---
library_name: transformers
license: apache-2.0
base_model: alignment-handbook/zephyr-7b-sft-full
tags:
- alignment-handbook
- trl
- dpo
- generated_from_trainer
- trl
- dpo
- generated_from_trainer
datasets:
- HuggingFaceH4/hh-rlhf-h4
model-index:
- name: zephyr-7b-hh-dpo
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# zephyr-7b-hh-dpo

This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the HuggingFaceH4/hh-rlhf-h4 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5453
- Rewards/chosen: -2.4749
- Rewards/rejected: -3.2088
- Rewards/accuracies: 0.7174
- Rewards/margins: 0.7340
- Logps/rejected: -474.4292
- Logps/chosen: -394.6965
- Logits/rejected: 3.9043
- Logits/chosen: 3.4245

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-07
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- total_eval_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1

### Training results



### Framework versions

- Transformers 4.44.1
- Pytorch 2.1.2+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1