free-61f53dcb
done · peak unseen_mismatch 97.0
α schedule
α 4.898 ×13 α 4.498 ×25 α 4.634 ×18 α 2.667 ×22 α 2.878 ×55 α 2.445 ×26 α 1.997 ×27 α 0.2762 ×95 α 3.231 ×19 α 3.231 ×100
Results
accuracy by training iteration, mean ± sd over seeds · schedule ends at α = 3.231
seen_* probed at reference α = 1.4 · unseen_* = held-out pairings, uniform (α-independent) · 1000 pairs per condition
| iteration | seen_match | seen_mismatch | unseen_match | unseen_mismatch |
|---|---|---|---|---|
| 100 | 59.2 ± 9.1 | 55.6 ± 6.5 | 54.4 ± 7.5 | 51.5 ± 4.1 |
| 110 | 63.4 ± 10.3 | 59.1 ± 8.9 | 56.9 ± 8.8 | 53.4 ± 5.9 |
| 120 | 68.8 ± 12.4 | 64.0 ± 10.9 | 60.5 ± 10.9 | 56.1 ± 7.8 |
| 130 | 73.9 ± 13.0 | 68.9 ± 11.8 | 64.2 ± 12.7 | 59.4 ± 9.3 |
| 140 | 77.8 ± 12.7 | 73.4 ± 11.9 | 67.3 ± 13.2 | 63.0 ± 10.3 |
| 150 | 81.6 ± 11.4 | 77.2 ± 11.7 | 70.2 ± 12.9 | 65.8 ± 10.5 |
| 160 | 84.5 ± 9.5 | 81.0 ± 10.0 | 73.1 ± 11.5 | 68.6 ± 9.9 |
| 170 | 87.9 ± 8.7 | 84.5 ± 9.0 | 77.0 ± 10.8 | 72.2 ± 9.7 |
| 180 | 90.4 ± 6.8 | 87.6 ± 7.4 | 80.0 ± 9.6 | 75.4 ± 8.9 |
| 190 | 92.6 ± 6.1 | 89.5 ± 7.6 | 83.6 ± 8.6 | 78.7 ± 8.6 |
| 200 | 97.0 ± 5.1 | 94.1 ± 8.2 | 94.3 ± 7.2 | 89.7 ± 10.0 |
| 210 | 98.3 ± 3.2 | 88.0 ± 17.7 | 97.4 ± 4.6 | 85.8 ± 17.8 |
| 220 | 99.8 ± 0.5 | 67.0 ± 22.9 | 99.4 ± 0.9 | 66.0 ± 22.7 |
| 230 | 99.9 ± 0.2 | 57.5 ± 21.7 | 99.7 ± 0.9 | 56.9 ± 21.3 |
| 240 | 100.0 ± 0.0 | 55.5 ± 28.9 | 100.0 ± 0.0 | 54.9 ± 28.4 |
| 250 | 100.0 ± 0.0 | 57.1 ± 27.4 | 100.0 ± 0.0 | 56.2 ± 26.7 |
| 260 | 100.0 ± 0.0 | 47.6 ± 20.5 | 100.0 ± 0.0 | 47.6 ± 20.4 |
| 270 | 100.0 ± 0.0 | 50.0 ± 20.0 | 100.0 ± 0.0 | 49.4 ± 19.9 |
| 280 | 100.0 ± 0.0 | 51.9 ± 16.6 | 100.0 ± 0.0 | 52.5 ± 14.8 |
| 290 | 100.0 ± 0.0 | 63.1 ± 27.2 | 100.0 ± 0.0 | 63.8 ± 27.2 |
| 300 | 100.0 ± 0.0 | 74.1 ± 20.4 | 100.0 ± 0.0 | 74.2 ± 19.8 |
| 310 | 100.0 ± 0.0 | 74.7 ± 20.9 | 100.0 ± 0.0 | 75.7 ± 19.8 |
| 320 | 100.0 ± 0.0 | 78.0 ± 17.1 | 100.0 ± 0.0 | 78.2 ± 17.6 |
| 330 | 100.0 ± 0.0 | 73.9 ± 0.0 | 100.0 ± 0.0 | 74.2 ± 0.0 |
| 340 | 100.0 ± 0.0 | 75.5 ± 0.0 | 100.0 ± 0.0 | 77.6 ± 0.0 |
| 350 | 100.0 ± 0.0 | 95.5 ± 0.0 | 100.0 ± 0.0 | 97.0 ± 0.0 |