gpt2_small_TR_superbpe_32768_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 5.0929
- Accuracy: 0.2528
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 |
|---|
| 7.8027 | 1.0 | 1155 | 6.7092 | 0.1741 |
| 6.452 | 2.0 | 2310 | 5.9110 | 0.2070 |
| 5.8945 | 3.0 | 3465 | 5.5823 | 0.2217 |
| 5.6113 | 4.0 | 4620 | 5.4083 | 0.2310 |
| 5.4372 | 5.0 | 5775 | 5.2941 | 0.2379 |
| 5.3148 | 6.0 | 6930 | 5.2175 | 0.2435 |
| 5.1496 | 7.0 | 8085 | 5.1632 | 0.2469 |
| 5.0914 | 8.0 | 9240 | 5.1264 | 0.2500 |
| 5.0476 | 9.0 | 10395 | 5.1021 | 0.2525 |
| 5.0186 | 10.0 | 11550 | 5.0929 | 0.2528 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
