gpt2_small_ZH_bpe_8192_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 4.1737
- Accuracy: 0.3324
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: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- lr_scheduler_warmup_steps: 1000
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|
| 6.3765 | 1.0 | 1494 | 5.2041 | 0.2468 |
| 5.283 | 2.0 | 2988 | 4.7282 | 0.2792 |
| 4.6935 | 3.0 | 4482 | 4.5185 | 0.2966 |
| 4.5593 | 4.0 | 5976 | 4.3982 | 0.3094 |
| 4.3769 | 5.0 | 7470 | 4.3194 | 0.3166 |
| 4.3182 | 6.0 | 8964 | 4.2640 | 0.3216 |
| 4.218 | 7.0 | 10458 | 4.2271 | 0.3258 |
| 4.1856 | 8.0 | 11952 | 4.1999 | 0.3292 |
| 4.1233 | 9.0 | 13446 | 4.1811 | 0.3313 |
| 4.1074 | 10.0 | 14940 | 4.1737 | 0.3324 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
