gpt2_small_ZH_superbpe_32768_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 5.7362
- Accuracy: 0.2326
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 |
|---|
| 7.9396 | 1.0 | 1120 | 7.2211 | 0.1768 |
| 7.0134 | 2.0 | 2240 | 6.6447 | 0.1897 |
| 6.6006 | 3.0 | 3360 | 6.3292 | 0.2009 |
| 6.3313 | 4.0 | 4480 | 6.1152 | 0.2113 |
| 6.1343 | 5.0 | 5600 | 5.9778 | 0.2178 |
| 5.9886 | 6.0 | 6720 | 5.8833 | 0.2237 |
| 5.8815 | 7.0 | 7840 | 5.8178 | 0.2274 |
| 5.7993 | 8.0 | 8960 | 5.7741 | 0.2302 |
| 5.6855 | 9.0 | 10080 | 5.7470 | 0.2320 |
| 5.6525 | 10.0 | 11200 | 5.7362 | 0.2326 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
