metadata
license: apache-2.0
base_model: bert-base-uncased
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: boss-toxicity-6000-bert-base-uncased
results: []
boss-toxicity-6000-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: 0.7068
- Acc: 0.8555
- Loss: 0.9111
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 |
|---|---|---|---|---|---|
| No log | 1.0 | 375 | 0.6887 | 0.8526 | 0.3768 |
| 0.5463 | 2.0 | 750 | 0.7002 | 0.8500 | 0.3333 |
| 0.27 | 3.0 | 1125 | 0.6929 | 0.8417 | 0.3775 |
| 0.1582 | 4.0 | 1500 | 0.7082 | 0.8564 | 0.5663 |
| 0.1582 | 5.0 | 1875 | 0.6730 | 0.8230 | 0.7996 |
| 0.0617 | 6.0 | 2250 | 0.6581 | 0.8051 | 1.1498 |
| 0.0267 | 7.0 | 2625 | 0.7068 | 0.8555 | 0.9111 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1