mtasic85 commited on
Commit
12444b3
·
1 Parent(s): c238f37
Files changed (1) hide show
  1. README.md +27 -0
README.md CHANGED
@@ -92,6 +92,33 @@ Epoch 1 | iter 4608 step 18 | loss train: 8.654, val: n/a | iter time: 1246.47 m
92
  Epoch 1 | iter 4864 step 19 | loss train: 8.651, val: n/a | iter time: 1246.56 ms (step) remaining time: 3 days, 23:41:11
93
  Epoch 1 | iter 5120 step 20 | loss train: 8.639, val: n/a | iter time: 1219.66 ms (step) remaining time: 3 days, 23:35:38
94
  # ...
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
95
  ```
96
 
97
  Backup `wandb`:
 
92
  Epoch 1 | iter 4864 step 19 | loss train: 8.651, val: n/a | iter time: 1246.56 ms (step) remaining time: 3 days, 23:41:11
93
  Epoch 1 | iter 5120 step 20 | loss train: 8.639, val: n/a | iter time: 1219.66 ms (step) remaining time: 3 days, 23:35:38
94
  # ...
95
+ Epoch 1 | iter 442880 step 1730 | loss train: 2.740, val: 2.863 | iter time: 1340.98 ms (step) remaining time: 0:51:28
96
+ Epoch 1 | iter 443136 step 1731 | loss train: 2.734, val: 2.863 | iter time: 1387.92 ms (step) remaining time: 0:48:00
97
+ Epoch 1 | iter 443392 step 1732 | loss train: 2.730, val: 2.863 | iter time: 1309.36 ms (step) remaining time: 0:44:31
98
+ Epoch 1 | iter 443648 step 1733 | loss train: 2.715, val: 2.863 | iter time: 1292.23 ms (step) remaining time: 0:41:03
99
+ Epoch 1 | iter 443904 step 1734 | loss train: 2.718, val: 2.863 | iter time: 1311.24 ms (step) remaining time: 0:37:35
100
+ Epoch 1 | iter 444160 step 1735 | loss train: 2.709, val: 2.863 | iter time: 1291.09 ms (step) remaining time: 0:34:07
101
+ Epoch 1 | iter 444416 step 1736 | loss train: 2.723, val: 2.863 | iter time: 1304.14 ms (step) remaining time: 0:30:39
102
+ Epoch 1 | iter 444672 step 1737 | loss train: 2.721, val: 2.863 | iter time: 1278.33 ms (step) remaining time: 0:27:10
103
+ Epoch 1 | iter 444928 step 1738 | loss train: 2.697, val: 2.863 | iter time: 1292.86 ms (step) remaining time: 0:23:42
104
+ Epoch 1 | iter 445184 step 1739 | loss train: 2.763, val: 2.863 | iter time: 1284.40 ms (step) remaining time: 0:20:14
105
+ Epoch 1 | iter 445440 step 1740 | loss train: 2.775, val: 2.863 | iter time: 1302.58 ms (step) remaining time: 0:16:46
106
+ Epoch 1 | iter 445696 step 1741 | loss train: 2.756, val: 2.863 | iter time: 1298.86 ms (step) remaining time: 0:13:18
107
+ Epoch 1 | iter 445952 step 1742 | loss train: 2.728, val: 2.863 | iter time: 1279.11 ms (step) remaining time: 0:09:49
108
+ Epoch 1 | iter 446208 step 1743 | loss train: 2.637, val: 2.863 | iter time: 1308.11 ms (step) remaining time: 0:06:21
109
+ Epoch 1 | iter 446464 step 1744 | loss train: 2.638, val: 2.863 | iter time: 1294.08 ms (step) remaining time: 0:02:53
110
+ Validating ...
111
+ Final evaluation | val loss: 2.862 | val ppl: 17.494
112
+ Saving checkpoint to '../out/pretrain-core/final/lit_model.pth'
113
+ ----------------------------------------
114
+ | Performance
115
+ | - Total tokens : 7,318,355,968
116
+ | - Training Time : 363457.29 s
117
+ | - Tok/sec : 2103064.60 tok/s
118
+ | ----------------------------------------
119
+ | Memory Usage
120
+ | - Memory Used : 20.93 GB
121
+ ----------------------------------------
122
  ```
123
 
124
  Backup `wandb`: