gpt2_small_RU_bpe_81920_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.8495
- Accuracy: 0.4009
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 |
|---|
| 6.1932 | 1.0 | 1509 | 4.7615 | 0.3449 |
| 4.5028 | 2.0 | 3018 | 4.3585 | 0.3654 |
| 4.3127 | 3.0 | 4527 | 4.1765 | 0.3758 |
| 4.105 | 4.0 | 6036 | 4.0619 | 0.3834 |
| 4.0164 | 5.0 | 7545 | 3.9846 | 0.3896 |
| 3.9134 | 6.0 | 9054 | 3.9337 | 0.3938 |
| 3.8585 | 7.0 | 10563 | 3.8960 | 0.3974 |
| 3.8008 | 8.0 | 12072 | 3.8720 | 0.3992 |
| 3.7656 | 9.0 | 13581 | 3.8557 | 0.4004 |
| 3.7367 | 10.0 | 15090 | 3.8495 | 0.4009 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.9.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
