train.log

train.log

@@ -480,3 +480,14 @@
 [14:29:49] step 7350/8000 | loss=1.7458 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=3880800/0/0 q=0 p=0 | 517 pos/s | vram=35.72GB
 [14:29:50] ckpt latest.pt (live weights) steps=7350 status=mid
 [14:30:12] step 7375/8000 | loss=1.7680 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=3894000/0/0 q=0 p=0 | 517 pos/s | vram=35.72GB
+[14:30:34] step 7400/8000 | loss=1.7627 | lr=2.00e-03,3.00e-05 | clip=3/25 | mix s/d/b=3907200/0/0 q=0 p=0 | 517 pos/s | vram=35.72GB
+[14:30:43] val/lichess_opening hard_ce=1.4139 soft_ce=1.4139 soft_temp_ce=1.4139 teacher_entropy=0.0000 teacher_kl=1.4139 wdl_ce=1.0345 value_rows=8192.0000
+[14:31:04] step 7425/8000 | loss=1.7510 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=3920400/0/0 q=0 p=0 | 517 pos/s | vram=35.72GB
+[14:31:26] step 7450/8000 | loss=1.7519 | lr=2.00e-03,3.00e-05 | clip=3/25 | mix s/d/b=3933600/0/0 q=0 p=0 | 517 pos/s | vram=35.72GB
+[14:31:48] step 7475/8000 | loss=1.7748 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=3946800/0/0 q=0 p=0 | 517 pos/s | vram=35.72GB
+[14:32:10] step 7500/8000 | loss=1.7359 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=3960000/0/0 q=0 p=0 | 517 pos/s | vram=35.72GB
+[14:32:11] ckpt step_007500.pt + latest.pt (live) steps=7500 status=mid kind=full
+[14:32:20] val/lichess_opening hard_ce=1.4156 soft_ce=1.4156 soft_temp_ce=1.4156 teacher_entropy=0.0000 teacher_kl=1.4156 wdl_ce=1.0341 value_rows=8192.0000
+[14:32:42] step 7525/8000 | loss=1.7756 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=3973200/0/0 q=0 p=0 | 517 pos/s | vram=35.72GB
+[14:33:03] step 7550/8000 | loss=1.7623 | lr=2.00e-03,3.00e-05 | clip=3/25 | mix s/d/b=3986400/0/0 q=0 p=0 | 517 pos/s | vram=35.72GB
+[14:33:25] step 7575/8000 | loss=1.7640 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=3999600/0/0 q=0 p=0 | 518 pos/s | vram=35.72GB