Pittawat Taveekitworachai
commited on
Commit
•
c629928
1
Parent(s):
e7922dc
🍻 cheers
Browse files- all_results.json +13 -0
- runs/Mar20_12-25-02_cde0d66f7c83/events.out.tfevents.1679318292.cde0d66f7c83.23.2 +3 -0
- test_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +2263 -0
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9880769230769231,
|
4 |
+
"eval_loss": 0.051459942013025284,
|
5 |
+
"eval_runtime": 20.2768,
|
6 |
+
"eval_samples_per_second": 128.225,
|
7 |
+
"eval_steps_per_second": 8.039,
|
8 |
+
"total_flos": 8.0609004822528e+18,
|
9 |
+
"train_loss": 0.11576663036481334,
|
10 |
+
"train_runtime": 3043.19,
|
11 |
+
"train_samples_per_second": 34.175,
|
12 |
+
"train_steps_per_second": 1.069
|
13 |
+
}
|
runs/Mar20_12-25-02_cde0d66f7c83/events.out.tfevents.1679318292.cde0d66f7c83.23.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ee9b0b797115af58ca6cb093c39c560dc6db971644a3fffda53e9daef0066daa
|
3 |
+
size 363
|
test_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9880769230769231,
|
4 |
+
"eval_loss": 0.051459942013025284,
|
5 |
+
"eval_runtime": 20.2768,
|
6 |
+
"eval_samples_per_second": 128.225,
|
7 |
+
"eval_steps_per_second": 8.039
|
8 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"total_flos": 8.0609004822528e+18,
|
4 |
+
"train_loss": 0.11576663036481334,
|
5 |
+
"train_runtime": 3043.19,
|
6 |
+
"train_samples_per_second": 34.175,
|
7 |
+
"train_steps_per_second": 1.069
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,2263 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.051459942013025284,
|
3 |
+
"best_model_checkpoint": "./vit-base-letter/checkpoint-2900",
|
4 |
+
"epoch": 4.0,
|
5 |
+
"global_step": 3252,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"learning_rate": 0.0001993849938499385,
|
13 |
+
"loss": 3.1242,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.02,
|
18 |
+
"learning_rate": 0.000198769987699877,
|
19 |
+
"loss": 2.6937,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.04,
|
24 |
+
"learning_rate": 0.00019815498154981552,
|
25 |
+
"loss": 2.1871,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.05,
|
30 |
+
"learning_rate": 0.000197539975399754,
|
31 |
+
"loss": 1.7584,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.06,
|
36 |
+
"learning_rate": 0.0001969249692496925,
|
37 |
+
"loss": 1.428,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.07,
|
42 |
+
"learning_rate": 0.000196309963099631,
|
43 |
+
"loss": 1.1022,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.09,
|
48 |
+
"learning_rate": 0.0001956949569495695,
|
49 |
+
"loss": 0.862,
|
50 |
+
"step": 70
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.1,
|
54 |
+
"learning_rate": 0.000195079950799508,
|
55 |
+
"loss": 0.7286,
|
56 |
+
"step": 80
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.11,
|
60 |
+
"learning_rate": 0.00019446494464944652,
|
61 |
+
"loss": 0.6413,
|
62 |
+
"step": 90
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.12,
|
66 |
+
"learning_rate": 0.000193849938499385,
|
67 |
+
"loss": 0.5539,
|
68 |
+
"step": 100
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.12,
|
72 |
+
"eval_accuracy": 0.9307692307692308,
|
73 |
+
"eval_loss": 0.5576168298721313,
|
74 |
+
"eval_runtime": 20.0625,
|
75 |
+
"eval_samples_per_second": 129.595,
|
76 |
+
"eval_steps_per_second": 8.125,
|
77 |
+
"step": 100
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 0.14,
|
81 |
+
"learning_rate": 0.0001932349323493235,
|
82 |
+
"loss": 0.5329,
|
83 |
+
"step": 110
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 0.15,
|
87 |
+
"learning_rate": 0.000192619926199262,
|
88 |
+
"loss": 0.4424,
|
89 |
+
"step": 120
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.16,
|
93 |
+
"learning_rate": 0.0001920049200492005,
|
94 |
+
"loss": 0.3445,
|
95 |
+
"step": 130
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 0.17,
|
99 |
+
"learning_rate": 0.00019138991389913902,
|
100 |
+
"loss": 0.448,
|
101 |
+
"step": 140
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 0.18,
|
105 |
+
"learning_rate": 0.00019077490774907748,
|
106 |
+
"loss": 0.3616,
|
107 |
+
"step": 150
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.2,
|
111 |
+
"learning_rate": 0.000190159901599016,
|
112 |
+
"loss": 0.422,
|
113 |
+
"step": 160
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 0.21,
|
117 |
+
"learning_rate": 0.0001895448954489545,
|
118 |
+
"loss": 0.3273,
|
119 |
+
"step": 170
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 0.22,
|
123 |
+
"learning_rate": 0.000188929889298893,
|
124 |
+
"loss": 0.2256,
|
125 |
+
"step": 180
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 0.23,
|
129 |
+
"learning_rate": 0.0001883148831488315,
|
130 |
+
"loss": 0.1926,
|
131 |
+
"step": 190
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 0.25,
|
135 |
+
"learning_rate": 0.00018769987699876998,
|
136 |
+
"loss": 0.2688,
|
137 |
+
"step": 200
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 0.25,
|
141 |
+
"eval_accuracy": 0.9665384615384616,
|
142 |
+
"eval_loss": 0.23705258965492249,
|
143 |
+
"eval_runtime": 19.9301,
|
144 |
+
"eval_samples_per_second": 130.456,
|
145 |
+
"eval_steps_per_second": 8.179,
|
146 |
+
"step": 200
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.26,
|
150 |
+
"learning_rate": 0.0001870848708487085,
|
151 |
+
"loss": 0.2367,
|
152 |
+
"step": 210
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.27,
|
156 |
+
"learning_rate": 0.000186469864698647,
|
157 |
+
"loss": 0.2625,
|
158 |
+
"step": 220
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.28,
|
162 |
+
"learning_rate": 0.0001858548585485855,
|
163 |
+
"loss": 0.2304,
|
164 |
+
"step": 230
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.3,
|
168 |
+
"learning_rate": 0.000185239852398524,
|
169 |
+
"loss": 0.2889,
|
170 |
+
"step": 240
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.31,
|
174 |
+
"learning_rate": 0.0001846248462484625,
|
175 |
+
"loss": 0.1779,
|
176 |
+
"step": 250
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.32,
|
180 |
+
"learning_rate": 0.00018400984009840098,
|
181 |
+
"loss": 0.1514,
|
182 |
+
"step": 260
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.33,
|
186 |
+
"learning_rate": 0.0001833948339483395,
|
187 |
+
"loss": 0.2225,
|
188 |
+
"step": 270
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.34,
|
192 |
+
"learning_rate": 0.00018277982779827797,
|
193 |
+
"loss": 0.1753,
|
194 |
+
"step": 280
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.36,
|
198 |
+
"learning_rate": 0.0001821648216482165,
|
199 |
+
"loss": 0.1848,
|
200 |
+
"step": 290
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.37,
|
204 |
+
"learning_rate": 0.00018154981549815499,
|
205 |
+
"loss": 0.1568,
|
206 |
+
"step": 300
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.37,
|
210 |
+
"eval_accuracy": 0.9688461538461538,
|
211 |
+
"eval_loss": 0.18294188380241394,
|
212 |
+
"eval_runtime": 19.7383,
|
213 |
+
"eval_samples_per_second": 131.724,
|
214 |
+
"eval_steps_per_second": 8.258,
|
215 |
+
"step": 300
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 0.38,
|
219 |
+
