checkpoint @ game=3890 steps=3890

training_log.txt

@@ -4189,3 +4189,13 @@
 [2026-06-23 07:13:45] game= 3878 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=3,878 | loss=8.8675 (p=8.4493 v=0.4182) | game=6.4s train=91.3s
 [2026-06-23 07:15:23] game= 3879 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=3,879 | loss=8.8637 (p=8.4493 v=0.4144) | game=6.6s train=91.7s
 [2026-06-23 07:17:01] game= 3880 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=3,880 | loss=8.7931 (p=8.4493 v=0.3438) | game=5.8s train=91.9s
+[2026-06-23 07:18:42] game= 3881 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=3,881 | loss=8.6684 (p=8.4493 v=0.2191) | game=5.5s train=91.6s
+[2026-06-23 07:20:19] game= 3882 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=3,882 | loss=8.7184 (p=8.4493 v=0.2691) | game=5.2s train=92.0s
+[2026-06-23 07:21:57] game= 3883 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=3,883 | loss=8.6900 (p=8.4493 v=0.2407) | game=6.3s train=91.6s
+[2026-06-23 07:23:38] game= 3884 | 1/2-1/2 | moves= 66 | buf=50,000 | steps=3,884 | loss=8.6800 (p=8.4493 v=0.2307) | game=7.3s train=94.3s
+[2026-06-23 07:25:18] game= 3885 | 1/2-1/2 | moves= 77 | buf=50,000 | steps=3,885 | loss=8.6559 (p=8.4493 v=0.2066) | game=8.5s train=91.3s
+[2026-06-23 07:26:55] game= 3886 | 1/2-1/2 | moves= 48 | buf=50,000 | steps=3,886 | loss=8.6514 (p=8.4493 v=0.2021) | game=5.3s train=91.8s
+[2026-06-23 07:28:33] game= 3887 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,887 | loss=8.6428 (p=8.4493 v=0.1935) | game=5.7s train=92.0s
+[2026-06-23 07:30:10] game= 3888 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,888 | loss=8.6300 (p=8.4493 v=0.1807) | game=5.3s train=92.2s
+[2026-06-23 07:31:52] game= 3889 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=3,889 | loss=8.6206 (p=8.4493 v=0.1713) | game=5.9s train=95.5s
+[2026-06-23 07:33:30] game= 3890 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=3,890 | loss=8.6190 (p=8.4493 v=0.1697) | game=6.2s train=91.7s