checkpoint @ game=4090 steps=4090

training_log.txt

@@ -4389,3 +4389,13 @@
 [2026-06-24 01:54:30] game= 4078 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=4,078 | loss=8.6408 (p=8.4493 v=0.1915) | game=5.0s train=87.7s
 [2026-06-24 01:56:03] game= 4079 | 1/2-1/2 | moves= 63 | buf=50,000 | steps=4,079 | loss=8.6321 (p=8.4493 v=0.1828) | game=6.2s train=87.1s
 [2026-06-24 01:57:35] game= 4080 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=4,080 | loss=8.6081 (p=8.4493 v=0.1588) | game=4.8s train=87.2s
+[2026-06-24 01:59:10] game= 4081 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=4,081 | loss=8.5828 (p=8.4493 v=0.1335) | game=5.0s train=87.2s
+[2026-06-24 02:00:45] game= 4082 | 1/2-1/2 | moves= 79 | buf=50,000 | steps=4,082 | loss=8.6398 (p=8.4493 v=0.1905) | game=7.7s train=87.3s
+[2026-06-24 02:02:18] game= 4083 | 1/2-1/2 | moves= 51 | buf=50,000 | steps=4,083 | loss=8.5975 (p=8.4493 v=0.1482) | game=5.0s train=87.3s
+[2026-06-24 02:03:54] game= 4084 | 1/2-1/2 | moves= 67 | buf=50,000 | steps=4,084 | loss=9.1823 (p=8.4493 v=0.7330) | game=9.3s train=87.4s
+[2026-06-24 02:05:27] game= 4085 | 1/2-1/2 | moves= 53 | buf=50,000 | steps=4,085 | loss=9.0360 (p=8.4493 v=0.5867) | game=5.2s train=87.4s
+[2026-06-24 02:06:59] game= 4086 | 1/2-1/2 | moves= 46 | buf=50,000 | steps=4,086 | loss=8.9325 (p=8.4493 v=0.4832) | game=4.5s train=87.4s
+[2026-06-24 02:08:31] game= 4087 | 1/2-1/2 | moves= 50 | buf=50,000 | steps=4,087 | loss=8.8465 (p=8.4493 v=0.3972) | game=4.9s train=87.5s
+[2026-06-24 02:10:04] game= 4088 | 1/2-1/2 | moves= 59 | buf=50,000 | steps=4,088 | loss=8.8293 (p=8.4493 v=0.3800) | game=5.8s train=86.7s
+[2026-06-24 02:11:38] game= 4089 | 1/2-1/2 | moves= 62 | buf=50,000 | steps=4,089 | loss=8.7802 (p=8.4493 v=0.3309) | game=6.2s train=87.5s
+[2026-06-24 02:13:10] game= 4090 | 1/2-1/2 | moves= 49 | buf=50,000 | steps=4,090 | loss=8.7026 (p=8.4493 v=0.2533) | game=4.9s train=87.5s