gpt2_medium_EN_bpe_65536_parallel3_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.6175
- Accuracy: 0.4942
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 |
|---|
| 2.9891 | 1.0 | 30432 | 2.9370 | 0.4477 |
| 2.8459 | 2.0 | 60864 | 2.8137 | 0.4623 |
| 2.7655 | 3.0 | 91296 | 2.7525 | 0.4706 |
| 2.7099 | 4.0 | 121728 | 2.7133 | 0.4766 |
| 2.6641 | 5.0 | 152160 | 2.6859 | 0.4815 |
| 2.6254 | 6.0 | 182592 | 2.6647 | 0.4847 |
| 2.5859 | 7.0 | 213024 | 2.6457 | 0.4879 |
| 2.5526 | 8.0 | 243456 | 2.6316 | 0.4906 |
| 2.5139 | 9.0 | 273888 | 2.6232 | 0.4925 |
| 2.4788 | 10.0 | 304320 | 2.6175 | 0.4942 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
