---
library_name: transformers
license: apache-2.0
base_model: jackf857/qwen3-8b-base-sft-hh-helpful-4xh200-batch-64-20260417-214452
tags:
- alignment-handbook
- new-dpo
- generated_from_trainer
datasets:
- Anthropic/hh-rlhf
model-index:
- name: qwen3-8b-base-new-dpo-hh-helpful-4xh200-batch-64-q_t-0.45-s_star-0.6
  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-new-dpo-hh-helpful-4xh200-batch-64-q_t-0.45-s_star-0.6

This model is a fine-tuned version of [jackf857/qwen3-8b-base-sft-hh-helpful-4xh200-batch-64-20260417-214452](https://huggingface.co/jackf857/qwen3-8b-base-sft-hh-helpful-4xh200-batch-64-20260417-214452) on the Anthropic/hh-rlhf dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5224
- Fcm Dpo/beta: 0.0073
- Margin Dpo/margin Mean: 73.6023
- Margin Dpo/margin Std: 100.2118
- Logps/chosen: -265.1147
- Logps/rejected: -332.2911
- Logps/ref Chosen: -100.4936
- Logps/ref Rejected: -94.0678
- Logits/chosen: -1.8407
- Logits/rejected: -1.4555

## 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: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- total_eval_batch_size: 32
- 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 | Fcm Dpo/beta | Margin Dpo/margin Mean | Margin Dpo/margin Std | Logps/chosen | Logps/rejected | Logps/ref Chosen | Logps/ref Rejected | Logits/chosen | Logits/rejected |
|:-------------:|:------:|:----:|:---------------:|:------------:|:----------------------:|:---------------------:|:------------:|:--------------:|:----------------:|:------------------:|:-------------:|:---------------:|
| 0.9831        | 0.1468 | 100  | 0.5851          | 0.1290       | 3.0273                 | 6.0741                | -103.4172    | -100.0187      | -100.4936        | -94.0678           | 0.0310        | 0.2743          |
| 0.9603        | 0.2937 | 200  | 0.5345          | 0.0330       | 15.0670                | 21.7615               | -123.5883    | -132.2295      | -100.4936        | -94.0678           | -1.4398       | -1.1262         |
| 1.0079        | 0.4405 | 300  | 0.5175          | 0.0128       | 42.6819                | 56.6599               | -199.0951    | -235.3512      | -100.4936        | -94.0678           | -1.7159       | -1.3591         |
| 0.9031        | 0.5874 | 400  | 0.5321          | 0.0072       | 66.1747                | 89.8270               | -255.0186    | -314.7675      | -100.4936        | -94.0678           | -1.9564       | -1.5863         |
| 1.0372        | 0.7342 | 500  | 0.5229          | 0.0074       | 72.2195                | 99.1610               | -264.5698    | -330.3635      | -100.4936        | -94.0678           | -1.9734       | -1.5979         |
| 0.9868        | 0.8811 | 600  | 0.5224          | 0.0073       | 73.6023                | 100.2118              | -265.1147    | -332.2911      | -100.4936        | -94.0678           | -1.8407       | -1.4555         |


### Framework versions

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