checkpoint @ game=3970 steps=3970

training_log.txt

@@ -4269,3 +4269,13 @@
 [2026-06-23 09:28:05] game= 3958 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=3,958 | loss=8.6100 (p=8.4493 v=0.1607) | game=6.1s train=93.7s
 [2026-06-23 09:29:44] game= 3959 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=3,959 | loss=8.5921 (p=8.4493 v=0.1428) | game=5.7s train=93.4s
 [2026-06-23 09:31:27] game= 3960 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,960 | loss=9.1933 (p=8.4493 v=0.7440) | game=8.4s train=94.9s
+[2026-06-23 09:33:12] game= 3961 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=3,961 | loss=9.0214 (p=8.4493 v=0.5721) | game=5.8s train=94.9s
+[2026-06-23 09:34:53] game= 3962 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=3,962 | loss=9.0180 (p=8.4493 v=0.5687) | game=5.7s train=94.8s
+[2026-06-23 09:36:34] game= 3963 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=3,963 | loss=8.9073 (p=8.4493 v=0.4580) | game=6.6s train=94.6s
+[2026-06-23 09:38:15] game= 3964 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,964 | loss=8.8366 (p=8.4493 v=0.3873) | game=5.6s train=95.2s
+[2026-06-23 09:39:58] game= 3965 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=3,965 | loss=8.8071 (p=8.4493 v=0.3578) | game=9.1s train=94.2s
+[2026-06-23 09:41:40] game= 3966 | 1/2-1/2 | moves= 68 | buf=50,000 | steps=3,966 | loss=8.7590 (p=8.4493 v=0.3097) | game=7.6s train=94.5s
+[2026-06-23 09:43:22] game= 3967 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=3,967 | loss=8.7043 (p=8.4493 v=0.2550) | game=5.5s train=96.0s
+[2026-06-23 09:45:03] game= 3968 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=3,968 | loss=8.6827 (p=8.4493 v=0.2334) | game=6.8s train=95.1s
+[2026-06-23 09:46:45] game= 3969 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=3,969 | loss=8.6682 (p=8.4493 v=0.2189) | game=6.1s train=95.5s
+[2026-06-23 09:48:26] game= 3970 | 1/2-1/2 | moves= 58 | buf=50,000 | steps=3,970 | loss=8.6556 (p=8.4493 v=0.2063) | game=6.0s train=95.3s