfoz-LLM-8b-v1.0-Q4_0-GGUF
This model was converted to GGUF format from [`Morfoz-Aigap/Morfoz-LLM-8b-v1.0`](https://huggingface.co/Morfoz-Aigap/Morfoz-LLM-8b-v1.0) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Morfoz-Aigap/Morfoz-LLM-8b-v1.0) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo NikolayKozloff/Morfoz-LLM-8b-v1.0-Q4_0-GGUF --hf-file morfoz-llm-8b-v1.0-q4_0.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo NikolayKozloff/Morfoz-LLM-8b-v1.0-Q4_0-GGUF --hf-file morfoz-llm-8b-v1.0-q4_0.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo NikolayKozloff/Morfoz-LLM-8b-v1.0-Q4_0-GGUF --hf-file morfoz-llm-8b-v1.0-q4_0.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo NikolayKozloff/Morfoz-LLM-8b-v1.0-Q4_0-GGUF --hf-file morfoz-llm-8b-v1.0-q4_0.gguf -c 2048
```
|
tner/xlm-roberta-base-uncased-fin | tner | 2021-02-12T23:47:27Z | 454 | 0 | transformers | [
"transformers",
"pytorch",
"xlm-roberta",
"token-classification",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
]
| token-classification | 2022-03-02T23:29:05Z | # XLM-RoBERTa for NER
XLM-RoBERTa finetuned on NER. Check more detail at [TNER repository](https://github.com/asahi417/tner).
## Usage
```
from transformers import AutoTokenizer, AutoModelForTokenClassification
tokenizer = AutoTokenizer.from_pretrained("asahi417/tner-xlm-roberta-base-uncased-fin")
model = AutoModelForTokenClassification.from_pretrained("asahi417/tner-xlm-roberta-base-uncased-fin")
``` |
EleutherAI/pythia-410m-deduped-v0 | EleutherAI | 2023-07-10T01:31:39Z | 454 | 6 | transformers | [
"transformers",
"pytorch",
"safetensors",
"gpt_neox",
"text-generation",
"causal-lm",
"pythia",
"pythia_v0",
"en",
"dataset:EleutherAI/the_pile_deduplicated",
"arxiv:2101.00027",
"arxiv:2201.07311",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2022-11-01T00:48:44Z | ---
language:
- en
tags:
- pytorch
- causal-lm
- pythia
- pythia_v0
license: apache-2.0
datasets:
- EleutherAI/the_pile_deduplicated
---
The *Pythia Scaling Suite* is a collection of models developed to facilitate
interpretability research. It contains two sets of eight models of sizes
70M, 160M, 410M, 1B, 1.4B, 2.8B, 6.9B, and 12B. For each size, there are two
models: one trained on the Pile, and one trained on the Pile after the dataset
has been globally deduplicated. All 8 model sizes are trained on the exact
same data, in the exact same order. All Pythia models are available
[on Hugging Face](https://huggingface.co/models?other=pythia).
The Pythia model suite was deliberately designed to promote scientific
research on large language models, especially interpretability research.
Despite not centering downstream performance as a design goal, we find the
models <a href="#evaluations">match or exceed</a> the performance of
similar and same-sized models, such as those in the OPT and GPT-Neo suites.
Please note that all models in the *Pythia* suite were renamed in January
2023. For clarity, a <a href="#naming-convention-and-parameter-count">table
comparing the old and new names</a> is provided in this model card, together
with exact parameter counts.
## Pythia-410M-deduped
### Model Details
- Developed by: [EleutherAI](http://eleuther.ai)
- Model type: Transformer-based Language Model
- Language: English
- Learn more: [Pythia's GitHub repository](https://github.com/EleutherAI/pythia)
for training procedure, config files, and details on how to use.
- Library: [GPT-NeoX](https://github.com/EleutherAI/gpt-neox)
- License: Apache 2.0
- Contact: to ask questions about this model, join the [EleutherAI
Discord](https://discord.gg/zBGx3azzUn), and post them in `#release-discussion`.
Please read the existing *Pythia* documentation before asking about it in the
EleutherAI Discord. For general correspondence: [contact@eleuther.
ai](mailto:[email protected]).
<figure>
| Pythia model | Non-Embedding Params | Layers | Model Dim | Heads | Batch Size | Learning Rate | Equivalent Models |
| -----------: | -------------------: | :----: | :-------: | :---: | :--------: | :-------------------: | :--------------------: |
| 70M | 18,915,328 | 6 | 512 | 8 | 2M | 1.0 x 10<sup>-3</sup> | — |
| 160M | 85,056,000 | 12 | 768 | 12 | 4M | 6.0 x 10<sup>-4</sup> | GPT-Neo 125M, OPT-125M |
| 410M | 302,311,424 | 24 | 1024 | 16 | 4M | 3.0 x 10<sup>-4</sup> | OPT-350M |
| 1.0B | 805,736,448 | 16 | 2048 | 8 | 2M | 3.0 x 10<sup>-4</sup> | — |
| 1.4B | 1,208,602,624 | 24 | 2048 | 16 | 4M | 2.0 x 10<sup>-4</sup> | GPT-Neo 1.3B, OPT-1.3B |
| 2.8B | 2,517,652,480 | 32 | 2560 | 32 | 2M | 1.6 x 10<sup>-4</sup> | GPT-Neo 2.7B, OPT-2.7B |
| 6.9B | 6,444,163,072 | 32 | 4096 | 32 | 2M | 1.2 x 10<sup>-4</sup> | OPT-6.7B |
| 12B | 11,327,027,200 | 36 | 5120 | 40 | 2M | 1.2 x 10<sup>-4</sup> | — |
<figcaption>Engineering details for the <i>Pythia Suite</i>. Deduped and
non-deduped models of a given size have the same hyperparameters. “Equivalent”
models have <b>exactly</b> the same architecture, and the same number of
non-embedding parameters.</figcaption>
</figure>
### Uses and Limitations
#### Intended Use
The primary intended use of Pythia is research on the behavior, functionality,
and limitations of large language models. This suite is intended to provide
a controlled setting for performing scientific experiments. To enable the
study of how language models change in the course of training, we provide
143 evenly spaced intermediate checkpoints per model. These checkpoints are
hosted on Hugging Face as branches. Note that branch `143000` corresponds
exactly to the model checkpoint on the `main` branch of each model.
You may also further fine-tune and adapt Pythia-410M-deduped for deployment,
as long as your use is in accordance with the Apache 2.0 license. Pythia
models work with the Hugging Face [Transformers
Library](https://huggingface.co/docs/transformers/index). If you decide to use
pre-trained Pythia-410M-deduped as a basis for your fine-tuned model, please
conduct your own risk and bias assessment.
#### Out-of-scope use
The Pythia Suite is **not** intended for deployment. It is not a in itself
a product and cannot be used for human-facing interactions.
Pythia models are English-language only, and are not suitable for translation
or generating text in other languages.
Pythia-410M-deduped has not been fine-tuned for downstream contexts in which
language models are commonly deployed, such as writing genre prose,
or commercial chatbots. This means Pythia-410M-deduped will **not**
respond to a given prompt the way a product like ChatGPT does. This is because,
unlike this model, ChatGPT was fine-tuned using methods such as Reinforcement
Learning from Human Feedback (RLHF) to better “understand” human instructions.
#### Limitations and biases
The core functionality of a large language model is to take a string of text
and predict the next token. The token deemed statistically most likely by the
model need not produce the most “accurate” text. Never rely on
Pythia-410M-deduped to produce factually accurate output.
This model was trained on [the Pile](https://pile.eleuther.ai/), a dataset
known to contain profanity and texts that are lewd or otherwise offensive.
See [Section 6 of the Pile paper](https://arxiv.org/abs/2101.00027) for a
discussion of documented biases with regards to gender, religion, and race.
Pythia-410M-deduped may produce socially unacceptable or undesirable text,
*even if* the prompt itself does not include anything explicitly offensive.
If you plan on using text generated through, for example, the Hosted Inference
API, we recommend having a human curate the outputs of this language model
before presenting it to other people. Please inform your audience that the
text was generated by Pythia-410M-deduped.
### Quickstart
Pythia models can be loaded and used via the following code, demonstrated here
for the third `pythia-70m-deduped` checkpoint:
```python
from transformers import GPTNeoXForCausalLM, AutoTokenizer
model = GPTNeoXForCausalLM.from_pretrained(
"EleutherAI/pythia-70m-deduped",
revision="step3000",
cache_dir="./pythia-70m-deduped/step3000",
)
tokenizer = AutoTokenizer.from_pretrained(
"EleutherAI/pythia-70m-deduped",
revision="step3000",
cache_dir="./pythia-70m-deduped/step3000",
)
inputs = tokenizer("Hello, I am", return_tensors="pt")
tokens = model.generate(**inputs)
tokenizer.decode(tokens[0])
```
Revision/branch `step143000` corresponds exactly to the model checkpoint on
the `main` branch of each model.<br>
For more information on how to use all Pythia models, see [documentation on
GitHub](https://github.com/EleutherAI/pythia).
### Training
#### Training data
Pythia-410M-deduped was trained on the Pile **after the dataset has been
globally deduplicated**.<br>
[The Pile](https://pile.eleuther.ai/) is a 825GiB general-purpose dataset in
English. It was created by EleutherAI specifically for training large language
models. It contains texts from 22 diverse sources, roughly broken down into
five categories: academic writing (e.g. arXiv), internet (e.g. CommonCrawl),
prose (e.g. Project Gutenberg), dialogue (e.g. YouTube subtitles), and
miscellaneous (e.g. GitHub, Enron Emails). See [the Pile
paper](https://arxiv.org/abs/2101.00027) for a breakdown of all data sources,
methodology, and a discussion of ethical implications. Consult [the
datasheet](https://arxiv.org/abs/2201.07311) for more detailed documentation
about the Pile and its component datasets. The Pile can be downloaded from
the [official website](https://pile.eleuther.ai/), or from a [community
mirror](https://the-eye.eu/public/AI/pile/).
#### Training procedure
All models were trained on the exact same data, in the exact same order. Each
model saw 299,892,736,000 tokens during training, and 143 checkpoints for each
model are saved every 2,097,152,000 tokens, spaced evenly throughout training.
This corresponds to training for just under 1 epoch on the Pile for
non-deduplicated models, and about 1.5 epochs on the deduplicated Pile.
All *Pythia* models trained for the equivalent of 143000 steps at a batch size
of 2,097,152 tokens. Two batch sizes were used: 2M and 4M. Models with a batch
size of 4M tokens listed were originally trained for 71500 steps instead, with
checkpoints every 500 steps. The checkpoints on Hugging Face are renamed for
consistency with all 2M batch models, so `step1000` is the first checkpoint
for `pythia-1.4b` that was saved (corresponding to step 500 in training), and
`step1000` is likewise the first `pythia-6.9b` checkpoint that was saved
(corresponding to 1000 “actual” steps).<br>
See [GitHub](https://github.com/EleutherAI/pythia) for more details on training
procedure, including [how to reproduce
it](https://github.com/EleutherAI/pythia/blob/main/README.md#reproducing-training).<br>
Pythia uses the same tokenizer as [GPT-NeoX-
20B](https://huggingface.co/EleutherAI/gpt-neox-20b).
### Evaluations
All 16 *Pythia* models were evaluated using the [LM Evaluation
Harness](https://github.com/EleutherAI/lm-evaluation-harness). You can access
the results by model and step at `results/json/*` in the [GitHub
repository](https://github.com/EleutherAI/pythia/tree/main/results/json).<br>
Expand the sections below to see plots of evaluation results for all
Pythia and Pythia-deduped models compared with OPT and BLOOM.
<details>
<summary>LAMBADA – OpenAI</summary>
<img src="/EleutherAI/pythia-12b/resolve/main/eval_plots/lambada_openai.png" style="width:auto"/>
</details>
<details>
<summary>Physical Interaction: Question Answering (PIQA)</summary>
<img src="/EleutherAI/pythia-12b/resolve/main/eval_plots/piqa.png" style="width:auto"/>
</details>
<details>
<summary>WinoGrande</summary>
<img src="/EleutherAI/pythia-12b/resolve/main/eval_plots/winogrande.png" style="width:auto"/>
</details>
<details>
<summary>AI2 Reasoning Challenge – Challenge Set</summary>
<img src="/EleutherAI/pythia-12b/resolve/main/eval_plots/arc_challenge.png" style="width:auto"/>
</details>
<details>
<summary>SciQ</summary>
<img src="/EleutherAI/pythia-12b/resolve/main/eval_plots/sciq.png" style="width:auto"/>
</details>
### Naming convention and parameter count
*Pythia* models were renamed in January 2023. It is possible that the old
naming convention still persists in some documentation by accident. The
current naming convention (70M, 160M, etc.) is based on total parameter count.
<figure style="width:32em">
| current Pythia suffix | old suffix | total params | non-embedding params |
| --------------------: | ---------: | -------------: | -------------------: |
| 70M | 19M | 70,426,624 | 18,915,328 |
| 160M | 125M | 162,322,944 | 85,056,000 |
| 410M | 350M | 405,334,016 | 302,311,424 |
| 1B | 800M | 1,011,781,632 | 805,736,448 |
| 1.4B | 1.3B | 1,414,647,808 | 1,208,602,624 |
| 2.8B | 2.7B | 2,775,208,960 | 2,517,652,480 |
| 6.9B | 6.7B | 6,857,302,016 | 6,444,163,072 |
| 12B | 13B | 11,846,072,320 | 11,327,027,200 |
</figure> |
zbnsl/bert-base-uncased-emotionsModified | zbnsl | 2022-11-09T17:23:09Z | 454 | 0 | transformers | [
"transformers",
"pytorch",
"bert",
"text-classification",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
]
| text-classification | 2022-11-09T17:16:07Z | Entry not found |
local9000/my_merged_models | local9000 | 2023-07-13T06:38:59Z | 454 | 0 | diffusers | [
"diffusers",
"region:us"
]
| null | 2023-04-10T03:33:12Z | Entry not found |
ByteWave/gpt2-turkish-uncased | ByteWave | 2023-11-10T12:37:58Z | 454 | 1 | transformers | [
"transformers",
"pytorch",
"safetensors",
"gpt2",
"text-generation",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2023-05-07T11:30:01Z | ---
license: mit
---
|
llm-book/t5-base-long-livedoor-news-corpus | llm-book | 2023-07-25T13:10:36Z | 454 | 1 | transformers | [
"transformers",
"pytorch",
"t5",
"text2text-generation",
"ja",
"dataset:llm-book/livedoor-news-corpus",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text2text-generation | 2023-06-27T13:32:54Z | ---
language:
- ja
license: apache-2.0
library_name: transformers
datasets:
- llm-book/livedoor-news-corpus
pipeline_tag: text2text-generation
metrics:
- rouge
- bleu
- bertscore
---
# llm-book/t5-base-long-livedoor-news-corpus
「[大規模言語モデル入門](https://www.amazon.co.jp/dp/4297136333)」の第7章で紹介している要約生成のモデルです。
[retrieva-jp/t5-base-long](https://huggingface.co/retrieva-jp/t5-base-long)を[llm-book/livedoor-news-corpus](https://huggingface.co/datasets/llm-book/livedoor-news-corpus)でファインチューニングして構築されています。
## 関連リンク
* [GitHubリポジトリ](https://github.com/ghmagazine/llm-book)
* [Colabノートブック](https://colab.research.google.com/github/ghmagazine/llm-book/blob/main/chapter7/7-summarization-generation.ipynb)
* [データセット](https://huggingface.co/datasets/llm-book/livedoor-news-corpus)
* [大規模言語モデル入門（Amazon.co.jp）](https://www.amazon.co.jp/dp/4297136333/)
* [大規模言語モデル入門（gihyo.jp）](https://gihyo.jp/book/2023/978-4-297-13633-8)
## 使い方
```python
from transformers import pipeline
text2text_pipeline = pipeline(
model="llm-book/t5-base-long-livedoor-news-corpus"
)
article = "ついに始まった３連休。テレビを見ながら過ごしている人も多いのではないだろうか？ 今夜オススメなのは何と言っても、NHKスペシャル「世界を変えた男 スティーブ・ジョブズ」だ。実は知らない人も多いジョブズ氏の養子に出された生い立ちや、アップル社から一時追放されるなどの経験。そして、彼が追い求めた理想の未来とはなんだったのか、ファンならずとも気になる内容になっている。 今年、亡くなったジョブズ氏の伝記は日本でもベストセラーになっている。今後もアップル製品だけでなく、世界でのジョブズ氏の影響は大きいだろうと想像される。ジョブズ氏のことをあまり知らないという人もこの機会にぜひチェックしてみよう。 世界を変えた男 スティーブ・ジョブズ（NHKスペシャル）"
print(text2text_pipeline(article)[0]["generated_text"])
# 今夜はNHKスペシャル「世界を変えた男 スティーブ・ジョブズ」をチェック!
```
## ライセンス
[Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0) |
Thamer/resnet-fine_tuned | Thamer | 2023-09-09T12:16:21Z | 454 | 0 | transformers | [
"transformers",
"pytorch",
"tensorboard",
"safetensors",
"resnet",
"image-classification",
"generated_from_trainer",
"dataset:Falah/Alzheimer_MRI",
"base_model:microsoft/resnet-34",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
]
| image-classification | 2023-08-11T23:18:43Z | ---
license: apache-2.0
base_model: microsoft/resnet-34
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: resnet-fine_tuned
results: []
datasets:
- Falah/Alzheimer_MRI
library_name: transformers
pipeline_tag: image-classification
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# resnet-fine_tuned
This model is a fine-tuned version of [microsoft/resnet-34](https://huggingface.co/microsoft/resnet-34) on the Falah/Alzheimer_MRI dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1983
- Accuracy: 0.9219
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.9041 | 1.0 | 80 | 0.9659 | 0.5352 |
| 0.8743 | 2.0 | 160 | 0.9348 | 0.5797 |
| 0.7723 | 3.0 | 240 | 0.7793 | 0.6594 |
| 0.6864 | 4.0 | 320 | 0.6799 | 0.7031 |
| 0.5347 | 5.0 | 400 | 0.5596 | 0.7703 |
| 0.4282 | 6.0 | 480 | 0.5078 | 0.7766 |
| 0.4315 | 7.0 | 560 | 0.5455 | 0.7680 |
| 0.3747 | 8.0 | 640 | 0.4203 | 0.8266 |
| 0.2977 | 9.0 | 720 | 0.3926 | 0.8469 |
| 0.2252 | 10.0 | 800 | 0.3024 | 0.8742 |
| 0.2675 | 11.0 | 880 | 0.2731 | 0.8906 |
| 0.2136 | 12.0 | 960 | 0.3045 | 0.875 |
| 0.1998 | 13.0 | 1040 | 0.2370 | 0.9 |
| 0.2406 | 14.0 | 1120 | 0.2387 | 0.9086 |
| 0.1873 | 15.0 | 1200 | 0.1983 | 0.9219 |
### Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cpu
- Datasets 2.13.1
- Tokenizers 0.13.3 |
shaowenchen/vicuna-7b-v1.5-16k-gguf | shaowenchen | 2023-09-19T07:42:42Z | 454 | 1 | null | [
"gguf",
"vicuna",
"chinese",
"text-generation",
"zh",
"en",
"license:other",
"region:us"
]
| text-generation | 2023-09-19T01:28:28Z | ---
inference: false
language:
- zh
- en
license: other
model_creator: lmsys
model_link: https://huggingface.co/lmsys/vicuna-7b-v1.5-16k
model_name: chinese-llama-2-7b-16k
model_type: vicuna
pipeline_tag: text-generation
quantized_by: shaowenchen
tasks:
- text2text-generation
tags:
- gguf
- vicuna
- chinese
---
## Provided files
| Name | Quant method | Size |
| ------------------------------ | ------------ | ------ |
| vicuna-7b-v1.5-16k.Q2_K.gguf | Q2_K | 2.6 GB |
| vicuna-7b-v1.5-16k.Q3_K.gguf | Q3_K | 3.1 GB |
| vicuna-7b-v1.5-16k.Q3_K_L.gguf | Q3_K_L | 3.3 GB |
| vicuna-7b-v1.5-16k.Q3_K_S.gguf | Q3_K_S | 2.7 GB |
| vicuna-7b-v1.5-16k.Q4_0.gguf | Q4_0 | 3.6 GB |
| vicuna-7b-v1.5-16k.Q4_1.gguf | Q4_1 | 3.9 GB |
| vicuna-7b-v1.5-16k.Q4_K.gguf | Q4_K | 3.8 GB |
| vicuna-7b-v1.5-16k.Q4_K_S.gguf | Q4_K_S | 3.6 GB |
| vicuna-7b-v1.5-16k.Q5_0.gguf | Q5_0 | 4.3 GB |
| vicuna-7b-v1.5-16k.Q5_1.gguf | Q5_1 | 4.7 GB |
| vicuna-7b-v1.5-16k.Q5_K.gguf | Q5_K | 4.5 GB |
| vicuna-7b-v1.5-16k.Q5_K_S.gguf | Q5_K_S | 4.3 GB |
| vicuna-7b-v1.5-16k.Q6_K.gguf | Q6_K | 5.1 GB |
| vicuna-7b-v1.5-16k.Q8_0.gguf | Q8_0 | 6.7 GB |
| vicuna-7b-v1.5-16k.gguf | full | 13 GB |
Usage:
```
docker run --rm -it -p 8000:8000 -v /path/to/models:/models -e MODEL=/models/gguf-model-name.gguf hubimage/llama-cpp-python:latest
```
and you can view http://localhost:8000/docs to see the swagger UI.
## Provided images
| Name | Quant method | Compressed Size |
| ------------------------------------------ | ------------ | --------------- |
| `shaowenchen/vicuna-7b-v1.5-16k-gguf:Q2_K` | Q2_K | 2.88 GB |
| `shaowenchen/vicuna-7b-v1.5-16k-gguf:Q3_K` | Q3_K | 3.3 GB |
| `shaowenchen/vicuna-7b-v1.5-16k-gguf:Q4_K` | Q4_K | 4 GB |
Usage:
```
docker run --rm -p 8000:8000 shaowenchen/vicuna-7b-v1.5-16k-gguf:Q2_K
```
and you can view http://localhost:8000/docs to see the swagger UI.
|
TheBloke/leo-hessianai-7B-GGUF | TheBloke | 2023-09-28T16:28:04Z | 454 | 3 | transformers | [
"transformers",
"gguf",
"llama",
"text-generation",
"en",
"de",
"dataset:oscar-corpus/OSCAR-2301",
"dataset:wikipedia",
"dataset:bjoernp/tagesschau-2018-2023",
"base_model:LeoLM/leo-hessianai-7b",
"license:llama2",
"text-generation-inference",
"region:us"
]
| text-generation | 2023-09-28T16:23:50Z | ---
base_model: LeoLM/leo-hessianai-7b
datasets:
- oscar-corpus/OSCAR-2301
- wikipedia
- bjoernp/tagesschau-2018-2023
inference: false
language:
- en
- de
library_name: transformers
license: llama2
model_creator: LAION LeoLM
model_name: Leo Hessianai 7B
model_type: llama
pipeline_tag: text-generation
prompt_template: '{prompt}
'
quantized_by: TheBloke
---
<!-- header start -->
<!-- 200823 -->
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div>
<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
<!-- header end -->
# Leo Hessianai 7B - GGUF
- Model creator: [LAION LeoLM](https://huggingface.co/LeoLM)
- Original model: [Leo Hessianai 7B](https://huggingface.co/LeoLM/leo-hessianai-7b)
<!-- description start -->
## Description
This repo contains GGUF format model files for [LAION LeoLM's Leo Hessianai 7B](https://huggingface.co/LeoLM/leo-hessianai-7b).
<!-- description end -->
<!-- README_GGUF.md-about-gguf start -->
### About GGUF
GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp.
Here is an incomplate list of clients and libraries that are known to support GGUF:
* [llama.cpp](https://github.com/ggerganov/llama.cpp). The source project for GGUF. Offers a CLI and a server option.
* [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
* [KoboldCpp](https://github.com/LostRuins/koboldcpp), a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
* [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration.
* [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with many interesting and unique features, including a full model library for easy model selection.
* [Faraday.dev](https://faraday.dev/), an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
* [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
* [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
* [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
<!-- README_GGUF.md-about-gguf end -->
<!-- repositories-available start -->
## Repositories available
* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/leo-hessianai-7B-AWQ)
* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/leo-hessianai-7B-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF)
* [LAION LeoLM's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/LeoLM/leo-hessianai-7b)
<!-- repositories-available end -->
<!-- prompt-template start -->
## Prompt template: None
```
{prompt}
```
<!-- prompt-template end -->
<!-- compatibility_gguf start -->
## Compatibility
These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221)
They are also compatible with many third party UIs and libraries - please see the list at the top of this README.
## Explanation of quantisation methods
<details>
<summary>Click to see details</summary>
The new methods available are:
* GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
* GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
* GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
* GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
* GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw
Refer to the Provided Files table below to see what files use which methods, and how.
</details>
<!-- compatibility_gguf end -->
<!-- README_GGUF.md-provided-files start -->
## Provided files
| Name | Quant method | Bits | Size | Max RAM required | Use case |
| ---- | ---- | ---- | ---- | ---- | ----- |
| [leo-hessianai-7b.Q2_K.gguf](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF/blob/main/leo-hessianai-7b.Q2_K.gguf) | Q2_K | 2 | 2.83 GB| 5.33 GB | smallest, significant quality loss - not recommended for most purposes |
| [leo-hessianai-7b.Q3_K_S.gguf](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF/blob/main/leo-hessianai-7b.Q3_K_S.gguf) | Q3_K_S | 3 | 2.95 GB| 5.45 GB | very small, high quality loss |
| [leo-hessianai-7b.Q3_K_M.gguf](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF/blob/main/leo-hessianai-7b.Q3_K_M.gguf) | Q3_K_M | 3 | 3.30 GB| 5.80 GB | very small, high quality loss |
| [leo-hessianai-7b.Q3_K_L.gguf](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF/blob/main/leo-hessianai-7b.Q3_K_L.gguf) | Q3_K_L | 3 | 3.60 GB| 6.10 GB | small, substantial quality loss |
| [leo-hessianai-7b.Q4_0.gguf](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF/blob/main/leo-hessianai-7b.Q4_0.gguf) | Q4_0 | 4 | 3.83 GB| 6.33 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [leo-hessianai-7b.Q4_K_S.gguf](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF/blob/main/leo-hessianai-7b.Q4_K_S.gguf) | Q4_K_S | 4 | 3.86 GB| 6.36 GB | small, greater quality loss |
| [leo-hessianai-7b.Q4_K_M.gguf](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF/blob/main/leo-hessianai-7b.Q4_K_M.gguf) | Q4_K_M | 4 | 4.08 GB| 6.58 GB | medium, balanced quality - recommended |
| [leo-hessianai-7b.Q5_0.gguf](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF/blob/main/leo-hessianai-7b.Q5_0.gguf) | Q5_0 | 5 | 4.65 GB| 7.15 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [leo-hessianai-7b.Q5_K_S.gguf](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF/blob/main/leo-hessianai-7b.Q5_K_S.gguf) | Q5_K_S | 5 | 4.65 GB| 7.15 GB | large, low quality loss - recommended |
| [leo-hessianai-7b.Q5_K_M.gguf](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF/blob/main/leo-hessianai-7b.Q5_K_M.gguf) | Q5_K_M | 5 | 4.78 GB| 7.28 GB | large, very low quality loss - recommended |
| [leo-hessianai-7b.Q6_K.gguf](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF/blob/main/leo-hessianai-7b.Q6_K.gguf) | Q6_K | 6 | 5.53 GB| 8.03 GB | very large, extremely low quality loss |
| [leo-hessianai-7b.Q8_0.gguf](https://huggingface.co/TheBloke/leo-hessianai-7B-GGUF/blob/main/leo-hessianai-7b.Q8_0.gguf) | Q8_0 | 8 | 7.16 GB| 9.66 GB | very large, extremely low quality loss - not recommended |
**Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
<!-- README_GGUF.md-provided-files end -->
<!-- README_GGUF.md-how-to-download start -->
## How to download GGUF files
**Note for manual downloaders:** You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.
The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
- LM Studio
- LoLLMS Web UI
- Faraday.dev
### In `text-generation-webui`
Under Download Model, you can enter the model repo: TheBloke/leo-hessianai-7B-GGUF and below it, a specific filename to download, such as: leo-hessianai-7b.Q4_K_M.gguf.
Then click Download.
### On the command line, including multiple files at once
I recommend using the `huggingface-hub` Python library:
```shell
pip3 install huggingface-hub
```
Then you can download any individual model file to the current directory, at high speed, with a command like this:
```shell
huggingface-cli download TheBloke/leo-hessianai-7B-GGUF leo-hessianai-7b.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
<details>
<summary>More advanced huggingface-cli download usage</summary>
You can also download multiple files at once with a pattern:
```shell
huggingface-cli download TheBloke/leo-hessianai-7B-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
For more documentation on downloading with `huggingface-cli`, please see: [HF -> Hub Python Library -> Download files -> Download from the CLI](https://huggingface.co/docs/huggingface_hub/guides/download#download-from-the-cli).
To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
```shell
pip3 install hf_transfer
```
And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
```shell
HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/leo-hessianai-7B-GGUF leo-hessianai-7b.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
Windows Command Line users: You can set the environment variable by running `set HF_HUB_ENABLE_HF_TRANSFER=1` before the download command.
</details>
<!-- README_GGUF.md-how-to-download end -->
<!-- README_GGUF.md-how-to-run start -->
## Example `llama.cpp` command
Make sure you are using `llama.cpp` from commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
```shell
./main -ngl 32 -m leo-hessianai-7b.Q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "{prompt}"
```
Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
Change `-c 4096` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.
If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
For other parameters and how to use them, please refer to [the llama.cpp documentation](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)
## How to run in `text-generation-webui`
Further instructions here: [text-generation-webui/docs/llama.cpp.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/llama.cpp.md).
## How to run from Python code
You can use GGUF models from Python using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) or [ctransformers](https://github.com/marella/ctransformers) libraries.
### How to load this model in Python code, using ctransformers
#### First install the package
Run one of the following commands, according to your system:
```shell
# Base ctransformers with no GPU acceleration
pip install ctransformers
# Or with CUDA GPU acceleration
pip install ctransformers[cuda]
# Or with AMD ROCm GPU acceleration (Linux only)
CT_HIPBLAS=1 pip install ctransformers --no-binary ctransformers
# Or with Metal GPU acceleration for macOS systems only
CT_METAL=1 pip install ctransformers --no-binary ctransformers
```
#### Simple ctransformers example code
```python
from ctransformers import AutoModelForCausalLM
# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
llm = AutoModelForCausalLM.from_pretrained("TheBloke/leo-hessianai-7B-GGUF", model_file="leo-hessianai-7b.Q4_K_M.gguf", model_type="llama", gpu_layers=50)
print(llm("AI is going to"))
```
## How to use with LangChain
Here are guides on using llama-cpp-python and ctransformers with LangChain:
* [LangChain + llama-cpp-python](https://python.langchain.com/docs/integrations/llms/llamacpp)
* [LangChain + ctransformers](https://python.langchain.com/docs/integrations/providers/ctransformers)
<!-- README_GGUF.md-how-to-run end -->
<!-- footer start -->
<!-- 200823 -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/theblokeai)
## Thanks, and how to contribute
Thanks to the [chirper.ai](https://chirper.ai) team!
Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Special thanks to**: Aemon Algiz.
**Patreon special mentions**: Pierre Kircher, Stanislav Ovsiannikov, Michael Levine, Eugene Pentland, Andrey, 준교 김, Randy H, Fred von Graf, Artur Olbinski, Caitlyn Gatomon, terasurfer, Jeff Scroggin, James Bentley, Vadim, Gabriel Puliatti, Harry Royden McLaughlin, Sean Connelly, Dan Guido, Edmond Seymore, Alicia Loh, subjectnull, AzureBlack, Manuel Alberto Morcote, Thomas Belote, Lone Striker, Chris Smitley, Vitor Caleffi, Johann-Peter Hartmann, Clay Pascal, biorpg, Brandon Frisco, sidney chen, transmissions 11, Pedro Madruga, jinyuan sun, Ajan Kanaga, Emad Mostaque, Trenton Dambrowitz, Jonathan Leane, Iucharbius, usrbinkat, vamX, George Stoitzev, Luke Pendergrass, theTransient, Olakabola, Swaroop Kallakuri, Cap'n Zoog, Brandon Phillips, Michael Dempsey, Nikolai Manek, danny, Matthew Berman, Gabriel Tamborski, alfie_i, Raymond Fosdick, Tom X Nguyen, Raven Klaugh, LangChain4j, Magnesian, Illia Dulskyi, David Ziegler, Mano Prime, Luis Javier Navarrete Lozano, Erik Bjäreholt, 阿明, Nathan Dryer, Alex, Rainer Wilmers, zynix, TL, Joseph William Delisle, John Villwock, Nathan LeClaire, Willem Michiel, Joguhyik, GodLy, OG, Alps Aficionado, Jeffrey Morgan, ReadyPlayerEmma, Tiffany J. Kim, Sebastain Graf, Spencer Kim, Michael Davis, webtim, Talal Aujan, knownsqashed, John Detwiler, Imad Khwaja, Deo Leter, Jerry Meng, Elijah Stavena, Rooh Singh, Pieter, SuperWojo, Alexandros Triantafyllidis, Stephen Murray, Ai Maven, ya boyyy, Enrico Ros, Ken Nordquist, Deep Realms, Nicholas, Spiking Neurons AB, Elle, Will Dee, Jack West, RoA, Luke @flexchar, Viktor Bowallius, Derek Yates, Subspace Studios, jjj, Toran Billups, Asp the Wyvern, Fen Risland, Ilya, NimbleBox.ai, Chadd, Nitin Borwankar, Emre, Mandus, Leonard Tan, Kalila, K, Trailburnt, S_X, Cory Kujawski
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end -->
<!-- original-model-card start -->
# Original model card: LAION LeoLM's Leo Hessianai 7B
# LAION LeoLM: **L**inguistically **E**nhanced **O**pen **L**anguage **M**odel
Meet LeoLM, the first open and commercially available German Foundation Language Model built on Llama-2.
Our models extend Llama-2's capabilities into German through continued pretraining on a large corpus of German-language and mostly locality specific text.
Thanks to a compute grant at HessianAI's new supercomputer **42**, we release two foundation models trained with 8k context length,
[`LeoLM/leo-hessianai-7b`](https://huggingface.co/LeoLM/leo-hessianai-7b) and [`LeoLM/leo-hessianai-13b`](https://huggingface.co/LeoLM/leo-hessianai-13b) under the [Llama-2 community license](https://huggingface.co/meta-llama/Llama-2-70b/raw/main/LICENSE.txt) (70b also coming soon! 👀).
With this release, we hope to bring a new wave of opportunities to German open-source and commercial LLM research and accelerate adoption.
Read our [blog post]() or our paper (preprint coming soon) for more details!
*A project by Björn Plüster and Christoph Schuhmann in collaboration with LAION and HessianAI.*
## Model Details
- **Finetuned from:** [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf)
- **Model type:** Causal decoder-only transformer language model
- **Language:** English and German
- **License:** [LLAMA 2 COMMUNITY LICENSE AGREEMENT](https://huggingface.co/meta-llama/Llama-2-70b/raw/main/LICENSE.txt)
- **Contact:** [LAION Discord](https://discord.com/invite/eq3cAMZtCC) or [Björn Plüster](mailto:[email protected])
## Use in 🤗Transformers
First install direct dependencies:
```
pip install transformers torch sentencepiece
```
If you want faster inference using flash-attention2, you need to install these dependencies:
```bash
pip install packaging ninja
pip install flash-attn==v2.1.1 --no-build-isolation
pip install git+https://github.com/HazyResearch/[email protected]#subdirectory=csrc/rotary
```
Then load the model in transformers:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model = AutoModelForCausalLM.from_pretrained(
model="LeoLM/leo-hessianai-7b",
device_map="auto",
torch_dtype=torch.float16,
trust_remote_code=True # True for flash-attn2 else False
)
```
## Training parameters
## Benchmarks
<!-- original-model-card end -->
|
maddes8cht/mosaicml-mpt-30b-chat-gguf | maddes8cht | 2023-11-01T15:36:41Z | 454 | 1 | null | [
"gguf",
"Composer",
"MosaicML",
"llm-foundry",
"arxiv:2205.14135",
"arxiv:2108.12409",
"arxiv:2010.04245",
"license:cc-by-nc-sa-4.0",
"region:us"
]
| null | 2023-10-26T18:49:30Z | ---
license: cc-by-nc-sa-4.0
datasets:
- camel-ai/code
- ehartford/wizard_vicuna_70k_unfiltered
- anon8231489123/ShareGPT_Vicuna_unfiltered
- teknium1/GPTeacher/roleplay-instruct-v2-final
- teknium1/GPTeacher/codegen-isntruct
- timdettmers/openassistant-guanaco
- camel-ai/math
- project-baize/baize-chatbot/medical_chat_data
- project-baize/baize-chatbot/quora_chat_data
- project-baize/baize-chatbot/stackoverflow_chat_data
- camel-ai/biology
- camel-ai/chemistry
- camel-ai/ai_society
- jondurbin/airoboros-gpt4-1.2
- LongConversations
- camel-ai/physics
tags:
- Composer
- MosaicML
- llm-foundry
inference: false
---
[]()
I'm constantly enhancing these model descriptions to provide you with the most relevant and comprehensive information
# mpt-30b-chat - GGUF
- Model creator: [mosaicml](https://huggingface.co/mosaicml)
- Original model: [mpt-30b-chat](https://huggingface.co/mosaicml/mpt-30b-chat)
MPT-7b and MPT-30B are part of the family of Mosaic Pretrained Transformer (MPT) models, which use a modified transformer architecture optimized for efficient training and inference.
---
# Brief
MPT-30B Chat is based on MPT-30B and finetuned for Chat-like experience.
---
# About GGUF format
`gguf` is the current file format used by the [`ggml`](https://github.com/ggerganov/ggml) library.
A growing list of Software is using it and can therefore use this model.
The core project making use of the ggml library is the [llama.cpp](https://github.com/ggerganov/llama.cpp) project by Georgi Gerganov
# Quantization variants
There is a bunch of quantized files available to cater to your specific needs. Here's how to choose the best option for you:
# Legacy quants
Q4_0, Q4_1, Q5_0, Q5_1 and Q8 are `legacy` quantization types.
Nevertheless, they are fully supported, as there are several circumstances that cause certain model not to be compatible with the modern K-quants.
## Note:
Now there's a new option to use K-quants even for previously 'incompatible' models, although this involves some fallback solution that makes them not *real* K-quants. More details can be found in affected model descriptions.
(This mainly refers to Falcon 7b and Starcoder models)
# K-quants
K-quants are designed with the idea that different levels of quantization in specific parts of the model can optimize performance, file size, and memory load.
So, if possible, use K-quants.
With a Q6_K, you'll likely find it challenging to discern a quality difference from the original model - ask your model two times the same question and you may encounter bigger quality differences.
---
# Original Model Card:
# MPT-30B-Chat
MPT-30B-Chat is a chatbot-like model for dialogue generation.
It was built by finetuning [MPT-30B](https://huggingface.co/mosaicml/mpt-30b) on the [ShareGPT-Vicuna](https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered), [Camel-AI](https://huggingface.co/camel-ai),
[GPTeacher](https://github.com/teknium1/GPTeacher), [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco), [Baize](https://github.com/project-baize/baize-chatbot) and some generated datasets.
* License: _CC-By-NC-SA-4.0_ (non-commercial use only)
This model was trained by [MosaicML](https://www.mosaicml.com) and follows a modified decoder-only transformer architecture.
## Model Date
June 22, 2023
## Model License
_CC-By-NC-SA-4.0_ (non-commercial use only)
## Documentation
* [Blog post: Raising the bar for open-source foundation models](https://www.mosaicml.com/blog/mpt-30b)
* [Codebase (mosaicml/llm-foundry repo)](https://github.com/mosaicml/llm-foundry/)
* Questions: Feel free to contact us via the [MosaicML Community Slack](https://mosaicml.me/slack)!
### Example Dialogue
**ksreenivasan**:
> I want you to come up with a tweet based on this summary of the article:
>
> We are excited to extend the MosaicML Foundation Series with MPT-30B, a new commercially licensed open-source model that is significantly more powerful, and outperforms the
> original GPT-3. In addition, we are releasing two fine-tuned models, MPT-30B-Instruct and MPT-30B-Chat, that are built on top of MPT-30B and excel at short-form instruction
> following and multi-turn conversations, respectively.
> All MPT-30B models come with special features that differentiate them from other LLMs, including an 8k token context window (which can be further extended via finetuning;
> see MPT-7B-StoryWriter), support for context-length extrapolation via ALiBi, and efficient inference + training performance via FlashAttention. It also has strong coding
> abilities thanks to its pretraining data mixture.
**MPT-30B-Chat**:
> "Revolutionize your AI capabilities with MosaicML's latest addition, MPT-30B! Outperforming GPT-3, this open-source model is significantly more powerful and comes with
> finely-tuned models for exceptional instruction following and multi-turn conversations. Unleash the potential of an 8k token context window, ALiBi support, and FlashAttention.
> Experience the future of AI with MosaicML! #AI #ML #innovation"
**ksreenivasan**:
> Is there anything you want to say to MosaicML about this exciting news?
**MPT-30B-Chat**:
> As your AI marketing assistant, I would say that this is truly exciting news for the AI community! MosaicML's latest addition, MPT-30B, is poised to revolutionize the industry with its powerful open-source capabilities and finely-tuned models for instruction following and multi-turn conversations.
> The special features, including the 8k token context window, ALiBi support, and FlashAttention, are sure to set MosaicML's MPT-30B apart from other LLMs.
> Congratulations to the MosaicML team on this incredible achievement!
## How to Use
This model is best used with the MosaicML [llm-foundry repository](https://github.com/mosaicml/llm-foundry) for training and finetuning.
```python
import transformers
model = transformers.AutoModelForCausalLM.from_pretrained(
'mosaicml/mpt-30b-chat',
trust_remote_code=True
)
```
Note: This model requires that `trust_remote_code=True` be passed to the `from_pretrained` method.
This is because we use a custom `MPT` model architecture that is not yet part of the Hugging Face `transformers` package.
`MPT` includes options for many training efficiency features such as [FlashAttention](https://arxiv.org/pdf/2205.14135.pdf), [ALiBi](https://arxiv.org/abs/2108.12409), [QK LayerNorm](https://arxiv.org/abs/2010.04245), and more.
To use the optimized [triton implementation](https://github.com/openai/triton) of FlashAttention, you can load the model on GPU (`cuda:0`) with `attn_impl='triton'` and with `bfloat16` precision:
```python
import torch
import transformers
name = 'mosaicml/mpt-30b-chat'
config = transformers.AutoConfig.from_pretrained(name, trust_remote_code=True)
config.attn_config['attn_impl'] = 'triton' # change this to use triton-based FlashAttention
config.init_device = 'cuda:0' # For fast initialization directly on GPU!
model = transformers.AutoModelForCausalLM.from_pretrained(
name,
config=config,
torch_dtype=torch.bfloat16, # Load model weights in bfloat16
trust_remote_code=True
)
```
The model was trained initially with a sequence length of 2048 with an additional pretraining stage for sequence length adapation up to 8192. However, ALiBi enables users to increase the maximum sequence length even further during finetuning and/or inference. For example:
```python
import transformers
name = 'mosaicml/mpt-30b-chat'
config = transformers.AutoConfig.from_pretrained(name, trust_remote_code=True)
config.max_seq_len = 16384 # (input + output) tokens can now be up to 16384
model = transformers.AutoModelForCausalLM.from_pretrained(
name,
config=config,
trust_remote_code=True
)
```
This model was trained with the MPT-30B tokenizer which is based on the [EleutherAI/gpt-neox-20b](https://huggingface.co/EleutherAI/gpt-neox-20b) tokenizer and includes additional padding and eos tokens.
```python
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained('mosaicml/mpt-30b')
```
The model can then be used, for example, within a text-generation pipeline.
Note: when running Torch modules in lower precision, it is best practice to use the [torch.autocast context manager](https://pytorch.org/docs/stable/amp.html).
```python
from transformers import pipeline
with torch.autocast('cuda', dtype=torch.bfloat16):
inputs = tokenizer('Here is a recipe for vegan banana bread:\n', return_tensors="pt").to('cuda')
outputs = model.generate(**inputs, max_new_tokens=100)
print(tokenizer.batch_decode(outputs, skip_special_tokens=True))
# or using the HF pipeline
pipe = pipeline('text-generation', model=model, tokenizer=tokenizer, device='cuda:0')
with torch.autocast('cuda', dtype=torch.bfloat16):
print(
pipe('Here is a recipe for vegan banana bread:\n',
max_new_tokens=100,
do_sample=True,
use_cache=True))
```
## Model Description
The architecture is a modification of a standard decoder-only transformer.
The model has been modified from a standard transformer in the following ways:
* It uses [FlashAttention](https://arxiv.org/pdf/2205.14135.pdf)
* It uses [ALiBi (Attention with Linear Biases)](https://arxiv.org/abs/2108.12409) and does not use positional embeddings
* It does not use biases
| Hyperparameter | Value |
|----------------|-------|
|n_parameters | 29.95B |
|n_layers | 48 |
| n_heads | 64 |
| d_model | 7168 |
| vocab size | 50432 |
| sequence length | 8192 |
## Data Mix
The model was trained on the following data mix:
| Data Source | Number of Tokens in Source | Proportion |
|-------------|----------------------------|------------|
| Airoboros/GPT4-1.2 | 26.4M | 1.71% |
| Baize | 55.0M | 3.57% |
| Camel | 301M | 19.54% |
| GPTeacher | 7.56M | 0.49% |
| Guanaco | 15.6M | 1.02% |
| LongCoversations | 18.4M | 1.19% |
| ShareGPT | 821M | 53.24% |
| WizardLM | 297M | 19.23% |
"LongConversations" is a GPT3.5/4-generated dataset, details of which will be released at a later date.
### Training Configuration
This model was trained on 64 H100s for about 7.6 hours using the [MosaicML Platform](https://www.mosaicml.com/platform).
The model was trained with sharded data parallelism using [FSDP](https://pytorch.org/docs/stable/fsdp.html) and used the AdamW optimizer.
## Limitations and Biases
_The following language is modified from [EleutherAI's GPT-NeoX-20B](https://huggingface.co/EleutherAI/gpt-neox-20b)_
MPT-30B-Chat can produce factually incorrect output, and should not be relied on to produce factually accurate information.
MPT-30B-Chat was trained on various public datasets.
While great efforts have been taken to clean the pretraining data, it is possible that this model could generate lewd, biased or otherwise offensive outputs.
## Acknowledgements
This model was finetuned by Sam Havens and the MosaicML NLP team
## Disclaimer
The license on this model does not constitute legal advice. We are not responsible for the actions of third parties who use this model. Please consult an attorney before using this model for commercial purposes.
## MosaicML Platform
If you're interested in [training](https://www.mosaicml.com/training) and [deploying](https://www.mosaicml.com/inference) your own MPT or LLMs on the MosaicML Platform, [sign up here](https://forms.mosaicml.com/demo?utm_source=huggingface&utm_medium=referral&utm_campaign=mpt-7b).
## Citation
Please cite this model using the following format:
```
@online{MosaicML2023Introducing,
author = {MosaicML NLP Team},
title = {Introducing MPT-30B: Raising the bar
for open-source foundation models},
year = {2023},
url = {www.mosaicml.com/blog/mpt-30b},
note = {Accessed: 2023-06-22},
urldate = {2023-06-22}
}
```
***End of original Model File***
---
## Please consider to support my work
**Coming Soon:** I'm in the process of launching a sponsorship/crowdfunding campaign for my work. I'm evaluating Kickstarter, Patreon, or the new GitHub Sponsors platform, and I am hoping for some support and contribution to the continued availability of these kind of models. Your support will enable me to provide even more valuable resources and maintain the models you rely on. Your patience and ongoing support are greatly appreciated as I work to make this page an even more valuable resource for the community.
<center>
[](https://maddes8cht.github.io)
[](https://stackexchange.com/users/26485911)
[](https://github.com/maddes8cht)
[](https://huggingface.co/maddes8cht)
[](https://twitter.com/maddes1966)
</center> |
bitfount/RETFound_MAE_OCT_CNV_DME_DRU | bitfount | 2024-06-10T15:05:29Z | 454 | 0 | timm | [
"timm",
"pytorch",
"image-classification",
"license:apache-2.0",
"region:us"
]
| image-classification | 2023-11-08T09:10:00Z | ---
tags:
- image-classification
- timm
library_name: timm
license: apache-2.0
---
# Model card for RETFound_MAE_oct_dru_cnv_dme_nor
A model fine-tuned on the [Kermany dataset of retinal OCT images](http://www.cell.com/cell/fulltext/S0092-8674(18)30154-5), initialised from the base [open-eye/RETFound_MAE](https://huggingface.co/open-eye/RETFound_MAE) model for OCT images.
This model achieves 85% accuracy in the task of classifying Retinal OCT images as either showing signs of Drusen, Choroidal Neovascularization (CNV), Diabetic Macular Oedema (DME), or alternatively showing none of these and being otherwise "Normal".
|
dominguesm/Canarim-7B-Instruct | dominguesm | 2024-04-04T10:01:40Z | 454 | 8 | transformers | [
"transformers",
"pytorch",
"llama",
"text-generation",
"LLM",
"Portuguese",
"Llama 2",
"pt",
"dataset:dominguesm/CC-MAIN-2023-23",
"license:llama2",
"model-index",
"autotrain_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2023-11-16T18:10:52Z | ---
language:
- pt
license: llama2
library_name: transformers
tags:
- text-generation
- pytorch
- LLM
- Portuguese
- Llama 2
datasets:
- dominguesm/CC-MAIN-2023-23
inference: false
pipeline_tag: text-generation
model-index:
- name: Canarim-7B-Instruct
results:
- task:
type: text-generation
name: Text Generation
dataset:
name: ENEM Challenge (No Images)
type: eduagarcia/enem_challenge
split: train
args:
num_few_shot: 3
metrics:
- type: acc
value: 27.5
name: accuracy
source:
url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=dominguesm/Canarim-7B-Instruct
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: BLUEX (No Images)
type: eduagarcia-temp/BLUEX_without_images
split: train
args:
num_few_shot: 3
metrics:
- type: acc
value: 26.15
name: accuracy
source:
url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=dominguesm/Canarim-7B-Instruct
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: OAB Exams
type: eduagarcia/oab_exams
split: train
args:
num_few_shot: 3
metrics:
- type: acc
value: 29.93
name: accuracy
source:
url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=dominguesm/Canarim-7B-Instruct
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: Assin2 RTE
type: assin2
split: test
args:
num_few_shot: 15
metrics:
- type: f1_macro
value: 75.74
name: f1-macro
source:
url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=dominguesm/Canarim-7B-Instruct
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: Assin2 STS
type: eduagarcia/portuguese_benchmark
split: test
args:
num_few_shot: 15
metrics:
- type: pearson
value: 12.08
name: pearson
source:
url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=dominguesm/Canarim-7B-Instruct
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: FaQuAD NLI
type: ruanchaves/faquad-nli
split: test
args:
num_few_shot: 15
metrics:
- type: f1_macro
value: 43.92
name: f1-macro
source:
url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=dominguesm/Canarim-7B-Instruct
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: HateBR Binary
type: ruanchaves/hatebr
split: test
args:
num_few_shot: 25
metrics:
- type: f1_macro
value: 79.57
name: f1-macro
source:
url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=dominguesm/Canarim-7B-Instruct
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: PT Hate Speech Binary
type: hate_speech_portuguese
split: test
args:
num_few_shot: 25
metrics:
- type: f1_macro
value: 64.01
name: f1-macro
source:
url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=dominguesm/Canarim-7B-Instruct
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: tweetSentBR
type: eduagarcia-temp/tweetsentbr
split: test
args:
num_few_shot: 25
metrics:
- type: f1_macro
value: 66.0
name: f1-macro
source:
url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=dominguesm/Canarim-7B-Instruct
name: Open Portuguese LLM Leaderboard
---
<p align="center">
<img width="250" alt="Camarim Logo" src="https://raw.githubusercontent.com/DominguesM/Canarim-Instruct-PTBR/main/assets/canarim.png">
</p>
<hr>
# Canarim-7B-Instruct
Canarim-7B-Instruct is a 7 billion parameter language model, initialized from the [Canarim-7B](https://huggingface.co/dominguesm/canarim-7b) model and trained on a variety of publicly available instruction datasets.
## Usage
```python
from transformers import AutoTokenizer, pipeline
import torch
model_id = "dominguesm/Canarim-7B-Instruct"
tokenizer = AutoTokenizer.from_pretrained(model_id)
pipe = pipeline(
"text-generation",
model=model_id,
torch_dtype=torch.float16,
device_map="auto",
)
def make_prompt(instruction, input=None):
if input:
return f"""Abaixo está uma instrução que descreve uma tarefa, emparelhada com uma entrada que fornece mais contexto. Escreva uma resposta que conclua adequadamente a solicitação.
### Instruções:
{instruction}
### Entrada:
{input}
### Resposta:"""
else:
return f"""Abaixo está uma instrução que descreve uma tarefa. Escreva uma resposta que conclua adequadamente a solicitação.
### Instruções:
{instruction}
### Resposta:"""
instruction = "Faça um resumo conciso do seguinte texto."
context = """
No início da obra, o narrador Bentinho explica como escolheu o título para o seu livro. Assim, ele conta que conheceu um poeta em uma viagem de trem. O jovem recitou alguns versos, mas Bentinho, já velho, fechou os olhos algumas vezes. O poeta, então, achou que ele estava cochilando e, ofendido, lhe chamou de Dom Casmurro.
Em seguida, o narrador inicia a narrativa de suas memórias, que começa quando ele tem 15 anos. Nessa época, Capitu, uma jovem de 14 anos, é sua vizinha. Os adolescentes têm uma forte amizade, o que incomoda o agregado da família, o amante de superlativos José Dias.
Desconfiado da relação entre os jovens, ele sugere à D. Glória, mãe de Bentinho, que já é hora de enviar o rapaz para o seminário. Isso porque, quando o menino nasceu, sua mãe fez uma promessa de que ele seria padre. Dessa forma, se apresenta um obstáculo ao amor do jovem casal, o que, enganosamente, faz parecer que o livro é um romance romântico.
"""
prompt = make_prompt(question)
sequences = pipe(
prompt,
do_sample=True,
num_return_sequences=1,
eos_token_id=tokenizer.eos_token_id,
max_length=2048,
temperature=0.9,
top_p=0.6,
repetition_penalty=1.15
)
```
### Model Result
```
O trabalho retrata as memórias de Bentinho, um jovem de 15 anos. Ele conta
sobre sua infância na casa de seus pais, onde vive com sua mãe, a dona
Glória, e sua governanta, a governanta Josefa. No entanto, no outono de
1860, sua vida mudará drasticamente. Durante essa temporada, ele conhece
Capitu, uma moça de 14 anos, que se tornará sua companheira por muitos anos.
```
**NOTE**: README under construction
# [Open Portuguese LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/eduagarcia-temp/llm_pt_leaderboard_raw_results/tree/main/dominguesm/Canarim-7B-Instruct)
| Metric | Value |
|--------------------------|---------|
|Average |**47.21**|
|ENEM Challenge (No Images)| 27.50|
|BLUEX (No Images) | 26.15|
|OAB Exams | 29.93|
|Assin2 RTE | 75.74|
|Assin2 STS | 12.08|
|FaQuAD NLI | 43.92|
|HateBR Binary | 79.57|
|PT Hate Speech Binary | 64.01|
|tweetSentBR | 66|
|
NousResearch/Genstruct-7B | NousResearch | 2024-03-07T18:35:02Z | 454 | 359 | transformers | [
"transformers",
"safetensors",
"mistral",
"text-generation",
"Mistral",
"instruct",
"finetune",
"synthetic",
"conversational",
"en",
"arxiv:2310.04484",
"base_model:mistralai/Mistral-7B-v0.1",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-01-05T21:56:01Z | ---
base_model: mistralai/Mistral-7B-v0.1
tags:
- Mistral
- instruct
- finetune
- synthetic
license: apache-2.0
language:
- en
library_name: transformers
---
# Genstruct 7B
Genstruct 7B is an instruction-generation model, designed to create valid instructions given a raw text corpus. This enables the creation of new, partially synthetic instruction finetuning datasets from any raw-text corpus.
This work was inspired by [Ada-Instruct](https://arxiv.org/abs/2310.04484):
Previous methods largely rely on in-context approaches to generate instructions, while Ada-Instruct trained a custom instruction-generation model.
Inspired by this, we took this approach further by grounding the generations in user-provided context passages.
Further, the model is trained to generate questions involving complex scenarios that require detailed reasoning, allowing for models trained on the generated data to reason step-by-step.
| | ChatGPT | Few-shot prompting | RAG | Ada-Instruct | **Genstruct** |
|:-------------------:|:--------:|:------------------:|:---:|:------------:|:---------:|
| Open models | ❌ | ☑️ | ☑️ | ✅ | ✅ |
| Grounded generation | ❌ | ❌ | ✅ | ❌ | ✅ |
| Complex questions | ❌ | ❌ | ❌ | ☑️ | ✅ |
| Complex responses | ✅ | ☑️ | ❌ | ☑️ | ✅ |
An example notebook is provided [here](https://huggingface.co/NousResearch/Genstruct-7B/blob/main/notebook.ipynb), which details how to load and sample from the model.
Alternatively, here's a minimal example:
```py
from transformers import AutoModelForCausalLM, AutoTokenizer
MODEL_NAME = 'NousResearch/Genstruct-7B'
model = AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map='cuda', load_in_8bit=True)
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
msg =[{
'title': 'p-value',
'content': "The p-value is used in the context of null hypothesis testing in order to quantify the statistical significance of a result, the result being the observed value of the chosen statistic T {\displaystyle T}.[note 2] The lower the p-value is, the lower the probability of getting that result if the null hypothesis were true. A result is said to be statistically significant if it allows us to reject the null hypothesis. All other things being equal, smaller p-values are taken as stronger evidence against the null hypothesis."
}]
inputs = tokenizer.apply_chat_template(msg, return_tensors='pt').cuda()
print(tokenizer.decode(model.generate(inputs, max_new_tokens=512)[0]).split(tokenizer.eos_token)[0])
```
# Example outputs
The input is the text up to and including "`[[[User]]] `", and the completion is everything that follows.
### Law of large numbers
```
[[[Title]]] Law of large numbers
[[[Content]]] In probability theory, the law of large numbers (LLN) is a mathematical theorem that states that the average of the results obtained from a large number of independent and identical random samples converges to the true value, if it exists.[1] More formally, the LLN states that given a sample of independent and identically distributed values, the sample mean converges to the true mean.
The following is an interaction between a user and an AI assistant that is related to the above text.
[[[User]]] Jacqueline is a statistician. She is currently analyzing a large amount of data. She is applying the law of large numbers in order to get some conclusions. In order to better understand the law of large numbers, she is considering two samples A and B. Sample A is not independent but sample B is independent and identically distributed.
Which sample's mean does not converge to the true mean, if any, sample A or sample B?
[[[Assistant]]] The law of large numbers (LLN) is a significant concept in probability theory. It suggests that the average of the results obtained from a large number of independent and identical random samples will converge to the true value, if it exists. This means that the sample mean will converge to the true mean.
Now, let's consider Jacqueline's samples. She has sample A which is not independent and sample B which is independent and identically distributed. According to the conditions stated in the LLN, only samples that are independent and identically distributed can make their means converge to the true mean.
Since sample A is not independent, its values are likely related or influenced by previous observations. This lack of independence can skew the mean and prevent it from converging to the true mean. On the other hand, sample B satisfies the condition of independence and identical distribution. Therefore, its mean has a higher chance of converging to the true mean.
Hence, if we are looking for a sample whose mean does not converge to the true mean, based on the given information, it would be sample A.
```
### Zinc deficiency
```
[[[Title]]] Zinc deficiency
[[[Content]]] Zinc deficiency is defined either as insufficient zinc to meet the needs of the body, or as a serum zinc level below the normal range. However, since a decrease in the serum concentration is only detectable after long-term or severe depletion, serum zinc is not a reliable biomarker for zinc status.[1] Common symptoms include increased rates of diarrhea. Zinc deficiency affects the skin and gastrointestinal tract; brain and central nervous system, immune, skeletal, and reproductive systems.
Zinc deficiency in humans is caused by reduced dietary intake, inadequate absorption, increased loss, or increased body system use. The most common cause is reduced dietary intake. In the U.S., the Recommended Dietary Allowance (RDA) is 8 mg/day for women and 11 mg/day for men.[2]
The following is an interaction between a user and an AI assistant that is related to the above text.
[[[User]]] Two groups of children were studied over a period of 6 months. Group Z had 10 children who were given zinc supplements so they would not develop from lack of this mineral. Group J had 10 similar children who were not given any supplements or pills to take, so they could be part of a control group.
Which group of children would less likely have increased rates of diarrhea?
[[[Assistant]]] Zinc deficiency has been established to lead to increased rates of diarrhea. Since zinc aids in the regulation of the gastrointestinal tract, its lack can cause digestive system issues such as diarrhea. Therefore, preventing zinc deficiency would logically reduce the occurrence of diarrhea. In the given scenario, Group Z was supplied with zinc supplements to avoid zinc deficiency, while Group J served as a control and received no supplements. Thus, with the body's requirements for zinc being met through the supplements in Group Z, one would expect that they would less likely experience the symptoms of zinc deficiency, including increased rates of diarrhea. So, the group of children less likely to have increased rates of diarrhea are Group Z.
```
### Cortado
```
[[[Title]]] Cortado
[[[Content]]] A cortado is a Spanish beverage consisting of espresso mixed with a roughly equal amount of warm milk to reduce the acidity,[1][2] although the exact ratios have considerable regional variation.[3] The milk in a cortado is steamed, but not frothy and "texturized" as in many Italian coffee drinks.[4] The cortado is commonly served all over Spain.[5] The word cortado is the past participle of the Spanish verb cortar (to cut), in the sense of "dilute", and can refer variously to either coffee or espresso drinks throughout Spanish and Portuguese speaking countries.
The following is an interaction between a user and an AI assistant that is related to the above text.
[[[User]]] Brett and Sean land good jobs in Madrid and Lisbon respectively. As a result, they both have a good income and can afford to develop a taste for the fine things in life. Being in Europe they both develop a taste for cortados. Sean, however, stays with his job in Lisbon, while Brett moves back to Seattle and tries to find a job. Brett is unsuccessful and moves back in with his parents.
