<!-- 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_sa_GLUE_Experiment_logit_kd_data_aug_rte
This model is a fine-tuned version of distilbert-base-uncased on the GLUE RTE dataset. It achieves the following results on the evaluation set:
- Loss: 0.5391
- Accuracy: 0.5523
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: 256
- eval_batch_size: 256
- 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
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.3193 | 1.0 | 568 | 0.5516 | 0.5018 |
0.285 | 2.0 | 1136 | 0.5504 | 0.4910 |
0.282 | 3.0 | 1704 | 0.5442 | 0.5271 |
0.2804 | 4.0 | 2272 | 0.5491 | 0.5126 |
0.2794 | 5.0 | 2840 | 0.5582 | 0.5379 |
0.279 | 6.0 | 3408 | 0.5391 | 0.5523 |
0.2786 | 7.0 | 3976 | 0.5446 | 0.5126 |
0.2784 | 8.0 | 4544 | 0.5440 | 0.5271 |
0.2782 | 9.0 | 5112 | 0.5522 | 0.5090 |
0.2781 | 10.0 | 5680 | 0.5519 | 0.5271 |
0.278 | 11.0 | 6248 | 0.5595 | 0.5054 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2