train.log

train.log

@@ -324,3 +324,15 @@
 [16:27:11] step 5500/8000 | loss=1.8037 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2904000/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
 [16:27:20] val/lichess_middlegame hard_ce=1.6268 soft_ce=1.6268 soft_temp_ce=1.6268 teacher_entropy=0.0000 teacher_kl=1.6268 wdl_ce=0.8305 value_rows=8192.0000
 [16:27:41] step 5525/8000 | loss=1.8364 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=2917200/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:28:03] step 5550/8000 | loss=1.8063 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2930400/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:28:04] ckpt latest.pt (live weights) steps=5550 status=mid
+[16:28:25] step 5575/8000 | loss=1.8305 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=2943600/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:28:47] step 5600/8000 | loss=1.8482 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=2956800/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:28:56] val/lichess_middlegame hard_ce=1.6249 soft_ce=1.6249 soft_temp_ce=1.6249 teacher_entropy=0.0000 teacher_kl=1.6249 wdl_ce=0.8273 value_rows=8192.0000
+[16:29:18] step 5625/8000 | loss=1.8089 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=2970000/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:29:40] step 5650/8000 | loss=1.8206 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2983200/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:30:02] step 5675/8000 | loss=1.8187 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=2996400/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:30:24] step 5700/8000 | loss=1.8218 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=3009600/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB
+[16:30:24] ckpt step_005700.pt + latest.pt (live) steps=5700 status=mid kind=full
+[16:30:33] val/lichess_middlegame hard_ce=1.6264 soft_ce=1.6264 soft_temp_ce=1.6264 teacher_entropy=0.0000 teacher_kl=1.6264 wdl_ce=0.8264 value_rows=8192.0000
+[16:30:55] step 5725/8000 | loss=1.8291 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=3022800/0/0 q=0 p=0 | 546 pos/s | vram=35.73GB