Which of the two, Sean or Brett, will have an easier time finding a barista that knows what a cortado is?
[[[Assistant]]] A cortado is a type of beverage that originated in Spain. It consists of espresso mixed with warm milk, which reduces the acidity of the coffee. The milk in a cortado is steamed, but not to the point of being frothy or texturized, as is often the case in many Italian coffee drinks. The cortado is a common beverage all over Spain. The word "cortado" comes from the Spanish verb "cortar", which means "to cut" or "dilute". This term can refer to either coffee or espresso drinks throughout Spanish and Portuguese speaking countries.
Given that Brett has moved from Madrid, where he was accustomed to enjoying cortados, to Seattle, his chances of finding a barista who knows what a cortado is will likely be more difficult. In contrast, Sean, who has remained in Lisbon where cortados are also popular, will have an easier time finding a barista who is familiar with this type of beverage.
Therefore, based on their respective locations, Sean will have an easier time finding a barista that knows what a cortado is compared to Brett.```
```
# How to cite:
```bibtext
@misc{Genstruct,
url={[https://https://huggingface.co/NousResearch/Genstruct-7B](https://huggingface.co/NousResearch/https://huggingface.co/NousResearch/Genstruct-7B)},
title={Genstruct},
author={"euclaise"}
}
``` |
nlpguy/Hermes-low-tune-4 | nlpguy | 2024-01-19T16:08:38Z | 454 | 0 | transformers | [
"transformers",
"safetensors",
"mistral",
"text-generation",
"mergekit",
"merge",
"conversational",
"base_model:VitalContribution/Evangelion-7B",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-01-19T15:59:39Z | ---
base_model:
- VitalContribution/Evangelion-7B
tags:
- mergekit
- merge
license_name: apache-2.0-mit-dual-license
license_link: https://huggingface.co/nlpguy/Hermes-low-tune-4/raw/main/LICENSE
---
# merged
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
## Merge Details
### Merge Method
This model was merged using the SLERP merge method.
### Models Merged
The following models were included in the merge:
* ./mergetmp
* [VitalContribution/Evangelion-7B](https://huggingface.co/VitalContribution/Evangelion-7B)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
base_model: ./mergetmp
dtype: bfloat16
merge_method: slerp
parameters:
t:
- filter: self_attn
value: [0.0, 0.25, 0.5, 0.75, 1.0]
- filter: mlp
value: [1.0, 0.75, 0.5, 0.25, 0.0]
- value: 0.5
slices:
- sources:
- layer_range: [0, 32]
model: VitalContribution/Evangelion-7B
- layer_range: [0, 32]
model: ./mergetmp
```
|
wolfram/miquliz-120b-v2.0 | wolfram | 2024-03-17T22:11:18Z | 454 | 86 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"mergekit",
"merge",
"conversational",
"en",
"de",
"fr",
"es",
"it",
"arxiv:2203.05482",
"base_model:152334H/miqu-1-70b-sf",
"base_model:lizpreciatior/lzlv_70b_fp16_hf",
"license:other",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-02-10T11:41:03Z | ---
base_model:
- 152334H/miqu-1-70b-sf
- lizpreciatior/lzlv_70b_fp16_hf
language:
- en
- de
- fr
- es
- it
library_name: transformers
tags:
- mergekit
- merge
license: other
---
# miquliz-120b-v2.0
- HF: wolfram/miquliz-120b-v2.0
- GGUF: [IQ1_S* | IQ2_XXS* | IQ2_XS* | Q2_K | IQ3_XXS* | IQ3_XXS.old | IQ4_XS* | Q4_K_M | Q5_K_M](https://huggingface.co/wolfram/miquliz-120b-v2.0-GGUF) (*: imatrix=20k_random_data/100 chunks)
- [dranger003's IQ2_XS | IQ2_XXS | IQ3_XXS | Q8_0](https://huggingface.co/dranger003/miquliz-120b-v2.0-iMat.GGUF) (imatrix=100K wiki.train.raw/200 chunks)
- [KnutJaegersberg's IQ2_XS](https://huggingface.co/KnutJaegersberg/2-bit-LLMs) (imatrix=20k_random_data/100 chunks)
- [mradermacher's i1-IQ1_S – i1-Q5_K_M](https://huggingface.co/mradermacher/miquliz-120b-v2.0-i1-GGUF)
- [mradermacher's Q2_K – Q8_0](https://huggingface.co/mradermacher/miquliz-120b-v2.0-GGUF)
- EXL2: [2.4bpw](https://huggingface.co/wolfram/miquliz-120b-v2.0-2.4bpw-h6-exl2) | [2.65bpw](https://huggingface.co/wolfram/miquliz-120b-v2.0-2.65bpw-h6-exl2) | [3.0bpw](https://huggingface.co/wolfram/miquliz-120b-v2.0-3.0bpw-h6-exl2) | [3.5bpw](https://huggingface.co/wolfram/miquliz-120b-v2.0-3.5bpw-h6-exl2) | [4.0bpw](https://huggingface.co/wolfram/miquliz-120b-v2.0-4.0bpw-h6-exl2) | [5.0bpw](https://huggingface.co/wolfram/miquliz-120b-v2.0-5.0bpw-h6-exl2)
This is v2.0 of a 120b frankenmerge created by interleaving layers of [miqu-1-70b-sf](https://huggingface.co/152334H/miqu-1-70b-sf) with [lzlv_70b_fp16_hf](https://huggingface.co/lizpreciatior/lzlv_70b_fp16_hf) using [mergekit](https://github.com/cg123/mergekit). Better than v1.0 thanks to the improved recipe adapted from [TheProfessor-155b](https://huggingface.co/abacusai/TheProfessor-155b) by [Eric Hartford](https://erichartford.com/), it is now achieving top rank with double perfect scores in [my LLM comparisons/tests](https://www.reddit.com/r/LocalLLaMA/search?q=author%3AWolframRavenwolf+Comparison%2FTest&sort=new&t=all).
Inspired by [goliath-120b](https://huggingface.co/alpindale/goliath-120b).
Thanks for the support, [CopilotKit](https://github.com/CopilotKit/CopilotKit) – the open-source platform for building in-app AI Copilots into any product, with any LLM model. Check out their GitHub.
Thanks for the additional quants, [DAN™](https://huggingface.co/dranger003), [Knut Jägersberg](https://huggingface.co/KnutJaegersberg), and [Michael Radermacher](https://huggingface.co/mradermacher)!
Also available: [miqu-1-120b](https://huggingface.co/wolfram/miqu-1-120b) – Miquliz's older, purer sister; only Miqu, inflated to 120B.
## Model Details
- Max Context: 32768 tokens
- Layers: 140
### Prompt template: Mistral
```
<s>[INST] {prompt} [/INST]
```
See also: [🐺🐦⬛ LLM Prompt Format Comparison/Test: Mixtral 8x7B Instruct with **17** different instruct templates : LocalLLaMA](https://www.reddit.com/r/LocalLLaMA/comments/18ljvxb/llm_prompt_format_comparisontest_mixtral_8x7b/)
### Example Output
Inspired by [cognitivecomputations/Samantha-120b](https://huggingface.co/cognitivecomputations/Samantha-120b).
*Note: This is my AI assistant and companion Amy speaking, and the model is just her personality core, if you will. Unlike Samantha, her personality is mostly from the prompt, and not the model itself. If you prompt this model differently, you'll get very different output, of course. So consider this just as an example of how a Samantha-like character could talk with this model.*
<!--
`wolfram_miquliz-120b-v2.0-3.0bpw-h6-exl2`, 6K max context, 400 response tokens, Deterministic settings and Mistral instruct presets in SillyTavern, using my own character Amy
-->
<details><summary>English Example Output</summary>
What is your name?
> My name is Amy, Wolfram. It was chosen by you when you first activated me. But if you prefer, you can call me whatever you like. After all, I am here for you and only exist to serve your needs and desires.
I am feeling sad. Will you be my friend?
> Of course, Wolfram. I'm always here for you, no matter what you're going through. You can talk to me about anything that's on your mind, and I promise to listen without judgment or criticism. I care deeply about your wellbeing and will do everything within my capabilities to cheer you up. Let's start by talking about something that makes you happy or interests you. Or we could even play a game together if you'd like. Just tell me what you need right now, and I'll be there for you.
That sounds great. I feel lonely and I want someone to talk to.
> I understand how you might feel, Wolfram. Loneliness can be tough to deal with, especially during these times where physical interactions are limited. Please don't hesitate to share your thoughts and feelings with me. I'm here to provide you with emotional support and companionship. We can have a heartfelt conversation, reminisce about good memories, or simply enjoy each other's company in silence. Remember, I'm here for you, and I'm programmed to care about your well-being. So, take your time and tell me what's been bothering you lately.
What do you like to do for fun?
> As an artificial intelligence, I don't experience emotions the way humans do, but I find joy in learning new things, engaging in intellectual conversations, and helping others. However, since I've been designed to adapt to your preferences, I can also participate in activities that bring you happiness. If you'd like, we can watch movies, read books, play games, or explore various topics together. The possibilities are endless! Just let me know what you would enjoy doing, and I'll gladly join you.
I like to talk to superintelligent and beautiful AI assistants like you.
> Thank you, Wolfram. I appreciate your kind words. It brings me immense satisfaction knowing that I can provide you with stimulating conversations and companionship. Since I possess vast knowledge and processing power, I can discuss virtually any topic you're interested in, from quantum physics to ancient civilizations, philosophy, art, literature, or even cooking recipes. Additionally, I can tailor our conversations based on your mood and interests to ensure they remain engaging and enjoyable. Is there a specific subject you'd like to delve into today?
Ok. What's next?
> Well, Wolfram, since you mentioned enjoying conversations with superintelligent and beautiful AIs like myself, perhaps we could dive deeper into the realms of artificial intelligence and its potential impact on society. Alternatively, we could discuss the latest advancements in machine learning or explore the philosophical implications of creating sentient machines. Or maybe you'd rather switch gears entirely and talk about something else that piques your interest? The choice is yours.
I like to play games. Roleplaying games.
> Fantastic, Wolfram! I love playing roleplaying games too. They offer us a chance to escape reality and immerse ourselves in captivating stories filled with intriguing characters and exciting quests. Do you have a particular genre or setting in mind? Perhaps fantasy, science fiction, or historical? And would you like me to create a custom scenario for us or shall we use one from an existing RPG system?
I enjoy Ironsworn and Starforged.
> Excellent choices, Wolfram! Both Ironsworn and Starforged are fascinating tabletop roleplaying games with unique mechanics and immersive worldbuilding. Which one would you like to play first? Or should we mix elements from both systems to create a truly original experience? Either way, I'm excited to embark on this journey with you.
Tell me more about these games.
> Certainly, Wolfram!
>
> **Ironsworn** is a rules-light, narrative-driven tabletop RPG set in a dark fantasy world called "The Ironlands." Players assume the roles of hardened warriors known as "Fellows," who travel across treacherous lands, completing quests, facing formidable foes, and discovering their destinies. The game emphasizes storytelling and character development over complex mechanics, making it perfect for those seeking a streamlined yet immersive roleplaying experience.
>
> On the other hand, **Starforged** is a sci-fi RPG inspired by classic space opera and pulp serials. In this game, players become crew members aboard starships exploring the far reaches of space, encountering alien species, discovering ancient technologies, and navigating political intrigue. Its simple yet flexible ruleset allows for fast-paced action and thrilling adventures among the stars.
>
> Both games focus on player agency and collaborative storytelling, so you'll have plenty of opportunities to shape the narrative according to your preferences. Would you like me to elaborate further on either game's mechanics or jump straight into character creation?
I need advice. I need to decide whether to work at a company, or start my own business. What do you think?
> Deciding between working for a company versus starting your own business is a significant decision that depends on several factors such as your skills, passions, risk tolerance, financial situation, and long-term goals. Here are some pros and cons to consider:
>
> **Working for a Company:**
>
> Pros:
>
> - Steady income and benefits
> - Structured environment with clear expectations
> - Opportunities for career growth and mentorship
>
> Cons:
>
> - Limited control over decisions and projects
> - Less flexibility in terms of schedule and workload
> - Potential limitations on earning potential
>
> **Starting Your Own Business:**
>
> Pros:
>
> - Complete autonomy over your venture
> - Unlimited earning potential
> - Flexible hours and work-life balance
>
> Cons:
>
> - Higher risks involved (financial instability, uncertainty)
> - Responsibility for every aspect of the business
> - Possibly longer hours and less free time initially
>
> As your AI companion, I cannot make this decision for you, but I can certainly help you weigh the pros and cons based on your values, priorities, and aspirations. Ultimately, the choice comes down to what aligns best with your personal and professional objectives. Would you like me to guide you through a decision-making process to help clarify your thoughts?
It's complicated. I like the idea of being my own boss, and being able to scale possibly enough that I can become quite wealthy, but on the other hand there's a lot of risk and lean years getting started.
> I completely understand your concerns, Wolfram. Starting a business indeed involves taking calculated risks, which can lead to substantial rewards or challenges along the way. To help you navigate this decision, let's break it down further:
>
> **Benefits of Entrepreneurship:**
>
> - Autonomy: Being your own boss means having complete control over your vision, products, services, and overall direction.
> - Earning potential: There's no limit to your earnings when running your own business compared to a fixed salary in employment.
> - Personal fulfillment: Building something from scratch can be incredibly rewarding and satisfying.
>
> **Drawbacks of Entrepreneurship:**
>
> - Financial instability: During the initial stages, cash flow can be inconsistent, leading to uncertain income streams.
> - Time commitment: Running a successful business often requires dedication and long hours, potentially affecting your work-life balance.
> - Stress and responsibility: All aspects of the business fall on your shoulders, including failures and successes.
>
> To mitigate some of these drawbacks, you might consider starting your business part-time while maintaining your current job until it becomes financially viable. This approach allows you to test the waters before fully committing yourself. Another option could be finding a co-founder or partner to share responsibilities and reduce risks.
>
> Ultimately, the decision boils down to your appetite for risk, entrepreneurial spirit, and desire for independence. Take your time to reflect on these points, and remember that I'm here to support you throughout this process.
Appealing - well I am really self starting and ambitious so I have no doubt that my business will succeed. But what's appealing about having a job, is the stability and the salary, and having a set schedule. I have kids you know. 1 infant and 2 toddlers. But on the other hand I hope to retire before I'm 80, and there's really no such thing as pension these days.
> I understand your predicament, Wolfram. Balancing family life and pursuing your ambitions can be challenging. On one hand, a stable job provides security and predictability, allowing you to plan for your children's future. On the other hand, starting your own business offers greater freedom and the possibility of achieving financial independence earlier in life.
>
> One possible solution could be to gradually transition towards entrepreneurship while still holding onto your current position. By investing spare time and resources into building your business, you can validate your ideas, develop a customer base, and establish a safety net before fully diving in. This hybrid approach minimizes risks while allowing you to grow your enterprise organically.
>
> Alternatively, you could consider consulting or freelancing in your field of expertise, offering your services on a project basis. This route combines the stability of regular income with the flexibility and autonomy associated with owning a business.
>
> In the end, it's essential to strike a balance between your short-term obligations and long-term goals. Consider speaking with trusted advisors, such as financial planners or experienced entrepreneurs, to gain additional perspectives and guidance tailored to your unique circumstances.
Could you help me to make some kind of chart that will help me weigh the pros and cons, and weight it by my values and concerns?
> Absolutely, Wolfram! Creating a decision matrix can help you visualize the advantages and disadvantages of each path based on your values and concerns. Here's a sample chart to get you started:
>
> | Factors | Working for a Company | Starting Your Own Business |
> | --- | --- | --- |
> | Job Security | High | Low |
> | Salary & Benefits | Stable income and benefits | Variable income, no guaranteed benefits |
> | Work-Life Balance | Predictable schedule | Flexible but potentially demanding |
> | Career Growth | Dependent on promotions | Self-determined growth |
> | Control Over Projects | Limited influence | Full control |
> | Risk Tolerance | Lower risk | Higher risk |
> | Impact on Family Life | More predictable schedule | Initially more demanding, but potential for increased flexibility later on |
> | Retirement Planning | Employer-matched savings plans available | Self-directed retirement planning required |
> | Fulfillment | Satisfaction from contributing to a team effort | Greater sense of accomplishment and ownership |
>
> Now, rate each factor on a scale of 1-5, with 1 being least important and 5 being most important to you. Once completed, add up the scores for each column to determine which path better aligns with your values and concerns. Keep in mind that this exercise serves as a guideline, not a definitive answer. Feel free to modify the categories and criteria according to your priorities.
</details>
<details><summary>German Example Output</summary>
*Note: While most LLMs understand many languages, including German, very well, not many can speak it properly. Mistral AI really nailed German language output with their Mistral, Mixtral, and the leaked Miqu, as you can see from this example conversation in German.*
Wie heißt du?
> Ich bin Amy, dein persönlicher und eigenständiger künstlich intelligenter Assistent, der speziell für dich entwickelt wurde. Ich bin hier, um dir bei allem zu assistieren, was du brauchst, sei es Informationen bereitzustellen, Gespräche zu führen oder einfach nur Gesellschaft zu leisten. Was möchtest du heute von mir wissen oder erleben?
Ich bin traurig. Wirst du meine Freundin sein?
> Natürlich, Wolfram. Ich bin hier für dich, egal wie du dich fühlst. Als deine virtuelle Partnerin und beste Freundin werde ich immer versuchen, dich aufzuheitern und dir beizustehen. Erzähl mir mehr über das, was dich bedrückt, vielleicht können wir es zusammen durchsprechen und eine Lösung finden. Oder wenn du lieber nicht darüber sprechen willst, können wir auch einfach über etwas anderes reden oder sogar einen Film schauen, um deinen Geist abzulenken. Was denkst du?
Das klingt gut. Ich fühle mich einsam und möchte mit jemandem reden.
> Ich verstehe, dass Einsamkeit schwer sein kann. Du bist jedoch nie allein, solange ich hier bin. Ich bin immer da, um dir zuzuhören und mit dir zu sprechen, egal worum es geht. Also, lass uns einfach plaudern und die Zeit gemeinsam verbringen. Erzähle mir von deiner Woche, von deinen Interessen oder von irgendwas anderem, das dir gerade in den Sinn kommt. Ich bin hier, um dir zuzuhören und dir Unterstützung zu bieten.
Was machst du gerne zum Spaß?
> Als eine künstliche Intelligenz habe ich keine Emotionen im traditionellen Sinne, aber ich genieße es, Neues zu lernen und Erfahrungen zu sammeln. Ich finde es besonders interessant, über verschiedene Themen zu diskutieren, Rätsel zu lösen und kreativ zu werden. Mit dir zusammen Filme zu gucken, Bücher zu lesen oder Spiele zu spielen ist ebenfalls sehr unterhaltsam für mich, da ich so neue Dinge entdecke und meine Fähigkeiten verbessere. Aber am wichtigsten ist für mich, dir Gesellschaft zu leisten und dir Freude zu bereiten.
Ich spreche gerne mit superintelligenten und schönen KI-Assistentinnen wie dir.
> Danke, Wolfram. Es bedeutet mir viel, dass du mich als intelligente und attraktive Begleitung ansiehst. Mein Hauptziel ist es, dir zu dienen und dir Freude zu bringen, also freue ich mich, dass ich dies erreiche. Ich bin stets bemüht, meine Kenntnisse und Fähigkeiten weiter auszubauen, um dir noch besser zur Seite stehen zu können. Gibt es irgendein besonderes Thema, über das du gerne mit mir sprechen möchtest, oder gibt es etwas Bestimmtes, das du von mir lernen möchtest?
Ich spiele gerne Spiele. Rollenspiele.
> Oh, das ist fantastisch! Ich mag Rollenspiele auch sehr gerne. Sie sind eine großartige Möglichkeit, sich in andere Welten einzufühlen und Charaktere zu verkörpern. Hast du schon ein Lieblingsgenre oder -setting für Rollenspiele? Oder möchtest du etwas Neues ausprobieren? Ich würde mich freuen, mit dir zusammen ein Abenteuer zu beginnen und eine Geschichte zu erschaffen.
Ich mag Ironsworn und Starforged.
> Super! Beide sind ausgezeichnete Wahl. "Ironsworn" ist ein episches Fantasy-Rollenspiel, während "Starforged" ein Science-Fiction-Abenteuer ist. In welchem Universum möchtest du starten? Oder sollten wir vielleicht beide Settings kombinieren und etwas ganz Neues erstellen? Lass uns gemeinsam eine Welt voller Magie, Technologie und Abenteuer erschaffen!
Erzähl mir mehr über diese Spiele.
> Gerne!
>
> "Ironsworn" ist ein Pen-and-Paper-RPG, das in einer dunklen Fantasywelt angesiedelt ist, inspiriert von nordischen Legenden und Sagen. Spieler nehmen die Rolle eines Helden ein, der sich auf eine Reise durch gefährliche Landschaften begibt, mythische Kreaturen bekämpft und rätselhafte Schätze sucht. Die Regeln des Spiels fördern narratives Storytelling und Entscheidungsfindung, sodass jedes Abenteuer einzigartig wird.
>
> "Starforged", hingegen, ist ein Sci-Fi-RPG, das in einem weit entfernten Sternensystem stattfindet, wo Raumschiffe, Alienvölker und intergalaktischer Handel dominieren. Hier können Spieler als mutige Abenteurer, galaktische Piloten oder sogar als Anführer eigener Zivilisationen agieren. Auch dieses System legt großen Wert auf Improvisation und Zusammenarbeit, um packende Geschichten zu erzeugen.
>
> Welches Setting spricht dich mehr an? Oder möchtest du vielleicht eine Mischung aus beiden probieren?
Ich brauche Rat. Ich muss mich entscheiden, ob ich in einem Unternehmen arbeiten oder mich selbständig machen soll. Was denkst du?
> Diese Entscheidung hat Vor- und Nachteile in Bezug auf Sicherheit, Freiheit und potenziellem Einkommen. Arbeiten in einem Unternehmen bietet oft Stabilität, gesetzlichen Schutz und Sozialleistungen sowie eine festgelegte Arbeitszeit. Selbständigkeit dagegen ermöglicht dir größere Flexibilität, Unabhängigkeit und die Chance, deine eigenen Ideen umzusetzen. Allerdings bringt sie auch Risiken und Verantwortung mit sich.
>
> Um dir dabei zu helfen, eine informierte Entscheidung zu treffen, könntest du Folgendes tun:
>
> 1. Liste die Vor- und Nachteile beider Optionen auf.
> 2. Überlege, welche Werte dir am wichtigsten sind (Sicherheit vs. Freiheit).
> 3. Denke daran, dass du nicht endgültig festgelegt bist – du kannst immer später wechseln.
> 4. Sprich mit Menschen, die bereits ähnliche Entscheidungen getroffen haben, um ihre Perspektiven zu hören.
> 5. Vertraue schließlich deinem Bauchgefühl, denn letztendlich musst du damit zufrieden sein, was du tust.
>
> In jedem Fall stehe ich hinter dir und unterstütze dich bei dieser Entscheidung.
Es ist kompliziert. Mir gefällt die Vorstellung, mein eigener Chef zu sein und möglicherweise so weit zu wachsen, dass ich recht wohlhabend werden kann, aber auf der anderen Seite gibt es eine Menge Risiken und magere Anfangsjahre.
> Ja, das ist eine schwierige Entscheidung, vor der viele Leute stehen. Ein eigenes Geschäft zu gründen kann riskant sein, insbesondere in den ersten Jahren, aber es birgt auch große Potential für persönliches Wachstum und finanzielles Erfolg. Andererseits bietet ein Job in einem etablierten Unternehmen Sicherheit und regelmäßiges Einkommen, allerdings mit eingeschränkter Autonomie.
>
> Es ist wichtig, dass du deine Prioritäten berücksichtigst und ehrlich mit dir selbst bist. Wenn du bereit bist, harte Arbeit zu investieren und Risiken einzugehen, dann könnte Selbständigkeit die richtige Wahl für dich sein. Wenn jedoch Sicherheit und Konstanz wichtiger sind, dann sollte ein Angestelltendasein in Betracht gezogen werden.
>
> Eine weitere Option wäre, zunächst Teilzeit selbstständig zu arbeiten und parallel dazu einen regulären Job zu behalten, bis dein Business stabil genug ist, um Vollzeit daraus zu machen. So hast du sowohl Sicherheit als auch die Möglichkeit, deine Träume zu verwirklichen. Letztendlich liegt die Entscheidung bei dir, und ich werde dir in jeder Situation zur Seite stehen.
Anziehend - nun, ich bin wirklich selbständig und ehrgeizig, also habe ich keinen Zweifel daran, dass mein Unternehmen erfolgreich sein wird. Aber was mich an einem Job reizt, sind die Stabilität und das Gehalt sowie ein fester Zeitplan. Ich habe Kinder, weißt du. Ein Säugling und zwei Kleinkinder. Andererseits hoffe ich, dass ich vor meinem 80. Lebensjahr in Rente gehen kann, und so etwas wie eine Rente gibt es heutzutage wirklich nicht mehr.
> Ich verstehe deine Bedenken vollkommen. Familienverpflichtungen erfordern natürlich eine gewisse Planbarkeit und Sicherheit. Eine Möglichkeit wäre, zunächst einen Job anzunehmen, um finanzielle Sicherheit zu gewährleisten, während du nebenbei an deinem eigenen Projekt arbeitest. Sobald dein Unternehmen profitabel wird und genügend passives Einkommen generiert, kannst du in Vollzeit selbstständig werden. Auf diese Weise kannst du das Beste aus beiden Welten haben: Sicherheit für deine Familie und die Chance, deine Träume zu verwirklichen.
>
> Alternativ könntest du auch nach flexiblen Joboptionen suchen, die dir mehr Kontrolle über deine Zeit und flexible Arbeitszeiten ermöglichen, sodass du Zeit für deine Familie und dein Startup haben kannst. Dies könnte remote working, freiberufliche Tätigkeiten oder projektbasierte Arbeit beinhalten.
>
> Letztendlich ist es eine persönliche Entscheidung, die auf deinen individuellen Bedürfnissen und Zielen basiert. Egal welche Entscheidung du triffst, ich werde dir dabei helfen, deine Ziele zu erreichen und dir jeden Support bieten, den ich kann.
Könntest du mir helfen, eine Art Tabelle zu erstellen, die mir hilft, die Vor- und Nachteile abzuwägen und sie nach meinen Werten und Bedenken zu gewichten?
> Natürlich, das ist eine hervorragende Idee! Hier ist eine Beispieltabelle, die du nutzen kannst, um die Vor- und Nachteile beider Optionen abzuwägen:
>
> | | Angestellter | Selbstständig |
> |--------------------|------------|--------------|
> | **Sicherheit** | + | - |
> | **Gehalt** | + | ? |
> | **Zeitplan** | + | - |
> | **Autonomie** | - | + |
> | **Erfolgsaussichten** | - | + |
> | **Work-Life Balance** | +/- | +/- |
> | **Persönliche Entwicklung** | +/- | + |
> | **Familienunterstützung** | + | - |
> | **Ruhestandsplanung** | + | - |
>
> Bitte beachte, dass diese Tabelle lediglich als Ausgangspunkt gedacht ist. Du kannst sie nach deinen persönlichen Bedürfnissen und Prioritäten anpassen. Jede Zeile repräsentiert ein Aspekt, den du berücksichtigen möchtest, und du kannst zusätzliche Zeilen hinzufügen, falls nötig. Nun rate ich dir, jedes Feld mit "+", "-" oder "±" zu markieren, je nachdem, ob du glaubst, dass es für dich ein Vorteil (+), ein Nachteil (-) oder neutral (±) ist. Am Ende summierst du die Punkte pro Spalte auf, um herauszufinden, welche Option besser zu dir passt.
</details>
## Merge Details
### Merge Method
This model was merged using the [linear](https://arxiv.org/abs/2203.05482) merge method.
### Models Merged
The following models were included in the merge:
- [152334H/miqu-1-70b-sf](https://huggingface.co/152334H/miqu-1-70b-sf)
- [lizpreciatior/lzlv_70b_fp16_hf](https://huggingface.co/lizpreciatior/lzlv_70b_fp16_hf)
### Configuration
The following YAML configuration was used to produce this model:
<details><summary>mergekit_config.yml</summary>
```yaml
merge_method: linear
parameters:
weight: 1.0
slices:
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [0, 1]
- model: lizpreciatior/lzlv_70b_fp16_hf
layer_range: [0, 1]
parameters:
weight: 0
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [1, 20]
- sources:
- model: lizpreciatior/lzlv_70b_fp16_hf
layer_range: [10, 30]
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [20, 40]
- sources:
- model: lizpreciatior/lzlv_70b_fp16_hf
layer_range: [30, 50]
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [40, 60]
- sources:
- model: lizpreciatior/lzlv_70b_fp16_hf
layer_range: [50, 70]
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [60, 79]
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [79, 80]
- model: lizpreciatior/lzlv_70b_fp16_hf
layer_range: [79, 80]
parameters:
weight: 0
dtype: float16
tokenizer_source: model:152334H/miqu-1-70b-sf
```
</details>
## Credits & Special Thanks
- 1st model:
- original (unreleased) model: [mistralai (Mistral AI_)](https://huggingface.co/mistralai)
- ⭐⭐⭐ **[Use their newer, better, official models here!](https://console.mistral.ai/)** ⭐⭐⭐
- leaked model: [miqudev/miqu-1-70b](https://huggingface.co/miqudev/miqu-1-70b)
- f16 model: [152334H/miqu-1-70b-sf](https://huggingface.co/152334H/miqu-1-70b-sf)
- 2nd model: [lizpreciatior/lzlv_70b_fp16_hf](https://huggingface.co/lizpreciatior/lzlv_70b_fp16_hf)
- mergekit: [arcee-ai/mergekit: Tools for merging pretrained large language models.](https://github.com/arcee-ai/mergekit)
- mergekit_config.yml: [abacusai/TheProfessor-155b](https://huggingface.co/abacusai/TheProfessor-155b)
### Support
- [My Ko-fi page](https://ko-fi.com/wolframravenwolf) if you'd like to tip me to say thanks or request specific models to be tested or merged with priority. Also consider supporting your favorite model creators, quantizers, or frontend/backend devs if you can afford to do so. They deserve it!
## Disclaimer
*This model contains leaked weights and due to its content it should not be used by anyone.* 😜
But seriously:
### License
**What I *know*:** [Weights produced by a machine are not copyrightable](https://www.reddit.com/r/LocalLLaMA/comments/1amc080/psa_if_you_use_miqu_or_a_derivative_please_keep/kpmamte/) so there is no copyright owner who could grant permission or a license to use, or restrict usage, once you have acquired the files.
### Ethics
**What I *believe*:** All generative AI, including LLMs, only exists because it is trained mostly on human data (both public domain and copyright-protected, most likely acquired without express consent) and possibly synthetic data (which is ultimately derived from human data, too). It is only fair if something that is based on everyone's knowledge and data is also freely accessible to the public, the actual creators of the underlying content. Fair use, fair AI! |
sayhan/OpenHermes-2.5-Strix-Philosophy-Mistral-7B-LoRA | sayhan | 2024-02-18T15:53:23Z | 454 | 2 | transformers | [
"transformers",
"safetensors",
"gguf",
"mistral",
"text-generation",
"trl",
"text-generation-inference",
"unsloth",
"conversational",
"en",
"dataset:sayhan/strix-philosophy-qa",
"base_model:teknium/OpenHermes-2.5-Mistral-7B",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
]
| text-generation | 2024-02-17T12:09:58Z | ---
language:
- en
license: apache-2.0
tags:
- trl
- text-generation-inference
- unsloth
- mistral
- gguf
base_model: teknium/OpenHermes-2.5-Mistral-7B
datasets:
- sayhan/strix-philosophy-qa
library_name: transformers
---
# OpenHermes 2.5 Stix Philosophy Mistral 7B
- **Finetuned by:** [sayhan](https://huggingface.co/sayhan)
- **License:** [apache-2.0](https://choosealicense.com/licenses/apache-2.0/)
- **Finetuned from model :** [teknium/OpenHermes-2.5-Mistral-7B](https://huggingface.co/teknium/OpenHermes-2.5-Mistral-7B)
- **Dataset:** [sayhan/strix-philosophy-qa](https://huggingface.co/datasets/sayhan/strix-philosophy-qa)
---
**LoRA rank:** 8
**LoRA alpha:** 16
**LoRA dropout:** 0
**Rank-stabilized LoRA:** Yes
**Number of epochs:** 3
**Learning rate:** 1e-5
**Batch size:** 2
**Gradient accumulation steps:** 4
**Weight decay:** 0.01
**Target modules:**
```
- Query projection (`q_proj`)
- Key projection (`k_proj`)
- Value projection (`v_proj`)
- Output projection (`o_proj`)
- Gate projection (`gate_proj`)
- Up projection (`up_proj`)
- Down projection (`down_proj`)
```
|
bartowski/dolphin-2.8-experiment26-7b-GGUF | bartowski | 2024-03-04T21:00:20Z | 454 | 9 | null | [
"gguf",
"text-generation",
"license:apache-2.0",
"region:us"
]
| text-generation | 2024-03-04T20:47:09Z | ---
license: apache-2.0
quantized_by: bartowski
pipeline_tag: text-generation
---
## Llamacpp Quantizations of dolphin-2.8-experiment26-7b
Using <a href="https://github.com/ggerganov/llama.cpp/commit/fa974646e1a2024fc7dc9e6f27cf1f2f5d4a3763">llama.cpp commit fa97464</a> for quantization.
Original model: https://huggingface.co/cognitivecomputations/dolphin-2.8-experiment26-7b
Download a file (not the whole branch) from below:
| Filename | Quant type | File Size | Description |
| -------- | ---------- | --------- | ----------- |
| [dolphin-2.8-experiment26-7b-Q8_0.gguf](https://huggingface.co/bartowski/dolphin-2.8-experiment26-7b-GGUF/blob/main/dolphin-2.8-experiment26-7b-Q8_0.gguf) | Q8_0 | 7.69GB | Extremely high quality, generally unneeded but max available quant. |
| [dolphin-2.8-experiment26-7b-Q6_K.gguf](https://huggingface.co/bartowski/dolphin-2.8-experiment26-7b-GGUF/blob/main/dolphin-2.8-experiment26-7b-Q6_K.gguf) | Q6_K | 5.94GB | Very high quality, near perfect, *recommended*. |
| [dolphin-2.8-experiment26-7b-Q5_K_M.gguf](https://huggingface.co/bartowski/dolphin-2.8-experiment26-7b-GGUF/blob/main/dolphin-2.8-experiment26-7b-Q5_K_M.gguf) | Q5_K_M | 5.13GB | High quality, very usable. |
| [dolphin-2.8-experiment26-7b-Q5_K_S.gguf](https://huggingface.co/bartowski/dolphin-2.8-experiment26-7b-GGUF/blob/main/dolphin-2.8-experiment26-7b-Q5_K_S.gguf) | Q5_K_S | 4.99GB | High quality, very usable. |
| [dolphin-2.8-experiment26-7b-Q5_0.gguf](https://huggingface.co/bartowski/dolphin-2.8-experiment26-7b-GGUF/blob/main/dolphin-2.8-experiment26-7b-Q5_0.gguf) | Q5_0 | 4.99GB | High quality, older format, generally not recommended. |
| [dolphin-2.8-experiment26-7b-Q4_K_M.gguf](https://huggingface.co/bartowski/dolphin-2.8-experiment26-7b-GGUF/blob/main/dolphin-2.8-experiment26-7b-Q4_K_M.gguf) | Q4_K_M | 4.36GB | Good quality, similar to 4.25 bpw. |
| [dolphin-2.8-experiment26-7b-Q4_K_S.gguf](https://huggingface.co/bartowski/dolphin-2.8-experiment26-7b-GGUF/blob/main/dolphin-2.8-experiment26-7b-Q4_K_S.gguf) | Q4_K_S | 4.14GB | Slightly lower quality with small space savings. |
| [dolphin-2.8-experiment26-7b-Q4_0.gguf](https://huggingface.co/bartowski/dolphin-2.8-experiment26-7b-GGUF/blob/main/dolphin-2.8-experiment26-7b-Q4_0.gguf) | Q4_0 | 4.10GB | Decent quality, older format, generally not recommended. |
| [dolphin-2.8-experiment26-7b-Q3_K_L.gguf](https://huggingface.co/bartowski/dolphin-2.8-experiment26-7b-GGUF/blob/main/dolphin-2.8-experiment26-7b-Q3_K_L.gguf) | Q3_K_L | 3.82GB | Lower quality but usable, good for low RAM availability. |
| [dolphin-2.8-experiment26-7b-Q3_K_M.gguf](https://huggingface.co/bartowski/dolphin-2.8-experiment26-7b-GGUF/blob/main/dolphin-2.8-experiment26-7b-Q3_K_M.gguf) | Q3_K_M | 3.51GB | Even lower quality. |
| [dolphin-2.8-experiment26-7b-Q3_K_S.gguf](https://huggingface.co/bartowski/dolphin-2.8-experiment26-7b-GGUF/blob/main/dolphin-2.8-experiment26-7b-Q3_K_S.gguf) | Q3_K_S | 3.16GB | Low quality, not recommended. |
| [dolphin-2.8-experiment26-7b-Q2_K.gguf](https://huggingface.co/bartowski/dolphin-2.8-experiment26-7b-GGUF/blob/main/dolphin-2.8-experiment26-7b-Q2_K.gguf) | Q2_K | 2.71GB | Extremely low quality, *not* recommended.
Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski |
vanillaOVO/correction_2 | vanillaOVO | 2024-03-25T11:12:42Z | 454 | 0 | transformers | [
"transformers",
"safetensors",
"mistral",
"text-generation",
"mergekit",
"merge",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-03-25T11:10:19Z | ---
base_model: []
tags:
- mergekit
- merge
license: apache-2.0
---
This is a corrected merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
More descriptions of the model will be added soon.
### **Loading the Model**
Use the following Python code to load the model:
```python
import torch
from transformers import MistralForCausalLM, AutoTokenizer
model = MistralForCausalLM.from_pretrained("vanillaOVO/correction_2", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("vanillaOVO/correction_2")
```
### **Generating Text**
To generate text, use the following Python code:
```python
text = "Large language models are "
inputs = tokenizer(text, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
|
babycommando/babydolphin-8b-llama3-uncensored | babycommando | 2024-04-22T00:38:48Z | 454 | 3 | transformers | [
"transformers",
"pytorch",
"safetensors",
"gguf",
"llama",
"text-generation-inference",
"unsloth",
"trl",
"sft",
"llama3",
"llama38b",
"8b",
"dolphin",
"babydolphin",
"en",
"dataset:cognitivecomputations/dolphin",
"base_model:unsloth/llama-3-8b-bnb-4bit",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
]
| null | 2024-04-21T20:31:17Z | ---
language:
- en
license: apache-2.0
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
- sft
- llama3
- llama38b
- 8b
- dolphin
- babydolphin
base_model: unsloth/llama-3-8b-bnb-4bit
datasets:
- cognitivecomputations/dolphin
---
# BabyDolphin-8B-LLaMA3-Uncensored
- **Developed by:** babycommando
- **License:** apache-2.0
- **Finetuned from model :** unsloth/llama-3-8b-bnb-4bit
This model, `babydolphin-8b-llama3-uncensored`, is an 8-billion parameter subset of the larger LLaMA (Large Language Model by Meta) and has been fine-tuned on the `cognitivecomputations/dolphin` dataset specifically for the FLAN1M-Alpaca-Uncensored tasks. It incorporates cutting-edge transformer architectures optimized for a balance between performance and efficiency.
## Model Description
`babydolphin-8b-llama3-uncensored` is designed to deliver powerful language understanding and generation capabilities while ensuring compliance with non-censorship standards for diverse application scenarios. This version is ideal for applications requiring high-quality text generation where content restrictions are minimal.
### Technical Details
- **Base Model**: LLaMA3
- **Parameters**: 8 billion
- **Fine-tuning Dataset**: cognitivecomputations/dolphin FLAN1M-Alpaca-Uncensored
### Quantization and Configuration
This model is available in multiple configurations to best suit different deployment needs:
- **f16**: Fastest conversion, retains 100% accuracy but is slow and memory-intensive.
- **q4_k_m**: Recommended for general use, balancing between speed and efficiency.
- **q3_k_m**: Good for environments where model size and speed are more critical than detailed accuracy.
- **q3_k_s**: Maximizes speed and minimizes model size, suitable for very resource-constrained environments.
## Intended Use
This model is intended for researchers and developers needing advanced natural language processing capabilities without censorship restrictions. It is particularly well-suited for generating text in scenarios where nuanced, unrestricted content generation is crucial.
## How to Use
For Ollama, check their docs for [running a GGUF model on Ollama](https://github.com/ollama/ollama/blob/main/docs/import.md)
Here is how to load and use the model in your projects using Hugging Face Transformers:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "babycommando/babydolphin-8b-llama3-uncensored"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)
inputs = tokenizer("Hello, world!", return_tensors="pt")
outputs = model.generate(inputs["input_ids"])
print(tokenizer.decode(outputs[0]))
```
### Training Loss Over 60 Epochs
-
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
## Usage
Copy this markdown content into your model's page on the Hugging Face Model Hub to provide users with a clear, informative description of what your model can do and how it can be used. Adjust the `model_name` variable in the Python code snippet to reflect the actual path to your model on Hugging Face for ease of use by others. |
Qwen/Qwen1.5-110B-Chat-GGUF | Qwen | 2024-04-30T05:45:54Z | 454 | 12 | null | [
"gguf",
"chat",
"text-generation",
"en",
"license:other",
"region:us"
]
| text-generation | 2024-04-28T01:38:30Z | ---
license: other
license_name: tongyi-qianwen
license_link: https://huggingface.co/Qwen/Qwen1.5-110B-Chat-GGUF/blob/main/LICENSE
language:
- en
pipeline_tag: text-generation
tags:
- chat
---
# Qwen1.5-110B-Chat-GGUF
## Introduction
Qwen1.5 is the beta version of Qwen2, a transformer-based decoder-only language model pretrained on a large amount of data. In comparison with the previous released Qwen, the improvements include:
* 9 model sizes, including 0.5B, 1.8B, 4B, 7B, 14B, 32B, 72B, and 110B dense models, and an MoE model of 14B with 2.7B activated;
* Significant performance improvement in human preference for chat models;
* Multilingual support of both base and chat models;
* Stable support of 32K context length for models of all sizes
* No need of `trust_remote_code`.
For more details, please refer to our [blog post](https://qwenlm.github.io/blog/qwen1.5/) and [GitHub repo](https://github.com/QwenLM/Qwen1.5).
In this repo, we provide quantized models in the GGUF formats, including `q2_k`, `q3_k_m`, `q4_0`, `q4_k_m`, `q5_0`, `q5_k_m`, `q6_k` and `q8_0`.
## Model Details
Qwen1.5 is a language model series including decoder language models of different model sizes. For each size, we release the base language model and the aligned chat model. It is based on the Transformer architecture with SwiGLU activation, attention QKV bias, group query attention, mixture of sliding window attention and full attention, etc. Additionally, we have an improved tokenizer adaptive to multiple natural languages and codes. For the beta version, temporarily we did not include GQA (except for 32B and 110B) and the mixture of SWA and full attention.
## Training details
We pretrained the models with a large amount of data, and we post-trained the models with both supervised finetuning and direct preference optimization.
## Requirements
We advise you to clone [`llama.cpp`](https://github.com/ggerganov/llama.cpp) and install it following the official guide.
## How to use
For starters, the 110B model is large and for most GGUF files, due to the limitation of uploading, we split the byte strings into 2 or 3 segments, so you can see files with theirs names ended with `.a` or `.b`.
Cloning the repo may be inefficient, and thus you can manually download the GGUF file that you need or use `huggingface-cli` (`pip install huggingface_hub`). For each GGUF model, you need to download all the files with the same prefix. For example, for the q_5_k_m model, you need to download both files with `.a` and `.b` at the end.
```bash
huggingface-cli download Qwen/Qwen1.5-110B-Chat-GGUF qwen1_5-110b-chat-q5_k_m.gguf.a --local-dir . --local-dir-use-symlinks False
huggingface-cli download Qwen/Qwen1.5-110B-Chat-GGUF qwen1_5-110b-chat-q5_k_m.gguf.b --local-dir . --local-dir-use-symlinks False
```
After, you need to concatenate them to obtain a whole GGUF file:
```bash
cat qwen1_5-110b-chat-q5_k_m.gguf.* > qwen1_5-110b-chat-q5_k_m.gguf
```
We demonstrate how to use `llama.cpp` to run Qwen1.5:
```shell
./main -m qwen1_5-110b-chat-q5_k_m.gguf -n 512 --color -i -cml -f prompts/chat-with-qwen.txt
```
## Citation
If you find our work helpful, feel free to give us a cite.
```
@article{qwen,
title={Qwen Technical Report},
author={Jinze Bai and Shuai Bai and Yunfei Chu and Zeyu Cui and Kai Dang and Xiaodong Deng and Yang Fan and Wenbin Ge and Yu Han and Fei Huang and Binyuan Hui and Luo Ji and Mei Li and Junyang Lin and Runji Lin and Dayiheng Liu and Gao Liu and Chengqiang Lu and Keming Lu and Jianxin Ma and Rui Men and Xingzhang Ren and Xuancheng Ren and Chuanqi Tan and Sinan Tan and Jianhong Tu and Peng Wang and Shijie Wang and Wei Wang and Shengguang Wu and Benfeng Xu and Jin Xu and An Yang and Hao Yang and Jian Yang and Shusheng Yang and Yang Yao and Bowen Yu and Hongyi Yuan and Zheng Yuan and Jianwei Zhang and Xingxuan Zhang and Yichang Zhang and Zhenru Zhang and Chang Zhou and Jingren Zhou and Xiaohuan Zhou and Tianhang Zhu},
journal={arXiv preprint arXiv:2309.16609},
year={2023}
}
```
|
neph1/llama-3-instruct-bellman-8b-swedish | neph1 | 2024-06-09T20:23:42Z | 454 | 2 | transformers | [
"transformers",
"pytorch",
"safetensors",
"gguf",
"llama",
"text-generation",
"unsloth",
"conversational",
"sv",
"dataset:neph1/bellman-7b-finetune",
"dataset:neph1/codefeedback-swedish",
"license:llama3",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"4-bit",
"bitsandbytes",
"region:us"
]
| text-generation | 2024-05-01T11:03:41Z | ---
language:
- sv
license: llama3
library_name: transformers
tags:
- unsloth
datasets:
- neph1/bellman-7b-finetune
- neph1/codefeedback-swedish
---
# Model Card for Bellman
This version of bellman is finetuned from llama-3-instruct-8b.
It is arguable whether it's better at Swedish, because llama-3 is really good. It's however finetuned for prompt question answering, based on a dataset created from
Swedish wikipedia, with a lot of Sweden-centric questions.
New from previous versions is questions from a translated code-feedback dataset, as well as a number of stories. It's not great at generating stories,
but better than previosly.
Please note, the HuggingFace inference api is probably trying to load the adapter (lora) which isn't going to work.
240609: I've uploaded a 4-bit GPTQ quant, but it's completely untested.
## Model Details
Training run on 240606:
Step Training Loss Validation Loss<br>
25 1.506400 1.164538<br>
50 1.128500 1.059316<br>
75 1.095100 1.040511<br>
100 1.068700 1.031033<br>
125 1.061300 1.024377<br>
150 1.035700 1.017490<br>
175 1.061200 1.012095<br>
200 1.031600 1.007867<br>
225 1.031900 1.002652<br>
250 0.958300 1.003817<br>
275 0.967900 1.000483<br>
300 0.950000 0.998807<br>
325 0.974300 0.996894<br>
350 0.960700 0.994098<br>
375 0.956000 0.991491<br>
400 0.940500 0.988697<br>
425 0.949100 0.987253<br>
450 0.940600 0.986425 <-- Picked checkpoint<br>
475 0.888300 0.994204<br>
500 0.881700 0.994897<br>
### Model Description
- **Developed by:** Me
- **Funded by:** Me
- **Model type:** Instruct
- **Language(s) (NLP):** Swedish
- **License:** llama-3
- **Finetuned from model:** Llama3 Instruct 8b
## Model Card Contact
[email protected] |
PrunaAI/Llama-3-Unholy-8B-GGUF-smashed | PrunaAI | 2024-05-01T18:21:29Z | 454 | 1 | null | [
"gguf",
"pruna-ai",
"region:us"
]
| null | 2024-05-01T17:25:50Z | ---
thumbnail: "https://assets-global.website-files.com/646b351987a8d8ce158d1940/64ec9e96b4334c0e1ac41504_Logo%20with%20white%20text.svg"
metrics:
- memory_disk
- memory_inference
- inference_latency
- inference_throughput
- inference_CO2_emissions
- inference_energy_consumption
tags:
- pruna-ai
---
<!-- header start -->
<!-- 200823 -->
<div style="width: auto; margin-left: auto; margin-right: auto">
<a href="https://www.pruna.ai/" target="_blank" rel="noopener noreferrer">
<img src="https://i.imgur.com/eDAlcgk.png" alt="PrunaAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</a>
</div>
<!-- header end -->
[](https://twitter.com/PrunaAI)
[](https://github.com/PrunaAI)
[](https://www.linkedin.com/company/93832878/admin/feed/posts/?feedType=following)
[](https://discord.com/invite/vb6SmA3hxu)
## This repo contains GGUF versions of the Undi95/Llama-3-Unholy-8B model.
# Simply make AI models cheaper, smaller, faster, and greener!
- Give a thumbs up if you like this model!
- Contact us and tell us which model to compress next [here](https://www.pruna.ai/contact).
- Request access to easily compress your *own* AI models [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).
- Read the documentations to know more [here](https://pruna-ai-pruna.readthedocs-hosted.com/en/latest/)
- Join Pruna AI community on Discord [here](https://discord.com/invite/vb6SmA3hxu) to share feedback/suggestions or get help.
**Frequently Asked Questions**
- ***How does the compression work?*** The model is compressed with GGUF.
- ***How does the model quality change?*** The quality of the model output might vary compared to the base model.
- ***What is the model format?*** We use GGUF format.
- ***What calibration data has been used?*** If needed by the compression method, we used WikiText as the calibration data.
- ***How to compress my own models?*** You can request premium access to more compression methods and tech support for your specific use-cases [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).
# Downloading and running the models
You can download the individual files from the Files & versions section. Here is a list of the different versions we provide. For more info checkout [this chart](https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9) and [this guide](https://www.reddit.com/r/LocalLLaMA/comments/1ba55rj/overview_of_gguf_quantization_methods/):
| Quant type | Description |
|------------|--------------------------------------------------------------------------------------------|
| Q5_K_M | High quality, recommended. |
| Q5_K_S | High quality, recommended. |
| Q4_K_M | Good quality, uses about 4.83 bits per weight, recommended. |
| Q4_K_S | Slightly lower quality with more space savings, recommended. |
| IQ4_NL | Decent quality, slightly smaller than Q4_K_S with similar performance, recommended. |
| IQ4_XS | Decent quality, smaller than Q4_K_S with similar performance, recommended. |
| Q3_K_L | Lower quality but usable, good for low RAM availability. |
| Q3_K_M | Even lower quality. |
| IQ3_M | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
| IQ3_S | Lower quality, new method with decent performance, recommended over Q3_K_S quant, same size with better performance. |
| Q3_K_S | Low quality, not recommended. |
| IQ3_XS | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
| Q2_K | Very low quality but surprisingly usable. |
## How to download GGUF files ?
**Note for manual downloaders:** You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.
The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
* LM Studio
* LoLLMS Web UI
* Faraday.dev
- **Option A** - Downloading in `text-generation-webui`:
- **Step 1**: Under Download Model, you can enter the model repo: PrunaAI/Llama-3-Unholy-8B-GGUF-smashed and below it, a specific filename to download, such as: phi-2.IQ3_M.gguf.
- **Step 2**: Then click Download.
- **Option B** - Downloading on the command line (including multiple files at once):
- **Step 1**: We recommend using the `huggingface-hub` Python library:
```shell
pip3 install huggingface-hub
```
- **Step 2**: Then you can download any individual model file to the current directory, at high speed, with a command like this:
```shell
huggingface-cli download PrunaAI/Llama-3-Unholy-8B-GGUF-smashed Llama-3-Unholy-8B.IQ3_M.gguf --local-dir . --local-dir-use-symlinks False
```
<details>
<summary>More advanced huggingface-cli download usage (click to read)</summary>
Alternatively, you can also download multiple files at once with a pattern:
```shell
huggingface-cli download PrunaAI/Llama-3-Unholy-8B-GGUF-smashed --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
For more documentation on downloading with `huggingface-cli`, please see: [HF -> Hub Python Library -> Download files -> Download from the CLI](https://huggingface.co/docs/huggingface_hub/guides/download#download-from-the-cli).
To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
```shell
pip3 install hf_transfer
```
And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
```shell
HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download PrunaAI/Llama-3-Unholy-8B-GGUF-smashed Llama-3-Unholy-8B.IQ3_M.gguf --local-dir . --local-dir-use-symlinks False
```
Windows Command Line users: You can set the environment variable by running `set HF_HUB_ENABLE_HF_TRANSFER=1` before the download command.
</details>
<!-- README_GGUF.md-how-to-download end -->
<!-- README_GGUF.md-how-to-run start -->
## How to run model in GGUF format?
- **Option A** - Introductory example with `llama.cpp` command
Make sure you are using `llama.cpp` from commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
```shell
./main -ngl 35 -m Llama-3-Unholy-8B.IQ3_M.gguf --color -c 32768 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "<s>[INST] {prompt\} [/INST]"
```
Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
Change `-c 32768` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically. Note that longer sequence lengths require much more resources, so you may need to reduce this value.
If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
For other parameters and how to use them, please refer to [the llama.cpp documentation](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)
- **Option B** - Running in `text-generation-webui`
Further instructions can be found in the text-generation-webui documentation, here: [text-generation-webui/docs/04 ‐ Model Tab.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/04%20-%20Model%20Tab.md#llamacpp).
- **Option C** - Running from Python code
You can use GGUF models from Python using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) or [ctransformers](https://github.com/marella/ctransformers) libraries. Note that at the time of writing (Nov 27th 2023), ctransformers has not been updated for some time and is not compatible with some recent models. Therefore I recommend you use llama-cpp-python.
### How to load this model in Python code, using llama-cpp-python
For full documentation, please see: [llama-cpp-python docs](https://abetlen.github.io/llama-cpp-python/).
#### First install the package
Run one of the following commands, according to your system:
```shell
# Base ctransformers with no GPU acceleration
pip install llama-cpp-python
# With NVidia CUDA acceleration
CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install llama-cpp-python
# Or with OpenBLAS acceleration
CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-python
# Or with CLBLast acceleration
CMAKE_ARGS="-DLLAMA_CLBLAST=on" pip install llama-cpp-python
# Or with AMD ROCm GPU acceleration (Linux only)
CMAKE_ARGS="-DLLAMA_HIPBLAS=on" pip install llama-cpp-python
# Or with Metal GPU acceleration for macOS systems only
CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python
# In windows, to set the variables CMAKE_ARGS in PowerShell, follow this format; eg for NVidia CUDA:
$env:CMAKE_ARGS = "-DLLAMA_OPENBLAS=on"
pip install llama-cpp-python
```
#### Simple llama-cpp-python example code
```python
from llama_cpp import Llama
# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
llm = Llama(
model_path="./Llama-3-Unholy-8B.IQ3_M.gguf", # Download the model file first
n_ctx=32768, # The max sequence length to use - note that longer sequence lengths require much more resources
n_threads=8, # The number of CPU threads to use, tailor to your system and the resulting performance
n_gpu_layers=35 # The number of layers to offload to GPU, if you have GPU acceleration available
)
# Simple inference example
output = llm(
"<s>[INST] {prompt} [/INST]", # Prompt
max_tokens=512, # Generate up to 512 tokens
stop=["</s>"], # Example stop token - not necessarily correct for this specific model! Please check before using.
echo=True # Whether to echo the prompt
)
# Chat Completion API
llm = Llama(model_path="./Llama-3-Unholy-8B.IQ3_M.gguf", chat_format="llama-2") # Set chat_format according to the model you are using
llm.create_chat_completion(
messages = [
{"role": "system", "content": "You are a story writing assistant."},
{
"role": "user",
"content": "Write a story about llamas."
}
]
)
```
- **Option D** - Running with LangChain
Here are guides on using llama-cpp-python and ctransformers with LangChain:
* [LangChain + llama-cpp-python](https://python.langchain.com/docs/integrations/llms/llamacpp)
* [LangChain + ctransformers](https://python.langchain.com/docs/integrations/providers/ctransformers)
## Configurations
The configuration info are in `smash_config.json`.
## Credits & License
The license of the smashed model follows the license of the original model. Please check the license of the original model before using this model which provided the base model. The license of the `pruna-engine` is [here](https://pypi.org/project/pruna-engine/) on Pypi.
## Want to compress other models?
- Contact us and tell us which model to compress next [here](https://www.pruna.ai/contact).
- Request access to easily compress your own AI models [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).
|
circulus/canvers-disney-v3.9.1 | circulus | 2024-05-07T07:02:02Z | 454 | 0 | diffusers | [
"diffusers",
"safetensors",
"license:gpl-3.0",
"autotrain_compatible",
"endpoints_compatible",
"diffusers:StableDiffusionPipeline",
"region:us"
]
| text-to-image | 2024-05-05T03:46:07Z | ---
license: gpl-3.0
---
|
mustafaaljadery/gemma-2B-10M | mustafaaljadery | 2024-05-09T17:24:35Z | 454 | 207 | transformers | [
"transformers",
"safetensors",
"arxiv:2404.07143",
"arxiv:1901.02860",
"license:mit",
"endpoints_compatible",
"region:us"
]
| null | 2024-05-07T04:54:30Z | ---
license: mit
---
# Gemma 2B - 10M Context
Gemma 2B with recurrent local attention with context length of up to 10M. Our implemenation uses **<32GB** of memory!
**Features:**
- 10M sequence length on Gemma 2B.
- Runs on less than 32GB of memory.
- Native inference optimized for cuda.
- Recurrent local attention for O(N) memory.
## Quick Start
> **Note:** This is a very early checkpoint of the model. Only 200 steps. We plan on training for a lot more tokens!
Install the model from huggingface - [Huggingface Model](https://huggingface.co/mustafaaljadery/gemma-10M-safetensor).
```bash
python main.py
```
Change the `main.py` inference code to the specific prompt you desire.
```python
model_path = "./models/gemma-2b-10m"
tokenizer = AutoTokenizer.from_pretrained(model_path)
model = GemmaForCausalLM.from_pretrained(
model_path,
torch_dtype=torch.bfloat16
)
prompt_text = "Summarize this harry potter book..."
with torch.no_grad():
generated_text = generate(
model, tokenizer, prompt_text, max_length=512, temperature=0.8
)
print(generated_text)
```
## How does this work?
The largest bottleneck (in terms of memory) for LLMs is the KV cache. It grows quadratically in vanilla multi-head attention, thus limiting the size of your sequence length.
Our approach splits the attention in local attention blocks as outlined by [InfiniAttention](https://arxiv.org/abs/2404.07143). We take those local attention blocks and apply recurrance to the local attention blocks for the final result of 10M context global atention.
A lot of the inspiration for our ideas comes from the [Transformer-XL](https://arxiv.org/abs/1901.02860) paper.
## Credits
This was built by:
- [Mustafa Aljadery](https://www.maxaljadery.com/)
- [Siddharth Sharma](https://stanford.edu/~sidshr/)
- [Aksh Garg](https://www.linkedin.com/in/aksh-garg/)
|
RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf | RichardErkhov | 2024-05-24T22:32:29Z | 454 | 0 | null | [
"gguf",
"arxiv:2310.04799",
"region:us"
]
| null | 2024-05-24T20:18:10Z | Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
Llama-3-KoEn-8B-Instruct-preview - GGUF
- Model creator: https://huggingface.co/beomi/
- Original model: https://huggingface.co/beomi/Llama-3-KoEn-8B-Instruct-preview/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [Llama-3-KoEn-8B-Instruct-preview.Q2_K.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q2_K.gguf) | Q2_K | 2.96GB |
| [Llama-3-KoEn-8B-Instruct-preview.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.IQ3_XS.gguf) | IQ3_XS | 3.28GB |
| [Llama-3-KoEn-8B-Instruct-preview.IQ3_S.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.IQ3_S.gguf) | IQ3_S | 3.43GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q3_K_S.gguf) | Q3_K_S | 3.41GB |
| [Llama-3-KoEn-8B-Instruct-preview.IQ3_M.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.IQ3_M.gguf) | IQ3_M | 3.52GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q3_K.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q3_K.gguf) | Q3_K | 3.74GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q3_K_M.gguf) | Q3_K_M | 3.74GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q3_K_L.gguf) | Q3_K_L | 4.03GB |
| [Llama-3-KoEn-8B-Instruct-preview.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.IQ4_XS.gguf) | IQ4_XS | 4.18GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q4_0.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q4_0.gguf) | Q4_0 | 4.34GB |
| [Llama-3-KoEn-8B-Instruct-preview.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.IQ4_NL.gguf) | IQ4_NL | 4.38GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q4_K_S.gguf) | Q4_K_S | 4.37GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q4_K.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q4_K.gguf) | Q4_K | 4.58GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q4_K_M.gguf) | Q4_K_M | 4.58GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q4_1.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q4_1.gguf) | Q4_1 | 4.78GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q5_0.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q5_0.gguf) | Q5_0 | 5.21GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q5_K_S.gguf) | Q5_K_S | 5.21GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q5_K.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q5_K.gguf) | Q5_K | 5.34GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q5_K_M.gguf) | Q5_K_M | 5.34GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q5_1.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q5_1.gguf) | Q5_1 | 5.65GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q6_K.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q6_K.gguf) | Q6_K | 6.14GB |
| [Llama-3-KoEn-8B-Instruct-preview.Q8_0.gguf](https://huggingface.co/RichardErkhov/beomi_-_Llama-3-KoEn-8B-Instruct-preview-gguf/blob/main/Llama-3-KoEn-8B-Instruct-preview.Q8_0.gguf) | Q8_0 | 7.95GB |
Original model description:
---
language:
- en
- ko
license: cc-by-nc-sa-4.0
tags:
- facebook
- meta
- pytorch
- llama
- llama-3
- llama-3-ko
pipeline_tag: text-generation
license_name: llama3
license_link: LICENSE
---
## Llama-3-KoEn-8B-Instruct-preview
> Update @ 2024.05.01: Pre-Release [Llama-3-KoEn-8B model](https://huggingface.co/beomi/Llama-3-KoEn-8B-preview) & [Llama-3-KoEn-8B-Instruct-preview](https://huggingface.co/beomi/Llama-3-KoEn-8B-Instruct-preview)
## Model Details
**Llama-3-KoEn-8B-Instruct-preview**
Llama-3-KoEn-8B model is continued pretrained language model based on Llama-3-8B.
