gpt2_small_TR_superbpe_81920_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 5.8592
- Accuracy: 0.2330
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 |
|---|
| 8.3877 | 1.0 | 1025 | 7.5628 | 0.1771 |
| 7.2227 | 2.0 | 2050 | 6.8679 | 0.1949 |
| 6.712 | 3.0 | 3075 | 6.4889 | 0.2054 |
| 6.4051 | 4.0 | 4100 | 6.2683 | 0.2136 |
| 6.2033 | 5.0 | 5125 | 6.1188 | 0.2202 |
| 6.0567 | 6.0 | 6150 | 6.0158 | 0.2245 |
| 5.9486 | 7.0 | 7175 | 5.9466 | 0.2287 |
| 5.8682 | 8.0 | 8200 | 5.9000 | 0.2305 |
| 5.8094 | 9.0 | 9225 | 5.8700 | 0.2325 |
| 5.7718 | 10.0 | 10250 | 5.8592 | 0.2330 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
