train.log

train.log

@@ -183,3 +183,14 @@
 [13:04:41] ckpt step_002100.pt + latest.pt (live) steps=2100 status=mid kind=full
 [13:04:49] val/lichess_opening hard_ce=1.4748 soft_ce=1.4748 soft_temp_ce=1.4748 teacher_entropy=0.0000 teacher_kl=1.4748 wdl_ce=1.0382 value_rows=8192.0000
 [13:05:12] step 2125/8000 | loss=1.7960 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=1122000/0/0 q=0 p=0 | 461 pos/s | vram=35.72GB
+[13:05:34] step 2150/8000 | loss=1.8058 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=1135200/0/0 q=0 p=0 | 462 pos/s | vram=35.72GB
+[13:05:55] step 2175/8000 | loss=1.7962 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=1148400/0/0 q=0 p=0 | 463 pos/s | vram=35.72GB
+[13:06:17] step 2200/8000 | loss=1.8061 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=1161600/0/0 q=0 p=0 | 465 pos/s | vram=35.72GB
+[13:06:26] val/lichess_opening hard_ce=1.4697 soft_ce=1.4697 soft_temp_ce=1.4697 teacher_entropy=0.0000 teacher_kl=1.4697 wdl_ce=1.0374 value_rows=8192.0000
+[13:06:48] step 2225/8000 | loss=1.8121 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=1174800/0/0 q=0 p=0 | 464 pos/s | vram=35.72GB
+[13:07:10] step 2250/8000 | loss=1.8013 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=1188000/0/0 q=0 p=0 | 465 pos/s | vram=35.72GB
+[13:07:11] ckpt latest.pt (live weights) steps=2250 status=mid
+[13:07:33] step 2275/8000 | loss=1.7889 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=1201200/0/0 q=0 p=0 | 466 pos/s | vram=35.72GB
+[13:07:55] step 2300/8000 | loss=1.8014 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=1214400/0/0 q=0 p=0 | 467 pos/s | vram=35.72GB
+[13:08:03] val/lichess_opening hard_ce=1.4756 soft_ce=1.4756 soft_temp_ce=1.4756 teacher_entropy=0.0000 teacher_kl=1.4756 wdl_ce=1.0368 value_rows=8192.0000
+[13:08:25] step 2325/8000 | loss=1.8094 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=1227600/0/0 q=0 p=0 | 467 pos/s | vram=35.72GB