gpt2_small_RU_superbpe_49152_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 5.0814
- Accuracy: 0.2631
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 |
|---|
| 7.8676 | 1.0 | 1184 | 6.7707 | 0.1815 |
| 6.5519 | 2.0 | 2368 | 5.9949 | 0.2097 |
| 5.9945 | 3.0 | 3552 | 5.6396 | 0.2275 |
| 5.6859 | 4.0 | 4736 | 5.4324 | 0.2402 |
| 5.4871 | 5.0 | 5920 | 5.3045 | 0.2477 |
| 5.2379 | 6.0 | 7104 | 5.2150 | 0.2548 |
| 5.1525 | 7.0 | 8288 | 5.1567 | 0.2579 |
| 5.0914 | 8.0 | 9472 | 5.1167 | 0.2609 |
| 5.0427 | 9.0 | 10656 | 5.0909 | 0.2622 |
| 5.0093 | 10.0 | 11840 | 5.0814 | 0.2631 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
