phi4star
done · peak unseen_mismatch 94.5
Referenced by
α schedule
α 5.0 ×25 α 5.0 ×25 α 5.0 ×25 α 1.0634 ×25 α 1.0536 ×25 α 5.0 ×25 α 5.0 ×25 α 5.0 ×25 α 0.9043 ×25 α 0.8215 ×25 α 0.7122 ×25 α 3.084 ×25 α 2.1282 ×25 α 0.1755 ×25 α 0.1755 ×50
Figures
Results
accuracy by training iteration, mean ± sd over seeds · schedule ends at α = 0.1755
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 | 57.0 ± 8.2 | 52.4 ± 4.4 | 53.8 ± 7.6 | 50.4 ± 3.4 |
| 110 | 59.7 ± 11.6 | 53.0 ± 5.8 | 56.8 ± 11.2 | 51.2 ± 4.4 |
| 120 | 64.0 ± 15.2 | 54.4 ± 7.1 | 61.3 ± 15.3 | 52.6 ± 6.1 |
| 130 | 70.6 ± 17.5 | 57.1 ± 10.4 | 67.8 ± 18.1 | 54.6 ± 9.3 |
| 140 | 77.1 ± 17.5 | 64.3 ± 13.6 | 73.7 ± 18.4 | 60.8 ± 13.0 |
| 150 | 82.2 ± 16.2 | 72.1 ± 14.4 | 78.3 ± 17.7 | 68.0 ± 14.3 |
| 160 | 85.5 ± 13.7 | 78.2 ± 12.9 | 80.8 ± 15.7 | 72.7 ± 13.6 |
| 170 | 87.6 ± 11.4 | 82.1 ± 11.1 | 81.9 ± 13.9 | 75.5 ± 12.4 |
| 180 | 88.5 ± 9.0 | 84.1 ± 8.9 | 81.8 ± 11.6 | 76.5 ± 10.8 |
| 190 | 88.8 ± 7.8 | 85.6 ± 7.5 | 81.7 ± 10.5 | 76.9 ± 9.8 |
| 200 | 88.4 ± 6.6 | 85.8 ± 6.4 | 79.7 ± 9.8 | 75.7 ± 8.8 |
| 210 | 92.7 ± 7.0 | 90.1 ± 7.3 | 87.3 ± 10.2 | 83.4 ± 10.3 |
| 220 | 96.0 ± 7.2 | 93.5 ± 8.1 | 93.3 ± 9.7 | 89.5 ± 10.1 |
| 230 | 95.6 ± 9.4 | 91.8 ± 10.6 | 93.5 ± 11.0 | 88.3 ± 12.1 |
| 240 | 93.7 ± 11.8 | 85.2 ± 12.8 | 92.4 ± 13.1 | 81.8 ± 12.6 |
| 250 | 95.1 ± 11.0 | 80.8 ± 14.6 | 94.0 ± 12.7 | 78.0 ± 14.1 |
| 260 | 95.5 ± 11.1 | 79.9 ± 22.7 | 94.6 ± 13.8 | 76.7 ± 21.7 |
| 270 | 95.4 ± 9.9 | 66.8 ± 27.9 | 93.8 ± 13.6 | 66.6 ± 26.9 |
| 280 | 96.3 ± 7.3 | 71.8 ± 35.2 | 95.6 ± 8.8 | 70.2 ± 34.0 |
| 290 | 98.7 ± 1.8 | 72.6 ± 26.1 | 98.1 ± 2.7 | 70.2 ± 24.9 |
| 300 | 98.3 ± 1.7 | 87.4 ± 0.8 | 98.0 ± 1.9 | 87.7 ± 2.6 |
| 310 | 98.7 ± 1.3 | 93.7 ± 1.7 | 98.0 ± 2.0 | 92.9 ± 2.2 |
| 320 | 98.2 ± 0.0 | 95.9 ± 0.0 | 97.0 ± 0.0 | 94.5 ± 0.0 |