ModernBERT-large-FT
This model is a fine-tuned version of answerdotai/ModernBERT-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5883
- Accuracy: 0.8958
- F1: 0.8910
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.8974 | 1.0 | 477 | 0.2649 | 0.9368 | 0.9359 |
| 0.1073 | 2.0 | 954 | 0.2305 | 0.9558 | 0.9557 |
| 0.0481 | 3.0 | 1431 | 0.2310 | 0.9610 | 0.9602 |
| 0.022 | 4.0 | 1908 | 0.1921 | 0.9645 | 0.9641 |
| 0.007 | 5.0 | 2385 | 0.1973 | 0.9684 | 0.9678 |
| 0.0072 | 6.0 | 2862 | 0.1938 | 0.9697 | 0.9694 |
| 0.006 | 7.0 | 3339 | 0.1998 | 0.97 | 0.9696 |
| 0.002 | 8.0 | 3816 | 0.2106 | 0.9658 | 0.9654 |
| 0.0007 | 9.0 | 4293 | 0.2084 | 0.9665 | 0.9661 |
Framework versions
- Transformers 4.56.2
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1