<!-- 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_lr5e-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: 0.1003
- Accuracy: 0.8811
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: 0.0005
- 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 |
---|---|---|---|---|
0.1376 | 0.17 | 2500 | 0.1103 | 0.8277 |
0.1149 | 0.34 | 5000 | 0.1001 | 0.845 |
0.1085 | 0.51 | 7500 | 0.0953 | 0.8503 |
0.1055 | 0.68 | 10000 | 0.0942 | 0.8523 |
0.1011 | 0.85 | 12500 | 0.0922 | 0.8558 |
0.0983 | 1.02 | 15000 | 0.0923 | 0.8567 |
0.0831 | 1.18 | 17500 | 0.0886 | 0.863 |
0.0831 | 1.35 | 20000 | 0.0883 | 0.8637 |
0.0838 | 1.52 | 22500 | 0.0860 | 0.8667 |
0.0826 | 1.69 | 25000 | 0.0864 | 0.8666 |
0.0819 | 1.86 | 27500 | 0.0859 | 0.869 |
0.0782 | 2.03 | 30000 | 0.0880 | 0.8678 |
0.0623 | 2.2 | 32500 | 0.0867 | 0.8702 |
0.0632 | 2.37 | 35000 | 0.0870 | 0.8705 |
0.0627 | 2.54 | 37500 | 0.0878 | 0.8729 |
0.0637 | 2.71 | 40000 | 0.0841 | 0.8745 |
0.0627 | 2.88 | 42500 | 0.0841 | 0.8766 |
0.0579 | 3.05 | 45000 | 0.0925 | 0.8755 |
0.0456 | 3.22 | 47500 | 0.0915 | 0.8765 |
0.0456 | 3.38 | 50000 | 0.0909 | 0.8778 |
0.0452 | 3.55 | 52500 | 0.0890 | 0.8795 |
0.0462 | 3.72 | 55000 | 0.0873 | 0.8807 |
0.0453 | 3.89 | 57500 | 0.0892 | 0.8791 |
0.0401 | 4.06 | 60000 | 0.0988 | 0.8793 |
0.0319 | 4.23 | 62500 | 0.1013 | 0.8805 |
0.0323 | 4.4 | 65000 | 0.1017 | 0.8799 |
0.0323 | 4.57 | 67500 | 0.1012 | 0.8809 |
0.0321 | 4.74 | 70000 | 0.1011 | 0.8809 |
0.0315 | 4.91 | 72500 | 0.1003 | 0.8811 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.0+cu117
- Datasets 2.10.1
- Tokenizers 0.12.1