metadata
library_name: peft
license: apache-2.0
base_model: Qwen/Qwen3-8B
tags:
- generated_from_trainer
model-index:
- name: Qwen3-8B-009-3000
results: []
Qwen3-8B-009-3000
This model is a fine-tuned version of Qwen/Qwen3-8B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.7001
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: 2e-05
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Use paged_adamw_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.3438 | 0.8658 | 100 | 1.3453 |
| 1.2099 | 1.7273 | 200 | 1.1676 |
| 1.0514 | 2.5887 | 300 | 1.0531 |
| 0.949 | 3.4502 | 400 | 0.9713 |
| 0.9212 | 4.3117 | 500 | 0.9078 |
| 0.8159 | 5.1732 | 600 | 0.8502 |
| 0.7702 | 6.0346 | 700 | 0.8033 |
| 0.7158 | 6.9004 | 800 | 0.7620 |
| 0.6779 | 7.7619 | 900 | 0.7308 |
| 0.6509 | 8.6234 | 1000 | 0.7105 |
| 0.6163 | 9.4848 | 1100 | 0.7001 |
Framework versions
- PEFT 0.15.2
- Transformers 4.51.3
- Pytorch 2.7.0+cu126
- Datasets 3.5.1
- Tokenizers 0.21.1