checkpoint @ game=3950 steps=3950

training_log.txt

@@ -4249,3 +4249,13 @@
 [2026-06-23 08:54:04] game= 3938 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=3,938 | loss=8.6405 (p=8.4493 v=0.1912) | game=7.4s train=94.4s
 [2026-06-23 08:55:44] game= 3939 | 1/2-1/2 | moves= 47 | buf=50,000 | steps=3,939 | loss=8.6556 (p=8.4493 v=0.2063) | game=5.2s train=94.7s
 [2026-06-23 08:57:25] game= 3940 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=3,940 | loss=8.7112 (p=8.4493 v=0.2619) | game=5.9s train=94.8s
+[2026-06-23 08:59:09] game= 3941 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=3,941 | loss=8.6021 (p=8.4493 v=0.1528) | game=6.8s train=93.4s
+[2026-06-23 09:00:55] game= 3942 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=3,942 | loss=8.6160 (p=8.4493 v=0.1667) | game=6.0s train=100.0s
+[2026-06-23 09:02:37] game= 3943 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,943 | loss=9.1052 (p=8.4493 v=0.6559) | game=8.3s train=93.5s
+[2026-06-23 09:04:18] game= 3944 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=3,944 | loss=9.0584 (p=8.4493 v=0.6091) | game=6.0s train=95.9s
+[2026-06-23 09:06:00] game= 3945 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=3,945 | loss=8.9661 (p=8.4493 v=0.5168) | game=7.1s train=94.9s
+[2026-06-23 09:07:42] game= 3946 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=3,946 | loss=8.9119 (p=8.4493 v=0.4626) | game=6.0s train=95.4s
+[2026-06-23 09:09:26] game= 3947 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=3,947 | loss=8.8205 (p=8.4493 v=0.3712) | game=9.2s train=94.9s
+[2026-06-23 09:11:06] game= 3948 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=3,948 | loss=8.7430 (p=8.4493 v=0.2937) | game=5.0s train=95.0s
+[2026-06-23 09:12:51] game= 3949 | 1/2-1/2 | moves= 62 | buf=50,000 | steps=3,949 | loss=8.7584 (p=8.4493 v=0.3091) | game=6.9s train=98.0s
+[2026-06-23 09:14:34] game= 3950 | 1/2-1/2 | moves= 75 | buf=50,000 | steps=3,950 | loss=8.6775 (p=8.4493 v=0.2282) | game=8.5s train=95.2s