[📄 Paper] | [🤗 Lion Weights] | [:desktop_computer: Demo]
- [May 26, 2023] We released the model weights. Check out the 7B model!
- [May 25, 2023] We released an online demo, try our model here!
- [May 23, 2023] We released the code for training and inference.
The high-level overview of our adversarial distillation framework, where we craft a compact Student LLM based on a superior closed-source LLM that serves three roles: the Teacher, the Referee, and the Generator. From left to right, there are three stages in an iteration:
- an imitation stage to align the student’s response with the teacher’s response;
- a discrimination stage to identify hard samples;
- a generation stage to produce new hard samples for escalating the challenges presented to the student model.
We will provide our latest models for you to try for as long as possible. You may ask some questions to Lion and we are happy to hear your feedback!
Demo Link (the UI interface is shown below)
Since the training data are English instruction-following examples, You'd better ask questions in English. However, we found Lion can also understand instructions in other languages to some extent. See the following case:
We release Lion weights as delta weights to comply with the LLaMA model license.
You can add our delta to the original LLaMA weights to obtain the Lion weights. Instructions:
- Get the original LLaMA weights in the huggingface format by following the instructions here
- Please download our delta model from Hugging Face
- Use the following scripts to get Lion weights by applying our delta:
python src/weight_diff.py recover --path_raw huggyllama/llama-7b --path_diff YuxinJiang/Lion --path_tuned <path_to_store_recovered_weights>
For inference and training of Lion, please first install the requirements:
pip install -r requirements.txt
We provide the decoding script for Lion, which reads a input file and generates corresponding responses for each sample, and finally consolidates them into an output file. It can be run on a single machine with 16GB GPU.
python src/lion_inference.py \
--model_dir <path_to_hf_converted_lion_ckpt_and_tokenizer> \
--data_dir <path_to_input_json_file> \
--output_dir <path_to_output_json_file> \
--num_gpus 1
Below shows one iteration of our adversarial distillation framework.
python src/chatgpt_inference.py \
-q <path_to_json_file_for_the_Train_Pool> \
-o <path_to_chatgpt_inference_for_the_Train_Pool> \
--api_key <your_openai_api_key>
Fine-tuning was conducted on on a machine with 8 A100 80G GPUs.
torchrun --nproc_per_node=8 --master_port=<your_random_port> src/train.py \
--model_name_or_path <path_to_hf_converted_ckpt_and_tokenizer> \
--data_path <path_to_chatgpt_inference_for_the_Train_Pool> \
--bf16 True \
--output_dir result \
--num_train_epochs 3 \
--model_max_length 1024 \
--per_device_train_batch_size 2 \
--per_device_eval_batch_size 2 \
--gradient_accumulation_steps 8 \
--evaluation_strategy "no" \
--save_strategy "steps" \
--save_steps 500 \
--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
python src/chatgpt_inference.py \
-q <path_to_json_file_for_the_Cache_Pool> \
-o <path_to_chatgpt_inference_for_the_Cache_Pool> \
--api_key <your_openai_api_key>
python src/lion_inference.py \
--model_dir <path_to_hf_converted_lion_ckpt_and_tokenizer> \
--data_dir <path_to_json_file_for_the_Cache_Pool> \
--output_dir <path_to_lion_inference_for_the_Cache_Pool> \
--num_gpus 8
2.3 Ask the referee to output two scores according to the respose quality of the teacher and the student
python src/chatgpt_referee.py \
-a <path_to_chatgpt_inference_for_the_Cache_Pool> <path_to_lion_inference_for_the_Cache_Pool> \
-o <path_to_output_review_file> \
--api_key <your_openai_api_key>
python src/discrimination.py \
--review_path <path_to_output_review_file> \
--chatgpt_inference_path <path_to_chatgpt_inference_for_the_Cache_Pool> \
--lion_inference_path path_to_lion_inference_for_the_Cache_Pool \
--hard_save_path <path_to_identified_hard_instructions> \
--easy_save_path <path_to_identified_easy_instructions>
Fill the openai.api_key = "<you_openai_api_key>"
in src/utils.py.
python -m src/generate_hard_instruction generate_instruction_following_data \
--seed_tasks_path <path_to_identified_hard_instructions> \
--output_dir <path_to_generated_hard_instructions> \
--num_instructions_to_generate 3000
python -m src/generate_easy_instruction generate_instruction_following_data \
--seed_tasks_path <path_to_identified_easy_instructions> \
--output_dir <path_to_generated_easy_instructions> \
--num_instructions_to_generate 3000
we leverage GPT-4 to automatically rate the response quality (with scores from 1 to 10) between two models on 80 unseen Vicuna-Instructions. ChatGPT has been chosen as the reference model to estimate the relative capability of diverse LLMs against it. The relative score is reported in percentage, computed as the ratio of the sum of scores.
Relative Overall Response Quality:
Relative Response Quality of Diverse Task Categories:
We employ the alignment criteria proposed by Askell et al. (2021), which define that an assistant is considered aligned if it is characterized by being helpful, honest, and harmless (HHH). We performed a human evaluation on 252 UserOriented-Instructions. To estimate the won rate, we compare the frequency of won, tie, and lost between each pair of models below.
Please cite our paper if you use the code in this repo.
@article{DBLP:journals/corr/abs-2305-12870,
author = {Yuxin Jiang and
Chunkit Chan and
Mingyang Chen and
Wei Wang},
title = {Lion: Adversarial Distillation of Closed-Source Large Language Model},
journal = {CoRR},
volume = {abs/2305.12870},
year = {2023},
url = {https://doi.org/10.48550/arXiv.2305.12870},
doi = {10.48550/arXiv.2305.12870},
eprinttype = {arXiv},
eprint = {2305.12870},
timestamp = {Fri, 26 May 2023 11:29:33 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2305-12870.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}