---
tags:
- trl
- dpo
- generated_from_trainer
model-index:
- name: zephyr-7b-dpo-full
  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-dpo-full

This model was trained from scratch on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0594
- Rewards/chosen: -5.1568
- Rewards/rejected: -5.4496
- Rewards/accuracies: 0.5625
- Rewards/margins: 0.2928
- Logps/rejected: -917.5723
- Logps/chosen: -895.0606
- Logits/rejected: -4.8109
- Logits/chosen: -4.6185

## 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: 2
- total_train_batch_size: 128
- 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

| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
| 0.5292        | 0.26  | 100  | 0.6958          | -1.2149        | -1.3146          | 0.5312             | 0.0997          | -504.0709      | -500.8688    | -3.0075         | -2.9813       |
| 0.3733        | 0.53  | 200  | 0.9588          | -4.2477        | -4.5339          | 0.5273             | 0.2862          | -826.0007      | -804.1498    | -4.5463         | -4.4070       |
| 0.3689        | 0.79  | 300  | 1.0594          | -5.1568        | -5.4496          | 0.5625             | 0.2928          | -917.5723      | -895.0606    | -4.8109         | -4.6185       |


### Framework versions

- Transformers 4.38.2
- Pytorch 2.1.2+cu118
- Datasets 2.16.1
- Tokenizers 0.15.2