<!-- 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. -->
FLAN-T5_GLUE_finetuning_lr8e-4
This model is a fine-tuned version of google/flan-t5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1016
- Accuracy: 0.8763
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: 0.0008
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.1513 | 0.17 | 2500 | 0.1262 | 0.7956 |
0.1251 | 0.34 | 5000 | 0.1114 | 0.8327 |
0.118 | 0.51 | 7500 | 0.1029 | 0.8377 |
0.1144 | 0.68 | 10000 | 0.1000 | 0.8431 |
0.1102 | 0.85 | 12500 | 0.0995 | 0.8448 |
0.1073 | 1.02 | 15000 | 0.0981 | 0.8453 |
0.0936 | 1.18 | 17500 | 0.0947 | 0.8522 |
0.0929 | 1.35 | 20000 | 0.0936 | 0.8545 |
0.0924 | 1.52 | 22500 | 0.0922 | 0.8547 |
0.0913 | 1.69 | 25000 | 0.0929 | 0.8586 |
0.0908 | 1.86 | 27500 | 0.0886 | 0.8619 |
0.0868 | 2.03 | 30000 | 0.0917 | 0.8596 |
0.0718 | 2.2 | 32500 | 0.0901 | 0.8647 |
0.0725 | 2.37 | 35000 | 0.0904 | 0.8636 |
0.0728 | 2.54 | 37500 | 0.0892 | 0.868 |
0.0729 | 2.71 | 40000 | 0.0892 | 0.8692 |
0.0718 | 2.88 | 42500 | 0.0870 | 0.8707 |
0.0658 | 3.05 | 45000 | 0.0952 | 0.8697 |
0.0523 | 3.22 | 47500 | 0.0938 | 0.8709 |
0.0524 | 3.38 | 50000 | 0.0943 | 0.8716 |
0.0517 | 3.55 | 52500 | 0.0929 | 0.8718 |
0.0523 | 3.72 | 55000 | 0.0905 | 0.8744 |
0.0519 | 3.89 | 57500 | 0.0903 | 0.8736 |
0.0452 | 4.06 | 60000 | 0.1016 | 0.8741 |
0.0351 | 4.23 | 62500 | 0.1025 | 0.8738 |
0.0356 | 4.4 | 65000 | 0.1028 | 0.8751 |
0.0352 | 4.57 | 67500 | 0.1028 | 0.8755 |
0.0344 | 4.74 | 70000 | 0.1021 | 0.8756 |
0.0342 | 4.91 | 72500 | 0.1016 | 0.8763 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.0+cu117
- Datasets 2.10.1
- Tokenizers 0.12.1