gpt2_small_KO_bpe_65536_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 4.9687
- Accuracy: 0.3211
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 |
|---|
| 7.2424 | 1.0 | 1263 | 6.0717 | 0.2768 |
| 5.9898 | 2.0 | 2526 | 5.5903 | 0.2913 |
| 5.6229 | 3.0 | 3789 | 5.3740 | 0.3005 |
| 5.2899 | 4.0 | 5052 | 5.2371 | 0.3062 |
| 5.1753 | 5.0 | 6315 | 5.1418 | 0.3113 |
| 5.09 | 6.0 | 7578 | 5.0736 | 0.3148 |
| 5.0227 | 7.0 | 8841 | 5.0282 | 0.3174 |
| 4.9215 | 8.0 | 10104 | 4.9958 | 0.3195 |
| 4.8819 | 9.0 | 11367 | 4.9755 | 0.3209 |
| 4.8578 | 10.0 | 12630 | 4.9687 | 0.3211 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
