gpt2_small_KO_bpe_8192_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.9289
- Accuracy: 0.3383
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 |
|---|
| 6.2119 | 1.0 | 1571 | 4.8288 | 0.2597 |
| 4.5868 | 2.0 | 3142 | 4.4100 | 0.2907 |
| 4.3979 | 3.0 | 4713 | 4.2278 | 0.3073 |
| 4.1929 | 4.0 | 6284 | 4.1236 | 0.3174 |
| 4.1196 | 5.0 | 7855 | 4.0555 | 0.3240 |
| 4.0191 | 6.0 | 9426 | 4.0092 | 0.3290 |
| 3.981 | 7.0 | 10997 | 3.9748 | 0.3322 |
| 3.914 | 8.0 | 12568 | 3.9519 | 0.3347 |
| 3.8765 | 9.0 | 14139 | 3.9360 | 0.3368 |
| 3.853 | 10.0 | 15710 | 3.9289 | 0.3383 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
