gpt2_medium_ZH_superbpe_65536_parallel3_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 4.6060
- Accuracy: 0.2820
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 |
|---|
| 5.2845 | 1.0 | 18703 | 5.1870 | 0.2281 |
| 4.9869 | 2.0 | 37406 | 4.9343 | 0.2474 |
| 4.8331 | 3.0 | 56109 | 4.8190 | 0.2579 |
| 4.7318 | 4.0 | 74812 | 4.7509 | 0.2644 |
| 4.6494 | 5.0 | 93515 | 4.7041 | 0.2692 |
| 4.5772 | 6.0 | 112218 | 4.6697 | 0.2731 |
| 4.5192 | 7.0 | 130921 | 4.6427 | 0.2771 |
| 4.4625 | 8.0 | 149624 | 4.6262 | 0.2792 |
| 4.4112 | 9.0 | 168327 | 4.6117 | 0.2814 |
| 4.3608 | 10.0 | 187030 | 4.6060 | 0.2820 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
