checkpoint @ game=870 steps=870

training_log.txt

@@ -787,3 +787,13 @@
 [2026-06-12 11:26:25] game=  858 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=858 | loss=nan (p=nan v=0.5646) | game=4.2s train=27.6s
 [2026-06-12 11:26:58] game=  859 | 1/2-1/2 | moves= 75 | buf=50,000 | steps=859 | loss=nan (p=nan v=0.5876) | game=5.8s train=27.7s
 [2026-06-12 11:27:31] game=  860 | 1/2-1/2 | moves= 46 | buf=50,000 | steps=860 | loss=nan (p=nan v=0.5957) | game=3.6s train=28.7s
+[2026-06-12 11:28:05] game=  861 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=861 | loss=nan (p=nan v=0.5156) | game=4.0s train=28.1s
+[2026-06-12 11:28:39] game=  862 | 1/2-1/2 | moves= 56 | buf=50,000 | steps=862 | loss=nan (p=nan v=0.5566) | game=4.4s train=29.4s
+[2026-06-12 11:29:12] game=  863 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=863 | loss=nan (p=nan v=0.5956) | game=4.4s train=29.0s
+[2026-06-12 11:29:46] game=  864 | 1/2-1/2 | moves= 56 | buf=50,000 | steps=864 | loss=nan (p=nan v=0.5333) | game=4.3s train=29.3s
+[2026-06-12 11:30:21] game=  865 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=865 | loss=nan (p=nan v=0.5527) | game=5.4s train=29.8s
+[2026-06-12 11:30:56] game=  866 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=866 | loss=nan (p=nan v=0.5408) | game=4.2s train=30.9s
+[2026-06-12 11:31:33] game=  867 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=867 | loss=nan (p=nan v=0.5605) | game=4.9s train=32.0s
+[2026-06-12 11:32:09] game=  868 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=868 | loss=nan (p=nan v=0.5663) | game=4.1s train=31.8s
+[2026-06-12 11:32:47] game=  869 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=869 | loss=nan (p=nan v=0.5725) | game=3.9s train=33.4s
+[2026-06-12 11:33:29] game=  870 | 1/2-1/2 | moves= 56 | buf=50,000 | steps=870 | loss=nan (p=nan v=0.5859) | game=9.0s train=32.8s