<img src="assets/logo.png" alt="Stanford-Alpaca" style="width: 50%; min-width: 300px; display: block; margin: auto;">
Sign in to like and favorite skills
This is the repo for the Stanford Alpaca project, which aims to build and share an instruction-following LLaMA model. The repo contains:
Note: We thank the community for feedback on Stanford-Alpaca and supporting our research. Our live demo is suspended until further notice.
Usage and License Notices: Alpaca is intended and licensed for research use only. The dataset is CC BY NC 4.0 (allowing only non-commercial use) and models trained using the dataset should not be used outside of research purposes. The weight diff is also CC BY NC 4.0 (allowing only non-commercial use).
The current Alpaca model is fine-tuned from a 7B LLaMA model [1] on 52K instruction-following data generated by the techniques in the Self-Instruct [2] paper, with some modifications that we discuss in the next section. In a preliminary human evaluation, we found that the Alpaca 7B model behaves similarly to the
text-davinci-003Alpaca is still under development, and there are many limitations that have to be addressed. Importantly, we have not yet fine-tuned the Alpaca model to be safe and harmless. We thus encourage users to be cautious when interacting with Alpaca, and to report any concerning behavior to help improve the safety and ethical considerations of the model.
Our initial release contains the data generation procedure, dataset, and training recipe. We intend to release the model weights if we are given permission to do so by the creators of LLaMA. For now, we have chosen to host a live demo to help readers better understand the capabilities and limits of Alpaca, as well as a way to help us better evaluate Alpaca's performance on a broader audience.
Please read our release blog post for more details about the model, our discussion of the potential harm and limitations of Alpaca models, and our thought process for releasing a reproducible model.
[1]: LLaMA: Open and Efficient Foundation Language Models. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, Aurelien Rodriguez, Armand Joulin, Edouard Grave, Guillaume Lample. https://arxiv.org/abs/2302.13971v1
[2]: Self-Instruct: Aligning Language Model with Self Generated Instructions. Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A. Smith, Daniel Khashabi, Hannaneh Hajishirzi. https://arxiv.org/abs/2212.10560
contains 52K instruction-following data we used for fine-tuning the Alpaca model.
This JSON file is a list of dictionaries, each dictionary contains the following fields:alpaca_data.json
instructionstrinputstroutputstrtext-davinci-003We used the following prompts for fine-tuning the Alpaca model:
Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request. ### Instruction: {instruction} ### Input: {input} ### Response:
Below is an instruction that describes a task. Write a response that appropriately completes the request. ### Instruction: {instruction} ### Response:
During inference (eg for the web demo), we use the user instruction with an empty input field (second option).
OPENAI_API_KEYpip install -r requirements.txtpython -m generate_instruction generate_instruction_following_dataWe built on the data generation pipeline from self-instruct and made the following modifications:
text-davinci-003davinciprompt.txttext-davinci-003This produced an instruction-following dataset with 52K examples obtained at a much lower cost (less than $500). In a preliminary study, we also find our 52K generated data to be much more diverse than the data released by self-instruct. We plot the below figure (in the style of Figure 2 in the self-instruct paper to demonstrate the diversity of our data. The inner circle of the plot represents the root verb of the instructions, and the outer circle represents the direct objects.