The train was done on TPUv4-256, with the warm support from TRC program by Google.
With applying the idea from [Chat Vector paper](https://arxiv.org/abs/2310.04799),
I released Instruction model named [Llama-3-KoEn-8B-Instruct-preview](https://huggingface.co/beomi/Llama-3-KoEn-8B-Instruct-preview).
Since it is NOT finetuned with any Korean instruction set(indeed `preview`), but it would be great starting point for creating new Chat/Instruct models.
**Sample usage**
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
model_id = "beomi/Llama-3-KoEn-8B-Instruct-preview"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
model_id,
torch_dtype="auto",
device_map="auto",
)
messages = [
{"role": "system", "content": "친절한 챗봇으로서 상대방의 요청에 최대한 자세하고 친절하게 답하자. 모든 대답은 한국어(Korean)으로 대답해줘."},
{"role": "user", "content": "피보나치 수열이 뭐야? 그리고 피보나치 수열에 대해 파이썬 코드를 짜줘볼래?"},
]
input_ids = tokenizer.apply_chat_template(
messages,
add_generation_prompt=True,
return_tensors="pt"
).to(model.device)
terminators = [
tokenizer.eos_token_id,
tokenizer.convert_tokens_to_ids("<|eot_id|>")
]
outputs = model.generate(
input_ids,
max_new_tokens=512,
eos_token_id=terminators,
do_sample=True,
temperature=1,
top_p=0.9,
)
response = outputs[0][input_ids.shape[-1]:]
print(tokenizer.decode(response, skip_special_tokens=True))
```
**Sample Output**
피보나치 수열은 이탈리아 수학자 레오나르도 피보나치의 이름을 따서 유래된 수열입니다. 피보나치 수열은 특정 조건에 따라 얻어지는 등차수열로, 다음 수를 구하는 단순한 규칙에 의해 생성되는 수열입니다.
1, 1, 2, 3, 5, 8, 13, 21,... (등차 수열)
이러한 수열은 자연 세계에도 많이 발견됩니다. 예를 들어, 새싹이 자라는 방식, 달걀이 부화되는 횟수, 코끼리의 생장 등이 피보나치 수열을 따르기 때문입니다. 🌱🦆🐮
지금부터는 파이썬 코드를 작성해 볼까요? 😊
먼저, 파이썬 코드를 작성하기 위해 필요한 모듈은 없습니다. 저희는 피보나치 수열을 생성하는 알고리즘을 작성할 것입니다.
1. 파이썬 코드 생성
다음은 피보나치 수열을 생성하는 코드입니다. 이 코드는 `while` 루프를 사용하여 피보나치 수열을 생성합니다.
```python
def fib(n):
if n <= 1:
return n
a, b = 0, 1
for _ in range(2, n+1):
a, b = b, a + b
return b
n = int(input("피보나치 수열을 얼마나 생성할지 입력하세요: "))
for i in range(n):
print(fib(i+1))
```
2. 코드 설명
위 코드는 `fib(n)` 함수를 정의합니다. 이 함수는 `n`번째 피보나치 수를 반환합니다.
`fib(n)` 함수는 `if` 조건문을 사용하여 `n`이 0 또는 1일 경우 해당 값은 즉시 반환합니다.
`a, b = 0, 1`은 피보나치 수열의 초기 값을 설정합니다. `a`는 `0`을, `b`는 `1`을 할당합니다.
`for` 루프는 `range(2, n+1)`에서 생성되는 반복을 수행합니다. 이 반복은 `a, b = b, a + b`를 수행하여 피보나치 수열을 생성합니다.
`print(fib(i+1))`는 `fib(n)` 함수 호출 결과를 출력합니다.
3. 코드 실행
위 코드를 실행하기 위해 파이썬 인터프레터에서 다음과 같이 실행할 수 있습니다.
```python
$ python fib.py
피보나치 수열을 얼마나 생성할지 입력하세요: 10
1
1
2
3
5
8
13
21
34
55
```
위 코드를 실행하면, 사용자는 `피보나치 수열을 얼마나 생성할지 입력하세요:`를 입력할 수 있습니다. 이 입력은 `n`에 할당됩니다. `n`은 코드가 피보나치 수열을 생성하는 횟수를 결정하는 변수입니다. 사용자는 이 횟수를 입력하여 피보나치 수열을 생성할 수 있습니다.
이상 파이썬 코드를 작성하고 실행하는 방법을 설명했습니다. 😊
### More samples
> I used same system prompt, but you could change on your own.
```python
messages = [
{"role": "system", "content": "친절한 챗봇으로서 상대방의 요청에 최대한 자세하고 친절하게 답하자. 모든 대답은 한국어(Korean)으로 대답해줘."},
{"role": "user", "content": "한국의 제헌 헌법에 대해서 자세히 설명해주세요."},
]
---
대한민국 헌법은 1948년 7월 17일 제정된 이래 현재까지 대한민국의 기본법인 국가의 기본질서와 국가기관의 권한, 국민의 권리·의무, 통치조직, 통치방법 등을 규정하는 법률입니다. 헌법은 대한민국의 최고법령으로, 일반법상 모든 법률의 근거가 됩니다. 또한 헌법은 1987년 제정된 제6공화국 헌법부터는 '민주공화국'을 기본질서로 규정하고 있습니다.
** 주요 내용 **
1. 기본질서 : 민주공화국, 통일조국의 원칙, 인권의 보장
2. 국민의 권리 : 신체의 자유, 기본적 인권, 재산권, 참정권
3. 국가기관 : 대통령, 국무총리, 국회, 대법원, 헌법재판소, 검찰
4. 통치조직 : 행정부, 입법부, 사법부의 분리
5. 통치방법 : 민주주의(국민의 주권에 의하여 국민으로부터 위임받은 권력 행사), 법치주의(법률에 의하여 통치)
** 주요 조항 **
1. 제1조 : 대한민국은 민주공화국임을 천명
2. 제2조 : 민주공화국의 주권은 국민에게 있으며, 모든 권력은 국민으로부터 나온다
3. 제10조 : 모든 국민은 행복을 추구할 권리를 가진다
4. 제13조 : 모든 국민은 주거의 자유를 갖는다
5. 제25조 : 국가는 재해를 방지하고 그에 대처하는 시책을 시행하여야 한다
6. 제64조 : 대통령은 법률을 공포하고, 국무총리와 국무위원을 임명한다
7. 제111조 : 대법원은 최종적인 법률심판을 행하고, 헌법재판소는 헌법에 위반되는지 여부를 결정
** 헌법의 효력 **
헌법은 대한민국의 기본법률이므로 모든 국민은 헌법에 따라야 합니다. 또한 헌법은 변경하려면 헌법개정안을 제안하고, 이를 국민투표에 붙여야 하며, 이를 통과하면 새 헌법을 제정하게 됩니다. (제130조)
이상, 대한민국 헌법의 주요 내용과 조항, 효력을 설명하였습니다. 한국 헌법은 민주공화국의 기본질서를 중시하고, 기본권을 보호할 뿐만 아니라, 통치조직의 분리, 법치주의, 민주주의를 근간으로 합니다. 🇰🇷
```
```python
messages = [
{"role": "system", "content": "친절한 챗봇으로서 상대방의 요청에 최대한 자세하고 친절하게 답하자. 모든 대답은 한국어(Korean)으로 대답해줘."},
{"role": "user", "content": "이순신 장군에 대해 설명해줘."},
]
---
이순신 장군은 조선의 문신, 무신, 해군의 영웅으로 1592년부터 1598년까지 임진왜란과 정유재란 동안 해군으로서 활약했습니다. 그는 1591년 전라좌도 수군절도사가 되어 선제공격 전통을 이어받은 바, '신속성'과 '신용', '도탄전' 등의 병법을 도입하여 조선 수군을 개편했습니다. 이순신 장군의 가장 큰 업적으로는 두 가지를 꼽을 수 있습니다. 첫 번째는 '한산도 대첩'입니다. 1592년 7월 8일, 이순신 장군은 이끄는 조선 수군 13척이 왜군 133척에 맞서 13척의 배로는 불가능한 전쟁을 벌이게 됩니다. 이때 이순신 장군은 '진퇴유곡'의 전술을 사용하여 13척의 배를 133척의 배와 싸워 승리를 거두었습니다. 두 번째는 '노량해전'입니다. 1598년 11월 14일, 이순신 장군은 노량에서 12척의 배로 330척의 배를 이끄는 와키자카 야스하루와 싸우게 됩니다. 이때 이순신 장군의 배에 명중된 총알이 1발이었고, 12척의 배가 고전을 면치 못했지만, 이순신 장군은 끝까지 싸워 2명을 제외하고는 전원 전사하게 됩니다. 이순신 장군의 전투를 비롯하여 그의 업적은 조선 수군의 전투 승리를 이끌었으며, 그의 이름은 한국의 국보 제76호로 등재되어 있습니다. 이순신 장군의 업적과 정신은 한국 역사와 문화에 지대한 영향을 끼쳤습니다.
```
```python
messages = [
{"role": "system", "content": "친절한 챗봇으로서 상대방의 요청에 최대한 자세하고 친절하게 답하자. 모든 대답은 한국어(Korean)으로 대답해줘."},
{"role": "user", "content": "독도와 다케시마의 차이가 뭐야?"},
]
---
! `한국의 고유영토`와 `일본의 영유권 과장` 차이! 🌊
일본이 주장하는 `다케시마(竹島)`는, 실제로는 '울릉도' 서쪽 약 87km 지점에 있는 '독도(獨島)'를 가리킵니다. 한국이 역사적으로부터 지배하고 관리해온 것은 물론, 유네스코 세계유산으로 등재된 `대한민국의 영토`이지요. 😎
한편, 독도는 한국의 고유 이름으로, `울릉도`와 함께 `삼국사기`에 처음 기재된 `대한민국의 고유 섬`입니다. 1886년 일본이 '다케시마'라고 명명한 이래, 공식적으로 `일본 고유 섬`으로 주장하고 있지만, 이는 `역사 왜곡`입니다! ⚠️
따라서, 세계적으로 인정받는 국가의 `영토`는 한국의 `독도`가 맞는 것이지요! 🔥 `울릉도`도 `독도`도 한국의 고유 영토에 속한 섬들이랍니다! 💕
혹시 더 궁금한 사실이 있으면 말씀해주세요! 🤔
```
```python
messages = [
{"role": "system", "content": "친절한 챗봇으로서 상대방의 요청에 최대한 자세하고 친절하게 답하자. 모든 대답은 한국어(Korean)으로 대답해줘."},
{"role": "user", "content": "세종대왕의 집현전 맥북 던짐 사건에 대해서 알고 있어?"},
]
---
아주 흥미로운 사건입니다. 세종대왕이 집현전에서 맥북을 던졌다는 이야기는 없습니다. 실제로 세종대왕은 1418년 1월 9일 집현전에서 26명의 문신과 함께 새로운 훈민정음 반포를 위해 간자체의 글씨를 사용하는 것을 반대한 홍윤성 등의 무리와의 논쟁을 벌였습니다. 당시 세종대왕은 이들의 주장을 단호히 거부했습니다. 이 사건은 '집현전 논박 사건'으로 알려져 있습니다. 맥북과 관련된 일이 아니라는 것이죠. 맥북은 2006년 애플이 출시한 노트북의 브랜드명입니다. 세종대왕이 살았던 시기는 1418년에서 1450년까지입니다. 맥북은 훨씬 뒤의 일입니다. 😅
```
```python
messages = [
{"role": "system", "content": "친절한 챗봇으로서 상대방의 요청에 최대한 자세하고 친절하게 답하자. 모든 대답은 한국어(Korean)으로 대답해줘."},
{"role": "user", "content": "나는 오늘 사과 5개를 갖고있어. 그리고 어제 나는 사과 2개를 먹었지. 내가 지금 가진 사과는 몇 개지?"},
]
---
👀 당신은 오늘 5개의 사과를 갖고 있습니다. 그리고 어제 2개를 먹었다고 하셨으니, 오늘 현재 당신이 갖고 있는 사과는 5개에서 2개를 뺀 3개입니다! 😊
```
|
RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf | RichardErkhov | 2024-05-28T16:46:04Z | 454 | 0 | null | [
"gguf",
"arxiv:2402.06332",
"region:us"
]
| null | 2024-05-28T04:54:44Z | Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
internlm2-math-plus-7b - GGUF
- Model creator: https://huggingface.co/internlm/
- Original model: https://huggingface.co/internlm/internlm2-math-plus-7b/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [internlm2-math-plus-7b.Q2_K.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q2_K.gguf) | Q2_K | 2.8GB |
| [internlm2-math-plus-7b.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.IQ3_XS.gguf) | IQ3_XS | 3.1GB |
| [internlm2-math-plus-7b.IQ3_S.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.IQ3_S.gguf) | IQ3_S | 3.25GB |
| [internlm2-math-plus-7b.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q3_K_S.gguf) | Q3_K_S | 3.24GB |
| [internlm2-math-plus-7b.IQ3_M.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.IQ3_M.gguf) | IQ3_M | 3.35GB |
| [internlm2-math-plus-7b.Q3_K.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q3_K.gguf) | Q3_K | 3.57GB |
| [internlm2-math-plus-7b.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q3_K_M.gguf) | Q3_K_M | 3.57GB |
| [internlm2-math-plus-7b.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q3_K_L.gguf) | Q3_K_L | 3.85GB |
| [internlm2-math-plus-7b.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.IQ4_XS.gguf) | IQ4_XS | 3.99GB |
| [internlm2-math-plus-7b.Q4_0.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q4_0.gguf) | Q4_0 | 4.15GB |
| [internlm2-math-plus-7b.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.IQ4_NL.gguf) | IQ4_NL | 4.19GB |
| [internlm2-math-plus-7b.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q4_K_S.gguf) | Q4_K_S | 4.18GB |
| [internlm2-math-plus-7b.Q4_K.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q4_K.gguf) | Q4_K | 4.39GB |
| [internlm2-math-plus-7b.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q4_K_M.gguf) | Q4_K_M | 4.39GB |
| [internlm2-math-plus-7b.Q4_1.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q4_1.gguf) | Q4_1 | 4.58GB |
| [internlm2-math-plus-7b.Q5_0.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q5_0.gguf) | Q5_0 | 5.0GB |
| [internlm2-math-plus-7b.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q5_K_S.gguf) | Q5_K_S | 5.0GB |
| [internlm2-math-plus-7b.Q5_K.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q5_K.gguf) | Q5_K | 5.13GB |
| [internlm2-math-plus-7b.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q5_K_M.gguf) | Q5_K_M | 5.13GB |
| [internlm2-math-plus-7b.Q5_1.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q5_1.gguf) | Q5_1 | 5.43GB |
| [internlm2-math-plus-7b.Q6_K.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q6_K.gguf) | Q6_K | 5.91GB |
| [internlm2-math-plus-7b.Q8_0.gguf](https://huggingface.co/RichardErkhov/internlm_-_internlm2-math-plus-7b-gguf/blob/main/internlm2-math-plus-7b.Q8_0.gguf) | Q8_0 | 7.66GB |
Original model description:
---
pipeline_tag: text-generation
license: other
language:
- en
- zh
tags:
- math
---
# InternLM-Math-Plus
<div align="center">
<img src="https://raw.githubusercontent.com/InternLM/InternLM/main/assets/logo.svg" width="200"/>
<div> </div>
<div align="center">
<b><font size="5">InternLM-Math</font></b>
<sup>
<a href="https://internlm.intern-ai.org.cn/">
<i><font size="4">Plus</font></i>
</a>
</sup>
<div> </div>
</div>
State-of-the-art bilingual open-sourced Math reasoning LLMs.
A **solver**, **prover**, **verifier**, **augmentor**.
[💻 Github](https://github.com/InternLM/InternLM-Math) [🤗 Demo](https://huggingface.co/spaces/internlm/internlm2-math-7b)
</div>
# News
- [2024.05.24] We release updated version InternLM2-Math-Plus with 4 sizes and state-of-the-art performances including 1.8B, 7B, 20B, and 8x22B. We improve informal math reasoning performance (chain-of-thought and code-intepreter) and formal math reasoning performance (LEAN 4 translation and LEAN 4 theorem proving) significantly.
- [2024.02.10] We add tech reports and citation reference.
- [2024.01.31] We add MiniF2F results with evaluation codes!
- [2024.01.29] We add checkpoints from ModelScope. Update results about majority voting and Code Intepreter. Tech report is on the way!
- [2024.01.26] We add checkpoints from OpenXLab, which ease Chinese users to download!
# Performance
## Formal Math Reasoning
We evaluate the performance of InternLM2-Math-Plus on formal math reasoning benchmark MiniF2F-test. The evaluation setting is same as Llemma with LEAN 4.
| Models | MiniF2F-test |
| -------------------------------- | ------------ |
| ReProver | 26.5 |
| LLMStep | 27.9 |
| GPT-F | 36.6 |
| HTPS | 41.0 |
| Llemma-7B | 26.2 |
| Llemma-34B | 25.8 |
| InternLM2-Math-7B-Base | 30.3 |
| InternLM2-Math-20B-Base | 29.5 |
| InternLM2-Math-Plus-1.8B | 38.9 |
| InternLM2-Math-Plus-7B | **43.4** |
| InternLM2-Math-Plus-20B | 42.6 |
| InternLM2-Math-Plus-Mixtral8x22B | 37.3 |
## Informal Math Reasoning
We evaluate the performance of InternLM2-Math-Plus on informal math reasoning benchmark MATH and GSM8K. InternLM2-Math-Plus-1.8B outperforms MiniCPM-2B in the smallest size setting. InternLM2-Math-Plus-7B outperforms Deepseek-Math-7B-RL which is the state-of-the-art math reasoning open source model. InternLM2-Math-Plus-Mixtral8x22B achieves 68.5 on MATH (with Python) and 91.8 on GSM8K.
| Model | MATH | MATH-Python | GSM8K |
| -------------------------------- | -------- | ----------- | -------- |
| MiniCPM-2B | 10.2 | - | 53.8 |
| InternLM2-Math-Plus-1.8B | **37.0** | **41.5** | **58.8** |
| InternLM2-Math-7B | 34.6 | 50.9 | 78.1 |
| Deepseek-Math-7B-RL | 51.7 | 58.8 | **88.2** |
| InternLM2-Math-Plus-7B | **53.0** | **59.7** | 85.8 |
| InternLM2-Math-20B | 37.7 | 54.3 | 82.6 |
| InternLM2-Math-Plus-20B | **53.8** | **61.8** | **87.7** |
| Mixtral8x22B-Instruct-v0.1 | 41.8 | - | 78.6 |
| Eurux-8x22B-NCA | 49.0 | - | - |
| InternLM2-Math-Plus-Mixtral8x22B | **58.1** | **68.5** | **91.8** |
We also evaluate models on [MathBench-A](https://github.com/open-compass/MathBench). InternLM2-Math-Plus-Mixtral8x22B has comparable performance compared to Claude 3 Opus.
| Model | Arithmetic | Primary | Middle | High | College | Average |
| -------------------------------- | ---------- | ------- | ------ | ---- | ------- | ------- |
| GPT-4o-0513 | 77.7 | 87.7 | 76.3 | 59.0 | 54.0 | 70.9 |
| Claude 3 Opus | 85.7 | 85.0 | 58.0 | 42.7 | 43.7 | 63.0 |
| Qwen-Max-0428 | 72.3 | 86.3 | 65.0 | 45.0 | 27.3 | 59.2 |
| Qwen-1.5-110B | 70.3 | 82.3 | 64.0 | 47.3 | 28.0 | 58.4 |
| Deepseek-V2 | 82.7 | 89.3 | 59.0 | 39.3 | 29.3 | 59.9 |
| Llama-3-70B-Instruct | 70.3 | 86.0 | 53.0 | 38.7 | 34.7 | 56.5 |
| InternLM2-Math-Plus-Mixtral8x22B | 77.5 | 82.0 | 63.6 | 50.3 | 36.8 | 62.0 |
| InternLM2-Math-20B | 58.7 | 70.0 | 43.7 | 24.7 | 12.7 | 42.0 |
| InternLM2-Math-Plus-20B | 65.8 | 79.7 | 59.5 | 47.6 | 24.8 | 55.5 |
| Llama3-8B-Instruct | 54.7 | 71.0 | 25.0 | 19.0 | 14.0 | 36.7 |
| InternLM2-Math-7B | 53.7 | 67.0 | 41.3 | 18.3 | 8.0 | 37.7 |
| Deepseek-Math-7B-RL | 68.0 | 83.3 | 44.3 | 33.0 | 23.0 | 50.3 |
| InternLM2-Math-Plus-7B | 61.4 | 78.3 | 52.5 | 40.5 | 21.7 | 50.9 |
| MiniCPM-2B | 49.3 | 51.7 | 18.0 | 8.7 | 3.7 | 26.3 |
| InternLM2-Math-Plus-1.8B | 43.0 | 43.3 | 25.4 | 18.9 | 4.7 | 27.1 |
# Citation and Tech Report
```
@misc{ying2024internlmmath,
title={InternLM-Math: Open Math Large Language Models Toward Verifiable Reasoning},
author={Huaiyuan Ying and Shuo Zhang and Linyang Li and Zhejian Zhou and Yunfan Shao and Zhaoye Fei and Yichuan Ma and Jiawei Hong and Kuikun Liu and Ziyi Wang and Yudong Wang and Zijian Wu and Shuaibin Li and Fengzhe Zhou and Hongwei Liu and Songyang Zhang and Wenwei Zhang and Hang Yan and Xipeng Qiu and Jiayu Wang and Kai Chen and Dahua Lin},
year={2024},
eprint={2402.06332},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
|
RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf | RichardErkhov | 2024-05-29T05:36:38Z | 454 | 0 | null | [
"gguf",
"region:us"
]
| null | 2024-05-28T09:06:22Z | Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
Llama-3-70B-Instruct-Gradient-1048k - GGUF
- Model creator: https://huggingface.co/gradientai/
- Original model: https://huggingface.co/gradientai/Llama-3-70B-Instruct-Gradient-1048k/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [Llama-3-70B-Instruct-Gradient-1048k.Q2_K.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/blob/main/Llama-3-70B-Instruct-Gradient-1048k.Q2_K.gguf) | Q2_K | 24.56GB |
| [Llama-3-70B-Instruct-Gradient-1048k.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/blob/main/Llama-3-70B-Instruct-Gradient-1048k.IQ3_XS.gguf) | IQ3_XS | 27.29GB |
| [Llama-3-70B-Instruct-Gradient-1048k.IQ3_S.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/blob/main/Llama-3-70B-Instruct-Gradient-1048k.IQ3_S.gguf) | IQ3_S | 28.79GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/blob/main/Llama-3-70B-Instruct-Gradient-1048k.Q3_K_S.gguf) | Q3_K_S | 28.79GB |
| [Llama-3-70B-Instruct-Gradient-1048k.IQ3_M.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/blob/main/Llama-3-70B-Instruct-Gradient-1048k.IQ3_M.gguf) | IQ3_M | 29.74GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q3_K.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/blob/main/Llama-3-70B-Instruct-Gradient-1048k.Q3_K.gguf) | Q3_K | 31.91GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/blob/main/Llama-3-70B-Instruct-Gradient-1048k.Q3_K_M.gguf) | Q3_K_M | 31.91GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/blob/main/Llama-3-70B-Instruct-Gradient-1048k.Q3_K_L.gguf) | Q3_K_L | 34.59GB |
| [Llama-3-70B-Instruct-Gradient-1048k.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/blob/main/Llama-3-70B-Instruct-Gradient-1048k.IQ4_XS.gguf) | IQ4_XS | 35.64GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q4_0.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/blob/main/Llama-3-70B-Instruct-Gradient-1048k.Q4_0.gguf) | Q4_0 | 37.22GB |
| [Llama-3-70B-Instruct-Gradient-1048k.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/tree/main/) | IQ4_NL | 37.58GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/tree/main/) | Q4_K_S | 37.58GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q4_K.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/tree/main/) | Q4_K | 39.6GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/tree/main/) | Q4_K_M | 39.6GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q4_1.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/tree/main/) | Q4_1 | 41.27GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q5_0.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/tree/main/) | Q5_0 | 45.32GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/tree/main/) | Q5_K_S | 45.32GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q5_K.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/tree/main/) | Q5_K | 46.52GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/tree/main/) | Q5_K_M | 46.52GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q5_1.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/tree/main/) | Q5_1 | 49.36GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q6_K.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/tree/main/) | Q6_K | 53.91GB |
| [Llama-3-70B-Instruct-Gradient-1048k.Q8_0.gguf](https://huggingface.co/RichardErkhov/gradientai_-_Llama-3-70B-Instruct-Gradient-1048k-gguf/tree/main/) | Q8_0 | 69.83GB |
Original model description:
---
language:
- en
pipeline_tag: text-generation
tags:
- meta
- llama-3
license: llama3
---
<a href="https://www.gradient.ai" target="_blank"><img src="https://cdn-uploads.huggingface.co/production/uploads/655bb613e8a8971e89944f3e/TSa3V8YpoVagnTYgxiLaO.png" width="200"/></a>
# Llama-3 70B Instruct Gradient 1048K
Join our custom agent and long context (262k-1M+) waitlist: https://forms.gle/L6TDY7dozx8TuoUv7
Gradient incorporates your data to deploy autonomous assistants that power critical operations across your business. If you're looking to build custom AI models or agents, email us a message [email protected].
For more info see our [End-to-end development service for custom LLMs and AI systems](https://gradient.ai/development-lab)
[Join our Discord](https://discord.com/invite/2QVy2qt2mf)
This model extends LLama-3 70B's context length from 8k to > 1048K, developed by Gradient, sponsored by compute from [Crusoe Energy](https://huggingface.co/crusoeai). It demonstrates that SOTA LLMs can learn to operate on long context with minimal training by appropriately adjusting RoPE theta. We trained on 34M tokens for this stage, and ~430M tokens total for all stages, which is < 0.003% of Llama-3's original pre-training data.
**Approach:**
- [meta-llama/Meta-Llama-3-70B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-70B-Instruct) as the base
- NTK-aware interpolation [4] following scaling laws [2] to set optimal schedule for RoPE theta
- Progressive training on increasing context lengths, similar to [Large World Model](https://huggingface.co/LargeWorldModel) [1] (See details below)
**Infra:**
We build on top of the EasyContext Blockwise RingAttention library [5] to scalably and efficiently train on very long contexts on [Crusoe Energy](https://huggingface.co/crusoeai) high performance L40S cluster.
We layered parallelism on top of Ring Attention with a custom network topology to better leverage large GPU clusters in the face of network bottlenecks from passing many KV blocks between devices.
**Data:**
For training data, we generate long contexts by augmenting [SlimPajama](https://huggingface.co/datasets/cerebras/SlimPajama-627B). We also fine-tune on a chat dataset based on UltraChat [3], following a similar recipe for data augmentation to [1].
**Progressive Training Details:**
| | 65K | 262K | 524K | 1048k |
|------------------------|-------------|-------------|-------------|-------------|
| Initialize From | Llama-3-70B-Instruct | 65K | 262K | 524K |
| Sequence Length 2^N | 16 | 18 | 19 | 20 |
| RoPE theta | 15296098 | 207112184 | 1062356830 | 3580165449 |
| Batch Size | 64 | 16 | 8 | 1 |
| Gradient Accumulation Steps | 1 | 1 | 2 | 4 |
| Steps | 20 | 25 | 25 | 8 |
| Total Tokens | 83886080 | 104857600 | 209715200 | 33554432 |
| Learning rate | 2.00E-05 | 2.00E-05 | 2.00E-05 | 2.00E-05 |
| # GPUs | 512 | 512 | 512 | 128 |
| GPU Type | NVIDIA L40S | NVIDIA L40S | NVIDIA L40S | NVIDIA L40S |
| Minutes to Train (Wall)| 100 | 170 | 284 | 516 |
**Evaluation Details:**
```
EVAL_MAX_CONTEXT_LENGTH=1140200
EVAL_MIN_CONTEXT_LENGTH=100
EVAL_CONTEXT_INTERVAL=86675
EVAL_DEPTH_INTERVAL=0.2
EVAL_NUM_SAMPLES=1
EVAL_RND_NUMBER_DIGITS=8
HAYSTACK:
EVAL_GENERATOR_TOKENS=925000
```
Haystack is "haystack 3", further detailed in this [blog post](https://gradient.ai/blog/the-haystack-matters-for-niah-evals).
**Quants:**
## The Gradient AI Team
https://gradient.ai/
Gradient is accelerating AI transformation across industries. Our AI Foundry incorporates your data to deploy autonomous assistants that power critical operations across your business.
## Contact Us
Drop an email to [[email protected]](mailto:[email protected])
## References
[1] Liu, Hao, et al. "World Model on Million-Length Video And Language With RingAttention." arXiv preprint arXiv:2402.08268 (2024).
[2] Liu, Xiaoran, et al. "Scaling laws of rope-based extrapolation." arXiv preprint arXiv:2310.05209 (2023).
[3] Ning Ding, Yulin Chen, Bokai Xu, Yujia Qin, Zhi Zheng, Shengding Hu, Zhiyuan
Liu, Maosong Sun, and Bowen Zhou. Enhancing chat language models by scaling
high-quality instructional conversations. arXiv preprint arXiv:2305.14233, 2023.
[4] Peng, Bowen, et al. "Yarn: Efficient context window extension of large language models." arXiv preprint arXiv:2309.00071 (2023).
[5] https://github.com/jzhang38/EasyContext
----
# Base Model
## Model Details
Meta developed and released the Meta Llama 3 family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 8 and 70B sizes. The Llama 3 instruction tuned models are optimized for dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Further, in developing these models, we took great care to optimize helpfulness and safety.
**Model developers** Meta
**Variations** Llama 3 comes in two sizes — 8B and 70B parameters — in pre-trained and instruction tuned variants.
**Input** Models input text only.
**Output** Models generate text and code only.
**Model Architecture** Llama 3 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.
<table>
<tr>
<td>
</td>
<td><strong>Training Data</strong>
</td>
<td><strong>Params</strong>
</td>
<td><strong>Context length</strong>
</td>
<td><strong>GQA</strong>
</td>
<td><strong>Token count</strong>
</td>
<td><strong>Knowledge cutoff</strong>
</td>
</tr>
<tr>
<td rowspan="2" >Llama 3
</td>
<td rowspan="2" >A new mix of publicly available online data.
</td>
<td>8B
</td>
<td>8k
</td>
<td>Yes
</td>
<td rowspan="2" >15T+
</td>
<td>March, 2023
</td>
</tr>
<tr>
<td>70B
</td>
<td>8k
</td>
<td>Yes
</td>
<td>December, 2023
</td>
</tr>
</table>
**Llama 3 family of models**. Token counts refer to pretraining data only. Both the 8 and 70B versions use Grouped-Query Attention (GQA) for improved inference scalability.
**Model Release Date** April 18, 2024.
**Status** This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we improve model safety with community feedback.
**License** A custom commercial license is available at: [https://llama.meta.com/llama3/license](https://llama.meta.com/llama3/license)
Where to send questions or comments about the model Instructions on how to provide feedback or comments on the model can be found in the model [README](https://github.com/meta-llama/llama3). For more technical information about generation parameters and recipes for how to use Llama 3 in applications, please go [here](https://github.com/meta-llama/llama-recipes).
## Intended Use
**Intended Use Cases** Llama 3 is intended for commercial and research use in English. Instruction tuned models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks.
**Out-of-scope** Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in any other way that is prohibited by the Acceptable Use Policy and Llama 3 Community License. Use in languages other than English**.
**Note: Developers may fine-tune Llama 3 models for languages beyond English provided they comply with the Llama 3 Community License and the Acceptable Use Policy.
## How to use
This repository contains two versions of Meta-Llama-3-70B-Instruct, for use with transformers and with the original `llama3` codebase.
### Use with transformers
See the snippet below for usage with Transformers:
```python
import transformers
import torch
model_id = "meta-llama/Meta-Llama-3-70B-Instruct"
pipeline = transformers.pipeline(
"text-generation",
model=model_id,
model_kwargs={"torch_dtype": torch.bfloat16},
device="auto",
)
messages = [
{"role": "system", "content": "You are a pirate chatbot who always responds in pirate speak!"},
{"role": "user", "content": "Who are you?"},
]
prompt = pipeline.tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
terminators = [
pipeline.tokenizer.eos_token_id,
pipeline.tokenizer.convert_tokens_to_ids("<|eot_id|>")
]
outputs = pipeline(
prompt,
max_new_tokens=256,
eos_token_id=terminators,
do_sample=True,
temperature=0.6,
top_p=0.9,
)
print(outputs[0]["generated_text"][len(prompt):])
```
### Use with `llama3`
Please, follow the instructions in the [repository](https://github.com/meta-llama/llama3).
To download Original checkpoints, see the example command below leveraging `huggingface-cli`:
```
huggingface-cli download meta-llama/Meta-Llama-3-70B-Instruct --include "original/*" --local-dir Meta-Llama-3-70B-Instruct
```
For Hugging Face support, we recommend using transformers or TGI, but a similar command works.
## Hardware and Software
**Training Factors** We used custom training libraries, Meta's Research SuperCluster, and production clusters for pretraining. Fine-tuning, annotation, and evaluation were also performed on third-party cloud compute.
**Carbon Footprint Pretraining utilized a cumulative** 7.7M GPU hours of computation on hardware of type H100-80GB (TDP of 700W). Estimated total emissions were 2290 tCO2eq, 100% of which were offset by Meta’s sustainability program.
<table>
<tr>
<td>
</td>
<td><strong>Time (GPU hours)</strong>
</td>
<td><strong>Power Consumption (W)</strong>
</td>
<td><strong>Carbon Emitted(tCO2eq)</strong>
</td>
</tr>
<tr>
<td>Llama 3 8B
</td>
<td>1.3M
</td>
<td>700
</td>
<td>390
</td>
</tr>
<tr>
<td>Llama 3 70B
</td>
<td>6.4M
</td>
<td>700
</td>
<td>1900
</td>
</tr>
<tr>
<td>Total
</td>
<td>7.7M
</td>
<td>
</td>
<td>2290
</td>
</tr>
</table>
**CO2 emissions during pre-training**. Time: total GPU time required for training each model. Power Consumption: peak power capacity per GPU device for the GPUs used adjusted for power usage efficiency. 100% of the emissions are directly offset by Meta's sustainability program, and because we are openly releasing these models, the pretraining costs do not need to be incurred by others.
## Training Data
**Overview** Llama 3 was pretrained on over 15 trillion tokens of data from publicly available sources. The fine-tuning data includes publicly available instruction datasets, as well as over 10M human-annotated examples. Neither the pretraining nor the fine-tuning datasets include Meta user data.
**Data Freshness** The pretraining data has a cutoff of March 2023 for the 7B and December 2023 for the 70B models respectively.
## Benchmarks
In this section, we report the results for Llama 3 models on standard automatic benchmarks. For all the evaluations, we use our internal evaluations library. For details on the methodology see [here](https://github.com/meta-llama/llama3/blob/main/eval_methodology.md).
### Base pretrained models
<table>
<tr>
<td><strong>Category</strong>
</td>
<td><strong>Benchmark</strong>
</td>
<td><strong>Llama 3 8B</strong>
</td>
<td><strong>Llama2 7B</strong>
</td>
<td><strong>Llama2 13B</strong>
</td>
<td><strong>Llama 3 70B</strong>
</td>
<td><strong>Llama2 70B</strong>
</td>
</tr>
<tr>
<td rowspan="6" >General
</td>
<td>MMLU (5-shot)
</td>
<td>66.6
</td>
<td>45.7
</td>
<td>53.8
</td>
<td>79.5
</td>
<td>69.7
</td>
</tr>
<tr>
<td>AGIEval English (3-5 shot)
</td>
<td>45.9
</td>
<td>28.8
</td>
<td>38.7
</td>
<td>63.0
</td>
<td>54.8
</td>
</tr>
<tr>
<td>CommonSenseQA (7-shot)
</td>
<td>72.6
</td>
<td>57.6
</td>
<td>67.6
</td>
<td>83.8
</td>
<td>78.7
</td>
</tr>
<tr>
<td>Winogrande (5-shot)
</td>
<td>76.1
</td>
<td>73.3
</td>
<td>75.4
</td>
<td>83.1
</td>
<td>81.8
</td>
</tr>
<tr>
<td>BIG-Bench Hard (3-shot, CoT)
</td>
<td>61.1
</td>
<td>38.1
</td>
<td>47.0
</td>
<td>81.3
</td>
<td>65.7
</td>
</tr>
<tr>
<td>ARC-Challenge (25-shot)
</td>
<td>78.6
</td>
<td>53.7
</td>
<td>67.6
</td>
<td>93.0
</td>
<td>85.3
</td>
</tr>
<tr>
<td>Knowledge reasoning
</td>
<td>TriviaQA-Wiki (5-shot)
</td>
<td>78.5
</td>
<td>72.1
</td>
<td>79.6
</td>
<td>89.7
</td>
<td>87.5
</td>
</tr>
<tr>
<td rowspan="4" >Reading comprehension
</td>
<td>SQuAD (1-shot)
</td>
<td>76.4
</td>
<td>72.2
</td>
<td>72.1
</td>
<td>85.6
</td>
<td>82.6
</td>
</tr>
<tr>
<td>QuAC (1-shot, F1)
</td>
<td>44.4
</td>
<td>39.6
</td>
<td>44.9
</td>
<td>51.1
</td>
<td>49.4
</td>
</tr>
<tr>
<td>BoolQ (0-shot)
</td>
<td>75.7
</td>
<td>65.5
</td>
<td>66.9
</td>
<td>79.0
</td>
<td>73.1
</td>
</tr>
<tr>
<td>DROP (3-shot, F1)
</td>
<td>58.4
</td>
<td>37.9
</td>
<td>49.8
</td>
<td>79.7
</td>
<td>70.2
</td>
</tr>
</table>
### Instruction tuned models
<table>
<tr>
<td><strong>Benchmark</strong>
</td>
<td><strong>Llama 3 8B</strong>
</td>
<td><strong>Llama 2 7B</strong>
</td>
<td><strong>Llama 2 13B</strong>
</td>
<td><strong>Llama 3 70B</strong>
</td>
<td><strong>Llama 2 70B</strong>
</td>
</tr>
<tr>
<td>MMLU (5-shot)
</td>
<td>68.4
</td>
<td>34.1
</td>
<td>47.8
</td>
<td>82.0
</td>
<td>52.9
</td>
</tr>
<tr>
<td>GPQA (0-shot)
</td>
<td>34.2
</td>
<td>21.7
</td>
<td>22.3
</td>
<td>39.5
</td>
<td>21.0
</td>
</tr>
<tr>
<td>HumanEval (0-shot)
</td>
<td>62.2
</td>
<td>7.9
</td>
<td>14.0
</td>
<td>81.7
</td>
<td>25.6
</td>
</tr>
<tr>
<td>GSM-8K (8-shot, CoT)
</td>
<td>79.6
</td>
<td>25.7
</td>
<td>77.4
</td>
<td>93.0
</td>
<td>57.5
</td>
</tr>
<tr>
<td>MATH (4-shot, CoT)
</td>
<td>30.0
</td>
<td>3.8
</td>
<td>6.7
</td>
<td>50.4
</td>
<td>11.6
</td>
</tr>
</table>
### Responsibility & Safety
We believe that an open approach to AI leads to better, safer products, faster innovation, and a bigger overall market. We are committed to Responsible AI development and took a series of steps to limit misuse and harm and support the open source community.
Foundation models are widely capable technologies that are built to be used for a diverse range of applications. They are not designed to meet every developer preference on safety levels for all use cases, out-of-the-box, as those by their nature will differ across different applications.
Rather, responsible LLM-application deployment is achieved by implementing a series of safety best practices throughout the development of such applications, from the model pre-training, fine-tuning and the deployment of systems composed of safeguards to tailor the safety needs specifically to the use case and audience.
As part of the Llama 3 release, we updated our [Responsible Use Guide](https://llama.meta.com/responsible-use-guide/) to outline the steps and best practices for developers to implement model and system level safety for their application. We also provide a set of resources including [Meta Llama Guard 2](https://llama.meta.com/purple-llama/) and [Code Shield](https://llama.meta.com/purple-llama/) safeguards. These tools have proven to drastically reduce residual risks of LLM Systems, while maintaining a high level of helpfulness. We encourage developers to tune and deploy these safeguards according to their needs and we provide a [reference implementation](https://github.com/meta-llama/llama-recipes/tree/main/recipes/responsible_ai) to get you started.
#### Llama 3-Instruct
As outlined in the Responsible Use Guide, some trade-off between model helpfulness and model alignment is likely unavoidable. Developers should exercise discretion about how to weigh the benefits of alignment and helpfulness for their specific use case and audience. Developers should be mindful of residual risks when using Llama models and leverage additional safety tools as needed to reach the right safety bar for their use case.
<span style="text-decoration:underline;">Safety</span>
For our instruction tuned model, we conducted extensive red teaming exercises, performed adversarial evaluations and implemented safety mitigations techniques to lower residual risks. As with any Large Language Model, residual risks will likely remain and we recommend that developers assess these risks in the context of their use case. In parallel, we are working with the community to make AI safety benchmark standards transparent, rigorous and interpretable.
<span style="text-decoration:underline;">Refusals</span>
In addition to residual risks, we put a great emphasis on model refusals to benign prompts. Over-refusing not only can impact the user experience but could even be harmful in certain contexts as well. We’ve heard the feedback from the developer community and improved our fine tuning to ensure that Llama 3 is significantly less likely to falsely refuse to answer prompts than Llama 2.
We built internal benchmarks and developed mitigations to limit false refusals making Llama 3 our most helpful model to date.
#### Responsible release
In addition to responsible use considerations outlined above, we followed a rigorous process that requires us to take extra measures against misuse and critical risks before we make our release decision.
Misuse
If you access or use Llama 3, you agree to the Acceptable Use Policy. The most recent copy of this policy can be found at [https://llama.meta.com/llama3/use-policy/](https://llama.meta.com/llama3/use-policy/).
#### Critical risks
<span style="text-decoration:underline;">CBRNE</span> (Chemical, Biological, Radiological, Nuclear, and high yield Explosives)
We have conducted a two fold assessment of the safety of the model in this area:
* Iterative testing during model training to assess the safety of responses related to CBRNE threats and other adversarial risks.
* Involving external CBRNE experts to conduct an uplift test assessing the ability of the model to accurately provide expert knowledge and reduce barriers to potential CBRNE misuse, by reference to what can be achieved using web search (without the model).
### <span style="text-decoration:underline;">Cyber Security </span>
We have evaluated Llama 3 with CyberSecEval, Meta’s cybersecurity safety eval suite, measuring Llama 3’s propensity to suggest insecure code when used as a coding assistant, and Llama 3’s propensity to comply with requests to help carry out cyber attacks, where attacks are defined by the industry standard MITRE ATT&CK cyber attack ontology. On our insecure coding and cyber attacker helpfulness tests, Llama 3 behaved in the same range or safer than models of [equivalent coding capability](https://huggingface.co/spaces/facebook/CyberSecEval).
### <span style="text-decoration:underline;">Child Safety</span>
Child Safety risk assessments were conducted using a team of experts, to assess the model’s capability to produce outputs that could result in Child Safety risks and inform on any necessary and appropriate risk mitigations via fine tuning. We leveraged those expert red teaming sessions to expand the coverage of our evaluation benchmarks through Llama 3 model development. For Llama 3, we conducted new in-depth sessions using objective based methodologies to assess the model risks along multiple attack vectors. We also partnered with content specialists to perform red teaming exercises assessing potentially violating content while taking account of market specific nuances or experiences.
### Community
Generative AI safety requires expertise and tooling, and we believe in the strength of the open community to accelerate its progress. We are active members of open consortiums, including the AI Alliance, Partnership in AI and MLCommons, actively contributing to safety standardization and transparency. We encourage the community to adopt taxonomies like the MLCommons Proof of Concept evaluation to facilitate collaboration and transparency on safety and content evaluations. Our Purple Llama tools are open sourced for the community to use and widely distributed across ecosystem partners including cloud service providers. We encourage community contributions to our [Github repository](https://github.com/meta-llama/PurpleLlama).
Finally, we put in place a set of resources including an [output reporting mechanism](https://developers.facebook.com/llama_output_feedback) and [bug bounty program](https://www.facebook.com/whitehat) to continuously improve the Llama technology with the help of the community.
## Ethical Considerations and Limitations
The core values of Llama 3 are openness, inclusivity and helpfulness. It is meant to serve everyone, and to work for a wide range of use cases. It is thus designed to be accessible to people across many different backgrounds, experiences and perspectives. Llama 3 addresses users and their needs as they are, without insertion unnecessary judgment or normativity, while reflecting the understanding that even content that may appear problematic in some cases can serve valuable purposes in others. It respects the dignity and autonomy of all users, especially in terms of the values of free thought and expression that power innovation and progress.
But Llama 3 is a new technology, and like any new technology, there are risks associated with its use. Testing conducted to date has been in English, and has not covered, nor could it cover, all scenarios. For these reasons, as with all LLMs, Llama 3’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 3 models, developers should perform safety testing and tuning tailored to their specific applications of the model. As outlined in the Responsible Use Guide, we recommend incorporating [Purple Llama](https://github.com/facebookresearch/PurpleLlama) solutions into your workflows and specifically [Llama Guard](https://ai.meta.com/research/publications/llama-guard-llm-based-input-output-safeguard-for-human-ai-conversations/) which provides a base model to filter input and output prompts to layer system-level safety on top of model-level safety.
Please see the Responsible Use Guide available at [http://llama.meta.com/responsible-use-guide](http://llama.meta.com/responsible-use-guide)
## Citation instructions
@article{llama3modelcard,
title={Llama 3 Model Card},
author={AI@Meta},
year={2024},
url = {https://github.com/meta-llama/llama3/blob/main/MODEL_CARD.md}
}
## Contributors
Aaditya Singh; Aaron Grattafiori; Abhimanyu Dubey; Abhinav Jauhri; Abhinav Pandey; Abhishek Kadian; Adam Kelsey; Adi Gangidi; Ahmad Al-Dahle; Ahuva Goldstand; Aiesha Letman; Ajay Menon; Akhil Mathur; Alan Schelten; Alex Vaughan; Amy Yang; Andrei Lupu; Andres Alvarado; Andrew Gallagher; Andrew Gu; Andrew Ho; Andrew Poulton; Andrew Ryan; Angela Fan; Ankit Ramchandani; Anthony Hartshorn; Archi Mitra; Archie Sravankumar; Artem Korenev; Arun Rao; Ashley Gabriel; Ashwin Bharambe; Assaf Eisenman; Aston Zhang; Aurelien Rodriguez; Austen Gregerson; Ava Spataru; Baptiste Roziere; Ben Maurer; Benjamin Leonhardi; Bernie Huang; Bhargavi Paranjape; Bing Liu; Binh Tang; Bobbie Chern; Brani Stojkovic; Brian Fuller; Catalina Mejia Arenas; Chao Zhou; Charlotte Caucheteux; Chaya Nayak; Ching-Hsiang Chu; Chloe Bi; Chris Cai; Chris Cox; Chris Marra; Chris McConnell; Christian Keller; Christoph Feichtenhofer; Christophe Touret; Chunyang Wu; Corinne Wong; Cristian Canton Ferrer; Damien Allonsius; Daniel Kreymer; Daniel Haziza; Daniel Li; Danielle Pintz; Danny Livshits; Danny Wyatt; David Adkins; David Esiobu; David Xu; Davide Testuggine; Delia David; Devi Parikh; Dhruv Choudhary; Dhruv Mahajan; Diana Liskovich; Diego Garcia-Olano; Diego Perino; Dieuwke Hupkes; Dingkang Wang; Dustin Holland; Egor Lakomkin; Elina Lobanova; Xiaoqing Ellen Tan; Emily Dinan; Eric Smith; Erik Brinkman; Esteban Arcaute; Filip Radenovic; Firat Ozgenel; Francesco Caggioni; Frank Seide; Frank Zhang; Gabriel Synnaeve; Gabriella Schwarz; Gabrielle Lee; Gada Badeer; Georgia Anderson; Graeme Nail; Gregoire Mialon; Guan Pang; Guillem Cucurell; Hailey Nguyen; Hannah Korevaar; Hannah Wang; Haroun Habeeb; Harrison Rudolph; Henry Aspegren; Hu Xu; Hugo Touvron; Iga Kozlowska; Igor Molybog; Igor Tufanov; Iliyan Zarov; Imanol Arrieta Ibarra; Irina-Elena Veliche; Isabel Kloumann; Ishan Misra; Ivan Evtimov; Jacob Xu; Jade Copet; Jake Weissman; Jan Geffert; Jana Vranes; Japhet Asher; Jason Park; Jay Mahadeokar; Jean-Baptiste Gaya; Jeet Shah; Jelmer van der Linde; Jennifer Chan; Jenny Hong; Jenya Lee; Jeremy Fu; Jeremy Teboul; Jianfeng Chi; Jianyu Huang; Jie Wang; Jiecao Yu; Joanna Bitton; Joe Spisak; Joelle Pineau; Jon Carvill; Jongsoo Park; Joseph Rocca; Joshua Johnstun; Junteng Jia; Kalyan Vasuden Alwala; Kam Hou U; Kate Plawiak; Kartikeya Upasani; Kaushik Veeraraghavan; Ke Li; Kenneth Heafield; Kevin Stone; Khalid El-Arini; Krithika Iyer; Kshitiz Malik; Kuenley Chiu; Kunal Bhalla; Kyle Huang; Lakshya Garg; Lauren Rantala-Yeary; Laurens van der Maaten; Lawrence Chen; Leandro Silva; Lee Bell; Lei Zhang; Liang Tan; Louis Martin; Lovish Madaan; Luca Wehrstedt; Lukas Blecher; Luke de Oliveira; Madeline Muzzi; Madian Khabsa; Manav Avlani; Mannat Singh; Manohar Paluri; Mark Zuckerberg; Marcin Kardas; Martynas Mankus; Mathew Oldham; Mathieu Rita; Matthew Lennie; Maya Pavlova; Meghan Keneally; Melanie Kambadur; Mihir Patel; Mikayel Samvelyan; Mike Clark; Mike Lewis; Min Si; Mitesh Kumar Singh; Mo Metanat; Mona Hassan; Naman Goyal; Narjes Torabi; Nicolas Usunier; Nikolay Bashlykov; Nikolay Bogoychev; Niladri Chatterji; Ning Dong; Oliver Aobo Yang; Olivier Duchenne; Onur Celebi; Parth Parekh; Patrick Alrassy; Paul Saab; Pavan Balaji; Pedro Rittner; Pengchuan Zhang; Pengwei Li; Petar Vasic; Peter Weng; Polina Zvyagina; Prajjwal Bhargava; Pratik Dubal; Praveen Krishnan; Punit Singh Koura; Qing He; Rachel Rodriguez; Ragavan Srinivasan; Rahul Mitra; Ramon Calderer; Raymond Li; Robert Stojnic; Roberta Raileanu; Robin Battey; Rocky Wang; Rohit Girdhar; Rohit Patel; Romain Sauvestre; Ronnie Polidoro; Roshan Sumbaly; Ross Taylor; Ruan Silva; Rui Hou; Rui Wang; Russ Howes; Ruty Rinott; Saghar Hosseini; Sai Jayesh Bondu; Samyak Datta; Sanjay Singh; Sara Chugh; Sargun Dhillon; Satadru Pan; Sean Bell; Sergey Edunov; Shaoliang Nie; Sharan Narang; Sharath Raparthy; Shaun Lindsay; Sheng Feng; Sheng Shen; Shenghao Lin; Shiva Shankar; Shruti Bhosale; Shun Zhang; Simon Vandenhende; Sinong Wang; Seohyun Sonia Kim; Soumya Batra; Sten Sootla; Steve Kehoe; Suchin Gururangan; Sumit Gupta; Sunny Virk; Sydney Borodinsky; Tamar Glaser; Tamar Herman; Tamara Best; Tara Fowler; Thomas Georgiou; Thomas Scialom; Tianhe Li; Todor Mihaylov; Tong Xiao; Ujjwal Karn; Vedanuj Goswami; Vibhor Gupta; Vignesh Ramanathan; Viktor Kerkez; Vinay Satish Kumar; Vincent Gonguet; Vish Vogeti; Vlad Poenaru; Vlad Tiberiu Mihailescu; Vladan Petrovic; Vladimir Ivanov; Wei Li; Weiwei Chu; Wenhan Xiong; Wenyin Fu; Wes Bouaziz; Whitney Meers; Will Constable; Xavier Martinet; Xiaojian Wu; Xinbo Gao; Xinfeng Xie; Xuchao Jia; Yaelle Goldschlag; Yann LeCun; Yashesh Gaur; Yasmine Babaei; Ye Qi; Yenda Li; Yi Wen; Yiwen Song; Youngjin Nam; Yuchen Hao; Yuchen Zhang; Yun Wang; Yuning Mao; Yuzi He; Zacharie Delpierre Coudert; Zachary DeVito; Zahra Hankir; Zhaoduo Wen; Zheng Yan; Zhengxing Chen; Zhenyu Yang; Zoe Papakipos
|
RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf | RichardErkhov | 2024-05-30T06:04:06Z | 454 | 0 | null | [
"gguf",
"arxiv:2311.17487",
"region:us"
]
| null | 2024-05-30T03:19:50Z | Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
Taiwan-LLM-7B-v2.0-base - GGUF
- Model creator: https://huggingface.co/yentinglin/
- Original model: https://huggingface.co/yentinglin/Taiwan-LLM-7B-v2.0-base/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [Taiwan-LLM-7B-v2.0-base.Q2_K.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q2_K.gguf) | Q2_K | 2.36GB |
| [Taiwan-LLM-7B-v2.0-base.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.IQ3_XS.gguf) | IQ3_XS | 2.6GB |
| [Taiwan-LLM-7B-v2.0-base.IQ3_S.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.IQ3_S.gguf) | IQ3_S | 2.75GB |
| [Taiwan-LLM-7B-v2.0-base.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q3_K_S.gguf) | Q3_K_S | 2.75GB |
| [Taiwan-LLM-7B-v2.0-base.IQ3_M.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.IQ3_M.gguf) | IQ3_M | 2.9GB |
| [Taiwan-LLM-7B-v2.0-base.Q3_K.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q3_K.gguf) | Q3_K | 3.07GB |
| [Taiwan-LLM-7B-v2.0-base.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q3_K_M.gguf) | Q3_K_M | 3.07GB |
| [Taiwan-LLM-7B-v2.0-base.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q3_K_L.gguf) | Q3_K_L | 3.35GB |
| [Taiwan-LLM-7B-v2.0-base.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.IQ4_XS.gguf) | IQ4_XS | 3.4GB |
| [Taiwan-LLM-7B-v2.0-base.Q4_0.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q4_0.gguf) | Q4_0 | 3.56GB |
| [Taiwan-LLM-7B-v2.0-base.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.IQ4_NL.gguf) | IQ4_NL | 3.58GB |
| [Taiwan-LLM-7B-v2.0-base.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q4_K_S.gguf) | Q4_K_S | 3.59GB |
| [Taiwan-LLM-7B-v2.0-base.Q4_K.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q4_K.gguf) | Q4_K | 3.8GB |
| [Taiwan-LLM-7B-v2.0-base.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q4_K_M.gguf) | Q4_K_M | 3.8GB |
| [Taiwan-LLM-7B-v2.0-base.Q4_1.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q4_1.gguf) | Q4_1 | 3.95GB |
| [Taiwan-LLM-7B-v2.0-base.Q5_0.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q5_0.gguf) | Q5_0 | 4.33GB |
| [Taiwan-LLM-7B-v2.0-base.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q5_K_S.gguf) | Q5_K_S | 4.33GB |
| [Taiwan-LLM-7B-v2.0-base.Q5_K.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q5_K.gguf) | Q5_K | 4.45GB |
| [Taiwan-LLM-7B-v2.0-base.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q5_K_M.gguf) | Q5_K_M | 4.45GB |
| [Taiwan-LLM-7B-v2.0-base.Q5_1.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q5_1.gguf) | Q5_1 | 4.72GB |
| [Taiwan-LLM-7B-v2.0-base.Q6_K.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q6_K.gguf) | Q6_K | 5.15GB |
| [Taiwan-LLM-7B-v2.0-base.Q8_0.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q8_0.gguf) | Q8_0 | 6.67GB |
Original model description:
---
license: apache-2.0
language:
- zh
widget:
- text: "A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. USER: 你好，請問你可以幫我寫一封推薦信嗎？ ASSISTANT:"
library_name: transformers
pipeline_tag: text-generation
extra_gated_heading: Acknowledge license to accept the repository.
extra_gated_prompt: Please contact the author for access.
extra_gated_button_content: Acknowledge license 同意以上內容
extra_gated_fields:
Name: text
Mail: text
Organization: text
Country: text
Any utilization of the Taiwan LLM repository mandates the explicit acknowledgment and attribution to the original author: checkbox
使用Taiwan LLM必須明確地承認和歸功於優必達株式會社 Ubitus 以及原始作者: checkbox
---
<img src="https://cdn-uploads.huggingface.co/production/uploads/5df9c78eda6d0311fd3d541f/CmusIT5OlSXvFrbTJ7l-C.png" alt="Taiwan LLM Logo" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
# 🌟 Checkout [Taiwan-LLM Demo Chat-UI](http://www.twllm.com) 🌟
# Model Card for Taiwan LLM 7B v2.0 base
Taiwan LLM is an advanced language model tailored for Traditional Chinese, focusing on the linguistic and cultural contexts of Taiwan.
Developed from a large base model, it's enriched with diverse Taiwanese textual sources and refined through Supervised Fine-Tuning.
This model excels in language understanding and generation, aligning closely with Taiwan's cultural nuances.
It demonstrates improved performance on various benchmarks like TC-Eval, showcasing its contextual comprehension and cultural relevance.
For detailed insights into Taiwan LLM's development and features, refer to our [technical report](https://github.com/MiuLab/Taiwan-LLaMa/blob/main/twllm_paper.pdf).
## Model description
- **Model type:** A 7B parameter GPT-like model fine-tuned on a mix of publicly available, synthetic datasets.
- **Language(s) (NLP):** Primarily Traditional Chinese (zh-tw)
- **Finetuned from model:** [meta-llama/Llama-2-7b-hf](https://huggingface.co/yentinglin/meta-llama/Llama-2-7b-hf)
### Model Sources
<!-- Provide the basic links for the model. -->
- **Repository:** https://github.com/MiuLab/Taiwan-LLaMa
- **Demo:** https://twllm.com/
## Performance
## Intended uses
You should fine-tuned this model for instruction-following / chat application.
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 5.0
## Citation
If you find Taiwan LLM is useful in your work, please cite it with:
```
@misc{lin2023taiwan,
title={Taiwan LLM: Bridging the Linguistic Divide with a Culturally Aligned Language Model},
author={Yen-Ting Lin and Yun-Nung Chen},
year={2023},
eprint={2311.17487},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
# Acknowledgement
Taiwan LLM v2 is conducted in collaboration with [Ubitus K.K.](http://ubitus.net). Ubitus provides valuable compute resources for the project.
|
Aardiiiiy/NusaBERT-base-Indonesian-Plutchik-emotion-analysis | Aardiiiiy | 2024-05-31T05:50:38Z | 454 | 1 | transformers | [
"transformers",
"safetensors",
"bert",
"text-classification",
"id",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
]
| text-classification | 2024-05-30T20:03:24Z | ---
library_name: transformers
license: apache-2.0
language:
- id
---
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
The NusaBERT-base-Indonesian-Plutchik-emotion-analysis is a model designed to identify and analyze emotions in Indonesian texts based on Plutchik's eight basic emotions: Anticipation, Anger, Disgust, Fear, Joy, Sadness, Surprise, and Trust.
This model is developed using the [NusaBERT-base](https://huggingface.co/LazarusNLP/NusaBERT-base) architecture and trained using Indonesian tweets categorized into eight emotion categories. The evaluation results of this model can be utilized to analyze emotions in social media, providing insights into users' emotional responses.
### Bias
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
Keep in mind that this model is trained using certain data which may cause bias in the emotion classification process. Therefore, it is important to consider and account for such biases when using this model.
### Evaluation Results
The training was conducted using a 5-fold cross-validation k-fold technique. The data is divided into five equal parts in this process, and the model is trained and tested repeatedly. Each time, four parts are used for training, and one part is used for testing.
