20230824164051
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: 0.8794
- Accuracy: 0.7437
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: 0.005
- train_batch_size: 16
- eval_batch_size: 8
- seed: 11
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 80.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
No log | 1.0 | 156 | 0.6969 | 0.5307 |
No log | 2.0 | 312 | 1.5361 | 0.4693 |
No log | 3.0 | 468 | 0.6771 | 0.5235 |
0.9695 | 4.0 | 624 | 0.6518 | 0.5776 |
0.9695 | 5.0 | 780 | 0.6275 | 0.5921 |
0.9695 | 6.0 | 936 | 0.6502 | 0.5668 |
0.8132 | 7.0 | 1092 | 0.8188 | 0.6137 |
0.8132 | 8.0 | 1248 | 0.6405 | 0.6570 |
0.8132 | 9.0 | 1404 | 0.5421 | 0.7076 |
0.7231 | 10.0 | 1560 | 0.7011 | 0.6751 |
0.7231 | 11.0 | 1716 | 0.6935 | 0.5993 |
0.7231 | 12.0 | 1872 | 0.5169 | 0.7365 |
0.6369 | 13.0 | 2028 | 0.5523 | 0.7329 |
0.6369 | 14.0 | 2184 | 0.5481 | 0.7292 |
0.6369 | 15.0 | 2340 | 0.7431 | 0.6606 |
0.6369 | 16.0 | 2496 | 0.6122 | 0.6787 |
0.5638 | 17.0 | 2652 | 0.5637 | 0.6931 |
0.5638 | 18.0 | 2808 | 0.5423 | 0.7437 |
0.5638 | 19.0 | 2964 | 0.5347 | 0.7401 |
0.5228 | 20.0 | 3120 | 1.6782 | 0.6354 |
0.5228 | 21.0 | 3276 | 0.7799 | 0.6715 |
0.5228 | 22.0 | 3432 | 0.6873 | 0.7581 |
0.4829 | 23.0 | 3588 | 0.6712 | 0.7329 |
0.4829 | 24.0 | 3744 | 0.7390 | 0.7329 |
0.4829 | 25.0 | 3900 | 0.6802 | 0.7509 |
0.4251 | 26.0 | 4056 | 0.5530 | 0.7076 |
0.4251 | 27.0 | 4212 | 0.6421 | 0.7112 |
0.4251 | 28.0 | 4368 | 0.9956 | 0.6859 |
0.395 | 29.0 | 4524 | 0.6741 | 0.7545 |
0.395 | 30.0 | 4680 | 0.8871 | 0.7437 |
0.395 | 31.0 | 4836 | 0.9265 | 0.7040 |
0.395 | 32.0 | 4992 | 0.7189 | 0.7401 |
0.3336 | 33.0 | 5148 | 1.1324 | 0.7040 |
0.3336 | 34.0 | 5304 | 0.8782 | 0.7437 |
0.3336 | 35.0 | 5460 | 0.7878 | 0.7329 |
0.3015 | 36.0 | 5616 | 1.1890 | 0.7040 |
0.3015 | 37.0 | 5772 | 1.2719 | 0.7112 |
0.3015 | 38.0 | 5928 | 1.3208 | 0.6931 |
0.2669 | 39.0 | 6084 | 0.9818 | 0.7437 |
0.2669 | 40.0 | 6240 | 0.8321 | 0.7292 |
0.2669 | 41.0 | 6396 | 0.8419 | 0.7292 |
0.2429 | 42.0 | 6552 | 0.9276 | 0.7365 |
0.2429 | 43.0 | 6708 | 0.9748 | 0.7401 |
0.2429 | 44.0 | 6864 | 0.8934 | 0.7473 |
0.2131 | 45.0 | 7020 | 0.9008 | 0.7473 |
0.2131 | 46.0 | 7176 | 1.0459 | 0.7437 |
0.2131 | 47.0 | 7332 | 1.0222 | 0.7256 |
0.2131 | 48.0 | 7488 | 0.9317 | 0.7545 |
0.1962 | 49.0 | 7644 | 0.8401 | 0.7473 |
0.1962 | 50.0 | 7800 | 0.9513 | 0.7401 |
0.1962 | 51.0 | 7956 | 0.9327 | 0.7401 |
0.1794 | 52.0 | 8112 | 1.0218 | 0.7509 |
0.1794 | 53.0 | 8268 | 1.1332 | 0.7473 |
0.1794 | 54.0 | 8424 | 0.8851 | 0.7365 |
0.1566 | 55.0 | 8580 | 0.8323 | 0.7473 |
0.1566 | 56.0 | 8736 | 0.8375 | 0.7437 |
0.1566 | 57.0 | 8892 | 0.8490 | 0.7509 |
0.15 | 58.0 | 9048 | 0.9740 | 0.7509 |
0.15 | 59.0 | 9204 | 1.1271 | 0.7473 |
0.15 | 60.0 | 9360 | 1.1190 | 0.7437 |
0.1377 | 61.0 | 9516 | 1.0394 | 0.7509 |
0.1377 | 62.0 | 9672 | 0.9735 | 0.7509 |
0.1377 | 63.0 | 9828 | 0.9987 | 0.7437 |
0.1377 | 64.0 | 9984 | 0.9496 | 0.7473 |
0.1283 | 65.0 | 10140 | 1.0721 | 0.7365 |
0.1283 | 66.0 | 10296 | 0.8997 | 0.7617 |
0.1283 | 67.0 | 10452 | 1.0014 | 0.7581 |
0.1212 | 68.0 | 10608 | 1.0382 | 0.7509 |
0.1212 | 69.0 | 10764 | 0.9417 | 0.7437 |
0.1212 | 70.0 | 10920 | 0.9328 | 0.7437 |
0.1101 | 71.0 | 11076 | 0.9084 | 0.7509 |
0.1101 | 72.0 | 11232 | 0.9051 | 0.7545 |
0.1101 | 73.0 | 11388 | 0.8080 | 0.7581 |
0.1147 | 74.0 | 11544 | 0.9505 | 0.7437 |
0.1147 | 75.0 | 11700 | 0.8757 | 0.7437 |
0.1147 | 76.0 | 11856 | 0.9067 | 0.7509 |
0.1095 | 77.0 | 12012 | 0.8988 | 0.7473 |
0.1095 | 78.0 | 12168 | 0.8956 | 0.7473 |
0.1095 | 79.0 | 12324 | 0.8622 | 0.7473 |
0.1095 | 80.0 | 12480 | 0.8794 | 0.7437 |
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.