checkpoint @ game=900 steps=900

training_log.txt

@@ -817,3 +817,13 @@
 [2026-06-12 11:46:30] game=  888 | 1/2-1/2 | moves= 48 | buf=50,000 | steps=888 | loss=nan (p=nan v=0.6581) | game=3.7s train=43.3s
 [2026-06-12 11:47:19] game=  889 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=889 | loss=nan (p=nan v=0.6800) | game=4.3s train=44.2s
 [2026-06-12 11:48:10] game=  890 | 1/2-1/2 | moves= 64 | buf=50,000 | steps=890 | loss=nan (p=nan v=0.6699) | game=5.0s train=45.9s
+[2026-06-12 11:49:04] game=  891 | 1/2-1/2 | moves= 70 | buf=50,000 | steps=891 | loss=nan (p=nan v=0.6816) | game=5.5s train=46.0s
+[2026-06-12 11:49:55] game=  892 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=892 | loss=nan (p=nan v=0.6582) | game=5.2s train=46.4s
+[2026-06-12 11:50:49] game=  893 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=893 | loss=nan (p=nan v=0.6992) | game=4.3s train=48.8s
+[2026-06-12 11:51:41] game=  894 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=894 | loss=nan (p=nan v=0.6152) | game=3.9s train=48.0s
+[2026-06-12 11:52:34] game=  895 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=895 | loss=nan (p=nan v=0.6855) | game=4.1s train=49.2s
+[2026-06-12 11:53:30] game=  896 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=896 | loss=nan (p=nan v=0.6448) | game=4.7s train=51.3s
+[2026-06-12 11:54:26] game=  897 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=897 | loss=nan (p=nan v=0.6425) | game=5.2s train=50.1s
+[2026-06-12 11:55:21] game=  898 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=898 | loss=nan (p=nan v=0.6092) | game=3.9s train=51.3s
+[2026-06-12 11:56:20] game=  899 | 1/2-1/2 | moves= 69 | buf=50,000 | steps=899 | loss=nan (p=nan v=0.6190) | game=5.4s train=53.6s
+[2026-06-12 11:57:18] game=  900 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=900 | loss=nan (p=nan v=0.6366) | game=4.0s train=53.4s