15 lines
398 B
Markdown
15 lines
398 B
Markdown
---
|
|
license: mit
|
|
---
|
|
|
|
|
|
* [DPO Trainer](https://huggingface.co/docs/trl/main/en/dpo_trainer) with jondurbin/truthy-dpo-v0.1
|
|
```
|
|
DPO Trainer
|
|
TRL supports the DPO Trainer for training language models from preference data, as described in the paper Direct Preference Optimization: Your Language Model is Secretly a Reward Model by Rafailov et al., 2023.
|
|
```
|
|
|
|
```
|
|
"num_experts_per_tok": 4
|
|
|
|
``` |