avewright commited on
Commit
08080f6
·
verified ·
1 Parent(s): 8afc664

train.log

Browse files
Files changed (1) hide show
  1. train.log +15 -0
train.log CHANGED
@@ -66,3 +66,18 @@
66
  [12:34:55] step 550/8000 | loss=1.8243 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=290400/0/0 q=0 p=0 | 470 pos/s | vram=35.72GB
67
  [12:35:02] soft inbox +480,000 train_n=12,314,337 shards=shard_000149,shard_000150,shard_000151,shard_000152,shard_000153,shard_000154
68
  [12:35:24] step 575/8000 | loss=1.8222 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=303600/0/0 q=0 p=0 | 469 pos/s | vram=35.72GB
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
66
  [12:34:55] step 550/8000 | loss=1.8243 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=290400/0/0 q=0 p=0 | 470 pos/s | vram=35.72GB
67
  [12:35:02] soft inbox +480,000 train_n=12,314,337 shards=shard_000149,shard_000150,shard_000151,shard_000152,shard_000153,shard_000154
68
  [12:35:24] step 575/8000 | loss=1.8222 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=303600/0/0 q=0 p=0 | 469 pos/s | vram=35.72GB
69
+ [12:35:31] soft inbox +491,287 train_n=12,805,624 shards=shard_000155,shard_000156,shard_000157,shard_000158,shard_000159,shard_000160,shard_000161
70
+ [12:35:53] step 600/8000 | loss=1.8233 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=316800/0/0 q=0 p=0 | 469 pos/s | vram=35.72GB
71
+ [12:35:57] soft inbox +320,000 train_n=13,125,624 shards=shard_000162,shard_000163,shard_000164,shard_000165
72
+ [12:35:58] ckpt step_000600.pt + latest.pt (live) steps=600 status=mid kind=full
73
+ [12:36:07] val/lichess_opening hard_ce=1.5066 soft_ce=1.5066 soft_temp_ce=1.5066 teacher_entropy=0.0000 teacher_kl=1.5066 wdl_ce=1.0391 value_rows=8192.0000
74
+ [12:36:28] step 625/8000 | loss=1.8204 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=330000/0/0 q=0 p=0 | 464 pos/s | vram=35.72GB
75
+ [12:36:37] soft inbox +640,000 train_n=13,765,624 shards=shard_000166,shard_000167,shard_000168,shard_000169,shard_000170,shard_000171,shard_000172,shard_000173
76
+ [12:36:59] step 650/8000 | loss=1.8283 | lr=2.00e-03,3.00e-05 | clip=4/25 | mix s/d/b=343200/0/0 q=0 p=0 | 462 pos/s | vram=35.72GB
77
+ [12:37:07] soft inbox +480,000 train_n=14,245,624 shards=shard_000174,shard_000175,shard_000176,shard_000177,shard_000178,shard_000179
78
+ [12:37:29] step 675/8000 | loss=1.8030 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=356400/0/0 q=0 p=0 | 461 pos/s | vram=35.72GB
79
+ [12:37:38] soft inbox +480,000 train_n=14,725,624 shards=shard_000180,shard_000181,shard_000182,shard_000183,shard_000184,shard_000185
80
+ [12:38:00] step 700/8000 | loss=1.8285 | lr=2.00e-03,3.00e-05 | clip=3/25 | mix s/d/b=369600/0/0 q=0 p=0 | 460 pos/s | vram=35.72GB
81
+ [12:38:10] soft inbox +480,000 train_n=15,205,624 shards=shard_000186,shard_000187,shard_000188,shard_000189,shard_000190,shard_000191
82
+ [12:38:19] val/lichess_opening hard_ce=1.5085 soft_ce=1.5085 soft_temp_ce=1.5085 teacher_entropy=0.0000 teacher_kl=1.5085 wdl_ce=1.0377 value_rows=8192.0000
83
+ [12:38:41] step 725/8000 | loss=1.8271 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=382800/0/0 q=0 p=0 | 454 pos/s | vram=35.72GB