checkpoint @ game=4590 steps=4590

training_log.txt

@@ -4889,3 +4889,13 @@
 [2026-06-25 04:36:15] game= 4578 | 1/2-1/2 | moves= 58 | buf=50,000 | steps=4,578 | loss=8.6086 (p=8.4493 v=0.1593) | game=6.4s train=89.2s
 [2026-06-25 04:37:56] game= 4579 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=4,579 | loss=9.1521 (p=8.4493 v=0.7028) | game=10.2s train=91.1s
 [2026-06-25 04:39:31] game= 4580 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=4,580 | loss=9.0102 (p=8.4493 v=0.5609) | game=5.7s train=89.4s
+[2026-06-25 04:41:12] game= 4581 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=4,581 | loss=8.9157 (p=8.4493 v=0.4664) | game=6.3s train=89.3s
+[2026-06-25 04:42:50] game= 4582 | 1/2-1/2 | moves= 75 | buf=50,000 | steps=4,582 | loss=8.8584 (p=8.4493 v=0.4091) | game=7.5s train=90.6s
+[2026-06-25 04:44:25] game= 4583 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=4,583 | loss=8.8858 (p=8.4493 v=0.4365) | game=6.3s train=88.6s
+[2026-06-25 04:46:00] game= 4584 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=4,584 | loss=8.7788 (p=8.4493 v=0.3295) | game=5.7s train=89.2s
+[2026-06-25 04:47:38] game= 4585 | 1/2-1/2 | moves= 60 | buf=50,000 | steps=4,585 | loss=8.7226 (p=8.4493 v=0.2733) | game=6.9s train=90.7s
+[2026-06-25 04:49:14] game= 4586 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=4,586 | loss=8.6966 (p=8.4493 v=0.2473) | game=6.7s train=89.8s
+[2026-06-25 04:50:49] game= 4587 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=4,587 | loss=8.6657 (p=8.4493 v=0.2164) | game=5.8s train=88.7s
+[2026-06-25 04:52:29] game= 4588 | 1/2-1/2 | moves= 75 | buf=50,000 | steps=4,588 | loss=8.6548 (p=8.4493 v=0.2055) | game=8.4s train=91.4s
+[2026-06-25 04:54:05] game= 4589 | 1/2-1/2 | moves= 60 | buf=50,000 | steps=4,589 | loss=8.6818 (p=8.4493 v=0.2326) | game=6.1s train=90.2s
+[2026-06-25 04:55:41] game= 4590 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=4,590 | loss=8.6164 (p=8.4493 v=0.1671) | game=7.2s train=88.7s