"learning_rate": 0.00018093480934809348,
|
220 |
+
"loss": 0.1902,
|
221 |
+
"step": 310
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 0.39,
|
225 |
+
"learning_rate": 0.00018031980319803198,
|
226 |
+
"loss": 0.1647,
|
227 |
+
"step": 320
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 0.41,
|
231 |
+
"learning_rate": 0.0001797047970479705,
|
232 |
+
"loss": 0.1953,
|
233 |
+
"step": 330
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.42,
|
237 |
+
"learning_rate": 0.000179089790897909,
|
238 |
+
"loss": 0.2202,
|
239 |
+
"step": 340
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 0.43,
|
243 |
+
"learning_rate": 0.0001784747847478475,
|
244 |
+
"loss": 0.1543,
|
245 |
+
"step": 350
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 0.44,
|
249 |
+
"learning_rate": 0.00017785977859778598,
|
250 |
+
"loss": 0.1254,
|
251 |
+
"step": 360
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 0.46,
|
255 |
+
"learning_rate": 0.00017724477244772448,
|
256 |
+
"loss": 0.225,
|
257 |
+
"step": 370
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 0.47,
|
261 |
+
"learning_rate": 0.000176629766297663,
|
262 |
+
"loss": 0.2148,
|
263 |
+
"step": 380
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 0.48,
|
267 |
+
"learning_rate": 0.00017601476014760147,
|
268 |
+
"loss": 0.2064,
|
269 |
+
"step": 390
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.49,
|
273 |
+
"learning_rate": 0.00017539975399754,
|
274 |
+
"loss": 0.1684,
|
275 |
+
"step": 400
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.49,
|
279 |
+
"eval_accuracy": 0.9661538461538461,
|
280 |
+
"eval_loss": 0.16110503673553467,
|
281 |
+
"eval_runtime": 19.5434,
|
282 |
+
"eval_samples_per_second": 133.037,
|
283 |
+
"eval_steps_per_second": 8.34,
|
284 |
+
"step": 400
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 0.5,
|
288 |
+
"learning_rate": 0.00017478474784747848,
|
289 |
+
"loss": 0.1313,
|
290 |
+
"step": 410
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 0.52,
|
294 |
+
"learning_rate": 0.00017416974169741698,
|
295 |
+
"loss": 0.1437,
|
296 |
+
"step": 420
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.53,
|
300 |
+
"learning_rate": 0.0001736162361623616,
|
301 |
+
"loss": 0.1067,
|
302 |
+
"step": 430
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 0.54,
|
306 |
+
"learning_rate": 0.00017300123001230013,
|
307 |
+
"loss": 0.1696,
|
308 |
+
"step": 440
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 0.55,
|
312 |
+
"learning_rate": 0.00017238622386223863,
|
313 |
+
"loss": 0.1877,
|
314 |
+
"step": 450
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 0.57,
|
318 |
+
"learning_rate": 0.00017177121771217712,
|
319 |
+
"loss": 0.1683,
|
320 |
+
"step": 460
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 0.58,
|
324 |
+
"learning_rate": 0.00017115621156211564,
|
325 |
+
"loss": 0.1345,
|
326 |
+
"step": 470
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 0.59,
|
330 |
+
"learning_rate": 0.00017054120541205414,
|
331 |
+
"loss": 0.1548,
|
332 |
+
"step": 480
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 0.6,
|
336 |
+
"learning_rate": 0.00016992619926199263,
|
337 |
+
"loss": 0.1133,
|
338 |
+
"step": 490
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.62,
|
342 |
+
"learning_rate": 0.00016931119311193113,
|
343 |
+
"loss": 0.1584,
|
344 |
+
"step": 500
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 0.62,
|
348 |
+
"eval_accuracy": 0.9673076923076923,
|
349 |
+
"eval_loss": 0.134025439620018,
|
350 |
+
"eval_runtime": 19.5947,
|
351 |
+
"eval_samples_per_second": 132.689,
|
352 |
+
"eval_steps_per_second": 8.319,
|
353 |
+
"step": 500
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 0.63,
|
357 |
+
"learning_rate": 0.00016869618696186962,
|
358 |
+
"loss": 0.0944,
|
359 |
+
"step": 510
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.64,
|
363 |
+
"learning_rate": 0.00016808118081180812,
|
364 |
+
"loss": 0.1339,
|
365 |
+
"step": 520
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 0.65,
|
369 |
+
"learning_rate": 0.00016746617466174664,
|
370 |
+
"loss": 0.1224,
|
371 |
+
"step": 530
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 0.66,
|
375 |
+
"learning_rate": 0.0001668511685116851,
|
376 |
+
"loss": 0.0884,
|
377 |
+
"step": 540
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 0.68,
|
381 |
+
"learning_rate": 0.00016623616236162363,
|
382 |
+
"loss": 0.0793,
|
383 |
+
"step": 550
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 0.69,
|
387 |
+
"learning_rate": 0.00016562115621156212,
|
388 |
+
"loss": 0.1709,
|
389 |
+
"step": 560
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 0.7,
|
393 |
+
"learning_rate": 0.00016500615006150062,
|
394 |
+
"loss": 0.1247,
|
395 |
+
"step": 570
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 0.71,
|
399 |
+
"learning_rate": 0.00016439114391143914,
|
400 |
+
"loss": 0.0932,
|
401 |
+
"step": 580
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.73,
|
405 |
+
"learning_rate": 0.0001637761377613776,
|
406 |
+
"loss": 0.1186,
|
407 |
+
"step": 590
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 0.74,
|
411 |
+
"learning_rate": 0.00016316113161131613,
|
412 |
+
"loss": 0.1569,
|
413 |
+
"step": 600
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 0.74,
|
417 |
+
"eval_accuracy": 0.953076923076923,
|
418 |
+
"eval_loss": 0.19328586757183075,
|
419 |
+
"eval_runtime": 19.7536,
|
420 |
+
"eval_samples_per_second": 131.621,
|
421 |
+
"eval_steps_per_second": 8.252,
|
422 |
+
"step": 600
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.75,
|
426 |
+
"learning_rate": 0.00016254612546125462,
|
427 |
+
"loss": 0.1604,
|
428 |
+
"step": 610
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 0.76,
|
432 |
+
"learning_rate": 0.00016193111931119312,
|
433 |
+
"loss": 0.0663,
|
434 |
+
"step": 620
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 0.77,
|
438 |
+
"learning_rate": 0.00016131611316113161,
|
439 |
+
"loss": 0.2167,
|
440 |
+
"step": 630
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 0.79,
|
444 |
+
"learning_rate": 0.00016070110701107014,
|
445 |
+
"loss": 0.095,
|
446 |
+
"step": 640
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 0.8,
|
450 |
+
"learning_rate": 0.0001600861008610086,
|
451 |
+
"loss": 0.0874,
|
452 |
+
"step": 650
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 0.81,
|
456 |
+
"learning_rate": 0.00015947109471094713,
|
457 |
+
"loss": 0.1631,
|
458 |
+
"step": 660
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 0.82,
|
462 |
+
"learning_rate": 0.00015885608856088562,
|
463 |
+
"loss": 0.1154,
|
464 |
+
"step": 670
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.84,
|
468 |
+
"learning_rate": 0.00015824108241082411,
|
469 |
+
"loss": 0.1664,
|
470 |
+
"step": 680
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 0.85,
|
474 |
+
"learning_rate": 0.00015762607626076264,
|
475 |
+
"loss": 0.0859,
|
476 |
+
"step": 690
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.86,
|
480 |
+
"learning_rate": 0.0001570110701107011,
|
481 |
+
"loss": 0.0992,
|
482 |
+
"step": 700
