metadata
license: apache-2.0
base_model: bert-base-uncased
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: boss-toxicity-48000-bert-base-uncased
results: []
boss-toxicity-48000-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: 0.7261
- Acc: 0.8698
- Loss: 0.8738
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 |
|---|---|---|---|---|---|
| 0.2892 | 1.0 | 3000 | 0.7239 | 0.8667 | 0.3107 |
| 0.2632 | 2.0 | 6000 | 0.7030 | 0.8463 | 0.3864 |
| 0.2114 | 3.0 | 9000 | 0.7417 | 0.8815 | 0.3007 |
| 0.143 | 4.0 | 12000 | 0.7330 | 0.8755 | 0.4253 |
| 0.0924 | 5.0 | 15000 | 0.7083 | 0.8532 | 0.6377 |
| 0.0622 | 6.0 | 18000 | 0.7261 | 0.8698 | 0.8738 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1