checkpoint @ game=4470 steps=4470

training_log.txt

@@ -4769,3 +4769,13 @@
 [2026-06-24 12:55:07] game= 4458 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=4,458 | loss=8.7629 (p=8.4493 v=0.3136) | game=4.9s train=95.4s
 [2026-06-24 12:56:48] game= 4459 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=4,459 | loss=8.6683 (p=8.4493 v=0.2190) | game=5.2s train=95.5s
 [2026-06-24 12:58:34] game= 4460 | 1/2-1/2 | moves= 73 | buf=50,000 | steps=4,460 | loss=8.6542 (p=8.4493 v=0.2049) | game=7.3s train=98.4s
+[2026-06-25 01:28:58] game= 4461 | 1/2-1/2 | moves= 47 | buf=33,870 | steps=4,461 | loss=9.3181 (p=8.4493 v=0.8688) | game=6.3s train=98.8s
+[2026-06-25 01:30:32] game= 4462 | 1/2-1/2 | moves= 59 | buf=33,929 | steps=4,462 | loss=9.1627 (p=8.4493 v=0.7134) | game=7.2s train=87.4s
+[2026-06-25 01:32:05] game= 4463 | 1/2-1/2 | moves= 52 | buf=33,981 | steps=4,463 | loss=9.0120 (p=8.4493 v=0.5627) | game=5.6s train=87.2s
+[2026-06-25 01:33:38] game= 4464 | 1/2-1/2 | moves= 56 | buf=34,037 | steps=4,464 | loss=8.8832 (p=8.4493 v=0.4339) | game=5.8s train=87.3s
+[2026-06-25 01:35:11] game= 4465 | 1/2-1/2 | moves= 50 | buf=34,087 | steps=4,465 | loss=8.8590 (p=8.4493 v=0.4097) | game=5.4s train=87.0s
+[2026-06-25 01:36:45] game= 4466 | 1/2-1/2 | moves= 61 | buf=34,148 | steps=4,466 | loss=8.7465 (p=8.4493 v=0.2972) | game=6.3s train=87.4s
+[2026-06-25 01:38:17] game= 4467 | 1/2-1/2 | moves= 49 | buf=34,197 | steps=4,467 | loss=8.6810 (p=8.4493 v=0.2317) | game=5.2s train=87.3s
+[2026-06-25 01:39:50] game= 4468 | 1/2-1/2 | moves= 49 | buf=34,246 | steps=4,468 | loss=8.6363 (p=8.4493 v=0.1870) | game=4.9s train=87.6s
+[2026-06-25 01:41:23] game= 4469 | 1/2-1/2 | moves= 57 | buf=34,303 | steps=4,469 | loss=8.6573 (p=8.4493 v=0.2080) | game=5.7s train=87.2s
+[2026-06-25 01:42:56] game= 4470 | 1/2-1/2 | moves= 57 | buf=34,360 | steps=4,470 | loss=8.6183 (p=8.4493 v=0.1690) | game=6.5s train=87.4s