checkpoint @ game=1400 steps=1400

training_log.txt

@@ -1698,3 +1698,13 @@
 [2026-06-17 20:40:34] game= 1388 | 1/2-1/2 | moves= 57 | buf=70,000 | steps=1,388 | loss=8.6946 (p=8.4493 v=0.2453) | game=5.7s train=163.0s
 [2026-06-17 20:43:23] game= 1389 | 1/2-1/2 | moves= 67 | buf=70,000 | steps=1,389 | loss=8.6791 (p=8.4493 v=0.2298) | game=6.6s train=162.8s
 [2026-06-17 20:46:11] game= 1390 | 1/2-1/2 | moves= 48 | buf=70,000 | steps=1,390 | loss=8.6512 (p=8.4493 v=0.2019) | game=4.7s train=162.9s
+[2026-06-17 20:49:02] game= 1391 | 1/2-1/2 | moves= 59 | buf=70,000 | steps=1,391 | loss=8.6206 (p=8.4493 v=0.1713) | game=5.8s train=162.9s
+[2026-06-17 20:51:56] game= 1392 | 1/2-1/2 | moves= 81 | buf=70,000 | steps=1,392 | loss=8.6154 (p=8.4493 v=0.1661) | game=11.2s train=162.9s
+[2026-06-17 20:54:47] game= 1393 | 1/2-1/2 | moves= 81 | buf=70,000 | steps=1,393 | loss=8.5961 (p=8.4493 v=0.1468) | game=8.0s train=163.1s
+[2026-06-17 20:57:36] game= 1394 | 1/2-1/2 | moves= 62 | buf=70,000 | steps=1,394 | loss=8.5768 (p=8.4493 v=0.1275) | game=6.2s train=163.0s
+[2026-06-17 21:00:24] game= 1395 | 1/2-1/2 | moves= 50 | buf=70,000 | steps=1,395 | loss=8.5666 (p=8.4493 v=0.1173) | game=4.9s train=162.9s
+[2026-06-17 21:03:17] game= 1396 | 1/2-1/2 | moves= 66 | buf=70,000 | steps=1,396 | loss=8.5587 (p=8.4493 v=0.1094) | game=9.7s train=163.1s
+[2026-06-17 21:06:05] game= 1397 | 1/2-1/2 | moves= 49 | buf=70,000 | steps=1,397 | loss=8.5570 (p=8.4493 v=0.1077) | game=4.8s train=162.9s
+[2026-06-17 21:08:53] game= 1398 | 1/2-1/2 | moves= 57 | buf=70,000 | steps=1,398 | loss=8.5374 (p=8.4493 v=0.0881) | game=5.6s train=162.9s
+[2026-06-17 21:11:41] game= 1399 | 1/2-1/2 | moves= 53 | buf=70,000 | steps=1,399 | loss=8.5398 (p=8.4493 v=0.0905) | game=5.2s train=163.0s
+[2026-06-17 21:14:33] game= 1400 | 1/2-1/2 | moves= 63 | buf=70,000 | steps=1,400 | loss=8.5248 (p=8.4493 v=0.0755) | game=6.2s train=165.6s