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 0.86,
|
486 |
+
"eval_accuracy": 0.9780769230769231,
|
487 |
+
"eval_loss": 0.10314588993787766,
|
488 |
+
"eval_runtime": 19.6708,
|
489 |
+
"eval_samples_per_second": 132.176,
|
490 |
+
"eval_steps_per_second": 8.286,
|
491 |
+
"step": 700
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 0.87,
|
495 |
+
"learning_rate": 0.00015639606396063963,
|
496 |
+
"loss": 0.0465,
|
497 |
+
"step": 710
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 0.89,
|
501 |
+
"learning_rate": 0.00015578105781057812,
|
502 |
+
"loss": 0.0781,
|
503 |
+
"step": 720
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 0.9,
|
507 |
+
"learning_rate": 0.00015516605166051662,
|
508 |
+
"loss": 0.1373,
|
509 |
+
"step": 730
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 0.91,
|
513 |
+
"learning_rate": 0.0001545510455104551,
|
514 |
+
"loss": 0.1474,
|
515 |
+
"step": 740
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 0.92,
|
519 |
+
"learning_rate": 0.0001539360393603936,
|
520 |
+
"loss": 0.1383,
|
521 |
+
"step": 750
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 0.93,
|
525 |
+
"learning_rate": 0.0001533210332103321,
|
526 |
+
"loss": 0.1276,
|
527 |
+
"step": 760
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.95,
|
531 |
+
"learning_rate": 0.00015270602706027062,
|
532 |
+
"loss": 0.1304,
|
533 |
+
"step": 770
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 0.96,
|
537 |
+
"learning_rate": 0.00015209102091020912,
|
538 |
+
"loss": 0.1041,
|
539 |
+
"step": 780
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 0.97,
|
543 |
+
"learning_rate": 0.0001514760147601476,
|
544 |
+
"loss": 0.0746,
|
545 |
+
"step": 790
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 0.98,
|
549 |
+
"learning_rate": 0.0001508610086100861,
|
550 |
+
"loss": 0.0573,
|
551 |
+
"step": 800
|
552 |
+
},
|
553 |
+
{
|
554 |
+
"epoch": 0.98,
|
555 |
+
"eval_accuracy": 0.9780769230769231,
|
556 |
+
"eval_loss": 0.1024000346660614,
|
557 |
+
"eval_runtime": 19.7326,
|
558 |
+
"eval_samples_per_second": 131.762,
|
559 |
+
"eval_steps_per_second": 8.26,
|
560 |
+
"step": 800
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 1.0,
|
564 |
+
"learning_rate": 0.0001502460024600246,
|
565 |
+
"loss": 0.068,
|
566 |
+
"step": 810
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 1.01,
|
570 |
+
"learning_rate": 0.00014963099630996312,
|
571 |
+
"loss": 0.0615,
|
572 |
+
"step": 820
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 1.02,
|
576 |
+
"learning_rate": 0.0001490159901599016,
|
577 |
+
"loss": 0.0928,
|
578 |
+
"step": 830
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 1.03,
|
582 |
+
"learning_rate": 0.0001484009840098401,
|
583 |
+
"loss": 0.1008,
|
584 |
+
"step": 840
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 1.05,
|
588 |
+
"learning_rate": 0.0001477859778597786,
|
589 |
+
"loss": 0.0815,
|
590 |
+
"step": 850
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 1.06,
|
594 |
+
"learning_rate": 0.0001471709717097171,
|
595 |
+
"loss": 0.0865,
|
596 |
+
"step": 860
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 1.07,
|
600 |
+
"learning_rate": 0.0001465559655596556,
|
601 |
+
"loss": 0.0468,
|
602 |
+
"step": 870
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 1.08,
|
606 |
+
"learning_rate": 0.00014594095940959412,
|
607 |
+
"loss": 0.0811,
|
608 |
+
"step": 880
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 1.09,
|
612 |
+
"learning_rate": 0.0001453259532595326,
|
613 |
+
"loss": 0.0475,
|
614 |
+
"step": 890
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 1.11,
|
618 |
+
"learning_rate": 0.0001447109471094711,
|
619 |
+
"loss": 0.0359,
|
620 |
+
"step": 900
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 1.11,
|
624 |
+
"eval_accuracy": 0.9803846153846154,
|
625 |
+
"eval_loss": 0.09497525542974472,
|
626 |
+
"eval_runtime": 19.8135,
|
627 |
+
"eval_samples_per_second": 131.224,
|
628 |
+
"eval_steps_per_second": 8.227,
|
629 |
+
"step": 900
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 1.12,
|
633 |
+
"learning_rate": 0.0001440959409594096,
|
634 |
+
"loss": 0.1011,
|
635 |
+
"step": 910
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 1.13,
|
639 |
+
"learning_rate": 0.0001434809348093481,
|
640 |
+
"loss": 0.1215,
|
641 |
+
"step": 920
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 1.14,
|
645 |
+
"learning_rate": 0.00014286592865928662,
|
646 |
+
"loss": 0.1678,
|
647 |
+
"step": 930
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 1.16,
|
651 |
+
"learning_rate": 0.00014225092250922509,
|
652 |
+
"loss": 0.0816,
|
653 |
+
"step": 940
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 1.17,
|
657 |
+
"learning_rate": 0.0001416359163591636,
|
658 |
+
"loss": 0.1048,
|
659 |
+
"step": 950
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 1.18,
|
663 |
+
"learning_rate": 0.0001410209102091021,
|
664 |
+
"loss": 0.0386,
|
665 |
+
"step": 960
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 1.19,
|
669 |
+
"learning_rate": 0.0001404059040590406,
|
670 |
+
"loss": 0.0852,
|
671 |
+
"step": 970
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 1.21,
|
675 |
+
"learning_rate": 0.0001397908979089791,
|
676 |
+
"loss": 0.0613,
|
677 |
+
"step": 980
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 1.22,
|
681 |
+
"learning_rate": 0.00013917589175891759,
|
682 |
+
"loss": 0.0738,
|
683 |
+
"step": 990
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 1.23,
|
687 |
+
"learning_rate": 0.0001385608856088561,
|
688 |
+
"loss": 0.0961,
|
689 |
+
"step": 1000
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 1.23,
|
693 |
+
"eval_accuracy": 0.9723076923076923,
|
694 |
+
"eval_loss": 0.11998701840639114,
|
695 |
+
"eval_runtime": 19.8193,
|
696 |
+
"eval_samples_per_second": 131.185,
|
697 |
+
"eval_steps_per_second": 8.224,
|
698 |
+
"step": 1000
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 1.24,
|
702 |
+
"learning_rate": 0.0001379458794587946,
|
703 |
+
"loss": 0.0705,
|
704 |
+
"step": 1010
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 1.25,
|
708 |
+
"learning_rate": 0.0001373308733087331,
|
709 |
+
"loss": 0.0944,
|
710 |
+
"step": 1020
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 1.27,
|
714 |
+
"learning_rate": 0.0001367158671586716,
|
715 |
+
"loss": 0.0626,
|
716 |
+
"step": 1030
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 1.28,
|
720 |
+
"learning_rate": 0.0001361008610086101,
|
721 |
+
"loss": 0.0442,
|
722 |
+
"step": 1040
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 1.29,
|
726 |
+
"learning_rate": 0.00013548585485854858,
|
727 |
+
"loss": 0.1229,
|
728 |
+
"step": 1050
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 1.3,
|
732 |
+
"learning_rate": 0.0001348708487084871,
|
733 |
+
"loss": 0.0466,
|
734 |
+
"step": 1060
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 1.32,
|
738 |
+
"learning_rate": 0.00013425584255842557,
|
739 |
+
"loss": 0.0718,
|
740 |
+
"step": 1070
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 1.33,
|
744 |
+
"learning_rate": 0.0001336408364083641,
|
745 |
+
"loss": 0.0657,
|
746 |
+
