gpt2_medium_AR_superbpe_65536_parallel3_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 4.0172
- Accuracy: 0.3240
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 |
|---|
| 4.6177 | 1.0 | 22627 | 4.5372 | 0.2702 |
| 4.3672 | 2.0 | 45254 | 4.3199 | 0.2895 |
| 4.2405 | 3.0 | 67881 | 4.2206 | 0.2994 |
| 4.1475 | 4.0 | 90508 | 4.1600 | 0.3055 |
| 4.0782 | 5.0 | 113135 | 4.1154 | 0.3107 |
| 4.0192 | 6.0 | 135762 | 4.0820 | 0.3145 |
| 3.9653 | 7.0 | 158389 | 4.0573 | 0.3185 |
| 3.9166 | 8.0 | 181016 | 4.0384 | 0.3205 |
| 3.8695 | 9.0 | 203643 | 4.0231 | 0.3234 |
| 3.8248 | 10.0 | 226270 | 4.0172 | 0.3240 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
