train.log

train.log

@@ -370,3 +370,14 @@
 [16:40:02] ckpt step_006300.pt + latest.pt (live) steps=6300 status=mid kind=full
 [16:40:11] val/lichess_middlegame hard_ce=1.6117 soft_ce=1.6117 soft_temp_ce=1.6117 teacher_entropy=0.0000 teacher_kl=1.6117 wdl_ce=0.8249 value_rows=8192.0000
 [16:40:33] step 6325/8000 | loss=1.8411 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=3339600/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:40:55] step 6350/8000 | loss=1.8262 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=3352800/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:41:16] step 6375/8000 | loss=1.8238 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=3366000/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:41:38] step 6400/8000 | loss=1.8276 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=3379200/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:41:47] val/lichess_middlegame hard_ce=1.6195 soft_ce=1.6195 soft_temp_ce=1.6195 teacher_entropy=0.0000 teacher_kl=1.6195 wdl_ce=0.8262 value_rows=8192.0000
+[16:42:08] step 6425/8000 | loss=1.8324 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=3392400/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:42:30] step 6450/8000 | loss=1.8147 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=3405600/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:42:31] ckpt latest.pt (live weights) steps=6450 status=mid
+[16:42:52] step 6475/8000 | loss=1.8208 | lr=2.00e-03,3.00e-05 | clip=3/25 | mix s/d/b=3418800/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:43:14] step 6500/8000 | loss=1.8249 | lr=2.00e-03,3.00e-05 | clip=3/25 | mix s/d/b=3432000/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:43:23] val/lichess_middlegame hard_ce=1.6179 soft_ce=1.6179 soft_temp_ce=1.6179 teacher_entropy=0.0000 teacher_kl=1.6179 wdl_ce=0.8244 value_rows=8192.0000
+[16:43:45] step 6525/8000 | loss=1.8190 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=3445200/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB