checkpoint @ game=3820 steps=3820

training_log.txt

@@ -4119,3 +4119,13 @@
 [2026-06-23 05:18:38] game= 3808 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=3,808 | loss=9.0167 (p=8.4493 v=0.5674) | game=6.4s train=89.0s
 [2026-06-23 05:20:15] game= 3809 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,809 | loss=8.9793 (p=8.4493 v=0.5300) | game=5.5s train=91.7s
 [2026-06-23 05:21:50] game= 3810 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=3,810 | loss=8.8803 (p=8.4493 v=0.4310) | game=6.0s train=89.1s
+[2026-06-23 05:23:30] game= 3811 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=3,811 | loss=8.7905 (p=8.4493 v=0.3412) | game=6.9s train=89.8s
+[2026-06-23 05:25:06] game= 3812 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=3,812 | loss=8.7132 (p=8.4493 v=0.2639) | game=6.1s train=89.6s
+[2026-06-23 05:26:44] game= 3813 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=3,813 | loss=8.7356 (p=8.4493 v=0.2863) | game=8.4s train=89.2s
+[2026-06-23 05:28:24] game= 3814 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=3,814 | loss=8.7517 (p=8.4493 v=0.3024) | game=10.8s train=89.4s
+[2026-06-23 05:30:01] game= 3815 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=3,815 | loss=8.7251 (p=8.4493 v=0.2758) | game=7.4s train=90.1s
+[2026-06-23 05:31:37] game= 3816 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=3,816 | loss=8.6687 (p=8.4493 v=0.2194) | game=5.5s train=90.4s
+[2026-06-23 05:33:13] game= 3817 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=3,817 | loss=8.7455 (p=8.4493 v=0.2962) | game=5.8s train=90.4s
+[2026-06-23 05:34:51] game= 3818 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=3,818 | loss=8.6385 (p=8.4493 v=0.1892) | game=8.3s train=89.7s
+[2026-06-23 05:36:27] game= 3819 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=3,819 | loss=8.6434 (p=8.4493 v=0.1941) | game=6.1s train=89.7s
+[2026-06-23 05:38:03] game= 3820 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,820 | loss=8.6114 (p=8.4493 v=0.1621) | game=5.5s train=90.3s