metadata

base_model: XueyingJia/pythia-1b-sft-HH-3-merge
datasets: XueyingJia/online_dpo_repo
library_name: transformers
model_name: pythia-1b-online-dpo-HH-merge-rewardmodel
tags:
  - generated_from_trainer
  - trl
  - online-dpo
licence: license

Model Card for pythia-1b-online-dpo-HH-merge-rewardmodel

This model is a fine-tuned version of XueyingJia/pythia-1b-sft-HH-3-merge on the XueyingJia/online_dpo_repo dataset. It has been trained using TRL.

Quick start

from transformers import pipeline

question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="XueyingJia/pythia-1b-online-dpo-HH-merge-rewardmodel", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])

Training procedure

This model was trained with Online DPO, a method introduced in Direct Language Model Alignment from Online AI Feedback.

Framework versions

TRL: 0.13.0.dev0
Transformers: 4.46.3
Pytorch: 2.5.1
Datasets: 3.1.0
Tokenizers: 0.20.3

Citations

Cite Online DPO as:

@article{guo2024direct,
    title        = {{Direct Language Model Alignment from Online AI Feedback}},
    author       = {Shangmin Guo and Biao Zhang and Tianlin Liu and Tianqi Liu and Misha Khalman and Felipe Llinares and Alexandre Ram{'{e}} and Thomas Mesnard and Yao Zhao and Bilal Piot and Johan Ferret and Mathieu Blondel},
    year         = 2024,
    eprint       = {arXiv:2402.04792}
}

Cite TRL as:

@misc{vonwerra2022trl,
    title        = {{TRL: Transformer Reinforcement Learning}},
    author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
    year         = 2020,
    journal      = {GitHub repository},
    publisher    = {GitHub},
    howpublished = {\url{https://github.com/huggingface/trl}}
}