gpt2_small_EN_bpe_49152_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.2687
- Accuracy: 0.4137
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 |
|---|
| 5.3888 | 1.0 | 1765 | 3.8140 | 0.3672 |
| 3.7267 | 2.0 | 3530 | 3.5759 | 0.3845 |
| 3.538 | 3.0 | 5295 | 3.4699 | 0.3932 |
| 3.435 | 4.0 | 7060 | 3.4010 | 0.3995 |
| 3.3798 | 5.0 | 8825 | 3.3563 | 0.4039 |
| 3.3149 | 6.0 | 10590 | 3.3215 | 0.4081 |
| 3.2678 | 7.0 | 12355 | 3.2996 | 0.4095 |
| 3.2279 | 8.0 | 14120 | 3.2828 | 0.4122 |
| 3.2028 | 9.0 | 15885 | 3.2724 | 0.4128 |
| 3.18 | 10.0 | 17650 | 3.2687 | 0.4137 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
