Initial GPTQ model commit

README.md

@@ -37,15 +37,13 @@ Multiple GPTQ parameter permutations are provided; see Provided Files below for
 * [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/stablecode-instruct-alpha-3b-GGML)
 * [StabilityAI's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/stabilityai/stablecode-instruct-alpha-3b)
 
-## Prompt template: Alpaca
+## Prompt template: StableCode
 
 ```
-Below is an instruction that describes a task. Write a response that appropriately completes the request.
-
-### Instruction:
+###Instruction:
 {prompt}
 
-### Response:
+###Response:
 ```
 
 ## Provided files and GPTQ parameters
@@ -76,7 +74,7 @@ All GPTQ files are made with AutoGPTQ.
 | [gptq-4bit-64g-actorder_True](https://huggingface.co/TheBloke/stablecode-instruct-alpha-3b-GPTQ/tree/gptq-4bit-64g-actorder_True) | 4 | 64 | Yes | 0.1 | [Evol Instruct Code](https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1) | 4096 | 1.86 GB | No | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. | 
 | [gptq-4bit-128g-actorder_True](https://huggingface.co/TheBloke/stablecode-instruct-alpha-3b-GPTQ/tree/gptq-4bit-128g-actorder_True) | 4 | 128 | Yes | 0.1 | [Evol Instruct Code](https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1) | 4096 | 1.82 GB | No | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. | 
 | [gptq-8bit-128g-actorder_True](https://huggingface.co/TheBloke/stablecode-instruct-alpha-3b-GPTQ/tree/gptq-8bit-128g-actorder_True) | 8 | 128 | Yes | 0.1 | [Evol Instruct Code](https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1) | 4096 | 3.08 GB | No | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. Poor AutoGPTQ CUDA speed. | 
-| [gptq-8bit-64g-actorder_True](https://huggingface.co/TheBloke/stablecode-instruct-alpha-3b-GPTQ/tree/gptq-8bit-64g-actorder_True) | 8 | 64 | Yes | 0.1 | [Evol Instruct Code](https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1) | 4096 | 3.14 GB | No | 8-bit, with group size 64g and Act Order for maximum inference quality. Poor AutoGPTQ CUDA speed. |
+| [gptq-8bit-64g-actorder_True](https://huggingface.co/TheBloke/stablecode-instruct-alpha-3b-GPTQ/tree/gptq-8bit-64g-actorder_True) | 8 | 64 | Yes | 0.1 | [Evol Instruct Code](https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1) | 4096 | 3.14 GB | No | 8-bit, with group size 64g and Act Order for even higher inference quality. Poor AutoGPTQ CUDA speed. |
 
 ## How to download from branches
 
@@ -154,12 +152,10 @@ model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
 """
 
 prompt = "Tell me about AI"
-prompt_template=f'''Below is an instruction that describes a task. Write a response that appropriately completes the request.
-
-### Instruction:
+prompt_template=f'''###Instruction:
 {prompt}
 
-### Response:
+###Response:
 '''
 
 print("\n\n*** Generate:")
@@ -224,4 +220,75 @@ Thank you to all my generous patrons and donaters!
 
 # Original model card: StabilityAI's Stablecode Instruct Alpha 3B
 
-No original model card was provided.
+# `StableCode-Instruct-Alpha-3B`
+
+## Model Description
+
+`StableCode-Instruct-Alpha-3B` is a 3 billion parameter decoder-only instruction tuned code model pre-trained on diverse set of programming languages that topped the stackoverflow developer survey.
+
+## Usage
+The model is intended to follow instruction to generate code. The dataset used to train the model is formatted in Alpaca format.
+Get started generating code with `StableCode-Instruct-Alpha-3B` by using the following code snippet:
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+tokenizer = AutoTokenizer.from_pretrained("stabilityai/stablecode-instruct-alpha-3b")
+model = AutoModelForCausalLM.from_pretrained(
+  "stabilityai/stablecode-instruct-alpha-3b",
+  trust_remote_code=True,
+  torch_dtype="auto",
+)
+model.cuda()
+inputs = tokenizer("###Instruction\nGenerate a python function to find number of CPU cores###Response\n", return_tensors="pt").to("cuda")
+tokens = model.generate(
+  **inputs,
+  max_new_tokens=48,
+  temperature=0.2,
+  do_sample=True,
+)
+print(tokenizer.decode(tokens[0], skip_special_tokens=True))
+```
+
+## Model Details
+
+* **Developed by**: [Stability AI](https://stability.ai/)
+* **Model type**: `StableCode-Instruct-Alpha-3B` models are auto-regressive language models based on the transformer decoder architecture.
+* **Language(s)**: Code
+* **Library**: [GPT-NeoX](https://github.com/EleutherAI/gpt-neox)
+* **License** : Model checkpoints are licensed under the [StableCode Research License](https://huggingface.co/stabilityai/stablecode-instruct-alpha-3b/blob/main/LICENSE.md) Copyright (c) Stability AI Ltd. All Rights Reserved
+* **Contact**: For questions and comments about the model, please email `[email protected]`
+
+### Model Architecture
+
+| Parameters     | Hidden Size | Layers | Heads | Sequence Length |
+|----------------|-------------|--------|-------|-----------------|
+| 2,796,431,360  | 2560        | 32     | 32    | 4096            |
+
+
+* **Decoder Layer**: Parallel Attention and MLP residuals with a single input LayerNorm ([Wang & Komatsuzaki, 2021](https://github.com/kingoflolz/mesh-transformer-jax/tree/master))
+* **Position Embeddings**: Rotary Position Embeddings ([Su et al., 2021](https://arxiv.org/abs/2104.09864))
+* **Bias**: LayerNorm bias terms only
+
+## Training
+
+`StableCode-Instruct-Alpha-3B` is the instruction finetuned version on [StableCode-Completion-Alpha-3B](https://huggingface.co/stabilityai/stablecode-completion-alpha-3b) with code instruction datasets.
+
+## Use and Limitations
+
+### Intended Use
+
+StableCode-Instruct-Alpha-3B independently generates new code completions, but we recommend that you use StableCode-Instruct-Alpha-3B together with the tool developed by BigCode and HuggingFace [(huggingface/huggingface-vscode: Code completion VSCode extension for OSS models (github.com))](https://github.com/huggingface/huggingface-vscode), to identify and, if necessary, attribute any outputs that match training code.
+
+### Limitations and bias
+
+This model is intended to be used responsibly. It is not intended to be used to create unlawful content of any kind, to further any unlawful activity, or to engage in activities with a high risk of physical or economic harm.
+
+## How to cite
+
+```bibtex
+@misc{StableCodeInstructAlpha, 
+      url={[https://huggingface.co/stabilityai/stablecode-instruct-alpha-3b](https://huggingface.co/stabilityai/stablecode-instruct-alpha-3b)}, 
+      title={Stable Code Instruct Alpha}, 
+      author={Adithyan, Reshinth and Phung, Duy and Cooper, Nathan and Pinnaparaju, Nikhil and Laforte, Christian}
+}
+```