checkpoint @ game=4640 steps=4640

training_log.txt

@@ -4939,3 +4939,13 @@
 [2026-06-25 05:56:13] game= 4628 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=4,628 | loss=8.6053 (p=8.4493 v=0.1560) | game=8.1s train=88.5s
 [2026-06-25 05:57:48] game= 4629 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=4,629 | loss=8.5828 (p=8.4493 v=0.1335) | game=6.1s train=88.5s
 [2026-06-25 05:59:22] game= 4630 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=4,630 | loss=8.5477 (p=8.4493 v=0.0984) | game=5.5s train=88.4s
+[2026-06-25 06:01:10] game= 4631 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=4,631 | loss=8.5772 (p=8.4493 v=0.1279) | game=5.0s train=90.7s
+[2026-06-25 06:02:44] game= 4632 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=4,632 | loss=8.5775 (p=8.4493 v=0.1282) | game=5.1s train=88.5s
+[2026-06-25 06:04:17] game= 4633 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=4,633 | loss=8.6019 (p=8.4493 v=0.1526) | game=5.0s train=88.5s
+[2026-06-25 06:05:51] game= 4634 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=4,634 | loss=8.5765 (p=8.4493 v=0.1272) | game=5.0s train=88.4s
+[2026-06-25 06:07:26] game= 4635 | 1/2-1/2 | moves= 46 | buf=50,000 | steps=4,635 | loss=8.5855 (p=8.4493 v=0.1362) | game=4.7s train=90.6s
+[2026-06-25 06:09:00] game= 4636 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=4,636 | loss=8.5742 (p=8.4493 v=0.1249) | game=5.0s train=88.6s
+[2026-06-25 06:10:34] game= 4637 | 1/2-1/2 | moves= 62 | buf=50,000 | steps=4,637 | loss=8.5660 (p=8.4493 v=0.1167) | game=6.2s train=88.5s
+[2026-06-25 06:12:08] game= 4638 | 1/2-1/2 | moves= 55 | buf=50,000 | steps=4,638 | loss=8.5511 (p=8.4493 v=0.1018) | game=5.5s train=88.6s
+[2026-06-25 06:13:44] game= 4639 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=4,639 | loss=8.5973 (p=8.4493 v=0.1480) | game=5.3s train=90.7s
+[2026-06-25 06:15:18] game= 4640 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=4,640 | loss=8.5634 (p=8.4493 v=0.1141) | game=5.1s train=88.7s