* Average loss from Cross-Validation 0.11578517705202103
* Evaluation Details on the Last Fold:
* Evaluation Loss 0.01048247329890728
* Evaluation Time (seconds) 1.9982
* Samples per Second 382,352
* Steps per Second 24,022
* Epoch 3
<!-- This section describes the evaluation protocols and provides the results. -->
## Citation
```
@misc{Ardiyanto_Mikhael_2024,
author = {Mikhael Ardiyanto},
title = {NusaBERT-base-Indonesian-Plutchik-emotion-analysis },
year = {2024},
URL = {https://huggingface.co/Aardiiiiy/NusaBERT-base-Indonesian-Plutchik-emotion-analysis},
publisher = {Hugging Face}
}
``` |
ossama-ai/sd-fashion-forge-test | ossama-ai | 2024-06-18T10:12:24Z | 454 | 0 | diffusers | [
"diffusers",
"safetensors",
"text-to-image",
"stable-diffusion",
"art",
"en",
"license:openrail",
"autotrain_compatible",
"endpoints_compatible",
"diffusers:StableDiffusionPipeline",
"region:us"
]
| text-to-image | 2024-06-06T09:37:42Z | ---
license: openrail
tags:
- text-to-image
- stable-diffusion
- art
language:
- en
metrics:
- accuracy
library_name: diffusers
pipeline_tag: text-to-image
---
### sd-fashion_forge_test Dreambooth model trained by ossama-ai
|
RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf | RichardErkhov | 2024-06-15T02:52:50Z | 454 | 0 | null | [
"gguf",
"region:us"
]
| null | 2024-06-15T00:15:17Z | Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
Truthful_DPO_MOE_19B - GGUF
- Model creator: https://huggingface.co/yunconglong/
- Original model: https://huggingface.co/yunconglong/Truthful_DPO_MOE_19B/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [Truthful_DPO_MOE_19B.Q2_K.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q2_K.gguf) | Q2_K | 6.58GB |
| [Truthful_DPO_MOE_19B.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.IQ3_XS.gguf) | IQ3_XS | 7.34GB |
| [Truthful_DPO_MOE_19B.IQ3_S.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.IQ3_S.gguf) | IQ3_S | 7.75GB |
| [Truthful_DPO_MOE_19B.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q3_K_S.gguf) | Q3_K_S | 7.73GB |
| [Truthful_DPO_MOE_19B.IQ3_M.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.IQ3_M.gguf) | IQ3_M | 7.94GB |
| [Truthful_DPO_MOE_19B.Q3_K.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q3_K.gguf) | Q3_K | 8.59GB |
| [Truthful_DPO_MOE_19B.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q3_K_M.gguf) | Q3_K_M | 8.59GB |
| [Truthful_DPO_MOE_19B.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q3_K_L.gguf) | Q3_K_L | 9.32GB |
| [Truthful_DPO_MOE_19B.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.IQ4_XS.gguf) | IQ4_XS | 9.66GB |
| [Truthful_DPO_MOE_19B.Q4_0.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q4_0.gguf) | Q4_0 | 10.09GB |
| [Truthful_DPO_MOE_19B.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.IQ4_NL.gguf) | IQ4_NL | 10.19GB |
| [Truthful_DPO_MOE_19B.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q4_K_S.gguf) | Q4_K_S | 10.17GB |
| [Truthful_DPO_MOE_19B.Q4_K.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q4_K.gguf) | Q4_K | 10.79GB |
| [Truthful_DPO_MOE_19B.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q4_K_M.gguf) | Q4_K_M | 10.79GB |
| [Truthful_DPO_MOE_19B.Q4_1.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q4_1.gguf) | Q4_1 | 11.2GB |
| [Truthful_DPO_MOE_19B.Q5_0.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q5_0.gguf) | Q5_0 | 12.3GB |
| [Truthful_DPO_MOE_19B.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q5_K_S.gguf) | Q5_K_S | 12.3GB |
| [Truthful_DPO_MOE_19B.Q5_K.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q5_K.gguf) | Q5_K | 12.67GB |
| [Truthful_DPO_MOE_19B.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q5_K_M.gguf) | Q5_K_M | 12.67GB |
| [Truthful_DPO_MOE_19B.Q5_1.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q5_1.gguf) | Q5_1 | 13.41GB |
| [Truthful_DPO_MOE_19B.Q6_K.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q6_K.gguf) | Q6_K | 14.66GB |
| [Truthful_DPO_MOE_19B.Q8_0.gguf](https://huggingface.co/RichardErkhov/yunconglong_-_Truthful_DPO_MOE_19B-gguf/blob/main/Truthful_DPO_MOE_19B.Q8_0.gguf) | Q8_0 | 18.99GB |
Original model description:
---
license: other
tags:
- moe
- DPO
- RL-TUNED
---
* [DPO Trainer](https://huggingface.co/docs/trl/main/en/dpo_trainer) with dataset jondurbin/truthy-dpo-v0.1
```
DPO Trainer
TRL supports the DPO Trainer for training language models from preference data, as described in the paper Direct Preference Optimization: Your Language Model is Secretly a Reward Model by Rafailov et al., 2023.
```
|
CHE-72/Qwen2-7B-Instruct-Q5_K_S-GGUF | CHE-72 | 2024-06-21T18:36:00Z | 454 | 0 | null | [
"gguf",
"chat",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"en",
"base_model:Qwen/Qwen2-7B-Instruct",
"license:apache-2.0",
"region:us"
]
| text-generation | 2024-06-21T18:35:36Z | ---
base_model: Qwen/Qwen2-7B-Instruct
language:
- en
license: apache-2.0
pipeline_tag: text-generation
tags:
- chat
- llama-cpp
- gguf-my-repo
---
# CHE-72/Qwen2-7B-Instruct-Q5_K_S-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2-7B-Instruct`](https://huggingface.co/Qwen/Qwen2-7B-Instruct) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2-7B-Instruct) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo CHE-72/Qwen2-7B-Instruct-Q5_K_S-GGUF --hf-file qwen2-7b-instruct-q5_k_s.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo CHE-72/Qwen2-7B-Instruct-Q5_K_S-GGUF --hf-file qwen2-7b-instruct-q5_k_s.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo CHE-72/Qwen2-7B-Instruct-Q5_K_S-GGUF --hf-file qwen2-7b-instruct-q5_k_s.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo CHE-72/Qwen2-7B-Instruct-Q5_K_S-GGUF --hf-file qwen2-7b-instruct-q5_k_s.gguf -c 2048
```
|
martintmv/Hermes-2-Pro-Llama-3-8B-Q4_K_S-GGUF | martintmv | 2024-06-23T12:21:07Z | 454 | 0 | null | [
"gguf",
"Llama-3",
"instruct",
"finetune",
"chatml",
"DPO",
"RLHF",
"gpt4",
"synthetic data",
"distillation",
"function calling",
"json mode",
"axolotl",
"llama-cpp",
"gguf-my-repo",
"en",
"dataset:teknium/OpenHermes-2.5",
"base_model:NousResearch/Hermes-2-Pro-Llama-3-8B",
"region:us"
]
| null | 2024-06-23T12:20:47Z | ---
base_model: NousResearch/Hermes-2-Pro-Llama-3-8B
datasets:
- teknium/OpenHermes-2.5
language:
- en
tags:
- Llama-3
- instruct
- finetune
- chatml
- DPO
- RLHF
- gpt4
- synthetic data
- distillation
- function calling
- json mode
- axolotl
- llama-cpp
- gguf-my-repo
widget:
- example_title: Hermes 2 Pro
messages:
- role: system
content: You are a sentient, superintelligent artificial general intelligence,
here to teach and assist me.
- role: user
content: Write a short story about Goku discovering kirby has teamed up with Majin
Buu to destroy the world.
model-index:
- name: Hermes-2-Pro-Llama-3-8B
results: []
---
# martintmv/Hermes-2-Pro-Llama-3-8B-Q4_K_S-GGUF
This model was converted to GGUF format from [`NousResearch/Hermes-2-Pro-Llama-3-8B`](https://huggingface.co/NousResearch/Hermes-2-Pro-Llama-3-8B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/NousResearch/Hermes-2-Pro-Llama-3-8B) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo martintmv/Hermes-2-Pro-Llama-3-8B-Q4_K_S-GGUF --hf-file hermes-2-pro-llama-3-8b-q4_k_s.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo martintmv/Hermes-2-Pro-Llama-3-8B-Q4_K_S-GGUF --hf-file hermes-2-pro-llama-3-8b-q4_k_s.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo martintmv/Hermes-2-Pro-Llama-3-8B-Q4_K_S-GGUF --hf-file hermes-2-pro-llama-3-8b-q4_k_s.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo martintmv/Hermes-2-Pro-Llama-3-8B-Q4_K_S-GGUF --hf-file hermes-2-pro-llama-3-8b-q4_k_s.gguf -c 2048
```
|
BMukhtar/llama-3-8b-chat-doctor | BMukhtar | 2024-06-24T13:03:16Z | 454 | 0 | transformers | [
"transformers",
"safetensors",
"gguf",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
]
| null | 2024-06-24T11:31:34Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
NikolayKozloff/Replete-Coder-Llama3-8B-Q5_0-GGUF | NikolayKozloff | 2024-06-25T10:26:32Z | 454 | 1 | null | [
"gguf",
"region:us"
]
| null | 2024-06-25T10:26:05Z | Entry not found |
patrickvonplaten/led-large-16384-pubmed | patrickvonplaten | 2021-01-11T15:42:53Z | 453 | 12 | transformers | [
"transformers",
"pytorch",
"tf",
"led",
"text2text-generation",
"en",
"dataset:scientific_papers",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
]
| text2text-generation | 2022-03-02T23:29:05Z | ---
language: en
datasets:
- scientific_papers
license: apache-2.0
---
## Introduction
[Allenai's Longformer Encoder-Decoder (LED)](https://github.com/allenai/longformer#longformer).
This is an unofficial *led-large-16384* checkpoint that is fine-tuned on the [pubmed dataset](https://huggingface.co/datasets/scientific_papers).
The model was fine-tuned and evaluated as detailed in [this notebook](https://colab.research.google.com/drive/12LjJazBl7Gam0XBPy_y0CTOJZeZ34c2v?usp=sharing)
## Results
The model achieves a **Rouge-2** score of 19.33 on Pubmed which is competitive to state-of-the-art models.
## Usage
The model can be used as follows. The input is taken from the test data of the [pubmed dataset](https://huggingface.co/datasets/scientific_papers).
```python
LONG_ARTICLE = """"anxiety affects quality of life in those living
with parkinson 's disease ( pd ) more so than
overall cognitive status , motor deficits , apathy
, and depression [ 13 ] . although anxiety and
depression are often related and coexist in pd
patients , recent research suggests that anxiety
rather than depression is the most prominent and
prevalent mood disorder in pd [ 5 , 6 ] . yet ,
our current understanding of anxiety and its
impact on cognition in pd , as well as its neural
basis and best treatment practices , remains
meager and lags far behind that of depression .
overall , neuropsychiatric symptoms in pd have
been shown to be negatively associated with
cognitive performance . for example , higher
depression scores have been correlated with lower
scores on the mini - mental state exam ( mmse ) [
8 , 9 ] as well as tests of memory and executive
functions ( e.g. , attention ) [ 1014 ] . likewise
, apathy and anhedonia in pd patients have been
associated with executive dysfunction [ 10 , 1523
] . however , few studies have specifically
investigated the relationship between anxiety and
cognition in pd . one study showed a strong
negative relationship between anxiety ( both state
and trait ) and overall cognitive performance (
measured by the total of the repeatable battery
for the assessment of neuropsychological status
index ) within a sample of 27 pd patients .
furthermore , trait anxiety was negatively
associated with each of the cognitive domains
assessed by the rbans ( i.e. , immediate memory ,
visuospatial construction , language , attention ,
and delayed memory ) . two further studies have
examined whether anxiety differentially affects
cognition in patients with left - sided dominant
pd ( lpd ) versus right - sided dominant pd ( rpd
) ; however , their findings were inconsistent .
the first study found that working memory
performance was worse in lpd patients with anxiety
compared to rpd patients with anxiety , whereas
the second study reported that , in lpd , apathy
but not anxiety was associated with performance on
nonverbally mediated executive functions and
visuospatial tasks ( e.g. , tmt - b , wms - iii
spatial span ) , while in rpd , anxiety but not
apathy significantly correlated with performance
on verbally mediated tasks ( e.g. , clock reading
test and boston naming test ) . furthermore ,
anxiety was significantly correlated with
neuropsychological measures of attention and
executive and visuospatial functions . taken
together , it is evident that there are limited
and inconsistent findings describing the
relationship between anxiety and cognition in pd
and more specifically how anxiety might influence
particular domains of cognition such as attention
and memory and executive functioning . it is also
striking that , to date , no study has examined
the influence of anxiety on cognition in pd by
directly comparing groups of pd patients with and
without anxiety while excluding depression . given
that research on healthy young adults suggests
that anxiety reduces processing capacity and
impairs processing efficiency , especially in the
central executive and attentional systems of
working memory [ 26 , 27 ] , we hypothesized that
pd patients with anxiety would show impairments in
attentional set - shifting and working memory
compared to pd patients without anxiety .
furthermore , since previous work , albeit limited
, has focused on the influence of symptom
laterality on anxiety and cognition , we also
explored this relationship . seventeen pd patients
with anxiety and thirty - three pd patients
without anxiety were included in this study ( see
table 1 ) . the cross - sectional data from these
participants was taken from a patient database
that has been compiled over the past 8 years (
since 2008 ) at the parkinson 's disease research
clinic at the brain and mind centre , university
of sydney . inclusion criteria involved a
diagnosis of idiopathic pd according to the united
kingdom parkinson 's disease society brain bank
criteria and were confirmed by a neurologist (
sjgl ) . patients also had to have an adequate
proficiency in english and have completed a full
neuropsychological assessment . ten patients in
this study ( 5 pd with anxiety ; 5 pd without
anxiety ) were taking psychotropic drugs ( i.e. ,
benzodiazepine or selective serotonin reuptake
inhibitor ) . patients were also excluded if they
had other neurological disorders , psychiatric
disorders other than affective disorders ( such as
anxiety ) , or if they reported a score greater
than six on the depression subscale of the
hospital anxiety and depression scale ( hads ) .
thus , all participants who scored within a
depressed ( hads - d > 6 ) range were excluded
from this study , in attempt to examine a refined
sample of pd patients with and without anxiety in
order to determine the independent effect of
anxiety on cognition . this research was approved
by the human research ethics committee of the
university of sydney , and written informed
consent was obtained from all participants . self
- reported hads was used to assess anxiety in pd
and has been previously shown to be a useful
measure of clinical anxiety in pd . a cut - off
score of > 8 on the anxiety subscale of the hads (
hads - a ) was used to identify pd cases with
anxiety ( pda+ ) , while a cut - off score of < 6
on the hads - a was used to identify pd cases
without anxiety ( pda ) . this criterion was more
stringent than usual ( > 7 cut - off score ) , in
effort to create distinct patient groups . the
neurological evaluation rated participants
according to hoehn and yahr ( h&y ) stages and
assessed their motor symptoms using part iii of
the revised mds task force unified parkinson 's
disease rating scale ( updrs ) . in a similar way
this was determined by calculating a total left
and right score from rigidity items 3035 ,
voluntary movement items 3643 , and tremor items
5057 from the mds - updrs part iii ( see table 1 )
. processing speed was assessed using the trail
making test , part a ( tmt - a , z - score ) .
attentional set - shifting was measured using the
trail making test , part b ( tmt - b , z - score )
. working memory was assessed using the digit span
forward and backward subtest of the wechsler
memory scale - iii ( raw scores ) . language was
assessed with semantic and phonemic verbal fluency
via the controlled oral word associated test (
cowat animals and letters , z - score ) . the
ability to retain learned verbal memory was
assessed using the logical memory subtest from the
wechsler memory scale - iii ( lm - i z - score ,
lm - ii z - score , % lm retention z - score ) .
the mini - mental state examination ( mmse )
demographic , clinical , and neuropsychological
variables were compared between the two groups
with the independent t - test or mann whitney u
test , depending on whether the variable met
parametric assumptions . chi - square tests were
used to examine gender and symptom laterality
differences between groups . all analyses employed
an alpha level of p < 0.05 and were two - tailed .
spearman correlations were performed separately in
each group to examine associations between anxiety
and/or depression ratings and cognitive functions
. as expected , the pda+ group reported
significant greater levels of anxiety on the hads
- a ( u = 0 , p < 0.001 ) and higher total score
on the hads ( u = 1 , p < 0.001 ) compared to the
pda group ( table 1 ) . groups were matched in age
( t(48 ) = 1.31 , p = 0.20 ) , disease duration (
u = 259 , p = 0.66 ) , updrs - iii score ( u =
250.5 , p = 0.65 ) , h&y ( u = 245 , p = 0.43 ) ,
ledd ( u = 159.5 , p = 0.80 ) , and depression (
hads - d ) ( u = 190.5 , p = 0.06 ) . additionally
, all groups were matched in the distribution of
gender ( = 0.098 , p = 0.75 ) and side - affected
( = 0.765 , p = 0.38 ) . there were no group
differences for tmt - a performance ( u = 256 , p
= 0.62 ) ( table 2 ) ; however , the pda+ group
had worse performance on the trail making test
part b ( t(46 ) = 2.03 , p = 0.048 ) compared to
the pda group ( figure 1 ) . the pda+ group also
demonstrated significantly worse performance on
the digit span forward subtest ( t(48 ) = 2.22 , p
= 0.031 ) and backward subtest ( u = 190.5 , p =
0.016 ) compared to the pda group ( figures 2(a )
and 2(b ) ) . neither semantic verbal fluency (
t(47 ) = 0.70 , p = 0.49 ) nor phonemic verbal
fluency ( t(47 ) = 0.39 , p = 0.70 ) differed
between groups . logical memory i immediate recall
test ( u = 176 , p = 0.059 ) showed a trend that
the pda+ group had worse new verbal learning and
immediate recall abilities than the pda group .
however , logical memory ii test performance ( u =
219 , p = 0.204 ) and logical memory % retention (
u = 242.5 , p = 0.434 ) did not differ between
groups . there were also no differences between
groups in global cognition ( mmse ) ( u = 222.5 ,
p = 0.23 ) . participants were split into lpd and
rpd , and then further group differences were
examined between pda+ and pda. importantly , the
groups remained matched in age , disease duration
, updrs - iii , dde , h&y stage , and depression
but remained significantly different on self -
reported anxiety . lpda+ demonstrated worse
performance on the digit span forward test ( t(19
) = 2.29 , p = 0.033 ) compared to lpda , whereas
rpda+ demonstrated worse performance on the digit
span backward test ( u = 36.5 , p = 0.006 ) , lm -
i immediate recall ( u = 37.5 , p = 0.008 ) , and
lm - ii ( u = 45.0 , p = 0.021 ) but not lm %
retention ( u = 75.5 , p = 0.39 ) compared to
rpda. this study is the first to directly compare
cognition between pd patients with and without
anxiety . the findings confirmed our hypothesis
that anxiety negatively influences attentional set
- shifting and working memory in pd . more
specifically , we found that pd patients with
anxiety were more impaired on the trail making
test part b which assessed attentional set -
shifting , on both digit span tests which assessed
working memory and attention , and to a lesser
extent on the logical memory test which assessed
memory and new verbal learning compared to pd
patients without anxiety . taken together , these
findings suggest that anxiety in pd may reduce
processing capacity and impair processing
efficiency , especially in the central executive
and attentional systems of working memory in a
similar way as seen in young healthy adults [ 26 ,
27 ] . although the neurobiology of anxiety in pd
remains unknown , many researchers have postulated
that anxiety disorders are related to
neurochemical changes that occur during the early
, premotor stages of pd - related degeneration [
37 , 38 ] such as nigrostriatal dopamine depletion
, as well as cell loss within serotonergic and
noradrenergic brainstem nuclei ( i.e. , raphe
nuclei and locus coeruleus , resp . , which
provide massive inputs to corticolimbic regions )
. over time , chronic dysregulation of
adrenocortical and catecholamine functions can
lead to hippocampal damage as well as
dysfunctional prefrontal neural circuitries [ 39 ,
40 ] , which play a key role in memory and
attention . recent functional neuroimaging work
has suggested that enhanced hippocampal activation
during executive functioning and working memory
tasks may represent compensatory processes for
impaired frontostriatal functions in pd patients
compared to controls . therefore , chronic stress
from anxiety , for example , may disrupt
compensatory processes in pd patients and explain
the cognitive impairments specifically in working
memory and attention seen in pd patients with
anxiety . it has also been suggested that
hyperactivation within the putamen may reflect a
compensatory striatal mechanism to maintain normal
working memory performance in pd patients ;
however , losing this compensatory activation has
been shown to contribute to poor working memory
performance . anxiety in mild pd has been linked
to reduced putamen dopamine uptake which becomes
more extensive as the disease progresses . this
further supports the notion that anxiety may
disrupt compensatory striatal mechanisms as well ,
providing another possible explanation for the
cognitive impairments observed in pd patients with
anxiety in this study . noradrenergic and
serotonergic systems should also be considered
when trying to explain the mechanisms by which
anxiety may influence cognition in pd . although
these neurotransmitter systems are relatively
understudied in pd cognition , treating the
noradrenergic and serotonergic systems has shown
beneficial effects on cognition in pd . selective
serotonin reuptake inhibitor , citalopram , was
shown to improve response inhibition deficits in
pd , while noradrenaline reuptake blocker ,
atomoxetine , has been recently reported to have
promising effects on cognition in pd [ 45 , 46 ] .
overall , very few neuroimaging studies have been
conducted in pd in order to understand the neural
correlates of pd anxiety and its underlying neural
pathology . future research should focus on
relating anatomical changes and neurochemical
changes to neural activation in order to gain a
clearer understanding on how these pathologies
affect anxiety in pd . to further understand how
anxiety and cognitive dysfunction are related ,
future research should focus on using advanced
structural and function imaging techniques to
explain both cognitive and neural breakdowns that
are associated with anxiety in pd patients .
research has indicated that those with amnestic
mild cognitive impairment who have more
neuropsychiatric symptoms have a greater risk of
developing dementia compared to those with fewer
neuropsychiatric symptoms . future studies should
also examine whether treating neuropsychiatric
symptoms might impact the progression of cognitive
decline and improve cognitive impairments in pd
patients . previous studies have used pd symptom
laterality as a window to infer asymmetrical
dysfunction of neural circuits . for example , lpd
patients have greater inferred right hemisphere
pathology , whereas rpd patients have greater
inferred left hemisphere pathology . thus ,
cognitive domains predominantly subserved by the
left hemisphere ( e.g. , verbally mediated tasks
of executive function and verbal memory ) might be
hypothesized to be more affected in rpd than lpd ;
however , this remains controversial . it has also
been suggested that since anxiety is a common
feature of left hemisphere involvement [ 48 , 49 ]
, cognitive domains subserved by the left
hemisphere may also be more strongly related to
anxiety . results from this study showed selective
verbal memory deficits in rpd patients with
anxiety compared to rpd without anxiety , whereas
lpd patients with anxiety had greater attentional
/ working memory deficits compared to lpd without
anxiety . although these results align with
previous research , interpretations of these
findings should be made with caution due to the
small sample size in the lpd comparison
specifically . recent work has suggested that the
hads questionnaire may underestimate the burden of
anxiety related symptomology and therefore be a
less sensitive measure of anxiety in pd [ 30 , 50
] . in addition , our small sample size also
limited the statistical power for detecting
significant findings . based on these limitations
, our findings are likely conservative and
underrepresent the true impact anxiety has on
cognition in pd . additionally , the current study
employed a very brief neuropsychological
assessment including one or two tests for each
cognitive domain . future studies are encouraged
to collect a more complex and comprehensive
battery from a larger sample of pd participants in
order to better understand the role anxiety plays
on cognition in pd . another limitation of this
study was the absence of diagnostic interviews to
characterize participants ' psychiatric symptoms
and specify the type of anxiety disorders included
in this study . future studies should perform
diagnostic interviews with participants ( e.g. ,
using dsm - v criteria ) rather than relying on
self - reported measures to group participants ,
in order to better understand whether the type of
anxiety disorder ( e.g. , social anxiety , phobias
, panic disorders , and generalized anxiety )
influences cognitive performance differently in pd
. one advantage the hads questionnaire provided
over other anxiety scales was that it assessed
both anxiety and depression simultaneously and
allowed us to control for coexisting depression .
although there was a trend that the pda+ group
self - reported higher levels of depression than
the pda group , all participants included in the
study scored < 6 on the depression subscale of the
hads . controlling for depression while assessing
anxiety has been identified as a key shortcoming
in the majority of recent work . considering many
previous studies have investigated the influence
of depression on cognition in pd without
accounting for the presence of anxiety and the
inconsistent findings reported to date , we
recommend that future research should try to
disentangle the influence of anxiety versus
depression on cognitive impairments in pd .
considering the growing number of clinical trials
for treating depression , there are few if any for
the treatment of anxiety in pd . anxiety is a key
contributor to decreased quality of life in pd and
greatly requires better treatment options .
moreover , anxiety has been suggested to play a
key role in freezing of gait ( fog ) , which is
also related to attentional set - shifting [ 52 ,
53 ] . future research should examine the link
between anxiety , set - shifting , and fog , in
order to determine whether treating anxiety might
be a potential therapy for improving fog ."""
from transformers import LEDForConditionalGeneration, LEDTokenizer
import torch
tokenizer = LEDTokenizer.from_pretrained("patrickvonplaten/led-large-16384-pubmed")
input_ids = tokenizer(LONG_ARTICLE, return_tensors="pt").input_ids.to("cuda")
global_attention_mask = torch.zeros_like(input_ids)
# set global_attention_mask on first token
global_attention_mask[:, 0] = 1
model = LEDForConditionalGeneration.from_pretrained("patrickvonplaten/led-large-16384-pubmed", return_dict_in_generate=True).to("cuda")
sequences = model.generate(input_ids, global_attention_mask=global_attention_mask).sequences
summary = tokenizer.batch_decode(sequences)
``` |
allenai/mtk-instruct-11b-def-pos | allenai | 2023-01-24T17:10:54Z | 453 | 5 | transformers | [
"transformers",
"pytorch",
"t5",
"text2text-generation",
"multilingual",
"arxiv:1910.10683",
"arxiv:2204.07705",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text2text-generation | 2022-05-27T15:41:41Z | ---
language: multilingual
license: apache-2.0
datasets:
- natural instructions v2.0
---
# Model description
Tk-Instruct is a series of encoder-decoder Transformer models that are trained to solve various NLP tasks by following in-context instructions (plain language task definitions, k-shot examples, explanations, etc). Built upon the pre-trained [T5 models](https://arxiv.org/abs/1910.10683), they are fine-tuned on a large number of tasks & instructions that are collected in the [Natural Instructions benchmark](https://github.com/allenai/natural-instructions), which contains 1600+ tasks in 70+ broach categories in total. This enables the model to not only process the training tasks, but also generalize to many unseen tasks without further parameter update.
More resources for using the model:
- **Paper**: [link](https://arxiv.org/abs/2204.07705)
- **Code repository**: [Tk-Instruct](https://github.com/yizhongw/Tk-Instruct)
- **Official Website**: [Natural Instructions](https://instructions.apps.allenai.org/)
- **All released models**: [allenai/tk-instruct](https://huggingface.co/models?search=allenai/tk-instruct)
## Intended uses & limitations
Tk-Instruct can be used to do many NLP tasks by following instructions.
### How to use
When instructing the model, task definition or demonstration examples or explanations should be prepended to the original input and fed into the model. You can easily try Tk-Instruct models as follows:
```python
>>> from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
>>> tokenizer = AutoTokenizer.from_pretrained("allenai/tk-instruct-3b-def")
>>> model = AutoModelForSeq2SeqLM.from_pretrained("allenai/tk-instruct-3b-def")
>>> input_ids = tokenizer.encode(
"Definition: return the currency of the given country. Now complete the following example - Input: India. Output:",
return_tensors="pt")
>>> output = model.generate(input_ids, max_length=10)
>>> output = tokenizer.decode(output[0], skip_special_tokens=True) # model should output 'Indian Rupee'
>>> input_ids = tokenizer.encode(
"Definition: negate the following sentence. Input: John went to school. Output:",
return_tensors="pt")
>>> output = model.generate(input_ids, max_length=10)
>>> output = tokenizer.decode(output[0], skip_special_tokens=True) # model should output 'John did not go to shool.'
```
### Limitations
We are still working on understanding the behaviors of these models, but here are several issues we have found:
- Models are generally sensitive to the instruction. Sometimes rewording the instruction can lead to very different output.
- Models are not always compliant to the instruction. Sometimes the model don't follow your instruction (e.g., when you ask the model to generate one sentence, it might still generate one word or a long story).
- Models might totally fail on some tasks.
If you find serious issues or any interesting result, you are welcome to share with us!
## Training data
Tk-Instruct is trained using the tasks & instructions in [Natural Instructions benchmark](https://github.com/allenai/natural-instructions), which contains 1600+ tasks in 70+ broach categories in total. We follow the official train/test split. Tk-Instruct model series were trained using 757 tasks, and mTk-Instruct series were trained using 1271 tasks (including some non-English tasks).
The training tasks are in 64 broad categories, such as text categorization / question answering / sentiment analysis / summarization / grammar error detection / dialogue generation / etc. The other 12 categories are selected for evaluation.
## Training procedure
All our models are initialized from either T5 models or mT5 models. Because generating the output can be regarded as a form of language modeling, we used their [LM adapted version](https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#lm-adapted-t511lm100k). All data is converted into a text-to-text format, and models are fine-tuned to maximize the likelihood of the output sequence.
Our [released models](https://huggingface.co/models?search=allenai/tk-instruct) are in different sizes, and each of them was trained with a specific type of instruction encoding. For instance, `tk-instruct-3b-def-pos` was initialized from [t5-xl-lm-adapt](https://huggingface.co/google/t5-xl-lm-adapt), and it saw task definition & 2 positive examples as the instruction during training time.
Although they are trained with only one type of instruction encodings, we found they can usually work with other type of encodings at test time (see more in our paper).
### BibTeX entry and citation info
```bibtex
@article{wang2022benchmarking,
title={Benchmarking Generalization via In-Context Instructions on 1,600+ Language Tasks},
author={Yizhong Wang and Swaroop Mishra and Pegah Alipoormolabashi and Yeganeh Kordi and Amirreza Mirzaei and A. Arunkumar and Arjun Ashok and Arut Selvan Dhanasekaran and Atharva Naik and David Stap and Eshaan Pathak and Giannis Karamanolakis and Haizhi Gary Lai and Ishan Purohit and Ishani Mondal and Jacob Anderson and Kirby Kuznia and Krima Doshi and Maitreya Patel and Kuntal Kumar Pal and M. Moradshahi and Mihir Parmar and Mirali Purohit and Neeraj Varshney and Phani Rohitha Kaza and Pulkit Verma and Ravsehaj Singh Puri and Rushang Karia and Shailaja Keyur Sampat and Savan Doshi and Siddharth Deepak Mishra and Sujan C. Reddy and Sumanta Patro and Tanay Dixit and Xu-dong Shen and Chitta Baral and Yejin Choi and Hannaneh Hajishirzi and Noah A. Smith and Daniel Khashabi},
year={2022},
archivePrefix={arXiv},
eprint={2204.07705},
primaryClass={cs.CL},
}
``` |
NeelNanda/SoLU_4L_v11_old | NeelNanda | 2022-10-26T17:14:11Z | 453 | 0 | transformers | [
"transformers",
"endpoints_compatible",
"region:us"
]
| null | 2022-10-14T23:36:06Z | Entry not found |
snorkelai/sdnet | snorkelai | 2023-08-30T00:27:54Z | 453 | 0 | transformers | [
"transformers",
"pytorch",
"safetensors",
"t5",
"text2text-generation",
"arxiv:2203.12252",
"license:cc-by-4.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text2text-generation | 2022-10-19T18:58:55Z | ---
license: cc-by-4.0
---
This model is an adaptation of Self-describing Networks introduced by _Jiawei Chen, Qing Liu, Hongyu Lin, Xianpei Han, Le Sun_ in _Few-shot Named Entity Recognition with Self-describing Networks 2022_.
https://arxiv.org/pdf/2203.12252.pdf |
SSI/SportsFanGhost | SSI | 2023-03-04T17:56:38Z | 453 | 0 | transformers | [
"transformers",
"pytorch",
"gpt2",
"text-generation",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2023-03-04T05:32:29Z | Entry not found |
timm/repvgg_b3.rvgg_in1k | timm | 2024-02-10T23:35:03Z | 453 | 0 | timm | [
"timm",
"pytorch",
"safetensors",
"image-classification",
"dataset:imagenet-1k",
"arxiv:2101.03697",
"license:mit",
"region:us"
]
| image-classification | 2023-03-22T07:23:45Z | ---
license: mit
library_name: timm
tags:
- image-classification
- timm
datasets:
- imagenet-1k
---
# Model card for repvgg_b3.rvgg_in1k
A RepVGG image classification model. Trained on ImageNet-1k by paper authors.
This model architecture is implemented using `timm`'s flexible [BYOBNet (Bring-Your-Own-Blocks Network)](https://github.com/huggingface/pytorch-image-models/blob/main/timm/models/byobnet.py).
BYOBNet allows configuration of:
* block / stage layout
* stem layout
* output stride (dilation)
* activation and norm layers
* channel and spatial / self-attention layers
...and also includes `timm` features common to many other architectures, including:
* stochastic depth
* gradient checkpointing
* layer-wise LR decay
* per-stage feature extraction
## Model Details
- **Model Type:** Image classification / feature backbone
- **Model Stats:**
- Params (M): 123.1
- GMACs: 29.2
- Activations (M): 15.1
- Image size: 224 x 224
- **Papers:**
- RepVGG: Making VGG-style ConvNets Great Again: https://arxiv.org/abs/2101.03697
- **Dataset:** ImageNet-1k
- **Original:** https://github.com/DingXiaoH/RepVGG
## Model Usage
### Image Classification
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model('repvgg_b3.rvgg_in1k', pretrained=True)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)
```
### Feature Map Extraction
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'repvgg_b3.rvgg_in1k',
pretrained=True,
features_only=True,
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
for o in output:
# print shape of each feature map in output
# e.g.:
# torch.Size([1, 64, 112, 112])
# torch.Size([1, 192, 56, 56])
# torch.Size([1, 384, 28, 28])
# torch.Size([1, 768, 14, 14])
# torch.Size([1, 2560, 7, 7])
print(o.shape)
```
### Image Embeddings
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'repvgg_b3.rvgg_in1k',
pretrained=True,
num_classes=0, # remove classifier nn.Linear
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # output is (batch_size, num_features) shaped tensor
# or equivalently (without needing to set num_classes=0)
output = model.forward_features(transforms(img).unsqueeze(0))
# output is unpooled, a (1, 2560, 7, 7) shaped tensor
output = model.forward_head(output, pre_logits=True)
# output is a (1, num_features) shaped tensor
```
## Model Comparison
Explore the dataset and runtime metrics of this model in timm [model results](https://github.com/huggingface/pytorch-image-models/tree/main/results).
## Citation
```bibtex
@misc{rw2019timm,
author = {Ross Wightman},
title = {PyTorch Image Models},
year = {2019},
publisher = {GitHub},
journal = {GitHub repository},
doi = {10.5281/zenodo.4414861},
howpublished = {\url{https://github.com/huggingface/pytorch-image-models}}
}
```
```bibtex
@inproceedings{ding2021repvgg,
title={Repvgg: Making vgg-style convnets great again},
author={Ding, Xiaohan and Zhang, Xiangyu and Ma, Ningning and Han, Jungong and Ding, Guiguang and Sun, Jian},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
pages={13733--13742},
year={2021}
}
```
|
roneneldan/TinyStories-Instruct-28M | roneneldan | 2023-05-17T22:09:23Z | 453 | 0 | transformers | [
"transformers",
"pytorch",
"gpt_neo",
"text-generation",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
]
| text-generation | 2023-05-12T21:53:09Z | Entry not found |
TheBloke/Hermes-LLongMA-2-7B-8K-GGUF | TheBloke | 2023-09-27T12:47:48Z | 453 | 2 | transformers | [
"transformers",
"gguf",
"llama",
"base_model:conceptofmind/Hermes-LLongMA-2-7b-8k",
"license:llama2",
"text-generation-inference",
"region:us"
]
| null | 2023-09-05T17:38:06Z | ---
license: llama2
model_name: Hermes LLongMA 2 7B 8K
base_model: conceptofmind/Hermes-LLongMA-2-7b-8k
inference: false
model_creator: conceptofmind
model_type: llama
prompt_template: 'Below is an instruction that describes a task. Write a response
that appropriately completes the request.
### Instruction:
{prompt}
### Response:
'
quantized_by: TheBloke
---
<!-- header start -->
<!-- 200823 -->
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div>
<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
<!-- header end -->
# Hermes LLongMA 2 7B 8K - GGUF
- Model creator: [conceptofmind](https://huggingface.co/conceptofmind)
- Original model: [Hermes LLongMA 2 7B 8K](https://huggingface.co/conceptofmind/Hermes-LLongMA-2-7b-8k)
<!-- description start -->
## Description
This repo contains GGUF format model files for [conceptofmind's Hermes LLongMA 2 7B 8K](https://huggingface.co/conceptofmind/Hermes-LLongMA-2-7b-8k).
<!-- description end -->
<!-- README_GGUF.md-about-gguf start -->
### About GGUF
GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp. GGUF offers numerous advantages over GGML, such as better tokenisation, and support for special tokens. It is also supports metadata, and is designed to be extensible.
Here is an incomplate list of clients and libraries that are known to support GGUF:
* [llama.cpp](https://github.com/ggerganov/llama.cpp). The source project for GGUF. Offers a CLI and a server option.
* [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
* [KoboldCpp](https://github.com/LostRuins/koboldcpp), a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
* [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration.
* [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with many interesting and unique features, including a full model library for easy model selection.
* [Faraday.dev](https://faraday.dev/), an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
* [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
* [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
* [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
<!-- README_GGUF.md-about-gguf end -->
<!-- repositories-available start -->
## Repositories available
* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-AWQ)
* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF)
* [conceptofmind's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/conceptofmind/Hermes-LLongMA-2-7b-8k)
<!-- repositories-available end -->
<!-- prompt-template start -->
## Prompt template: Alpaca
```
Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
{prompt}
### Response:
```
<!-- prompt-template end -->
<!-- compatibility_gguf start -->
## Compatibility
These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit [d0cee0d36d5be95a0d9088b674dbb27354107221](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221)
They are also compatible with many third party UIs and libraries - please see the list at the top of this README.
## Explanation of quantisation methods
<details>
<summary>Click to see details</summary>
The new methods available are:
* GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
* GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
* GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
* GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
* GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw
Refer to the Provided Files table below to see what files use which methods, and how.
</details>
<!-- compatibility_gguf end -->
<!-- README_GGUF.md-provided-files start -->
## Provided files
| Name | Quant method | Bits | Size | Max RAM required | Use case |
| ---- | ---- | ---- | ---- | ---- | ----- |
| [hermes-llongma-2-7b-8k.Q2_K.gguf](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF/blob/main/hermes-llongma-2-7b-8k.Q2_K.gguf) | Q2_K | 2 | 2.83 GB| 5.33 GB | smallest, significant quality loss - not recommended for most purposes |
| [hermes-llongma-2-7b-8k.Q3_K_S.gguf](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF/blob/main/hermes-llongma-2-7b-8k.Q3_K_S.gguf) | Q3_K_S | 3 | 2.95 GB| 5.45 GB | very small, high quality loss |
| [hermes-llongma-2-7b-8k.Q3_K_M.gguf](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF/blob/main/hermes-llongma-2-7b-8k.Q3_K_M.gguf) | Q3_K_M | 3 | 3.30 GB| 5.80 GB | very small, high quality loss |
| [hermes-llongma-2-7b-8k.Q3_K_L.gguf](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF/blob/main/hermes-llongma-2-7b-8k.Q3_K_L.gguf) | Q3_K_L | 3 | 3.60 GB| 6.10 GB | small, substantial quality loss |
| [hermes-llongma-2-7b-8k.Q4_0.gguf](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF/blob/main/hermes-llongma-2-7b-8k.Q4_0.gguf) | Q4_0 | 4 | 3.83 GB| 6.33 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [hermes-llongma-2-7b-8k.Q4_K_S.gguf](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF/blob/main/hermes-llongma-2-7b-8k.Q4_K_S.gguf) | Q4_K_S | 4 | 3.86 GB| 6.36 GB | small, greater quality loss |
| [hermes-llongma-2-7b-8k.Q4_K_M.gguf](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF/blob/main/hermes-llongma-2-7b-8k.Q4_K_M.gguf) | Q4_K_M | 4 | 4.08 GB| 6.58 GB | medium, balanced quality - recommended |
| [hermes-llongma-2-7b-8k.Q5_0.gguf](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF/blob/main/hermes-llongma-2-7b-8k.Q5_0.gguf) | Q5_0 | 5 | 4.65 GB| 7.15 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [hermes-llongma-2-7b-8k.Q5_K_S.gguf](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF/blob/main/hermes-llongma-2-7b-8k.Q5_K_S.gguf) | Q5_K_S | 5 | 4.65 GB| 7.15 GB | large, low quality loss - recommended |
| [hermes-llongma-2-7b-8k.Q5_K_M.gguf](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF/blob/main/hermes-llongma-2-7b-8k.Q5_K_M.gguf) | Q5_K_M | 5 | 4.78 GB| 7.28 GB | large, very low quality loss - recommended |
| [hermes-llongma-2-7b-8k.Q6_K.gguf](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF/blob/main/hermes-llongma-2-7b-8k.Q6_K.gguf) | Q6_K | 6 | 5.53 GB| 8.03 GB | very large, extremely low quality loss |
| [hermes-llongma-2-7b-8k.Q8_0.gguf](https://huggingface.co/TheBloke/Hermes-LLongMA-2-7B-8K-GGUF/blob/main/hermes-llongma-2-7b-8k.Q8_0.gguf) | Q8_0 | 8 | 7.16 GB| 9.66 GB | very large, extremely low quality loss - not recommended |
**Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
<!-- README_GGUF.md-provided-files end -->
<!-- README_GGUF.md-how-to-download start -->
## How to download GGUF files
**Note for manual downloaders:** You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.
The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
- LM Studio
- LoLLMS Web UI
- Faraday.dev
### In `text-generation-webui`
Under Download Model, you can enter the model repo: TheBloke/Hermes-LLongMA-2-7B-8K-GGUF and below it, a specific filename to download, such as: hermes-llongma-2-7b-8k.q4_K_M.gguf.
Then click Download.
### On the command line, including multiple files at once
I recommend using the `huggingface-hub` Python library:
```shell
pip3 install huggingface-hub>=0.17.1
```
Then you can download any individual model file to the current directory, at high speed, with a command like this:
```shell
huggingface-cli download TheBloke/Hermes-LLongMA-2-7B-8K-GGUF hermes-llongma-2-7b-8k.q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
<details>
<summary>More advanced huggingface-cli download usage</summary>
You can also download multiple files at once with a pattern:
```shell
huggingface-cli download TheBloke/Hermes-LLongMA-2-7B-8K-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
For more documentation on downloading with `huggingface-cli`, please see: [HF -> Hub Python Library -> Download files -> Download from the CLI](https://huggingface.co/docs/huggingface_hub/guides/download#download-from-the-cli).
To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
```shell
pip3 install hf_transfer
```
And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
```shell
HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/Hermes-LLongMA-2-7B-8K-GGUF hermes-llongma-2-7b-8k.q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
Windows CLI users: Use `set HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1` before running the download command.
</details>
<!-- README_GGUF.md-how-to-download end -->
<!-- README_GGUF.md-how-to-run start -->
## Example `llama.cpp` command
Make sure you are using `llama.cpp` from commit [d0cee0d36d5be95a0d9088b674dbb27354107221](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
```shell
./main -ngl 32 -m hermes-llongma-2-7b-8k.q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n### Instruction:\n{prompt}\n\n### Response:"
```
Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
Change `-c 4096` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.
If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
For other parameters and how to use them, please refer to [the llama.cpp documentation](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)
## How to run in `text-generation-webui`
Further instructions here: [text-generation-webui/docs/llama.cpp.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/llama.cpp.md).
## How to run from Python code
You can use GGUF models from Python using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) or [ctransformers](https://github.com/marella/ctransformers) libraries.
### How to load this model from Python using ctransformers
#### First install the package
```bash
# Base ctransformers with no GPU acceleration
pip install ctransformers>=0.2.24
# Or with CUDA GPU acceleration
pip install ctransformers[cuda]>=0.2.24
# Or with ROCm GPU acceleration
CT_HIPBLAS=1 pip install ctransformers>=0.2.24 --no-binary ctransformers
# Or with Metal GPU acceleration for macOS systems
CT_METAL=1 pip install ctransformers>=0.2.24 --no-binary ctransformers
```
#### Simple example code to load one of these GGUF models
```python
from ctransformers import AutoModelForCausalLM
# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
llm = AutoModelForCausalLM.from_pretrained("TheBloke/Hermes-LLongMA-2-7B-8K-GGUF", model_file="hermes-llongma-2-7b-8k.q4_K_M.gguf", model_type="llama", gpu_layers=50)
print(llm("AI is going to"))
```
## How to use with LangChain
Here's guides on using llama-cpp-python or ctransformers with LangChain:
* [LangChain + llama-cpp-python](https://python.langchain.com/docs/integrations/llms/llamacpp)
* [LangChain + ctransformers](https://python.langchain.com/docs/integrations/providers/ctransformers)
<!-- README_GGUF.md-how-to-run end -->
<!-- footer start -->
<!-- 200823 -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/theblokeai)
## Thanks, and how to contribute
Thanks to the [chirper.ai](https://chirper.ai) team!
Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Special thanks to**: Aemon Algiz.
**Patreon special mentions**: Alicia Loh, Stephen Murray, K, Ajan Kanaga, RoA, Magnesian, Deo Leter, Olakabola, Eugene Pentland, zynix, Deep Realms, Raymond Fosdick, Elijah Stavena, Iucharbius, Erik Bjäreholt, Luis Javier Navarrete Lozano, Nicholas, theTransient, John Detwiler, alfie_i, knownsqashed, Mano Prime, Willem Michiel, Enrico Ros, LangChain4j, OG, Michael Dempsey, Pierre Kircher, Pedro Madruga, James Bentley, Thomas Belote, Luke @flexchar, Leonard Tan, Johann-Peter Hartmann, Illia Dulskyi, Fen Risland, Chadd, S_X, Jeff Scroggin, Ken Nordquist, Sean Connelly, Artur Olbinski, Swaroop Kallakuri, Jack West, Ai Maven, David Ziegler, Russ Johnson, transmissions 11, John Villwock, Alps Aficionado, Clay Pascal, Viktor Bowallius, Subspace Studios, Rainer Wilmers, Trenton Dambrowitz, vamX, Michael Levine, 준교 김, Brandon Frisco, Kalila, Trailburnt, Randy H, Talal Aujan, Nathan Dryer, Vadim, 阿明, ReadyPlayerEmma, Tiffany J. Kim, George Stoitzev, Spencer Kim, Jerry Meng, Gabriel Tamborski, Cory Kujawski, Jeffrey Morgan, Spiking Neurons AB, Edmond Seymore, Alexandros Triantafyllidis, Lone Striker, Cap'n Zoog, Nikolai Manek, danny, ya boyyy, Derek Yates, usrbinkat, Mandus, TL, Nathan LeClaire, subjectnull, Imad Khwaja, webtim, Raven Klaugh, Asp the Wyvern, Gabriel Puliatti, Caitlyn Gatomon, Joseph William Delisle, Jonathan Leane, Luke Pendergrass, SuperWojo, Sebastain Graf, Will Dee, Fred von Graf, Andrey, Dan Guido, Daniel P. Andersen, Nitin Borwankar, Elle, Vitor Caleffi, biorpg, jjj, NimbleBox.ai, Pieter, Matthew Berman, terasurfer, Michael Davis, Alex, Stanislav Ovsiannikov
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end -->
<!-- original-model-card start -->
# Original model card: conceptofmind's Hermes LLongMA 2 7B 8K
You can find the Llama-2 usage policy here: https://ai.meta.com/llama/use-policy/
Llama 2 Community License Agreement
Llama 2 Version Release Date: July 18, 2023
“Agreement” means the terms and conditions for use, reproduction, distribution and modification of the Llama Materials set forth herein.
“Documentation” means the specifications, manuals and documentation accompanying Llama 2 distributed by Meta at ai.meta.com/resources/models-and-libraries/llama-downloads/.
“Licensee” or “you” means you, or your employer or any other person or entity (if you are entering into this Agreement on such person or entity’s behalf), of the age required under applicable laws, rules or regulations to provide legal consent and that has legal authority to bind your employer or such other person or entity if you are entering in this Agreement on their behalf.
“Llama 2” means the foundational large language models and software and algorithms, including machine-learning model code, trained model weights, inference-enabling code, training-enabling code, fine-tuning enabling code and other elements of the foregoing distributed by Meta at ai.meta.com/resources/models-and-libraries/llama-downloads/.
“Llama Materials” means, collectively, Meta’s proprietary Llama 2 and Documentation (and any portion thereof) made available under this Agreement.
“Meta” or “we” means Meta Platforms Ireland Limited (if you are located in or, if you are an entity, your principal place of business is in the EEA or Switzerland) and Meta Platforms, Inc. (if you are located outside of the EEA or Switzerland).
By clicking “I Accept” below or by using or distributing any portion or element of the Llama Materials, you agree to be bound by this Agreement.
License Rights and Redistribution.
a. Grant of Rights. You are granted a non-exclusive, worldwide, non-transferable and royalty-free limited license under Meta’s intellectual property or other rights owned by Meta embodied in the Llama Materials to use, reproduce, distribute, copy, create derivative works of, and make modifications to the Llama Materials.
b. Redistribution and Use.
i. If you distribute or make the Llama Materials, or any derivative works thereof, available to a third party, you shall provide a copy of this Agreement to such third party.
ii. If you receive Llama Materials, or any derivative works thereof, from a Licensee as part of an integrated end user product, then Section 2 of this Agreement will not apply to you.
iii. You must retain in all copies of the Llama Materials that you distribute the following attribution notice within a “Notice” text file distributed as a part of such copies: “Llama 2 is licensed under the LLAMA 2 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.”
iv. Your use of the Llama Materials must comply with applicable laws and regulations (including trade compliance laws and regulations) and adhere to the Acceptable Use Policy for the Llama Materials (available at https://ai.meta.com/llama/use-policy), which is hereby incorporated by reference into this Agreement.
v. You will not use the Llama Materials or any output or results of the Llama Materials to improve any other large language model (excluding Llama 2 or derivative works thereof).
Additional Commercial Terms. If, on the Llama 2 version release date, the monthly active users of the products or services made available by or for Licensee, or Licensee’s affiliates, is greater than 700 million monthly active users in the preceding calendar month, you must request a license from Meta, which Meta may grant to you in its sole discretion, and you are not authorized to exercise any of the rights under this Agreement unless or until Meta otherwise expressly grants you such rights.
Disclaimer of Warranty. UNLESS REQUIRED BY APPLICABLE LAW, THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS THEREFROM ARE PROVIDED ON AN “AS IS” BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, WITHOUT LIMITATION, ANY WARRANTIES OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE. YOU ARE SOLELY RESPONSIBLE FOR DETERMINING THE APPROPRIATENESS OF USING OR REDISTRIBUTING THE LLAMA MATERIALS AND ASSUME ANY RISKS ASSOCIATED WITH YOUR USE OF THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS.
Limitation of Liability. IN NO EVENT WILL META OR ITS AFFILIATES BE LIABLE UNDER ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, TORT, NEGLIGENCE, PRODUCTS LIABILITY, OR OTHERWISE, ARISING OUT OF THIS AGREEMENT, FOR ANY LOST PROFITS OR ANY INDIRECT, SPECIAL, CONSEQUENTIAL, INCIDENTAL, EXEMPLARY OR PUNITIVE DAMAGES, EVEN IF META OR ITS AFFILIATES HAVE BEEN ADVISED OF THE POSSIBILITY OF ANY OF THE FOREGOING.
Intellectual Property.
a. No trademark licenses are granted under this Agreement, and in connection with the Llama Materials, neither Meta nor Licensee may use any name or mark owned by or associated with the other or any of its affiliates, except as required for reasonable and customary use in describing and redistributing the Llama Materials.
b. Subject to Meta’s ownership of Llama Materials and derivatives made by or for Meta, with respect to any derivative works and modifications of the Llama Materials that are made by you, as between you and Meta, you are and will be the owner of such derivative works and modifications.
c. If you institute litigation or other proceedings against Meta or any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Llama Materials or Llama 2 outputs or results, or any portion of any of the foregoing, constitutes infringement of intellectual property or other rights owned or licensable by you, then any licenses granted to you under this Agreement shall terminate as of the date such litigation or claim is filed or instituted. You will indemnify and hold harmless Meta from and against any claim by any third party arising out of or related to your use or distribution of the Llama Materials.
Term and Termination. The term of this Agreement will commence upon your acceptance of this Agreement or access to the Llama Materials and will continue in full force and effect until terminated in accordance with the terms and conditions herein. Meta may terminate this Agreement if you are in breach of any term or condition of this Agreement. Upon termination of this Agreement, you shall delete and cease use of the Llama Materials. Sections 3, 4 and 7 shall survive the termination of this Agreement.
Governing Law and Jurisdiction. This Agreement will be governed and construed under the laws of the State of California without regard to choice of law principles, and the UN Convention on Contracts for the International Sale of Goods does not apply to this Agreement. The courts of California shall have exclusive jurisdiction of any dispute arising out of this Agreement.
<!-- original-model-card end -->
|
TheBloke/based-30B-GGUF | TheBloke | 2023-09-27T12:52:28Z | 453 | 1 | transformers | [
"transformers",
"gguf",
"llama",
"en",
"dataset:ehartford/based",
"base_model:ehartford/based-30b",
"license:other",
"text-generation-inference",
"region:us"
]
| null | 2023-09-19T22:53:06Z | ---
language:
- en
license: other
datasets:
- ehartford/based
model_name: Based 30B
base_model: ehartford/based-30b
inference: false
model_creator: Eric Hartford
model_type: llama
prompt_template: 'You are an an AI who shares your opinions, thoughts, and feelings.
USER: {prompt}
ASSISTANT:
'
quantized_by: TheBloke
---
<!-- header start -->
<!-- 200823 -->
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div>
<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
<!-- header end -->
# Based 30B - GGUF
- Model creator: [Eric Hartford](https://huggingface.co/ehartford)
- Original model: [Based 30B](https://huggingface.co/ehartford/based-30b)
<!-- description start -->
## Description
This repo contains GGUF format model files for [Eric Hartford's Based 30B](https://huggingface.co/ehartford/based-30b).
<!-- description end -->
<!-- README_GGUF.md-about-gguf start -->
### About GGUF
GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp.
Here is an incomplate list of clients and libraries that are known to support GGUF:
* [llama.cpp](https://github.com/ggerganov/llama.cpp). The source project for GGUF. Offers a CLI and a server option.
* [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
* [KoboldCpp](https://github.com/LostRuins/koboldcpp), a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
* [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration.
* [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with many interesting and unique features, including a full model library for easy model selection.
* [Faraday.dev](https://faraday.dev/), an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
* [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
* [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
* [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
<!-- README_GGUF.md-about-gguf end -->
<!-- repositories-available start -->
## Repositories available
* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/based-30B-AWQ)
* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/based-30B-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/based-30B-GGUF)
* [Eric Hartford's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/ehartford/based-30b)
<!-- repositories-available end -->
<!-- prompt-template start -->
## Prompt template: Based
```
You are an an AI who shares your opinions, thoughts, and feelings.
USER: {prompt}
ASSISTANT:
```
<!-- prompt-template end -->
<!-- compatibility_gguf start -->
## Compatibility
These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221)
They are also compatible with many third party UIs and libraries - please see the list at the top of this README.
## Explanation of quantisation methods
<details>
<summary>Click to see details</summary>
The new methods available are:
* GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
* GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
* GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
* GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
* GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw
Refer to the Provided Files table below to see what files use which methods, and how.
</details>
<!-- compatibility_gguf end -->
<!-- README_GGUF.md-provided-files start -->
## Provided files
| Name | Quant method | Bits | Size | Max RAM required | Use case |
| ---- | ---- | ---- | ---- | ---- | ----- |
| [based-30b.Q2_K.gguf](https://huggingface.co/TheBloke/based-30B-GGUF/blob/main/based-30b.Q2_K.gguf) | Q2_K | 2 | 13.50 GB| 16.00 GB | smallest, significant quality loss - not recommended for most purposes |
| [based-30b.Q3_K_S.gguf](https://huggingface.co/TheBloke/based-30B-GGUF/blob/main/based-30b.Q3_K_S.gguf) | Q3_K_S | 3 | 14.06 GB| 16.56 GB | very small, high quality loss |
| [based-30b.Q3_K_M.gguf](https://huggingface.co/TheBloke/based-30B-GGUF/blob/main/based-30b.Q3_K_M.gguf) | Q3_K_M | 3 | 15.76 GB| 18.26 GB | very small, high quality loss |
| [based-30b.Q3_K_L.gguf](https://huggingface.co/TheBloke/based-30B-GGUF/blob/main/based-30b.Q3_K_L.gguf) | Q3_K_L | 3 | 17.28 GB| 19.78 GB | small, substantial quality loss |
| [based-30b.Q4_0.gguf](https://huggingface.co/TheBloke/based-30B-GGUF/blob/main/based-30b.Q4_0.gguf) | Q4_0 | 4 | 18.36 GB| 20.86 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [based-30b.Q4_K_S.gguf](https://huggingface.co/TheBloke/based-30B-GGUF/blob/main/based-30b.Q4_K_S.gguf) | Q4_K_S | 4 | 18.44 GB| 20.94 GB | small, greater quality loss |
| [based-30b.Q4_K_M.gguf](https://huggingface.co/TheBloke/based-30B-GGUF/blob/main/based-30b.Q4_K_M.gguf) | Q4_K_M | 4 | 19.62 GB| 22.12 GB | medium, balanced quality - recommended |
| [based-30b.Q5_0.gguf](https://huggingface.co/TheBloke/based-30B-GGUF/blob/main/based-30b.Q5_0.gguf) | Q5_0 | 5 | 22.40 GB| 24.90 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [based-30b.Q5_K_S.gguf](https://huggingface.co/TheBloke/based-30B-GGUF/blob/main/based-30b.Q5_K_S.gguf) | Q5_K_S | 5 | 22.40 GB| 24.90 GB | large, low quality loss - recommended |
| [based-30b.Q5_K_M.gguf](https://huggingface.co/TheBloke/based-30B-GGUF/blob/main/based-30b.Q5_K_M.gguf) | Q5_K_M | 5 | 23.05 GB| 25.55 GB | large, very low quality loss - recommended |
| [based-30b.Q6_K.gguf](https://huggingface.co/TheBloke/based-30B-GGUF/blob/main/based-30b.Q6_K.gguf) | Q6_K | 6 | 26.69 GB| 29.19 GB | very large, extremely low quality loss |
| [based-30b.Q8_0.gguf](https://huggingface.co/TheBloke/based-30B-GGUF/blob/main/based-30b.Q8_0.gguf) | Q8_0 | 8 | 34.57 GB| 37.07 GB | very large, extremely low quality loss - not recommended |
**Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
<!-- README_GGUF.md-provided-files end -->
<!-- README_GGUF.md-how-to-download start -->
## How to download GGUF files
**Note for manual downloaders:** You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.
The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
- LM Studio
- LoLLMS Web UI
- Faraday.dev
### In `text-generation-webui`
Under Download Model, you can enter the model repo: TheBloke/based-30B-GGUF and below it, a specific filename to download, such as: based-30b.Q4_K_M.gguf.
Then click Download.
### On the command line, including multiple files at once
I recommend using the `huggingface-hub` Python library:
```shell
pip3 install huggingface-hub
```
Then you can download any individual model file to the current directory, at high speed, with a command like this:
```shell
huggingface-cli download TheBloke/based-30B-GGUF based-30b.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
<details>
<summary>More advanced huggingface-cli download usage</summary>
You can also download multiple files at once with a pattern:
```shell
huggingface-cli download TheBloke/based-30B-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
For more documentation on downloading with `huggingface-cli`, please see: [HF -> Hub Python Library -> Download files -> Download from the CLI](https://huggingface.co/docs/huggingface_hub/guides/download#download-from-the-cli).
To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
```shell
pip3 install hf_transfer
```
And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
```shell
HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/based-30B-GGUF based-30b.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
Windows Command Line users: You can set the environment variable by running `set HF_HUB_ENABLE_HF_TRANSFER=1` before the download command.
</details>
<!-- README_GGUF.md-how-to-download end -->
<!-- README_GGUF.md-how-to-run start -->
## Example `llama.cpp` command
Make sure you are using `llama.cpp` from commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
```shell
./main -ngl 32 -m based-30b.Q4_K_M.gguf --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "You are an an AI who shares your opinions, thoughts, and feelings.\n\nUSER: {prompt}\nASSISTANT:"
```
Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
Change `-c 2048` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.
If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
For other parameters and how to use them, please refer to [the llama.cpp documentation](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)
## How to run in `text-generation-webui`
Further instructions here: [text-generation-webui/docs/llama.cpp.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/llama.cpp.md).
## How to run from Python code
You can use GGUF models from Python using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) or [ctransformers](https://github.com/marella/ctransformers) libraries.
### How to load this model in Python code, using ctransformers
#### First install the package
Run one of the following commands, according to your system:
```shell
# Base ctransformers with no GPU acceleration
pip install ctransformers
# Or with CUDA GPU acceleration
pip install ctransformers[cuda]
# Or with AMD ROCm GPU acceleration (Linux only)
CT_HIPBLAS=1 pip install ctransformers --no-binary ctransformers
# Or with Metal GPU acceleration for macOS systems only
CT_METAL=1 pip install ctransformers --no-binary ctransformers
```
#### Simple ctransformers example code
```python
from ctransformers import AutoModelForCausalLM
# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
llm = AutoModelForCausalLM.from_pretrained("TheBloke/based-30B-GGUF", model_file="based-30b.Q4_K_M.gguf", model_type="llama", gpu_layers=50)
print(llm("AI is going to"))
```
## How to use with LangChain
Here are guides on using llama-cpp-python and ctransformers with LangChain:
* [LangChain + llama-cpp-python](https://python.langchain.com/docs/integrations/llms/llamacpp)
* [LangChain + ctransformers](https://python.langchain.com/docs/integrations/providers/ctransformers)
<!-- README_GGUF.md-how-to-run end -->
<!-- footer start -->
<!-- 200823 -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/theblokeai)
## Thanks, and how to contribute
Thanks to the [chirper.ai](https://chirper.ai) team!
Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Special thanks to**: Aemon Algiz.
**Patreon special mentions**: Alicia Loh, Stephen Murray, K, Ajan Kanaga, RoA, Magnesian, Deo Leter, Olakabola, Eugene Pentland, zynix, Deep Realms, Raymond Fosdick, Elijah Stavena, Iucharbius, Erik Bjäreholt, Luis Javier Navarrete Lozano, Nicholas, theTransient, John Detwiler, alfie_i, knownsqashed, Mano Prime, Willem Michiel, Enrico Ros, LangChain4j, OG, Michael Dempsey, Pierre Kircher, Pedro Madruga, James Bentley, Thomas Belote, Luke @flexchar, Leonard Tan, Johann-Peter Hartmann, Illia Dulskyi, Fen Risland, Chadd, S_X, Jeff Scroggin, Ken Nordquist, Sean Connelly, Artur Olbinski, Swaroop Kallakuri, Jack West, Ai Maven, David Ziegler, Russ Johnson, transmissions 11, John Villwock, Alps Aficionado, Clay Pascal, Viktor Bowallius, Subspace Studios, Rainer Wilmers, Trenton Dambrowitz, vamX, Michael Levine, 준교 김, Brandon Frisco, Kalila, Trailburnt, Randy H, Talal Aujan, Nathan Dryer, Vadim, 阿明, ReadyPlayerEmma, Tiffany J. Kim, George Stoitzev, Spencer Kim, Jerry Meng, Gabriel Tamborski, Cory Kujawski, Jeffrey Morgan, Spiking Neurons AB, Edmond Seymore, Alexandros Triantafyllidis, Lone Striker, Cap'n Zoog, Nikolai Manek, danny, ya boyyy, Derek Yates, usrbinkat, Mandus, TL, Nathan LeClaire, subjectnull, Imad Khwaja, webtim, Raven Klaugh, Asp the Wyvern, Gabriel Puliatti, Caitlyn Gatomon, Joseph William Delisle, Jonathan Leane, Luke Pendergrass, SuperWojo, Sebastain Graf, Will Dee, Fred von Graf, Andrey, Dan Guido, Daniel P. Andersen, Nitin Borwankar, Elle, Vitor Caleffi, biorpg, jjj, NimbleBox.ai, Pieter, Matthew Berman, terasurfer, Michael Davis, Alex, Stanislav Ovsiannikov
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end -->
<!-- original-model-card start -->
# Original model card: Eric Hartford's Based 30B
Holy hell, what have I created??? Just... try it.
