checkpoint @ game=4020 steps=4020

training_log.txt

@@ -4319,3 +4319,13 @@
 [2026-06-23 10:52:00] game= 4008 | 1/2-1/2 | moves= 56 | buf=50,000 | steps=4,008 | loss=8.6621 (p=8.4493 v=0.2128) | game=5.6s train=91.4s
 [2026-06-23 10:53:36] game= 4009 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=4,009 | loss=8.6425 (p=8.4493 v=0.1932) | game=5.0s train=91.0s
 [2026-06-23 10:55:18] game= 4010 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=4,010 | loss=8.6179 (p=8.4493 v=0.1686) | game=5.2s train=97.4s
+[2026-06-23 10:57:00] game= 4011 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=4,011 | loss=9.1160 (p=8.4493 v=0.6667) | game=7.7s train=91.5s
+[2026-06-23 10:58:37] game= 4012 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=4,012 | loss=9.1011 (p=8.4493 v=0.6519) | game=5.3s train=91.1s
+[2026-06-23 11:00:14] game= 4013 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=4,013 | loss=8.9544 (p=8.4493 v=0.5051) | game=6.0s train=91.6s
+[2026-06-23 11:01:50] game= 4014 | 1/2-1/2 | moves= 47 | buf=50,000 | steps=4,014 | loss=8.8401 (p=8.4493 v=0.3908) | game=4.8s train=91.3s
+[2026-06-23 11:03:32] game= 4015 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=4,015 | loss=8.8159 (p=8.4493 v=0.3666) | game=6.0s train=96.0s
+[2026-06-23 11:05:18] game= 4016 | 1/2-1/2 | moves= 68 | buf=50,000 | steps=4,016 | loss=8.7535 (p=8.4493 v=0.3042) | game=7.5s train=98.4s
+[2026-06-23 11:07:06] game= 4017 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=4,017 | loss=8.7304 (p=8.4493 v=0.2811) | game=9.6s train=98.2s
+[2026-06-23 11:08:51] game= 4018 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=4,018 | loss=8.7071 (p=8.4493 v=0.2578) | game=6.3s train=98.6s
+[2026-06-23 11:10:41] game= 4019 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=4,019 | loss=8.7236 (p=8.4493 v=0.2743) | game=7.6s train=103.1s
+[2026-06-23 11:12:27] game= 4020 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=4,020 | loss=8.7184 (p=8.4493 v=0.2691) | game=7.2s train=98.7s