| --- |
| library_name: transformers |
| license: apache-2.0 |
| base_model: deepvk/RuModernBERT-base |
| tags: |
| - generated_from_trainer |
| metrics: |
| - accuracy |
| model-index: |
| - name: RuHalluBERT-base-v3 |
| results: [] |
| --- |
| |
| <!-- This model card has been generated automatically according to the information the Trainer had access to. You |
| should probably proofread and complete it, then remove this comment. --> |
|
|
| # RuHalluBERT-base-v3 |
|
|
| This model is a fine-tuned version of [deepvk/RuModernBERT-base](https://huggingface.co/deepvk/RuModernBERT-base) on an unknown dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.7977 |
| - F1 Macro: 0.6248 |
| - F1 Class1: 0.6341 |
| - F1 Class0: 0.6154 |
| - Accuracy: 0.625 |
| - Precision Macro: 0.6290 |
| - Recall Macro: 0.6302 |
|
|
| ## 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: 2 |
| - eval_batch_size: 2 |
| - seed: 42 |
| - distributed_type: multi-GPU |
| - gradient_accumulation_steps: 16 |
| - total_train_batch_size: 32 |
| - 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: cosine |
| - lr_scheduler_warmup_steps: 0.06 |
| - num_epochs: 10 |
|
|
| ### Training results |
|
|
| | Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1 Class1 | F1 Class0 | Accuracy | Precision Macro | Recall Macro | |
| |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:---------:|:--------:|:---------------:|:------------:| |
| | 13.9495 | 1.0 | 98 | 0.7467 | 0.4752 | 0.6256 | 0.3248 | 0.5183 | 0.6315 | 0.5669 | |
| | 11.2511 | 2.0 | 196 | 0.6250 | 0.6699 | 0.6860 | 0.6538 | 0.6707 | 0.6968 | 0.6897 | |
| | 9.3975 | 3.0 | 294 | 0.6134 | 0.6450 | 0.5758 | 0.7143 | 0.6585 | 0.6513 | 0.6439 | |
| | 8.1226 | 4.0 | 392 | 0.6403 | 0.6089 | 0.5373 | 0.6804 | 0.6220 | 0.6124 | 0.6084 | |
| | 7.3214 | 5.0 | 490 | 0.6810 | 0.7291 | 0.7027 | 0.7556 | 0.7317 | 0.7285 | 0.7318 | |
| | 4.5838 | 6.0 | 588 | 0.7438 | 0.7066 | 0.6207 | 0.7925 | 0.7317 | 0.7529 | 0.7051 | |
| | 2.3476 | 7.0 | 686 | 1.3592 | 0.6559 | 0.5593 | 0.7524 | 0.6829 | 0.6887 | 0.6571 | |
| | 0.8999 | 8.0 | 784 | 1.5965 | 0.6723 | 0.5984 | 0.7463 | 0.6890 | 0.6865 | 0.6708 | |
|
|
|
|
| ### Framework versions |
|
|
| - Transformers 5.8.1 |
| - Pytorch 2.11.0+cu130 |
| - Datasets 4.8.5 |
| - Tokenizers 0.22.2 |
|
|
