gpt2_small_KO_superbpe_4263_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.8095
- Accuracy: 0.2991
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 |
|---|
| 6.165 | 1.0 | 1627 | 4.7121 | 0.2059 |
| 4.4795 | 2.0 | 3254 | 4.2761 | 0.2481 |
| 4.2842 | 3.0 | 4881 | 4.1009 | 0.2659 |
| 4.0793 | 4.0 | 6508 | 4.0023 | 0.2771 |
| 3.9673 | 5.0 | 8135 | 3.9356 | 0.2845 |
| 3.9115 | 6.0 | 9762 | 3.8922 | 0.2896 |
| 3.844 | 7.0 | 11389 | 3.8578 | 0.2935 |
| 3.7965 | 8.0 | 13016 | 3.8326 | 0.2965 |
| 3.7624 | 9.0 | 14643 | 3.8168 | 0.2982 |
| 3.7352 | 10.0 | 16270 | 3.8095 | 0.2991 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
