metadata
library_name: transformers
license: mit
base_model: BRlkl/BingoGuard-bert-large-base-benchmarks
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: BingoGuard-bert-large-base-plus-custom
results: []
BingoGuard-bert-large-base-plus-custom
This model is a fine-tuned version of BRlkl/BingoGuard-bert-large-base-benchmarks on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6046
- Accuracy: 0.8915
- F1: 0.8884
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
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Use 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: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.3266 | 1.0 | 67 | 0.2604 | 0.8979 | 0.8970 |
| 0.2215 | 2.0 | 134 | 0.2803 | 0.8851 | 0.8831 |
| 0.096 | 3.0 | 201 | 0.3214 | 0.8894 | 0.8874 |
| 0.0439 | 4.0 | 268 | 0.4621 | 0.8915 | 0.8903 |
| 0.0324 | 5.0 | 335 | 0.5207 | 0.8936 | 0.8913 |
| 0.0119 | 6.0 | 402 | 0.5807 | 0.8915 | 0.8884 |
| 0.0076 | 7.0 | 469 | 0.5984 | 0.8936 | 0.8913 |
| 0.0032 | 8.0 | 536 | 0.6046 | 0.8915 | 0.8884 |
Framework versions
- Transformers 4.55.4
- Pytorch 2.8.0+cu128
- Datasets 3.6.0
- Tokenizers 0.21.4