gpt2_medium_ZH_bpe_32768_parallel3-100_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 4.1152
- Accuracy: 0.3377
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 |
|---|
| 5.0444 | 1.0 | 5916 | 4.7743 | 0.2769 |
| 4.5497 | 2.0 | 11832 | 4.4379 | 0.3034 |
| 4.3448 | 3.0 | 17748 | 4.3028 | 0.3148 |
| 4.2064 | 4.0 | 23664 | 4.2230 | 0.3224 |
| 4.0984 | 5.0 | 29580 | 4.1785 | 0.3275 |
| 4.0088 | 6.0 | 35496 | 4.1478 | 0.3302 |
| 3.9274 | 7.0 | 41412 | 4.1290 | 0.3337 |
| 3.8572 | 8.0 | 47328 | 4.1195 | 0.3355 |
| 3.7943 | 9.0 | 53244 | 4.1154 | 0.3371 |
| 3.7402 | 10.0 | 59160 | 4.1152 | 0.3377 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
