checkpoint @ game=1840 steps=1840

training_log.txt

@@ -2138,3 +2138,13 @@
 [2026-06-18 17:25:35] game= 1828 | 0-1 | moves= 18 | buf=50,000 | steps=1,828 | loss=8.6817 (p=8.4493 v=0.2324) | game=1.9s train=90.9s
 [2026-06-18 17:27:13] game= 1829 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=1,829 | loss=8.6853 (p=8.4493 v=0.2360) | game=6.4s train=91.1s
 [2026-06-18 17:28:51] game= 1830 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=1,830 | loss=8.6566 (p=8.4493 v=0.2073) | game=6.4s train=91.4s
+[2026-06-18 17:30:30] game= 1831 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=1,831 | loss=8.6499 (p=8.4493 v=0.2006) | game=5.1s train=91.0s
+[2026-06-18 17:32:07] game= 1832 | 1/2-1/2 | moves= 60 | buf=50,000 | steps=1,832 | loss=8.6144 (p=8.4493 v=0.1651) | game=6.2s train=91.2s
+[2026-06-18 17:33:44] game= 1833 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=1,833 | loss=8.6191 (p=8.4493 v=0.1698) | game=5.9s train=91.1s
+[2026-06-18 17:35:28] game= 1834 | 1/2-1/2 | moves= 69 | buf=50,000 | steps=1,834 | loss=8.6057 (p=8.4493 v=0.1564) | game=7.0s train=96.6s
+[2026-06-18 17:37:04] game= 1835 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=1,835 | loss=8.5943 (p=8.4493 v=0.1450) | game=5.2s train=91.6s
+[2026-06-18 17:38:42] game= 1836 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=1,836 | loss=8.6084 (p=8.4493 v=0.1591) | game=6.3s train=91.3s
+[2026-06-18 17:40:22] game= 1837 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=1,837 | loss=8.5855 (p=8.4493 v=0.1362) | game=8.3s train=91.6s
+[2026-06-18 17:42:04] game= 1838 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=1,838 | loss=8.9235 (p=8.4493 v=0.4742) | game=10.4s train=91.4s
+[2026-06-18 17:43:40] game= 1839 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=1,839 | loss=8.8817 (p=8.4493 v=0.4324) | game=5.6s train=90.7s
+[2026-06-18 17:45:16] game= 1840 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=1,840 | loss=8.8595 (p=8.4493 v=0.4102) | game=5.2s train=91.0s