metadata
license: apache-2.0
base_model: bert-base-uncased
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: boss-toxicity-24000-bert-base-uncased
results: []
boss-toxicity-24000-bert-base-uncased
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- F1: 0.7003
- Acc: 0.8460
- Loss: 1.2310
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | F1 | Acc | Validation Loss |
|---|---|---|---|---|---|
| 0.3305 | 1.0 | 1500 | 0.6780 | 0.8250 | 0.3952 |
| 0.2821 | 2.0 | 3000 | 0.6773 | 0.8219 | 0.3777 |
| 0.2 | 3.0 | 4500 | 0.7331 | 0.8766 | 0.3716 |
| 0.1396 | 4.0 | 6000 | 0.6900 | 0.8356 | 0.6738 |
| 0.0914 | 5.0 | 7500 | 0.7011 | 0.8472 | 0.6766 |
| 0.0548 | 6.0 | 9000 | 0.7480 | 0.8905 | 0.6274 |
| 0.0365 | 7.0 | 10500 | 0.6624 | 0.8068 | 1.2739 |
| 0.0324 | 8.0 | 12000 | 0.7222 | 0.8655 | 1.0403 |
| 0.0295 | 9.0 | 13500 | 0.7003 | 0.8460 | 1.2310 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1