"step": 1080
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 1.34,
|
750 |
+
"learning_rate": 0.0001330258302583026,
|
751 |
+
"loss": 0.0509,
|
752 |
+
"step": 1090
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 1.35,
|
756 |
+
"learning_rate": 0.00013241082410824108,
|
757 |
+
"loss": 0.0334,
|
758 |
+
"step": 1100
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 1.35,
|
762 |
+
"eval_accuracy": 0.975,
|
763 |
+
"eval_loss": 0.09949370473623276,
|
764 |
+
"eval_runtime": 19.8958,
|
765 |
+
"eval_samples_per_second": 130.681,
|
766 |
+
"eval_steps_per_second": 8.193,
|
767 |
+
"step": 1100
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 1.37,
|
771 |
+
"learning_rate": 0.0001317958179581796,
|
772 |
+
"loss": 0.0408,
|
773 |
+
"step": 1110
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 1.38,
|
777 |
+
"learning_rate": 0.0001311808118081181,
|
778 |
+
"loss": 0.0814,
|
779 |
+
"step": 1120
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 1.39,
|
783 |
+
"learning_rate": 0.0001305658056580566,
|
784 |
+
"loss": 0.0706,
|
785 |
+
"step": 1130
|
786 |
+
},
|
787 |
+
{
|
788 |
+
"epoch": 1.4,
|
789 |
+
"learning_rate": 0.0001299507995079951,
|
790 |
+
"loss": 0.0224,
|
791 |
+
"step": 1140
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 1.41,
|
795 |
+
"learning_rate": 0.00012933579335793358,
|
796 |
+
"loss": 0.1008,
|
797 |
+
"step": 1150
|
798 |
+
},
|
799 |
+
{
|
800 |
+
"epoch": 1.43,
|
801 |
+
"learning_rate": 0.00012872078720787208,
|
802 |
+
"loss": 0.0706,
|
803 |
+
"step": 1160
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 1.44,
|
807 |
+
"learning_rate": 0.0001281057810578106,
|
808 |
+
"loss": 0.1359,
|
809 |
+
"step": 1170
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 1.45,
|
813 |
+
"learning_rate": 0.00012749077490774907,
|
814 |
+
"loss": 0.1311,
|
815 |
+
"step": 1180
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 1.46,
|
819 |
+
"learning_rate": 0.0001268757687576876,
|
820 |
+
"loss": 0.053,
|
821 |
+
"step": 1190
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 1.48,
|
825 |
+
"learning_rate": 0.00012626076260762608,
|
826 |
+
"loss": 0.0855,
|
827 |
+
"step": 1200
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 1.48,
|
831 |
+
"eval_accuracy": 0.9815384615384616,
|
832 |
+
"eval_loss": 0.07914509624242783,
|
833 |
+
"eval_runtime": 20.4038,
|
834 |
+
"eval_samples_per_second": 127.427,
|
835 |
+
"eval_steps_per_second": 7.989,
|
836 |
+
"step": 1200
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 1.49,
|
840 |
+
"learning_rate": 0.00012564575645756458,
|
841 |
+
"loss": 0.0763,
|
842 |
+
"step": 1210
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 1.5,
|
846 |
+
"learning_rate": 0.0001250307503075031,
|
847 |
+
"loss": 0.0159,
|
848 |
+
"step": 1220
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 1.51,
|
852 |
+
"learning_rate": 0.00012441574415744157,
|
853 |
+
"loss": 0.0918,
|
854 |
+
"step": 1230
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 1.53,
|
858 |
+
"learning_rate": 0.0001238007380073801,
|
859 |
+
"loss": 0.0489,
|
860 |
+
"step": 1240
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 1.54,
|
864 |
+
"learning_rate": 0.00012318573185731858,
|
865 |
+
"loss": 0.0499,
|
866 |
+
"step": 1250
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 1.55,
|
870 |
+
"learning_rate": 0.00012257072570725708,
|
871 |
+
"loss": 0.0372,
|
872 |
+
"step": 1260
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 1.56,
|
876 |
+
"learning_rate": 0.00012195571955719559,
|
877 |
+
"loss": 0.0526,
|
878 |
+
"step": 1270
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 1.57,
|
882 |
+
"learning_rate": 0.00012134071340713408,
|
883 |
+
"loss": 0.0638,
|
884 |
+
"step": 1280
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 1.59,
|
888 |
+
"learning_rate": 0.00012072570725707258,
|
889 |
+
"loss": 0.0662,
|
890 |
+
"step": 1290
|
891 |
+
},
|
892 |
+
{
|
893 |
+
"epoch": 1.6,
|
894 |
+
"learning_rate": 0.00012011070110701108,
|
895 |
+
"loss": 0.0902,
|
896 |
+
"step": 1300
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 1.6,
|
900 |
+
"eval_accuracy": 0.9765384615384616,
|
901 |
+
"eval_loss": 0.09805391728878021,
|
902 |
+
"eval_runtime": 20.4238,
|
903 |
+
"eval_samples_per_second": 127.302,
|
904 |
+
"eval_steps_per_second": 7.981,
|
905 |
+
"step": 1300
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 1.61,
|
909 |
+
"learning_rate": 0.00011949569495694957,
|
910 |
+
"loss": 0.0736,
|
911 |
+
"step": 1310
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 1.62,
|
915 |
+
"learning_rate": 0.00011888068880688807,
|
916 |
+
"loss": 0.0986,
|
917 |
+
"step": 1320
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 1.64,
|
921 |
+
"learning_rate": 0.00011826568265682658,
|
922 |
+
"loss": 0.0817,
|
923 |
+
"step": 1330
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 1.65,
|
927 |
+
"learning_rate": 0.00011765067650676506,
|
928 |
+
"loss": 0.0586,
|
929 |
+
"step": 1340
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 1.66,
|
933 |
+
"learning_rate": 0.00011703567035670357,
|
934 |
+
"loss": 0.0788,
|
935 |
+
"step": 1350
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 1.67,
|
939 |
+
"learning_rate": 0.00011642066420664208,
|
940 |
+
"loss": 0.0452,
|
941 |
+
"step": 1360
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 1.69,
|
945 |
+
"learning_rate": 0.00011580565805658057,
|
946 |
+
"loss": 0.0125,
|
947 |
+
"step": 1370
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 1.7,
|
951 |
+
"learning_rate": 0.00011519065190651908,
|
952 |
+
"loss": 0.1138,
|
953 |
+
"step": 1380
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 1.71,
|
957 |
+
"learning_rate": 0.00011457564575645756,
|
958 |
+
"loss": 0.0824,
|
959 |
+
"step": 1390
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 1.72,
|
963 |
+
"learning_rate": 0.00011396063960639607,
|
964 |
+
"loss": 0.0583,
|
965 |
+
"step": 1400
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 1.72,
|
969 |
+
"eval_accuracy": 0.9711538461538461,
|
970 |
+
"eval_loss": 0.11919466406106949,
|
971 |
+
"eval_runtime": 20.5384,
|
972 |
+
"eval_samples_per_second": 126.592,
|
973 |
+
"eval_steps_per_second": 7.936,
|
974 |
+
"step": 1400
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 1.73,
|
978 |
+
"learning_rate": 0.00011334563345633458,
|
979 |
+
"loss": 0.0746,
|
980 |
+
"step": 1410
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 1.75,
|
984 |
+
"learning_rate": 0.00011273062730627306,
|
985 |
+
"loss": 0.1108,
|
986 |
+
"step": 1420
|
987 |
+
},
|
988 |
+
{
|
989 |
+
"epoch": 1.76,
|
990 |
+
"learning_rate": 0.00011211562115621157,
|
991 |
+
"loss": 0.1226,
|
992 |
+
"step": 1430
|
993 |
+
},
|
994 |
+
{
|
995 |
+
"epoch": 1.77,
|
996 |
+
"learning_rate": 0.00011150061500615008,
|
997 |
+
"loss": 0.1026,
|
998 |
+
"step": 1440
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 1.78,
|
1002 |
+
"learning_rate": 0.00011088560885608856,
|
1003 |
+
"loss": 0.1145,
|
1004 |
+
"step": 1450
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 1.8,
|
1008 |
+
"learning_rate": 0.00011027060270602707,
