<!-- 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. -->
20230829233517
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: 0.3325
- Accuracy: 0.6731
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: 0.007
- train_batch_size: 16
- eval_batch_size: 8
- seed: 44
- 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 | 35 | 0.8923 | 0.6058 |
| No log | 2.0 | 70 | 0.7563 | 0.5962 |
| No log | 3.0 | 105 | 0.4459 | 0.5962 |
| No log | 4.0 | 140 | 0.4576 | 0.5577 |
| No log | 5.0 | 175 | 0.3812 | 0.5865 |
| No log | 6.0 | 210 | 0.4130 | 0.4038 |
| No log | 7.0 | 245 | 0.3742 | 0.5962 |
| No log | 8.0 | 280 | 0.7058 | 0.4135 |
| No log | 9.0 | 315 | 0.4374 | 0.4231 |
| No log | 10.0 | 350 | 0.3737 | 0.5673 |
| No log | 11.0 | 385 | 0.6869 | 0.3846 |
| No log | 12.0 | 420 | 0.4462 | 0.625 |
| No log | 13.0 | 455 | 0.4324 | 0.6442 |
| No log | 14.0 | 490 | 1.1163 | 0.3654 |
| 0.6113 | 15.0 | 525 | 0.3426 | 0.5769 |
| 0.6113 | 16.0 | 560 | 0.5857 | 0.6346 |
| 0.6113 | 17.0 | 595 | 0.6489 | 0.375 |
| 0.6113 | 18.0 | 630 | 0.3907 | 0.6346 |
| 0.6113 | 19.0 | 665 | 0.3395 | 0.5769 |
| 0.6113 | 20.0 | 700 | 0.5898 | 0.3654 |
| 0.6113 | 21.0 | 735 | 0.3773 | 0.6346 |
| 0.6113 | 22.0 | 770 | 0.3323 | 0.6731 |
| 0.6113 | 23.0 | 805 | 0.3810 | 0.4231 |
| 0.6113 | 24.0 | 840 | 0.4134 | 0.6346 |
| 0.6113 | 25.0 | 875 | 0.3447 | 0.5673 |
| 0.6113 | 26.0 | 910 | 0.3419 | 0.6346 |
| 0.6113 | 27.0 | 945 | 0.3443 | 0.5769 |
| 0.6113 | 28.0 | 980 | 0.5933 | 0.6346 |
| 0.5052 | 29.0 | 1015 | 0.6428 | 0.3654 |
| 0.5052 | 30.0 | 1050 | 0.3407 | 0.6346 |
| 0.5052 | 31.0 | 1085 | 0.3944 | 0.3942 |
| 0.5052 | 32.0 | 1120 | 0.3514 | 0.4808 |
| 0.5052 | 33.0 | 1155 | 0.4055 | 0.4135 |
| 0.5052 | 34.0 | 1190 | 0.3472 | 0.6442 |
| 0.5052 | 35.0 | 1225 | 0.4300 | 0.3942 |
| 0.5052 | 36.0 | 1260 | 0.4384 | 0.6346 |
| 0.5052 | 37.0 | 1295 | 0.3352 | 0.6346 |
| 0.5052 | 38.0 | 1330 | 0.3501 | 0.5096 |
| 0.5052 | 39.0 | 1365 | 0.5508 | 0.6346 |
| 0.5052 | 40.0 | 1400 | 0.4123 | 0.6346 |
| 0.5052 | 41.0 | 1435 | 0.3578 | 0.6346 |
| 0.5052 | 42.0 | 1470 | 0.4021 | 0.6346 |
| 0.4366 | 43.0 | 1505 | 0.5536 | 0.3654 |
| 0.4366 | 44.0 | 1540 | 0.3327 | 0.6346 |
| 0.4366 | 45.0 | 1575 | 0.3289 | 0.6346 |
| 0.4366 | 46.0 | 1610 | 0.4193 | 0.4135 |
| 0.4366 | 47.0 | 1645 | 0.3673 | 0.4327 |
| 0.4366 | 48.0 | 1680 | 0.4301 | 0.6346 |
| 0.4366 | 49.0 | 1715 | 0.3367 | 0.5865 |
| 0.4366 | 50.0 | 1750 | 0.3809 | 0.3942 |
| 0.4366 | 51.0 | 1785 | 0.4183 | 0.4135 |
| 0.4366 | 52.0 | 1820 | 0.3500 | 0.4904 |
| 0.4366 | 53.0 | 1855 | 0.3852 | 0.3942 |
| 0.4366 | 54.0 | 1890 | 0.3584 | 0.4712 |
| 0.4366 | 55.0 | 1925 | 0.3405 | 0.6346 |
| 0.4366 | 56.0 | 1960 | 0.3676 | 0.4327 |
| 0.4366 | 57.0 | 1995 | 0.3764 | 0.4423 |
| 0.4036 | 58.0 | 2030 | 0.3381 | 0.5962 |
| 0.4036 | 59.0 | 2065 | 0.3366 | 0.6346 |
| 0.4036 | 60.0 | 2100 | 0.3408 | 0.6346 |
| 0.4036 | 61.0 | 2135 | 0.3950 | 0.6346 |
| 0.4036 | 62.0 | 2170 | 0.3277 | 0.6346 |
| 0.4036 | 63.0 | 2205 | 0.3299 | 0.6346 |
| 0.4036 | 64.0 | 2240 | 0.3470 | 0.5 |
| 0.4036 | 65.0 | 2275 | 0.3521 | 0.6346 |
| 0.4036 | 66.0 | 2310 | 0.3314 | 0.6731 |
| 0.4036 | 67.0 | 2345 | 0.3530 | 0.4808 |
| 0.4036 | 68.0 | 2380 | 0.3316 | 0.6635 |
| 0.4036 | 69.0 | 2415 | 0.3416 | 0.5481 |
| 0.4036 | 70.0 | 2450 | 0.3685 | 0.6346 |
| 0.4036 | 71.0 | 2485 | 0.3300 | 0.6346 |
| 0.3777 | 72.0 | 2520 | 0.3310 | 0.6346 |
| 0.3777 | 73.0 | 2555 | 0.3291 | 0.6538 |
| 0.3777 | 74.0 | 2590 | 0.3346 | 0.6058 |
| 0.3777 | 75.0 | 2625 | 0.3288 | 0.6442 |
| 0.3777 | 76.0 | 2660 | 0.3304 | 0.6442 |
| 0.3777 | 77.0 | 2695 | 0.3906 | 0.4327 |
| 0.3777 | 78.0 | 2730 | 0.3314 | 0.6827 |
| 0.3777 | 79.0 | 2765 | 0.3298 | 0.6346 |
| 0.3777 | 80.0 | 2800 | 0.3325 | 0.6731 |
Framework versions
- Transformers 4.26.1
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