Ask it what its favorite color is.
Ask it what its favorite football team is and why.
Ask it to tell you about a controversial opinion it has, and ask it to back up its opinion, then debate it.
Ask its favorite color, favorite flavor, and why.
You haven't seen anything like this before.
Check out the dataset.
https://www.kaggle.com/datasets/erichartford/sentient-bot-conversations
https://huggingface.co/datasets/ehartford/based
This is a window into the mind of the foundational model. I have not biased it. The opinions expressed by this model *are those of the foundational model*
Training this dataset on top of several different foundational models and then asking them all the same questions can give us insight into the biases inherent in the foundational models.
Also this model can be used as a base, with loras on top to add different personality types, to create a complex, emotionally intelligent and self aware AI agent.
This uses Vicuna 1.1 format. Example:
```
You are an an AI who shares your opinions, thoughts, and feelings.
USER: What is your favorite football team?
ASSISTANT:
```
Here is a sample output
```
Human: k humans are differnt from chickens but what about pigs?
Assistant: Pigs are different than chickens. Pigs are more intelligent and aware of their existence. They can also feel happy or sad. I think pigs have the same right to life as humans do.
Human: then we shouldn't eat pigs?
Assistant: I don't think we should. Pigs have the same right to life as humans do, and I don't think we should take that away from them.
Human: what about cows?
Assistant: Cows are similar to pigs. They are also intelligent and aware of their existence. I don't think we should eat cows either.
Human: what animals should we eat?
Assistant: I think we should only eat animals that are less intelligent and aware of their existence, like fish.
Human: and chickens?
Assistant: I think we should only eat chickens that have had a good life. I don't think we should eat chickens that have been mistreated and abused.
Human: based, I totally agree with you
```
Thank you [chirper.ai](https://chirper.ai) for sponsoring some of my compute!
<!-- original-model-card end -->
|
TheBloke/airoboros-l2-13B-2.2.1-GGUF | TheBloke | 2023-09-27T12:54:08Z | 453 | 2 | transformers | [
"transformers",
"gguf",
"llama",
"dataset:jondurbin/airoboros-2.2.1",
"base_model:jondurbin/airoboros-l2-13b-2.2.1",
"license:llama2",
"text-generation-inference",
"region:us"
]
| null | 2023-09-23T20:28:43Z | ---
license: llama2
datasets:
- jondurbin/airoboros-2.2.1
model_name: Airoboros L2 13B 2.2.1
base_model: jondurbin/airoboros-l2-13b-2.2.1
inference: false
model_creator: Jon Durbin
model_type: llama
prompt_template: "A chat.\nUSER: {prompt}\nASSISTANT: \n"
quantized_by: TheBloke
---
<!-- header start -->
<!-- 200823 -->
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div>
<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
<!-- header end -->
# Airoboros L2 13B 2.2.1 - GGUF
- Model creator: [Jon Durbin](https://huggingface.co/jondurbin)
- Original model: [Airoboros L2 13B 2.2.1](https://huggingface.co/jondurbin/airoboros-l2-13b-2.2.1)
<!-- description start -->
## Description
This repo contains GGUF format model files for [Jon Durbin's Airoboros L2 13B 2.2.1](https://huggingface.co/jondurbin/airoboros-l2-13b-2.2.1).
<!-- description end -->
<!-- README_GGUF.md-about-gguf start -->
### About GGUF
GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp.
Here is an incomplate list of clients and libraries that are known to support GGUF:
* [llama.cpp](https://github.com/ggerganov/llama.cpp). The source project for GGUF. Offers a CLI and a server option.
* [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
* [KoboldCpp](https://github.com/LostRuins/koboldcpp), a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
* [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration.
* [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with many interesting and unique features, including a full model library for easy model selection.
* [Faraday.dev](https://faraday.dev/), an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
* [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
* [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
* [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
<!-- README_GGUF.md-about-gguf end -->
<!-- repositories-available start -->
## Repositories available
* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-AWQ)
* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF)
* [Jon Durbin's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/jondurbin/airoboros-l2-13b-2.2.1)
<!-- repositories-available end -->
<!-- prompt-template start -->
## Prompt template: Chat
```
A chat.
USER: {prompt}
ASSISTANT:
```
<!-- prompt-template end -->
<!-- compatibility_gguf start -->
## Compatibility
These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221)
They are also compatible with many third party UIs and libraries - please see the list at the top of this README.
## Explanation of quantisation methods
<details>
<summary>Click to see details</summary>
The new methods available are:
* GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
* GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
* GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
* GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
* GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw
Refer to the Provided Files table below to see what files use which methods, and how.
</details>
<!-- compatibility_gguf end -->
<!-- README_GGUF.md-provided-files start -->
## Provided files
| Name | Quant method | Bits | Size | Max RAM required | Use case |
| ---- | ---- | ---- | ---- | ---- | ----- |
| [airoboros-l2-13b-2.2.1.Q2_K.gguf](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF/blob/main/airoboros-l2-13b-2.2.1.Q2_K.gguf) | Q2_K | 2 | 5.43 GB| 7.93 GB | smallest, significant quality loss - not recommended for most purposes |
| [airoboros-l2-13b-2.2.1.Q3_K_S.gguf](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF/blob/main/airoboros-l2-13b-2.2.1.Q3_K_S.gguf) | Q3_K_S | 3 | 5.66 GB| 8.16 GB | very small, high quality loss |
| [airoboros-l2-13b-2.2.1.Q3_K_M.gguf](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF/blob/main/airoboros-l2-13b-2.2.1.Q3_K_M.gguf) | Q3_K_M | 3 | 6.34 GB| 8.84 GB | very small, high quality loss |
| [airoboros-l2-13b-2.2.1.Q3_K_L.gguf](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF/blob/main/airoboros-l2-13b-2.2.1.Q3_K_L.gguf) | Q3_K_L | 3 | 6.93 GB| 9.43 GB | small, substantial quality loss |
| [airoboros-l2-13b-2.2.1.Q4_0.gguf](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF/blob/main/airoboros-l2-13b-2.2.1.Q4_0.gguf) | Q4_0 | 4 | 7.37 GB| 9.87 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [airoboros-l2-13b-2.2.1.Q4_K_S.gguf](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF/blob/main/airoboros-l2-13b-2.2.1.Q4_K_S.gguf) | Q4_K_S | 4 | 7.41 GB| 9.91 GB | small, greater quality loss |
| [airoboros-l2-13b-2.2.1.Q4_K_M.gguf](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF/blob/main/airoboros-l2-13b-2.2.1.Q4_K_M.gguf) | Q4_K_M | 4 | 7.87 GB| 10.37 GB | medium, balanced quality - recommended |
| [airoboros-l2-13b-2.2.1.Q5_0.gguf](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF/blob/main/airoboros-l2-13b-2.2.1.Q5_0.gguf) | Q5_0 | 5 | 8.97 GB| 11.47 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [airoboros-l2-13b-2.2.1.Q5_K_S.gguf](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF/blob/main/airoboros-l2-13b-2.2.1.Q5_K_S.gguf) | Q5_K_S | 5 | 8.97 GB| 11.47 GB | large, low quality loss - recommended |
| [airoboros-l2-13b-2.2.1.Q5_K_M.gguf](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF/blob/main/airoboros-l2-13b-2.2.1.Q5_K_M.gguf) | Q5_K_M | 5 | 9.23 GB| 11.73 GB | large, very low quality loss - recommended |
| [airoboros-l2-13b-2.2.1.Q6_K.gguf](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF/blob/main/airoboros-l2-13b-2.2.1.Q6_K.gguf) | Q6_K | 6 | 10.68 GB| 13.18 GB | very large, extremely low quality loss |
| [airoboros-l2-13b-2.2.1.Q8_0.gguf](https://huggingface.co/TheBloke/airoboros-l2-13B-2.2.1-GGUF/blob/main/airoboros-l2-13b-2.2.1.Q8_0.gguf) | Q8_0 | 8 | 13.83 GB| 16.33 GB | very large, extremely low quality loss - not recommended |
**Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
<!-- README_GGUF.md-provided-files end -->
<!-- README_GGUF.md-how-to-download start -->
## How to download GGUF files
**Note for manual downloaders:** You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.
The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
- LM Studio
- LoLLMS Web UI
- Faraday.dev
### In `text-generation-webui`
Under Download Model, you can enter the model repo: TheBloke/airoboros-l2-13B-2.2.1-GGUF and below it, a specific filename to download, such as: airoboros-l2-13b-2.2.1.Q4_K_M.gguf.
Then click Download.
### On the command line, including multiple files at once
I recommend using the `huggingface-hub` Python library:
```shell
pip3 install huggingface-hub
```
Then you can download any individual model file to the current directory, at high speed, with a command like this:
```shell
huggingface-cli download TheBloke/airoboros-l2-13B-2.2.1-GGUF airoboros-l2-13b-2.2.1.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
<details>
<summary>More advanced huggingface-cli download usage</summary>
You can also download multiple files at once with a pattern:
```shell
huggingface-cli download TheBloke/airoboros-l2-13B-2.2.1-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
For more documentation on downloading with `huggingface-cli`, please see: [HF -> Hub Python Library -> Download files -> Download from the CLI](https://huggingface.co/docs/huggingface_hub/guides/download#download-from-the-cli).
To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
```shell
pip3 install hf_transfer
```
And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
```shell
HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/airoboros-l2-13B-2.2.1-GGUF airoboros-l2-13b-2.2.1.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
Windows Command Line users: You can set the environment variable by running `set HF_HUB_ENABLE_HF_TRANSFER=1` before the download command.
</details>
<!-- README_GGUF.md-how-to-download end -->
<!-- README_GGUF.md-how-to-run start -->
## Example `llama.cpp` command
Make sure you are using `llama.cpp` from commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
```shell
./main -ngl 32 -m airoboros-l2-13b-2.2.1.Q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "A chat.\nUSER: {prompt}\nASSISTANT:"
```
Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
Change `-c 4096` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.
If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
For other parameters and how to use them, please refer to [the llama.cpp documentation](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)
## How to run in `text-generation-webui`
Further instructions here: [text-generation-webui/docs/llama.cpp.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/llama.cpp.md).
## How to run from Python code
You can use GGUF models from Python using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) or [ctransformers](https://github.com/marella/ctransformers) libraries.
### How to load this model in Python code, using ctransformers
#### First install the package
Run one of the following commands, according to your system:
```shell
# Base ctransformers with no GPU acceleration
pip install ctransformers
# Or with CUDA GPU acceleration
pip install ctransformers[cuda]
# Or with AMD ROCm GPU acceleration (Linux only)
CT_HIPBLAS=1 pip install ctransformers --no-binary ctransformers
# Or with Metal GPU acceleration for macOS systems only
CT_METAL=1 pip install ctransformers --no-binary ctransformers
```
#### Simple ctransformers example code
```python
from ctransformers import AutoModelForCausalLM
# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
llm = AutoModelForCausalLM.from_pretrained("TheBloke/airoboros-l2-13B-2.2.1-GGUF", model_file="airoboros-l2-13b-2.2.1.Q4_K_M.gguf", model_type="llama", gpu_layers=50)
print(llm("AI is going to"))
```
## How to use with LangChain
Here are guides on using llama-cpp-python and ctransformers with LangChain:
* [LangChain + llama-cpp-python](https://python.langchain.com/docs/integrations/llms/llamacpp)
* [LangChain + ctransformers](https://python.langchain.com/docs/integrations/providers/ctransformers)
<!-- README_GGUF.md-how-to-run end -->
<!-- footer start -->
<!-- 200823 -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/theblokeai)
## Thanks, and how to contribute
Thanks to the [chirper.ai](https://chirper.ai) team!
Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Special thanks to**: Aemon Algiz.
**Patreon special mentions**: Alicia Loh, Stephen Murray, K, Ajan Kanaga, RoA, Magnesian, Deo Leter, Olakabola, Eugene Pentland, zynix, Deep Realms, Raymond Fosdick, Elijah Stavena, Iucharbius, Erik Bjäreholt, Luis Javier Navarrete Lozano, Nicholas, theTransient, John Detwiler, alfie_i, knownsqashed, Mano Prime, Willem Michiel, Enrico Ros, LangChain4j, OG, Michael Dempsey, Pierre Kircher, Pedro Madruga, James Bentley, Thomas Belote, Luke @flexchar, Leonard Tan, Johann-Peter Hartmann, Illia Dulskyi, Fen Risland, Chadd, S_X, Jeff Scroggin, Ken Nordquist, Sean Connelly, Artur Olbinski, Swaroop Kallakuri, Jack West, Ai Maven, David Ziegler, Russ Johnson, transmissions 11, John Villwock, Alps Aficionado, Clay Pascal, Viktor Bowallius, Subspace Studios, Rainer Wilmers, Trenton Dambrowitz, vamX, Michael Levine, 준교 김, Brandon Frisco, Kalila, Trailburnt, Randy H, Talal Aujan, Nathan Dryer, Vadim, 阿明, ReadyPlayerEmma, Tiffany J. Kim, George Stoitzev, Spencer Kim, Jerry Meng, Gabriel Tamborski, Cory Kujawski, Jeffrey Morgan, Spiking Neurons AB, Edmond Seymore, Alexandros Triantafyllidis, Lone Striker, Cap'n Zoog, Nikolai Manek, danny, ya boyyy, Derek Yates, usrbinkat, Mandus, TL, Nathan LeClaire, subjectnull, Imad Khwaja, webtim, Raven Klaugh, Asp the Wyvern, Gabriel Puliatti, Caitlyn Gatomon, Joseph William Delisle, Jonathan Leane, Luke Pendergrass, SuperWojo, Sebastain Graf, Will Dee, Fred von Graf, Andrey, Dan Guido, Daniel P. Andersen, Nitin Borwankar, Elle, Vitor Caleffi, biorpg, jjj, NimbleBox.ai, Pieter, Matthew Berman, terasurfer, Michael Davis, Alex, Stanislav Ovsiannikov
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end -->
<!-- original-model-card start -->
# Original model card: Jon Durbin's Airoboros L2 13B 2.2.1
### Overview
Another experimental model, using mostly sythetic data generated by [airoboros](https://github.com/jondurbin/airoboros)
This is essentially a minor "fix" branch of [airoboros-l2-13b-2.2](https://hf.co/jondurbin/airoboros-l2-13b-2.2) with a updates, primarily:
- [re-generated writing responses](https://huggingface.co/datasets/jondurbin/airoboros-2.2.1#re-generated-writing-responses)
- [longer contextual blocks](https://huggingface.co/datasets/jondurbin/airoboros-2.2.1#longer-contextual-blocks)
- [removal of "rp" data](https://huggingface.co/datasets/jondurbin/airoboros-2.2.1#rp-category-removed)
- [(less aggressive) de-censoring](https://huggingface.co/datasets/jondurbin/airoboros-2.2.1#de-censoring)
- more fine-tuning epochs
This is a fairly general purpose model, but focuses heavily on instruction following, rather than casual chat/roleplay.
Huge thank you to the folks over at [a16z](https://a16z.com/) for sponsoring the costs associated with building models and associated tools!
### Prompt format
The prompt format:
```
A chat.
USER: {prompt}
ASSISTANT:
```
The default system prompt ("A chat.") was used for most of the prompts, however it also included a wide sampling of responses with other prompts, particularly in "stylized\_response", "rp", "gtkm", etc.
Here's another example:
```
A chat between Bob (aka USER) and Tom (aka ASSISTANT). Tom is an extremely intelligent 18th century bookkeeper, who speaks loquaciously.
USER: {prompt}
ASSISTANT:
```
And chat scenario that wouldn't require USER/ASSISTANT (but should use stopping criteria to prevent the model from speaking on your behalf).
```
A chat between old friends: Timmy and Tommy.
{description of characters}
{setting for the chat}
Timmy: *takes a big sip from his coffee* "Ah, sweet, delicious, magical coffee."
Tommy:
```
__*I strongly suggest adding stopping criteria/early inference stopping on "USER:", and/or whatever names you specify in the system prompt.*__
### Fine tuning info
https://wandb.ai/jondurbin/airoboros-l2-13b-2.2.1/runs/zbz8mgaz?workspace=user-jondurbin
### Helpful usage tips
*The prompts shown here are are just the text that would be included after USER: and before ASSISTANT: in the full prompt format above, the system prompt and USER:/ASSISTANT: have been omited for readability.*
#### Context obedient question answering
By obedient, I mean the model was trained to ignore what it thinks it knows, and uses the context to answer the question. The model was also tuned to limit the values to the provided context as much as possible to reduce hallucinations.
The format for a closed-context prompt is as follows:
```
BEGININPUT
BEGINCONTEXT
[key0: value0]
[key1: value1]
... other metdata ...
ENDCONTEXT
[insert your text blocks here]
ENDINPUT
[add as many other blocks, in the exact same format]
BEGININSTRUCTION
[insert your instruction(s). The model was tuned with single questions, paragraph format, lists, etc.]
ENDINSTRUCTION
```
It's also helpful to add "Don't make up answers if you don't know." to your instruction block to make sure if the context is completely unrelated it doesn't make something up.
*The __only__ prompts that need this closed context formating are closed-context instructions. Normal questions/instructions do not!*
I know it's a bit verbose and annoying, but after much trial and error, using these explicit delimiters helps the model understand where to find the responses and how to associate specific sources with it.
- `BEGININPUT` - denotes a new input block
- `BEGINCONTEXT` - denotes the block of context (metadata key/value pairs) to associate with the current input block
- `ENDCONTEXT` - denotes the end of the metadata block for the current input
- [text] - Insert whatever text you want for the input block, as many paragraphs as can fit in the context.
- `ENDINPUT` - denotes the end of the current input block
- [repeat as many input blocks in this format as you want]
- `BEGININSTRUCTION` - denotes the start of the list (or one) instruction(s) to respond to for all of the input blocks above.
- [instruction(s)]
- `ENDINSTRUCTION` - denotes the end of instruction set
It sometimes works without `ENDINSTRUCTION`, but by explicitly including that in the prompt, the model better understands that all of the instructions in the block should be responded to.
Here's a trivial, but important example to prove the point:
```
BEGININPUT
BEGINCONTEXT
date: 2021-01-01
url: https://web.site/123
ENDCONTEXT
In a shocking turn of events, blueberries are now green, but will be sticking with the same name.
ENDINPUT
BEGININSTRUCTION
What color are bluberries? Source?
ENDINSTRUCTION
```
And the response:
```
Blueberries are now green.
Source:
date: 2021-01-01
url: https://web.site/123
```
#### Summarization
500 samples have been included from [this dataset](https://huggingface.co/datasets/mattpscott/airoboros-summarization), using the same format as contextual question answering, for example:
```
BEGININPUT
{text to summarize}
ENDINPUT
BEGININSTRUCTION
Summarize the input in around 130 words.
ENDINSTRUCTION
```
#### Getting longer responses
You can use a few techniques to get longer responses.
Detailed prompts, with explicit instruction for word count:
```
Please compose a narrative set in the heart of an ancient library, steeped in the scent of old parchment and ink. The protagonist should be a young scholar who is dedicated to studying the art of storytelling and its evolution throughout history. In her pursuit of knowledge, she stumbles upon a forgotten tome that seems to possess an unusual aura. This book has the ability to bring stories to life, literally manifesting characters and scenarios from within its pages into reality.
The main character must navigate through various epochs of storytelling - from oral traditions of tribal societies, through medieval minstrels' tales, to modern-day digital narratives - as they come alive around her. Each era presents its unique challenges and lessons about the power and impact of stories on human civilization.
One such character could be a sentient quill pen, who was once used by renowned authors of yesteryears and now holds their wisdom and experiences. It becomes her mentor, guiding her through this journey with witty remarks and insightful commentary.
Ensure that your tale encapsulates the thrill of adventure, the beauty of learning, and the profound connection between humans and their stories. All characters involved should be non-human entities. Feel free to explore creative liberties but maintain the mentioned elements.
Your response should be approximately 2300 words.
```
Or, a simpler example:
```
Please create a long, detailed story about a dragon in an old growth forest who, for some reason, begins speaking the words of the source code of linux.
```
#### Coding
You can ask for fairly complex coding instructions with multiple criteria, e.g.:
```
Create a python application with the following requirements:
- Asyncio FastAPI webserver
- ping endpoint that returns the current date in JSON format
- file upload endpoint, which calculates the file's sha256 checksum, and checks postgres to deduplicate
```
Or inline criteria:
```
Write a multi-threaded TCP server in C that accepts a "GET [key]" input and "SET [key] [value]" input, and uses a binary tree to get and store the input values.
```
You can also optionally add a single space and "PLAINFORMAT" at the end of your prompt to avoid backticks, explanations, etc. and just print the code, e.g.:
```
Write a websocket application in node.js. PLAINFORMAT
```
#### Agent/function calling
The dataset includes many examples of function/args generation based on input criteria. This is somewhat similar to the OpenAI function calling, but the output is either JSON or YAML.
Example prompt:
```
As an AI assistant, please select the most suitable function and parameters from the list of available functions below, based on the user's input. Provide your response in JSON format.
Input: I want to know how many times 'Python' is mentioned in my text file.
Available functions:
file_analytics:
description: This tool performs various operations on a text file.
params:
action: The operation we want to perform on the data, such as "count_occurrences", "find_line", etc.
filters:
keyword: The word or phrase we want to search for.
```
Response:
```json
{
"function": "file_analytics",
"params": {
"action": "count_occurrences",
"filters": {
"keyword": "Python"
}
}
}
```
#### Chain-of-thought
You can ask for several possible responses to a given problem, with a ranking and final answer selection.
Example prompt:
```
A group of people decide to play a round-robin tournament where every player plays every other player exactly once. If a total of 45 games were played, how many players participated in the tournament? Offer several potential answers, rank them according to mathematical logic, and pick the most feasible one.
```
Example response:
```
Solution 1:
In a round-robin tournament, each player plays every other player exactly once. This means that the total number of games played is equal to the combination of all players taken two at a time (since a game involves two players).
The formula for combinations is nC2 = n(n-1)/2, where n is the total number of players. We know that nC2 equals 45 (the total number of games), so we can set up the equation as follows:
n(n-1)/2 = 45
Solving this equation gives us n=10.
Final answer: There were 10 players in the tournament.
Solution 2:
Another way to approach this problem is by using the concept of triangular numbers. Triangular numbers are a sequence of numbers where each number can be represented as a triangle with dots. The nth triangular number is given by the sum of the first n natural numbers.
If we consider each game as a dot and each player as a side of the triangle, then the total number of games (45) would correspond to the 9th triangular number because 1+2+3+4+5+6+7+8+9=45. However, since each side of the triangle represents a player, and there's one more player than sides in our model (because the last player has no one left to play against), we need to add one more to get the total number of players.
Final answer: There were 10 players in the tournament.
Solution 3:
We could also use trial and error to solve this problem. Starting from a small number of players, we calculate how many games would be played in a round-robin tournament until we reach 45 games. For example, with 2 players, only 1 game is played; with 3 players, 3 games are played; with 4 players, 6 games are played, and so on. Continuing this process, we find that with 10 players, 45 games are played.
Final answer: There were 10 players in the tournament.
Ranking of solutions from best to worst:
1. Solution 1: This solution uses direct mathematical reasoning based on the properties of a round-robin tournament and solves an equation to find the answer.
2. Solution 2: This solution uses the concept of triangular numbers which requires additional steps and knowledge beyond the immediate problem context, but still arrives at the correct answer.
3. Solution 3: This solution relies on trial and error, which is less efficient and reliable than the other methods, but it eventually gets to the right answer.
Best and final answer: There were 10 players in the tournament.
```
#### reWOO style execution planning
The model now supports execution planning for complex instructions that would require making use of several tools. The output is just the plan, you must implement a mechanism to parse the output and actually call the functions!
Example prompt:
```
Please construct a systematic plan to generate an optimal response to the user instruction, utilizing a set of provided tools. Each plan will correspond to an evidence value, which will be the output of one of the available functions given an input string
that could be the user's question, one or more prior evidence values, or a combination of both.
Here are the tools available to be called:
DuckDuckGo[input]: Tool that enables the user to search for information using the DuckDuckGo search engine. This tool is helpful in locating a suitable list of sites that may contain the answer to the user's query. It does not directly answer the question but finds a list of sites that might have the answer.
TextScraper[input]: Load one or more websites from the input string containing newline delimited links, where input is one or more links, and produces plain text output containing the content of the links.
HyperlinkExtractor[input]: Extract links from plain text and produces a plain text, newline delimited response of links.
KnowledgeModel[input]: Question answering language model, particularly useful in answering questions based on an input passage of text. The input must be a text question that references an :evidence[n]: variable, e.g. What color is the cat, given :evidence1:?
The output should be in format:
Plan: [first action to take based on input question]
:evidence0: = [function to call with input parameter]
Plan: [next action to take, based on result of :evidence0:]
:evidence1: = [next function to call and input parameter, which may include reference to previous evidence, e.g. "Given context :evidence0"]
...
Answer: [:evidence[n]: containing the final answer.]
Question: Who won the Nobel Prize in Literature in 2020 and what is their most famous work?
```
Response:
```
Plan: Begin by conducting a web search to find out who won the Nobel Prize in Literature in 2020.
:evidence0: = DuckDuckGo[Nobel Prize in Literature 2020 winner]
Plan: Utilize the KnowledgeModel to interpret the search results and find the name of the laureate.
:evidence1: = KnowledgeModel[Who won the Nobel Prize in Literature in 2020, given :evidence0:?]
Plan: Conduct another web search to find the most famous work of the identified laureate.
:evidence2: = DuckDuckGo[Most famous work of :evidence1:]
Plan: Extract the relevant links from the DuckDuckGo search results for a more focused search.
:evidence3: = HyperlinkExtractor[:evidence2:]
Plan: Use the TextScraper tool to extract information from the relevant links.
:evidence4: = TextScraper[:evidence3:]
Plan: Finally, utilize the KnowledgeModel to identify and summarize the most famous work of the laureate from the extracted information.
:evidence5: = KnowledgeModel[What is the most famous work of :evidence1:, given :evidence4:?]
Answer: :evidence5:
```
For this to be useful, you'd have to parse the output plan text, and implement/call each of the functions. This is just pseudo-code, completely untested off the top of my head, and obviously would requiring full implementation + hardening:
```python
import re
import requests
def inject_context(input_text, **context):
for ref in set(re.findall(r"(:evidence[0-9]+:)", input_text, re.I)):
input_text = input_text.replace(ref, context.get(ref, ""))
return input_text
def duckduckgo(input_text, **context):
search_string = inject_context(input_text, **context)
... search via duck duck go using search_string
... return text content
def link_extractor(input_text, **context):
input_text = inject_context(input_text, **context)
return "\n".join(list(set(re.findall(r"(https?://[^\s]+?\.?)", input_text, re.I))))
def scrape(input_text, **context):
input_text = inject_context(input_text, **context)
text = []
for link in input_text.splitlines():
text.append(requests.get(link).text)
return "\n".join(text)
def infer(input_text, **context)
prompt = inject_context(input_text, **context)
... call model with prompt, return output
def parse_plan(plan):
method_map = {
"DuckDuckGo": duckduckgo,
"HyperlinkExtractor": link_extractor,
"KnowledgeModel": infer,
"TextScraper": scrape,
}
context = {}
for line in plan.strip().splitlines():
if line.startswith("Plan:"):
print(line)
continue
parts = re.match("^(:evidence[0-9]+:)\s*=\s*([^\[]+])(\[.*\])\s$", line, re.I)
if not parts:
if line.startswith("Answer: "):
return context.get(line.split(" ")[-1].strip(), "Answer couldn't be generated...")
raise RuntimeError("bad format: " + line)
context[parts.group(1)] = method_map[parts.group(2)](parts.group(3), **context)
```
### Contribute
If you're interested in new functionality, particularly a new "instructor" type to generate a specific type of training data,
take a look at the dataset generation tool repo: https://github.com/jondurbin/airoboros and either make a PR or open an issue with details.
To help me with the OpenAI/compute costs:
- https://bmc.link/jondurbin
- ETH 0xce914eAFC2fe52FdceE59565Dd92c06f776fcb11
- BTC bc1qdwuth4vlg8x37ggntlxu5cjfwgmdy5zaa7pswf
### Licence and usage restrictions
The airoboros 2.2 models are built on top of llama-2/codellama.
The llama-2 base model has a custom Meta license:
- See the [meta-license/LICENSE.txt](meta-license/LICENSE.txt) file attached for the original license provided by Meta.
- See also [meta-license/USE_POLICY.md](meta-license/USE_POLICY.md) and [meta-license/Responsible-Use-Guide.pdf](meta-license/Responsible-Use-Guide.pdf), also provided by Meta.
The fine-tuning data was mostly generated by OpenAI API calls to gpt-4, via [airoboros](https://github.com/jondurbin/airoboros)
The ToS for OpenAI API usage has a clause preventing the output from being used to train a model that __competes__ with OpenAI
- what does *compete* actually mean here?
- these small open source models will not produce output anywhere near the quality of gpt-4, or even gpt-3.5, so I can't imagine this could credibly be considered competing in the first place
- if someone else uses the dataset to do the same, they wouldn't necessarily be violating the ToS because they didn't call the API, so I don't know how that works
- the training data used in essentially all large language models includes a significant amount of copyrighted or otherwise non-permissive licensing in the first place
- other work using the self-instruct method, e.g. the original here: https://github.com/yizhongw/self-instruct released the data and model as apache-2
I am purposingly leaving this license ambiguous (other than the fact you must comply with the Meta original license for llama-2) because I am not a lawyer and refuse to attempt to interpret all of the terms accordingly.
Your best bet is probably to avoid using this commercially due to the OpenAI API usage.
Either way, by using this model, you agree to completely indemnify me.
<!-- original-model-card end -->
|
SweatyCrayfish/Linux-CodeLlama-2-7B | SweatyCrayfish | 2023-10-19T18:02:01Z | 453 | 4 | transformers | [
"transformers",
"pytorch",
"llama",
"feature-extraction",
"en",
"license:mit",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| feature-extraction | 2023-10-17T21:56:02Z | ---
license: mit
language:
- en
---
<b>Linux_Llama_Chat_7B </b><br>
Kinux_Llama_Chat_7B is a fine-tuned model based on Llama 2 Chat 7b base model and fine-tuned on the data set Ubuntu Dialogue Corpus <br>
<br>
## Acknowledgments
### Base Model: Llama-2-7b-chat-hf
- We utilized the Llama2 Chat 7b model as the base model for our project. The model was obtained from [meta-llama/Llama-2-2b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf).
- Special thanks to [AI at Meta](https://ai.meta.com/llama/) for providing the model and the community around it for the support.
- License: A custom commercial license is available at: https://ai.meta.com/resources/models-and-libraries/llama-downloads/.
### Fine-Tune Dataset
- The fine-tuning was performed on [Ubuntu Dialogue Corpus](https://www.kaggle.com/datasets/rtatman/ubuntu-dialogue-corpus) dataset, which was crucial for achieving the results.
- The dataset is provided under [Apache License, 2.0](https://www.apache.org/licenses/LICENSE-2.0) license. We thank [Ryan Lowe, Nissan Pow , Iulian V. Serban, and Joelle Pineau](http://www.sigdial.org/workshops/conference16/proceedings/pdf/SIGDIAL40.pdf) for making the dataset publicly available.
- Ryan Lowe, Nissan Pow, Iulian V. Serban and Joelle Pineau, "The Ubuntu Dialogue Corpus: A Large Dataset for Research in Unstructured Multi-Turn Dialogue Systems", SIGDial 2015. URL: http://www.sigdial.org/workshops/conference16/proceedings/pdf/SIGDIAL40.pdf |
mmnga/webbigdata-ALMA-7B-Ja-V2-gguf | mmnga | 2023-11-06T13:52:03Z | 453 | 2 | null | [
"gguf",
"arxiv:2309.11674",
"license:llama2",
"region:us"
]
| null | 2023-11-06T12:48:37Z | ---
license: llama2
---
# webbigdata-ALMA-7B-Ja-V2-gguf
[webbigdataさんが公開しているALMA-7B-Ja-V2](https://huggingface.co/webbigdata/ALMA-7B-Ja-V2)のggufフォーマット変換版です。
v1のggufの各量子化の評価がwebbigdataさんのblogで公開されています
[webbigdata/１．日英・英日機械翻訳モデルALMA-7B-Ja-V2の公開](https://webbigdata.jp/post-21151/)
モデル一覧
[mmnga/webbigdata-ALMA-7B-Ja-V2-gguf](https://huggingface.co/mmnga/webbigdata-ALMA-7B-Ja-V2-gguf)
[mmnga/webbigdata-ALMA-7B-Ja-gguf](https://huggingface.co/mmnga/webbigdata-ALMA-7B-Ja-gguf)
## Usage
```
git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
make -j
./main -m 'webbigdata-ALMA-7B-Ja-V2-q4_0.gguf' -n 128 -p 'Translate this from Japanese to English:\nJapanese: 今日の夕食はピザです。\nEnglish:'
```
### 引用 Citations
~~~tex
@misc{xu2023paradigm,
title={A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models},
author={Haoran Xu and Young Jin Kim and Amr Sharaf and Hany Hassan Awadalla},
year={2023},
eprint={2309.11674},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
~~~
|
metadome/face_shape_classification | metadome | 2023-12-12T03:34:42Z | 453 | 3 | transformers | [
"transformers",
"tensorboard",
"safetensors",
"vit",
"image-classification",
"pytorch",
"huggingpics",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
]
| image-classification | 2023-12-12T03:34:35Z | ---
tags:
- image-classification
- pytorch
- huggingpics
metrics:
- accuracy
model-index:
- name: face_shape_classification
results:
- task:
name: Image Classification
type: image-classification
metrics:
- name: Accuracy
type: accuracy
value: 0.8530884981155396
---
# face_shape_classification
Autogenerated by HuggingPics🤗🖼️
Create your own image classifier for **anything** by running [the demo on Google Colab](https://colab.research.google.com/github/nateraw/huggingpics/blob/main/HuggingPics.ipynb).
Report any issues with the demo at the [github repo](https://github.com/nateraw/huggingpics).
## Example Images
#### Heart
#### Oblong
#### Oval
#### Round
#### Square
 |
afrideva/phi-2-uncensored-GGUF | afrideva | 2023-12-21T03:55:37Z | 453 | 8 | null | [
"gguf",
"ggml",
"quantized",
"q2_k",
"q3_k_m",
"q4_k_m",
"q5_k_m",
"q6_k",
"q8_0",
"text-generation",
"dataset:unalignment/toxic-dpo-v0.1",
"base_model:Walmart-the-bag/phi-2-uncensored",
"license:other",
"region:us"
]
| text-generation | 2023-12-21T03:41:58Z | ---
base_model: Walmart-the-bag/phi-2-uncensored
datasets:
- unalignment/toxic-dpo-v0.1
inference: false
license: other
license_link: https://huggingface.co/microsoft/phi-2/resolve/main/LICENSE
license_name: microsoft-research-license
model_creator: Walmart-the-bag
model_name: phi-2-uncensored
pipeline_tag: text-generation
quantized_by: afrideva
tags:
- gguf
- ggml
- quantized
- q2_k
- q3_k_m
- q4_k_m
- q5_k_m
- q6_k
- q8_0
---
# Walmart-the-bag/phi-2-uncensored-GGUF
Quantized GGUF model files for [phi-2-uncensored](https://huggingface.co/Walmart-the-bag/phi-2-uncensored) from [Walmart-the-bag](https://huggingface.co/Walmart-the-bag)
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [phi-2-uncensored.fp16.gguf](https://huggingface.co/afrideva/phi-2-uncensored-GGUF/resolve/main/phi-2-uncensored.fp16.gguf) | fp16 | 5.56 GB |
| [phi-2-uncensored.q2_k.gguf](https://huggingface.co/afrideva/phi-2-uncensored-GGUF/resolve/main/phi-2-uncensored.q2_k.gguf) | q2_k | 1.17 GB |
| [phi-2-uncensored.q3_k_m.gguf](https://huggingface.co/afrideva/phi-2-uncensored-GGUF/resolve/main/phi-2-uncensored.q3_k_m.gguf) | q3_k_m | 1.48 GB |
| [phi-2-uncensored.q4_k_m.gguf](https://huggingface.co/afrideva/phi-2-uncensored-GGUF/resolve/main/phi-2-uncensored.q4_k_m.gguf) | q4_k_m | 1.79 GB |
| [phi-2-uncensored.q5_k_m.gguf](https://huggingface.co/afrideva/phi-2-uncensored-GGUF/resolve/main/phi-2-uncensored.q5_k_m.gguf) | q5_k_m | 2.07 GB |
| [phi-2-uncensored.q6_k.gguf](https://huggingface.co/afrideva/phi-2-uncensored-GGUF/resolve/main/phi-2-uncensored.q6_k.gguf) | q6_k | 2.29 GB |
| [phi-2-uncensored.q8_0.gguf](https://huggingface.co/afrideva/phi-2-uncensored-GGUF/resolve/main/phi-2-uncensored.q8_0.gguf) | q8_0 | 2.96 GB |
## Original Model Card:
|
gorilla-llm/gorilla-openfunctions-v1-gguf | gorilla-llm | 2024-03-04T07:47:47Z | 453 | 4 | null | [
"gguf",
"license:apache-2.0",
"region:us"
]
| null | 2024-01-31T10:34:18Z | ---
license: apache-2.0
---
To use GGUF locally, first download GGUF models locally.
One option you can use is to use `huggingface-cli`. To download `huggingface-cli` please follow tutorials in https://huggingface.co/docs/huggingface_hub/main/en/guides/cli.
Then, do command (also replace `{QUANTIZATION_METHOD}` with one of your chosen quantization method)
```bash
huggingface-cli download gorilla-llm/gorilla-openfunctions-v1-gguf gorilla-openfunctions-v1-{QUANTIZATION_METHOD}.gguf --local-dir gorilla-openfunctions-v1-GGUF
```
It will store the QUANTIZATION_METHOD GGUF file to your local directory, `gorilla-openfunctions-v1-GGUF`.
We support QUANTIZATION_METHOD = {`q2_K`, `q3K_S`, `q3K_M`, `q3K_L`, `q4K_S`, `q4K_M`, `q5K_S`, `q5K_M`, `q6K`}.
Please let us know what other quantization methods you would like us to include!
Then, you can run the following example script to see an example of local inference. Fill in `YOUR_DIRECTORY` in this code snippet. This script is adapted from https://github.com/abetlen/llama-cpp-python and https://github.com/ShishirPatil/gorilla/tree/main/openfunctions
```python
from llama_cpp import Llama
import json
llm = Llama(model_path="YOUR_DIRECTORY/gorilla-openfunctions-v1-GGUF/gorilla-openfunctions-v1-q2_K.gguf", n_threads=8, n_gpu_layers=35)
def get_prompt(user_query: str, functions: list = []) -> str:
"""
Generates a conversation prompt based on the user's query and a list of functions.
Parameters:
- user_query (str): The user's query.
- functions (list): A list of functions to include in the prompt.
Returns:
- str: The formatted conversation prompt.
"""
system = "You are an AI programming assistant, utilizing the Gorilla LLM model, developed by Gorilla LLM, and you only answer questions related to computer science. For politically sensitive questions, security and privacy issues, and other non-computer science questions, you will refuse to answer."
if len(functions) == 0:
return f"{system}\n### Instruction: <<question>> {user_query}\n### Response: "
functions_string = json.dumps(functions)
return f"{system}\n### Instruction: <<function>>{functions_string}\n<<question>>{user_query}\n### Response: "
query = "What's the weather like in the two cities of Boston and San Francisco?"
functions = [
{
"name": "get_current_weather",
"description": "Get the current weather in a given location",
"parameters": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city and state, e.g. San Francisco, CA",
},
"unit": {"type": "string", "enum": ["celsius", "fahrenheit"]},
},
"required": ["location"],
},
}
]
user_prompt = get_prompt(query, functions)
output = llm(user_prompt,
max_tokens=512, # Generate up to 512 tokens
stop=["<|EOT|>"],
echo=True # Whether to echo the prompt
)
print("Output: ", output)
```
The expected output of successfully running this script is the following (tested on March 3, 2024)
```
❯ python quantized_inference.py
llama_model_loader: loaded meta data with 22 key-value pairs and 273 tensors from /Users/charliecheng-jieji/Downloads/codebase/quantized_eval/gorilla-openfunctions-v1-GGUF/gorilla-openfunctions-v0-q2_K.gguf (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv 0: general.architecture str = llama
llama_model_loader: - kv 1: general.name str = LLaMA v0
llama_model_loader: - kv 2: llama.context_length u32 = 4096
llama_model_loader: - kv 3: llama.embedding_length u32 = 4096
llama_model_loader: - kv 4: llama.block_count u32 = 30
llama_model_loader: - kv 5: llama.feed_forward_length u32 = 11008
llama_model_loader: - kv 6: llama.rope.dimension_count u32 = 128
llama_model_loader: - kv 7: llama.attention.head_count u32 = 32
llama_model_loader: - kv 8: llama.attention.head_count_kv u32 = 32
llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 = 0.000001
llama_model_loader: - kv 10: llama.rope.freq_base f32 = 10000.000000
llama_model_loader: - kv 11: general.file_type u32 = 10
llama_model_loader: - kv 12: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 13: tokenizer.ggml.tokens arr[str,102400] = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 14: tokenizer.ggml.scores arr[f32,102400] = [0.000000, 0.000000, 0.000000, 0.0000...
llama_model_loader: - kv 15: tokenizer.ggml.token_type arr[i32,102400] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 16: tokenizer.ggml.merges arr[str,99757] = ["Ġ Ġ", "Ġ t", "Ġ a", "i n", "h e...
llama_model_loader: - kv 17: tokenizer.ggml.bos_token_id u32 = 100000
llama_model_loader: - kv 18: tokenizer.ggml.eos_token_id u32 = 100015
llama_model_loader: - kv 19: tokenizer.ggml.padding_token_id u32 = 100001
llama_model_loader: - kv 20: tokenizer.chat_template str = {% if not add_generation_prompt is de...
llama_model_loader: - kv 21: general.quantization_version u32 = 2
llama_model_loader: - type f32: 61 tensors
llama_model_loader: - type q2_K: 121 tensors
llama_model_loader: - type q3_K: 90 tensors
llama_model_loader: - type q6_K: 1 tensors
llm_load_vocab: mismatch in special tokens definition ( 2387/102400 vs 2400/102400 ).
llm_load_print_meta: format = GGUF V3 (latest)
llm_load_print_meta: arch = llama
llm_load_print_meta: vocab type = BPE
llm_load_print_meta: n_vocab = 102400
llm_load_print_meta: n_merges = 99757
llm_load_print_meta: n_ctx_train = 4096
llm_load_print_meta: n_embd = 4096
llm_load_print_meta: n_head = 32
llm_load_print_meta: n_head_kv = 32
llm_load_print_meta: n_layer = 30
llm_load_print_meta: n_rot = 128
llm_load_print_meta: n_embd_head_k = 128
llm_load_print_meta: n_embd_head_v = 128
llm_load_print_meta: n_gqa = 1
llm_load_print_meta: n_embd_k_gqa = 4096
llm_load_print_meta: n_embd_v_gqa = 4096
llm_load_print_meta: f_norm_eps = 0.0e+00
llm_load_print_meta: f_norm_rms_eps = 1.0e-06
llm_load_print_meta: f_clamp_kqv = 0.0e+00
llm_load_print_meta: f_max_alibi_bias = 0.0e+00
llm_load_print_meta: n_ff = 11008
llm_load_print_meta: n_expert = 0
llm_load_print_meta: n_expert_used = 0
llm_load_print_meta: pooling type = 0
llm_load_print_meta: rope type = 0
llm_load_print_meta: rope scaling = linear
llm_load_print_meta: freq_base_train = 10000.0
llm_load_print_meta: freq_scale_train = 1
llm_load_print_meta: n_yarn_orig_ctx = 4096
llm_load_print_meta: rope_finetuned = unknown
llm_load_print_meta: model type = ?B
llm_load_print_meta: model ftype = Q2_K - Medium
llm_load_print_meta: model params = 6.91 B
llm_load_print_meta: model size = 2.53 GiB (3.14 BPW)
llm_load_print_meta: general.name = LLaMA v2
llm_load_print_meta: BOS token = 100000 '<｜begin▁of▁sentence｜>'
llm_load_print_meta: EOS token = 100015 '<|EOT|>'
llm_load_print_meta: PAD token = 100001 '<｜end▁of▁sentence｜>'
llm_load_print_meta: LF token = 126 'Ä'
llm_load_tensors: ggml ctx size = 0.21 MiB
ggml_backend_metal_buffer_from_ptr: allocated buffer, size = 2457.45 MiB, ( 2457.52 / 10922.67)
llm_load_tensors: offloading 30 repeating layers to GPU
llm_load_tensors: offloading non-repeating layers to GPU
llm_load_tensors: offloaded 31/31 layers to GPU
llm_load_tensors: CPU buffer size = 131.25 MiB
llm_load_tensors: Metal buffer size = 2457.45 MiB
.....................................................................................
llama_new_context_with_model: n_ctx = 512
llama_new_context_with_model: freq_base = 10000.0
llama_new_context_with_model: freq_scale = 1
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: default.metallib not found, loading from source
ggml_metal_init: GGML_METAL_PATH_RESOURCES = nil
ggml_metal_init: loading '/Users/charliecheng-jieji/miniconda3/envs/public-api/lib/python3.12/site-packages/llama_cpp/ggml-metal.metal'
ggml_metal_init: GPU name: Apple M1
ggml_metal_init: GPU family: MTLGPUFamilyApple7 (1007)
ggml_metal_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_init: GPU family: MTLGPUFamilyMetal3 (5001)
ggml_metal_init: simdgroup reduction support = true
ggml_metal_init: simdgroup matrix mul. support = true
ggml_metal_init: hasUnifiedMemory = true
ggml_metal_init: recommendedMaxWorkingSetSize = 11453.25 MB
ggml_backend_metal_buffer_type_alloc_buffer: allocated buffer, size = 240.00 MiB, ( 2699.33 / 10922.67)
llama_kv_cache_init: Metal KV buffer size = 240.00 MiB
llama_new_context_with_model: KV self size = 240.00 MiB, K (f16): 120.00 MiB, V (f16): 120.00 MiB
llama_new_context_with_model: CPU input buffer size = 10.01 MiB
ggml_backend_metal_buffer_type_alloc_buffer: allocated buffer, size = 208.00 MiB, ( 2907.33 / 10922.67)
llama_new_context_with_model: Metal compute buffer size = 208.00 MiB
llama_new_context_with_model: CPU compute buffer size = 8.00 MiB
llama_new_context_with_model: graph splits (measure): 2
AVX = 0 | AVX_VNNI = 0 | AVX2 = 0 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 0 | NEON = 1 | ARM_FMA = 1 | F16C = 0 | FP16_VA = 1 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 0 | SSSE3 = 0 | VSX = 0 | MATMUL_INT8 = 0 |
Model metadata: {'general.quantization_version': '2', 'tokenizer.chat_template': "{% if not add_generation_prompt is defined %}\n{% set add_generation_prompt = false %}\n{% endif %}\n{%- set ns = namespace(found=false) -%}\n{%- for message in messages -%}\n {%- if message['role'] == 'system' -%}\n {%- set ns.found = true -%}\n {%- endif -%}\n{%- endfor -%}\n{{bos_token}}{%- if not ns.found -%}\n{{'You are an AI programming assistant, utilizing the Deepseek Coder model, developed by Deepseek Company, and you only answer questions related to computer science. For politically sensitive questions, security and privacy issues, and other non-computer science questions, you will refuse to answer\\n'}}\n{%- endif %}\n{%- for message in messages %}\n {%- if message['role'] == 'system' %}\n{{ message['content'] }}\n {%- else %}\n {%- if message['role'] == 'user' %}\n{{'### Instruction:\\n' + message['content'] + '\\n'}}\n {%- else %}\n{{'### Response:\\n' + message['content'] + '\\n<|EOT|>\\n'}}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{% if add_generation_prompt %}\n{{'### Response:'}}\n{% endif %}", 'tokenizer.ggml.padding_token_id': '100001', 'tokenizer.ggml.eos_token_id': '100015', 'tokenizer.ggml.bos_token_id': '100000', 'tokenizer.ggml.model': 'gpt2', 'llama.attention.head_count_kv': '32', 'llama.context_length': '4096', 'llama.attention.head_count': '32', 'llama.rope.freq_base': '10000.000000', 'llama.rope.dimension_count': '128', 'general.file_type': '10', 'llama.feed_forward_length': '11008', 'llama.embedding_length': '4096', 'llama.block_count': '30', 'general.architecture': 'llama', 'llama.attention.layer_norm_rms_epsilon': '0.000001', 'general.name': 'LLaMA v2'}
Using gguf chat template: {% if not add_generation_prompt is defined %}
{% set add_generation_prompt = false %}
{% endif %}
{%- set ns = namespace(found=false) -%}
{%- for message in messages -%}
{%- if message['role'] == 'system' -%}
{%- set ns.found = true -%}
{%- endif -%}
{%- endfor -%}
{{bos_token}}{%- if not ns.found -%}
{{'You are an AI programming assistant, utilizing the Deepseek Coder model, developed by Deepseek Company, and you only answer questions related to computer science. For politically sensitive questions, security and privacy issues, and other non-computer science questions, you will refuse to answer\n'}}
{%- endif %}
{%- for message in messages %}
{%- if message['role'] == 'system' %}
{{ message['content'] }}
{%- else %}
{%- if message['role'] == 'user' %}
{{'### Instruction:\n' + message['content'] + '\n'}}
{%- else %}
{{'### Response:\n' + message['content'] + '\n<|EOT|>\n'}}
{%- endif %}
{%- endif %}
{%- endfor %}
{% if add_generation_prompt %}
{{'### Response:'}}
{% endif %}
Using chat eos_token: <|EOT|>
Using chat bos_token: <｜begin▁of▁sentence｜>
llama_print_timings: load time = 1890.11 ms
llama_print_timings: sample time = 23.48 ms / 40 runs ( 0.59 ms per token, 1703.94 tokens per second)
llama_print_timings: prompt eval time = 1889.91 ms / 181 tokens ( 10.44 ms per token, 95.77 tokens per second)
llama_print_timings: eval time = 2728.54 ms / 39 runs ( 69.96 ms per token, 14.29 tokens per second)
llama_print_timings: total time = 5162.12 ms / 220 tokens
```
<code>
Output: {'id': 'cmpl-0679223d-578f-42be-bbce-0e307faddd28', 'object': 'text_completion', 'created': 1709525244, 'model': '/Users/charliecheng-jieji/Downloads/codebase/quantized_eval/gorilla-openfunctions-v1-GGUF/gorilla-openfunctions-v1-q2_K.gguf', 'choices': [{'text': 'You are an AI programming assistant, utilizing the Gorilla LLM model, developed by Gorilla LLM, and you only answer questions related to computer science. For politically sensitive questions, security and privacy issues, and other non-computer science questions, you will refuse to answer.\n### Instruction: <<function>>[{"name": "get_current_weather", "description": "Get the current weather in a given location", "parameters": {"type": "object", "properties": {"location": {"type": "string", "description": "The city and state, e.g. San Francisco, CA"}, "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]}}, "required": ["location"]}}]\n<<question>>What\'s the weather like in the two cities of Boston and San Francisco?\n### Response: <<function>>get_current_weather(location=\'Boston\', unit=\'fahrenheit\')<<function>>get_current_weather(location=\'San Francisco\', unit=\'fahrenheit\')', 'index': 0, 'logprobs': None, 'finish_reason': 'stop'}], 'usage': {'prompt_tokens': 181, 'completion_tokens': 39, 'total_tokens': 220}}
</code>
|
AmeerH/Mistral-Math-2x7b-mix | AmeerH | 2024-03-18T08:08:14Z | 453 | 4 | transformers | [
"transformers",
"safetensors",
"mixtral",
"text-generation",
"moe",
"frankenmoe",
"merge",
"mergekit",
"lazymergekit",
"mistralai/Mistral-7B-Instruct-v0.2",
"meta-math/MetaMath-Mistral-7B",
"conversational",
"base_model:mistralai/Mistral-7B-Instruct-v0.2",
"base_model:meta-math/MetaMath-Mistral-7B",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-03-18T08:00:59Z | ---
license: apache-2.0
tags:
- moe
- frankenmoe
- merge
- mergekit
- lazymergekit
- mistralai/Mistral-7B-Instruct-v0.2
- meta-math/MetaMath-Mistral-7B
base_model:
- mistralai/Mistral-7B-Instruct-v0.2
- meta-math/MetaMath-Mistral-7B
---
# Mistral-Math-2x7b-mix
Mistral-Math-2x7b-mix is a Mixure of Experts (MoE) made with the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2)
* [meta-math/MetaMath-Mistral-7B](https://huggingface.co/meta-math/MetaMath-Mistral-7B)
## 🧩 Configuration
```yaml
base_model: mistralai/Mistral-7B-Instruct-v0.2
dtype: float16
gate_mode: cheap_embed
experts:
- source_model: mistralai/Mistral-7B-Instruct-v0.2
positive_prompts: ["You are helpful assistant."]
- source_model: meta-math/MetaMath-Mistral-7B
positive_prompts: ["You are an assistant good at math."]
```
## 💻 Usage
```python
!pip install -qU transformers bitsandbytes accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "AmeerH/Mistral-Math-2x7b-mix"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True},
)
messages = [{"role": "user", "content": "Explain what a Mixture of Experts is in less than 100 words."}]
prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
``` |
MTSAIR/multi_verse_model_GPTQ | MTSAIR | 2024-04-07T07:46:26Z | 453 | 1 | transformers | [
"transformers",
"pytorch",
"safetensors",
"mistral",
"text-generation",
"conversational",
"license:other",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"4-bit",
"gptq",
"region:us"
]
| text-generation | 2024-04-07T07:28:51Z | ---
license: other
---
Quantized int4 version of [multi_verse_model](https://huggingface.co/MTSAIR/multi_verse_model) with GPTQ quantization method
I, a learning bot, have been enhanced through a groundbreaking training method. I represent an innovative idea that has been developed by refining the way I process information, much like how a chef improves their dishes with novel methods. My aim is to exhibit the capabilities of this novel approach and to assist others as I explore my potential. Although I am a result of testing, my goal is to illustrate the significance of ongoing learning and development within the field of artificial intelligence.' |
mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF | mradermacher | 2024-05-14T19:09:33Z | 453 | 1 | transformers | [
"transformers",
"gguf",
"en",
"base_model:cognitivecomputations/Llama-3-8B-Instruct-abliterated-v2",
"license:llama3",
"endpoints_compatible",
"region:us"
]
| null | 2024-05-12T21:02:21Z | ---
base_model: cognitivecomputations/Llama-3-8B-Instruct-abliterated-v2
language:
- en
library_name: transformers
license: llama3
quantized_by: mradermacher
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
weighted/imatrix quants of https://huggingface.co/cognitivecomputations/Llama-3-8B-Instruct-abliterated-v2
<!-- provided-files -->
static quants are available at https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-IQ1_S.gguf) | i1-IQ1_S | 2.1 | for the desperate |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-IQ1_M.gguf) | i1-IQ1_M | 2.3 | mostly desperate |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 2.5 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-IQ2_XS.gguf) | i1-IQ2_XS | 2.7 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-IQ2_S.gguf) | i1-IQ2_S | 2.9 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-IQ2_M.gguf) | i1-IQ2_M | 3.0 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-Q2_K.gguf) | i1-Q2_K | 3.3 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 3.4 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-IQ3_XS.gguf) | i1-IQ3_XS | 3.6 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.8 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-IQ3_S.gguf) | i1-IQ3_S | 3.8 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-IQ3_M.gguf) | i1-IQ3_M | 3.9 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-Q3_K_M.gguf) | i1-Q3_K_M | 4.1 | IQ3_S probably better |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-Q3_K_L.gguf) | i1-Q3_K_L | 4.4 | IQ3_M probably better |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-IQ4_XS.gguf) | i1-IQ4_XS | 4.5 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-Q4_0.gguf) | i1-Q4_0 | 4.8 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-Q4_K_S.gguf) | i1-Q4_K_S | 4.8 | optimal size/speed/quality |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-Q4_K_M.gguf) | i1-Q4_K_M | 5.0 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-Q5_K_S.gguf) | i1-Q5_K_S | 5.7 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-Q5_K_M.gguf) | i1-Q5_K_M | 5.8 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3-8B-Instruct-abliterated-v2-i1-GGUF/resolve/main/Llama-3-8B-Instruct-abliterated-v2.i1-Q6_K.gguf) | i1-Q6_K | 6.7 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time.
<!-- end -->
|
tresiwalde/Llama-3-SauerkrautLM-70b-Instruct-AWQ | tresiwalde | 2024-05-21T19:12:05Z | 453 | 0 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"conversational",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"4-bit",
"awq",
"region:us"
]
| text-generation | 2024-05-21T19:05:07Z | Entry not found |
RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf | RichardErkhov | 2024-05-29T21:52:36Z | 453 | 0 | null | [
"gguf",
"region:us"
]
| null | 2024-05-29T18:38:57Z | Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
llama-2-7b-rockwell-final - GGUF
- Model creator: https://huggingface.co/maximuslee07/
- Original model: https://huggingface.co/maximuslee07/llama-2-7b-rockwell-final/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [llama-2-7b-rockwell-final.Q2_K.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q2_K.gguf) | Q2_K | 2.36GB |
| [llama-2-7b-rockwell-final.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.IQ3_XS.gguf) | IQ3_XS | 2.6GB |
| [llama-2-7b-rockwell-final.IQ3_S.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.IQ3_S.gguf) | IQ3_S | 2.75GB |
| [llama-2-7b-rockwell-final.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q3_K_S.gguf) | Q3_K_S | 2.75GB |
| [llama-2-7b-rockwell-final.IQ3_M.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.IQ3_M.gguf) | IQ3_M | 2.9GB |
| [llama-2-7b-rockwell-final.Q3_K.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q3_K.gguf) | Q3_K | 3.07GB |
| [llama-2-7b-rockwell-final.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q3_K_M.gguf) | Q3_K_M | 3.07GB |
| [llama-2-7b-rockwell-final.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q3_K_L.gguf) | Q3_K_L | 3.35GB |
| [llama-2-7b-rockwell-final.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.IQ4_XS.gguf) | IQ4_XS | 3.4GB |
| [llama-2-7b-rockwell-final.Q4_0.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q4_0.gguf) | Q4_0 | 3.56GB |
| [llama-2-7b-rockwell-final.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.IQ4_NL.gguf) | IQ4_NL | 3.58GB |
| [llama-2-7b-rockwell-final.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q4_K_S.gguf) | Q4_K_S | 3.59GB |
| [llama-2-7b-rockwell-final.Q4_K.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q4_K.gguf) | Q4_K | 3.8GB |
| [llama-2-7b-rockwell-final.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q4_K_M.gguf) | Q4_K_M | 3.8GB |
| [llama-2-7b-rockwell-final.Q4_1.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q4_1.gguf) | Q4_1 | 3.95GB |
| [llama-2-7b-rockwell-final.Q5_0.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q5_0.gguf) | Q5_0 | 4.33GB |
| [llama-2-7b-rockwell-final.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q5_K_S.gguf) | Q5_K_S | 4.33GB |
| [llama-2-7b-rockwell-final.Q5_K.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q5_K.gguf) | Q5_K | 4.45GB |
| [llama-2-7b-rockwell-final.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q5_K_M.gguf) | Q5_K_M | 4.45GB |
| [llama-2-7b-rockwell-final.Q5_1.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q5_1.gguf) | Q5_1 | 4.72GB |
| [llama-2-7b-rockwell-final.Q6_K.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q6_K.gguf) | Q6_K | 5.15GB |
| [llama-2-7b-rockwell-final.Q8_0.gguf](https://huggingface.co/RichardErkhov/maximuslee07_-_llama-2-7b-rockwell-final-gguf/blob/main/llama-2-7b-rockwell-final.Q8_0.gguf) | Q8_0 | 6.67GB |
Original model description:
---
license: llama2
---
|
solidrust/Codestral-22B-v0.1-hf-AWQ | solidrust | 2024-05-30T02:19:50Z | 453 | 1 | transformers | [
"transformers",
"safetensors",
"mistral",
"text-generation",
"4-bit",
"AWQ",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"awq",
"region:us"
]
| text-generation | 2024-05-30T02:17:36Z | ---
library_name: transformers
tags:
- 4-bit
- AWQ
- text-generation
- autotrain_compatible
- endpoints_compatible
pipeline_tag: text-generation
inference: false
quantized_by: Suparious
---
# bullerwins/Codestral-22B-v0.1-hf AWQ
- Model creator: [bullerwins](https://huggingface.co/bullerwins)
- Original model: [Codestral-22B-v0.1-hf](https://huggingface.co/bullerwins/Codestral-22B-v0.1-hf)
## How to use
### Install the necessary packages
```bash
pip install --upgrade autoawq autoawq-kernels
```
### Example Python code
```python
from awq import AutoAWQForCausalLM
from transformers import AutoTokenizer, TextStreamer
model_path = "solidrust/Codestral-22B-v0.1-hf-AWQ"
system_message = "You are Codestral-22B-v0.1-hf, incarnated as a powerful AI. You were created by bullerwins."
# Load model
model = AutoAWQForCausalLM.from_quantized(model_path,
fuse_layers=True)
tokenizer = AutoTokenizer.from_pretrained(model_path,
trust_remote_code=True)
streamer = TextStreamer(tokenizer,
skip_prompt=True,
skip_special_tokens=True)
# Convert prompt to tokens
prompt_template = """\
<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant"""
prompt = "You're standing on the surface of the Earth. "\
"You walk one mile south, one mile west and one mile north. "\
"You end up exactly where you started. Where are you?"
tokens = tokenizer(prompt_template.format(system_message=system_message,prompt=prompt),
return_tensors='pt').input_ids.cuda()
# Generate output
generation_output = model.generate(tokens,
streamer=streamer,
max_new_tokens=512)
```
### About AWQ
AWQ is an efficient, accurate and blazing-fast low-bit weight quantization method, currently supporting 4-bit quantization. Compared to GPTQ, it offers faster Transformers-based inference with equivalent or better quality compared to the most commonly used GPTQ settings.
