train.log
Browse files
train.log
CHANGED
|
@@ -94,3 +94,27 @@
|
|
| 94 |
[12:41:01] soft inbox +640,000 train_n=17,712,850 shards=shard_000216,shard_000217,shard_000218,shard_000219,shard_000220,shard_000221,shard_000222,shard_000223
|
| 95 |
[12:41:23] step 850/8000 | loss=1.8443 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=448800/0/0 q=0 p=0 | 446 pos/s | vram=35.72GB
|
| 96 |
[12:41:33] soft inbox +480,000 train_n=18,192,850 shards=shard_000224,shard_000225,shard_000226,shard_000227,shard_000228,shard_000229
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 94 |
[12:41:01] soft inbox +640,000 train_n=17,712,850 shards=shard_000216,shard_000217,shard_000218,shard_000219,shard_000220,shard_000221,shard_000222,shard_000223
|
| 95 |
[12:41:23] step 850/8000 | loss=1.8443 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=448800/0/0 q=0 p=0 | 446 pos/s | vram=35.72GB
|
| 96 |
[12:41:33] soft inbox +480,000 train_n=18,192,850 shards=shard_000224,shard_000225,shard_000226,shard_000227,shard_000228,shard_000229
|
| 97 |
+
[12:41:55] step 875/8000 | loss=1.8190 | lr=2.00e-03,3.00e-05 | clip=1/25 | mix s/d/b=462000/0/0 q=0 p=0 | 445 pos/s | vram=35.72GB
|
| 98 |
+
[12:42:02] soft inbox +462,199 train_n=18,655,049 shards=shard_000230,shard_000231,shard_000232,shard_000233,shard_000234,shard_000235
|
| 99 |
+
[12:42:24] step 900/8000 | loss=1.8286 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=475200/0/0 q=0 p=0 | 445 pos/s | vram=35.72GB
|
| 100 |
+
[12:42:32] soft inbox +400,000 train_n=19,055,049 shards=shard_000236,shard_000237,shard_000238,shard_000239,shard_000240
|
| 101 |
+
[12:42:33] ckpt step_000900.pt + latest.pt (live) steps=900 status=mid kind=full
|
| 102 |
+
[12:42:42] val/lichess_opening hard_ce=1.5039 soft_ce=1.5039 soft_temp_ce=1.5039 teacher_entropy=0.0000 teacher_kl=1.5039 wdl_ce=1.0374 value_rows=8192.0000
|
| 103 |
+
[12:43:04] step 925/8000 | loss=1.8387 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=488400/0/0 q=0 p=0 | 441 pos/s | vram=35.72GB
|
| 104 |
+
[12:43:16] soft inbox +560,000 train_n=19,615,049 shards=shard_000241,shard_000242,shard_000243,shard_000244,shard_000245,shard_000246,shard_000247
|
| 105 |
+
[12:43:38] step 950/8000 | loss=1.8129 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=501600/0/0 q=0 p=0 | 439 pos/s | vram=35.72GB
|
| 106 |
+
[12:43:51] soft inbox +560,000 train_n=20,175,049 shards=shard_000248,shard_000249,shard_000250,shard_000251,shard_000252,shard_000253,shard_000254
|
| 107 |
+
[12:44:13] step 975/8000 | loss=1.8227 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=514800/0/0 q=0 p=0 | 438 pos/s | vram=35.72GB
|
| 108 |
+
[12:44:23] soft inbox +480,000 train_n=20,655,049 shards=shard_000255,shard_000256,shard_000257,shard_000258,shard_000259,shard_000260
|
| 109 |
+
[12:44:45] step 1000/8000 | loss=1.8152 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=528000/0/0 q=0 p=0 | 437 pos/s | vram=35.72GB
|
| 110 |
+
[12:44:55] soft inbox +400,000 train_n=21,055,049 shards=shard_000261,shard_000262,shard_000263,shard_000264,shard_000265
|
| 111 |
+
[12:45:04] val/lichess_opening hard_ce=1.4990 soft_ce=1.4990 soft_temp_ce=1.4990 teacher_entropy=0.0000 teacher_kl=1.4990 wdl_ce=1.0374 value_rows=8192.0000
|
| 112 |
+
[12:45:26] step 1025/8000 | loss=1.8247 | lr=2.00e-03,3.00e-05 | clip=3/25 | mix s/d/b=541200/0/0 q=0 p=0 | 433 pos/s | vram=35.72GB
|
| 113 |
+
[12:45:35] soft inbox +517,282 train_n=21,572,331 shards=shard_000266,shard_000267,shard_000268,shard_000269,shard_000270,shard_000271,shard_000272
|
| 114 |
+
[12:45:57] step 1050/8000 | loss=1.8175 | lr=2.00e-03,3.00e-05 | clip=3/25 | mix s/d/b=554400/0/0 q=0 p=0 | 433 pos/s | vram=35.72GB
|
| 115 |
+
[12:46:06] soft inbox +399,996 train_n=21,972,327 shards=shard_000273,shard_000274,shard_000275,shard_000276,shard_000277
|
| 116 |
+
[12:46:06] ckpt latest.pt (live weights) steps=1050 status=mid
|
| 117 |
+
[12:46:28] step 1075/8000 | loss=1.8138 | lr=2.00e-03,3.00e-05 | clip=0/25 | mix s/d/b=567600/0/0 q=0 p=0 | 433 pos/s | vram=35.72GB
|
| 118 |
+
[12:46:37] soft inbox +400,000 train_n=22,372,327 shards=shard_000278,shard_000279,shard_000280,shard_000281,shard_000282
|
| 119 |
+
[12:46:59] step 1100/8000 | loss=1.8012 | lr=2.00e-03,3.00e-05 | clip=2/25 | mix s/d/b=580800/0/0 q=0 p=0 | 433 pos/s | vram=35.72GB
|
| 120 |
+
[12:47:09] soft inbox +400,000 train_n=22,772,327 shards=shard_000283,shard_000284,shard_000285,shard_000286,shard_000287
|