train.log

train.log

@@ -251,3 +251,14 @@
 [16:06:16] ckpt step_004200.pt + latest.pt (live) steps=4200 status=mid kind=full
 [16:06:25] val/lichess_middlegame hard_ce=1.6322 soft_ce=1.6322 soft_temp_ce=1.6322 teacher_entropy=0.0000 teacher_kl=1.6322 wdl_ce=0.8297 value_rows=8192.0000
 [16:06:47] step 4225/8000 | loss=1.8335 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=2230800/0/0 q=0 p=0 | 545 pos/s | vram=35.73GB
+[16:07:09] step 4250/8000 | loss=1.8639 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=2244000/0/0 q=0 p=0 | 545 pos/s | vram=35.73GB
+[16:07:30] step 4275/8000 | loss=1.8515 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2257200/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:07:52] step 4300/8000 | loss=1.8325 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=2270400/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:08:01] val/lichess_middlegame hard_ce=1.6345 soft_ce=1.6345 soft_temp_ce=1.6345 teacher_entropy=0.0000 teacher_kl=1.6345 wdl_ce=0.8299 value_rows=8192.0000
+[16:08:23] step 4325/8000 | loss=1.8261 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2283600/0/0 q=0 p=0 | 545 pos/s | vram=35.73GB
+[16:08:45] step 4350/8000 | loss=1.8384 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=2296800/0/0 q=0 p=0 | 545 pos/s | vram=35.73GB
+[16:08:45] ckpt latest.pt (live weights) steps=4350 status=mid
+[16:09:07] step 4375/8000 | loss=1.8480 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2310000/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:09:29] step 4400/8000 | loss=1.8481 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=2323200/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:09:37] val/lichess_middlegame hard_ce=1.6352 soft_ce=1.6352 soft_temp_ce=1.6352 teacher_entropy=0.0000 teacher_kl=1.6352 wdl_ce=0.8292 value_rows=8192.0000
+[16:09:59] step 4425/8000 | loss=1.8501 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=2336400/0/0 q=0 p=0 | 545 pos/s | vram=35.73GB