Update README.md

README.md

@@ -1,17 +1,37 @@
 ---
 license: agpl-3.0
 language:
-- en
+- zh
 tags:
-- moss
 - llm
 - reward model
+- moss
+- rlhf
 ---
 
 # MOSS-RLHF
 
-### *MOSS-RLHF & "Secrets of RLHF in Large Language Models Part I: PPO" <br>👉 <a href="https://arxiv.org/abs/2307.04964" target="_blank">[Technical report]</a>  <a href="https://github.com/OpenLMLab/MOSS-RLHF" target="_blank">[Open-source code]*
+### *MOSS-RLHF & "Secrets of RLHF in Large Language Models Part I: PPO" <br>👉 <a href="https://arxiv.org/abs/2307.04964" target="_blank">[Technical report]</a>  <a href="https://openlmlab.github.io/MOSS-RLHF/" target="_blank">[Home page]*
+
+
+## 🌟 News
+### 👉 Wed, 12. July 2023. We have released Chinese reward model based OpenChineseLlama-7B! 
+[moss-rlhf-reward-model-7B-zh](https://huggingface.co/Ablustrund/moss-rlhf-reward-model-7B-zh/tree/main)
+<br>
 
+### 👉 Thu, 13. July 2023. We have released English reward model and SFT model based Llama-7B! 
+[moss-rlhf-reward-model-7B-en](https://huggingface.co/fnlp/moss-rlhf-reward-model-7B-en)
+
+[moss-rlhf-sft-model-7B-en](https://huggingface.co/fnlp/moss-rlhf-sft-model-7B-en)
+<br>
+
+## 🧾 Open-source List
+- [x] Open source code for RL training in large language models.
+- [x] A 7B Chinese reward model based on openChineseLlama.
+- [x] A 7B English reward model based on Llama-7B.
+- [x] SFT model for English.
+- [ ] Policy model for English after RLHF.
+- ...
 
 ## 🌠 Introduction
 
@@ -24,34 +44,24 @@ Contributions are summarized as follows:
 3) We release the complete PPO-max codes to ensure that the LLMs in the current SFT stage can be better aligned with humans.
 
 
-## 🧾 Open-source List
-- A 7B Chinese reward model based on openChineseLlama.
-- A 7B English reward model based on Llama-7B.
-- Open source code for RL training in large language models.
-- ...
-
-## ✨ Start training your own model!
-
-Run code in a few steps.
-
-### 🔩 Requirements & Setup
+## 🔩 Requirements & Setup
 
 This repository works on Python 3.8 and PyTorch 1.13.1.
 
 We recommend using the **conda** virtual environment to run the code.
 
-#### Step 1: create a new Python virtual environment
+#### Step 1: Create a new Python virtual environment
 ```bash
 conda update conda -n base -c defaults
 conda create -n rlhf python=3.8
 conda activate rlhf
 ```
-#### Step 2: install PyTorch and TensorBoard
+#### Step 2: Install PyTorch and TensorBoard
 ```bash
 conda install pytorch==1.13.1 pytorch-cuda=11.7 tensorboard -c pytorch -c nvidia
 ```
 
-#### Step 3: install the remaining dependencies
+#### Step 3: Install the remaining dependencies
 ```bash
 conda install datasets accelerate safetensors chardet cchardet -c huggingface -c conda-forge
 pip3 install transformers sentencepiece einops triton==1.0.0 rouge jionlp==1.4.14 nltk sacrebleu cpm_kernels
@@ -60,16 +70,57 @@ apt install libaio-dev
 DS_BUILD_OPS=1 pip install deepspeed
 ```
 
-### 👉 Start Training
+## ✨ Start training your own model!
+Run code in a few steps.
+
+### Step 1: Recover Reward model weights
+We can not directly release the full weight of the reward model because of protocol restrictions.
+You can merge the diff weight with original Llama-7B to recover the reward model we used.
+
+We upload the diff models, thanks to tatsu-lab, you can recover the reward model follow these steps:
+```bash
+1) Download the weight diff into your local machine. The weight diff is located at:
+# For English:
+TODO
+# For Chinese:
+https://huggingface.co/Ablustrund/moss-rlhf-reward-model-7B-zh/tree/main
+
+2) Merge the weight diff with the original Llama-7B:
+# For English:
+# Reward model
+python merge_weight_en.py recover --path_raw decapoda-research/llama-7b-hf --path_diff ./models/moss-rlhf-reward-model-7B-en/diff --path_tuned ./models/moss-rlhf-reward-model-7B-en/recover --model_type reward
+# SFT model
+python merge_weight_en.py recover --path_raw decapoda-research/llama-7b-hf --path_diff ./models/moss-rlhf-sft-model-7B-en/diff --path_tuned ./models/moss-rlhf-sft-model-7B-en/recover --model_type sft
+# Policy model
+TODO
+# For Chinese:
+python merge_weight_zh.py recover --path_raw decapoda-research/llama-7b-hf --path_diff ./models/moss-rlhf-reward-model-7B-zh/diff --path_tuned ./models/moss-rlhf-reward-model-7B-zh/recover
+```
+### Step 2: Select your own SFT model.
+Because of some limitations, we can not release the **Chinese** SFT model (Currently).
+You can use your own SFT model, or a strong base model instead of our SFT model.
+
+### Step 3: Start training
+Run the command below.
+```
+# For Chinese:
+# You need to use your own sft model currently.
+bash run_zh.sh
 
-TODO, To be finalised before 15. July 2023
+# For English:
+# We have loaded the sft model and reward model to huggingface.
+bash run_en.sh
+```
 
 ## Citation
 
 ```bibtex
 @article{zheng2023secrets,
-  title={Secrets of RLHF in Large Language Models Part I: PPO}, 
-  author={Rui Zheng and Shihan Dou and Songyang Gao and Yuan Hua and Wei Shen and Binghai Wang and Yan Liu and Senjie Jin and Qin Liu and Yuhao Zhou and Limao Xiong and Lu Chen and Zhiheng Xi and Nuo Xu and Wenbin Lai and Minghao Zhu and Cheng Chang and Zhangyue Yin and Rongxiang Weng and Wensen Cheng and Haoran Huang and Tianxiang Sun and Hang Yan and Tao Gui and Qi Zhang and Xipeng Qiu and Xuanjing Huang},
-  year={2023}
+      title={Secrets of RLHF in Large Language Models Part I: PPO}, 
+      author={Rui Zheng and Shihan Dou and Songyang Gao and Wei Shen and Binghai Wang and Yan Liu and Senjie Jin and Qin Liu and Limao Xiong and Lu Chen and Zhiheng Xi and Yuhao Zhou and Nuo Xu and Wenbin Lai and Minghao Zhu and Rongxiang Weng and Wensen Cheng and Cheng Chang and Zhangyue Yin and Yuan Hua and Haoran Huang and Tianxiang Sun and Hang Yan and Tao Gui and Qi Zhang and Xipeng Qiu and Xuanjing Huang},
+      year={2023},
+      eprint={2307.04964},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
 }
-```
+```