checkpoint @ game=3550 steps=3550

training_log.txt

@@ -3849,3 +3849,13 @@
 [2026-06-22 11:44:56] game= 3538 | 1/2-1/2 | moves= 58 | buf=50,000 | steps=3,538 | loss=9.1143 (p=8.4493 v=0.6651) | game=5.8s train=87.1s
 [2026-06-22 11:46:33] game= 3539 | 1/2-1/2 | moves= 79 | buf=50,000 | steps=3,539 | loss=8.9398 (p=8.4493 v=0.4905) | game=7.9s train=89.3s
 [2026-06-22 11:48:05] game= 3540 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=3,540 | loss=8.9131 (p=8.4493 v=0.4638) | game=5.3s train=87.1s
+[2026-06-22 11:49:41] game= 3541 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=3,541 | loss=8.8550 (p=8.4493 v=0.4057) | game=5.4s train=87.2s
+[2026-06-22 11:51:13] game= 3542 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,542 | loss=8.7865 (p=8.4493 v=0.3372) | game=5.1s train=87.2s
+[2026-06-22 11:52:47] game= 3543 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=3,543 | loss=8.7790 (p=8.4493 v=0.3297) | game=6.1s train=87.0s
+[2026-06-22 11:54:23] game= 3544 | 1/2-1/2 | moves= 68 | buf=50,000 | steps=3,544 | loss=8.7676 (p=8.4493 v=0.3183) | game=6.8s train=89.3s
+[2026-06-22 11:55:55] game= 3545 | 1/2-1/2 | moves= 56 | buf=50,000 | steps=3,545 | loss=8.6878 (p=8.4493 v=0.2385) | game=5.6s train=87.1s
+[2026-06-22 11:57:29] game= 3546 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=3,546 | loss=8.7205 (p=8.4493 v=0.2712) | game=6.5s train=87.3s
+[2026-06-22 11:59:04] game= 3547 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=3,547 | loss=8.6556 (p=8.4493 v=0.2063) | game=8.1s train=87.2s
+[2026-06-22 12:00:37] game= 3548 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=3,548 | loss=8.6572 (p=8.4493 v=0.2079) | game=5.7s train=87.2s
+[2026-06-22 12:02:12] game= 3549 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=3,549 | loss=8.6976 (p=8.4493 v=0.2483) | game=7.6s train=87.2s
+[2026-06-22 12:03:45] game= 3550 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=3,550 | loss=8.6231 (p=8.4493 v=0.1738) | game=5.5s train=87.2s