We fine-tune our models using standard Hugging Face training code. We fine-tune LLaMA-7B and LLaMA-13B with the following hyperparameters:
| Hyperparameter | LLaMA-7B | LLaMA-13B |
|---|---|---|
| Batch size | 128 | 128 |
| Learning rate | 2e-5 | 1e-5 |
| Epochs | 3 | 5 |
| Max length | 512 | 512 |
| Weight decay | 0 | 0 |
To reproduce our fine-tuning runs for LLaMA, first install the requirements
pip install -r requirements.txt
Below is a command that fine-tunes LLaMA-7B with our dataset on a machine with 4 A100 80G GPUs in FSDP
full_shard<your_random_port><your_path_to_hf_converted_llama_ckpt_and_tokenizer><your_output_dir>torchrun --nproc_per_node=4 --master_port=<your_random_port> train.py \ --model_name_or_path <your_path_to_hf_converted_llama_ckpt_and_tokenizer> \ --data_path ./alpaca_data.json \ --bf16 True \ --output_dir <your_output_dir> \ --num_train_epochs 3 \ --per_device_train_batch_size 4 \ --per_device_eval_batch_size 4 \ --gradient_accumulation_steps 8 \ --evaluation_strategy "no" \ --save_strategy "steps" \ --save_steps 2000 \ --save_total_limit 1 \ --learning_rate 2e-5 \ --weight_decay 0. \ --warmup_ratio 0.03 \ --lr_scheduler_type "cosine" \ --logging_steps 1 \ --fsdp "full_shard auto_wrap" \ --fsdp_transformer_layer_cls_to_wrap 'LlamaDecoderLayer' \ --tf32 True
The same script also works for OPT fine-tuning. Here's an example for fine-tuning OPT-6.7B
torchrun --nproc_per_node=4 --master_port=<your_random_port> train.py \ --model_name_or_path "facebook/opt-6.7b" \ --data_path ./alpaca_data.json \ --bf16 True \ --output_dir <your_output_dir> \ --num_train_epochs 3 \ --per_device_train_batch_size 4 \ --per_device_eval_batch_size 4 \ --gradient_accumulation_steps 8 \ --evaluation_strategy "no" \ --save_strategy "steps" \ --save_steps 2000 \ --save_total_limit 1 \ --learning_rate 2e-5 \ --weight_decay 0. \ --warmup_ratio 0.03 \ --lr_scheduler_type "cosine" \ --logging_steps 1 \ --fsdp "full_shard auto_wrap" \ --fsdp_transformer_layer_cls_to_wrap 'OPTDecoderLayer' \ --tf32 True
Note the given training script is meant to be simple and easy to use, and is not particularly optimized. To run on more gpus, you may prefer to turn down
gradient_accumulation_stepsNaively, fine-tuning a 7B model requires about 7 x 4 x 4 = 112 GB of VRAM. Commands given above enable parameter sharding, so no redundant model copy is stored on any GPU. If you'd like to further reduce the memory footprint, here are some options:
--fsdp "full_shard auto_wrap offload"pip install deepspeed torchrun --nproc_per_node=4 --master_port=<your_random_port> train.py \ --model_name_or_path <your_path_to_hf_converted_llama_ckpt_and_tokenizer> \ --data_path ./alpaca_data.json \ --bf16 True \ --output_dir <your_output_dir> \ --num_train_epochs 3 \ --per_device_train_batch_size 4 \ --per_device_eval_batch_size 4 \ --gradient_accumulation_steps 8 \ --evaluation_strategy "no" \ --save_strategy "steps" \ --save_steps 2000 \ --save_total_limit 1 \ --learning_rate 2e-5 \ --weight_decay 0. \ --warmup_ratio 0.03 \ --deepspeed "./configs/default_offload_opt_param.json" \ --tf32 True
The weight diff between Alpaca-7B and LLaMA-7B is located here. To recover the original Alpaca-7B weights, follow these steps:
1. Convert Meta's released weights into huggingface format. Follow this guide: https://huggingface.co/docs/transformers/main/model_doc/llama 2. Make sure you cloned the released weight diff into your local machine. The weight diff is located at: https://huggingface.co/tatsu-lab/alpaca-7b/tree/main 3. Run this function with the correct paths. E.g., python weight_diff.py recover --path_raw <path_to_step_1_dir> --path_diff <path_to_step_2_dir> --path_tuned <path_to_store_recovered_weights>
Once step 3 completes, you should have a directory with the recovered weights, from which you can load the model like the following
import transformers alpaca_model = transformers.AutoModelForCausalLM.from_pretrained("<path_to_store_recovered_weights>") alpaca_tokenizer = transformers.AutoTokenizer.from_pretrained("<path_to_store_recovered_weights>")
All grad students below contributed equally and the order is determined by random draw.
All advised by Tatsunori B. Hashimoto. Yann is also advised by Percy Liang and Xuechen is also advised by Carlos Guestrin.
Please cite the repo if you use the data or code in this repo.
@misc{alpaca, author = {Rohan Taori and Ishaan Gulrajani and Tianyi Zhang and Yann Dubois and Xuechen Li and Carlos Guestrin and Percy Liang and Tatsunori B. Hashimoto }, title = {Stanford Alpaca: An Instruction-following LLaMA model}, year = {2023}, publisher = {GitHub}, journal = {GitHub repository}, howpublished = {\url{https://github.com/tatsu-lab/stanford_alpaca}}, }
Naturally, you should also cite the original LLaMA paper [1] and the Self-Instruct paper [2].
We thank Yizhong Wang for his help in explaining the data generation pipeline in Self-Instruct and providing the code for the parse analysis plot. We thank Yifan Mai for helpful support, and members of the Stanford NLP Group as well as the Center for Research on Foundation Models (CRFM) for their helpful feedback.