checkpoint @ game=840 steps=840

training_log.txt

@@ -757,3 +757,13 @@
 [2026-06-13 04:03:41] game=  828 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=828 | loss=nan (p=nan v=0.5311) | game=23.9s train=12.4s
 [2026-06-13 04:04:18] game=  829 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=829 | loss=nan (p=nan v=0.5390) | game=23.9s train=12.8s
 [2026-06-13 04:04:55] game=  830 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=830 | loss=nan (p=nan v=0.5862) | game=23.9s train=13.1s
+[2026-06-13 04:05:34] game=  831 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=831 | loss=nan (p=nan v=0.6081) | game=24.0s train=13.6s
+[2026-06-13 04:06:12] game=  832 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=832 | loss=nan (p=nan v=0.6098) | game=23.5s train=13.9s
+[2026-06-13 04:06:50] game=  833 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=833 | loss=nan (p=nan v=0.5235) | game=24.0s train=14.3s
+[2026-06-13 04:07:29] game=  834 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=834 | loss=nan (p=nan v=0.5231) | game=24.1s train=14.7s
+[2026-06-13 04:08:08] game=  835 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=835 | loss=nan (p=nan v=0.5390) | game=23.7s train=15.2s
+[2026-06-13 04:08:48] game=  836 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=836 | loss=nan (p=nan v=0.4058) | game=24.1s train=15.1s
+[2026-06-13 04:09:28] game=  837 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=837 | loss=nan (p=nan v=0.6568) | game=24.1s train=16.1s
+[2026-06-13 04:10:08] game=  838 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=838 | loss=nan (p=nan v=0.6643) | game=24.2s train=16.0s
+[2026-06-13 04:10:50] game=  839 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=839 | loss=nan (p=nan v=0.5233) | game=24.3s train=16.9s
+[2026-06-13 04:11:31] game=  840 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=840 | loss=nan (p=nan v=0.5764) | game=23.7s train=17.2s