olm-bert-tiny-december-2022-target-glue-qqp
This model is a fine-tuned version of muhtasham/olm-bert-tiny-december-2022 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5217
- Accuracy: 0.7433
- F1: 0.6048
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- training_steps: 5000
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.6283 | 0.04 | 500 | 0.5955 | 0.6795 | 0.5186 |
| 0.5875 | 0.09 | 1000 | 0.5763 | 0.6972 | 0.5596 |
| 0.5791 | 0.13 | 1500 | 0.5690 | 0.6975 | 0.6011 |
| 0.5666 | 0.18 | 2000 | 0.5536 | 0.7156 | 0.5520 |
| 0.5568 | 0.22 | 2500 | 0.5447 | 0.7230 | 0.5709 |
| 0.5489 | 0.26 | 3000 | 0.5386 | 0.7281 | 0.5665 |
| 0.5465 | 0.31 | 3500 | 0.5305 | 0.7329 | 0.5917 |
| 0.5384 | 0.35 | 4000 | 0.5262 | 0.7357 | 0.6231 |
| 0.5422 | 0.4 | 4500 | 0.5207 | 0.7409 | 0.6200 |
| 0.5299 | 0.44 | 5000 | 0.5217 | 0.7433 | 0.6048 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.9.1.dev0
- Tokenizers 0.13.2