|
1009 |
+
"loss": 0.0374,
|
1010 |
+
"step": 1460
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 1.81,
|
1014 |
+
"learning_rate": 0.00010965559655596556,
|
1015 |
+
"loss": 0.0525,
|
1016 |
+
"step": 1470
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 1.82,
|
1020 |
+
"learning_rate": 0.00010904059040590407,
|
1021 |
+
"loss": 0.0879,
|
1022 |
+
"step": 1480
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 1.83,
|
1026 |
+
"learning_rate": 0.00010842558425584256,
|
1027 |
+
"loss": 0.0417,
|
1028 |
+
"step": 1490
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 1.85,
|
1032 |
+
"learning_rate": 0.00010781057810578106,
|
1033 |
+
"loss": 0.0683,
|
1034 |
+
"step": 1500
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 1.85,
|
1038 |
+
"eval_accuracy": 0.9846153846153847,
|
1039 |
+
"eval_loss": 0.06915987282991409,
|
1040 |
+
"eval_runtime": 20.393,
|
1041 |
+
"eval_samples_per_second": 127.495,
|
1042 |
+
"eval_steps_per_second": 7.993,
|
1043 |
+
"step": 1500
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 1.86,
|
1047 |
+
"learning_rate": 0.00010719557195571957,
|
1048 |
+
"loss": 0.0444,
|
1049 |
+
"step": 1510
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 1.87,
|
1053 |
+
"learning_rate": 0.00010658056580565808,
|
1054 |
+
"loss": 0.0272,
|
1055 |
+
"step": 1520
|
1056 |
+
},
|
1057 |
+
{
|
1058 |
+
"epoch": 1.88,
|
1059 |
+
"learning_rate": 0.00010596555965559656,
|
1060 |
+
"loss": 0.0615,
|
1061 |
+
"step": 1530
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 1.89,
|
1065 |
+
"learning_rate": 0.00010535055350553507,
|
1066 |
+
"loss": 0.0254,
|
1067 |
+
"step": 1540
|
1068 |
+
},
|
1069 |
+
{
|
1070 |
+
"epoch": 1.91,
|
1071 |
+
"learning_rate": 0.00010473554735547355,
|
1072 |
+
"loss": 0.0911,
|
1073 |
+
"step": 1550
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 1.92,
|
1077 |
+
"learning_rate": 0.00010412054120541206,
|
1078 |
+
"loss": 0.0585,
|
1079 |
+
"step": 1560
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 1.93,
|
1083 |
+
"learning_rate": 0.00010350553505535056,
|
1084 |
+
"loss": 0.1168,
|
1085 |
+
"step": 1570
|
1086 |
+
},
|
1087 |
+
{
|
1088 |
+
"epoch": 1.94,
|
1089 |
+
"learning_rate": 0.00010289052890528906,
|
1090 |
+
"loss": 0.0898,
|
1091 |
+
"step": 1580
|
1092 |
+
},
|
1093 |
+
{
|
1094 |
+
"epoch": 1.96,
|
1095 |
+
"learning_rate": 0.00010227552275522757,
|
1096 |
+
"loss": 0.0568,
|
1097 |
+
"step": 1590
|
1098 |
+
},
|
1099 |
+
{
|
1100 |
+
"epoch": 1.97,
|
1101 |
+
"learning_rate": 0.00010166051660516606,
|
1102 |
+
"loss": 0.1188,
|
1103 |
+
"step": 1600
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 1.97,
|
1107 |
+
"eval_accuracy": 0.9784615384615385,
|
1108 |
+
"eval_loss": 0.09314684569835663,
|
1109 |
+
"eval_runtime": 20.4297,
|
1110 |
+
"eval_samples_per_second": 127.266,
|
1111 |
+
"eval_steps_per_second": 7.979,
|
1112 |
+
"step": 1600
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 1.98,
|
1116 |
+
"learning_rate": 0.00010104551045510456,
|
1117 |
+
"loss": 0.0545,
|
1118 |
+
"step": 1610
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 1.99,
|
1122 |
+
"learning_rate": 0.00010043050430504306,
|
1123 |
+
"loss": 0.0571,
|
1124 |
+
"step": 1620
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 2.0,
|
1128 |
+
"learning_rate": 9.981549815498156e-05,
|
1129 |
+
"loss": 0.0525,
|
1130 |
+
"step": 1630
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 2.02,
|
1134 |
+
"learning_rate": 9.920049200492005e-05,
|
1135 |
+
"loss": 0.0253,
|
1136 |
+
"step": 1640
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 2.03,
|
1140 |
+
"learning_rate": 9.858548585485855e-05,
|
1141 |
+
"loss": 0.0148,
|
1142 |
+
"step": 1650
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 2.04,
|
1146 |
+
"learning_rate": 9.797047970479706e-05,
|
1147 |
+
"loss": 0.0338,
|
1148 |
+
"step": 1660
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 2.05,
|
1152 |
+
"learning_rate": 9.735547355473555e-05,
|
1153 |
+
"loss": 0.0078,
|
1154 |
+
"step": 1670
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 2.07,
|
1158 |
+
"learning_rate": 9.674046740467405e-05,
|
1159 |
+
"loss": 0.0407,
|
1160 |
+
"step": 1680
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 2.08,
|
1164 |
+
"learning_rate": 9.612546125461255e-05,
|
1165 |
+
"loss": 0.0654,
|
1166 |
+
"step": 1690
|
1167 |
+
},
|
1168 |
+
{
|
1169 |
+
"epoch": 2.09,
|
1170 |
+
"learning_rate": 9.551045510455105e-05,
|
1171 |
+
"loss": 0.0366,
|
1172 |
+
"step": 1700
|
1173 |
+
},
|
1174 |
+
{
|
1175 |
+
"epoch": 2.09,
|
1176 |
+
"eval_accuracy": 0.9803846153846154,
|
1177 |
+
"eval_loss": 0.09191113710403442,
|
1178 |
+
"eval_runtime": 20.4659,
|
1179 |
+
"eval_samples_per_second": 127.041,
|
1180 |
+
"eval_steps_per_second": 7.964,
|
1181 |
+
"step": 1700
|
1182 |
+
},
|
1183 |
+
{
|
1184 |
+
"epoch": 2.1,
|
1185 |
+
"learning_rate": 9.489544895448956e-05,
|
1186 |
+
"loss": 0.0548,
|
1187 |
+
"step": 1710
|
1188 |
+
},
|
1189 |
+
{
|
1190 |
+
"epoch": 2.12,
|
1191 |
+
"learning_rate": 9.428044280442805e-05,
|
1192 |
+
"loss": 0.0473,
|
1193 |
+
"step": 1720
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 2.13,
|
1197 |
+
"learning_rate": 9.366543665436655e-05,
|
1198 |
+
"loss": 0.0664,
|
1199 |
+
"step": 1730
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 2.14,
|
1203 |
+
"learning_rate": 9.305043050430505e-05,
|
1204 |
+
"loss": 0.0333,
|
1205 |
+
"step": 1740
|
1206 |
+
},
|
1207 |
+
{
|
1208 |
+
"epoch": 2.15,
|
1209 |
+
"learning_rate": 9.243542435424355e-05,
|
1210 |
+
"loss": 0.0527,
|
1211 |
+
"step": 1750
|
1212 |
+
},
|
1213 |
+
{
|
1214 |
+
"epoch": 2.16,
|
1215 |
+
"learning_rate": 9.182041820418204e-05,
|
1216 |
+
"loss": 0.0532,
|
1217 |
+
"step": 1760
|
1218 |
+
},
|
1219 |
+
{
|
1220 |
+
"epoch": 2.18,
|
1221 |
+
"learning_rate": 9.120541205412054e-05,
|
1222 |
+
"loss": 0.0172,
|
1223 |
+
"step": 1770
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 2.19,
|
1227 |
+
"learning_rate": 9.059040590405905e-05,
|
1228 |
+
"loss": 0.0396,
|
1229 |
+
"step": 1780
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 2.2,
|
1233 |
+
"learning_rate": 8.997539975399754e-05,
|
1234 |
+
"loss": 0.0656,
|
1235 |
+
"step": 1790
|
1236 |
+
},
|
1237 |
+
{
|
1238 |
+
"epoch": 2.21,
|
1239 |
+
"learning_rate": 8.936039360393605e-05,
|
1240 |
+
"loss": 0.0276,
|
1241 |
+
"step": 1800
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 2.21,
|
1245 |
+
"eval_accuracy": 0.9846153846153847,
|
1246 |
+
"eval_loss": 0.06672599911689758,
|
1247 |
+
"eval_runtime": 20.3533,
|
1248 |
+
"eval_samples_per_second": 127.743,
|
1249 |
+
"eval_steps_per_second": 8.009,
|
1250 |
+
"step": 1800
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 2.23,
|
1254 |
+
"learning_rate": 8.874538745387454e-05,
|
1255 |
+
"loss": 0.043,
|
1256 |
+
"step": 1810
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 2.24,
|
1260 |
+
"learning_rate": 8.813038130381305e-05,
|
1261 |
+
"loss": 0.0156,
|
1262 |
+
"step": 1820
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 2.25,
|
1266 |
+
"learning_rate": 8.751537515375155e-05,
|
1267 |
+
