checkpoint @ game=950 steps=950

training_log.txt

@@ -867,3 +867,13 @@
 [2026-06-12 12:46:38] game=  938 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=938 | loss=nan (p=nan v=0.5164) | game=11.5s train=90.7s
 [2026-06-12 12:48:16] game=  939 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=939 | loss=nan (p=nan v=0.5877) | game=5.2s train=92.6s
 [2026-06-12 12:49:53] game=  940 | 1/2-1/2 | moves= 69 | buf=50,000 | steps=940 | loss=nan (p=nan v=0.5685) | game=5.4s train=91.5s
+[2026-06-12 12:51:34] game=  941 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=941 | loss=nan (p=nan v=0.5508) | game=4.8s train=94.1s
+[2026-06-12 12:53:14] game=  942 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=942 | loss=nan (p=nan v=0.5370) | game=3.9s train=96.0s
+[2026-06-12 12:54:53] game=  943 | 1/2-1/2 | moves= 64 | buf=50,000 | steps=943 | loss=nan (p=nan v=0.5623) | game=5.1s train=94.3s
+[2026-06-12 12:56:35] game=  944 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=944 | loss=nan (p=nan v=0.5992) | game=4.3s train=97.1s
+[2026-06-12 12:58:18] game=  945 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=945 | loss=nan (p=nan v=0.5578) | game=3.9s train=98.6s
+[2026-06-12 13:00:01] game=  946 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=946 | loss=nan (p=nan v=0.5545) | game=4.5s train=98.8s
+[2026-06-12 13:01:48] game=  947 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=947 | loss=nan (p=nan v=0.5723) | game=4.7s train=102.1s
+[2026-06-12 13:03:31] game=  948 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=948 | loss=nan (p=nan v=0.5330) | game=5.0s train=98.4s
+[2026-06-12 13:05:19] game=  949 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=949 | loss=nan (p=nan v=0.5666) | game=5.0s train=102.1s
+[2026-06-12 13:07:07] game=  950 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=950 | loss=nan (p=nan v=0.5650) | game=4.1s train=104.4s