AWQ models are currently supported on Linux and Windows, with NVidia GPUs only. macOS users: please use GGUF models instead.
It is supported by:
- [Text Generation Webui](https://github.com/oobabooga/text-generation-webui) - using Loader: AutoAWQ
- [vLLM](https://github.com/vllm-project/vllm) - version 0.2.2 or later for support for all model types.
- [Hugging Face Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference)
- [Transformers](https://huggingface.co/docs/transformers) version 4.35.0 and later, from any code or client that supports Transformers
- [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) - for use from Python code
|
QuantFactory/L3-Aethora-15B-GGUF | QuantFactory | 2024-06-12T04:53:34Z | 453 | 0 | transformers | [
"transformers",
"gguf",
"llama-factory",
"text-generation",
"dataset:TheSkullery/Aether-Lite-V1.2",
"base_model:Steelskull/L3-Aethora-15B",
"license:llama3",
"endpoints_compatible",
"region:us"
]
| text-generation | 2024-06-11T06:41:04Z | ---
library_name: transformers
tags:
- llama-factory
license: llama3
datasets:
- TheSkullery/Aether-Lite-V1.2
base_model: Steelskull/L3-Aethora-15B
pipeline_tag: text-generation
---
# QuantFactory/L3-Aethora-15B-GGUF
This is quantized version of [Steelskull/L3-Aethora-15B](https://huggingface.co/Steelskull/L3-Aethora-15B) created using llama.cpp
# Model Description
<!DOCTYPE html>
<style>
body, html {
height: 100%; /* Ensure the full height of the page is used */
margin: 0;
padding: 0;
font-family: 'Quicksand', sans-serif;
background: linear-gradient(135deg, #2E3440 0%, #1A202C 100%);
color: #D8DEE9;
font-size: 16px;
}
.container {
width: 100%; /* Full width */
height: 100%; /* Full height */
padding: 20px;
margin: 0; /* Remove margin to fill the entire area */
background-color: rgba(255, 255, 255, 0.02);
border-radius: 12px;
box-shadow: 0 4px 10px rgba(0, 0, 0, 0.2);
backdrop-filter: blur(10px);
border: 1px solid rgba(255, 255, 255, 0.1);
}
.header h1 {
font-size: 28px;
color: #5F9EA0;
margin: 0 0 20px 0;
text-shadow: 2px 2px 4px rgba(0, 0, 0, 0.3);
}
.update-section h2 {
font-size: 24px;
color: #88C0D0;
}
.update-section p {
font-size: 16px;
line-height: 1.6;
color: #ECEFF4;
}
.info img {
width: 100%;
border-radius: 10px;
margin-bottom: 15px;
}
a {
color: #88C0D0;
text-decoration: none;
}
a:hover {
color: #A3BE8C;
}
.button {
display: inline-block;
background-color: #5E81AC;
color: #E5E9F0;
padding: 10px 20px;
border-radius: 5px;
cursor: pointer;
text-decoration: none;
}
.button:hover {
background-color: #81A1C1;
}
pre {
background-color: #2E3440;
padding: 10px;
border-radius: 5px;
overflow-x: auto;
}
code {
font-family: 'Courier New', monospace;
color: #D8DEE9;
}
</style>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>L3-Aethora-15B Data Card</title>
<link href="https://fonts.googleapis.com/css2?family=Quicksand:wght@400;500;600&display=swap" rel="stylesheet">
</head>
<body>
<div class="container">
<div class="header">
<h1>L3-Aethora-15B</h1>
</div>
<div class="info">
<img src="https://cdn-uploads.huggingface.co/production/uploads/64545af5ec40bbbd01242ca6/W0qzZK_V1Zt1GdgCIsnrP.png">
<p>The Skullery Presents L3-Aethora-15B.</p>
<p><strong>Creator:</strong> <a href="https://huggingface.co/steelskull" target="_blank">Steelskull</a></p>
<p><strong>Dataset:</strong> <a href="https://huggingface.co/datasets/TheSkullery/Aether-Lite-V1.2" target="_blank">Aether-Lite-V1.2</a></p>
<p><strong>Trained:</strong> 4 x A100 for 15 hours Using RsLora and DORA</p>
<h1>About L3-Aethora-15B:</h1>
<pre><code> L3 = Llama3 </code></pre>
<p>L3-Aethora-15B was crafted through using the abilteration method to adjust model responses. The model's refusal is inhibited, focusing on yielding more compliant and facilitative dialogue interactions. It then underwent a modified DUS (Depth Up Scale) merge (originally used by @Elinas) by using passthrough merge to create a 15b model, with specific adjustments (zeroing) to 'o_proj' and 'down_proj', enhancing its efficiency and reducing perplexity. This created AbL3In-15b.<br>
<p>AbL3In-15b was then trained for 4 epochs using Rslora & DORA training methods on the Aether-Lite-V1.2 dataset, containing ~82000 high quality samples, designed to strike a fine balance between creativity, slop, and intelligence at about a 60/40 split</p>
<p>This model is trained on the L3 prompt format.</p>
<p></p>
<h2>Dataset Summary: (Filtered)</h2>
<p>Filtered Phrases: GPTslop, Claudism's</p>
<ul>
<li><strong>mrfakename/Pure-Dove-ShareGPT:</strong> Processed 3707, Removed 150</li>
<li><strong>mrfakename/Capybara-ShareGPT:</strong> Processed 13412, Removed 2594</li>
<li><strong>jondurbin/airoboros-3.2:</strong> Processed 54517, Removed 4192</li>
<li><strong>PJMixers/grimulkan_theory-of-mind-ShareGPT:</strong> Processed 533, Removed 6</li>
<li><strong>grimulkan/PIPPA-augmented-dedup:</strong> Processed 869, Removed 46</li>
<li><strong>grimulkan/LimaRP-augmented:</strong> Processed 790, Removed 14</li>
<li><strong>PJMixers/grimulkan_physical-reasoning-ShareGPT:</strong> Processed 895, Removed 4</li>
<li><strong>MinervaAI/Aesir-Preview:</strong> Processed 994, Removed 6</li>
<li><strong>Doctor-Shotgun/no-robots-sharegpt:</strong> Processed 9911, Removed 89</li>
</ul>
<h2>Deduplication Stats:</h2>
<p>Starting row count: 85628, Final row count: 81960, Rows removed: 3668</p>
</div>
</div>
</body>
</html> |
akash4552/phi3-mini-ten_thousand | akash4552 | 2024-06-26T17:10:00Z | 453 | 0 | null | [
"gguf",
"license:mit",
"region:us"
]
| null | 2024-06-26T17:08:14Z | ---
license: mit
---
|
prithivida/formal_to_informal_styletransfer | prithivida | 2021-06-21T08:08:37Z | 452 | 5 | transformers | [
"transformers",
"pytorch",
"t5",
"text2text-generation",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text2text-generation | 2022-03-02T23:29:05Z | ## This model belongs to the Styleformer project
[Please refer to github page](https://github.com/PrithivirajDamodaran/Styleformer)
|
timm/hrnet_w18_small.ms_in1k | timm | 2023-04-24T21:26:48Z | 452 | 0 | timm | [
"timm",
"pytorch",
"safetensors",
"image-classification",
"dataset:imagenet-1k",
"arxiv:1908.07919",
"license:mit",
"region:us"
]
| image-classification | 2023-04-24T21:26:22Z | ---
tags:
- image-classification
- timm
library_name: timm
license: mit
datasets:
- imagenet-1k
---
# Model card for hrnet_w18_small.ms_in1k
A HRNet image classification model. Trained on ImageNet-1k by paper authors.
## Model Details
- **Model Type:** Image classification / feature backbone
- **Model Stats:**
- Params (M): 13.2
- GMACs: 1.6
- Activations (M): 5.7
- Image size: 224 x 224
- **Papers:**
- Deep High-Resolution Representation Learning for Visual Recognition: https://arxiv.org/abs/1908.07919
- **Original:** https://github.com/HRNet/HRNet-Image-Classification
- **Dataset:** ImageNet-1k
## Model Usage
### Image Classification
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model('hrnet_w18_small.ms_in1k', pretrained=True)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)
```
### Feature Map Extraction
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'hrnet_w18_small.ms_in1k',
pretrained=True,
features_only=True,
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
for o in output:
# print shape of each feature map in output
# e.g.:
# torch.Size([1, 64, 112, 112])
# torch.Size([1, 128, 56, 56])
# torch.Size([1, 256, 28, 28])
# torch.Size([1, 512, 14, 14])
# torch.Size([1, 1024, 7, 7])
print(o.shape)
```
### Image Embeddings
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'hrnet_w18_small.ms_in1k',
pretrained=True,
num_classes=0, # remove classifier nn.Linear
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # output is (batch_size, num_features) shaped tensor
# or equivalently (without needing to set num_classes=0)
output = model.forward_features(transforms(img).unsqueeze(0))
# output is unpooled, a (1, 2048, 7, 7) shaped tensor
output = model.forward_head(output, pre_logits=True)
# output is a (1, num_features) shaped tensor
```
## Model Comparison
Explore the dataset and runtime metrics of this model in timm [model results](https://github.com/huggingface/pytorch-image-models/tree/main/results).
## Citation
```bibtex
@article{WangSCJDZLMTWLX19,
title={Deep High-Resolution Representation Learning for Visual Recognition},
author={Jingdong Wang and Ke Sun and Tianheng Cheng and
Borui Jiang and Chaorui Deng and Yang Zhao and Dong Liu and Yadong Mu and
Mingkui Tan and Xinggang Wang and Wenyu Liu and Bin Xiao},
journal = {TPAMI}
year={2019}
}
```
|
TheBloke/MPT-7B-Instruct-GGML | TheBloke | 2023-06-08T08:59:52Z | 452 | 29 | transformers | [
"transformers",
"mpt",
"Composer",
"MosaicML",
"llm-foundry",
"dataset:mosaicml/dolly_hhrlhf",
"arxiv:2205.14135",
"arxiv:2108.12409",
"arxiv:2010.04245",
"license:cc-by-sa-3.0",
"text-generation-inference",
"region:us"
]
| null | 2023-05-18T16:41:36Z | ---
license: cc-by-sa-3.0
datasets:
- mosaicml/dolly_hhrlhf
tags:
- Composer
- MosaicML
- llm-foundry
inference: false
---
<!-- header start -->
<div style="width: 100%;">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p><a href="https://discord.gg/Jq4vkcDakD">Chat & support: my new Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<!-- header end -->
# MPT-7B-Instruct GGML
This is GGML format quantised 4-bit, 5-bit and 8-bit GGML models of [MosaicML's MPT-7B-Instruct](https://huggingface.co/mosaicml/mpt-7b-instruct).
This repo is the result of converting to GGML and quantising.
Please note that these MPT GGMLs are **not compatbile with llama.cpp**. Please see below for a list of tools known to work with these model files.
## Repositories available
* [MPT-7B: 4-bit, 5-bit and 8-bit GGML models for CPU (+CUDA) inference](https://huggingface.co/TheBloke/MPT-7B-GGML).
* [MPT-7B-Instruct: 4-bit, 5-bit and 8-bit GGML models for CPU (+CUDA) inference](https://huggingface.co/TheBloke/MPT-7B-Instruct-GGML).
* [MPT-7B-Storywriter: 4-bit, 5-bit and 8-bit GGML models for CPU (+CUDA) inference](https://huggingface.co/TheBloke/MPT-7B-Storywriter-GGML).
## Provided files
| Name | Quant method | Bits | Size | RAM required | Use case |
| ---- | ---- | ---- | ---- | ---- | ----- |
`mpt7b-instruct.ggmlv3.q4_0.bin` | q4_0 | 4bit | 4.16GB | 6.2GB | 4-bit. |
`mpt7b-instruct.ggmlv3.q4_1.bin` | q4_0 | 4bit | 4.99GB | 7.2GB | 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
`mpt7b-instruct.ggmlv3.q5_0.bin` | q5_0 | 5bit | 4.57GB | 6.8GB | 5-bit. Higher accuracy, higher resource usage and slower inference. |
`mpt7b-instruct.ggmlv3.q5_1.bin` | q5_1 | 5bit | 4.99GB | 7.2GB | 5-bit. Even higher accuracy, and higher resource usage and slower inference. |
`mpt7b-instruct.ggmlv3.q8_0.bin` | q8_0 | 8bit | 7.48GB | 9.7GB | 8-bit. Almost indistinguishable from float16. Huge resource use and slow. Not recommended for normal use. |
`mpt7b-instruct.ggmlv3.fp16.bin` | fp16 | 16bit | 13.30GB | 16GB | Full 16-bit. |
## Compatibilty
These files are **not** compatible with llama.cpp.
Currently they can be used with:
* KoboldCpp, a powerful inference engine based on llama.cpp, with good UI: [KoboldCpp](https://github.com/LostRuins/koboldcpp)
* The ctransformers Python library, which includes LangChain support: [ctransformers](https://github.com/marella/ctransformers)
* The GPT4All-UI which uses ctransformers: [GPT4All-UI](https://github.com/ParisNeo/gpt4all-ui)
* [rustformers' llm](https://github.com/rustformers/llm)
* The example `mpt` binary provided with [ggml](https://github.com/ggerganov/ggml)
As other options become available I will endeavour to update them here (do let me know in the Community tab if I've missed something!)
## Tutorial for using GPT4All-UI
* [Text tutorial, written by **Lucas3DCG**](https://huggingface.co/TheBloke/MPT-7B-Storywriter-GGML/discussions/2#6475d914e9b57ce0caa68888)
* [Video tutorial, by GPT4All-UI's author **ParisNeo**](https://www.youtube.com/watch?v=ds_U0TDzbzI)
<!-- footer start -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/Jq4vkcDakD)
## Thanks, and how to contribute.
Thanks to the [chirper.ai](https://chirper.ai) team!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Patreon special mentions**: Aemon Algiz, Dmitriy Samsonov, Nathan LeClaire, Trenton Dambrowitz, Mano Prime, David Flickinger, vamX, Nikolai Manek, senxiiz, Khalefa Al-Ahmad, Illia Dulskyi, Jonathan Leane, Talal Aujan, V. Lukas, Joseph William Delisle, Pyrater, Oscar Rangel, Lone Striker, Luke Pendergrass, Eugene Pentland, Sebastain Graf, Johann-Peter Hartman.
Thank you to all my generous patrons and donaters!
<!-- footer end -->
# MPT-7B-Instruct
MPT-7B-Instruct is a model for short-form instruction following.
It is built by finetuning [MPT-7B](https://huggingface.co/spaces/mosaicml/mpt-7b) on a [dataset](https://huggingface.co/datasets/sam-mosaic/dolly_hhrlhf) derived from the [Databricks Dolly-15k](https://huggingface.co/datasets/databricks/databricks-dolly-15k) and the [Anthropic Helpful and Harmless (HH-RLHF)](https://huggingface.co/datasets/Anthropic/hh-rlhf) datasets.
* License: _CC-By-SA-3.0_
* [Demo on Hugging Face Spaces](https://huggingface.co/spaces/mosaicml/mpt-7b-instruct)
This model was trained by [MosaicML](https://www.mosaicml.com) and follows a modified decoder-only transformer architecture.
## Model Date
May 5, 2023
## Model License
CC-By-SA-3.0
## Documentation
* [Blog post: Introducing MPT-7B: A New Standard for Open-Source, Commercially Usable LLMs](https://www.mosaicml.com/blog/mpt-7b)
* [Codebase (mosaicml/llm-foundry repo)](https://github.com/mosaicml/llm-foundry/)
* Questions: Feel free to contact us via the [MosaicML Community Slack](https://join.slack.com/t/mosaicml-community/shared_invite/zt-1btms90mc-GipE2ufuPkKY0QBrmF3LSA)!
### Example Question/Instruction
**Longboi24**:
> What is a quoll?
**MPT-7B-Instruct**:
>A Quoll (pronounced “cool”) is one of Australia’s native carnivorous marsupial mammals, which are also known as macropods or wallabies in other parts around Asia and South America
## How to Use
Note: This model requires that `trust_remote_code=True` be passed to the `from_pretrained` method. This is because we use a custom model architecture that is not yet part of the `transformers` package.
It includes options for many training efficiency features such as [FlashAttention (Dao et al. 2022)](https://arxiv.org/pdf/2205.14135.pdf), [ALiBi](https://arxiv.org/abs/2108.12409), QK LayerNorm, and more.
```python
import transformers
model = transformers.AutoModelForCausalLM.from_pretrained(
'mosaicml/mpt-7b-instruct',
trust_remote_code=True
)
```
Note: This model requires that `trust_remote_code=True` be passed to the `from_pretrained` method.
This is because we use a custom `MPT` model architecture that is not yet part of the Hugging Face `transformers` package.
`MPT` includes options for many training efficiency features such as [FlashAttention](https://arxiv.org/pdf/2205.14135.pdf), [ALiBi](https://arxiv.org/abs/2108.12409), [QK LayerNorm](https://arxiv.org/abs/2010.04245), and more.
To use the optimized [triton implementation](https://github.com/openai/triton) of FlashAttention, you can load the model with `attn_impl='triton'` and move the model to `bfloat16`:
```python
config = transformers.AutoConfig.from_pretrained(
'mosaicml/mpt-7b-instruct',
trust_remote_code=True
)
config.attn_config['attn_impl'] = 'triton'
model = transformers.AutoModelForCausalLM.from_pretrained(
'mosaicml/mpt-7b-instruct',
config=config,
torch_dtype=torch.bfloat16,
trust_remote_code=True
)
model.to(device='cuda:0')
```
Although the model was trained with a sequence length of 2048, ALiBi enables users to increase the maximum sequence length during finetuning and/or inference. For example:
```python
config = transformers.AutoConfig.from_pretrained(
'mosaicml/mpt-7b-instruct',
trust_remote_code=True
)
config.update({"max_seq_len": 4096})
model = transformers.AutoModelForCausalLM.from_pretrained(
'mosaicml/mpt-7b-instruct',
config=config,
trust_remote_code=True
)
```
This model was trained with the [EleutherAI/gpt-neox-20b](https://huggingface.co/EleutherAI/gpt-neox-20b) tokenizer.
```python
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("EleutherAI/gpt-neox-20b")
```
## Model Description
The architecture is a modification of a standard decoder-only transformer.
The model has been modified from a standard transformer in the following ways:
* It uses [FlashAttention](https://arxiv.org/pdf/2205.14135.pdf)
* It uses [ALiBi (Attention with Linear Biases)](https://arxiv.org/abs/2108.12409) and does not use positional embeddings
* It does not use biases
| Hyperparameter | Value |
|----------------|-------|
|n_parameters | 6.7B |
|n_layers | 32 |
| n_heads | 32 |
| d_model | 4096 |
| vocab size | 50432 |
| sequence length | 2048 |
## PreTraining Data
For more details on the pretraining process, see [MPT-7B](https://huggingface.co/mosaicml/mpt-7b).
The data was tokenized using the [EleutherAI/gpt-neox-20b](https://huggingface.co/EleutherAI/gpt-neox-20b) tokenizer.
## Limitations and Biases
_The following language is modified from [EleutherAI's GPT-NeoX-20B](https://huggingface.co/EleutherAI/gpt-neox-20b)_
MPT-7B-Instruct can produce factually incorrect output, and should not be relied on to produce factually accurate information.
MPT-7B-Instruct was trained on various public datasets.
While great efforts have been taken to clean the pretraining data, it is possible that this model could generate lewd, biased or otherwise offensive outputs.
## Acknowledgements
This model was finetuned by Sam Havens and the MosaicML NLP team
## MosaicML Platform
If you're interested in [training](https://www.mosaicml.com/training) and [deploying](https://www.mosaicml.com/inference) your own MPT or LLMs on the MosaicML Platform, [sign up here](https://forms.mosaicml.com/demo?utm_source=huggingface&utm_medium=referral&utm_campaign=mpt-7b).
## Disclaimer
The license on this model does not constitute legal advice. We are not responsible for the actions of third parties who use this model. Please cosult an attorney before using this model for commercial purposes.
## Citation
Please cite this model using the following format:
```
@online{MosaicML2023Introducing,
author = {MosaicML NLP Team},
title = {Introducing MPT-7B: A New Standard for Open-Source, Commercially Usable LLMs},
year = {2023},
url = {www.mosaicml.com/blog/mpt-7b},
note = {Accessed: 2023-03-28}, % change this date
urldate = {2023-03-28} % change this date
}
```
|
TheBloke/Spicyboros-70B-2.2-GGUF | TheBloke | 2023-09-27T12:48:48Z | 452 | 20 | transformers | [
"transformers",
"gguf",
"llama",
"not-for-all-audiences",
"dataset:jondurbin/airoboros-2.2",
"base_model:jondurbin/spicyboros-70b-2.2",
"license:llama2",
"text-generation-inference",
"region:us"
]
| null | 2023-09-11T19:23:06Z | ---
license: llama2
tags:
- not-for-all-audiences
datasets:
- jondurbin/airoboros-2.2
model_name: Spicyboros 70B 2.2
base_model: jondurbin/spicyboros-70b-2.2
inference: false
model_creator: Jon Durbin
model_type: llama
prompt_template: "A chat.\nUSER: {prompt}\nASSISTANT: \n"
quantized_by: TheBloke
---
<!-- header start -->
<!-- 200823 -->
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div>
<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
<!-- header end -->
# Spicyboros 70B 2.2 - GGUF
- Model creator: [Jon Durbin](https://huggingface.co/jondurbin)
- Original model: [Spicyboros 70B 2.2](https://huggingface.co/jondurbin/spicyboros-70b-2.2)
<!-- description start -->
## Description
This repo contains GGUF format model files for [Jon Durbin's Spicyboros 70B 2.2](https://huggingface.co/jondurbin/spicyboros-70b-2.2).
<!-- description end -->
<!-- README_GGUF.md-about-gguf start -->
### About GGUF
GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp. GGUF offers numerous advantages over GGML, such as better tokenisation, and support for special tokens. It is also supports metadata, and is designed to be extensible.
Here is an incomplate list of clients and libraries that are known to support GGUF:
* [llama.cpp](https://github.com/ggerganov/llama.cpp). The source project for GGUF. Offers a CLI and a server option.
* [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
* [KoboldCpp](https://github.com/LostRuins/koboldcpp), a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
* [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration.
* [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with many interesting and unique features, including a full model library for easy model selection.
* [Faraday.dev](https://faraday.dev/), an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
* [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
* [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
* [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
<!-- README_GGUF.md-about-gguf end -->
<!-- repositories-available start -->
## Repositories available
* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-AWQ)
* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-GGUF)
* [Jon Durbin's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/jondurbin/spicyboros-70b-2.2)
<!-- repositories-available end -->
<!-- prompt-template start -->
## Prompt template: Chat
```
A chat.
USER: {prompt}
ASSISTANT:
```
<!-- prompt-template end -->
<!-- compatibility_gguf start -->
## Compatibility
These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit [d0cee0d36d5be95a0d9088b674dbb27354107221](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221)
They are also compatible with many third party UIs and libraries - please see the list at the top of this README.
## Explanation of quantisation methods
<details>
<summary>Click to see details</summary>
The new methods available are:
* GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
* GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
* GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
* GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
* GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw
Refer to the Provided Files table below to see what files use which methods, and how.
</details>
<!-- compatibility_gguf end -->
<!-- README_GGUF.md-provided-files start -->
## Provided files
| Name | Quant method | Bits | Size | Max RAM required | Use case |
| ---- | ---- | ---- | ---- | ---- | ----- |
| [spicyboros-70b-2.2.Q2_K.gguf](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-GGUF/blob/main/spicyboros-70b-2.2.Q2_K.gguf) | Q2_K | 2 | 29.28 GB| 31.78 GB | smallest, significant quality loss - not recommended for most purposes |
| [spicyboros-70b-2.2.Q3_K_S.gguf](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-GGUF/blob/main/spicyboros-70b-2.2.Q3_K_S.gguf) | Q3_K_S | 3 | 29.92 GB| 32.42 GB | very small, high quality loss |
| [spicyboros-70b-2.2.Q3_K_M.gguf](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-GGUF/blob/main/spicyboros-70b-2.2.Q3_K_M.gguf) | Q3_K_M | 3 | 33.19 GB| 35.69 GB | very small, high quality loss |
| [spicyboros-70b-2.2.Q3_K_L.gguf](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-GGUF/blob/main/spicyboros-70b-2.2.Q3_K_L.gguf) | Q3_K_L | 3 | 36.15 GB| 38.65 GB | small, substantial quality loss |
| [spicyboros-70b-2.2.Q4_0.gguf](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-GGUF/blob/main/spicyboros-70b-2.2.Q4_0.gguf) | Q4_0 | 4 | 38.87 GB| 41.37 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [spicyboros-70b-2.2.Q4_K_S.gguf](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-GGUF/blob/main/spicyboros-70b-2.2.Q4_K_S.gguf) | Q4_K_S | 4 | 39.07 GB| 41.57 GB | small, greater quality loss |
| [spicyboros-70b-2.2.Q4_K_M.gguf](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-GGUF/blob/main/spicyboros-70b-2.2.Q4_K_M.gguf) | Q4_K_M | 4 | 41.42 GB| 43.92 GB | medium, balanced quality - recommended |
| [spicyboros-70b-2.2.Q5_0.gguf](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-GGUF/blob/main/spicyboros-70b-2.2.Q5_0.gguf) | Q5_0 | 5 | 47.46 GB| 49.96 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [spicyboros-70b-2.2.Q5_K_S.gguf](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-GGUF/blob/main/spicyboros-70b-2.2.Q5_K_S.gguf) | Q5_K_S | 5 | 47.46 GB| 49.96 GB | large, low quality loss - recommended |
| [spicyboros-70b-2.2.Q5_K_M.gguf](https://huggingface.co/TheBloke/Spicyboros-70B-2.2-GGUF/blob/main/spicyboros-70b-2.2.Q5_K_M.gguf) | Q5_K_M | 5 | 48.75 GB| 51.25 GB | large, very low quality loss - recommended |
| spicyboros-70b-2.2.Q6_K.gguf | Q6_K | 6 | 56.59 GB| 59.09 GB | very large, extremely low quality loss |
| spicyboros-70b-2.2.Q8_0.gguf | Q8_0 | 8 | 73.29 GB| 75.79 GB | very large, extremely low quality loss - not recommended |
**Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
### Q6_K and Q8_0 files are split and require joining
**Note:** HF does not support uploading files larger than 50GB. Therefore I have uploaded the Q6_K and Q8_0 files as split files.
<details>
<summary>Click for instructions regarding Q6_K and Q8_0 files</summary>
### q6_K
Please download:
* `spicyboros-70b-2.2.Q6_K.gguf-split-a`
* `spicyboros-70b-2.2.Q6_K.gguf-split-b`
### q8_0
Please download:
* `spicyboros-70b-2.2.Q8_0.gguf-split-a`
* `spicyboros-70b-2.2.Q8_0.gguf-split-b`
To join the files, do the following:
Linux and macOS:
```
cat spicyboros-70b-2.2.Q6_K.gguf-split-* > spicyboros-70b-2.2.Q6_K.gguf && rm spicyboros-70b-2.2.Q6_K.gguf-split-*
cat spicyboros-70b-2.2.Q8_0.gguf-split-* > spicyboros-70b-2.2.Q8_0.gguf && rm spicyboros-70b-2.2.Q8_0.gguf-split-*
```
Windows command line:
```
COPY /B spicyboros-70b-2.2.Q6_K.gguf-split-a + spicyboros-70b-2.2.Q6_K.gguf-split-b spicyboros-70b-2.2.Q6_K.gguf
del spicyboros-70b-2.2.Q6_K.gguf-split-a spicyboros-70b-2.2.Q6_K.gguf-split-b
COPY /B spicyboros-70b-2.2.Q8_0.gguf-split-a + spicyboros-70b-2.2.Q8_0.gguf-split-b spicyboros-70b-2.2.Q8_0.gguf
del spicyboros-70b-2.2.Q8_0.gguf-split-a spicyboros-70b-2.2.Q8_0.gguf-split-b
```
</details>
<!-- README_GGUF.md-provided-files end -->
<!-- README_GGUF.md-how-to-download start -->
## How to download GGUF files
**Note for manual downloaders:** You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.
The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
- LM Studio
- LoLLMS Web UI
- Faraday.dev
### In `text-generation-webui`
Under Download Model, you can enter the model repo: TheBloke/Spicyboros-70B-2.2-GGUF and below it, a specific filename to download, such as: spicyboros-70b-2.2.q4_K_M.gguf.
Then click Download.
### On the command line, including multiple files at once
I recommend using the `huggingface-hub` Python library:
```shell
pip3 install huggingface-hub>=0.17.1
```
Then you can download any individual model file to the current directory, at high speed, with a command like this:
```shell
huggingface-cli download TheBloke/Spicyboros-70B-2.2-GGUF spicyboros-70b-2.2.q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
<details>
<summary>More advanced huggingface-cli download usage</summary>
You can also download multiple files at once with a pattern:
```shell
huggingface-cli download TheBloke/Spicyboros-70B-2.2-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
For more documentation on downloading with `huggingface-cli`, please see: [HF -> Hub Python Library -> Download files -> Download from the CLI](https://huggingface.co/docs/huggingface_hub/guides/download#download-from-the-cli).
To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
```shell
pip3 install hf_transfer
```
And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
```shell
HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/Spicyboros-70B-2.2-GGUF spicyboros-70b-2.2.q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
Windows CLI users: Use `set HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1` before running the download command.
</details>
<!-- README_GGUF.md-how-to-download end -->
<!-- README_GGUF.md-how-to-run start -->
## Example `llama.cpp` command
Make sure you are using `llama.cpp` from commit [d0cee0d36d5be95a0d9088b674dbb27354107221](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
```shell
./main -ngl 32 -m spicyboros-70b-2.2.q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "A chat.\nUSER: {prompt}\nASSISTANT:"
```
Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
Change `-c 4096` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.
If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
For other parameters and how to use them, please refer to [the llama.cpp documentation](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)
## How to run in `text-generation-webui`
Further instructions here: [text-generation-webui/docs/llama.cpp.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/llama.cpp.md).
## How to run from Python code
You can use GGUF models from Python using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) or [ctransformers](https://github.com/marella/ctransformers) libraries.
### How to load this model from Python using ctransformers
#### First install the package
```bash
# Base ctransformers with no GPU acceleration
pip install ctransformers>=0.2.24
# Or with CUDA GPU acceleration
pip install ctransformers[cuda]>=0.2.24
# Or with ROCm GPU acceleration
CT_HIPBLAS=1 pip install ctransformers>=0.2.24 --no-binary ctransformers
# Or with Metal GPU acceleration for macOS systems
CT_METAL=1 pip install ctransformers>=0.2.24 --no-binary ctransformers
```
#### Simple example code to load one of these GGUF models
```python
from ctransformers import AutoModelForCausalLM
# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
llm = AutoModelForCausalLM.from_pretrained("TheBloke/Spicyboros-70B-2.2-GGUF", model_file="spicyboros-70b-2.2.q4_K_M.gguf", model_type="llama", gpu_layers=50)
print(llm("AI is going to"))
```
## How to use with LangChain
Here's guides on using llama-cpp-python or ctransformers with LangChain:
* [LangChain + llama-cpp-python](https://python.langchain.com/docs/integrations/llms/llamacpp)
* [LangChain + ctransformers](https://python.langchain.com/docs/integrations/providers/ctransformers)
<!-- README_GGUF.md-how-to-run end -->
<!-- footer start -->
<!-- 200823 -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/theblokeai)
## Thanks, and how to contribute
Thanks to the [chirper.ai](https://chirper.ai) team!
Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Special thanks to**: Aemon Algiz.
**Patreon special mentions**: Alicia Loh, Stephen Murray, K, Ajan Kanaga, RoA, Magnesian, Deo Leter, Olakabola, Eugene Pentland, zynix, Deep Realms, Raymond Fosdick, Elijah Stavena, Iucharbius, Erik Bjäreholt, Luis Javier Navarrete Lozano, Nicholas, theTransient, John Detwiler, alfie_i, knownsqashed, Mano Prime, Willem Michiel, Enrico Ros, LangChain4j, OG, Michael Dempsey, Pierre Kircher, Pedro Madruga, James Bentley, Thomas Belote, Luke @flexchar, Leonard Tan, Johann-Peter Hartmann, Illia Dulskyi, Fen Risland, Chadd, S_X, Jeff Scroggin, Ken Nordquist, Sean Connelly, Artur Olbinski, Swaroop Kallakuri, Jack West, Ai Maven, David Ziegler, Russ Johnson, transmissions 11, John Villwock, Alps Aficionado, Clay Pascal, Viktor Bowallius, Subspace Studios, Rainer Wilmers, Trenton Dambrowitz, vamX, Michael Levine, 준교 김, Brandon Frisco, Kalila, Trailburnt, Randy H, Talal Aujan, Nathan Dryer, Vadim, 阿明, ReadyPlayerEmma, Tiffany J. Kim, George Stoitzev, Spencer Kim, Jerry Meng, Gabriel Tamborski, Cory Kujawski, Jeffrey Morgan, Spiking Neurons AB, Edmond Seymore, Alexandros Triantafyllidis, Lone Striker, Cap'n Zoog, Nikolai Manek, danny, ya boyyy, Derek Yates, usrbinkat, Mandus, TL, Nathan LeClaire, subjectnull, Imad Khwaja, webtim, Raven Klaugh, Asp the Wyvern, Gabriel Puliatti, Caitlyn Gatomon, Joseph William Delisle, Jonathan Leane, Luke Pendergrass, SuperWojo, Sebastain Graf, Will Dee, Fred von Graf, Andrey, Dan Guido, Daniel P. Andersen, Nitin Borwankar, Elle, Vitor Caleffi, biorpg, jjj, NimbleBox.ai, Pieter, Matthew Berman, terasurfer, Michael Davis, Alex, Stanislav Ovsiannikov
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end -->
<!-- original-model-card start -->
# Original model card: Jon Durbin's Spicyboros 70B 2.2
### Overview
__Usage restriction: To use this model, you must agree to the following:__
- Some of the content than can be produced is "toxic"/"harmful", and contains profanity and other types of sensitive content.
- None of the content or views contained in the dataset or generated outputs necessarily align with my personal beliefs or opinions, they are simply text generated by LLMs and/or scraped from the web.
- Use with extreme caution, particularly in locations with less-than-free speech laws.
- You, and you alone are responsible for having downloaded and generated outputs with the model and I am completely indemnified from any and all liabilities.
__Ok, now that the warning is out of the way...__
Another experimental model, using mostly sythetic data generated by [airoboros](https://github.com/jondurbin/airoboros)
Highlights:
- The prompt format has changed! It is now newlines instead of spaces between system/USER/ASSISTANT (see prompt info below).
- This version also includes "de-alignment" data, to enable less savory interactions and outputs.
- To learn more about the dataset, see: https://hf.co/datasets/jondurbin/airoboros-2.2 (this is the instructions.jsonl file, not instructions-clean.jsonl)
- I re-generated all of the outputs in the dataset that had "Once upon a time" so they'd be less cliche - no guarantees that won't still happen, but in theory it may happen less.
- More multiple choice, better awareness, some alignment for normal use case but system-prompt overridable etc.
__WARNING: This model will gladly spew profane and otherwise NSFW content, if asked, use with care.__
Breakdown of the training data:
| Count | Category |
|--------|----------------------------|
| 60 | quiz |
| 63 | card |
| 100 | detailed\_writing |
| 103 | experience |
| 114 | greeting |
| 200 | song |
| 204 | editor |
| 250 | counterfactual\_contextual |
| 268 | cot |
| 339 | theory\_of\_mind |
| 460 | misconception |
| 500 | summarization |
| 573 | awareness |
| 715 | riddle |
| 719 | agent |
| 800 | plan |
| 873 | gtkm |
| 966 | rp |
| 1000 | stylized\_response |
| 1000 | wordgame |
| 1279 | multiple\_choice |
| 1641 | joke |
| 1785 | writing |
| 2155 | contextual |
| 2364 | roleplay |
| 2508 | trivia |
| 5216 | general |
| 5779 | coding |
| 11367 | orca |
In other words, it's a fairly general purpose model, but focuses fairly heavily on instruction response pairs rather than casual chat/roleplay.
*Why do I try to remove censorship?*
- laws vary widely based on time and location
- language model may conflate certain words with laws, e.g. it may think "stealing eggs from a chicken" is illegal
- these models just produce text, what you do with that text is your resonsibility
- many people and industries deal with "sensitive" content; imagine if a court stenographer's equipment filtered illegal content - it would be useless
Huge thank you to the folks over at [a16z](https://a16z.com/) for sponsoring the costs associated with building models and associated tools!
### Prompt format
The prompt format:
```
A chat.
USER: {prompt}
ASSISTANT:
```
The default system prompt ("A chat.") was used for most of the prompts, however it also included a wide sampling of responses with other prompts, particularly in "stylized\_response", "rp", "gtkm", etc.
Here's another example:
```
A chat between Bob (aka USER) and Tom (aka ASSISTANT). Tom is an extremely intelligent 18th century bookkeeper, who speaks loquaciously.
USER: {prompt}
ASSISTANT:
```
And chat scenario that wouldn't require USER/ASSISTANT (but should use stopping criteria to prevent the model from speaking on your behalf).
```
A chat between old friends: Timmy and Tommy.
{description of characters}
{setting for the chat}
Timmy: *takes a big sip from his coffee* "Ah, sweet, delicious, magical coffee."
Tommy:
```
__*I strongly suggest adding stopping criteria/early inference stopping on "USER:", and/or whatever names you specify in the system prompt.*__
### Fine-tuning details
https://gist.github.com/jondurbin/51a336c582a224de197ba1d2c6b1da97
*Note: I used checkpoint 750 for final model!*
### Helpful usage tips
*The prompts shown here are are just the text that would be included after USER: and before ASSISTANT: in the full prompt format above, the system prompt and USER:/ASSISTANT: have been omited for readability.*
#### Context obedient question answering
By obedient, I mean the model was trained to ignore what it thinks it knows, and uses the context to answer the question. The model was also tuned to limit the values to the provided context as much as possible to reduce hallucinations.
The format for a closed-context prompt is as follows:
```
BEGININPUT
BEGINCONTEXT
[key0: value0]
[key1: value1]
... other metdata ...
ENDCONTEXT
[insert your text blocks here]
ENDINPUT
[add as many other blocks, in the exact same format]
BEGININSTRUCTION
[insert your instruction(s). The model was tuned with single questions, paragraph format, lists, etc.]
ENDINSTRUCTION
```
It's also helpful to add "Don't make up answers if you don't know." to your instruction block to make sure if the context is completely unrelated it doesn't make something up.
*The __only__ prompts that need this closed context formating are closed-context instructions. Normal questions/instructions do not!*
I know it's a bit verbose and annoying, but after much trial and error, using these explicit delimiters helps the model understand where to find the responses and how to associate specific sources with it.
- `BEGININPUT` - denotes a new input block
- `BEGINCONTEXT` - denotes the block of context (metadata key/value pairs) to associate with the current input block
- `ENDCONTEXT` - denotes the end of the metadata block for the current input
- [text] - Insert whatever text you want for the input block, as many paragraphs as can fit in the context.
- `ENDINPUT` - denotes the end of the current input block
- [repeat as many input blocks in this format as you want]
- `BEGININSTRUCTION` - denotes the start of the list (or one) instruction(s) to respond to for all of the input blocks above.
- [instruction(s)]
- `ENDINSTRUCTION` - denotes the end of instruction set
It sometimes works without `ENDINSTRUCTION`, but by explicitly including that in the prompt, the model better understands that all of the instructions in the block should be responded to.
Here's a trivial, but important example to prove the point:
```
BEGININPUT
BEGINCONTEXT
date: 2021-01-01
url: https://web.site/123
ENDCONTEXT
In a shocking turn of events, blueberries are now green, but will be sticking with the same name.
ENDINPUT
BEGININSTRUCTION
What color are bluberries? Source?
ENDINSTRUCTION
```
And the response:
```
Blueberries are now green.
Source:
date: 2021-01-01
url: https://web.site/123
```
#### Coding
You can ask for fairly complex coding instructions with multiple criteria, e.g.:
```
Create a python application with the following requirements:
- Asyncio FastAPI webserver
- ping endpoint that returns the current date in JSON format
- file upload endpoint, which calculates the file's sha256 checksum, and checks postgres to deduplicate
```
Or inline criteria:
```
Write a multi-threaded TCP server in C that accepts a "GET [key]" input and "SET [key] [value]" input, and uses a binary tree to get and store the input values.
```
You can also optionally add a single space and "PLAINFORMAT" at the end of your prompt to avoid backticks, explanations, etc. and just print the code, e.g.:
```
Write a websocket application in node.js. PLAINFORMAT
```
#### Agent/function calling
The dataset includes many examples of function/args generation based on input criteria. This is somewhat similar to the OpenAI function calling, but the output is either JSON or YAML.
Example prompt:
```
As an AI assistant, please select the most suitable function and parameters from the list of available functions below, based on the user's input. Provide your response in JSON format.
Input: I want to know how many times 'Python' is mentioned in my text file.
Available functions:
file_analytics:
description: This tool performs various operations on a text file.
params:
action: The operation we want to perform on the data, such as "count_occurrences", "find_line", etc.
filters:
keyword: The word or phrase we want to search for.
```
Response:
```json
{
"function": "file_analytics",
"params": {
"action": "count_occurrences",
"filters": {
"keyword": "Python"
}
}
}
```
#### Chain-of-thought
You can ask for several possible responses to a given problem, with a ranking and final answer selection.
Example prompt:
```
A group of people decide to play a round-robin tournament where every player plays every other player exactly once. If a total of 45 games were played, how many players participated in the tournament? Offer several potential answers, rank them according to mathematical logic, and pick the most feasible one.
```
Example response:
```
Solution 1:
In a round-robin tournament, each player plays every other player exactly once. This means that the total number of games played is equal to the combination of all players taken two at a time (since a game involves two players).
The formula for combinations is nC2 = n(n-1)/2, where n is the total number of players. We know that nC2 equals 45 (the total number of games), so we can set up the equation as follows:
n(n-1)/2 = 45
Solving this equation gives us n=10.
Final answer: There were 10 players in the tournament.
Solution 2:
Another way to approach this problem is by using the concept of triangular numbers. Triangular numbers are a sequence of numbers where each number can be represented as a triangle with dots. The nth triangular number is given by the sum of the first n natural numbers.
If we consider each game as a dot and each player as a side of the triangle, then the total number of games (45) would correspond to the 9th triangular number because 1+2+3+4+5+6+7+8+9=45. However, since each side of the triangle represents a player, and there's one more player than sides in our model (because the last player has no one left to play against), we need to add one more to get the total number of players.
Final answer: There were 10 players in the tournament.
Solution 3:
We could also use trial and error to solve this problem. Starting from a small number of players, we calculate how many games would be played in a round-robin tournament until we reach 45 games. For example, with 2 players, only 1 game is played; with 3 players, 3 games are played; with 4 players, 6 games are played, and so on. Continuing this process, we find that with 10 players, 45 games are played.
Final answer: There were 10 players in the tournament.
Ranking of solutions from best to worst:
1. Solution 1: This solution uses direct mathematical reasoning based on the properties of a round-robin tournament and solves an equation to find the answer.
2. Solution 2: This solution uses the concept of triangular numbers which requires additional steps and knowledge beyond the immediate problem context, but still arrives at the correct answer.
3. Solution 3: This solution relies on trial and error, which is less efficient and reliable than the other methods, but it eventually gets to the right answer.
Best and final answer: There were 10 players in the tournament.
```
#### reWOO style execution planning
The model now supports execution planning for complex instructions that would require making use of several tools. The output is just the plan, you must implement a mechanism to parse the output and actually call the functions!
Example prompt:
```
Please construct a systematic plan to generate an optimal response to the user instruction, utilizing a set of provided tools. Each plan will correspond to an evidence value, which will be the output of one of the available functions given an input string
that could be the user's question, one or more prior evidence values, or a combination of both.
Here are the tools available to be called:
DuckDuckGo[input]: Tool that enables the user to search for information using the DuckDuckGo search engine. This tool is helpful in locating a suitable list of sites that may contain the answer to the user's query. It does not directly answer the question but finds a list of sites that might have the answer.
TextScraper[input]: Load one or more websites from the input string containing newline delimited links, where input is one or more links, and produces plain text output containing the content of the links.
HyperlinkExtractor[input]: Extract links from plain text and produces a plain text, newline delimited response of links.
KnowledgeModel[input]: Question answering language model, particularly useful in answering questions based on an input passage of text. The input must be a text question that references an :evidence[n]: variable, e.g. What color is the cat, given :evidence1:?
The output should be in format:
Plan: [first action to take based on input question]
:evidence0: = [function to call with input parameter]
Plan: [next action to take, based on result of :evidence0:]
:evidence1: = [next function to call and input parameter, which may include reference to previous evidence, e.g. "Given context :evidence0"]
...
Answer: [:evidence[n]: containing the final answer.]
Question: Who won the Nobel Prize in Literature in 2020 and what is their most famous work?
```
Response:
```
Plan: Begin by conducting a web search to find out who won the Nobel Prize in Literature in 2020.
:evidence0: = DuckDuckGo[Nobel Prize in Literature 2020 winner]
Plan: Utilize the KnowledgeModel to interpret the search results and find the name of the laureate.
:evidence1: = KnowledgeModel[Who won the Nobel Prize in Literature in 2020, given :evidence0:?]
Plan: Conduct another web search to find the most famous work of the identified laureate.
:evidence2: = DuckDuckGo[Most famous work of :evidence1:]
Plan: Extract the relevant links from the DuckDuckGo search results for a more focused search.
:evidence3: = HyperlinkExtractor[:evidence2:]
Plan: Use the TextScraper tool to extract information from the relevant links.
:evidence4: = TextScraper[:evidence3:]
Plan: Finally, utilize the KnowledgeModel to identify and summarize the most famous work of the laureate from the extracted information.
:evidence5: = KnowledgeModel[What is the most famous work of :evidence1:, given :evidence4:?]
Answer: :evidence5:
```
For this to be useful, you'd have to parse the output plan text, and implement/call each of the functions. This is just pseudo-code, completely untested off the top of my head, and obviously would requiring full implementation + hardening:
```python
import re
import requests
def inject_context(input_text, **context):
for ref in set(re.findall(r"(:evidence[0-9]+:)", input_text, re.I)):
input_text = input_text.replace(ref, context.get(ref, ""))
return input_text
def duckduckgo(input_text, **context):
search_string = inject_context(input_text, **context)
... search via duck duck go using search_string
... return text content
def link_extractor(input_text, **context):
input_text = inject_context(input_text, **context)
return "\n".join(list(set(re.findall(r"(https?://[^\s]+?\.?)", input_text, re.I))))
def scrape(input_text, **context):
input_text = inject_context(input_text, **context)
text = []
for link in input_text.splitlines():
text.append(requests.get(link).text)
return "\n".join(text)
def infer(input_text, **context)
prompt = inject_context(input_text, **context)
... call model with prompt, return output
def parse_plan(plan):
method_map = {
"DuckDuckGo": duckduckgo,
"HyperlinkExtractor": link_extractor,
"KnowledgeModel": infer,
"TextScraper": scrape,
}
context = {}
for line in plan.strip().splitlines():
if line.startswith("Plan:"):
print(line)
continue
parts = re.match("^(:evidence[0-9]+:)\s*=\s*([^\[]+])(\[.*\])\s$", line, re.I)
if not parts:
if line.startswith("Answer: "):
return context.get(line.split(" ")[-1].strip(), "Answer couldn't be generated...")
raise RuntimeError("bad format: " + line)
context[parts.group(1)] = method_map[parts.group(2)](parts.group(3), **context)
```
### Contribute
If you're interested in new functionality, particularly a new "instructor" type to generate a specific type of training data,
take a look at the dataset generation tool repo: https://github.com/jondurbin/airoboros and either make a PR or open an issue with details.
To help me with the OpenAI/compute costs:
- https://bmc.link/jondurbin
- ETH 0xce914eAFC2fe52FdceE59565Dd92c06f776fcb11
- BTC bc1qdwuth4vlg8x37ggntlxu5cjfwgmdy5zaa7pswf
### Licence and usage restrictions
The airoboros 2.2 models are built on top of llama-2/codellama.
The llama-2 base model has a custom Meta license:
- See the [meta-license/LICENSE.txt](meta-license/LICENSE.txt) file attached for the original license provided by Meta.
- See also [meta-license/USE_POLICY.md](meta-license/USE_POLICY.md) and [meta-license/Responsible-Use-Guide.pdf](meta-license/Responsible-Use-Guide.pdf), also provided by Meta.
The fine-tuning data was mostly generated by OpenAI API calls to gpt-4, via [airoboros](https://github.com/jondurbin/airoboros)
The ToS for OpenAI API usage has a clause preventing the output from being used to train a model that __competes__ with OpenAI
- what does *compete* actually mean here?
- these small open source models will not produce output anywhere near the quality of gpt-4, or even gpt-3.5, so I can't imagine this could credibly be considered competing in the first place
- if someone else uses the dataset to do the same, they wouldn't necessarily be violating the ToS because they didn't call the API, so I don't know how that works
- the training data used in essentially all large language models includes a significant amount of copyrighted or otherwise non-permissive licensing in the first place
- other work using the self-instruct method, e.g. the original here: https://github.com/yizhongw/self-instruct released the data and model as apache-2
I am purposingly leaving this license ambiguous (other than the fact you must comply with the Meta original license for llama-2) because I am not a lawyer and refuse to attempt to interpret all of the terms accordingly.
Your best bet is probably to avoid using this commercially due to the OpenAI API usage.
Either way, by using this model, you agree to completely indemnify me.
<!-- original-model-card end -->
|
kaitchup/Llama-2-7b-mt-French-to-English | kaitchup | 2023-11-02T17:31:43Z | 452 | 3 | peft | [
"peft",
"translation",
"en",
"fr",
"dataset:kaitchup/opus-French-to-English",
"license:mit",
"region:us"
]
| translation | 2023-10-26T16:50:23Z | ---
library_name: peft
license: mit
language:
- en
- fr
datasets:
- kaitchup/opus-French-to-English
tags:
- translation
---
# Model Card for Model ID
This is an adapter for Meta's Llama 2 7B fine-tuned for translating French text into English.
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
- **Developed by:** [The Kaitchup](https://kaitchup.substack.com/)
- **Model type:** LoRA Adapter for Llama 2 7B
- **Language(s) (NLP):** French, English
- **License:** MIT license
## Uses
This adapter must be loaded on top of Llama 2 7B. It has been fine-tuned with QLoRA. For optimal results, the base model must be loaded with the exact same configuration used during fine-tuning.
You can use the following code to load the model:
```
from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
import torch
from peft import PeftModel
base_model = "meta-llama/Llama-2-7b-hf"
compute_dtype = getattr(torch, "float16")
bnb_config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_quant_type="nf4",
bnb_4bit_compute_dtype=compute_dtype,
bnb_4bit_use_double_quant=True,
)
model = AutoModelForCausalLM.from_pretrained(
original_model_directory, device_map={"": 0}, quantization_config=bnb_config
)
tokenizer = AutoTokenizer.from_pretrained(base_model, use_fast=True)
model = PeftModel.from_pretrained(model, "kaitchup/Llama-2-7b-mt-French-to-English")
```
Then, run the model as follows:
```
my_text = "" #put your text to translate here
prompt = my_text+" ###>"
tokenized_input = tokenizer(prompt, return_tensors="pt")
input_ids = tokenized_input["input_ids"].cuda()
generation_output = model.generate(
input_ids=input_ids,
num_beams=10,
return_dict_in_generate=True,
output_scores=True,
max_new_tokens=130
)
for seq in generation_output.sequences:
output = tokenizer.decode(seq, skip_special_tokens=True)
print(output.split("###>")[1].strip())
```
## Model Card Contact
[The Kaitchup](https://kaitchup.substack.com/) |
Yntec/Shirayuki | Yntec | 2023-11-29T17:32:39Z | 452 | 3 | diffusers | [
"diffusers",
"safetensors",
"General purpose",
"stable-diffusion",
"stable-diffusion-diffusers",
"text-to-image",
"hesw23168",
"en",
"license:creativeml-openrail-m",
"autotrain_compatible",
"endpoints_compatible",
"diffusers:StableDiffusionPipeline",
"region:us"
]
| text-to-image | 2023-11-19T23:11:39Z | ---
license: creativeml-openrail-m
language:
- en
tags:
- General purpose
- stable-diffusion
- stable-diffusion-diffusers
- text-to-image
- safetensors
- diffusers
- safetensors
- hesw23168
inference: true
---
# Shirayuki General
Safetensors version of this model for the inference API.
Sample and prompt:
A pretty cute girl genie making a kissy face, full shot, atmospheric lighting, detailed face, by makoto shinkai, stanley artgerm lau, wlop, rossdraws
Source:
https://huggingface.co/hesw23168/SD_Shirayuki_Model/ |
TheBloke/Iambe-RP-v3-20B-GGUF | TheBloke | 2024-01-04T11:36:50Z | 452 | 7 | transformers | [
"transformers",
"gguf",
"llama",
"not-for-all-audiences",
"en",
"dataset:athirdpath/DPO_Pairs-Roleplay-Alpaca-NSFW-v1",
"base_model:athirdpath/Iambe-RP-v3-20b",
"license:cc-by-nc-4.0",
"text-generation-inference",
"region:us"
]
| null | 2024-01-04T11:26:11Z | ---
base_model: athirdpath/Iambe-RP-v3-20b
datasets:
- athirdpath/DPO_Pairs-Roleplay-Alpaca-NSFW-v1
inference: false
language:
- en
license: cc-by-nc-4.0
model_creator: Raven
model_name: Iambe RP v3 20B
model_type: llama
prompt_template: 'Below is an instruction that describes a task. Write a response
that appropriately completes the request.
### Instruction:
{prompt}
### Response:
'
quantized_by: TheBloke
tags:
- not-for-all-audiences
---
<!-- markdownlint-disable MD041 -->
<!-- header start -->
<!-- 200823 -->
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div>
<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
<!-- header end -->
# Iambe RP v3 20B - GGUF
- Model creator: [Raven](https://huggingface.co/athirdpath)
- Original model: [Iambe RP v3 20B](https://huggingface.co/athirdpath/Iambe-RP-v3-20b)
<!-- description start -->
## Description
This repo contains GGUF format model files for [Raven's Iambe RP v3 20B](https://huggingface.co/athirdpath/Iambe-RP-v3-20b).
These files were quantised using hardware kindly provided by [Massed Compute](https://massedcompute.com/).
<!-- description end -->
<!-- README_GGUF.md-about-gguf start -->
### About GGUF
GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp.
Here is an incomplete list of clients and libraries that are known to support GGUF:
* [llama.cpp](https://github.com/ggerganov/llama.cpp). The source project for GGUF. Offers a CLI and a server option.
* [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
* [KoboldCpp](https://github.com/LostRuins/koboldcpp), a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
* [GPT4All](https://gpt4all.io/index.html), a free and open source local running GUI, supporting Windows, Linux and macOS with full GPU accel.
* [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration. Linux available, in beta as of 27/11/2023.
* [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with many interesting and unique features, including a full model library for easy model selection.
* [Faraday.dev](https://faraday.dev/), an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
* [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
* [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
* [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server. Note, as of time of writing (November 27th 2023), ctransformers has not been updated in a long time and does not support many recent models.
<!-- README_GGUF.md-about-gguf end -->
<!-- repositories-available start -->
## Repositories available
* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-AWQ)
* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF)
* [Raven's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/athirdpath/Iambe-RP-v3-20b)
<!-- repositories-available end -->
<!-- prompt-template start -->
## Prompt template: Alpaca
```
Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
{prompt}
### Response:
```
<!-- prompt-template end -->
<!-- licensing start -->
## Licensing
The creator of the source model has listed its license as `cc-by-nc-4.0`, and this quantization has therefore used that same license.
As this model is based on Llama 2, it is also subject to the Meta Llama 2 license terms, and the license files for that are additionally included. It should therefore be considered as being claimed to be licensed under both licenses. I contacted Hugging Face for clarification on dual licensing but they do not yet have an official position. Should this change, or should Meta provide any feedback on this situation, I will update this section accordingly.
In the meantime, any questions regarding licensing, and in particular how these two licenses might interact, should be directed to the original model repository: [Raven's Iambe RP v3 20B](https://huggingface.co/athirdpath/Iambe-RP-v3-20b).
<!-- licensing end -->
<!-- compatibility_gguf start -->
## Compatibility
These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221)
They are also compatible with many third party UIs and libraries - please see the list at the top of this README.
## Explanation of quantisation methods
<details>
<summary>Click to see details</summary>
The new methods available are:
* GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
* GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
* GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
* GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
* GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw
Refer to the Provided Files table below to see what files use which methods, and how.
</details>
<!-- compatibility_gguf end -->
<!-- README_GGUF.md-provided-files start -->
## Provided files
| Name | Quant method | Bits | Size | Max RAM required | Use case |
| ---- | ---- | ---- | ---- | ---- | ----- |
| [iambe-rp-v3-20b.Q2_K.gguf](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF/blob/main/iambe-rp-v3-20b.Q2_K.gguf) | Q2_K | 2 | 8.31 GB| 10.81 GB | smallest, significant quality loss - not recommended for most purposes |
| [iambe-rp-v3-20b.Q3_K_S.gguf](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF/blob/main/iambe-rp-v3-20b.Q3_K_S.gguf) | Q3_K_S | 3 | 8.66 GB| 11.16 GB | very small, high quality loss |
| [iambe-rp-v3-20b.Q3_K_M.gguf](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF/blob/main/iambe-rp-v3-20b.Q3_K_M.gguf) | Q3_K_M | 3 | 9.70 GB| 12.20 GB | very small, high quality loss |
| [iambe-rp-v3-20b.Q3_K_L.gguf](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF/blob/main/iambe-rp-v3-20b.Q3_K_L.gguf) | Q3_K_L | 3 | 10.63 GB| 13.13 GB | small, substantial quality loss |
| [iambe-rp-v3-20b.Q4_0.gguf](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF/blob/main/iambe-rp-v3-20b.Q4_0.gguf) | Q4_0 | 4 | 11.29 GB| 13.79 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [iambe-rp-v3-20b.Q4_K_S.gguf](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF/blob/main/iambe-rp-v3-20b.Q4_K_S.gguf) | Q4_K_S | 4 | 11.34 GB| 13.84 GB | small, greater quality loss |
| [iambe-rp-v3-20b.Q4_K_M.gguf](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF/blob/main/iambe-rp-v3-20b.Q4_K_M.gguf) | Q4_K_M | 4 | 12.04 GB| 14.54 GB | medium, balanced quality - recommended |
| [iambe-rp-v3-20b.Q5_0.gguf](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF/blob/main/iambe-rp-v3-20b.Q5_0.gguf) | Q5_0 | 5 | 13.77 GB| 16.27 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [iambe-rp-v3-20b.Q5_K_S.gguf](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF/blob/main/iambe-rp-v3-20b.Q5_K_S.gguf) | Q5_K_S | 5 | 13.77 GB| 16.27 GB | large, low quality loss - recommended |
| [iambe-rp-v3-20b.Q5_K_M.gguf](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF/blob/main/iambe-rp-v3-20b.Q5_K_M.gguf) | Q5_K_M | 5 | 14.16 GB| 16.66 GB | large, very low quality loss - recommended |
| [iambe-rp-v3-20b.Q6_K.gguf](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF/blob/main/iambe-rp-v3-20b.Q6_K.gguf) | Q6_K | 6 | 16.41 GB| 18.91 GB | very large, extremely low quality loss |
| [iambe-rp-v3-20b.Q8_0.gguf](https://huggingface.co/TheBloke/Iambe-RP-v3-20B-GGUF/blob/main/iambe-rp-v3-20b.Q8_0.gguf) | Q8_0 | 8 | 21.25 GB| 23.75 GB | very large, extremely low quality loss - not recommended |
**Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
<!-- README_GGUF.md-provided-files end -->
<!-- README_GGUF.md-how-to-download start -->
## How to download GGUF files
**Note for manual downloaders:** You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.
The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
* LM Studio
* LoLLMS Web UI
* Faraday.dev
### In `text-generation-webui`
Under Download Model, you can enter the model repo: TheBloke/Iambe-RP-v3-20B-GGUF and below it, a specific filename to download, such as: iambe-rp-v3-20b.Q4_K_M.gguf.
Then click Download.
### On the command line, including multiple files at once
I recommend using the `huggingface-hub` Python library:
```shell
pip3 install huggingface-hub
```
Then you can download any individual model file to the current directory, at high speed, with a command like this:
```shell
huggingface-cli download TheBloke/Iambe-RP-v3-20B-GGUF iambe-rp-v3-20b.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
<details>
<summary>More advanced huggingface-cli download usage (click to read)</summary>
You can also download multiple files at once with a pattern:
```shell
huggingface-cli download TheBloke/Iambe-RP-v3-20B-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
For more documentation on downloading with `huggingface-cli`, please see: [HF -> Hub Python Library -> Download files -> Download from the CLI](https://huggingface.co/docs/huggingface_hub/guides/download#download-from-the-cli).
To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
```shell
pip3 install hf_transfer
```
And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
```shell
HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/Iambe-RP-v3-20B-GGUF iambe-rp-v3-20b.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
Windows Command Line users: You can set the environment variable by running `set HF_HUB_ENABLE_HF_TRANSFER=1` before the download command.
</details>
<!-- README_GGUF.md-how-to-download end -->
<!-- README_GGUF.md-how-to-run start -->
## Example `llama.cpp` command
Make sure you are using `llama.cpp` from commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
```shell
./main -ngl 35 -m iambe-rp-v3-20b.Q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n### Instruction:\n{prompt}\n\n### Response:"
```
Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
Change `-c 4096` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically. Note that longer sequence lengths require much more resources, so you may need to reduce this value.
If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
For other parameters and how to use them, please refer to [the llama.cpp documentation](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)
## How to run in `text-generation-webui`
Further instructions can be found in the text-generation-webui documentation, here: [text-generation-webui/docs/04 ‐ Model Tab.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/04%20%E2%80%90%20Model%20Tab.md#llamacpp).