"loss": 0.0472,
|
1268 |
+
"step": 1830
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 2.26,
|
1272 |
+
"learning_rate": 8.690036900369004e-05,
|
1273 |
+
"loss": 0.0278,
|
1274 |
+
"step": 1840
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 2.28,
|
1278 |
+
"learning_rate": 8.628536285362854e-05,
|
1279 |
+
"loss": 0.0376,
|
1280 |
+
"step": 1850
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 2.29,
|
1284 |
+
"learning_rate": 8.567035670356705e-05,
|
1285 |
+
"loss": 0.0764,
|
1286 |
+
"step": 1860
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 2.3,
|
1290 |
+
"learning_rate": 8.505535055350554e-05,
|
1291 |
+
"loss": 0.0819,
|
1292 |
+
"step": 1870
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 2.31,
|
1296 |
+
"learning_rate": 8.444034440344403e-05,
|
1297 |
+
"loss": 0.0277,
|
1298 |
+
"step": 1880
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 2.32,
|
1302 |
+
"learning_rate": 8.382533825338253e-05,
|
1303 |
+
"loss": 0.0471,
|
1304 |
+
"step": 1890
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 2.34,
|
1308 |
+
"learning_rate": 8.321033210332104e-05,
|
1309 |
+
"loss": 0.0309,
|
1310 |
+
"step": 1900
|
1311 |
+
},
|
1312 |
+
{
|
1313 |
+
"epoch": 2.34,
|
1314 |
+
"eval_accuracy": 0.9857692307692307,
|
1315 |
+
"eval_loss": 0.059869736433029175,
|
1316 |
+
"eval_runtime": 20.2748,
|
1317 |
+
"eval_samples_per_second": 128.238,
|
1318 |
+
"eval_steps_per_second": 8.04,
|
1319 |
+
"step": 1900
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 2.35,
|
1323 |
+
"learning_rate": 8.259532595325955e-05,
|
1324 |
+
"loss": 0.0217,
|
1325 |
+
"step": 1910
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 2.36,
|
1329 |
+
"learning_rate": 8.198031980319804e-05,
|
1330 |
+
"loss": 0.02,
|
1331 |
+
"step": 1920
|
1332 |
+
},
|
1333 |
+
{
|
1334 |
+
"epoch": 2.37,
|
1335 |
+
"learning_rate": 8.136531365313654e-05,
|
1336 |
+
"loss": 0.0179,
|
1337 |
+
"step": 1930
|
1338 |
+
},
|
1339 |
+
{
|
1340 |
+
"epoch": 2.39,
|
1341 |
+
"learning_rate": 8.075030750307504e-05,
|
1342 |
+
"loss": 0.0511,
|
1343 |
+
"step": 1940
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 2.4,
|
1347 |
+
"learning_rate": 8.013530135301354e-05,
|
1348 |
+
"loss": 0.037,
|
1349 |
+
"step": 1950
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 2.41,
|
1353 |
+
"learning_rate": 7.952029520295203e-05,
|
1354 |
+
"loss": 0.0609,
|
1355 |
+
"step": 1960
|
1356 |
+
},
|
1357 |
+
{
|
1358 |
+
"epoch": 2.42,
|
1359 |
+
"learning_rate": 7.890528905289053e-05,
|
1360 |
+
"loss": 0.0392,
|
1361 |
+
"step": 1970
|
1362 |
+
},
|
1363 |
+
{
|
1364 |
+
"epoch": 2.44,
|
1365 |
+
"learning_rate": 7.829028290282904e-05,
|
1366 |
+
"loss": 0.0322,
|
1367 |
+
"step": 1980
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 2.45,
|
1371 |
+
"learning_rate": 7.767527675276753e-05,
|
1372 |
+
"loss": 0.0366,
|
1373 |
+
"step": 1990
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 2.46,
|
1377 |
+
"learning_rate": 7.706027060270603e-05,
|
1378 |
+
"loss": 0.0183,
|
1379 |
+
"step": 2000
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 2.46,
|
1383 |
+
"eval_accuracy": 0.9769230769230769,
|
1384 |
+
"eval_loss": 0.08916281163692474,
|
1385 |
+
"eval_runtime": 20.1987,
|
1386 |
+
"eval_samples_per_second": 128.721,
|
1387 |
+
"eval_steps_per_second": 8.07,
|
1388 |
+
"step": 2000
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 2.47,
|
1392 |
+
"learning_rate": 7.644526445264453e-05,
|
1393 |
+
"loss": 0.0555,
|
1394 |
+
"step": 2010
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 2.48,
|
1398 |
+
"learning_rate": 7.583025830258303e-05,
|
1399 |
+
"loss": 0.0125,
|
1400 |
+
"step": 2020
|
1401 |
+
},
|
1402 |
+
{
|
1403 |
+
"epoch": 2.5,
|
1404 |
+
"learning_rate": 7.521525215252154e-05,
|
1405 |
+
"loss": 0.0155,
|
1406 |
+
"step": 2030
|
1407 |
+
},
|
1408 |
+
{
|
1409 |
+
"epoch": 2.51,
|
1410 |
+
"learning_rate": 7.460024600246003e-05,
|
1411 |
+
"loss": 0.005,
|
1412 |
+
"step": 2040
|
1413 |
+
},
|
1414 |
+
{
|
1415 |
+
"epoch": 2.52,
|
1416 |
+
"learning_rate": 7.398523985239853e-05,
|
1417 |
+
"loss": 0.0193,
|
1418 |
+
"step": 2050
|
1419 |
+
},
|
1420 |
+
{
|
1421 |
+
"epoch": 2.53,
|
1422 |
+
"learning_rate": 7.337023370233703e-05,
|
1423 |
+
"loss": 0.0513,
|
1424 |
+
"step": 2060
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 2.55,
|
1428 |
+
"learning_rate": 7.275522755227553e-05,
|
1429 |
+
"loss": 0.0618,
|
1430 |
+
"step": 2070
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 2.56,
|
1434 |
+
"learning_rate": 7.214022140221402e-05,
|
1435 |
+
"loss": 0.0601,
|
1436 |
+
"step": 2080
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 2.57,
|
1440 |
+
"learning_rate": 7.152521525215252e-05,
|
1441 |
+
"loss": 0.02,
|
1442 |
+
"step": 2090
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 2.58,
|
1446 |
+
"learning_rate": 7.091020910209103e-05,
|
1447 |
+
"loss": 0.0431,
|
1448 |
+
"step": 2100
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 2.58,
|
1452 |
+
"eval_accuracy": 0.985,
|
1453 |
+
"eval_loss": 0.06632479280233383,
|
1454 |
+
"eval_runtime": 19.6556,
|
1455 |
+
"eval_samples_per_second": 132.278,
|
1456 |
+
"eval_steps_per_second": 8.293,
|
1457 |
+
"step": 2100
|
1458 |
+
},
|
1459 |
+
{
|
1460 |
+
"epoch": 2.6,
|
1461 |
+
"learning_rate": 7.029520295202952e-05,
|
1462 |
+
"loss": 0.0327,
|
1463 |
+
"step": 2110
|
1464 |
+
},
|
1465 |
+
{
|
1466 |
+
"epoch": 2.61,
|
1467 |
+
"learning_rate": 6.968019680196803e-05,
|
1468 |
+
"loss": 0.0119,
|
1469 |
+
"step": 2120
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 2.62,
|
1473 |
+
"learning_rate": 6.906519065190652e-05,
|
1474 |
+
"loss": 0.0558,
|
1475 |
+
"step": 2130
|
1476 |
+
},
|
1477 |
+
{
|
1478 |
+
"epoch": 2.63,
|
1479 |
+
"learning_rate": 6.845018450184503e-05,
|
1480 |
+
"loss": 0.0283,
|
1481 |
+
"step": 2140
|
1482 |
+
},
|
1483 |
+
{
|
1484 |
+
"epoch": 2.64,
|
1485 |
+
"learning_rate": 6.783517835178353e-05,
|
1486 |
+
"loss": 0.0451,
|
1487 |
+
"step": 2150
|
1488 |
+
},
|
1489 |
+
{
|
1490 |
+
"epoch": 2.66,
|
1491 |
+
"learning_rate": 6.722017220172202e-05,
|
1492 |
+
"loss": 0.0484,
|
1493 |
+
"step": 2160
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 2.67,
|
1497 |
+
"learning_rate": 6.660516605166052e-05,
|
1498 |
+
"loss": 0.0319,
|
1499 |
+
"step": 2170
|
1500 |
+
},
|
1501 |
+
{
|
1502 |
+
"epoch": 2.68,
|
1503 |
+
"learning_rate": 6.599015990159902e-05,
|
1504 |
+
"loss": 0.0496,
|
1505 |
+
"step": 2180
|
1506 |
+
},
|
1507 |
+
{
|
1508 |
+
"epoch": 2.69,
|
1509 |
+
"learning_rate": 6.537515375153752e-05,
|
1510 |
+
"loss": 0.0128,
|
1511 |
+
"step": 2190
|
1512 |
+
},
|
1513 |
+
{
|
1514 |
+
"epoch": 2.71,
|
1515 |
+
"learning_rate": 6.476014760147601e-05,
|
1516 |
+
"loss": 0.0424,
|
1517 |
+
"step": 2200
|
1518 |
+
},
|
1519 |
+
{
|
1520 |
+
"epoch": 2.71,
|
1521 |
+
"eval_accuracy": 0.9861538461538462,
|
1522 |
+
"eval_loss": 0.0642935037612915,
|
1523 |
+
"eval_runtime": 20.0743,
|
1524 |
+
"eval_samples_per_second": 129.519,
|
1525 |
+
"eval_steps_per_second": 8.12,
|
1526 |
+
"step": 2200
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 2.72,
|
