---
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/ultrafeedback_binarized
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 is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the HuggingFaceH4/ultrafeedback_binarized dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2784
- Rewards/chosen: 31.9054
- Rewards/rejected: 27.5130
- Rewards/accuracies: 0.5742
- Rewards/margins: 4.3924
- Logps/rejected: -235.1542
- Logps/chosen: -230.6881
- Logits/rejected: -2.6646
- Logits/chosen: -2.6896

## 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 |
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
| 0.3163        | 0.21  | 100  | 0.3129          | 27.2147        | 24.2237          | 0.5820             | 2.9910          | -238.4435      | -235.3788    | -2.6701         | -2.6969       |
| 0.2918        | 0.42  | 200  | 0.2923          | 30.3522        | 26.4568          | 0.5820             | 3.8954          | -236.2104      | -232.2412    | -2.6826         | -2.7085       |
| 0.286         | 0.63  | 300  | 0.2921          | 31.4394        | 27.2385          | 0.5820             | 4.2009          | -235.4286      | -231.1540    | -2.6717         | -2.6957       |
| 0.2819        | 0.84  | 400  | 0.2788          | 31.8547        | 27.4761          | 0.5781             | 4.3786          | -235.1911      | -230.7387    | -2.6619         | -2.6869       |


### Framework versions

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