checkpoint @ game=3700 steps=3700

training_log.txt

@@ -3999,3 +3999,13 @@
 [2026-06-23 02:07:04] game= 3688 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=3,688 | loss=9.1655 (p=8.4493 v=0.7162) | game=8.4s train=87.9s
 [2026-06-23 02:08:36] game= 3689 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=3,689 | loss=9.0774 (p=8.4493 v=0.6282) | game=5.3s train=86.4s
 [2026-06-23 02:10:10] game= 3690 | 1/2-1/2 | moves= 69 | buf=50,000 | steps=3,690 | loss=8.9961 (p=8.4493 v=0.5468) | game=7.3s train=86.6s
+[2026-06-23 02:11:46] game= 3691 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,691 | loss=8.8935 (p=8.4493 v=0.4442) | game=5.8s train=87.9s
+[2026-06-23 02:13:21] game= 3692 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=3,692 | loss=8.8448 (p=8.4493 v=0.3955) | game=7.0s train=87.5s
+[2026-06-23 02:14:59] game= 3693 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=3,693 | loss=8.8293 (p=8.4493 v=0.3800) | game=8.8s train=89.1s
+[2026-06-23 02:16:32] game= 3694 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=3,694 | loss=8.7325 (p=8.4493 v=0.2833) | game=6.2s train=87.3s
+[2026-06-23 02:18:07] game= 3695 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=3,695 | loss=8.7019 (p=8.4493 v=0.2526) | game=5.8s train=88.6s
+[2026-06-23 02:19:40] game= 3696 | 1/2-1/2 | moves= 60 | buf=50,000 | steps=3,696 | loss=8.6960 (p=8.4493 v=0.2467) | game=6.3s train=87.2s
+[2026-06-23 02:21:14] game= 3697 | 1/2-1/2 | moves= 56 | buf=50,000 | steps=3,697 | loss=8.6581 (p=8.4493 v=0.2088) | game=6.1s train=87.5s
+[2026-06-23 02:22:48] game= 3698 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=3,698 | loss=8.6438 (p=8.4493 v=0.1945) | game=7.1s train=87.1s
+[2026-06-23 02:24:21] game= 3699 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,699 | loss=8.6692 (p=8.4493 v=0.2199) | game=5.5s train=87.3s
+[2026-06-23 02:25:55] game= 3700 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=3,700 | loss=8.6942 (p=8.4493 v=0.2449) | game=5.9s train=87.9s