1530 |
+
"learning_rate": 6.414514145141451e-05,
|
1531 |
+
"loss": 0.063,
|
1532 |
+
"step": 2210
|
1533 |
+
},
|
1534 |
+
{
|
1535 |
+
"epoch": 2.73,
|
1536 |
+
"learning_rate": 6.353013530135302e-05,
|
1537 |
+
"loss": 0.0555,
|
1538 |
+
"step": 2220
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 2.74,
|
1542 |
+
"learning_rate": 6.291512915129151e-05,
|
1543 |
+
"loss": 0.0488,
|
1544 |
+
"step": 2230
|
1545 |
+
},
|
1546 |
+
{
|
1547 |
+
"epoch": 2.76,
|
1548 |
+
"learning_rate": 6.230012300123002e-05,
|
1549 |
+
"loss": 0.0197,
|
1550 |
+
"step": 2240
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 2.77,
|
1554 |
+
"learning_rate": 6.168511685116851e-05,
|
1555 |
+
"loss": 0.0157,
|
1556 |
+
"step": 2250
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 2.78,
|
1560 |
+
"learning_rate": 6.107011070110702e-05,
|
1561 |
+
"loss": 0.0359,
|
1562 |
+
"step": 2260
|
1563 |
+
},
|
1564 |
+
{
|
1565 |
+
"epoch": 2.79,
|
1566 |
+
"learning_rate": 6.045510455104552e-05,
|
1567 |
+
"loss": 0.0218,
|
1568 |
+
"step": 2270
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 2.8,
|
1572 |
+
"learning_rate": 5.984009840098401e-05,
|
1573 |
+
"loss": 0.0176,
|
1574 |
+
"step": 2280
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 2.82,
|
1578 |
+
"learning_rate": 5.922509225092251e-05,
|
1579 |
+
"loss": 0.0146,
|
1580 |
+
"step": 2290
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 2.83,
|
1584 |
+
"learning_rate": 5.8610086100861016e-05,
|
1585 |
+
"loss": 0.0453,
|
1586 |
+
"step": 2300
|
1587 |
+
},
|
1588 |
+
{
|
1589 |
+
"epoch": 2.83,
|
1590 |
+
"eval_accuracy": 0.9861538461538462,
|
1591 |
+
"eval_loss": 0.06456407159566879,
|
1592 |
+
"eval_runtime": 19.4018,
|
1593 |
+
"eval_samples_per_second": 134.008,
|
1594 |
+
"eval_steps_per_second": 8.401,
|
1595 |
+
"step": 2300
|
1596 |
+
},
|
1597 |
+
{
|
1598 |
+
"epoch": 2.84,
|
1599 |
+
"learning_rate": 5.799507995079951e-05,
|
1600 |
+
"loss": 0.0617,
|
1601 |
+
"step": 2310
|
1602 |
+
},
|
1603 |
+
{
|
1604 |
+
"epoch": 2.85,
|
1605 |
+
"learning_rate": 5.738007380073801e-05,
|
1606 |
+
"loss": 0.046,
|
1607 |
+
"step": 2320
|
1608 |
+
},
|
1609 |
+
{
|
1610 |
+
"epoch": 2.87,
|
1611 |
+
"learning_rate": 5.6765067650676506e-05,
|
1612 |
+
"loss": 0.0127,
|
1613 |
+
"step": 2330
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 2.88,
|
1617 |
+
"learning_rate": 5.6150061500615015e-05,
|
1618 |
+
"loss": 0.0337,
|
1619 |
+
"step": 2340
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 2.89,
|
1623 |
+
"learning_rate": 5.553505535055351e-05,
|
1624 |
+
"loss": 0.0428,
|
1625 |
+
"step": 2350
|
1626 |
+
},
|
1627 |
+
{
|
1628 |
+
"epoch": 2.9,
|
1629 |
+
"learning_rate": 5.492004920049201e-05,
|
1630 |
+
"loss": 0.0395,
|
1631 |
+
"step": 2360
|
1632 |
+
},
|
1633 |
+
{
|
1634 |
+
"epoch": 2.92,
|
1635 |
+
"learning_rate": 5.4305043050430506e-05,
|
1636 |
+
"loss": 0.0145,
|
1637 |
+
"step": 2370
|
1638 |
+
},
|
1639 |
+
{
|
1640 |
+
"epoch": 2.93,
|
1641 |
+
"learning_rate": 5.3690036900369014e-05,
|
1642 |
+
"loss": 0.0131,
|
1643 |
+
"step": 2380
|
1644 |
+
},
|
1645 |
+
{
|
1646 |
+
"epoch": 2.94,
|
1647 |
+
"learning_rate": 5.307503075030751e-05,
|
1648 |
+
"loss": 0.0316,
|
1649 |
+
"step": 2390
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 2.95,
|
1653 |
+
"learning_rate": 5.2460024600246e-05,
|
1654 |
+
"loss": 0.0528,
|
1655 |
+
"step": 2400
|
1656 |
+
},
|
1657 |
+
{
|
1658 |
+
"epoch": 2.95,
|
1659 |
+
"eval_accuracy": 0.985,
|
1660 |
+
"eval_loss": 0.05503399297595024,
|
1661 |
+
"eval_runtime": 19.633,
|
1662 |
+
"eval_samples_per_second": 132.43,
|
1663 |
+
"eval_steps_per_second": 8.302,
|
1664 |
+
"step": 2400
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 2.96,
|
1668 |
+
"learning_rate": 5.1845018450184505e-05,
|
1669 |
+
"loss": 0.0179,
|
1670 |
+
"step": 2410
|
1671 |
+
},
|
1672 |
+
{
|
1673 |
+
"epoch": 2.98,
|
1674 |
+
"learning_rate": 5.1230012300123006e-05,
|
1675 |
+
"loss": 0.0185,
|
1676 |
+
"step": 2420
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 2.99,
|
1680 |
+
"learning_rate": 5.061500615006151e-05,
|
1681 |
+
"loss": 0.0151,
|
1682 |
+
"step": 2430
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 3.0,
|
1686 |
+
"learning_rate": 5e-05,
|
1687 |
+
"loss": 0.0501,
|
1688 |
+
"step": 2440
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 3.01,
|
1692 |
+
"learning_rate": 4.93849938499385e-05,
|
1693 |
+
"loss": 0.0043,
|
1694 |
+
"step": 2450
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 3.03,
|
1698 |
+
"learning_rate": 4.8769987699877e-05,
|
1699 |
+
"loss": 0.0085,
|
1700 |
+
"step": 2460
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 3.04,
|
1704 |
+
"learning_rate": 4.81549815498155e-05,
|
1705 |
+
"loss": 0.0068,
|
1706 |
+
"step": 2470
|
1707 |
+
},
|
1708 |
+
{
|
1709 |
+
"epoch": 3.05,
|
1710 |
+
"learning_rate": 4.7539975399754e-05,
|
1711 |
+
"loss": 0.0222,
|
1712 |
+
"step": 2480
|
1713 |
+
},
|
1714 |
+
{
|
1715 |
+
"epoch": 3.06,
|
1716 |
+
"learning_rate": 4.6924969249692496e-05,
|
1717 |
+
"loss": 0.022,
|
1718 |
+
"step": 2490
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 3.08,
|
1722 |
+
"learning_rate": 4.6309963099631e-05,
|
1723 |
+
"loss": 0.0045,
|
1724 |
+
"step": 2500
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 3.08,
|
1728 |
+
"eval_accuracy": 0.9846153846153847,
|
1729 |
+
"eval_loss": 0.05792374908924103,
|
1730 |
+
"eval_runtime": 19.8321,
|
1731 |
+
"eval_samples_per_second": 131.101,
|
1732 |
+
"eval_steps_per_second": 8.219,
|
1733 |
+
"step": 2500
|
1734 |
+
},
|
1735 |
+
{
|
1736 |
+
"epoch": 3.09,
|
1737 |
+
"learning_rate": 4.569495694956949e-05,
|
1738 |
+
"loss": 0.0161,
|
1739 |
+
"step": 2510
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 3.1,
|
1743 |
+
"learning_rate": 4.5079950799507994e-05,
|
1744 |
+
"loss": 0.0409,
|
1745 |
+
"step": 2520
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 3.11,
|
1749 |
+
"learning_rate": 4.4464944649446495e-05,
|
1750 |
+
"loss": 0.0364,
|
1751 |
+
"step": 2530
|
1752 |
+
},
|
1753 |
+
{
|
1754 |
+
"epoch": 3.12,
|
1755 |
+
"learning_rate": 4.3849938499385e-05,
|
1756 |
+
"loss": 0.0069,
|
1757 |
+
"step": 2540
|
1758 |
+
},
|
1759 |
+
{
|
1760 |
+
"epoch": 3.14,
|
1761 |
+
"learning_rate": 4.323493234932349e-05,
|
1762 |
+
"loss": 0.0091,
|
1763 |
+
"step": 2550
|
1764 |
+
},
|
1765 |
+
{
|
1766 |
+
"epoch": 3.15,
|
1767 |
+
"learning_rate": 4.261992619926199e-05,
|
1768 |
+
"loss": 0.0037,
|
1769 |
+
"step": 2560
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 3.16,
|
1773 |
+
"learning_rate": 4.2004920049200495e-05,
|
1774 |
+
"loss": 0.045,
|
1775 |
+
"step": 2570
|
1776 |
+
},
|
1777 |
+
{
|
1778 |
+
"epoch": 3.17,
|
1779 |
+
"learning_rate": 4.1389913899138996e-05,
|
1780 |
+
"loss": 0.0038,
|
1781 |
+
"step": 2580
|
1782 |
+
},
|
1783 |
+
{
|
1784 |
+
"epoch": 3.19,
|
1785 |
+
"learning_rate": 4.077490774907749e-05,
|
1786 |
+
"loss": 0.0035,
|
1787 |
+
"step": 2590
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 3.2,
|
1791 |
+
"learning_rate": 4.015990159901599e-05,
|
1792 |
+
"loss": 0.007,
|
1793 |
+
"step": 2600
