---
library_name: transformers
base_model: W-61/llama-3-8b-base-sft-ultrachat-8xh200
tags:
- alignment-handbook
- r-dpo
- generated_from_trainer
datasets:
- HuggingFaceH4/ultrafeedback_binarized
model-index:
- name: llama-3-8b-base-r-dpo-ultrafeedback-4xh200-batch-128-20260426-105614
  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. -->

# llama-3-8b-base-r-dpo-ultrafeedback-4xh200-batch-128-20260426-105614

This model is a fine-tuned version of [W-61/llama-3-8b-base-sft-ultrachat-8xh200](https://huggingface.co/W-61/llama-3-8b-base-sft-ultrachat-8xh200) on the HuggingFaceH4/ultrafeedback_binarized dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5338
- R Dpo/chosen Len: 286.9760
- R Dpo/rejected Len: 246.0880
- R Dpo/length Delta: 40.8880
- R Dpo/regularization Term: 0.0
- Logps/chosen: -406.7590
- Logps/rejected: -443.3631
- Logps/ref Chosen: -288.6415
- Logps/ref Rejected: -265.9616
- Logits/chosen: -0.8653
- Logits/rejected: -0.8488

## 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: 4
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- total_eval_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1

### Training results

| Training Loss | Epoch  | Step | Validation Loss | R Dpo/chosen Len | R Dpo/rejected Len | R Dpo/length Delta | R Dpo/regularization Term | Logps/chosen | Logps/rejected | Logps/ref Chosen | Logps/ref Rejected | Logits/chosen | Logits/rejected |
|:-------------:|:------:|:----:|:---------------:|:----------------:|:------------------:|:------------------:|:-------------------------:|:------------:|:--------------:|:----------------:|:------------------:|:-------------:|:---------------:|
| 4.4915        | 0.4188 | 200  | 0.5614          | 286.9760         | 246.0880           | 40.8880            | 0.0                       | -383.4086    | -408.7636      | -288.6415        | -265.9616          | -0.9071       | -0.8845         |
| 4.2712        | 0.8377 | 400  | 0.5338          | 286.9760         | 246.0880           | 40.8880            | 0.0                       | -406.7590    | -443.3631      | -288.6415        | -265.9616          | -0.8653       | -0.8488         |


### Framework versions

- Transformers 4.51.0
- Pytorch 2.3.1+cu121
- Datasets 2.21.0
- Tokenizers 0.21.4