checkpoint @ game=2590 steps=2590

training_log.txt

@@ -2888,3 +2888,13 @@
 [2026-06-20 08:15:37] game= 2578 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=2,578 | loss=8.4888 (p=8.4493 v=0.0395) | game=6.7s train=94.8s
 [2026-06-20 08:17:18] game= 2579 | 1/2-1/2 | moves= 65 | buf=50,000 | steps=2,579 | loss=8.4797 (p=8.4493 v=0.0304) | game=6.5s train=94.9s
 [2026-06-20 08:19:02] game= 2580 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=2,580 | loss=8.4838 (p=8.4493 v=0.0345) | game=5.9s train=97.8s
+[2026-06-20 08:20:46] game= 2581 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=2,581 | loss=8.4800 (p=8.4493 v=0.0307) | game=6.4s train=95.0s
+[2026-06-20 08:22:28] game= 2582 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=2,582 | loss=8.4795 (p=8.4493 v=0.0302) | game=6.4s train=94.9s
+[2026-06-20 08:24:08] game= 2583 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=2,583 | loss=8.4906 (p=8.4493 v=0.0413) | game=5.7s train=94.8s
+[2026-06-20 08:25:49] game= 2584 | 1/2-1/2 | moves= 60 | buf=50,000 | steps=2,584 | loss=8.4631 (p=8.4493 v=0.0138) | game=6.0s train=94.9s
+[2026-06-20 08:27:33] game= 2585 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=2,585 | loss=8.4830 (p=8.4493 v=0.0337) | game=9.1s train=95.0s
+[2026-06-20 08:29:16] game= 2586 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=2,586 | loss=8.4943 (p=8.4493 v=0.0450) | game=4.9s train=98.1s
+[2026-06-20 08:30:58] game= 2587 | 1/2-1/2 | moves= 70 | buf=50,000 | steps=2,587 | loss=8.4958 (p=8.4493 v=0.0465) | game=7.1s train=95.2s
+[2026-06-20 08:32:39] game= 2588 | 1/2-1/2 | moves= 56 | buf=50,000 | steps=2,588 | loss=8.4872 (p=8.4493 v=0.0379) | game=5.6s train=95.1s
+[2026-06-20 08:34:20] game= 2589 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=2,589 | loss=8.4988 (p=8.4493 v=0.0495) | game=5.5s train=95.1s
+[2026-06-20 08:36:01] game= 2590 | 1/2-1/2 | moves= 60 | buf=50,000 | steps=2,590 | loss=8.4888 (p=8.4493 v=0.0395) | game=6.0s train=95.0s