gpt2_small_RU_bpe_32768_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.6330
- Accuracy: 0.4090
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 |
|---|
| 5.9887 | 1.0 | 1599 | 4.6300 | 0.3315 |
| 4.3686 | 2.0 | 3198 | 4.1692 | 0.3616 |
| 4.1546 | 3.0 | 4797 | 3.9601 | 0.3781 |
| 3.9205 | 4.0 | 6396 | 3.8422 | 0.3879 |
| 3.8417 | 5.0 | 7995 | 3.7692 | 0.3944 |
| 3.7266 | 6.0 | 9594 | 3.7171 | 0.3997 |
| 3.659 | 7.0 | 11193 | 3.6809 | 0.4026 |
| 3.6211 | 8.0 | 12792 | 3.6548 | 0.4068 |
| 3.5766 | 9.0 | 14391 | 3.6394 | 0.4077 |
| 3.5576 | 10.0 | 15990 | 3.6330 | 0.4090 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.9.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
