HuggingFaceTB
/

SmolLM2-360M-Instruct

@@ -1,87 +1,124 @@
 ---
-base_model: loubnabnl/smollm2-360M-8k-lc100k-mix1-ep2
-tags:
-- alignment-handbook
-- trl
-- dpo
-- generated_from_trainer
-- trl
-- dpo
-- generated_from_trainer
-datasets:
-- HuggingFaceH4/ultrafeedback_binarized
-model-index:
-- name: smollm2-360M-8k-lc100k-dpo-ultaf-ep2
-  results: []
 ---
-<!-- 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. -->
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/loubnabnl/huggingface/runs/3nedrmyg)
-# smollm2-360M-8k-lc100k-dpo-ultaf-ep2
-This model is a fine-tuned version of [loubnabnl/smollm2-360M-8k-lc100k-mix1-ep2](https://huggingface.co/loubnabnl/smollm2-360M-8k-lc100k-mix1-ep2) on the HuggingFaceH4/ultrafeedback_binarized dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.6348
-- Rewards/chosen: -0.0342
-- Rewards/rejected: -0.3910
-- Rewards/accuracies: 0.6190
-- Rewards/margins: 0.3568
-- Logps/rejected: -323.7198
-- Logps/chosen: -375.6464
-- Logits/rejected: -1.6969
-- Logits/chosen: -1.6408
-## Model description
-More information needed
-## Intended uses & limitations
-More information needed
-## Training and evaluation data
-More information needed
-## Training procedure
-### Training hyperparameters
-The following hyperparameters were used during training:
-- learning_rate: 1e-06
-- train_batch_size: 2
-- eval_batch_size: 4
-- seed: 42
-- distributed_type: multi-GPU
-- num_devices: 8
-- gradient_accumulation_steps: 8
-- total_train_batch_size: 128
-- total_eval_batch_size: 32
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: cosine
-- lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 2
-### Training results
-| Training Loss | Epoch  | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
-|:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
-| 0.7098        | 0.2094 | 100  | 0.7162          | -0.0109        | -0.0675          | 0.5278             | 0.0566          | -323.0727      | -375.5997    | -1.6983         | -1.6387       |
-| 0.6825        | 0.4187 | 200  | 0.6842          | -0.0010        | -0.1880          | 0.5794             | 0.1870          | -323.3139      | -375.5800    | -1.6938         | -1.6358       |
-| 0.663         | 0.6281 | 300  | 0.6617          | 0.0225         | -0.2389          | 0.6032             | 0.2614          | -323.4156      | -375.5330    | -1.6893         | -1.6317       |
-| 0.6547        | 0.8375 | 400  | 0.6591          | 0.0001         | -0.3516          | 0.6389             | 0.3517          | -323.6410      | -375.5778    | -1.6980         | -1.6414       |
-| 0.6456        | 1.0468 | 500  | 0.6430          | 0.0133         | -0.3566          | 0.6667             | 0.3699          | -323.6510      | -375.5514    | -1.6931         | -1.6365       |
-| 0.6054        | 1.2562 | 600  | 0.6423          | -0.0329        | -0.3895          | 0.6349             | 0.3566          | -323.7167      | -375.6438    | -1.6991         | -1.6431       |
-| 0.6129        | 1.4656 | 700  | 0.6431          | -0.0449        | -0.4183          | 0.6349             | 0.3735          | -323.7745      | -375.6677    | -1.6979         | -1.6414       |
-| 0.5972        | 1.6750 | 800  | 0.6384          | -0.0695        | -0.4139          | 0.6429             | 0.3444          | -323.7656      | -375.7169    | -1.6965         | -1.6399       |
-| 0.6207        | 1.8843 | 900  | 0.6362          | -0.0627        | -0.4222          | 0.6786             | 0.3595          | -323.7822      | -375.7033    | -1.6976         | -1.6407       |
-### Framework versions
-- Transformers 4.42.3
-- Pytorch 2.1.2
-- Datasets 2.20.0
-- Tokenizers 0.19.1

