checkpoint @ game=640 steps=640

training_log.txt

@@ -931,3 +931,13 @@
 [2026-06-14 11:04:21] game=  628 | 1/2-1/2 | moves= 60 | buf=70,000 | steps=628 | loss=8.8403 (p=8.4493 v=0.3910) | game=7.0s train=162.1s
 [2026-06-14 11:07:08] game=  629 | 1/2-1/2 | moves= 53 | buf=70,000 | steps=629 | loss=8.8186 (p=8.4493 v=0.3693) | game=5.2s train=161.9s
 [2026-06-14 11:09:57] game=  630 | 1/2-1/2 | moves= 57 | buf=70,000 | steps=630 | loss=8.8143 (p=8.4493 v=0.3651) | game=6.9s train=161.9s
+[2026-06-14 11:12:49] game=  631 | 1/2-1/2 | moves= 73 | buf=70,000 | steps=631 | loss=8.7772 (p=8.4493 v=0.3279) | game=7.1s train=162.0s
+[2026-06-14 11:15:37] game=  632 | 1/2-1/2 | moves= 51 | buf=70,000 | steps=632 | loss=8.7668 (p=8.4493 v=0.3175) | game=4.9s train=163.5s
+[2026-06-14 11:18:25] game=  633 | 1/2-1/2 | moves= 60 | buf=70,000 | steps=633 | loss=8.7165 (p=8.4493 v=0.2672) | game=5.9s train=162.2s
+[2026-06-14 11:21:14] game=  634 | 1/2-1/2 | moves= 53 | buf=70,000 | steps=634 | loss=8.7068 (p=8.4493 v=0.2575) | game=6.6s train=162.2s
+[2026-06-14 11:24:02] game=  635 | 1/2-1/2 | moves= 61 | buf=70,000 | steps=635 | loss=8.6715 (p=8.4493 v=0.2222) | game=6.0s train=162.2s
+[2026-06-14 11:26:50] game=  636 | 1/2-1/2 | moves= 61 | buf=70,000 | steps=636 | loss=8.6453 (p=8.4493 v=0.1960) | game=5.9s train=162.1s
+[2026-06-14 11:29:39] game=  637 | 1/2-1/2 | moves= 52 | buf=70,000 | steps=637 | loss=8.6425 (p=8.4493 v=0.1932) | game=6.4s train=162.2s
+[2026-06-14 11:32:26] game=  638 | 1/2-1/2 | moves= 48 | buf=70,000 | steps=638 | loss=8.6435 (p=8.4493 v=0.1942) | game=4.7s train=162.2s
+[2026-06-14 11:35:13] game=  639 | 1/2-1/2 | moves= 54 | buf=70,000 | steps=639 | loss=8.6119 (p=8.4493 v=0.1626) | game=5.2s train=162.2s
+[2026-06-14 11:38:02] game=  640 | 1/2-1/2 | moves= 52 | buf=70,000 | steps=640 | loss=8.6130 (p=8.4493 v=0.1637) | game=6.4s train=162.2s