FLAN-T5_GLUE_finetuning_lr1e-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: 0.0828
- Accuracy: 0.8829
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: 0.0001
- 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 |
---|---|---|---|---|
0.1325 | 0.17 | 2500 | 0.1085 | 0.8349 |
0.1061 | 0.34 | 5000 | 0.0972 | 0.851 |
0.0999 | 0.51 | 7500 | 0.0897 | 0.8584 |
0.0965 | 0.68 | 10000 | 0.0876 | 0.862 |
0.0931 | 0.85 | 12500 | 0.0844 | 0.8661 |
0.091 | 1.02 | 15000 | 0.0844 | 0.8662 |
0.0788 | 1.18 | 17500 | 0.0829 | 0.8707 |
0.0789 | 1.35 | 20000 | 0.0833 | 0.8695 |
0.0784 | 1.52 | 22500 | 0.0813 | 0.8709 |
0.0775 | 1.69 | 25000 | 0.0818 | 0.8731 |
0.0776 | 1.86 | 27500 | 0.0808 | 0.874 |
0.0752 | 2.03 | 30000 | 0.0821 | 0.8747 |
0.0655 | 2.2 | 32500 | 0.0798 | 0.8778 |
0.0659 | 2.37 | 35000 | 0.0831 | 0.8759 |
0.0649 | 2.54 | 37500 | 0.0806 | 0.8786 |
0.0664 | 2.71 | 40000 | 0.0793 | 0.8785 |
0.0652 | 2.88 | 42500 | 0.0787 | 0.8794 |
0.0631 | 3.05 | 45000 | 0.0827 | 0.88 |
0.0565 | 3.22 | 47500 | 0.0826 | 0.8808 |
0.056 | 3.38 | 50000 | 0.0840 | 0.8798 |
0.0562 | 3.55 | 52500 | 0.0813 | 0.8817 |
0.0572 | 3.72 | 55000 | 0.0789 | 0.8828 |
0.0567 | 3.89 | 57500 | 0.0804 | 0.8824 |
0.0542 | 4.06 | 60000 | 0.0823 | 0.8835 |
0.0496 | 4.23 | 62500 | 0.0846 | 0.8824 |
0.0503 | 4.4 | 65000 | 0.0836 | 0.8822 |
0.0508 | 4.57 | 67500 | 0.0832 | 0.883 |
0.0506 | 4.74 | 70000 | 0.0829 | 0.8828 |
0.0505 | 4.91 | 72500 | 0.0828 | 0.8829 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.0+cu117
- Datasets 2.10.1
- Tokenizers 0.12.1
- Downloads last month
- 7
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.