<!-- 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. -->
distilbert-base-uncased-finetuned-cola
This model is a fine-tuned version of distilbert-base-uncased on the glue dataset. It achieves the following results on the evaluation set:
- Loss: 1.1491
- Matthews Correlation: 0.5365
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
Training results
Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
---|---|---|---|---|
No log | 1.0 | 268 | 0.7051 | 0.5240 |
0.1591 | 2.0 | 536 | 0.7489 | 0.5047 |
0.1591 | 3.0 | 804 | 0.7951 | 0.5308 |
0.0931 | 4.0 | 1072 | 0.9231 | 0.4909 |
0.0931 | 5.0 | 1340 | 0.9489 | 0.5185 |
0.0693 | 6.0 | 1608 | 1.0154 | 0.5269 |
0.0693 | 7.0 | 1876 | 1.1369 | 0.5111 |
0.0493 | 8.0 | 2144 | 1.1491 | 0.5365 |
0.0493 | 9.0 | 2412 | 1.2824 | 0.5358 |
0.0288 | 10.0 | 2680 | 1.3411 | 0.5189 |
0.0288 | 11.0 | 2948 | 1.3368 | 0.5250 |
0.0248 | 12.0 | 3216 | 1.3929 | 0.5354 |
0.0248 | 13.0 | 3484 | 1.5103 | 0.5082 |
0.0145 | 14.0 | 3752 | 1.4857 | 0.5305 |
0.0108 | 15.0 | 4020 | 1.4900 | 0.5331 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2