train.log

train.log

@@ -335,3 +335,14 @@
 [13:48:24] step 4800/8000 | loss=1.7870 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2534400/0/0 q=0 p=0 | 504 pos/s | vram=35.72GB
 [13:48:25] ckpt step_004800.pt + latest.pt (live) steps=4800 status=mid kind=full
 [13:48:34] val/lichess_opening hard_ce=1.4479 soft_ce=1.4479 soft_temp_ce=1.4479 teacher_entropy=0.0000 teacher_kl=1.4479 wdl_ce=1.0345 value_rows=8192.0000
+[13:48:56] step 4825/8000 | loss=1.7853 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2547600/0/0 q=0 p=0 | 504 pos/s | vram=35.72GB
+[13:49:18] step 4850/8000 | loss=1.7490 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2560800/0/0 q=0 p=0 | 504 pos/s | vram=35.72GB
+[13:49:40] step 4875/8000 | loss=1.7564 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=2574000/0/0 q=0 p=0 | 504 pos/s | vram=35.72GB
+[13:50:02] step 4900/8000 | loss=1.7534 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2587200/0/0 q=0 p=0 | 505 pos/s | vram=35.72GB
+[13:50:10] val/lichess_opening hard_ce=1.4395 soft_ce=1.4395 soft_temp_ce=1.4395 teacher_entropy=0.0000 teacher_kl=1.4395 wdl_ce=1.0347 value_rows=8192.0000
+[13:50:32] step 4925/8000 | loss=1.7941 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2600400/0/0 q=0 p=0 | 504 pos/s | vram=35.72GB
+[13:50:55] step 4950/8000 | loss=1.7669 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2613600/0/0 q=0 p=0 | 505 pos/s | vram=35.72GB
+[13:50:56] ckpt latest.pt (live weights) steps=4950 status=mid
+[13:51:18] step 4975/8000 | loss=1.7636 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2626800/0/0 q=0 p=0 | 505 pos/s | vram=35.72GB
+[13:51:40] step 5000/8000 | loss=1.7778 | lr=2.00e-03,3.00e-05 | clip=3/25 | mix s/d/b=2640000/0/0 q=0 p=0 | 505 pos/s | vram=35.72GB
+[13:51:49] val/lichess_opening hard_ce=1.4362 soft_ce=1.4362 soft_temp_ce=1.4362 teacher_entropy=0.0000 teacher_kl=1.4362 wdl_ce=1.0361 value_rows=8192.0000