Fake-Citation-Detector-V2-bert
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8065
- Accuracy: 0.5964
- Macro F1: 0.5753
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: 2e-05
- train_batch_size: 16
- 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
- lr_scheduler_warmup_steps: 51
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Macro F1 |
|---|---|---|---|---|---|
| 0.7187 | 1.0 | 86 | 0.6909 | 0.5181 | 0.3986 |
| 0.6939 | 2.0 | 172 | 0.6925 | 0.5 | 0.3333 |
| 0.6628 | 3.0 | 258 | 0.6801 | 0.5482 | 0.5011 |
| 0.6166 | 4.0 | 344 | 0.6908 | 0.6024 | 0.585 |
| 0.4969 | 5.0 | 430 | 0.7657 | 0.5904 | 0.5677 |
| 0.4062 | 6.0 | 516 | 0.8065 | 0.5964 | 0.5753 |
Framework versions
- Transformers 5.17.0
- Pytorch 2.11.0+cu130
- Datasets 4.8.5
- Tokenizers 0.23.2