gpt2_small_TR_superbpe_8192_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 4.0159
- Accuracy: 0.3019
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 |
|---|
| 6.6772 | 1.0 | 1441 | 5.1299 | 0.1977 |
| 5.1737 | 2.0 | 2882 | 4.5829 | 0.2424 |
| 4.5146 | 3.0 | 4323 | 4.3599 | 0.2641 |
| 4.3742 | 4.0 | 5764 | 4.2391 | 0.2772 |
| 4.2061 | 5.0 | 7205 | 4.1621 | 0.2853 |
| 4.1397 | 6.0 | 8646 | 4.1085 | 0.2907 |
| 4.059 | 7.0 | 10087 | 4.0718 | 0.2954 |
| 4.0162 | 8.0 | 11528 | 4.0429 | 0.2988 |
| 3.9904 | 9.0 | 12969 | 4.0233 | 0.3011 |
| 3.9452 | 10.0 | 14410 | 4.0159 | 0.3019 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
