---
library_name: transformers
base_model: jackf857/qwen3-8b-base-sft-ultrachat-4xh200-batch-128
tags:
- alignment-handbook
- r-dpo
- generated_from_trainer
datasets:
- HuggingFaceH4/ultrafeedback_binarized
model-index:
- name: qwen3-8b-base-r-dpo-ultrafeedback-4xh200-batch-128
  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. -->

# qwen3-8b-base-r-dpo-ultrafeedback-4xh200-batch-128

This model is a fine-tuned version of [jackf857/qwen3-8b-base-sft-ultrachat-4xh200-batch-128](https://huggingface.co/jackf857/qwen3-8b-base-sft-ultrachat-4xh200-batch-128) on the HuggingFaceH4/ultrafeedback_binarized dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5469
- R Dpo/chosen Len: 294.4800
- R Dpo/rejected Len: 249.8700
- R Dpo/length Delta: 44.6100
- R Dpo/regularization Term: 4.4610
- Logps/chosen: -2936.7747
- Logps/rejected: -2579.1685
- Logps/ref Chosen: -281.4850
- Logps/ref Rejected: -261.8005
- Logits/chosen: 0.6320
- Logits/rejected: 0.5334

## 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: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- total_eval_batch_size: 16
- 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 |
|:-------------:|:------:|:----:|:---------------:|:----------------:|:------------------:|:------------------:|:-------------------------:|:------------:|:--------------:|:----------------:|:------------------:|:-------------:|:---------------:|
| 6.6608        | 0.4188 | 200  | 0.7288          | 294.4800         | 249.8700           | 44.6100            | 4.4610                    | -2819.1780   | -2438.3210     | -281.4850        | -261.8005          | 0.6077        | 0.5856          |
| 4.5388        | 0.8377 | 400  | 0.5469          | 294.4800         | 249.8700           | 44.6100            | 4.4610                    | -2936.7747   | -2579.1685     | -281.4850        | -261.8005          | 0.6320        | 0.5334          |


### Framework versions

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