checkpoint @ game=3850 steps=3850

training_log.txt

@@ -4149,3 +4149,13 @@
 [2026-06-23 06:07:36] game= 3838 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=3,838 | loss=8.6174 (p=8.4493 v=0.1681) | game=7.1s train=90.7s
 [2026-06-23 06:09:13] game= 3839 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=3,839 | loss=8.6387 (p=8.4493 v=0.1894) | game=6.4s train=90.2s
 [2026-06-23 06:10:52] game= 3840 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=3,840 | loss=8.5995 (p=8.4493 v=0.1502) | game=8.8s train=90.4s
+[2026-06-23 06:12:38] game= 3841 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=3,841 | loss=9.1574 (p=8.4493 v=0.7081) | game=10.1s train=92.9s
+[2026-06-23 06:14:16] game= 3842 | 1/2-1/2 | moves= 71 | buf=50,000 | steps=3,842 | loss=9.0499 (p=8.4493 v=0.6006) | game=7.9s train=90.9s
+[2026-06-23 06:15:54] game= 3843 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=3,843 | loss=8.9270 (p=8.4493 v=0.4777) | game=7.0s train=90.1s
+[2026-06-23 06:17:32] game= 3844 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=3,844 | loss=8.9125 (p=8.4493 v=0.4632) | game=7.5s train=91.0s
+[2026-06-23 06:19:10] game= 3845 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=3,845 | loss=8.8058 (p=8.4493 v=0.3565) | game=6.0s train=91.5s
+[2026-06-23 06:20:50] game= 3846 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,846 | loss=8.8280 (p=8.4493 v=0.3787) | game=6.0s train=95.0s
+[2026-06-23 06:22:29] game= 3847 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=3,847 | loss=8.7062 (p=8.4493 v=0.2569) | game=5.9s train=92.4s
+[2026-06-23 06:24:09] game= 3848 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=3,848 | loss=8.7268 (p=8.4493 v=0.2775) | game=7.3s train=92.5s
+[2026-06-23 06:25:48] game= 3849 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=3,849 | loss=8.7303 (p=8.4493 v=0.2810) | game=6.4s train=92.9s
+[2026-06-23 06:27:25] game= 3850 | 1/2-1/2 | moves= 58 | buf=50,000 | steps=3,850 | loss=8.6531 (p=8.4493 v=0.2038) | game=5.8s train=91.4s