tiny-mlm-glue-qqp-target-glue-sst2
This model is a fine-tuned version of muhtasham/tiny-mlm-glue-qqp on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5039
- Accuracy: 0.8291
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
- num_epochs: 200
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.5922 | 0.24 | 500 | 0.4935 | 0.7798 |
| 0.4475 | 0.48 | 1000 | 0.4672 | 0.7936 |
| 0.3948 | 0.71 | 1500 | 0.4418 | 0.7947 |
| 0.3742 | 0.95 | 2000 | 0.4701 | 0.7878 |
| 0.3364 | 1.19 | 2500 | 0.4464 | 0.8050 |
| 0.318 | 1.43 | 3000 | 0.4442 | 0.8108 |
| 0.2982 | 1.66 | 3500 | 0.4462 | 0.8062 |
| 0.2942 | 1.9 | 4000 | 0.4449 | 0.8211 |
| 0.2759 | 2.14 | 4500 | 0.4794 | 0.8062 |
| 0.2554 | 2.38 | 5000 | 0.4390 | 0.8200 |
| 0.2476 | 2.61 | 5500 | 0.4339 | 0.8303 |
| 0.2572 | 2.85 | 6000 | 0.4432 | 0.8268 |
| 0.2383 | 3.09 | 6500 | 0.4562 | 0.8291 |
| 0.2339 | 3.33 | 7000 | 0.4548 | 0.8349 |
| 0.2178 | 3.56 | 7500 | 0.4400 | 0.8349 |
| 0.2156 | 3.8 | 8000 | 0.4745 | 0.8337 |
| 0.2135 | 4.04 | 8500 | 0.5039 | 0.8291 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2