gpt2_small_ZH_unigram_65536_parallel3-100_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 4.8672
- Accuracy: 0.2815
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 |
|---|
| 6.0886 | 1.0 | 5143 | 5.9297 | 0.2224 |
| 5.5224 | 2.0 | 10286 | 5.4195 | 0.2446 |
| 5.2705 | 3.0 | 15429 | 5.2034 | 0.2568 |
| 5.1159 | 4.0 | 20572 | 5.0813 | 0.2645 |
| 5.0106 | 5.0 | 25715 | 5.0071 | 0.2695 |
| 4.9314 | 6.0 | 30858 | 4.9547 | 0.2737 |
| 4.8674 | 7.0 | 36001 | 4.9165 | 0.2767 |
| 4.8197 | 8.0 | 41144 | 4.8914 | 0.2789 |
| 4.7787 | 9.0 | 46287 | 4.8753 | 0.2805 |
| 4.7487 | 10.0 | 51430 | 4.8672 | 0.2815 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
