gpt2_medium_EN_unigram_65536_parallel3_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.7249
- Accuracy: 0.4759
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 |
|---|
| 3.1249 | 1.0 | 29350 | 3.0671 | 0.4283 |
| 2.976 | 2.0 | 58700 | 2.9324 | 0.4444 |
| 2.8878 | 3.0 | 88050 | 2.8663 | 0.4529 |
| 2.829 | 4.0 | 117400 | 2.8256 | 0.4587 |
| 2.7804 | 5.0 | 146750 | 2.7966 | 0.4632 |
| 2.7375 | 6.0 | 176100 | 2.7742 | 0.4665 |
| 2.696 | 7.0 | 205450 | 2.7560 | 0.4699 |
| 2.6602 | 8.0 | 234800 | 2.7418 | 0.4724 |
| 2.6229 | 9.0 | 264150 | 2.7306 | 0.4745 |
| 2.5875 | 10.0 | 293500 | 2.7249 | 0.4759 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
