checkpoint @ game=1590 steps=1590

training_log.txt

@@ -1888,3 +1888,13 @@
 [2026-06-18 10:45:46] game= 1578 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=1,578 | loss=8.6298 (p=8.4493 v=0.1805) | game=5.5s train=86.7s
 [2026-06-18 10:47:19] game= 1579 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=1,579 | loss=8.6021 (p=8.4493 v=0.1528) | game=6.6s train=86.5s
 [2026-06-18 10:48:50] game= 1580 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=1,580 | loss=8.6127 (p=8.4493 v=0.1634) | game=4.9s train=86.6s
+[2026-06-18 10:50:27] game= 1581 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=1,581 | loss=8.5642 (p=8.4493 v=0.1150) | game=5.7s train=88.5s
+[2026-06-18 10:51:58] game= 1582 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=1,582 | loss=8.6010 (p=8.4493 v=0.1517) | game=5.0s train=86.6s
+[2026-06-18 10:53:35] game= 1583 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=1,583 | loss=9.0029 (p=8.4493 v=0.5536) | game=9.7s train=86.7s
+[2026-06-18 10:55:06] game= 1584 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=1,584 | loss=8.9248 (p=8.4493 v=0.4755) | game=5.0s train=86.6s
+[2026-06-18 10:56:43] game= 1585 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=1,585 | loss=8.8575 (p=8.4493 v=0.4082) | game=10.0s train=86.8s
+[2026-06-18 10:58:15] game= 1586 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=1,586 | loss=8.8212 (p=8.4493 v=0.3719) | game=5.0s train=86.7s
+[2026-06-18 10:59:48] game= 1587 | 1/2-1/2 | moves= 62 | buf=50,000 | steps=1,587 | loss=8.8060 (p=8.4493 v=0.3567) | game=6.2s train=86.7s
+[2026-06-18 11:01:22] game= 1588 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=1,588 | loss=8.7344 (p=8.4493 v=0.2851) | game=5.3s train=89.1s
+[2026-06-18 11:02:55] game= 1589 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=1,589 | loss=8.7180 (p=8.4493 v=0.2687) | game=6.8s train=86.8s
+[2026-06-18 11:04:29] game= 1590 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=1,590 | loss=8.6971 (p=8.4493 v=0.2478) | game=6.5s train=86.8s