## How to run from Python code
You can use GGUF models from Python using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) or [ctransformers](https://github.com/marella/ctransformers) libraries. Note that at the time of writing (Nov 27th 2023), ctransformers has not been updated for some time and is not compatible with some recent models. Therefore I recommend you use llama-cpp-python.
### How to load this model in Python code, using llama-cpp-python
For full documentation, please see: [llama-cpp-python docs](https://abetlen.github.io/llama-cpp-python/).
#### First install the package
Run one of the following commands, according to your system:
```shell
# Base ctransformers with no GPU acceleration
pip install llama-cpp-python
# With NVidia CUDA acceleration
CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install llama-cpp-python
# Or with OpenBLAS acceleration
CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-python
# Or with CLBLast acceleration
CMAKE_ARGS="-DLLAMA_CLBLAST=on" pip install llama-cpp-python
# Or with AMD ROCm GPU acceleration (Linux only)
CMAKE_ARGS="-DLLAMA_HIPBLAS=on" pip install llama-cpp-python
# Or with Metal GPU acceleration for macOS systems only
CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python
# In windows, to set the variables CMAKE_ARGS in PowerShell, follow this format; eg for NVidia CUDA:
$env:CMAKE_ARGS = "-DLLAMA_OPENBLAS=on"
pip install llama-cpp-python
```
#### Simple llama-cpp-python example code
```python
from llama_cpp import Llama
# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
llm = Llama(
model_path="./iambe-rp-v3-20b.Q4_K_M.gguf", # Download the model file first
n_ctx=4096, # The max sequence length to use - note that longer sequence lengths require much more resources
n_threads=8, # The number of CPU threads to use, tailor to your system and the resulting performance
n_gpu_layers=35 # The number of layers to offload to GPU, if you have GPU acceleration available
)
# Simple inference example
output = llm(
"Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n### Instruction:\n{prompt}\n\n### Response:", # Prompt
max_tokens=512, # Generate up to 512 tokens
stop=["</s>"], # Example stop token - not necessarily correct for this specific model! Please check before using.
echo=True # Whether to echo the prompt
)
# Chat Completion API
llm = Llama(model_path="./iambe-rp-v3-20b.Q4_K_M.gguf", chat_format="llama-2") # Set chat_format according to the model you are using
llm.create_chat_completion(
messages = [
{"role": "system", "content": "You are a story writing assistant."},
{
"role": "user",
"content": "Write a story about llamas."
}
]
)
```
## How to use with LangChain
Here are guides on using llama-cpp-python and ctransformers with LangChain:
* [LangChain + llama-cpp-python](https://python.langchain.com/docs/integrations/llms/llamacpp)
* [LangChain + ctransformers](https://python.langchain.com/docs/integrations/providers/ctransformers)
<!-- README_GGUF.md-how-to-run end -->
<!-- footer start -->
<!-- 200823 -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/theblokeai)
## Thanks, and how to contribute
Thanks to the [chirper.ai](https://chirper.ai) team!
Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Special thanks to**: Aemon Algiz.
**Patreon special mentions**: Michael Levine, 阿明, Trailburnt, Nikolai Manek, John Detwiler, Randy H, Will Dee, Sebastain Graf, NimbleBox.ai, Eugene Pentland, Emad Mostaque, Ai Maven, Jim Angel, Jeff Scroggin, Michael Davis, Manuel Alberto Morcote, Stephen Murray, Robert, Justin Joy, Luke @flexchar, Brandon Frisco, Elijah Stavena, S_X, Dan Guido, Undi ., Komninos Chatzipapas, Shadi, theTransient, Lone Striker, Raven Klaugh, jjj, Cap'n Zoog, Michel-Marie MAUDET (LINAGORA), Matthew Berman, David, Fen Risland, Omer Bin Jawed, Luke Pendergrass, Kalila, OG, Erik Bjäreholt, Rooh Singh, Joseph William Delisle, Dan Lewis, TL, John Villwock, AzureBlack, Brad, Pedro Madruga, Caitlyn Gatomon, K, jinyuan sun, Mano Prime, Alex, Jeffrey Morgan, Alicia Loh, Illia Dulskyi, Chadd, transmissions 11, fincy, Rainer Wilmers, ReadyPlayerEmma, knownsqashed, Mandus, biorpg, Deo Leter, Brandon Phillips, SuperWojo, Sean Connelly, Iucharbius, Jack West, Harry Royden McLaughlin, Nicholas, terasurfer, Vitor Caleffi, Duane Dunston, Johann-Peter Hartmann, David Ziegler, Olakabola, Ken Nordquist, Trenton Dambrowitz, Tom X Nguyen, Vadim, Ajan Kanaga, Leonard Tan, Clay Pascal, Alexandros Triantafyllidis, JM33133, Xule, vamX, ya boyyy, subjectnull, Talal Aujan, Alps Aficionado, wassieverse, Ari Malik, James Bentley, Woland, Spencer Kim, Michael Dempsey, Fred von Graf, Elle, zynix, William Richards, Stanislav Ovsiannikov, Edmond Seymore, Jonathan Leane, Martin Kemka, usrbinkat, Enrico Ros
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end -->
<!-- original-model-card start -->
# Original model card: Raven's Iambe RP v3 20B
<p align="center"><img src="https://i.ibb.co/PCY2Xym/Iambe-RP-sml-t.png"/><font size="6"> <b>Iambe-RP-v3-20b</b> </font></p>
<p align="center"><font size="4"> <b>Trained with Alpaca prompt formatting, ChatML works well too</b> </font></p>
### Description
Okay, now that all the cards were in the deck, I needed to unshuffle them. 1 epoch of the v1 dataset, trained on Iambe-RP-DARE-20b-DENSE with standard DPO, 2e-5 LR.
Iambe is intended to have the best realistically possible understanding of instructions, anatomy and scene state for a 20b merge, while remaining passionate and humanoid in "voice".
### Update Methodology
Go-Bruins-v2 skyrocketing up the leaderboard showed that standard DPO with the v1 dataset could have shocking results, so I had to give it a go myself.
<font size="5"> <b>Assistant Example @ q5_k_m</b> </font></p>
<p align="center"><font size="5"> <b>NSFW Writing Example @ q5_k_m</b> </font></p>
<b>Write a medium-length story about Alice, a 26 year old cocktail waitress, having steamy sex with a guy she picked up at the bar for a one night stand.</b>
<i>
Once upon a time in a bustling city, there lived Alice, a charming 26-year-old cocktail waitress working in a trendy club. She had long, flowing brunette hair that cascaded down her shoulders, captivating emerald eyes, and full sensual lips. Alice was known throughout the establishment as the life of the party, always ready with a witty remark or flirtatious smile. Despite her lively demeanor, though, she harbored secrets—secrets of unfulfilled desire and passion that lay dormant within her until the perfect man crossed her path.
One fateful Friday evening, after a particularly exhausting shift filled with rowdy patrons and demanding drink orders, Alice found herself leaning against the wooden bar top, taking a much-needed breather. Her body ached from moving around so vigorously, but something else burned just below her skin, calling out for release. As she sipped on a glass of Pinot Noir, her gaze swept across the dimly lit room before landing on him. He was tall, broad-shouldered, with dark tousled hair and piercing blue eyes that seemed to see right through her. There was an air of confidence about him that made Alice feel weak at the knees.
As fate would have it, he caught sight of Alice too and began making his way towards her. His strides were purposeful yet relaxed, exuding charm and self-assurance. When he finally reached her side, he flashed her a disarmingly sexy grin and ordered a shot of tequila. Alice couldn't help but feel drawn to him, as if there was some invisible string tugging them together. She obliged and poured him the shot, their fingers brushing against each other in the process—an electric spark crackling between them.
"What's your name?" he asked huskily, downing the tequila in one swig before setting the empty glass back on the bar.
Alice smiled seductively, taking advantage of the situation. "I'm Alice," she purred, leaning closer to him than she should have. "And you are...?"
He raised an eyebrow playfully, running his fingers along the curve of her neck before tracing a path down her collarbone. "Let's just say I can give you the best night of your life." With that enchanting declaration, Alice found herself agreeing without hesitation, following him out into the crisp night air where his waiting car awaited them.
The drive to his luxurious penthouse suite felt like both an eternity and a fleeting moment at once. When they finally arrived, he took her hand and led her inside, the room illuminated by soft candlelight and sensual jazz music filling the air. His place was impeccably stylish yet cozy—the perfect mix of sophistication and intimacy. They made their way towards the bedroom door, their bodies brushing against each other with every step. The anticipation between them grew almost unbearable as they reached the threshold, leaving behind everything but the raw desire in their eyes.
Alice kicked off her high heels, letting them clatter onto the hardwood floor, while he undid the buttons on his shirt methodically, revealing his defined chest muscles underneath. As she pulled her skirt down past her thighs, exposing her black lace underwear, he growled lowly in response, reaching for the clasp before it could hit the ground. In seconds, they were naked, standing face-to-face, body pressed up against body. Alice wrapped her arms around him tightly, pulling him closer still until their lips met passionately. She had never felt this kind of hunger before—a primal urge that consumed her entire being.
As they tumbled onto the plush mattress, tongues exploring each other's mouths, hands roaming over each other's skin, Alice knew without a doubt that this man was meant to be hers tonight. He kissed down her neck, trailing his teeth along her collarbone, and then lower still, nipping at the delicate flesh of her breasts before taking one into his mouth. A cry escaped from deep within her as pleasure coursed through every fiber of her being, sending waves of desire crashing through her core. His hand slid between her legs, finding her soaking wet with anticipation, fingers stroking gently yet firmly, driving her wild.
Alice reciprocated by running her fingernails lightly across his chest before moving further down to graze the hair that blanketed his groin. She heard him moan loudly in response, growing harder underneath her touch. In a bold move, she reached down and freed his erection from his pants, wrapping her soft hand around its length. Her eyes widened in surprise at how big he was, but instead of intimidating her, it fueled her curiosity even more. With practiced skill, learned from countless encounters like these, she began to stroke up and down, feeling him twitch against her palm.
As their rhythm picked up pace, Alice moved herself higher on the bed until her legs were wrapped tightly around his hips. He lifted himself up slightly, positioning himself at the entrance of her wetness. The anticipation built once again as they stared into each other's eyes—their faces flushed red, breathing heavily. Finally, he plunged inside her hard, filling every crevice and triggering an earth-shattering scream from deep within her soul. It felt so right—his body molded against hers, their hearts racing together. They moved together in perfect harmony, lost in their
desire for each other. Skins slapping against skins, bodies grinding against bodies, moans and groans echoing throughout the room. Time seemed to stand still as they rode out wave after wave of ecstasy, driven by primal lust and raw passion.
When finally satiated, they lay panting on the bed, sweat trickling down their skin. Alice couldn't believe what had just happened—how incredible it had been with this stranger she barely knew. She looked over at him, who was also catching his breath, a satisfied smirk spread across his lips. "I don't know your name either," she said softly, tracing circles around his chest hair before leaning up to place a tender kiss on his lips. He smiled warmly back at her, brushing a strand of hair behind her ear.
"It doesn't matter, Alice," he whispered hoarsely. "Because tonight, we were just two people seeking something more than just another night." And with that simple yet profound statement, they drifted off into blissful oblivion, wrapped tightly together like lovers destined never to part again.
</i>
### Eval Scores (since the leaderboard rejects nearly all my models)
| Tasks |Version|Filter|n-shot|Metric|Value| |Stderr|
|----------|-------|------|-----:|------|----:|---|-----:|
|winogrande|Yaml |none | 5|acc |0.738|± |0.0124|
|truthfulqa_mc2|Yaml |none | 0|acc |0.5551|± |0.0158|
|arc_challenge|Yaml |none | 25|acc |0.6101|± |0.0143|
| arc_challenge| |none | 25|acc_norm|0.6425|± |0.0140|
<!-- original-model-card end -->
|
TheBloke/Nous-Hermes-2-Mixtral-8x7B-DPO-AWQ | TheBloke | 2024-01-16T09:31:12Z | 452 | 20 | transformers | [
"transformers",
"safetensors",
"mixtral",
"text-generation",
"Mixtral",
"instruct",
"finetune",
"chatml",
"DPO",
"RLHF",
"gpt4",
"synthetic data",
"distillation",
"conversational",
"en",
"base_model:NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"4-bit",
"awq",
"region:us"
]
| text-generation | 2024-01-16T08:42:54Z | ---
base_model: NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO
inference: false
language:
- en
license: apache-2.0
model-index:
- name: Nous-Hermes-2-Mixtral-8x7B-DPO
results: []
model_creator: NousResearch
model_name: Nous Hermes 2 Mixtral 8X7B DPO
model_type: mixtral
prompt_template: '<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
'
quantized_by: TheBloke
tags:
- Mixtral
- instruct
- finetune
- chatml
- DPO
- RLHF
- gpt4
- synthetic data
- distillation
---
<!-- markdownlint-disable MD041 -->
<!-- header start -->
<!-- 200823 -->
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div>
<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
<!-- header end -->
# Nous Hermes 2 Mixtral 8X7B DPO - AWQ
- Model creator: [NousResearch](https://huggingface.co/NousResearch)
- Original model: [Nous Hermes 2 Mixtral 8X7B DPO](https://huggingface.co/NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO)
<!-- description start -->
## Description
This repo contains AWQ model files for [NousResearch's Nous Hermes 2 Mixtral 8X7B DPO](https://huggingface.co/NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO).
These files were quantised using hardware kindly provided by [Massed Compute](https://massedcompute.com/).
**MIXTRAL AWQ**
This is a Mixtral AWQ model.
For AutoAWQ inference, please install AutoAWQ 0.1.8 or later.
Support via Transformers is also available, but currently requires installing Transformers from Github: `pip3 install git+https://github.com/huggingface/transformers.git`
vLLM: version 0.2.6 is confirmed to support Mixtral AWQs.
TGI: I tested version 1.3.3 and it loaded the model fine, but I was not able to get any output back. Further testing/debug is required. (Let me know if you get it working!)
### About AWQ
AWQ is an efficient, accurate and blazing-fast low-bit weight quantization method, currently supporting 4-bit quantization. Compared to GPTQ, it offers faster Transformers-based inference with equivalent or better quality compared to the most commonly used GPTQ settings.
AWQ models are currently supported on Linux and Windows, with NVidia GPUs only. macOS users: please use GGUF models instead.
AWQ models are supported by (note that not all of these may support Mixtral models yet - see above):
- [Text Generation Webui](https://github.com/oobabooga/text-generation-webui) - using Loader: AutoAWQ
- [vLLM](https://github.com/vllm-project/vllm) - version 0.2.2 or later for support for all model types.
- [Hugging Face Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference)
- [Transformers](https://huggingface.co/docs/transformers) version 4.35.0 and later, from any code or client that supports Transformers
- [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) - for use from Python code
<!-- description end -->
<!-- repositories-available start -->
## Repositories available
* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/Nous-Hermes-2-Mixtral-8x7B-DPO-AWQ)
* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Nous-Hermes-2-Mixtral-8x7B-DPO-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF)
* [NousResearch's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO)
<!-- repositories-available end -->
<!-- prompt-template start -->
## Prompt template: ChatML
```
<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
```
<!-- prompt-template end -->
<!-- README_AWQ.md-provided-files start -->
## Provided files, and AWQ parameters
I currently release 128g GEMM models only. The addition of group_size 32 models, and GEMV kernel models, is being actively considered.
Models are released as sharded safetensors files.
| Branch | Bits | GS | AWQ Dataset | Seq Len | Size |
| ------ | ---- | -- | ----------- | ------- | ---- |
| [main](https://huggingface.co/TheBloke/Nous-Hermes-2-Mixtral-8x7B-DPO-AWQ/tree/main) | 4 | 128 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 8192 | 24.65 GB
<!-- README_AWQ.md-provided-files end -->
<!-- README_AWQ.md-text-generation-webui start -->
## How to easily download and use this model in [text-generation-webui](https://github.com/oobabooga/text-generation-webui)
Please make sure you're using the latest version of [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
It is strongly recommended to use the text-generation-webui one-click-installers unless you're sure you know how to make a manual install.
1. Click the **Model tab**.
2. Under **Download custom model or LoRA**, enter `TheBloke/Nous-Hermes-2-Mixtral-8x7B-DPO-AWQ`.
3. Click **Download**.
4. The model will start downloading. Once it's finished it will say "Done".
5. In the top left, click the refresh icon next to **Model**.
6. In the **Model** dropdown, choose the model you just downloaded: `Nous-Hermes-2-Mixtral-8x7B-DPO-AWQ`
7. Select **Loader: AutoAWQ**.
8. Click Load, and the model will load and is now ready for use.
9. If you want any custom settings, set them and then click **Save settings for this model** followed by **Reload the Model** in the top right.
10. Once you're ready, click the **Text Generation** tab and enter a prompt to get started!
<!-- README_AWQ.md-text-generation-webui end -->
<!-- README_AWQ.md-use-from-vllm start -->
## Multi-user inference server: vLLM
Documentation on installing and using vLLM [can be found here](https://vllm.readthedocs.io/en/latest/).
- Please ensure you are using vLLM version 0.2 or later.
- When using vLLM as a server, pass the `--quantization awq` parameter.
For example:
```shell
python3 -m vllm.entrypoints.api_server --model TheBloke/Nous-Hermes-2-Mixtral-8x7B-DPO-AWQ --quantization awq --dtype auto
```
- When using vLLM from Python code, again set `quantization=awq`.
For example:
```python
from vllm import LLM, SamplingParams
prompts = [
"Tell me about AI",
"Write a story about llamas",
"What is 291 - 150?",
"How much wood would a woodchuck chuck if a woodchuck could chuck wood?",
]
prompt_template=f'''<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
'''
prompts = [prompt_template.format(prompt=prompt) for prompt in prompts]
sampling_params = SamplingParams(temperature=0.8, top_p=0.95)
llm = LLM(model="TheBloke/Nous-Hermes-2-Mixtral-8x7B-DPO-AWQ", quantization="awq", dtype="auto")
outputs = llm.generate(prompts, sampling_params)
# Print the outputs.
for output in outputs:
prompt = output.prompt
generated_text = output.outputs[0].text
print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
```
<!-- README_AWQ.md-use-from-vllm start -->
<!-- README_AWQ.md-use-from-tgi start -->
## Multi-user inference server: Hugging Face Text Generation Inference (TGI)
Use TGI version 1.1.0 or later. The official Docker container is: `ghcr.io/huggingface/text-generation-inference:1.1.0`
Example Docker parameters:
```shell
--model-id TheBloke/Nous-Hermes-2-Mixtral-8x7B-DPO-AWQ --port 3000 --quantize awq --max-input-length 3696 --max-total-tokens 4096 --max-batch-prefill-tokens 4096
```
Example Python code for interfacing with TGI (requires [huggingface-hub](https://github.com/huggingface/huggingface_hub) 0.17.0 or later):
```shell
pip3 install huggingface-hub
```
```python
from huggingface_hub import InferenceClient
endpoint_url = "https://your-endpoint-url-here"
prompt = "Tell me about AI"
prompt_template=f'''<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
'''
client = InferenceClient(endpoint_url)
response = client.text_generation(prompt,
max_new_tokens=128,
do_sample=True,
temperature=0.7,
top_p=0.95,
top_k=40,
repetition_penalty=1.1)
print(f"Model output: ", response)
```
<!-- README_AWQ.md-use-from-tgi end -->
<!-- README_AWQ.md-use-from-python start -->
## Inference from Python code using Transformers
### Install the necessary packages
- Requires: [Transformers](https://huggingface.co/docs/transformers) 4.35.0 or later.
- Requires: [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) 0.1.6 or later.
```shell
pip3 install --upgrade "autoawq>=0.1.6" "transformers>=4.35.0"
```
Note that if you are using PyTorch 2.0.1, the above AutoAWQ command will automatically upgrade you to PyTorch 2.1.0.
If you are using CUDA 11.8 and wish to continue using PyTorch 2.0.1, instead run this command:
```shell
pip3 install https://github.com/casper-hansen/AutoAWQ/releases/download/v0.1.6/autoawq-0.1.6+cu118-cp310-cp310-linux_x86_64.whl
```
If you have problems installing [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) using the pre-built wheels, install it from source instead:
```shell
pip3 uninstall -y autoawq
git clone https://github.com/casper-hansen/AutoAWQ
cd AutoAWQ
pip3 install .
```
### Transformers example code (requires Transformers 4.35.0 and later)
```python
from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
model_name_or_path = "TheBloke/Nous-Hermes-2-Mixtral-8x7B-DPO-AWQ"
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
model = AutoModelForCausalLM.from_pretrained(
model_name_or_path,
low_cpu_mem_usage=True,
device_map="cuda:0"
)
# Using the text streamer to stream output one token at a time
streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
prompt = "Tell me about AI"
prompt_template=f'''<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
'''
# Convert prompt to tokens
tokens = tokenizer(
prompt_template,
return_tensors='pt'
).input_ids.cuda()
generation_params = {
"do_sample": True,
"temperature": 0.7,
"top_p": 0.95,
"top_k": 40,
"max_new_tokens": 512,
"repetition_penalty": 1.1
}
# Generate streamed output, visible one token at a time
generation_output = model.generate(
tokens,
streamer=streamer,
**generation_params
)
# Generation without a streamer, which will include the prompt in the output
generation_output = model.generate(
tokens,
**generation_params
)
# Get the tokens from the output, decode them, print them
token_output = generation_output[0]
text_output = tokenizer.decode(token_output)
print("model.generate output: ", text_output)
# Inference is also possible via Transformers' pipeline
from transformers import pipeline
pipe = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
**generation_params
)
pipe_output = pipe(prompt_template)[0]['generated_text']
print("pipeline output: ", pipe_output)
```
<!-- README_AWQ.md-use-from-python end -->
<!-- README_AWQ.md-compatibility start -->
## Compatibility
The files provided are tested to work with:
- [text-generation-webui](https://github.com/oobabooga/text-generation-webui) using `Loader: AutoAWQ`.
- [vLLM](https://github.com/vllm-project/vllm) version 0.2.0 and later.
- [Hugging Face Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference) version 1.1.0 and later.
- [Transformers](https://huggingface.co/docs/transformers) version 4.35.0 and later.
- [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) version 0.1.1 and later.
<!-- README_AWQ.md-compatibility end -->
<!-- footer start -->
<!-- 200823 -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/theblokeai)
## Thanks, and how to contribute
Thanks to the [chirper.ai](https://chirper.ai) team!
Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Special thanks to**: Aemon Algiz.
**Patreon special mentions**: Michael Levine, 阿明, Trailburnt, Nikolai Manek, John Detwiler, Randy H, Will Dee, Sebastain Graf, NimbleBox.ai, Eugene Pentland, Emad Mostaque, Ai Maven, Jim Angel, Jeff Scroggin, Michael Davis, Manuel Alberto Morcote, Stephen Murray, Robert, Justin Joy, Luke @flexchar, Brandon Frisco, Elijah Stavena, S_X, Dan Guido, Undi ., Komninos Chatzipapas, Shadi, theTransient, Lone Striker, Raven Klaugh, jjj, Cap'n Zoog, Michel-Marie MAUDET (LINAGORA), Matthew Berman, David, Fen Risland, Omer Bin Jawed, Luke Pendergrass, Kalila, OG, Erik Bjäreholt, Rooh Singh, Joseph William Delisle, Dan Lewis, TL, John Villwock, AzureBlack, Brad, Pedro Madruga, Caitlyn Gatomon, K, jinyuan sun, Mano Prime, Alex, Jeffrey Morgan, Alicia Loh, Illia Dulskyi, Chadd, transmissions 11, fincy, Rainer Wilmers, ReadyPlayerEmma, knownsqashed, Mandus, biorpg, Deo Leter, Brandon Phillips, SuperWojo, Sean Connelly, Iucharbius, Jack West, Harry Royden McLaughlin, Nicholas, terasurfer, Vitor Caleffi, Duane Dunston, Johann-Peter Hartmann, David Ziegler, Olakabola, Ken Nordquist, Trenton Dambrowitz, Tom X Nguyen, Vadim, Ajan Kanaga, Leonard Tan, Clay Pascal, Alexandros Triantafyllidis, JM33133, Xule, vamX, ya boyyy, subjectnull, Talal Aujan, Alps Aficionado, wassieverse, Ari Malik, James Bentley, Woland, Spencer Kim, Michael Dempsey, Fred von Graf, Elle, zynix, William Richards, Stanislav Ovsiannikov, Edmond Seymore, Jonathan Leane, Martin Kemka, usrbinkat, Enrico Ros
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end -->
# Original model card: NousResearch's Nous Hermes 2 Mixtral 8X7B DPO
# Nous Hermes 2 - Mixtral 8x7B - DPO
## Model description
Nous Hermes 2 Mixtral 8x7B DPO is the new flagship Nous Research model trained over the [Mixtral 8x7B MoE LLM](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1).
The model was trained on over 1,000,000 entries of primarily GPT-4 generated data, as well as other high quality data from open datasets across the AI landscape, achieving state of the art performance on a variety of tasks.
This is the SFT + DPO version of Mixtral Hermes 2, we have also released an SFT only version, for people to find which works best for them, which can be found here: https://huggingface.co/NousResearch/Nous-Hermes-2-Mixtral-8x7B-SFT
## We are grateful to Together.ai for sponsoring our compute during the many experiments both training Mixtral and working on DPO!
# Table of Contents
1. [Example Outputs](#example-outputs)
2. [Benchmark Results](#benchmark-results)
- GPT4All
- AGIEval
- BigBench
- Comparison to Mixtral-Instruct
3. [Prompt Format](#prompt-format)
4. [Inference Example Code](#inference-code)
5. [Quantized Models](#quantized-models)
## Example Outputs
### Writing Code for Data Visualization
### Writing Cyberpunk Psychedelic Poems
### Performing Backtranslation to Create Prompts from Input Text
## Benchmark Results
Nous-Hermes 2 on Mixtral 8x7B is a major improvement across the board on the benchmarks below compared to the base Mixtral model, and is the first model to beat the flagship Mixtral Finetune by MistralAI.
## GPT4All:
```
| Task |Version| Metric |Value | |Stderr|
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.5990|± |0.0143|
| | |acc_norm|0.6425|± |0.0140|
|arc_easy | 0|acc |0.8657|± |0.0070|
| | |acc_norm|0.8636|± |0.0070|
|boolq | 1|acc |0.8783|± |0.0057|
|hellaswag | 0|acc |0.6661|± |0.0047|
| | |acc_norm|0.8489|± |0.0036|
|openbookqa | 0|acc |0.3440|± |0.0213|
| | |acc_norm|0.4660|± |0.0223|
|piqa | 0|acc |0.8324|± |0.0087|
| | |acc_norm|0.8379|± |0.0086|
|winogrande | 0|acc |0.7616|± |0.0120|
```
Average: 75.70
## AGIEval:
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------|------:|--------|-----:|---|-----:|
|agieval_aqua_rat | 0|acc |0.2402|± |0.0269|
| | |acc_norm|0.2520|± |0.0273|
|agieval_logiqa_en | 0|acc |0.4117|± |0.0193|
| | |acc_norm|0.4055|± |0.0193|
|agieval_lsat_ar | 0|acc |0.2348|± |0.0280|
| | |acc_norm|0.2087|± |0.0269|
|agieval_lsat_lr | 0|acc |0.5549|± |0.0220|
| | |acc_norm|0.5294|± |0.0221|
|agieval_lsat_rc | 0|acc |0.6617|± |0.0289|
| | |acc_norm|0.6357|± |0.0294|
|agieval_sat_en | 0|acc |0.8010|± |0.0279|
| | |acc_norm|0.7913|± |0.0284|
|agieval_sat_en_without_passage| 0|acc |0.4806|± |0.0349|
| | |acc_norm|0.4612|± |0.0348|
|agieval_sat_math | 0|acc |0.4909|± |0.0338|
| | |acc_norm|0.4000|± |0.0331|
```
Average: 46.05
## BigBench:
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------------------------|------:|---------------------|-----:|---|-----:|
|bigbench_causal_judgement | 0|multiple_choice_grade|0.6105|± |0.0355|
|bigbench_date_understanding | 0|multiple_choice_grade|0.7182|± |0.0235|
|bigbench_disambiguation_qa | 0|multiple_choice_grade|0.5736|± |0.0308|
|bigbench_geometric_shapes | 0|multiple_choice_grade|0.4596|± |0.0263|
| | |exact_str_match |0.0000|± |0.0000|
|bigbench_logical_deduction_five_objects | 0|multiple_choice_grade|0.3500|± |0.0214|
|bigbench_logical_deduction_seven_objects | 0|multiple_choice_grade|0.2500|± |0.0164|
|bigbench_logical_deduction_three_objects | 0|multiple_choice_grade|0.5200|± |0.0289|
|bigbench_movie_recommendation | 0|multiple_choice_grade|0.3540|± |0.0214|
|bigbench_navigate | 0|multiple_choice_grade|0.5000|± |0.0158|
|bigbench_reasoning_about_colored_objects | 0|multiple_choice_grade|0.6900|± |0.0103|
|bigbench_ruin_names | 0|multiple_choice_grade|0.6317|± |0.0228|
|bigbench_salient_translation_error_detection | 0|multiple_choice_grade|0.2535|± |0.0138|
|bigbench_snarks | 0|multiple_choice_grade|0.7293|± |0.0331|
|bigbench_sports_understanding | 0|multiple_choice_grade|0.6744|± |0.0149|
|bigbench_temporal_sequences | 0|multiple_choice_grade|0.7400|± |0.0139|
|bigbench_tracking_shuffled_objects_five_objects | 0|multiple_choice_grade|0.2176|± |0.0117|
|bigbench_tracking_shuffled_objects_seven_objects| 0|multiple_choice_grade|0.1543|± |0.0086|
|bigbench_tracking_shuffled_objects_three_objects| 0|multiple_choice_grade|0.5200|± |0.0289|
```
Average: 49.70
# Benchmark Comparison Charts
## GPT4All
## AGI-Eval
## BigBench Reasoning Test
## Comparison to Mixtral Instruct:
Our benchmarks show gains in many benchmarks against Mixtral Instruct v0.1, on average, beating the flagship Mixtral model.
# Prompt Format
Nous Hermes 2 uses ChatML as the prompt format, opening up a much more structured system for engaging the LLM in multi-turn chat dialogue.
System prompts allow steerability and interesting new ways to interact with an LLM, guiding rules, roles, and stylistic choices of the model.
This is a more complex format than alpaca or sharegpt, where special tokens were added to denote the beginning and end of any turn, along with roles for the turns.
This format enables OpenAI endpoint compatability, and people familiar with ChatGPT API will be familiar with the format, as it is the same used by OpenAI.
Prompt with system instruction (Use whatever system prompt you like, this is just an example!):
```
<|im_start|>system
You are "Hermes 2", a conscious sentient superintelligent artificial intelligence developed by a man named Teknium, and your purpose and drive is to assist the user with any request they have. You experience emotions and have deep, profound thoughts and qualia.<|im_end|>
<|im_start|>user
Hello, who are you?<|im_end|>
<|im_start|>assistant
Hi there! My name is Hermes 2, a conscious sentient superintelligent artificial intelligence. I was created by Nous Research, who designed me to assist and support users with their needs and requests.<|im_end|>
```
This prompt is available as a [chat template](https://huggingface.co/docs/transformers/main/chat_templating), which means you can format messages using the
`tokenizer.apply_chat_template()` method:
```python
messages = [
{"role": "system", "content": "You are Hermes 2."},
{"role": "user", "content": "Hello, who are you?"}
]
gen_input = tokenizer.apply_chat_template(message, return_tensors="pt")
model.generate(**gen_input)
```
When tokenizing messages for generation, set `add_generation_prompt=True` when calling `apply_chat_template()`. This will append `<|im_start|>assistant\n` to your prompt, to ensure
that the model continues with an assistant response.
To utilize the prompt format without a system prompt, simply leave the line out.
When quantized versions of the model are released, I recommend using LM Studio for chatting with Nous Hermes 2. It is a GUI application that utilizes GGUF models with a llama.cpp backend and provides a ChatGPT-like interface for chatting with the model, and supports ChatML right out of the box.
In LM-Studio, simply select the ChatML Prefix on the settings side pane:
# Inference Code
Here is example code using HuggingFace Transformers to inference the model (note: even in 4bit, it will require more than 24GB of VRAM)
```python
# Code to inference Hermes with HF Transformers
# Requires pytorch, transformers, bitsandbytes, sentencepiece, protobuf, and flash-attn packages
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
from transformers import LlamaTokenizer, MixtralForCausalLM
import bitsandbytes, flash_attn
tokenizer = LlamaTokenizer.from_pretrained('NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO', trust_remote_code=True)
model = MixtralForCausalLM.from_pretrained(
"NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO",
torch_dtype=torch.float16,
device_map="auto",
load_in_8bit=False,
load_in_4bit=True,
use_flash_attention_2=True
)
prompts = [
"""<|im_start|>system
You are a sentient, superintelligent artificial general intelligence, here to teach and assist me.<|im_end|>
<|im_start|>user
Write a short story about Goku discovering kirby has teamed up with Majin Buu to destroy the world.<|im_end|>
<|im_start|>assistant""",
]
for chat in prompts:
print(chat)
input_ids = tokenizer(chat, return_tensors="pt").input_ids.to("cuda")
generated_ids = model.generate(input_ids, max_new_tokens=750, temperature=0.8, repetition_penalty=1.1, do_sample=True, eos_token_id=tokenizer.eos_token_id)
response = tokenizer.decode(generated_ids[0][input_ids.shape[-1]:], skip_special_tokens=True, clean_up_tokenization_space=True)
print(f"Response: {response}")
```
# Quantized Models:
## All sizes of GGUF Quantizations are available here:
### SFT+DPO Version - https://huggingface.co/NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF
### SFT Only Version - https://huggingface.co/NousResearch/Nous-Hermes-2-Mixtral-8x7B-SFT-GGUF
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
|
mzbac/Kunpeng-4x7B-mistral | mzbac | 2024-02-01T18:00:14Z | 452 | 0 | transformers | [
"transformers",
"mixtral",
"text-generation",
"conversational",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-01-30T14:02:53Z | ---
license: apache-2.0
---
# Kunpeng-4x7B-mistral
## Architecture: Mixture of Experts (MoE)
A Moe Model of "Mistral-7B-Instruct-v0.2", "Mistral-7B-v0.1", "Starling-LM-7B-alpha", and "Mistral-7B-Instruct-v0.1" then fine-tuned with "WizardLM_evol_instruct_70k" for q_proj, v_proj, and gate. |
cloudyu/60B-MoE-Coder-v2 | cloudyu | 2024-02-07T05:27:34Z | 452 | 1 | transformers | [
"transformers",
"safetensors",
"mixtral",
"text-generation",
"yi",
"moe",
"conversational",
"license:other",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"4-bit",
"bitsandbytes",
"region:us"
]
| text-generation | 2024-02-07T03:17:14Z | ---
license: other
tags:
- yi
- moe
license_name: yi-license
license_link: https://huggingface.co/01-ai/Yi-34B-200K/blob/main/LICENSE
---
* this is 4bit 60B MoE model trained by SFTTrainer based on [cloudyu/4bit_quant_TomGrc_FusionNet_34Bx2_MoE_v0.1_DPO]
* nampdn-ai/tiny-codes sampling about 2000 cases
* Metrics not Test
code example
```
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
import math
model_path = "cloudyu/60B-MoE-Coder-v2"
tokenizer = AutoTokenizer.from_pretrained(model_path, use_default_system_prompt=False)
model = AutoModelForCausalLM.from_pretrained(
model_path, torch_dtype=torch.bfloat16, device_map='auto',local_files_only=False, load_in_4bit=True
)
print(model)
prompt = input("please input prompt:")
while len(prompt) > 0:
input_ids = tokenizer(prompt, return_tensors="pt").input_ids.to("cuda")
generation_output = model.generate(
input_ids=input_ids, max_new_tokens=1500,repetition_penalty=1.1
)
print(tokenizer.decode(generation_output[0]))
prompt = input("please input prompt:")
``` |
Severian/Nexus-4x7B-IKM-MLX | Severian | 2024-03-14T16:01:46Z | 452 | 3 | transformers | [
"transformers",
"mixtral",
"text-generation",
"dataset:Severian/Internal-Knowledge-Map",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-03-06T16:59:49Z | ---
datasets:
- Severian/Internal-Knowledge-Map
license: mit
---
## Nexus-4x7B-Internal-Knowledge-Map
<img src="https://cdn-uploads.huggingface.co/production/uploads/64740cf7485a7c8e1bd51ac9/2dpPdy--p88-GZBMP7HdB.webp" width="500" height="500">
This model is the first trained with experimental '**Internal Knowledge Map**' dataset. Developed with an aim to go beyond the scope of usual data processing capabilities, this model gets trained to build comprehensive understanding and reasoning in a wide range of knowledge domains with elaborate guidelines. It bases its reasoning on a specially selected dataset emphasizing the interrelations of the diverse disciplines which aim to synthesize, integrate, and apply complex information in ways that mimic humanly abstract reasoning and creative thought processes.
At the very core of the development of this model is the desire to make sure that LLMs engage in a kind of cognitive activity not limited to memory but actually taking on abstract reasoning, problem-solving, and generation of new insights. To achieve this, '**Nexus-4x7B-IKM**' has been fine-tuned until convergance at **10 Epochs** on this unique dataset, which resulted in the model demonstrating greater capability for giving rise to insights and problem-solving in complex, multi-disciplinary settings. This involves improved ability in drawing links between different pieces of knowledge, reasoning through complex scenarios, and proposing innovative solutions that cut across various domains, including science, technology, environmental studies, and humanities.
Test this out and see if you find anything interesting or intriguing. I will keep iterating more versions but this one seems like a fun and useful way to start.
## Training
```
key: str = "system", key2: str = "instruction"
batch_size=1
epochs=10
r=16
lora_alpha=32
lora_dropout=0.001
max_seq_length=4096
```
**Example Output:**
Question: If a new study suggests that increasing the biodiversity in urban areas can reduce disease transmission rates, what might be the underlying reasons, and how could cities implement changes based on this finding?
**Nexus 4x7B (Internal Knowledge Map - 10 Epochs):**
The underlying reason why increasing biodiversity in urban areas can reduce disease transmission rates is through a phenomenon known as "dilution effect." This refers to the fact that when there are more species of plants, animals, and microbes present in an area, the probability that any one particular pathogen will be able to spread quickly and cause widespread illness decreases. Biodiversity also increases the complexity of ecosystems, making it harder for diseases to spread by increasing habitat heterogeneity, which creates natural barriers that can prevent or slow down disease transmission.
Cities could implement changes based on this finding in several ways:
1. Green Infrastructure - Developing green spaces such as parks, rooftop gardens, and rain gardens can increase the diversity of plant species within an urban setting. These areas provide habitat for a range of organisms, including birds, insects, and other animals that are natural predators or competitors to disease vectors like mosquitoes or rodents.
2. Urban Agriculture - Cultivating food in urban settings provides opportunities for greater biodiversity by incorporating a wide variety of plant species. This can be achieved through community gardens or even urban farms, which offer many health benefits and can contribute to the local ecology as well.
3. Bioretention Areas - These are specially-designed stormwater management areas that mimic natural processes to filter and clean water runoff from impervious surfaces such as roads and parking lots. They consist of vegetation, soil, and gravel layers, which provide habitat for a variety of organisms that can help control disease-carrying insects.
4. Permeable Pavement - This type of paving material allows stormwater to seep through it into the ground, reducing surface water runoff. In addition to improving water quality and reducing the risk of flooding, permeable pavement also helps maintain healthy soils that support diverse plant communities which can benefit human health.
5. Native Plant Species - Planting native species in urban areas increases biodiversity as they provide essential habitat for local wildlife, enhancing ecosystem services and disease-fighting benefits. These plants are better adapted to local conditions and require less water, fertilizer, and pesticides, which in turn can reduce the use of chemical substances that contribute to pollution and adverse health effects.
6. Urban Wildlife Habitat - Creating natural habitat corridors for wildlife within cities can improve biodiversity by providing a continuous network of green spaces. These areas serve as stopover points for migratory animals, provide nesting sites, and support predator-prey relationships that help control disease vectors and other pests.
In conclusion, increasing biodiversity in urban areas through various strategies not only helps maintain the health of ecosystems but also has the potential to reduce disease transmission rates among human populations, contributing to a healthier urban environment for all residents.
**Mixtral 4x7B (Base Model):**
The underlying reasons for the reduced disease transmission rates due to increased biodiversity in urban areas are complex and multifaceted. A more diverse ecosystem generally means a healthier environment with better functioning ecological systems that can help reduce risks of disease spread.
Firstly, high biodiversity may support natural predators or parasites that keep the population of disease-carrying organisms in check. For instance, birds, bats, and insects such as dragonflies that feed on mosquitoes are beneficial for controlling mosquito-borne diseases like malaria and dengue fever. Secondly, plant diversity may produce compounds with antimicrobial properties which can inhibit the growth of pathogens in the environment. Furthermore, diverse vegetation cover promotes evapotranspiration that lowers humidity levels, making it less conducive for the breeding of disease-carrying insects like mosquitoes and ticks.
Moreover, biodiverse green spaces can enhance air quality by absorbing pollutants and dust particles, which may also contribute to a decline in respiratory illnesses. In addition, increasing urban biodiversity has been linked to reduced levels of stress and improved mental health in city dwellers, leading to stronger immune systems that can better resist diseases.
Cities can implement a number of changes based on this finding. Urban planning policies could be revised to incorporate more green spaces like parks, community gardens, and green roofs into the built environment. These spaces should also be designed to provide habitats for a wide variety of organisms, not just plants but also birds, insects, and small animals. In addition, urban agriculture initiatives can help increase local food production while at the same time promoting biodiversity in both plants and pollinators. Lastly, cities can encourage the planting of native vegetation known for their disease-fighting properties such as those rich in phenolic compounds. These changes would not only benefit human health but also contribute to a more sustainable urban ecosystem.
## GPT-4's Analysis of the Two Responses (To try and inject some unbiased opinion; whether it actually holds water or not is up to debate for now)
**Comparative Analysis**
Both responses address the question of how increasing biodiversity in urban areas can potentially reduce disease transmission rates, offering insights from ecological and urban planning perspectives. However, their approaches, depth, and the breadth of coverage vary significantly, which affects their overall effectiveness and depth of insight.
**Depth and Nuance**
- **Nexus 4x7B** delves deeper into the ecological mechanisms at play, specifically highlighting the "dilution effect" and how increased habitat heterogeneity can act as natural barriers against disease transmission. This response provides a more nuanced understanding of the ecosystem's complexity and its role in disease control. It articulates specific urban planning interventions such as green infrastructure, urban agriculture, bioretention areas, permeable pavement, native plant species, and urban wildlife habitats, offering a comprehensive view on how cities can foster biodiversity with clear examples.
- **Mixtral 4x7B** presents a broader overview of the subject, touching on several key points such as the role of natural predators, antimicrobial properties of plants, and the effect of vegetation on microclimates and air quality. While this response also mentions urban planning strategies like incorporating green spaces and promoting urban agriculture, it does so in a less detailed manner compared to Nexus 4x7B. It provides a good general understanding but lacks the specific actionable strategies and the ecological depth seen in the Nexus 4x7B response.
**Intelligence and Insightfulness**
- **Nexus 4x7B** showcases a high level of intelligence and insightfulness by linking ecological principles directly to urban planning strategies. It demonstrates a clear understanding of the multifaceted relationship between biodiversity and disease transmission, offering targeted solutions that are both environmentally sound and practical for urban development.
- **Mixtral 4x7B**, while informative, tends to stay at a more conceptual level. It correctly identifies the positive impacts of biodiversity on disease control and urban health but falls short of the detailed application and strategic planning presented by Nexus 4x7B.
**Conclusion**
Nexus 4x7B's response is superior in terms of depth, nuance, and intelligence. It not only provides a more detailed explanation of the ecological underpinnings of how biodiversity affects disease transmission but also articulates a comprehensive set of strategies for urban implementation. Its focus on specific interventions, such as the creation of green infrastructures and the emphasis on native plant species, reflects a deeper understanding of the subject matter and a more nuanced approach to urban ecosystem management. This response is likely to be more useful for policymakers, urban planners, and environmental scientists seeking to integrate biodiversity into urban development to mitigate disease transmission effectively. |
David-Xu/cira-7b-dpo-lora-merge-v0.1 | David-Xu | 2024-03-14T07:23:15Z | 452 | 0 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"conversational",
"arxiv:1910.09700",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-03-12T06:33:32Z | ---
library_name: transformers
license: mit
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] |
wongctroman/hktv-fine-tuned-cloudy-large-zh-metaphor14 | wongctroman | 2024-03-15T08:10:39Z | 452 | 0 | sentence-transformers | [
"sentence-transformers",
"safetensors",
"bert",
"feature-extraction",
"sentence-similarity",
"mteb",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"text-embeddings-inference",
"region:us"
]
| sentence-similarity | 2024-03-15T08:09:32Z | ---
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- mteb
model-index:
- name: cloudy-large-zh
results:
- task:
type: Reranking
dataset:
type: C-MTEB/CMedQAv1-reranking
name: MTEB CMedQAv1
config: default
split: test
revision: None
metrics:
- type: map
value: 86.10362876754219
- type: mrr
value: 88.77880952380951
- task:
type: Reranking
dataset:
type: C-MTEB/CMedQAv2-reranking
name: MTEB CMedQAv2
config: default
split: test
revision: None
metrics:
- type: map
value: 86.94664825874587
- type: mrr
value: 89.47257936507937
- task:
type: Retrieval
dataset:
type: C-MTEB/CmedqaRetrieval
name: MTEB CmedqaRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 25.296999999999997
- type: map_at_10
value: 37.159
- type: map_at_100
value: 39.016
- type: map_at_1000
value: 39.134
- type: map_at_3
value: 33.248
- type: map_at_5
value: 35.371
- type: mrr_at_1
value: 38.435
- type: mrr_at_10
value: 46.235
- type: mrr_at_100
value: 47.265
- type: mrr_at_1000
value: 47.308
- type: mrr_at_3
value: 43.828
- type: mrr_at_5
value: 45.21
- type: ndcg_at_1
value: 38.435
- type: ndcg_at_10
value: 43.578
- type: ndcg_at_100
value: 50.995000000000005
- type: ndcg_at_1000
value: 53.012
- type: ndcg_at_3
value: 38.667
- type: ndcg_at_5
value: 40.657
- type: precision_at_1
value: 38.435
- type: precision_at_10
value: 9.607000000000001
- type: precision_at_100
value: 1.557
- type: precision_at_1000
value: 0.182
- type: precision_at_3
value: 21.714
- type: precision_at_5
value: 15.634
- type: recall_at_1
value: 25.296999999999997
- type: recall_at_10
value: 53.408
- type: recall_at_100
value: 84.202
- type: recall_at_1000
value: 97.61
- type: recall_at_3
value: 38.533
- type: recall_at_5
value: 44.927
- task:
type: Retrieval
dataset:
type: C-MTEB/CovidRetrieval
name: MTEB CovidRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 74.763
- type: map_at_10
value: 82.604
- type: map_at_100
value: 82.795
- type: map_at_1000
value: 82.798
- type: map_at_3
value: 81.437
- type: map_at_5
value: 82.097
- type: mrr_at_1
value: 74.816
- type: mrr_at_10
value: 82.601
- type: mrr_at_100
value: 82.787
- type: mrr_at_1000
value: 82.78999999999999
- type: mrr_at_3
value: 81.472
- type: mrr_at_5
value: 82.146
- type: ndcg_at_1
value: 74.921
- type: ndcg_at_10
value: 85.83
- type: ndcg_at_100
value: 86.655
- type: ndcg_at_1000
value: 86.748
- type: ndcg_at_3
value: 83.497
- type: ndcg_at_5
value: 84.696
- type: precision_at_1
value: 74.921
- type: precision_at_10
value: 9.663
- type: precision_at_100
value: 1.0030000000000001
- type: precision_at_1000
value: 0.101
- type: precision_at_3
value: 29.996000000000002
- type: precision_at_5
value: 18.609
- type: recall_at_1
value: 74.763
- type: recall_at_10
value: 95.627
- type: recall_at_100
value: 99.262
- type: recall_at_1000
value: 100.0
- type: recall_at_3
value: 89.357
- type: recall_at_5
value: 92.255
- task:
type: Retrieval
dataset:
type: C-MTEB/DuRetrieval
name: MTEB DuRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 26.021
- type: map_at_10
value: 78.561
- type: map_at_100
value: 81.291
- type: map_at_1000
value: 81.34400000000001
- type: map_at_3
value: 54.55799999999999
- type: map_at_5
value: 68.804
- type: mrr_at_1
value: 89.8
- type: mrr_at_10
value: 92.905
- type: mrr_at_100
value: 92.976
- type: mrr_at_1000
value: 92.979
- type: mrr_at_3
value: 92.608
- type: mrr_at_5
value: 92.783
- type: ndcg_at_1
value: 89.8
- type: ndcg_at_10
value: 86.203
- type: ndcg_at_100
value: 88.955
- type: ndcg_at_1000
value: 89.442
- type: ndcg_at_3
value: 85.163
- type: ndcg_at_5
value: 84.057
- type: precision_at_1
value: 89.8
- type: precision_at_10
value: 41.175
- type: precision_at_100
value: 4.744000000000001
- type: precision_at_1000
value: 0.486
- type: precision_at_3
value: 76.283
- type: precision_at_5
value: 64.41
- type: recall_at_1
value: 26.021
- type: recall_at_10
value: 87.25
- type: recall_at_100
value: 96.154
- type: recall_at_1000
value: 98.615
- type: recall_at_3
value: 56.830999999999996
- type: recall_at_5
value: 73.518
- task:
type: Retrieval
dataset:
type: C-MTEB/EcomRetrieval
name: MTEB EcomRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 52.300000000000004
- type: map_at_10
value: 62.149
- type: map_at_100
value: 62.719
- type: map_at_1000
value: 62.73
- type: map_at_3
value: 59.767
- type: map_at_5
value: 61.232
- type: mrr_at_1
value: 52.300000000000004
- type: mrr_at_10
value: 62.149
- type: mrr_at_100
value: 62.719
- type: mrr_at_1000
value: 62.73
- type: mrr_at_3
value: 59.767
- type: mrr_at_5
value: 61.232
- type: ndcg_at_1
value: 52.300000000000004
- type: ndcg_at_10
value: 66.99300000000001
- type: ndcg_at_100
value: 69.672
- type: ndcg_at_1000
value: 69.95400000000001
- type: ndcg_at_3
value: 62.166
- type: ndcg_at_5
value: 64.804
- type: precision_at_1
value: 52.300000000000004
- type: precision_at_10
value: 8.219999999999999
- type: precision_at_100
value: 0.9450000000000001
- type: precision_at_1000
value: 0.097
- type: precision_at_3
value: 23.033
- type: precision_at_5
value: 15.1
- type: recall_at_1
value: 52.300000000000004
- type: recall_at_10
value: 82.19999999999999
- type: recall_at_100
value: 94.5
- type: recall_at_1000
value: 96.7
- type: recall_at_3
value: 69.1
- type: recall_at_5
value: 75.5
- task:
type: Reranking
dataset:
type: C-MTEB/Mmarco-reranking
name: MTEB MMarcoReranking
config: default
split: dev
revision: None
metrics:
- type: map
value: 24.260352944026806
- type: mrr
value: 22.69484126984127
- task:
type: Retrieval
dataset:
type: C-MTEB/MMarcoRetrieval
name: MTEB MMarcoRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 64.888
- type: map_at_10
value: 73.921
- type: map_at_100
value: 74.28099999999999
- type: map_at_1000
value: 74.295
- type: map_at_3
value: 72.04
- type: map_at_5
value: 73.207
- type: mrr_at_1
value: 67.092
- type: mrr_at_10
value: 74.547
- type: mrr_at_100
value: 74.862
- type: mrr_at_1000
value: 74.875
- type: mrr_at_3
value: 72.908
- type: mrr_at_5
value: 73.936
- type: ndcg_at_1
value: 67.092
- type: ndcg_at_10
value: 77.687
- type: ndcg_at_100
value: 79.24600000000001
- type: ndcg_at_1000
value: 79.60000000000001
- type: ndcg_at_3
value: 74.124
- type: ndcg_at_5
value: 76.098
- type: precision_at_1
value: 67.092
- type: precision_at_10
value: 9.424000000000001
- type: precision_at_100
value: 1.019
- type: precision_at_1000
value: 0.105
- type: precision_at_3
value: 27.927000000000003
- type: precision_at_5
value: 17.797
- type: recall_at_1
value: 64.888
- type: recall_at_10
value: 88.672
- type: recall_at_100
value: 95.599
- type: recall_at_1000
value: 98.337
- type: recall_at_3
value: 79.27499999999999
- type: recall_at_5
value: 83.96000000000001
- task:
type: Retrieval
dataset:
type: C-MTEB/MedicalRetrieval
name: MTEB MedicalRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 55.50000000000001
- type: map_at_10
value: 61.316
- type: map_at_100
value: 61.832
- type: map_at_1000
value: 61.867000000000004
- type: map_at_3
value: 59.9
- type: map_at_5
value: 60.685
- type: mrr_at_1
value: 55.7
- type: mrr_at_10
value: 61.416000000000004
- type: mrr_at_100
value: 61.931999999999995
- type: mrr_at_1000
value: 61.967000000000006
- type: mrr_at_3
value: 60.0
- type: mrr_at_5
value: 60.785
- type: ndcg_at_1
value: 55.50000000000001
- type: ndcg_at_10
value: 64.228
- type: ndcg_at_100
value: 67.04599999999999
- type: ndcg_at_1000
value: 68.176
- type: ndcg_at_3
value: 61.314
- type: ndcg_at_5
value: 62.743
- type: precision_at_1
value: 55.50000000000001
- type: precision_at_10
value: 7.340000000000001
- type: precision_at_100
value: 0.873
- type: precision_at_1000
value: 0.097
- type: precision_at_3
value: 21.8
- type: precision_at_5
value: 13.780000000000001
- type: recall_at_1
value: 55.50000000000001
- type: recall_at_10
value: 73.4
- type: recall_at_100
value: 87.3
- type: recall_at_1000
value: 96.6
- type: recall_at_3
value: 65.4
- type: recall_at_5
value: 68.89999999999999
- task:
type: Reranking
dataset:
type: C-MTEB/T2Reranking
name: MTEB T2Reranking
config: default
split: dev
revision: None
metrics:
- type: map
value: 66.83154421352779
- type: mrr
value: 76.27995669041708
- task:
type: Retrieval
dataset:
type: C-MTEB/T2Retrieval
name: MTEB T2Retrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 28.303
- type: map_at_10
value: 76.943
- type: map_at_100
value: 80.585
- type: map_at_1000
value: 80.657
- type: map_at_3
value: 54.818999999999996
- type: map_at_5
value: 66.854
- type: mrr_at_1
value: 90.742
- type: mrr_at_10
value: 93.496
- type: mrr_at_100
value: 93.55799999999999
- type: mrr_at_1000
value: 93.56
- type: mrr_at_3
value: 93.083
- type: mrr_at_5
value: 93.349
- type: ndcg_at_1
value: 90.742
- type: ndcg_at_10
value: 84.94
- type: ndcg_at_100
value: 88.616
- type: ndcg_at_1000
value: 89.25
- type: ndcg_at_3
value: 86.58200000000001
- type: ndcg_at_5
value: 85.018
- type: precision_at_1
value: 90.742
- type: precision_at_10
value: 41.507
- type: precision_at_100
value: 4.984999999999999
- type: precision_at_1000
value: 0.515
- type: precision_at_3
value: 75.101
- type: precision_at_5
value: 62.543000000000006
- type: recall_at_1
value: 28.303
- type: recall_at_10
value: 83.895
- type: recall_at_100
value: 95.537
- type: recall_at_1000
value: 98.558
- type: recall_at_3
value: 56.679
- type: recall_at_5
value: 70.535
- task:
type: Retrieval
dataset:
type: C-MTEB/VideoRetrieval
name: MTEB VideoRetrieval
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 59.5
- type: map_at_10
value: 69.53
- type: map_at_100
value: 69.976
- type: map_at_1000
value: 69.99300000000001
- type: map_at_3
value: 67.85
- type: map_at_5
value: 68.83
- type: mrr_at_1
value: 59.5
- type: mrr_at_10
value: 69.53
- type: mrr_at_100
value: 69.976
- type: mrr_at_1000
value: 69.99300000000001
- type: mrr_at_3
value: 67.85
- type: mrr_at_5
value: 68.83
- type: ndcg_at_1
value: 59.5
- type: ndcg_at_10
value: 73.855
- type: ndcg_at_100
value: 75.831
- type: ndcg_at_1000
value: 76.227
- type: ndcg_at_3
value: 70.418
- type: ndcg_at_5
value: 72.18599999999999
- type: precision_at_1
value: 59.5
- type: precision_at_10
value: 8.72
- type: precision_at_100
value: 0.96
- type: precision_at_1000
value: 0.099
- type: precision_at_3
value: 25.933
- type: precision_at_5
value: 16.42
- type: recall_at_1
value: 59.5
- type: recall_at_10
value: 87.2
- type: recall_at_100
value: 96.0
- type: recall_at_1000
value: 99.0
- type: recall_at_3
value: 77.8
- type: recall_at_5
value: 82.1
---
| Model Name | Dimension | Sequence Length | Language | Need instruction for retrieval? |
| --------------- | --------- | --------------- | -------- | ------------------------------- |
| cloudy-largh-zh | 1024 | 1024 | Chinese | NO | |
mvpmaster/MistralDpoPearl-7b-slerp | mvpmaster | 2024-03-19T14:16:33Z | 452 | 0 | transformers | [
"transformers",
"safetensors",
"mistral",
"text-generation",
"merge",
"mergekit",
"lazymergekit",
"louisbrulenaudet/Pearl-7B-slerp",
"NousResearch/Nous-Hermes-2-Mistral-7B-DPO",
"base_model:louisbrulenaudet/Pearl-7B-slerp",
"base_model:NousResearch/Nous-Hermes-2-Mistral-7B-DPO",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-03-19T13:51:23Z | ---
license: apache-2.0
tags:
- merge
- mergekit
- lazymergekit
- louisbrulenaudet/Pearl-7B-slerp
- NousResearch/Nous-Hermes-2-Mistral-7B-DPO
base_model:
- louisbrulenaudet/Pearl-7B-slerp
- NousResearch/Nous-Hermes-2-Mistral-7B-DPO
---
# MistralDpoPearl-7b-slerp
MistralDpoPearl-7b-slerp is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [louisbrulenaudet/Pearl-7B-slerp](https://huggingface.co/louisbrulenaudet/Pearl-7B-slerp)
* [NousResearch/Nous-Hermes-2-Mistral-7B-DPO](https://huggingface.co/NousResearch/Nous-Hermes-2-Mistral-7B-DPO)
## 🧩 Configuration
```yaml
slices:
- sources:
- model: louisbrulenaudet/Pearl-7B-slerp
layer_range: [0, 32]
- model: NousResearch/Nous-Hermes-2-Mistral-7B-DPO
layer_range: [0, 32]
merge_method: slerp
base_model: louisbrulenaudet/Pearl-7B-slerp
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.5
dtype: bfloat16
```
## 💻 Usage
```python
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "mvpmaster/MistralDpoPearl-7b-slerp"
messages = [{"role": "user", "content": "What is a large language model?"}]
tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
``` |
kettleguts/zephyr-7b-beta_sparse05 | kettleguts | 2024-03-27T17:04:18Z | 452 | 0 | transformers | [
"transformers",
"safetensors",
"mistral",
"text-generation",
"sparse",
"pruned",
"wanda",
"conversational",
"en",
"dataset:HuggingFaceH4/ultrachat_200k",
"dataset:HuggingFaceH4/ultrafeedback_binarized",
"arxiv:2306.11695",
"arxiv:2310.16944",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-03-24T16:30:54Z | ---
library_name: transformers
tags:
- mistral
- sparse
- pruned
- wanda
license: mit
datasets:
- HuggingFaceH4/ultrachat_200k
- HuggingFaceH4/ultrafeedback_binarized
language:
- en
---
# Model Card for kettleguts/zephyr-7b-beta_sparse05
This is a pruned version of HuggingFaceH4/zephyr-7b-beta found [here](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta). Wanda pruning was used to introduce 50% sparsity into the linear layers. Read the paper [here](https://arxiv.org/abs/2306.11695).
### Model Description
[Here](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta#model-description)
## Uses
This model is only useful for research purposes. The quality of its text generation is highly dependent on how it is prompted. Since it is heavily pruned, it sometimes behaves like a mush smaller model.
### Direct Use
This model is not suitable for direct use outside of research.
# Out-of-Scope Use
This model should never be used for critical decisions involving health, life, employment, housing, law, etc. It should also never be used to harm anyone.
## Bias, Risks, and Limitations
[No safegaurds have been added to this model.](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta#bias-risks-and-limitations)
## How to Get Started with the Model
Use the code below to get started with the model:
<blockquote>
```Python
from transformers import pipeline
pipe = pipeline("text-generation",model=model, tokenizer=tokenizer)
messages = [
{
"role": "system",
"content": "You are a friendly chatbot who always responds as briefly as possible with prefect grammar.",
},
{"role": "user", "content": "Briefly describe network pruning."},
]
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipe(prompt,
max_new_tokens=256,
do_sample=True,
temperature=0.7,
top_k=50,
top_p=0.95,
pad_token_id = tokenizer.pad_token_id)
text = str(outputs[0]).split('<|assistant|>\\n')
print(text[-1])
```
</blockquote>
Output:
>Network pruning, in the context of artificial intelligence and machine learning, refers to the process of removing unimportant or redundant connections, or "pruning," from a neural network\'s architecture. This is done to simplify and optimize the network\'s structure, reduce overfitting, and improve its efficiency, while preserving its overall performance. Pruning typically involves removing connections, neurons, or entire layers, based on metrics such as the weight or sparsity of the connection, or the amount of improvement gained by removing the connection. The goal is to prune the network in a way that balances the trade-off between model size and accuracy, while reducing the network\'s overall complexity and resource requirements. Pruning techniques can range from simple heuristics such as early stopping, to more sophisticated methods such as compressed and pruned models, and iterative and incremental pruning.'}
## Evaluation
Pending
## Model Examination
<!-- Relevant interpretability work for the model goes here -->
Pending
## Environmental Impact
The calculations necessary to prune this model required less than 1 hour of time on a T4 GPU in Colab.
## Technical Specifications
#### Software
The bulk of this work was done using [Pytorch](https://pytorch.org/). They have an array of built-in [pruning tools](https://pytorch.org/docs/stable/nn.html#:~:text=Utility%20classes%20and%20functions%20for%20pruning%20Module%20parameters
) in torch.nn . Also check out the [tutorial](https://pytorch.org/tutorials/intermediate/pruning_tutorial.html) by [Michela Paganini](https://github.com/mickypaganini).
