| epoch train_loss train_f1 train_exact dev_loss dev_f1 dev_exact test_loss test_f1 test_exact | |
| 1 0.7719278225432271 79.52947448859554 64.94490358126721 0.37337882444262505 91.59556560655463 84.61538461538461 0 0 0 | |
| 2 0.20413360582745593 94.44341525725213 89.60055096418733 0.33021537711222965 94.43942809327424 90.10989010989012 0 0 0 | |
| 3 0.0944768785316821 97.6436161742433 95.59228650137742 0.4031308473398288 92.82780711352142 89.01098901098901 0 0 0 | |
| 4 0.05001066402410683 98.68079576811634 97.38292011019284 0.36196490765238803 94.18621294671715 90.10989010989012 0 0 0 | |
| 5 0.030056177744763376 99.31963436642582 98.4159779614325 0.3473975524927179 94.3366157651872 90.38461538461539 0 0 0 | |
| 6 0.036456696730253854 98.9240347194893 98.14049586776859 0.4160988898947835 94.99857285571572 90.38461538461539 0 0 0 | |
| 7 0.020396103414342455 99.37515844753989 98.89807162534436 0.4281040045122306 94.67937816152103 92.03296703296704 0 0 0 | |
| 8 0.013906667820739804 99.59074213206443 99.31129476584022 0.3941942236075799 94.8467626695649 90.93406593406593 0 0 0 | |
| 9 0.01627988931242093 99.77043158861342 99.72451790633609 0.41736091890682775 94.61145996860283 91.20879120879121 0 0 0 | |
| 10 0.022256465652785468 99.17724536893198 98.55371900826447 0.4191143736243248 95.14768378213759 91.75824175824175 0 0 0 | |