checkpoint @ game=1640 steps=1640

training_log.txt

@@ -1938,3 +1938,13 @@
 [2026-06-18 12:04:09] game= 1628 | 1/2-1/2 | moves= 75 | buf=50,000 | steps=1,628 | loss=8.6276 (p=8.4493 v=0.1783) | game=7.6s train=87.3s
 [2026-06-18 12:05:43] game= 1629 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=1,629 | loss=8.5936 (p=8.4493 v=0.1443) | game=6.6s train=87.1s
 [2026-06-18 12:07:16] game= 1630 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=1,630 | loss=8.6064 (p=8.4493 v=0.1571) | game=5.9s train=87.3s
+[2026-06-18 12:08:56] game= 1631 | 1/2-1/2 | moves= 75 | buf=50,000 | steps=1,631 | loss=8.5685 (p=8.4493 v=0.1192) | game=7.6s train=89.9s
+[2026-06-18 12:10:29] game= 1632 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=1,632 | loss=8.5549 (p=8.4493 v=0.1056) | game=5.2s train=87.5s
+[2026-06-18 12:12:04] game= 1633 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=1,633 | loss=8.5836 (p=8.4493 v=0.1343) | game=8.3s train=87.3s
+[2026-06-18 12:13:43] game= 1634 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=1,634 | loss=8.9869 (p=8.4493 v=0.5376) | game=10.9s train=87.5s
+[2026-06-18 12:15:18] game= 1635 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=1,635 | loss=8.9915 (p=8.4493 v=0.5422) | game=5.1s train=90.1s
+[2026-06-18 12:16:51] game= 1636 | 1/2-1/2 | moves= 58 | buf=50,000 | steps=1,636 | loss=8.9016 (p=8.4493 v=0.4523) | game=5.9s train=87.6s
+[2026-06-18 12:18:26] game= 1637 | 1/2-1/2 | moves= 71 | buf=50,000 | steps=1,637 | loss=8.8249 (p=8.4493 v=0.3757) | game=7.2s train=87.6s
+[2026-06-18 12:19:59] game= 1638 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=1,638 | loss=8.7979 (p=8.4493 v=0.3486) | game=5.1s train=87.4s
+[2026-06-18 12:21:32] game= 1639 | 1/2-1/2 | moves= 56 | buf=50,000 | steps=1,639 | loss=8.7413 (p=8.4493 v=0.2920) | game=5.7s train=87.6s
+[2026-06-18 12:23:05] game= 1640 | 1/2-1/2 | moves= 46 | buf=50,000 | steps=1,640 | loss=8.7026 (p=8.4493 v=0.2533) | game=4.7s train=88.1s