<!-- 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. -->
bert-base-uncased-rte
This model is a fine-tuned version of bert-base-uncased on the GLUE RTE dataset. It achieves the following results on the evaluation set:
- Loss: 0.6540
- Accuracy: 0.6065
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: 5e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.7009 | 1.0 | 20 | 0.6781 | 0.5560 |
| 0.6393 | 2.0 | 40 | 0.6540 | 0.6065 |
| 0.4606 | 3.0 | 60 | 0.7134 | 0.6498 |
| 0.2597 | 4.0 | 80 | 0.8379 | 0.6751 |
| 0.1492 | 5.0 | 100 | 1.3531 | 0.6282 |
| 0.0954 | 6.0 | 120 | 1.2220 | 0.6354 |
| 0.0561 | 7.0 | 140 | 1.2282 | 0.6715 |
| 0.0379 | 8.0 | 160 | 1.4368 | 0.6679 |
| 0.0368 | 9.0 | 180 | 1.8559 | 0.6498 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2
