t5-small-sci
This model is a fine-tuned version of google-t5/t5-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.1430
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: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 72
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.1983 | 1.0 | 1452 | 3.8284 |
| 3.9959 | 2.0 | 2904 | 3.7109 |
| 3.9016 | 3.0 | 4356 | 3.6394 |
| 3.8153 | 4.0 | 5808 | 3.5881 |
| 3.7842 | 5.0 | 7260 | 3.5475 |
| 3.7176 | 6.0 | 8712 | 3.5119 |
| 3.6995 | 7.0 | 10164 | 3.4844 |
| 3.6475 | 8.0 | 11616 | 3.4581 |
| 3.6184 | 9.0 | 13068 | 3.4346 |
| 3.589 | 10.0 | 14520 | 3.4164 |
| 3.5805 | 11.0 | 15972 | 3.3966 |
| 3.5591 | 12.0 | 17424 | 3.3813 |
| 3.5254 | 13.0 | 18876 | 3.3663 |
| 3.5144 | 14.0 | 20328 | 3.3537 |
| 3.4929 | 15.0 | 21780 | 3.3406 |
| 3.4783 | 16.0 | 23232 | 3.3298 |
| 3.4564 | 17.0 | 24684 | 3.3153 |
| 3.4502 | 18.0 | 26136 | 3.3069 |
| 3.4252 | 19.0 | 27588 | 3.2974 |
| 3.403 | 20.0 | 29040 | 3.2892 |
| 3.3887 | 21.0 | 30492 | 3.2806 |
| 3.4035 | 22.0 | 31944 | 3.2715 |
| 3.3774 | 23.0 | 33396 | 3.2633 |
| 3.3555 | 24.0 | 34848 | 3.2575 |
| 3.3592 | 25.0 | 36300 | 3.2511 |
| 3.362 | 26.0 | 37752 | 3.2449 |
| 3.3483 | 27.0 | 39204 | 3.2397 |
| 3.3345 | 28.0 | 40656 | 3.2322 |
| 3.3254 | 29.0 | 42108 | 3.2275 |
| 3.3312 | 30.0 | 43560 | 3.2225 |
| 3.3051 | 31.0 | 45012 | 3.2186 |
| 3.303 | 32.0 | 46464 | 3.2134 |
| 3.2969 | 33.0 | 47916 | 3.2098 |
| 3.2713 | 34.0 | 49368 | 3.2047 |
| 3.2745 | 35.0 | 50820 | 3.2009 |
| 3.2677 | 36.0 | 52272 | 3.1974 |
| 3.2622 | 37.0 | 53724 | 3.1938 |
| 3.2549 | 38.0 | 55176 | 3.1901 |
| 3.2578 | 39.0 | 56628 | 3.1868 |
| 3.2549 | 40.0 | 58080 | 3.1846 |
| 3.2357 | 41.0 | 59532 | 3.1816 |
| 3.2582 | 42.0 | 60984 | 3.1790 |
| 3.2327 | 43.0 | 62436 | 3.1760 |
| 3.232 | 44.0 | 63888 | 3.1745 |
| 3.218 | 45.0 | 65340 | 3.1716 |
| 3.2248 | 46.0 | 66792 | 3.1686 |
| 3.2193 | 47.0 | 68244 | 3.1667 |
| 3.2095 | 48.0 | 69696 | 3.1653 |
| 3.2036 | 49.0 | 71148 | 3.1630 |
| 3.1974 | 50.0 | 72600 | 3.1613 |
| 3.2142 | 51.0 | 74052 | 3.1591 |
| 3.1993 | 52.0 | 75504 | 3.1571 |
| 3.202 | 53.0 | 76956 | 3.1552 |
| 3.2168 | 54.0 | 78408 | 3.1552 |
| 3.188 | 55.0 | 79860 | 3.1528 |
| 3.2123 | 56.0 | 81312 | 3.1519 |
| 3.1947 | 57.0 | 82764 | 3.1511 |
| 3.1835 | 58.0 | 84216 | 3.1497 |
| 3.1836 | 59.0 | 85668 | 3.1492 |
| 3.1791 | 60.0 | 87120 | 3.1481 |
| 3.1732 | 61.0 | 88572 | 3.1474 |
| 3.1835 | 62.0 | 90024 | 3.1459 |
| 3.1682 | 63.0 | 91476 | 3.1463 |
| 3.1693 | 64.0 | 92928 | 3.1450 |
| 3.1763 | 65.0 | 94380 | 3.1449 |
| 3.1724 | 66.0 | 95832 | 3.1445 |
| 3.1616 | 67.0 | 97284 | 3.1442 |
| 3.1662 | 68.0 | 98736 | 3.1435 |
| 3.1593 | 69.0 | 100188 | 3.1432 |
| 3.1644 | 70.0 | 101640 | 3.1435 |
| 3.1583 | 71.0 | 103092 | 3.1431 |
| 3.1395 | 72.0 | 104544 | 3.1430 |
Framework versions
- Transformers 4.43.4
- Pytorch 1.13.1
- Datasets 2.12.0
- Tokenizers 0.19.1