checkpoint @ game=3520 steps=3520

training_log.txt

@@ -3819,3 +3819,13 @@
 [2026-06-22 10:57:59] game= 3508 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=3,508 | loss=8.7361 (p=8.4493 v=0.2868) | game=8.1s train=86.6s
 [2026-06-22 10:59:32] game= 3509 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=3,509 | loss=8.8170 (p=8.4493 v=0.3677) | game=6.5s train=86.6s
 [2026-06-22 11:01:06] game= 3510 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,510 | loss=8.7754 (p=8.4493 v=0.3261) | game=6.9s train=86.6s
+[2026-06-22 11:02:42] game= 3511 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=3,511 | loss=8.7265 (p=8.4493 v=0.2772) | game=6.4s train=86.7s
+[2026-06-22 11:04:14] game= 3512 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=3,512 | loss=8.6969 (p=8.4493 v=0.2477) | game=5.4s train=86.7s
+[2026-06-22 11:05:48] game= 3513 | 1/2-1/2 | moves= 47 | buf=50,000 | steps=3,513 | loss=8.6747 (p=8.4493 v=0.2254) | game=4.7s train=89.3s
+[2026-06-22 11:07:20] game= 3514 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=3,514 | loss=8.6593 (p=8.4493 v=0.2100) | game=5.4s train=87.0s
+[2026-06-22 11:08:54] game= 3515 | 1/2-1/2 | moves= 64 | buf=50,000 | steps=3,515 | loss=8.6868 (p=8.4493 v=0.2375) | game=6.4s train=86.9s
+[2026-06-22 11:10:28] game= 3516 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,516 | loss=8.5826 (p=8.4493 v=0.1333) | game=5.2s train=89.5s
+[2026-06-22 11:12:01] game= 3517 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=3,517 | loss=8.6513 (p=8.4493 v=0.2020) | game=5.8s train=87.0s
+[2026-06-22 11:13:35] game= 3518 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=3,518 | loss=8.6291 (p=8.4493 v=0.1798) | game=6.5s train=87.0s
+[2026-06-22 11:15:07] game= 3519 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=3,519 | loss=8.6163 (p=8.4493 v=0.1671) | game=5.3s train=86.9s
+[2026-06-22 11:16:46] game= 3520 | 1/2-1/2 | moves= 66 | buf=50,000 | steps=3,520 | loss=9.2202 (p=8.4493 v=0.7709) | game=12.0s train=86.9s