20230903070300
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: 0.8203
- Accuracy: 0.6599
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: 0.0002
- 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 | 340 | 0.7251 | 0.5063 |
0.7449 | 2.0 | 680 | 0.7348 | 0.5 |
0.7388 | 3.0 | 1020 | 0.7304 | 0.5 |
0.7388 | 4.0 | 1360 | 0.7639 | 0.5 |
0.7384 | 5.0 | 1700 | 0.7316 | 0.5 |
0.7376 | 6.0 | 2040 | 0.7268 | 0.5 |
0.7376 | 7.0 | 2380 | 0.7263 | 0.5 |
0.7328 | 8.0 | 2720 | 0.7333 | 0.5 |
0.7266 | 9.0 | 3060 | 0.7533 | 0.5 |
0.7266 | 10.0 | 3400 | 0.7247 | 0.4984 |
0.7293 | 11.0 | 3740 | 0.7290 | 0.5172 |
0.7248 | 12.0 | 4080 | 0.7539 | 0.5 |
0.7248 | 13.0 | 4420 | 0.7395 | 0.5 |
0.7255 | 14.0 | 4760 | 0.7360 | 0.5031 |
0.7271 | 15.0 | 5100 | 0.7278 | 0.5 |
0.7271 | 16.0 | 5440 | 0.7314 | 0.5094 |
0.7265 | 17.0 | 5780 | 0.7417 | 0.4984 |
0.724 | 18.0 | 6120 | 0.7263 | 0.5 |
0.724 | 19.0 | 6460 | 0.7272 | 0.5031 |
0.723 | 20.0 | 6800 | 0.7283 | 0.5172 |
0.7254 | 21.0 | 7140 | 0.7284 | 0.5047 |
0.7254 | 22.0 | 7480 | 0.7346 | 0.4984 |
0.7254 | 23.0 | 7820 | 0.7295 | 0.5125 |
0.7259 | 24.0 | 8160 | 0.7322 | 0.5047 |
0.7235 | 25.0 | 8500 | 0.7327 | 0.5172 |
0.7235 | 26.0 | 8840 | 0.7300 | 0.5172 |
0.7241 | 27.0 | 9180 | 0.7345 | 0.5016 |
0.7227 | 28.0 | 9520 | 0.7263 | 0.5172 |
0.7227 | 29.0 | 9860 | 0.7341 | 0.5016 |
0.7212 | 30.0 | 10200 | 0.7302 | 0.5125 |
0.7226 | 31.0 | 10540 | 0.7346 | 0.5078 |
0.7226 | 32.0 | 10880 | 0.7606 | 0.4702 |
0.7195 | 33.0 | 11220 | 0.7357 | 0.5063 |
0.7226 | 34.0 | 11560 | 0.7356 | 0.5031 |
0.7226 | 35.0 | 11900 | 0.7397 | 0.5063 |
0.7224 | 36.0 | 12240 | 0.7340 | 0.5157 |
0.7216 | 37.0 | 12580 | 0.7319 | 0.5047 |
0.7216 | 38.0 | 12920 | 0.7298 | 0.5141 |
0.7225 | 39.0 | 13260 | 0.7438 | 0.5016 |
0.7197 | 40.0 | 13600 | 0.7306 | 0.5047 |
0.7197 | 41.0 | 13940 | 0.7279 | 0.5125 |
0.7206 | 42.0 | 14280 | 0.7181 | 0.5502 |
0.7079 | 43.0 | 14620 | 0.7566 | 0.5862 |
0.7079 | 44.0 | 14960 | 0.7480 | 0.6254 |
0.6794 | 45.0 | 15300 | 0.6922 | 0.6630 |
0.6556 | 46.0 | 15640 | 0.7232 | 0.6223 |
0.6556 | 47.0 | 15980 | 0.6961 | 0.6458 |
0.6438 | 48.0 | 16320 | 0.7193 | 0.6458 |
0.6249 | 49.0 | 16660 | 0.6663 | 0.6693 |
0.6117 | 50.0 | 17000 | 0.8045 | 0.6191 |
0.6117 | 51.0 | 17340 | 0.6984 | 0.6630 |
0.5961 | 52.0 | 17680 | 0.6973 | 0.6646 |
0.5831 | 53.0 | 18020 | 0.7606 | 0.6348 |
0.5831 | 54.0 | 18360 | 0.7159 | 0.6614 |
0.5624 | 55.0 | 18700 | 0.7947 | 0.6426 |
0.558 | 56.0 | 19040 | 0.8629 | 0.6238 |
0.558 | 57.0 | 19380 | 0.7299 | 0.6646 |
0.5461 | 58.0 | 19720 | 0.7642 | 0.6411 |
0.5322 | 59.0 | 20060 | 0.7357 | 0.6661 |
0.5322 | 60.0 | 20400 | 0.8926 | 0.6191 |
0.5253 | 61.0 | 20740 | 0.7845 | 0.6348 |
0.5193 | 62.0 | 21080 | 0.7580 | 0.6614 |
0.5193 | 63.0 | 21420 | 0.7705 | 0.6505 |
0.5169 | 64.0 | 21760 | 0.8464 | 0.6458 |
0.5021 | 65.0 | 22100 | 0.8002 | 0.6536 |
0.5021 | 66.0 | 22440 | 0.7595 | 0.6677 |
0.487 | 67.0 | 22780 | 0.7971 | 0.6458 |
0.4977 | 68.0 | 23120 | 0.8245 | 0.6270 |
0.4977 | 69.0 | 23460 | 0.8225 | 0.6379 |
0.4822 | 70.0 | 23800 | 0.8323 | 0.6364 |
0.4802 | 71.0 | 24140 | 0.8205 | 0.6364 |
0.4802 | 72.0 | 24480 | 0.8086 | 0.6520 |
0.4779 | 73.0 | 24820 | 0.7994 | 0.6567 |
0.4801 | 74.0 | 25160 | 0.8206 | 0.6520 |
0.4706 | 75.0 | 25500 | 0.8035 | 0.6442 |
0.4706 | 76.0 | 25840 | 0.8213 | 0.6364 |
0.4738 | 77.0 | 26180 | 0.8128 | 0.6630 |
0.4687 | 78.0 | 26520 | 0.8068 | 0.6567 |
0.4687 | 79.0 | 26860 | 0.8098 | 0.6630 |
0.4598 | 80.0 | 27200 | 0.8203 | 0.6599 |
Framework versions
- Transformers 4.26.1
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
- Downloads last month
- 2
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.