Filtered Corpus Training
Collection
All models from the paper "Filtered Corpus Training (FiCT) Shows...". Naming convention: `{filter}-{model}-{seed}`.
•
39 items
•
Updated
This model is a fine-tuned version of on the None dataset. It achieves the following results on the evaluation set:
More information needed
More information needed
More information needed
The following hyperparameters were used during training:
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
4.7774 | 0.03 | 76320 | 4.7429 |
4.4932 | 1.03 | 152640 | 4.4638 |
4.3479 | 0.03 | 228960 | 4.3308 |
4.2625 | 1.03 | 305280 | 4.2490 |
4.1981 | 0.03 | 381600 | 4.1928 |
4.1521 | 1.03 | 457920 | 4.1530 |
4.1154 | 0.03 | 534240 | 4.1230 |
4.0805 | 1.03 | 610560 | 4.0988 |
4.0514 | 0.03 | 686880 | 4.0793 |
4.0259 | 1.03 | 763200 | 4.0640 |
4.0056 | 0.03 | 839520 | 4.0506 |
3.9903 | 1.03 | 915840 | 4.0404 |
3.9761 | 0.03 | 992160 | 4.0308 |
3.9565 | 1.03 | 1068480 | 4.0234 |
3.9472 | 0.03 | 1144800 | 4.0168 |
3.9352 | 1.03 | 1221120 | 4.0114 |
3.9245 | 0.03 | 1297440 | 4.0061 |
3.9137 | 1.03 | 1373760 | 4.0013 |
3.9036 | 0.03 | 1450080 | 3.9982 |
3.8998 | 0.03 | 1526400 | 3.9949 |
3.8957 | 1.03 | 1602720 | 3.9922 |
3.891 | 0.03 | 1679040 | 3.9897 |
3.8872 | 1.03 | 1755360 | 3.9876 |
3.8784 | 0.03 | 1831680 | 3.9853 |
3.8704 | 1.03 | 1908000 | 3.9831 |
3.8615 | 0.03 | 1984320 | 3.9815 |
3.8584 | 0.03 | 2060640 | 3.9799 |
3.8554 | 1.03 | 2136960 | 3.9784 |
3.8507 | 0.03 | 2213280 | 3.9773 |
3.8436 | 0.03 | 2289600 | 3.9763 |
3.8417 | 1.03 | 2365920 | 3.9754 |
3.8366 | 0.03 | 2442240 | 3.9742 |
3.8328 | 1.03 | 2518560 | 3.9736 |
3.8293 | 0.03 | 2594880 | 3.9726 |
3.8258 | 0.03 | 2671200 | 3.9719 |
3.8263 | 0.03 | 2747520 | 3.9714 |
3.8265 | 1.03 | 2823840 | 3.9709 |
3.8291 | 0.03 | 2900160 | 3.9704 |
3.8271 | 1.03 | 2976480 | 3.9701 |
3.8234 | 0.02 | 3052726 | 3.9698 |