gpt2_small_KO_superbpe_32768_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 5.3553
- Accuracy: 0.2272
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: 64
- eval_batch_size: 64
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 128
- total_eval_batch_size: 128
- 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 |
|---|
| 7.9778 | 1.0 | 1187 | 6.8832 | 0.1627 |
| 6.6731 | 2.0 | 2374 | 6.1667 | 0.1861 |
| 6.1636 | 3.0 | 3561 | 5.8536 | 0.1990 |
| 5.8901 | 4.0 | 4748 | 5.6699 | 0.2089 |
| 5.7149 | 5.0 | 5935 | 5.5594 | 0.2143 |
| 5.4917 | 6.0 | 7122 | 5.4808 | 0.2188 |
| 5.4155 | 7.0 | 8309 | 5.4263 | 0.2218 |
| 5.3581 | 8.0 | 9496 | 5.3877 | 0.2243 |
| 5.3119 | 9.0 | 10683 | 5.3646 | 0.2264 |
| 5.2796 | 10.0 | 11870 | 5.3553 | 0.2272 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