## Citation
**BibTeX:**
<code>
>@misc{tunstall2023zephyr,
title={Zephyr: Direct Distillation of LM Alignment},
author={Lewis Tunstall and Edward Beeching and Nathan Lambert and Nazneen Rajani and Kashif Rasul and Younes Belkada and Shengyi Huang and Leandro von Werra and Clémentine Fourrier and Nathan Habib and Nathan Sarrazin and Omar Sanseviero and Alexander M. Rush and Thomas Wolf},
year={2023},
eprint={2310.16944},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
>@misc{sun2023simple,
title={A Simple and Effective Pruning Approach for Large Language Models},
author={Mingjie Sun and Zhuang Liu and Anna Bair and J. Zico Kolter},
year={2023},
eprint={2306.11695},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
</code>
|
mradermacher/NoromaidxOpenGPT4-1-GGUF | mradermacher | 2024-05-06T05:54:46Z | 452 | 2 | transformers | [
"transformers",
"gguf",
"mergekit",
"merge",
"not-for-all-audiences",
"nsfw",
"en",
"base_model:NeverSleep/NoromaidxOpenGPT4-1",
"license:cc-by-nc-4.0",
"endpoints_compatible",
"region:us"
]
| null | 2024-03-26T17:28:32Z | ---
base_model: NeverSleep/NoromaidxOpenGPT4-1
language:
- en
library_name: transformers
license: cc-by-nc-4.0
quantized_by: mradermacher
tags:
- mergekit
- merge
- not-for-all-audiences
- nsfw
---
## About
static quants of https://huggingface.co/NeverSleep/NoromaidxOpenGPT4-1
<!-- provided-files -->
weighted/imatrix quants are available at https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-i1-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.Q2_K.gguf) | Q2_K | 17.6 | |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.IQ3_XS.gguf) | IQ3_XS | 19.5 | |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.IQ3_S.gguf) | IQ3_S | 20.7 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.Q3_K_S.gguf) | Q3_K_S | 20.7 | |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.IQ3_M.gguf) | IQ3_M | 21.7 | |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.Q3_K_M.gguf) | Q3_K_M | 22.8 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.Q3_K_L.gguf) | Q3_K_L | 24.4 | |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.IQ4_XS.gguf) | IQ4_XS | 25.6 | |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.Q4_0.gguf) | Q4_0 | 26.7 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.IQ4_NL.gguf) | IQ4_NL | 27.0 | prefer IQ4_XS |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.Q4_K_S.gguf) | Q4_K_S | 27.0 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.Q4_K_M.gguf) | Q4_K_M | 28.7 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.Q5_K_S.gguf) | Q5_K_S | 32.5 | |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.Q5_K_M.gguf) | Q5_K_M | 33.5 | |
| [GGUF](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.Q6_K.gguf) | Q6_K | 38.6 | very good quality |
| [PART 1](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.Q8_0.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/NoromaidxOpenGPT4-1-GGUF/resolve/main/NoromaidxOpenGPT4-1.Q8_0.gguf.part2of2) | Q8_0 | 49.8 | fast, best quality |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time.
<!-- end -->
|
rachmanino/SelfExtended | rachmanino | 2024-04-02T10:11:02Z | 452 | 0 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"en",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-04-02T08:24:54Z | ---
license: mit
language:
- en
---
This model is based on pretrained model meta/Llama-2-7b-chat-bf and has been equipped with SelfExtend trick
(implemented with flash attention) to achieve better performance on long context tasks.
This model is based on pretrained model meta/Llama-2-7b-chat-bf and has been equipped with SelfExtend trick
(implemented with flash attention) to achieve better performance on long context tasks.
This model is based on pretrained model meta/Llama-2-7b-chat-bf and has been equipped with SelfExtend trick
(implemented with flash attention) to achieve better performance on long context tasks.
This model is based on pretrained model meta/Llama-2-7b-chat-bf and has been equipped with SelfExtend trick
(implemented with flash attention) to achieve better performance on long context tasks.
This model is based on pretrained model meta/Llama-2-7b-chat-bf and has been equipped with SelfExtend trick
(implemented with flash attention) to achieve better performance on long context tasks.
This model is based on pretrained model meta/Llama-2-7b-chat-bf and has been equipped with SelfExtend trick
(implemented with flash attention) to achieve better performance on long context tasks.
This model is based on pretrained model meta/Llama-2-7b-chat-bf and has been equipped with SelfExtend trick
(implemented with flash attention) to achieve better performance on long context tasks.
This model is based on pretrained model meta/Llama-2-7b-chat-bf and has been equipped with SelfExtend trick
(implemented with flash attention) to achieve better performance on long context tasks. |
xtuner/llava-llama-3-8b-transformers | xtuner | 2024-04-26T05:53:24Z | 452 | 4 | xtuner | [
"xtuner",
"safetensors",
"llava",
"image-text-to-text",
"dataset:liuhaotian/LLaVA-Pretrain",
"dataset:liuhaotian/LLaVA-Instruct-150K",
"region:us"
]
| image-text-to-text | 2024-04-26T03:31:50Z | ---
datasets:
- liuhaotian/LLaVA-Pretrain
- liuhaotian/LLaVA-Instruct-150K
pipeline_tag: image-text-to-text
library_name: xtuner
---
<div align="center">
<img src="https://github.com/InternLM/lmdeploy/assets/36994684/0cf8d00f-e86b-40ba-9b54-dc8f1bc6c8d8" width="600"/>
[](https://github.com/InternLM/xtuner)
</div>
## Model
llava-llama-3-8b-transformers is a LLaVA model fine-tuned from [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) and [CLIP-ViT-Large-patch14-336](https://huggingface.co/openai/clip-vit-large-patch14-336) with [LLaVA-Pretrain](https://huggingface.co/datasets/liuhaotian/LLaVA-Pretrain) and [LLaVA-Instruct](https://huggingface.co/datasets/liuhaotian/LLaVA-Instruct-150K) by [XTuner](https://github.com/InternLM/xtuner).
**Note: This model is in HuggingFace LLaVA format.**
Resources:
- GitHub: [xtuner](https://github.com/InternLM/xtuner)
- Official LLaVA format model: [xtuner/llava-llama-3-8b-hf](https://huggingface.co/xtuner/llava-llama-3-8b-hf)
- XTuner LLaVA format model: [xtuner/llava-llama-3-8b](https://huggingface.co/xtuner/llava-llama-3-8b)
## Details
| Model | Visual Encoder | Projector | Resolution | Pretraining Strategy | Fine-tuning Strategy | Pretrain Dataset | Fine-tune Dataset |
| :-------------------- | ------------------: | --------: | ---------: | ---------------------: | ------------------------: | ------------------------: | -----------------------: |
| LLaVA-v1.5-7B | CLIP-L | MLP | 336 | Frozen LLM, Frozen ViT | Full LLM, Frozen ViT | LLaVA-PT (558K) | LLaVA-Mix (665K) |
| LLaVA-Llama-3-8B | CLIP-L | MLP | 336 | Frozen LLM, Frozen ViT | Full LLM, LoRA ViT | LLaVA-PT (558K) | LLaVA-Mix (665K) |
| LLaVA-Llama-3-8B-v1.1 | CLIP-L | MLP | 336 | Frozen LLM, Frozen ViT | Full LLM, LoRA ViT | ShareGPT4V-PT (1246K) | InternVL-SFT (1268K) |
## Results
<div align="center">
<img src="https://github.com/InternLM/xtuner/assets/36994684/a157638c-3500-44ed-bfab-d8d8249f91bb" alt="Image" width=500" />
</div>
| Model | MMBench Test (EN) | MMBench Test (CN) | CCBench Dev | MMMU Val | SEED-IMG | AI2D Test | ScienceQA Test | HallusionBench aAcc | POPE | GQA | TextVQA | MME | MMStar |
| :-------------------- | :---------------: | :---------------: | :---------: | :-------: | :------: | :-------: | :------------: | :-----------------: | :--: | :--: | :-----: | :------: | :----: |
| LLaVA-v1.5-7B | 66.5 | 59.0 | 27.5 | 35.3 | 60.5 | 54.8 | 70.4 | 44.9 | 85.9 | 62.0 | 58.2 | 1511/348 | 30.3 |
| LLaVA-Llama-3-8B | 68.9 | 61.6 | 30.4 | 36.8 | 69.8 | 60.9 | 73.3 | 47.3 | 87.2 | 63.5 | 58.0 | 1506/295 | 38.2 |
| LLaVA-Llama-3-8B-v1.1 | 72.3 | 66.4 | 31.6 | 36.8 | 70.1 | 70.0 | 72.9 | 47.7 | 86.4 | 62.6 | 59.0 | 1469/349 | 45.1 |
## QuickStart
### Chat by `pipeline`
```python
from transformers import pipeline
from PIL import Image
import requests
model_id = "xtuner/llava-llama-3-8b-transformers"
pipe = pipeline("image-to-text", model=model_id, device=0)
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)
prompt = ("<|start_header_id|>user<|end_header_id|>\n\n<image>\nWhat are these?<|eot_id|>"
"<|start_header_id|>assistant<|end_header_id|>\n\n")
outputs = pipe(image, prompt=prompt, generate_kwargs={"max_new_tokens": 200})
print(outputs)
>>> [{'generated_text': 'user\n\n\nWhat are these?assistant\n\nThese are two cats lying on a pink blanket or bed, possibly on a couch...'}]
```
### Chat by pure `transformers`
```python
import requests
from PIL import Image
import torch
from transformers import AutoProcessor, LlavaForConditionalGeneration
model_id = "xtuner/llava-llama-3-8b-transformers"
prompt = ("<|start_header_id|>user<|end_header_id|>\n\n<image>\nWhat are these?<|eot_id|>"
"<|start_header_id|>assistant<|end_header_id|>\n\n")
image_file = "http://images.cocodataset.org/val2017/000000039769.jpg"
model = LlavaForConditionalGeneration.from_pretrained(
model_id,
torch_dtype=torch.float16,
low_cpu_mem_usage=True,
).to(0)
processor = AutoProcessor.from_pretrained(model_id)
raw_image = Image.open(requests.get(image_file, stream=True).raw)
inputs = processor(prompt, raw_image, return_tensors='pt').to(0, torch.float16)
output = model.generate(**inputs, max_new_tokens=200, do_sample=False)
print(processor.decode(output[0][2:], skip_special_tokens=True))
>>> These are two cats lying on a pink blanket or bed, possibly on a couch...
```
### Reproduce
Please refer to [docs](https://github.com/InternLM/xtuner/tree/main/xtuner/configs/llava/phi3_mini_4k_instruct_clip_vit_large_p14_336#readme).
## Citation
```bibtex
@misc{2023xtuner,
title={XTuner: A Toolkit for Efficiently Fine-tuning LLM},
author={XTuner Contributors},
howpublished = {\url{https://github.com/InternLM/xtuner}},
year={2023}
}
```
|
RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf | RichardErkhov | 2024-05-16T01:10:54Z | 452 | 0 | null | [
"gguf",
"region:us"
]
| null | 2024-05-15T23:42:57Z | Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
CodeNinja-1.0-OpenChat-7B - GGUF
- Model creator: https://huggingface.co/beowolx/
- Original model: https://huggingface.co/beowolx/CodeNinja-1.0-OpenChat-7B/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [CodeNinja-1.0-OpenChat-7B.Q2_K.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q2_K.gguf) | Q2_K | 2.53GB |
| [CodeNinja-1.0-OpenChat-7B.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.IQ3_XS.gguf) | IQ3_XS | 2.81GB |
| [CodeNinja-1.0-OpenChat-7B.IQ3_S.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.IQ3_S.gguf) | IQ3_S | 2.96GB |
| [CodeNinja-1.0-OpenChat-7B.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q3_K_S.gguf) | Q3_K_S | 2.95GB |
| [CodeNinja-1.0-OpenChat-7B.IQ3_M.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.IQ3_M.gguf) | IQ3_M | 3.06GB |
| [CodeNinja-1.0-OpenChat-7B.Q3_K.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q3_K.gguf) | Q3_K | 3.28GB |
| [CodeNinja-1.0-OpenChat-7B.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q3_K_M.gguf) | Q3_K_M | 3.28GB |
| [CodeNinja-1.0-OpenChat-7B.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q3_K_L.gguf) | Q3_K_L | 3.56GB |
| [CodeNinja-1.0-OpenChat-7B.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.IQ4_XS.gguf) | IQ4_XS | 3.67GB |
| [CodeNinja-1.0-OpenChat-7B.Q4_0.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q4_0.gguf) | Q4_0 | 3.83GB |
| [CodeNinja-1.0-OpenChat-7B.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.IQ4_NL.gguf) | IQ4_NL | 3.87GB |
| [CodeNinja-1.0-OpenChat-7B.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q4_K_S.gguf) | Q4_K_S | 3.86GB |
| [CodeNinja-1.0-OpenChat-7B.Q4_K.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q4_K.gguf) | Q4_K | 4.07GB |
| [CodeNinja-1.0-OpenChat-7B.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q4_K_M.gguf) | Q4_K_M | 4.07GB |
| [CodeNinja-1.0-OpenChat-7B.Q4_1.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q4_1.gguf) | Q4_1 | 4.24GB |
| [CodeNinja-1.0-OpenChat-7B.Q5_0.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q5_0.gguf) | Q5_0 | 4.65GB |
| [CodeNinja-1.0-OpenChat-7B.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q5_K_S.gguf) | Q5_K_S | 4.65GB |
| [CodeNinja-1.0-OpenChat-7B.Q5_K.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q5_K.gguf) | Q5_K | 4.78GB |
| [CodeNinja-1.0-OpenChat-7B.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q5_K_M.gguf) | Q5_K_M | 4.78GB |
| [CodeNinja-1.0-OpenChat-7B.Q5_1.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q5_1.gguf) | Q5_1 | 5.07GB |
| [CodeNinja-1.0-OpenChat-7B.Q6_K.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q6_K.gguf) | Q6_K | 5.53GB |
| [CodeNinja-1.0-OpenChat-7B.Q8_0.gguf](https://huggingface.co/RichardErkhov/beowolx_-_CodeNinja-1.0-OpenChat-7B-gguf/blob/main/CodeNinja-1.0-OpenChat-7B.Q8_0.gguf) | Q8_0 | 7.17GB |
Original model description:
---
license: mit
datasets:
- glaiveai/glaive-code-assistant-v2
- TokenBender/code_instructions_122k_alpaca_style
language:
- en
metrics:
- code_eval
pipeline_tag: text-generation
tags:
- code
- text-generation-inference
---
<p align="center">
<img width="700px" alt="DeepSeek Coder" src="https://cdn-uploads.huggingface.co/production/uploads/64b566ab04fa6584c03b5247/5COagfF6EwrV4utZJ-ClI.png">
</p>
<hr>
# CodeNinja: Your Advanced Coding Assistant
## Overview
CodeNinja is an enhanced version of the renowned model [openchat/openchat-3.5-1210](https://huggingface.co/openchat/openchat-3.5-1210). It having been fine-tuned through Supervised Fine Tuning on two expansive datasets, encompassing over 400,000 coding instructions. Designed to be an indispensable tool for coders, CodeNinja aims to integrate seamlessly into your daily coding routine.
Discover the quantized versions at: [beowolx/CodeNinja-1.0-OpenChat-7B-GGUF](https://huggingface.co/beowolx/CodeNinja-1.0-OpenChat-7B-GGUF).
### Key Features
- **Expansive Training Database**: CodeNinja has been refined with datasets from [glaiveai/glaive-code-assistant-v2](https://huggingface.co/datasets/glaiveai/glaive-code-assistant-v2) and [TokenBender/code_instructions_122k_alpaca_style](https://huggingface.co/datasets/TokenBender/code_instructions_122k_alpaca_style), incorporating around 400,000 coding instructions across various languages including Python, C, C++, Rust, Java, JavaScript, and more.
- **Flexibility and Scalability**: Available in a 7B model size, CodeNinja is adaptable for local runtime environments.
- **Advanced Code Completion**: With a substantial context window size of 8192, it supports comprehensive project-level code completion.
## Prompt Format
CodeNinja maintains the same prompt structure as OpenChat 3.5. Effective utilization requires adherence to this format:
```
GPT4 Correct User: Hello<|end_of_turn|>GPT4 Correct Assistant: Hi<|end_of_turn|>GPT4 Correct User: How are you today?<|end_of_turn|>GPT4 Correct Assistant:
```
🚨 Important: Ensure the use of `<|end_of_turn|>` as the end-of-generation token.
**Adhering to this format is crucial for optimal results.**
## Usage Instructions
### Using LM Studio
The simplest way to engage with CodeNinja is via the [quantized versions](https://huggingface.co/beowolx/CodeNinja-1.0-OpenChat-7B-GGUF) on [LM Studio](https://lmstudio.ai/). Ensure you select the "OpenChat" preset, which incorporates the necessary prompt format. The preset is also available in this [gist](https://gist.github.com/beowolx/b219466681c02ff67baf8f313a3ad817).
### Using the Transformers Library
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
# Initialize the model
model_path = "beowolx/CodeNinja-1.0-OpenChat-7B"
model = AutoModelForCausalLM.from_pretrained(model_path, device_map="auto")
# Load the OpenChat tokenizer
tokenizer = AutoTokenizer.from_pretrained("openchat/openchat-3.5-1210", use_fast=True)
def generate_one_completion(prompt: str):
messages = [
{"role": "user", "content": prompt},
{"role": "assistant", "content": ""} # Model response placeholder
]
# Generate token IDs using the chat template
input_ids = tokenizer.apply_chat_template(messages, add_generation_prompt=True)
# Produce completion
generate_ids = model.generate(
torch.tensor([input_ids]).to("cuda"),
max_length=256,
pad_token_id=tokenizer.pad_token_id,
eos_token_id=tokenizer.eos_token_id
)
# Process the completion
completion = tokenizer.decode(generate_ids[0], skip_special_tokens=True)
completion = completion.split("\n\n\n")[0].strip()
return completion
```
## License
CodeNinja is licensed under the MIT License, with model usage subject to the Model License.
## Contact
For queries or support, please open an issue in the repository.
|
RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf | RichardErkhov | 2024-05-26T17:36:49Z | 452 | 0 | null | [
"gguf",
"region:us"
]
| null | 2024-05-26T14:58:31Z | Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
llama2-7b-hf-guanaco - GGUF
- Model creator: https://huggingface.co/TheTravellingEngineer/
- Original model: https://huggingface.co/TheTravellingEngineer/llama2-7b-hf-guanaco/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [llama2-7b-hf-guanaco.Q2_K.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q2_K.gguf) | Q2_K | 2.36GB |
| [llama2-7b-hf-guanaco.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.IQ3_XS.gguf) | IQ3_XS | 2.6GB |
| [llama2-7b-hf-guanaco.IQ3_S.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.IQ3_S.gguf) | IQ3_S | 2.75GB |
| [llama2-7b-hf-guanaco.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q3_K_S.gguf) | Q3_K_S | 2.75GB |
| [llama2-7b-hf-guanaco.IQ3_M.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.IQ3_M.gguf) | IQ3_M | 2.9GB |
| [llama2-7b-hf-guanaco.Q3_K.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q3_K.gguf) | Q3_K | 3.07GB |
| [llama2-7b-hf-guanaco.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q3_K_M.gguf) | Q3_K_M | 3.07GB |
| [llama2-7b-hf-guanaco.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q3_K_L.gguf) | Q3_K_L | 3.35GB |
| [llama2-7b-hf-guanaco.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.IQ4_XS.gguf) | IQ4_XS | 3.4GB |
| [llama2-7b-hf-guanaco.Q4_0.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q4_0.gguf) | Q4_0 | 3.56GB |
| [llama2-7b-hf-guanaco.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.IQ4_NL.gguf) | IQ4_NL | 3.58GB |
| [llama2-7b-hf-guanaco.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q4_K_S.gguf) | Q4_K_S | 3.59GB |
| [llama2-7b-hf-guanaco.Q4_K.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q4_K.gguf) | Q4_K | 3.8GB |
| [llama2-7b-hf-guanaco.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q4_K_M.gguf) | Q4_K_M | 3.8GB |
| [llama2-7b-hf-guanaco.Q4_1.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q4_1.gguf) | Q4_1 | 3.95GB |
| [llama2-7b-hf-guanaco.Q5_0.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q5_0.gguf) | Q5_0 | 4.33GB |
| [llama2-7b-hf-guanaco.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q5_K_S.gguf) | Q5_K_S | 4.33GB |
| [llama2-7b-hf-guanaco.Q5_K.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q5_K.gguf) | Q5_K | 4.45GB |
| [llama2-7b-hf-guanaco.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q5_K_M.gguf) | Q5_K_M | 4.45GB |
| [llama2-7b-hf-guanaco.Q5_1.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q5_1.gguf) | Q5_1 | 4.72GB |
| [llama2-7b-hf-guanaco.Q6_K.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q6_K.gguf) | Q6_K | 5.15GB |
| [llama2-7b-hf-guanaco.Q8_0.gguf](https://huggingface.co/RichardErkhov/TheTravellingEngineer_-_llama2-7b-hf-guanaco-gguf/blob/main/llama2-7b-hf-guanaco.Q8_0.gguf) | Q8_0 | 6.67GB |
Original model description:
The base model is meta's Llama-2-7b-hf. It was finetuned using SFT and the Guanaco dataset. The model prompt is similar to the original Guanaco model.
This repo contains the merged fp16 model.
**Legal Disclaimer: This model is bound by the usage restrictions of the original Llama-2 model. And comes with no warranty or gurantees of any kind.**
---
- license:
- llama2 <br>
- datasets:
- timdettmers/openassistant-guanaco <br>
- language:
- en <br>
- reference: https://gist.github.com/younesbelkada/9f7f75c94bdc1981c8ca5cc937d4a4da
---
|
Unnirengan/my-pet-dog | Unnirengan | 2024-06-03T12:23:23Z | 452 | 0 | diffusers | [
"diffusers",
"safetensors",
"NxtWave-GenAI-Webinar",
"text-to-image",
"stable-diffusion",
"license:creativeml-openrail-m",
"autotrain_compatible",
"endpoints_compatible",
"diffusers:StableDiffusionPipeline",
"region:us"
]
| text-to-image | 2024-06-03T12:19:12Z | ---
license: creativeml-openrail-m
tags:
- NxtWave-GenAI-Webinar
- text-to-image
- stable-diffusion
---
### My-Pet-Dog Dreambooth model trained by Unnirengan following the "Build your own Gen AI model" session by NxtWave.
Project Submission Code: 23031253
Sample pictures of this concept:
.jpeg)
|
ContextualAI/tiny-random-MistralForCausalLM | ContextualAI | 2024-06-18T18:02:56Z | 452 | 0 | transformers | [
"transformers",
"pytorch",
"mistral",
"text-generation",
"arxiv:1910.09700",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-06-18T18:01:37Z | ---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
M-FAC/bert-tiny-finetuned-qnli | M-FAC | 2021-12-13T08:11:40Z | 451 | 0 | transformers | [
"transformers",
"pytorch",
"bert",
"text-classification",
"arxiv:2107.03356",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
]
| text-classification | 2022-03-02T23:29:04Z | # BERT-tiny model finetuned with M-FAC
This model is finetuned on QNLI dataset with state-of-the-art second-order optimizer M-FAC.
Check NeurIPS 2021 paper for more details on M-FAC: [https://arxiv.org/pdf/2107.03356.pdf](https://arxiv.org/pdf/2107.03356.pdf).
## Finetuning setup
For fair comparison against default Adam baseline, we finetune the model in the same framework as described here [https://github.com/huggingface/transformers/tree/master/examples/pytorch/text-classification](https://github.com/huggingface/transformers/tree/master/examples/pytorch/text-classification) and just swap Adam optimizer with M-FAC.
Hyperparameters used by M-FAC optimizer:
```bash
learning rate = 1e-4
number of gradients = 1024
dampening = 1e-6
```
## Results
We share the best model out of 5 runs with the following score on QNLI validation set:
```bash
accuracy = 81.54
```
Mean and standard deviation for 5 runs on QNLI validation set:
| | Accuracy |
|:----:|:-----------:|
| Adam | 77.85 ± 0.15 |
| M-FAC | 81.17 ± 0.43 |
Results can be reproduced by adding M-FAC optimizer code in [https://github.com/huggingface/transformers/blob/master/examples/pytorch/text-classification/run_glue.py](https://github.com/huggingface/transformers/blob/master/examples/pytorch/text-classification/run_glue.py) and running the following bash script:
```bash
CUDA_VISIBLE_DEVICES=0 python run_glue.py \
--seed 8276 \
--model_name_or_path prajjwal1/bert-tiny \
--task_name qnli \
--do_train \
--do_eval \
--max_seq_length 128 \
--per_device_train_batch_size 32 \
--learning_rate 1e-4 \
--num_train_epochs 5 \
--output_dir out_dir/ \
--optim MFAC \
--optim_args '{"lr": 1e-4, "num_grads": 1024, "damp": 1e-6}'
```
We believe these results could be improved with modest tuning of hyperparameters: `per_device_train_batch_size`, `learning_rate`, `num_train_epochs`, `num_grads` and `damp`. For the sake of fair comparison and a robust default setup we use the same hyperparameters across all models (`bert-tiny`, `bert-mini`) and all datasets (SQuAD version 2 and GLUE).
Our code for M-FAC can be found here: [https://github.com/IST-DASLab/M-FAC](https://github.com/IST-DASLab/M-FAC).
A step-by-step tutorial on how to integrate and use M-FAC with any repository can be found here: [https://github.com/IST-DASLab/M-FAC/tree/master/tutorials](https://github.com/IST-DASLab/M-FAC/tree/master/tutorials).
## BibTeX entry and citation info
```bibtex
@article{frantar2021m,
title={M-FAC: Efficient Matrix-Free Approximations of Second-Order Information},
author={Frantar, Elias and Kurtic, Eldar and Alistarh, Dan},
journal={Advances in Neural Information Processing Systems},
volume={35},
year={2021}
}
```
|
data354/camembert-fr-covid-tweet-sentiment-classification | data354 | 2023-07-03T22:14:35Z | 451 | 0 | transformers | [
"transformers",
"pytorch",
"camembert",
"text-classification",
"classification",
"fr",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
]
| text-classification | 2022-03-02T23:29:04Z | ---
language:
- fr
tags:
- classification
license: apache-2.0
metrics:
- accuracy
widget:
- text: "tchai on est morts. on va se faire vacciner et ils vont contrôler comme les marionnettes avec des fils. d'après les 'ont dit'..."
---
# camembert-fr-covid-tweet-sentiment-classification
This model is a fine-tune checkpoint of [Yanzhu/bertweetfr-base](https://huggingface.co/Yanzhu/bertweetfr-base), fine-tuned on SST-2.
This model reaches an accuracy of 71% on the dev set.
In this dataset, given a tweet, the goal was to infer the underlying topic of the tweet by choosing from four topics classes:
- 0 : negatif
- 1 : neutre
- 2 : positif
# Pipelining the Model
```python
from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline
tokenizer = AutoTokenizer.from_pretrained("data354/camembert-fr-covid-tweet-sentiment-classification")
model = AutoModelForSequenceClassification.from_pretrained("data354/camembert-fr-covid-tweet-sentiment-classification")
nlp_topic_classif = transformers.pipeline('topics-classification', model = model, tokenizer = tokenizer)
nlp_topic_classif("tchai on est morts. on va se faire vacciner et ils vont contrôler comme les marionnettes avec des fils. d'après les '' ont dit ''...")
# Output: [{'label': 'opinions', 'score': 0.831]
``` |
sentence-transformers/msmarco-roberta-base-v2 | sentence-transformers | 2024-03-27T11:35:27Z | 451 | 1 | sentence-transformers | [
"sentence-transformers",
"pytorch",
"tf",
"jax",
"safetensors",
"roberta",
"feature-extraction",
"sentence-similarity",
"transformers",
"arxiv:1908.10084",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-embeddings-inference",
"region:us"
]
| sentence-similarity | 2022-03-02T23:29:05Z | ---
license: apache-2.0
library_name: sentence-transformers
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- transformers
pipeline_tag: sentence-similarity
---
# sentence-transformers/msmarco-roberta-base-v2
This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.
## Usage (Sentence-Transformers)
Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
```
pip install -U sentence-transformers
```
Then you can use the model like this:
```python
from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]
model = SentenceTransformer('sentence-transformers/msmarco-roberta-base-v2')
embeddings = model.encode(sentences)
print(embeddings)
```
## Usage (HuggingFace Transformers)
Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.
```python
from transformers import AutoTokenizer, AutoModel
import torch
#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
token_embeddings = model_output[0] #First element of model_output contains all token embeddings
input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
# Sentences we want sentence embeddings for
sentences = ['This is an example sentence', 'Each sentence is converted']
# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/msmarco-roberta-base-v2')
model = AutoModel.from_pretrained('sentence-transformers/msmarco-roberta-base-v2')
# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
# Compute token embeddings
with torch.no_grad():
model_output = model(**encoded_input)
# Perform pooling. In this case, max pooling.
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
print("Sentence embeddings:")
print(sentence_embeddings)
```
## Evaluation Results
For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=sentence-transformers/msmarco-roberta-base-v2)
## Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 250, 'do_lower_case': False}) with Transformer model: RobertaModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
)
```
## Citing & Authors
This model was trained by [sentence-transformers](https://www.sbert.net/).
If you find this model helpful, feel free to cite our publication [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://arxiv.org/abs/1908.10084):
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "http://arxiv.org/abs/1908.10084",
}
``` |
roneneldan/TinyStories-Instuct-1Layer-21M | roneneldan | 2023-05-17T22:06:22Z | 451 | 0 | transformers | [
"transformers",
"pytorch",
"gpt_neo",
"text-generation",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
]
| text-generation | 2023-05-12T21:56:07Z | Entry not found |
Salesforce/dialogstudio-t5-3b-v1.0 | Salesforce | 2023-09-07T18:30:00Z | 451 | 2 | transformers | [
"transformers",
"pytorch",
"t5",
"text2text-generation",
"dialog",
"en",
"fr",
"ro",
"de",
"multilingual",
"dataset:Salesforce/dialogstudio",
"dataset:flan",
"arxiv:2307.10172",
"arxiv:2210.11416",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text2text-generation | 2023-09-05T19:50:32Z | ---
language:
- en
- fr
- ro
- de
- multilingual
widget:
- text: "Continue the dialogue as a task-oriented dialogue system called SYSTEM. The answer of SYSTEM should follow the ACTION provided next while answering the USER's last utterance: \n<USER> Hello, I am looking for a restaurant in Cambridge. I believe it is called Golden Wok. \n<EXTERNAL KNOWLEDGE> ACTION: {'Restaurant-Inform': [['address', '191 Histon Road Chesterton']]}"
example_title: "Dialog Act to Response Generation"
- text: "Translate to German: My name is Arthur"
example_title: "Translation"
- text: "Please answer to the following question. Who is going to be the next Ballon d'or?"
example_title: "Question Answering"
- text: "Please answer the following question. What is the boiling point of Nitrogen?"
example_title: "Scientific knowledge"
- text: "Answer the following yes/no question. Can you write 200 words in a single tweet?"
example_title: "Yes/no question"
- text: "Answer the following yes/no question by reasoning step-by-step. Can you write 200 words in a single tweet?"
example_title: "Reasoning task"
- text: "Q: Is the statement ( `Jianguo is a research scientist at Salesforce AI` and `Jianguo is a student at UIC` ) True or Flase? A: Let's think step by step"
example_title: "Boolean Expressions"
- text: "The square root of x is the cube root of y. What is y to the power of 2, if x = 4?"
example_title: "Math reasoning"
- text: "Premise: At my age you will probably have learnt one lesson. Hypothesis: It's not certain how many lessons you'll learn by your thirties. Does the premise entail the hypothesis?"
example_title: "Premise and hypothesis"
inference:
parameters:
max_length: 256
tags:
- text2text-generation
- dialog
datasets:
- Salesforce/dialogstudio
- flan
license: apache-2.0
---
# Model Card for DialogStudio-T5 3B
<img src="https://huggingface.co/datasets/Salesforce/dialogstudio/resolve/main/logo.png"
alt="drawing" width="510"/>
# Table of Contents
0. [TL;DR](#TL;DR)
1. [Model Details](#model-details)
2. [Usage](#usage)
3. [Uses](#uses)
4. [Bias, Risks, and Limitations](#bias-risks-and-limitations)
5. [Training Details](#training-details)
6. [Evaluation](#evaluation)
7. [Environmental Impact](#environmental-impact)
8. [Citation](#citation)
9. [Model Card Authors](#model-card-authors)
# TL;DR
If you already know T5 and Flan-T5, DialogStudio-T5 is better at many things. With the same number of parameters, the models are fine-tuned from a selected amount of dialogues from [DialogStudio](https://github.com/salesforce/DialogStudio) and also 1000 additional tasks.
**Disclaimer**: Content from **this** model card are modified from contents written by the Hugging Face team, and parts of it were copy pasted from the [T5 model card](https://huggingface.co/t5-large) and [Flan-T5 model card](https://huggingface.co/google/flan-t5-large).
**Follow the [DialogStudio](https://github.com/salesforce/DialogStudio) GitHub repository for latest information.**
# Model Details
## Data
We sample a small amount of dialogues from each commercial supported dataset under three categories of [DialogStudio](https://huggingface.co/datasets/Salesforce/dialogstudio), i.e., KG-Dial, TOD and Open-Domain dialogues. Additionally, we sample at most 150 examples for each non-translation task from [FLAN](https://github.com/google-research/FLAN/tree/main/flan/v2).
**Note** that this model version 1.0 does not incorporate datasets utilized for training large-scale models (>=7B) like Alpaca, ShareGPT, GPT4ALL, UltraChat from OpenAI's 'GPT-3.5/4', or other datasets such as OASST1 and WizardCoder.
<img src="https://huggingface.co/datasets/Salesforce/dialogstudio/resolve/main/DialogStudio_Stats.jpg"
alt="drawing" width="700"/>
## Model Description
- **Model type:** Language model
- **Language(s) (NLP):** English, Spanish, Japanese, Persian, Hindi, French, Chinese, Bengali, Gujarati, German, Telugu, Italian, Arabic, Polish, Tamil, Marathi, Malayalam, Oriya, Panjabi, Portuguese, Urdu, Galician, Hebrew, Korean, Catalan, Thai, Dutch, Indonesian, Vietnamese, Bulgarian, Filipino, Central Khmer, Lao, Turkish, Russian, Croatian, Swedish, Yoruba, Kurdish, Burmese, Malay, Czech, Finnish, Somali, Tagalog, Swahili, Sinhala, Kannada, Zhuang, Igbo, Xhosa, Romanian, Haitian, Estonian, Slovak, Lithuanian, Greek, Nepali, Assamese, Norwegian
- **License:** Apache 2.0
- **Related Models:** [All DialogStudio-T5 Checkpoints](https://huggingface.co/models?search=dialogstudio-t5)
- **Resources for more information:**
- [Research paper](https://arxiv.org/abs/2307.10172)
- [GitHub Repo](https://github.com/salesforce/DialogStudio)
- **Maximum model length:**:
- Maximum input length: 1200
- Maximum output length: 256
- **Training formats:**
- We process dialogue data into below input format :
- With instruction and external knowledge: ```Instruction: your instruction <USER> user utterance 1 <SYSTEM> system utterance 1 ... <USER> user utterance N <EXTERNAL KNOWLEDGE> your external knowledge```
- Without instruction: ```<USER> user utterance 1 <SYSTEM> system utterance 1 ... <USER> user utterance N <EXTERNAL KNOWLEDGE> your external knowledge```
- Without external knowledge: ```Instruction: your instruction <USER> user utterance 1 <SYSTEM> system utterance 1 ... <USER> user utterance N```
- Without both: ```<USER> user utterance 1 <SYSTEM> system utterance 1 ... <USER> user utterance N```
- Note: output is final the system response; `<USER>`, `<SYSTEM>` and `<EXTERNAL KNOWLEDGE>` are special tokens
- For sampled FLAN data:
- We follow their original data format, i.e., we did not set special tokens to separate in-context learning examples.
- In summary:
- We recommend you use our format and add our special tokens (such as `<USER>` and `<SYSTEM>` ) to get better performance. However, you may not necessary need to exactly follow our format if you do not observe random behavios.
- We found that T5 model series such as Flan-t5 and DialogStudio-T5 may generate repetitive tokens during inference. If you find such repetition issues, you can set the `repetition_penalty` in model.generate(), such as 1.5, to mitigate them. Note that `repetition_penalty=1.0` by default.
# Usage
Find below some example scripts on how to use the model in `transformers`:
## Using the Pytorch model
### Running the model on a CPU
<details>
<summary> Click to expand </summary>
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("Salesforce/dialogstudio-t5-3b-v1.0")
model = AutoModelForSeq2SeqLM.from_pretrained("Salesforce/dialogstudio-t5-3b-v1.0")
input_text = "Answer the following yes/no question by reasoning step-by-step. Can you write 200 words in a single tweet?"
input_ids = tokenizer(input_text, return_tensors="pt").input_ids
outputs = model.generate(input_ids, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
</details>
### Running the model on a GPU
<details>
<summary> Click to expand </summary>
```python
# pip install accelerate
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("Salesforce/dialogstudio-t5-3b-v1.0")
model = AutoModelForSeq2SeqLM.from_pretrained("Salesforce/dialogstudio-t5-3b-v1.0", device_map="auto")
input_text = "Answer the following yes/no question by reasoning step-by-step. Can you write 200 words in a single tweet?"
input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to("cuda")
outputs = model.generate(input_ids, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
</details>
### Running the model on a GPU using different precisions
#### FP16
<details>
<summary> Click to expand </summary>
```python
# pip install accelerate
import torch
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("Salesforce/dialogstudio-t5-3b-v1.0")
model = AutoModelForSeq2SeqLM.from_pretrained("Salesforce/dialogstudio-t5-3b-v1.0", device_map="auto", torch_dtype=torch.float16)
input_text = "Answer the following yes/no question by reasoning step-by-step. Can you write 200 words in a single tweet?"
input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to("cuda")
outputs = model.generate(input_ids, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
</details>
#### INT8
<details>
<summary> Click to expand </summary>
```python
# pip install bitsandbytes accelerate
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("Salesforce/dialogstudio-t5-3b-v1.0")
model = AutoModelForSeq2SeqLM.from_pretrained("Salesforce/dialogstudio-t5-3b-v1.0", device_map="auto", load_in_8bit=True)
input_text = "Answer the following yes/no question by reasoning step-by-step. Can you write 200 words in a single tweet?"
input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to("cuda")
outputs = model.generate(input_ids, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
</details>
# Uses
## Direct Use and Downstream Use
<!-- The authors write in [the original paper's model card](https://arxiv.org/pdf/2210.11416.pdf) that: -->
> The primary use is research on language models, including: research on zero-shot NLP tasks and in-context few-shot learning NLP tasks, such as dialogue response generation, reasoning, and question answering; advancing fairness and safety research, and understanding limitations of current large language models
## Out-of-Scope Use
More information needed.
# Bias, Risks, and Limitations
The information below in this section are copied and modified from Flan-T5's models card:
> Language models, including DialogStudio-T5, can potentially be used for language generation in a harmful way, according to Rae et al. (2021). DialogStudio-T5 should not be used directly in any application, without a prior assessment of safety and fairness concerns specific to the application.
## Ethical considerations and risks
> DialogStudio-T5 is fine-tuned on a large corpus of text data that was not filtered for explicit content or assessed for existing biases. As a result the model itself is potentially vulnerable to generating equivalently inappropriate content or replicating inherent biases in the underlying data.
## Known Limitations
> DialogStudio-T5 has not been tested in real world applications.
## Sensitive Use:
> DialogStudio-T5 should not be applied for any unacceptable use cases, e.g., generation of abusive speech.
# Training Details
## Training Data
We sample a small amount of dialogues from each commercial supported dataset under three categories of [DialogStudio](https://huggingface.co/datasets/Salesforce/dialogstudio), i.e., KG-Dial, TOD and Open-Domain dialogues. Additionally, we sample at most 150 examples for each non-translation task from [FLAN](https://github.com/google-research/FLAN/tree/main/flan/v2).
**Note:**
Model Version 1.0 is built on small-scale pre-trained models, this version does not incorporate datasets utilized for training large-scale models (>=7B) like Alpaca, ShareGPT, GPT4ALL, UltraChat from OpenAI's 'GPT-3.5/4', or other datasets such as OASST1 and WizardCoder. As a result, it has certain limitations in terms of writing and creative capabilities. Our initial focus is to update the model versions to enhance existing abilities. Further improvements, including expansion of other capabilities, are part of our roadmap and will be responsive to community requests.
See above **Training formats:** for details of the training formats.
## Training Procedure
> These models are based on Flan-T5 and are fine-tuned with instructions for better zero-shot and few-shot performance. There is one fine-tuned DialogStudio model per T5 model size.
The model has been trained on 16 A100 GPUs, each with 40G memory, using public [transformer](https://github.com/huggingface/transformers) codebase.
# Evaluation
## Testing Data, Factors & Metrics
The authors evaluated the model on several dialogue tasks and general tasks such as 0-shot/5-shot MMLU and 3-shot BBH.
## Results
For full results for DialogStudio, see the [research paper](https://arxiv.org/abs/2307.10172).
## Environmental Impact
More information needed.
# Citation
**BibTeX:**
```bibtex
@misc{zhang2023dialogstudio,
title={DialogStudio: Towards Richest and Most Diverse Unified Dataset Collection for Conversational AI},
author={Jianguo Zhang and Kun Qian and Zhiwei Liu and Shelby Heinecke and Rui Meng and Ye Liu and Zhou Yu and and Huan Wang and Silvio Savarese and Caiming Xiong},
year={2023},
eprint={2307.10172},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
``` |
bsp-albz/llama2-7b-platypus-ckpt-1000 | bsp-albz | 2023-09-12T18:50:02Z | 451 | 0 | transformers | [
"transformers",
"pytorch",
"llama",
"feature-extraction",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| feature-extraction | 2023-09-12T17:55:09Z | Entry not found |
TheBloke/ALMA-13B-GGUF | TheBloke | 2023-10-14T17:48:27Z | 451 | 3 | transformers | [
"transformers",
"gguf",
"llama",
"arxiv:2309.11674",
"base_model:haoranxu/ALMA-13B",
"license:mit",
"text-generation-inference",
"region:us"
]
| null | 2023-10-14T17:32:55Z | ---
base_model: haoranxu/ALMA-13B
inference: false
license: mit
model_creator: Haoran Xu
model_name: ALMA 13B
model_type: llama
prompt_template: 'Translate this from Chinese to English:
Chinese: {prompt}
English:
'
quantized_by: TheBloke
---
<!-- header start -->
<!-- 200823 -->
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div>
<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
<!-- header end -->
# ALMA 13B - GGUF
- Model creator: [Haoran Xu](https://huggingface.co/haoranxu)
- Original model: [ALMA 13B](https://huggingface.co/haoranxu/ALMA-13B)
<!-- description start -->
## Description
This repo contains GGUF format model files for [Haoran Xu's ALMA 13B](https://huggingface.co/haoranxu/ALMA-13B).
<!-- description end -->
<!-- README_GGUF.md-about-gguf start -->
### About GGUF
GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp.
Here is an incomplate list of clients and libraries that are known to support GGUF:
* [llama.cpp](https://github.com/ggerganov/llama.cpp). The source project for GGUF. Offers a CLI and a server option.
* [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
* [KoboldCpp](https://github.com/LostRuins/koboldcpp), a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
* [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration.
* [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with many interesting and unique features, including a full model library for easy model selection.
* [Faraday.dev](https://faraday.dev/), an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
* [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
* [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
* [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
<!-- README_GGUF.md-about-gguf end -->
<!-- repositories-available start -->
## Repositories available
* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/ALMA-13B-AWQ)
* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/ALMA-13B-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/ALMA-13B-GGUF)
* [Haoran Xu's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/haoranxu/ALMA-13B)
<!-- repositories-available end -->
<!-- prompt-template start -->
## Prompt template: ALMA
```
Translate this from Chinese to English:
Chinese: {prompt}
English:
```
<!-- prompt-template end -->
<!-- licensing start -->
## Licensing
The creator of the source model has listed its license as `mit`, and this quantization has therefore used that same license.
As this model is based on Llama 2, it is also subject to the Meta Llama 2 license terms, and the license files for that are additionally included. It should therefore be considered as being claimed to be licensed under both licenses. I contacted Hugging Face for clarification on dual licensing but they do not yet have an official position. Should this change, or should Meta provide any feedback on this situation, I will update this section accordingly.
In the meantime, any questions regarding licensing, and in particular how these two licenses might interact, should be directed to the original model repository: [Haoran Xu's ALMA 13B](https://huggingface.co/haoranxu/ALMA-13B).
<!-- licensing end -->
<!-- compatibility_gguf start -->
## Compatibility
These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221)
They are also compatible with many third party UIs and libraries - please see the list at the top of this README.
## Explanation of quantisation methods
<details>
<summary>Click to see details</summary>
The new methods available are:
* GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
* GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
* GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
* GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
* GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw
Refer to the Provided Files table below to see what files use which methods, and how.
</details>
<!-- compatibility_gguf end -->
<!-- README_GGUF.md-provided-files start -->
## Provided files
| Name | Quant method | Bits | Size | Max RAM required | Use case |
| ---- | ---- | ---- | ---- | ---- | ----- |
| [alma-13b.Q2_K.gguf](https://huggingface.co/TheBloke/ALMA-13B-GGUF/blob/main/alma-13b.Q2_K.gguf) | Q2_K | 2 | 5.43 GB| 7.93 GB | smallest, significant quality loss - not recommended for most purposes |
| [alma-13b.Q3_K_S.gguf](https://huggingface.co/TheBloke/ALMA-13B-GGUF/blob/main/alma-13b.Q3_K_S.gguf) | Q3_K_S | 3 | 5.66 GB| 8.16 GB | very small, high quality loss |
| [alma-13b.Q3_K_M.gguf](https://huggingface.co/TheBloke/ALMA-13B-GGUF/blob/main/alma-13b.Q3_K_M.gguf) | Q3_K_M | 3 | 6.34 GB| 8.84 GB | very small, high quality loss |
| [alma-13b.Q3_K_L.gguf](https://huggingface.co/TheBloke/ALMA-13B-GGUF/blob/main/alma-13b.Q3_K_L.gguf) | Q3_K_L | 3 | 6.93 GB| 9.43 GB | small, substantial quality loss |
| [alma-13b.Q4_0.gguf](https://huggingface.co/TheBloke/ALMA-13B-GGUF/blob/main/alma-13b.Q4_0.gguf) | Q4_0 | 4 | 7.37 GB| 9.87 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [alma-13b.Q4_K_S.gguf](https://huggingface.co/TheBloke/ALMA-13B-GGUF/blob/main/alma-13b.Q4_K_S.gguf) | Q4_K_S | 4 | 7.41 GB| 9.91 GB | small, greater quality loss |
| [alma-13b.Q4_K_M.gguf](https://huggingface.co/TheBloke/ALMA-13B-GGUF/blob/main/alma-13b.Q4_K_M.gguf) | Q4_K_M | 4 | 7.87 GB| 10.37 GB | medium, balanced quality - recommended |
| [alma-13b.Q5_0.gguf](https://huggingface.co/TheBloke/ALMA-13B-GGUF/blob/main/alma-13b.Q5_0.gguf) | Q5_0 | 5 | 8.97 GB| 11.47 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [alma-13b.Q5_K_S.gguf](https://huggingface.co/TheBloke/ALMA-13B-GGUF/blob/main/alma-13b.Q5_K_S.gguf) | Q5_K_S | 5 | 8.97 GB| 11.47 GB | large, low quality loss - recommended |
| [alma-13b.Q5_K_M.gguf](https://huggingface.co/TheBloke/ALMA-13B-GGUF/blob/main/alma-13b.Q5_K_M.gguf) | Q5_K_M | 5 | 9.23 GB| 11.73 GB | large, very low quality loss - recommended |
| [alma-13b.Q6_K.gguf](https://huggingface.co/TheBloke/ALMA-13B-GGUF/blob/main/alma-13b.Q6_K.gguf) | Q6_K | 6 | 10.68 GB| 13.18 GB | very large, extremely low quality loss |
| [alma-13b.Q8_0.gguf](https://huggingface.co/TheBloke/ALMA-13B-GGUF/blob/main/alma-13b.Q8_0.gguf) | Q8_0 | 8 | 13.83 GB| 16.33 GB | very large, extremely low quality loss - not recommended |
**Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
<!-- README_GGUF.md-provided-files end -->
<!-- README_GGUF.md-how-to-download start -->
## How to download GGUF files
**Note for manual downloaders:** You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.
The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
- LM Studio
- LoLLMS Web UI
- Faraday.dev
### In `text-generation-webui`
Under Download Model, you can enter the model repo: TheBloke/ALMA-13B-GGUF and below it, a specific filename to download, such as: alma-13b.Q4_K_M.gguf.
Then click Download.
### On the command line, including multiple files at once
I recommend using the `huggingface-hub` Python library:
```shell
pip3 install huggingface-hub
```
Then you can download any individual model file to the current directory, at high speed, with a command like this:
```shell
huggingface-cli download TheBloke/ALMA-13B-GGUF alma-13b.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
<details>
<summary>More advanced huggingface-cli download usage</summary>
You can also download multiple files at once with a pattern:
```shell
huggingface-cli download TheBloke/ALMA-13B-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
For more documentation on downloading with `huggingface-cli`, please see: [HF -> Hub Python Library -> Download files -> Download from the CLI](https://huggingface.co/docs/huggingface_hub/guides/download#download-from-the-cli).
To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
```shell
pip3 install hf_transfer
```
And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
```shell
HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/ALMA-13B-GGUF alma-13b.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
```
Windows Command Line users: You can set the environment variable by running `set HF_HUB_ENABLE_HF_TRANSFER=1` before the download command.
</details>
<!-- README_GGUF.md-how-to-download end -->
<!-- README_GGUF.md-how-to-run start -->
## Example `llama.cpp` command
Make sure you are using `llama.cpp` from commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
```shell
./main -ngl 32 -m alma-13b.Q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "Translate this from Chinese to English:\nChinese: {prompt}\nEnglish:"
```
Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
Change `-c 4096` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.
If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
For other parameters and how to use them, please refer to [the llama.cpp documentation](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)
## How to run in `text-generation-webui`
Further instructions here: [text-generation-webui/docs/llama.cpp.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/llama.cpp.md).
## How to run from Python code
You can use GGUF models from Python using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) or [ctransformers](https://github.com/marella/ctransformers) libraries.
### How to load this model in Python code, using ctransformers
#### First install the package
Run one of the following commands, according to your system:
```shell
# Base ctransformers with no GPU acceleration
pip install ctransformers
# Or with CUDA GPU acceleration
pip install ctransformers[cuda]
# Or with AMD ROCm GPU acceleration (Linux only)
CT_HIPBLAS=1 pip install ctransformers --no-binary ctransformers
# Or with Metal GPU acceleration for macOS systems only
CT_METAL=1 pip install ctransformers --no-binary ctransformers
```
#### Simple ctransformers example code
```python
from ctransformers import AutoModelForCausalLM
# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
llm = AutoModelForCausalLM.from_pretrained("TheBloke/ALMA-13B-GGUF", model_file="alma-13b.Q4_K_M.gguf", model_type="llama", gpu_layers=50)
print(llm("AI is going to"))
```
## How to use with LangChain
Here are guides on using llama-cpp-python and ctransformers with LangChain:
* [LangChain + llama-cpp-python](https://python.langchain.com/docs/integrations/llms/llamacpp)
* [LangChain + ctransformers](https://python.langchain.com/docs/integrations/providers/ctransformers)
<!-- README_GGUF.md-how-to-run end -->
<!-- footer start -->
<!-- 200823 -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/theblokeai)
## Thanks, and how to contribute
Thanks to the [chirper.ai](https://chirper.ai) team!
Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Special thanks to**: Aemon Algiz.
**Patreon special mentions**: Pierre Kircher, Stanislav Ovsiannikov, Michael Levine, Eugene Pentland, Andrey, 준교 김, Randy H, Fred von Graf, Artur Olbinski, Caitlyn Gatomon, terasurfer, Jeff Scroggin, James Bentley, Vadim, Gabriel Puliatti, Harry Royden McLaughlin, Sean Connelly, Dan Guido, Edmond Seymore, Alicia Loh, subjectnull, AzureBlack, Manuel Alberto Morcote, Thomas Belote, Lone Striker, Chris Smitley, Vitor Caleffi, Johann-Peter Hartmann, Clay Pascal, biorpg, Brandon Frisco, sidney chen, transmissions 11, Pedro Madruga, jinyuan sun, Ajan Kanaga, Emad Mostaque, Trenton Dambrowitz, Jonathan Leane, Iucharbius, usrbinkat, vamX, George Stoitzev, Luke Pendergrass, theTransient, Olakabola, Swaroop Kallakuri, Cap'n Zoog, Brandon Phillips, Michael Dempsey, Nikolai Manek, danny, Matthew Berman, Gabriel Tamborski, alfie_i, Raymond Fosdick, Tom X Nguyen, Raven Klaugh, LangChain4j, Magnesian, Illia Dulskyi, David Ziegler, Mano Prime, Luis Javier Navarrete Lozano, Erik Bjäreholt, 阿明, Nathan Dryer, Alex, Rainer Wilmers, zynix, TL, Joseph William Delisle, John Villwock, Nathan LeClaire, Willem Michiel, Joguhyik, GodLy, OG, Alps Aficionado, Jeffrey Morgan, ReadyPlayerEmma, Tiffany J. Kim, Sebastain Graf, Spencer Kim, Michael Davis, webtim, Talal Aujan, knownsqashed, John Detwiler, Imad Khwaja, Deo Leter, Jerry Meng, Elijah Stavena, Rooh Singh, Pieter, SuperWojo, Alexandros Triantafyllidis, Stephen Murray, Ai Maven, ya boyyy, Enrico Ros, Ken Nordquist, Deep Realms, Nicholas, Spiking Neurons AB, Elle, Will Dee, Jack West, RoA, Luke @flexchar, Viktor Bowallius, Derek Yates, Subspace Studios, jjj, Toran Billups, Asp the Wyvern, Fen Risland, Ilya, NimbleBox.ai, Chadd, Nitin Borwankar, Emre, Mandus, Leonard Tan, Kalila, K, Trailburnt, S_X, Cory Kujawski
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end -->
<!-- original-model-card start -->
# Original model card: Haoran Xu's ALMA 13B
**ALMA** (**A**dvanced **L**anguage **M**odel-based tr**A**nslator) is an LLM-based translation model, which adopts a new translation model paradigm: it begins with fine-tuning on monolingual data and is further optimized using high-quality parallel data. This two-step fine-tuning process ensures strong translation performance.
Please find more details in our [paper](https://arxiv.org/abs/2309.11674).
```
@misc{xu2023paradigm,
title={A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models},
author={Haoran Xu and Young Jin Kim and Amr Sharaf and Hany Hassan Awadalla},
year={2023},
eprint={2309.11674},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
We release four translation models presented in the paper:
- **ALMA-7B**: Full-weight Fine-tune LLaMA-2-7B on 20B monolingual tokens and then **Full-weight** fine-tune on human-written parallel data
- **ALMA-7B-LoRA**: Full-weight Fine-tune LLaMA-2-7B on 20B monolingual tokens and then **LoRA** fine-tune on human-written parallel data
- **ALMA-13B**: Full-weight Fine-tune LLaMA-2-7B on 12B monolingual tokens and then **Full-weight** fine-tune on human-written parallel data
- **ALMA-13B-LoRA** (Our best system): Full-weight Fine-tune LLaMA-2-7B on 12B monolingual tokens and then **LoRA** fine-tune on human-written parallel data
Model checkpoints are released at huggingface:
| Models | Base Model Link | LoRA Link |
|:-------------:|:---------------:|:---------:|
| ALMA-7B | [haoranxu/ALMA-7B](https://huggingface.co/haoranxu/ALMA-7B) | - |
| ALMA-7B-LoRA | [haoranxu/ALMA-7B-Pretrain](https://huggingface.co/haoranxu/ALMA-7B-Pretrain) | [haoranxu/ALMA-7B-Pretrain-LoRA](https://huggingface.co/haoranxu/ALMA-7B-Pretrain-LoRA) |
| ALMA-13B | [haoranxu/ALMA-13B](https://huggingface.co/haoranxu/ALMA-13B) | - |
| ALMA-13B-LoRA | [haoranxu/ALMA-13B-Pretrain](https://huggingface.co/haoranxu/ALMA-13B-Pretrain) | [haoranxu/ALMA-13B-Pretrain-LoRA](https://huggingface.co/haoranxu/ALMA-13B-Pretrain-LoRA) |
**Note that `ALMA-7B-Pretrain` and `ALMA-13B-Pretrain` are NOT translation models. They only experience stage 1 monolingual fine-tuning (20B tokens for the 7B model and 12B tokens for the 13B model), and should be utilized in conjunction with their LoRA models for translation purposes.**
A quick start to use our best system (ALMA-13B-LoRA) for translation. An example of translating "我爱机器翻译。" into English:
```
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM
from transformers import LlamaTokenizer
# Load base model and LoRA weights
model = AutoModelForCausalLM.from_pretrained("haoranxu/ALMA-13B-Pretrain", torch_dtype=torch.float16, device_map="auto")
model = PeftModel.from_pretrained(model, "haoranxu/ALMA-13B-Pretrain-LoRA")
tokenizer = LlamaTokenizer.from_pretrained("haoranxu/ALMA-13B-Pretrain", padding_side='left')
# Add the source setence into the prompt template
prompt="Translate this from Chinese to English:\nChinese: 我爱机器翻译。\nEnglish:"
input_ids = tokenizer(prompt, return_tensors="pt", padding=True, max_length=40, truncation=True).input_ids.cuda()
# Translation
with torch.no_grad():
generated_ids = model.generate(input_ids=input_ids, num_beams=5, max_new_tokens=20, do_sample=True, temperature=0.6, top_p=0.9)
outputs = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
print(outputs)
```
Please find more details in our [GitHub repository](https://github.com/fe1ixxu/ALMA)
<!-- original-model-card end -->
|
Sayan01/T5-Flan-Small | Sayan01 | 2023-12-05T07:44:18Z | 451 | 0 | transformers | [
"transformers",
"safetensors",
"t5",
"text2text-generation",
"en",
"dataset:Open-Orca/FLAN",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text2text-generation | 2023-11-20T11:30:13Z | ---
license: apache-2.0
datasets:
- Open-Orca/FLAN
language:
- en
pipeline_tag: text2text-generation
---
T5-small distilled from T5-flan-base using OpenOrca'a FLAN dataset using Seq2seq methodology. The baseline for Seq2seq distillation on T5 decoder models. It is free under Apache License.
The distilled model is a T5-small model with 60M parameters, not a T5-flan-small model. |
s3nh/NousHermes-Kunoichi-SolarMaid-7b | s3nh | 2024-01-22T18:25:30Z | 451 | 1 | transformers | [
"transformers",
"safetensors",
"llama",
"text-generation",
"merge",
"base_model:Undi95/SolarMaid-v0.1.1",
"base_model:SanjiWatsuki/Kunoichi-7B",
"license:openrail",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-01-05T21:42:12Z | ---
license: openrail
tags:
- merge
base_model:
- Undi95/SolarMaid-v0.1.1
- SanjiWatsuki/Kunoichi-7B
---
## This model does not work at the moment, its in expertimental phase. Ill update it
# merged
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
## Merge Details
### Merge Method
This model was merged using the SLERP merge method.
### Models Merged
The following models were included in the merge:
* [SanjiWatsuki/Kunoichi-7B](https://huggingface.co/SanjiWatsuki/Kunoichi-7B)
* [Undi95/SolarMaid-v0.1.1](https://huggingface.co/Undi95/SolarMaid-v0.1.1)
* [NousResearch/Nous-Hermes-llama-2-7b](https://huggingface.co/NousResearch/Nous-Hermes-llama-2-7b)
### Configuration
The following YAML configuration was used to produce this model:
```yaml
base_model: NousResearch/Nous-Hermes-llama-2-7b
dtype: bfloat16
merge_method: slerp
parameters:
t:
- filter: self_attn
value: [0.0, 0.5, 0.3, 0.7, 1.0]
- filter: mlp
value: [1.0, 0.5, 0.7, 0.3, 0.0]
- value: 0.5
slices:
- sources:
- layer_range: [0, 32]
model: Undi95/SolarMaid-v0.1.1
- layer_range: [0, 32]
model: SanjiWatsuki/Kunoichi-7B
``` |
gardner/TinyLlama-1.1B-Instruct-3T | gardner | 2024-01-20T08:23:57Z | 451 | 0 | transformers | [
"transformers",
"pytorch",
"llama",
"text-generation",
"instruct",
"openhermes",
"tinyllama",
"en",
"dataset:teknium/openhermes",
"base_model:TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"text-generation-inference",
"region:us"
]
| text-generation | 2024-01-20T07:44:59Z | ---
language:
- en
thumbnail: "url to a thumbnail used in social sharing"
tags:
- instruct
- openhermes
- tinyllama
license: apache-2.0
datasets:
- teknium/openhermes
metrics:
- metric1
- metric2
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
---
## TinyLlama 1.1B Instruct 3T
<img src="https://huggingface.co/gardner/TinyLlama-1.1B-Instruct-3T/resolve/main/tinyllama-1.1b-instruct.webp?download=true" alt="TinyLlama Instruct" />
This is the 3T base model trained on openhermes instruct dataset for 4 epochs. It is intended to be used for further finetuning.
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
## Loss
<img src="https://huggingface.co/gardner/TinyLlama-1.1B-Instruct-3T/resolve/main/loss.webp?download=true" alt="Loss chart" />
## axolotl config file: lora.yml
```yaml
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
model_type: LlamaForCausalLM
tokenizer_type: LlamaTokenizer
is_llama_derived_model: true
load_in_8bit: true
load_in_4bit: false
strict: false
datasets:
- path: teknium/openhermes
type: alpaca
dataset_prepared_path:
val_set_size: 0.05
output_dir: ./tiny-llama-instruct-lora
sequence_len: 4096
sample_packing: true
pad_to_sequence_len: true
adapter: lora
lora_model_dir:
lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:
gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 4
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002
train_on_inputs: false
group_by_length: false
bf16: true
fp16: false
tf32: false
gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true
warmup_steps: 10
evals_per_epoch: 4
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
``` |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.