File size: 2,429 Bytes
4eb4ea8 646b0cf 4eb4ea8 646b0cf 4eb4ea8 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 |
---
base_model: trl-lib/pythia-1b-deduped-tldr-sft
datasets: trl-lib/tldr
library_name: transformers
model_name: pythia-1b-tldr-online-dpo-mark-peft
tags:
- generated_from_trainer
- trl
- online-dpo
licence: license
---
# Model Card for pythia-1b-tldr-online-dpo-mark-peft
This model is a fine-tuned version of [trl-lib/pythia-1b-deduped-tldr-sft](https://huggingface.co/trl-lib/pythia-1b-deduped-tldr-sft) on the [trl-lib/tldr](https://huggingface.co/datasets/trl-lib/tldr) dataset.
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
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="ikimhope/pythia-1b-tldr-online-dpo-mark-peft", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/ikimhope-el/huggingface/runs/46biac5f)
This model was trained with Online DPO, a method introduced in [Direct Language Model Alignment from Online AI Feedback](https://huggingface.co/papers/2402.04792).
### Framework versions
- TRL: 0.12.0.dev0
- Transformers: 4.45.1
- Pytorch: 2.4.1
- Datasets: 3.0.1
- Tokenizers: 0.20.0
## Citations
Cite Online DPO as:
```bibtex
@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:
```bibtex
@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}}
}
``` |