gpt2_small_TR_bpe_49152_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 4.1055
- Accuracy: 0.4025
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 |
|---|
| 6.3471 | 1.0 | 1403 | 5.1252 | 0.3486 |
| 5.132 | 2.0 | 2806 | 4.6772 | 0.3669 |
| 4.5898 | 3.0 | 4209 | 4.4717 | 0.3778 |
| 4.453 | 4.0 | 5612 | 4.3426 | 0.3857 |
| 4.2786 | 5.0 | 7015 | 4.2548 | 0.3910 |
| 4.2034 | 6.0 | 8418 | 4.1970 | 0.3948 |
| 4.1542 | 7.0 | 9821 | 4.1574 | 0.3974 |
| 4.0779 | 8.0 | 11224 | 4.1291 | 0.4001 |
| 4.0468 | 9.0 | 12627 | 4.1130 | 0.4017 |
| 4.0121 | 10.0 | 14030 | 4.1055 | 0.4025 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.9.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
