<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
tiny-mlm-glue-qqp-target-glue-mnli
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: 0.8018
- Accuracy: 0.6440
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 |
|---|---|---|---|---|
| 1.0756 | 0.04 | 500 | 1.0341 | 0.4692 |
| 1.009 | 0.08 | 1000 | 0.9636 | 0.5497 |
| 0.9589 | 0.12 | 1500 | 0.9186 | 0.5777 |
| 0.9309 | 0.16 | 2000 | 0.9050 | 0.5885 |
| 0.9229 | 0.2 | 2500 | 0.8882 | 0.5992 |
| 0.9154 | 0.24 | 3000 | 0.8755 | 0.6048 |
| 0.8993 | 0.29 | 3500 | 0.8683 | 0.6136 |
| 0.8917 | 0.33 | 4000 | 0.8555 | 0.6193 |
| 0.8757 | 0.37 | 4500 | 0.8553 | 0.6165 |
| 0.8796 | 0.41 | 5000 | 0.8441 | 0.6235 |
| 0.8705 | 0.45 | 5500 | 0.8400 | 0.6239 |
| 0.8656 | 0.49 | 6000 | 0.8362 | 0.6289 |
| 0.8692 | 0.53 | 6500 | 0.8298 | 0.6307 |
| 0.8585 | 0.57 | 7000 | 0.8287 | 0.6315 |
| 0.8529 | 0.61 | 7500 | 0.8139 | 0.6400 |
| 0.8534 | 0.65 | 8000 | 0.8133 | 0.6391 |
| 0.8449 | 0.69 | 8500 | 0.8223 | 0.6342 |
| 0.8505 | 0.73 | 9000 | 0.7989 | 0.6473 |
| 0.841 | 0.77 | 9500 | 0.8106 | 0.6379 |
| 0.8302 | 0.81 | 10000 | 0.7966 | 0.6451 |
| 0.8338 | 0.86 | 10500 | 0.8018 | 0.6440 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2
