vietnamese-gpt2
This model is a fine-tuned version of nmthien/vietnamese-gpt2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.0493
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-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 500
- training_steps: 17000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.1841 | 0.0588 | 1000 | 3.0744 |
| 3.1802 | 0.1176 | 2000 | 3.0735 |
| 3.1664 | 0.1765 | 3000 | 3.0709 |
| 3.1513 | 0.2353 | 4000 | 3.0693 |
| 3.2179 | 0.2941 | 5000 | 3.0677 |
| 3.0193 | 0.3529 | 6000 | 3.0700 |
| 3.0308 | 0.4118 | 7000 | 3.0668 |
| 3.1744 | 0.4706 | 8000 | 3.0588 |
| 3.1756 | 0.5294 | 9000 | 3.0566 |
| 3.1679 | 0.5882 | 10000 | 3.0544 |
| 3.1626 | 0.6471 | 11000 | 3.0527 |
| 3.1515 | 0.0588 | 12000 | 3.0517 |
| 3.1470 | 0.1176 | 13000 | 3.0510 |
| 3.1325 | 0.1765 | 14000 | 3.0499 |
| 3.1195 | 0.2353 | 15000 | 3.0497 |
| 3.1889 | 0.2941 | 16000 | 3.0500 |
| 2.9972 | 0.3529 | 17000 | 3.0493 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2