20230824103950
This model is a fine-tuned version of bert-large-cased on the super_glue dataset. It achieves the following results on the evaluation set:
- Loss: 0.6377
- Accuracy: 0.7401
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: 0.003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 11
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 60.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
No log | 1.0 | 312 | 0.9784 | 0.5307 |
0.905 | 2.0 | 624 | 0.6756 | 0.5126 |
0.905 | 3.0 | 936 | 0.7039 | 0.5379 |
0.7844 | 4.0 | 1248 | 0.6938 | 0.5090 |
0.7863 | 5.0 | 1560 | 0.7988 | 0.5487 |
0.7863 | 6.0 | 1872 | 0.7152 | 0.5993 |
0.7505 | 7.0 | 2184 | 0.7856 | 0.6173 |
0.7505 | 8.0 | 2496 | 0.6053 | 0.6606 |
0.7043 | 9.0 | 2808 | 0.6424 | 0.5957 |
0.7083 | 10.0 | 3120 | 0.7874 | 0.6354 |
0.7083 | 11.0 | 3432 | 0.6513 | 0.6390 |
0.6321 | 12.0 | 3744 | 0.5910 | 0.7148 |
0.6204 | 13.0 | 4056 | 0.5993 | 0.7112 |
0.6204 | 14.0 | 4368 | 0.5440 | 0.7292 |
0.5835 | 15.0 | 4680 | 0.5542 | 0.7184 |
0.5835 | 16.0 | 4992 | 0.6144 | 0.7329 |
0.5634 | 17.0 | 5304 | 0.5821 | 0.6968 |
0.5461 | 18.0 | 5616 | 0.6826 | 0.5776 |
0.5461 | 19.0 | 5928 | 0.5617 | 0.7148 |
0.5275 | 20.0 | 6240 | 0.7824 | 0.6643 |
0.4726 | 21.0 | 6552 | 0.6157 | 0.7437 |
0.4726 | 22.0 | 6864 | 0.6498 | 0.7076 |
0.465 | 23.0 | 7176 | 0.6576 | 0.7292 |
0.465 | 24.0 | 7488 | 0.5731 | 0.7184 |
0.4375 | 25.0 | 7800 | 0.7370 | 0.7220 |
0.4182 | 26.0 | 8112 | 0.5957 | 0.7148 |
0.4182 | 27.0 | 8424 | 0.6041 | 0.7256 |
0.4008 | 28.0 | 8736 | 0.5790 | 0.7184 |
0.392 | 29.0 | 9048 | 0.6321 | 0.7329 |
0.392 | 30.0 | 9360 | 0.6253 | 0.7148 |
0.3691 | 31.0 | 9672 | 0.6031 | 0.7329 |
0.3691 | 32.0 | 9984 | 0.5903 | 0.7148 |
0.3659 | 33.0 | 10296 | 0.6663 | 0.7329 |
0.3375 | 34.0 | 10608 | 0.6000 | 0.7292 |
0.3375 | 35.0 | 10920 | 0.5734 | 0.7256 |
0.3372 | 36.0 | 11232 | 0.6547 | 0.7329 |
0.3242 | 37.0 | 11544 | 0.6508 | 0.7401 |
0.3242 | 38.0 | 11856 | 0.6472 | 0.7365 |
0.3199 | 39.0 | 12168 | 0.6785 | 0.7365 |
0.3199 | 40.0 | 12480 | 0.6019 | 0.7365 |
0.3014 | 41.0 | 12792 | 0.5783 | 0.7329 |
0.3011 | 42.0 | 13104 | 0.6245 | 0.7329 |
0.3011 | 43.0 | 13416 | 0.6497 | 0.7292 |
0.2909 | 44.0 | 13728 | 0.6170 | 0.7365 |
0.2725 | 45.0 | 14040 | 0.6515 | 0.7437 |
0.2725 | 46.0 | 14352 | 0.6511 | 0.7365 |
0.286 | 47.0 | 14664 | 0.6303 | 0.7292 |
0.286 | 48.0 | 14976 | 0.6408 | 0.7365 |
0.2713 | 49.0 | 15288 | 0.7056 | 0.7292 |
0.2574 | 50.0 | 15600 | 0.6540 | 0.7365 |
0.2574 | 51.0 | 15912 | 0.5996 | 0.7256 |
0.2735 | 52.0 | 16224 | 0.6616 | 0.7329 |
0.2646 | 53.0 | 16536 | 0.6601 | 0.7365 |
0.2646 | 54.0 | 16848 | 0.6284 | 0.7329 |
0.2494 | 55.0 | 17160 | 0.6420 | 0.7329 |
0.2494 | 56.0 | 17472 | 0.6434 | 0.7401 |
0.2512 | 57.0 | 17784 | 0.6324 | 0.7437 |
0.2452 | 58.0 | 18096 | 0.6028 | 0.7365 |
0.2452 | 59.0 | 18408 | 0.6412 | 0.7401 |
0.2491 | 60.0 | 18720 | 0.6377 | 0.7401 |
Framework versions
- Transformers 4.26.1
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
- 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.