gpt2_small_ZH_bpe_32768_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 5.0900
- Accuracy: 0.3045
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 |
|---|
| 7.1569 | 1.0 | 1253 | 6.3349 | 0.2520 |
| 6.2564 | 2.0 | 2506 | 5.8732 | 0.2630 |
| 5.9126 | 3.0 | 3759 | 5.6007 | 0.2743 |
| 5.5126 | 4.0 | 5012 | 5.4167 | 0.2839 |
| 5.3708 | 5.0 | 6265 | 5.2979 | 0.2912 |
| 5.2701 | 6.0 | 7518 | 5.2164 | 0.2959 |
| 5.1945 | 7.0 | 8771 | 5.1605 | 0.2989 |
| 5.0843 | 8.0 | 10024 | 5.1223 | 0.3021 |
| 5.0391 | 9.0 | 11277 | 5.0985 | 0.3037 |
| 5.0129 | 10.0 | 12530 | 5.0900 | 0.3045 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