 ---
+library_name: transformers
+license: apache-2.0
+language:
+- en
 ---
+# SmolLM2
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/61c141342aac764ce1654e43/7IzejwZJ62MfRwvDYvQXY.png)
+##  Table of Contents
+1. [Model Summary](##model-summary)
+2. [Limitations](##limitations)
+3. [Training](##training)
+4. [License](##license)
+5. [Citation](##citation)
+## Model Summary
+SmolLM2 is a family of compact language models available in three size: 135M, 360M, and 1.7B parameters. They are capable of solving a wide range of tasks while being lightweight enough to run on-device.
+SmolLM2 demonstrates significant advances over its predecessor SmolLM1, particularly in instruction following, knowledge, reasoning. The 135M model was trained on 2 trillion tokens using a diverse dataset combination: FineWeb-Edu, DCLM, The Stack, along with new filtered datasets we curated and will release soon.
+We developed the instruct version through supervised fine-tuning (SFT) using a combination of public datasets and our own curated datasets, designed to enhance instruction following, rewriting, and summarization capabilities. We then applied Direct Preference Optimization (DPO) using a mix of UltraFeedback and DPO-ORPO.
+### How to use
+### Transformers
+```bash
+pip install transformers
+```
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+checkpoint = "HuggingFaceTB/SmolLM2-360M-Instruct"
+device = "cuda" # for GPU usage or "cpu" for CPU usage
+tokenizer = AutoTokenizer.from_pretrained(checkpoint)
+# for multiple GPUs install accelerate and do `model = AutoModelForCausalLM.from_pretrained(checkpoint, device_map="auto")`
+model = AutoModelForCausalLM.from_pretrained(checkpoint).to(device)
+messages = [{"role": "user", "content": "What is the capital of France."}]
+input_text=tokenizer.apply_chat_template(messages, tokenize=False)
+print(input_text)
+inputs = tokenizer.encode(input_text, return_tensors="pt").to(device)
+outputs = model.generate(inputs, max_new_tokens=50, temperature=0.2, top_p=0.9, do_sample=True)
+print(tokenizer.decode(outputs[0]))
+```
+### Chat in TRL
+You can also use the TRL CLI to chat with the model from the terminal:
+```bash
+pip install trl
+trl chat --model_name_or_path HuggingFaceTB/SmolLM2-360M-Instruct --device cpu
+```
+## Evaluation
+In this section, we report the evaluation results of SmolLM2. All evaluations are zero-shot unless stated otherwise, and we use [lighteval](https://github.com/huggingface/lighteval) to run them.
+## Base Pre-Trained Model
+| Metrics            | SmolLM2-360M | Qwen2.5-0.5B | SmolLM-360M  |
+|:-------------------|:------------:|:------------:|:------------:|
+| HellaSwag          | **54.5**     | 51.2         | 51.8         |
+| ARC (Average)      | **53.0**     | 45.4         | 50.1         |
+| PIQA               | **71.7**     | 69.9         | 71.6         |
+| MMLU (cloze)       | **35.8**     | 33.7         | 34.4         |
+| CommonsenseQA      | **38.0**     | 31.6         | 35.3         |
+| TriviaQA           | **16.9**     | 4.3          | 9.1          |
+| Winogrande         | 52.5         | **54.1**     | 52.8         |
+| OpenBookQA         | **37.4**     | **37.4**     | 37.2         |
+| GSM8K (5-shot)     | 3.2          | **33.4**     | 1.6          |
+## Instruction Model
+| Metric                       | SmolLM2-360M-Instruct | Qwen2.5-0.5B-Instruct | SmolLM-360M-Instruct |
+|:-----------------------------|:---------------------:|:---------------------:|:---------------------:|
+| IFEval (Average prompt/inst) | **41.0**             | 31.6                 | 19.8                 |
+| MT-Bench                     | 3.66                 | **4.16**             | 3.37                 |
+| HellaSwag                    | **52.1**             | 48.0                 | 47.9                 |
+| ARC (Average)                | **43.7**             | 37.3                 | 38.8                 |
+| PIQA                         | **70.8**             | 67.2                 | 69.4                 |
+| MMLU (cloze)                 | **32.8**             | 31.7                 | 30.6                 |
+| BBH (3-shot)                 | 27.3                 | **30.7**             | 24.4                 |
+| GSM8K (5-shot)               | 7.43                 | **26.8**             | 1.36                 |
+## Limitations
+SmolLM2 models primarily understand and generate content in English. They can produce text on a variety of topics, but the generated content may not always be factually accurate, logically consistent, or free from biases present in the training data. These models should be used as assistive tools rather than definitive sources of information. Users should always verify important information and critically evaluate any generated content.
+## Training
+### Model
+- **Architecture:** Transformer decoder
+- **Pretraining tokens:** 4T
+- **Precision:** bfloat16
+### Hardware
+- **GPUs:** 64 H100
+### Software
+- **Training Framework:** [nanotron](https://github.com/huggingface/nanotron/tree/main)
+## License
+[Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)
+## Citation
+```bash
+@misc{allal2024SmolLM2,
+      title={SmolLM2 - with great data, comes great performance},
+      author={Loubna Ben Allal and Anton Lozhkov and Elie Bakouch and Gabriel Martín Blázquez and Lewis Tunstall and Agustín Piqueres and Andres Marafioti and Cyril Zakka and Leandro von Werra and Thomas Wolf},
+      year={2024},
+}
+```