checkpoint @ game=3170 steps=3170

training_log.txt

@@ -3469,3 +3469,13 @@
 [2026-06-21 11:48:26] game= 3158 | 1/2-1/2 | moves= 54 | buf=50,000 | steps=3,158 | loss=9.1852 (p=8.4493 v=0.7359) | game=8.2s train=97.0s
 [2026-06-21 11:50:11] game= 3159 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=3,159 | loss=9.1074 (p=8.4493 v=0.6581) | game=5.0s train=100.8s
 [2026-06-21 11:51:53] game= 3160 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=3,160 | loss=8.9951 (p=8.4493 v=0.5458) | game=5.1s train=96.8s
+[2026-06-21 11:53:39] game= 3161 | 1/2-1/2 | moves= 57 | buf=50,000 | steps=3,161 | loss=8.8766 (p=8.4493 v=0.4273) | game=5.9s train=96.8s
+[2026-06-21 11:55:24] game= 3162 | 1/2-1/2 | moves= 81 | buf=50,000 | steps=3,162 | loss=8.8620 (p=8.4493 v=0.4127) | game=8.3s train=97.1s
+[2026-06-21 11:57:08] game= 3163 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=3,163 | loss=8.8076 (p=8.4493 v=0.3583) | game=6.2s train=97.3s
+[2026-06-21 11:58:52] game= 3164 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=3,164 | loss=8.7714 (p=8.4493 v=0.3221) | game=6.9s train=97.3s
+[2026-06-21 12:00:34] game= 3165 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=3,165 | loss=8.7585 (p=8.4493 v=0.3092) | game=5.3s train=97.0s
+[2026-06-21 12:02:17] game= 3166 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=3,166 | loss=8.7295 (p=8.4493 v=0.2802) | game=5.0s train=97.3s
+[2026-06-21 12:04:00] game= 3167 | 1/2-1/2 | moves= 60 | buf=50,000 | steps=3,167 | loss=8.7069 (p=8.4493 v=0.2576) | game=6.1s train=97.3s
+[2026-06-21 12:05:48] game= 3168 | 1/2-1/2 | moves= 52 | buf=50,000 | steps=3,168 | loss=8.6873 (p=8.4493 v=0.2380) | game=10.3s train=97.6s
+[2026-06-21 12:07:31] game= 3169 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=3,169 | loss=8.7241 (p=8.4493 v=0.2748) | game=5.3s train=97.5s
+[2026-06-21 12:09:15] game= 3170 | 1/2-1/2 | moves= 61 | buf=50,000 | steps=3,170 | loss=8.6453 (p=8.4493 v=0.1960) | game=6.3s train=97.4s