checkpoint @ game=3870 steps=3870

training_log.txt

@@ -4169,3 +4169,13 @@
 [2026-06-23 06:40:41] game= 3858 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=3,858 | loss=9.1948 (p=8.4493 v=0.7455) | game=9.3s train=93.9s
 [2026-06-23 06:42:17] game= 3859 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=3,859 | loss=9.0410 (p=8.4493 v=0.5917) | game=5.5s train=90.5s
 [2026-06-23 06:43:54] game= 3860 | 1/2-1/2 | moves= 46 | buf=50,000 | steps=3,860 | loss=8.9051 (p=8.4493 v=0.4558) | game=5.1s train=91.3s
+[2026-06-23 06:45:37] game= 3861 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=3,861 | loss=8.8843 (p=8.4493 v=0.4350) | game=7.1s train=91.5s
+[2026-06-23 06:47:20] game= 3862 | 1/2-1/2 | moves= 69 | buf=50,000 | steps=3,862 | loss=8.8556 (p=8.4493 v=0.4063) | game=7.2s train=95.9s
+[2026-06-23 06:49:02] game= 3863 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=3,863 | loss=8.7153 (p=8.4493 v=0.2661) | game=9.5s train=92.0s
+[2026-06-23 06:50:41] game= 3864 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=3,864 | loss=8.7477 (p=8.4493 v=0.2984) | game=6.9s train=92.2s
+[2026-06-23 06:52:19] game= 3865 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=3,865 | loss=8.7765 (p=8.4493 v=0.3272) | game=6.0s train=91.9s
+[2026-06-23 06:53:56] game= 3866 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=3,866 | loss=8.7201 (p=8.4493 v=0.2708) | game=5.3s train=91.9s
+[2026-06-23 06:55:33] game= 3867 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,867 | loss=8.6788 (p=8.4493 v=0.2295) | game=5.5s train=91.2s
+[2026-06-23 06:57:14] game= 3868 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=3,868 | loss=8.6494 (p=8.4493 v=0.2001) | game=5.4s train=95.7s
+[2026-06-23 06:58:52] game= 3869 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=3,869 | loss=8.6295 (p=8.4493 v=0.1802) | game=5.8s train=92.6s
+[2026-06-23 07:00:31] game= 3870 | 1/2-1/2 | moves= 60 | buf=50,000 | steps=3,870 | loss=8.6379 (p=8.4493 v=0.1886) | game=7.0s train=91.7s