|
--- |
|
license: apache-2.0 |
|
datasets: |
|
- openbmb/UltraFeedback |
|
language: |
|
- en |
|
pipeline_tag: text-generation |
|
--- |
|
Self-Play Preference Optimization for Language Model Alignment (https://arxiv.org/abs/2405.00675) |
|
|
|
# Mistral7B-PairRM-SPPO |
|
|
|
This model was developed using [Self-Play Preference Optimization](https://arxiv.org/abs/2405.00675) at iteration 3, based on the [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) architecture as starting point. We utilized the prompt sets from the [openbmb/UltraFeedback](https://huggingface.co/datasets/openbmb/UltraFeedback) dataset, splited to 3 parts for 3 iterations by [snorkelai/Snorkel-Mistral-PairRM-DPO-Dataset](https://huggingface.co/datasets/snorkelai/Snorkel-Mistral-PairRM-DPO-Dataset). All responses used are synthetic. |
|
|
|
While K = 5 (generate 5 samples per iteration), this model uses 3 samples to estimate the soft probabilities P(y_w > y_l) and P(y_l > y_w). These samples include the winner, the loser, and another random sample. This approach has shown to deliver better performance on AlpacaEval 2.0 than the results reported in [the paper](https://arxiv.org/abs/2405.00675), but it might also lead to overfitting the PairRM core. |
|
|
|
❗Please refer to the original checkpoint at [**UCLA-AGI/Mistral7B-PairRM-SPPO-Iter3**](https://huggingface.co/UCLA-AGI/Mistral7B-PairRM-SPPO-Iter3) as **reported in our paper**. We anticipate that the version in paper demonstrates a more consistent performance improvement across all benchmark tasks. |
|
|
|
|
|
## Links to Other Models |
|
- [Mistral7B-PairRM-SPPO-Iter1](https://huggingface.co/UCLA-AGI/Mistral7B-PairRM-SPPO-Iter1) |
|
- [Mistral7B-PairRM-SPPO-Iter2](https://huggingface.co/UCLA-AGI/Mistral7B-PairRM-SPPO-Iter2) |
|
- [Mistral7B-PairRM-SPPO-Iter3](https://huggingface.co/UCLA-AGI/Mistral7B-PairRM-SPPO-Iter3) |
|
- [Mistral7B-PairRM-SPPO](https://huggingface.co/UCLA-AGI/Mistral7B-PairRM-SPPO) |
|
|
|
|
|
### Model Description |
|
|
|
- Model type: A 7B parameter GPT-like model fine-tuned on synthetic datasets. |
|
- Language(s) (NLP): Primarily English |
|
- License: Apache-2.0 |
|
- Finetuned from model: mistralai/Mistral-7B-Instruct-v0.2 |
|
|
|
|
|
## [AlpacaEval Leaderboard Evaluation Results](https://tatsu-lab.github.io/alpaca_eval/) |
|
Model | LC. Win Rate | Win Rate | Avg. Length | |
|
|------|-----------------------|---------------------------|------------| |
|
Mistral7B-PairRM-SPPO| 30.46 | 32.14 | 2114 | |
|
Mistral7B-PairRM-SPPO (best-of-16)| 32.90 | 34.67 | 2112 | |
|
|
|
|
|
### Training hyperparameters |
|
The following hyperparameters were used during training: |
|
|
|
- learning_rate: 5e-07 |
|
- eta: 1000 |
|
- per_device_train_batch_size: 8 |
|
- gradient_accumulation_steps: 1 |
|
- seed: 42 |
|
- distributed_type: deepspeed_zero3 |
|
- num_devices: 8 |
|
- optimizer: RMSProp |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_train_epochs: 18.0 (stop at epoch=1.0) |
|
|
|
|
|
|
|
|
|
## Citation |
|
``` |
|
@misc{wu2024self, |
|
title={Self-Play Preference Optimization for Language Model Alignment}, |
|
author={Wu, Yue and Sun, Zhiqing and Yuan, Huizhuo and Ji, Kaixuan and Yang, Yiming and Gu, Quanquan}, |
|
year={2024}, |
|
eprint={2405.00675}, |
|
archivePrefix={arXiv}, |
|
primaryClass={cs.LG} |
|
} |
|
``` |
