checkpoint @ game=850 steps=850

training_log.txt

@@ -767,3 +767,13 @@
 [2026-06-12 11:16:53] game=  838 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=838 | loss=nan (p=nan v=0.6426) | game=4.6s train=19.5s
 [2026-06-12 11:17:17] game=  839 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=839 | loss=nan (p=nan v=0.6563) | game=4.6s train=19.3s
 [2026-06-12 11:17:40] game=  840 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=840 | loss=nan (p=nan v=0.6603) | game=3.8s train=19.8s
+[2026-06-12 11:18:08] game=  841 | 1/2-1/2 | moves= 56 | buf=50,000 | steps=841 | loss=nan (p=nan v=0.6310) | game=4.3s train=20.7s
+[2026-06-12 11:18:35] game=  842 | 1/2-1/2 | moves= 75 | buf=50,000 | steps=842 | loss=nan (p=nan v=0.6719) | game=5.8s train=21.2s
+[2026-06-12 11:19:02] game=  843 | 1/2-1/2 | moves= 75 | buf=50,000 | steps=843 | loss=nan (p=nan v=0.6485) | game=5.8s train=21.7s
+[2026-06-12 11:19:28] game=  844 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=844 | loss=nan (p=nan v=0.6465) | game=3.8s train=21.7s
+[2026-06-12 11:19:54] game=  845 | 1/2-1/2 | moves= 46 | buf=50,000 | steps=845 | loss=nan (p=nan v=0.6893) | game=3.6s train=22.2s
+[2026-06-12 11:20:20] game=  846 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=846 | loss=nan (p=nan v=0.6739) | game=4.0s train=22.4s
+[2026-06-12 11:20:48] game=  847 | 1/2-1/2 | moves= 69 | buf=50,000 | steps=847 | loss=nan (p=nan v=0.5908) | game=5.3s train=22.4s
+[2026-06-12 11:21:16] game=  848 | 1/2-1/2 | moves= 56 | buf=50,000 | steps=848 | loss=nan (p=nan v=0.6326) | game=4.3s train=23.7s
+[2026-06-12 11:21:44] game=  849 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=849 | loss=nan (p=nan v=0.6330) | game=4.1s train=23.4s
+[2026-06-12 11:22:14] game=  850 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=850 | loss=nan (p=nan v=0.6371) | game=5.8s train=23.8s