checkpoint @ game=4570 steps=4570

training_log.txt

@@ -4869,3 +4869,13 @@
 [2026-06-25 04:03:43] game= 4558 | 1/2-1/2 | moves= 48 | buf=50,000 | steps=4,558 | loss=8.5932 (p=8.4493 v=0.1439) | game=5.4s train=89.7s
 [2026-06-25 04:05:23] game= 4559 | 1/2-1/2 | moves= 71 | buf=50,000 | steps=4,559 | loss=8.6226 (p=8.4493 v=0.1733) | game=10.1s train=89.8s
 [2026-06-25 04:06:58] game= 4560 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=4,560 | loss=8.6658 (p=8.4493 v=0.2165) | game=5.6s train=89.0s
+[2026-06-25 04:08:44] game= 4561 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=4,561 | loss=8.6230 (p=8.4493 v=0.1737) | game=6.4s train=91.0s
+[2026-06-25 04:10:22] game= 4562 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=4,562 | loss=9.1566 (p=8.4493 v=0.7073) | game=8.7s train=89.4s
+[2026-06-25 04:11:59] game= 4563 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=4,563 | loss=9.0284 (p=8.4493 v=0.5791) | game=6.1s train=90.9s
+[2026-06-25 04:13:38] game= 4564 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=4,564 | loss=9.0081 (p=8.4493 v=0.5588) | game=9.1s train=89.7s
+[2026-06-25 04:15:15] game= 4565 | 1/2-1/2 | moves= 73 | buf=50,000 | steps=4,565 | loss=8.8805 (p=8.4493 v=0.4312) | game=8.4s train=88.5s
+[2026-06-25 04:16:54] game= 4566 | 1/2-1/2 | moves= 69 | buf=50,000 | steps=4,566 | loss=8.7908 (p=8.4493 v=0.3415) | game=7.8s train=91.3s
+[2026-06-25 04:18:32] game= 4567 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=4,567 | loss=8.7866 (p=8.4493 v=0.3373) | game=8.9s train=89.4s
+[2026-06-25 04:20:08] game= 4568 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=4,568 | loss=8.7057 (p=8.4493 v=0.2564) | game=5.9s train=89.3s
+[2026-06-25 04:21:45] game= 4569 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=4,569 | loss=8.7001 (p=8.4493 v=0.2508) | game=7.4s train=90.2s
+[2026-06-25 04:23:20] game= 4570 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=4,570 | loss=8.6772 (p=8.4493 v=0.2279) | game=5.7s train=89.0s