checkpoint @ game=920 steps=920

training_log.txt

@@ -837,3 +837,13 @@
 [2026-06-12 12:05:36] game=  908 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=908 | loss=nan (p=nan v=0.5514) | game=4.2s train=60.6s
 [2026-06-12 12:06:41] game=  909 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=909 | loss=nan (p=nan v=0.5665) | game=4.0s train=61.2s
 [2026-06-12 12:07:49] game=  910 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=910 | loss=nan (p=nan v=0.6112) | game=3.8s train=64.4s
+[2026-06-12 12:09:00] game=  911 | 1/2-1/2 | moves= 62 | buf=50,000 | steps=911 | loss=nan (p=nan v=0.5923) | game=4.9s train=63.5s
+[2026-06-12 12:10:09] game=  912 | 1/2-1/2 | moves= 62 | buf=50,000 | steps=912 | loss=nan (p=nan v=0.5680) | game=4.8s train=63.9s
+[2026-06-12 12:11:21] game=  913 | 1/2-1/2 | moves= 66 | buf=50,000 | steps=913 | loss=nan (p=nan v=0.5196) | game=5.1s train=66.6s
+[2026-06-12 12:12:31] game=  914 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=914 | loss=nan (p=nan v=0.5528) | game=3.9s train=66.5s
+[2026-06-12 12:13:43] game=  915 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=915 | loss=nan (p=nan v=0.6032) | game=4.0s train=67.3s
+[2026-06-12 12:14:56] game=  916 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=916 | loss=nan (p=nan v=0.5682) | game=5.0s train=68.0s
+[2026-06-12 12:16:11] game=  917 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=917 | loss=nan (p=nan v=0.5859) | game=4.0s train=71.2s
+[2026-06-12 12:17:27] game=  918 | 1/2-1/2 | moves= 74 | buf=50,000 | steps=918 | loss=nan (p=nan v=0.5957) | game=5.8s train=70.1s
+[2026-06-12 12:18:45] game=  919 | 1/2-1/2 | moves= 79 | buf=50,000 | steps=919 | loss=nan (p=nan v=0.5469) | game=6.1s train=71.4s
+[2026-06-12 12:20:04] game=  920 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=920 | loss=nan (p=nan v=0.5467) | game=6.3s train=72.5s