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 3.2,
|
1797 |
+
"eval_accuracy": 0.9884615384615385,
|
1798 |
+
"eval_loss": 0.05168744549155235,
|
1799 |
+
"eval_runtime": 20.5897,
|
1800 |
+
"eval_samples_per_second": 126.277,
|
1801 |
+
"eval_steps_per_second": 7.917,
|
1802 |
+
"step": 2600
|
1803 |
+
},
|
1804 |
+
{
|
1805 |
+
"epoch": 3.21,
|
1806 |
+
"learning_rate": 3.954489544895449e-05,
|
1807 |
+
"loss": 0.0043,
|
1808 |
+
"step": 2610
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 3.22,
|
1812 |
+
"learning_rate": 3.892988929889299e-05,
|
1813 |
+
"loss": 0.0466,
|
1814 |
+
"step": 2620
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 3.23,
|
1818 |
+
"learning_rate": 3.831488314883149e-05,
|
1819 |
+
"loss": 0.0055,
|
1820 |
+
"step": 2630
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 3.25,
|
1824 |
+
"learning_rate": 3.769987699876999e-05,
|
1825 |
+
"loss": 0.0037,
|
1826 |
+
"step": 2640
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 3.26,
|
1830 |
+
"learning_rate": 3.7084870848708486e-05,
|
1831 |
+
"loss": 0.005,
|
1832 |
+
"step": 2650
|
1833 |
+
},
|
1834 |
+
{
|
1835 |
+
"epoch": 3.27,
|
1836 |
+
"learning_rate": 3.646986469864699e-05,
|
1837 |
+
"loss": 0.0082,
|
1838 |
+
"step": 2660
|
1839 |
+
},
|
1840 |
+
{
|
1841 |
+
"epoch": 3.28,
|
1842 |
+
"learning_rate": 3.585485854858548e-05,
|
1843 |
+
"loss": 0.0223,
|
1844 |
+
"step": 2670
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 3.3,
|
1848 |
+
"learning_rate": 3.5239852398523984e-05,
|
1849 |
+
"loss": 0.0034,
|
1850 |
+
"step": 2680
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 3.31,
|
1854 |
+
"learning_rate": 3.4624846248462485e-05,
|
1855 |
+
"loss": 0.0053,
|
1856 |
+
"step": 2690
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 3.32,
|
1860 |
+
"learning_rate": 3.400984009840099e-05,
|
1861 |
+
"loss": 0.0048,
|
1862 |
+
"step": 2700
|
1863 |
+
},
|
1864 |
+
{
|
1865 |
+
"epoch": 3.32,
|
1866 |
+
"eval_accuracy": 0.9865384615384616,
|
1867 |
+
"eval_loss": 0.058377496898174286,
|
1868 |
+
"eval_runtime": 20.5674,
|
1869 |
+
"eval_samples_per_second": 126.413,
|
1870 |
+
"eval_steps_per_second": 7.925,
|
1871 |
+
"step": 2700
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 3.33,
|
1875 |
+
"learning_rate": 3.339483394833948e-05,
|
1876 |
+
"loss": 0.0075,
|
1877 |
+
"step": 2710
|
1878 |
+
},
|
1879 |
+
{
|
1880 |
+
"epoch": 3.35,
|
1881 |
+
"learning_rate": 3.277982779827798e-05,
|
1882 |
+
"loss": 0.0092,
|
1883 |
+
"step": 2720
|
1884 |
+
},
|
1885 |
+
{
|
1886 |
+
"epoch": 3.36,
|
1887 |
+
"learning_rate": 3.2164821648216484e-05,
|
1888 |
+
"loss": 0.0164,
|
1889 |
+
"step": 2730
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 3.37,
|
1893 |
+
"learning_rate": 3.1549815498154986e-05,
|
1894 |
+
"loss": 0.0063,
|
1895 |
+
"step": 2740
|
1896 |
+
},
|
1897 |
+
{
|
1898 |
+
"epoch": 3.38,
|
1899 |
+
"learning_rate": 3.093480934809348e-05,
|
1900 |
+
"loss": 0.0195,
|
1901 |
+
"step": 2750
|
1902 |
+
},
|
1903 |
+
{
|
1904 |
+
"epoch": 3.39,
|
1905 |
+
"learning_rate": 3.0319803198031982e-05,
|
1906 |
+
"loss": 0.0563,
|
1907 |
+
"step": 2760
|
1908 |
+
},
|
1909 |
+
{
|
1910 |
+
"epoch": 3.41,
|
1911 |
+
"learning_rate": 2.970479704797048e-05,
|
1912 |
+
"loss": 0.003,
|
1913 |
+
"step": 2770
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 3.42,
|
1917 |
+
"learning_rate": 2.908979089790898e-05,
|
1918 |
+
"loss": 0.0423,
|
1919 |
+
"step": 2780
|
1920 |
+
},
|
1921 |
+
{
|
1922 |
+
"epoch": 3.43,
|
1923 |
+
"learning_rate": 2.8474784747847476e-05,
|
1924 |
+
"loss": 0.0035,
|
1925 |
+
"step": 2790
|
1926 |
+
},
|
1927 |
+
{
|
1928 |
+
"epoch": 3.44,
|
1929 |
+
"learning_rate": 2.7859778597785978e-05,
|
1930 |
+
"loss": 0.019,
|
1931 |
+
"step": 2800
|
1932 |
+
},
|
1933 |
+
{
|
1934 |
+
"epoch": 3.44,
|
1935 |
+
"eval_accuracy": 0.9873076923076923,
|
1936 |
+
"eval_loss": 0.05602681264281273,
|
1937 |
+
"eval_runtime": 20.8189,
|
1938 |
+
"eval_samples_per_second": 124.886,
|
1939 |
+
"eval_steps_per_second": 7.829,
|
1940 |
+
"step": 2800
|
1941 |
+
},
|
1942 |
+
{
|
1943 |
+
"epoch": 3.46,
|
1944 |
+
"learning_rate": 2.7244772447724476e-05,
|
1945 |
+
"loss": 0.0066,
|
1946 |
+
"step": 2810
|
1947 |
+
},
|
1948 |
+
{
|
1949 |
+
"epoch": 3.47,
|
1950 |
+
"learning_rate": 2.6629766297662977e-05,
|
1951 |
+
"loss": 0.0059,
|
1952 |
+
"step": 2820
|
1953 |
+
},
|
1954 |
+
{
|
1955 |
+
"epoch": 3.48,
|
1956 |
+
"learning_rate": 2.6014760147601475e-05,
|
1957 |
+
"loss": 0.0035,
|
1958 |
+
"step": 2830
|
1959 |
+
},
|
1960 |
+
{
|
1961 |
+
"epoch": 3.49,
|
1962 |
+
"learning_rate": 2.5399753997539977e-05,
|
1963 |
+
"loss": 0.0276,
|
1964 |
+
"step": 2840
|
1965 |
+
},
|
1966 |
+
{
|
1967 |
+
"epoch": 3.51,
|
1968 |
+
"learning_rate": 2.4784747847478475e-05,
|
1969 |
+
"loss": 0.0174,
|
1970 |
+
"step": 2850
|
1971 |
+
},
|
1972 |
+
{
|
1973 |
+
"epoch": 3.52,
|
1974 |
+
"learning_rate": 2.4169741697416977e-05,
|
1975 |
+
"loss": 0.0041,
|
1976 |
+
"step": 2860
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 3.53,
|
1980 |
+
"learning_rate": 2.3554735547355475e-05,
|
1981 |
+
"loss": 0.0174,
|
1982 |
+
"step": 2870
|
1983 |
+
},
|
1984 |
+
{
|
1985 |
+
"epoch": 3.54,
|
1986 |
+
"learning_rate": 2.2939729397293973e-05,
|
1987 |
+
"loss": 0.0059,
|
1988 |
+
"step": 2880
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 3.55,
|
1992 |
+
"learning_rate": 2.2324723247232474e-05,
|
1993 |
+
"loss": 0.0141,
|
1994 |
+
"step": 2890
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 3.57,
|
1998 |
+
"learning_rate": 2.1709717097170972e-05,
|
1999 |
+
"loss": 0.0038,
|
2000 |
+
"step": 2900
|
2001 |
+
},
|
2002 |
+
{
|
2003 |
+
"epoch": 3.57,
|
2004 |
+
"eval_accuracy": 0.9880769230769231,
|
2005 |
+
"eval_loss": 0.051459942013025284,
|
2006 |
+
"eval_runtime": 20.7979,
|
2007 |
+
"eval_samples_per_second": 125.013,
|
2008 |
+
"eval_steps_per_second": 7.837,
|
2009 |
+
"step": 2900
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 3.58,
|
2013 |
+
"learning_rate": 2.1094710947109474e-05,
|
2014 |
+
"loss": 0.0105,
|
2015 |
+
"step": 2910
|
2016 |
+
},
|
2017 |
+
{
|
2018 |
+
"epoch": 3.59,
|
2019 |
+
"learning_rate": 2.0479704797047972e-05,
|
2020 |
+
"loss": 0.0036,
|
2021 |
+
"step": 2920
|
2022 |
+
},
|
2023 |
+
{
|
2024 |
+
"epoch": 3.6,
|
2025 |
+
"learning_rate": 1.986469864698647e-05,
|
2026 |
+
"loss": 0.035,
|
2027 |
+
"step": 2930
|
2028 |
+
},
|
2029 |
+
{
|
2030 |
+
"epoch": 3.62,
|
2031 |
+
"learning_rate": 1.924969249692497e-05,
|
2032 |
+
"loss": 0.027,
|
2033 |
+
"step": 2940
|
2034 |
+
},
|
2035 |
+
{
|
2036 |
+
"epoch": 3.63,
|
2037 |
+
"learning_rate": 1.863468634686347e-05,
|
2038 |
+
"loss": 0.0062,
|
2039 |
+
"step": 2950
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 3.64,
|
2043 |
+
"learning_rate": 1.8019680196801968e-05,
|
2044 |
+
"loss": 0.0183,
|
2045 |
+
"step": 2960
|
2046 |
+
},
|
2047 |
+
{
|
2048 |
+
"epoch": 3.65,
|
2049 |
+
"learning_rate": 1.740467404674047e-05,
|
2050 |
+
"loss": 0.0031,
|
2051 |
+
"step": 2970
|
2052 |
+
},
|
2053 |
+
{
|
2054 |
+
"epoch": 3.67,
|
2055 |
+
"learning_rate": 1.6789667896678967e-05,
|
2056 |
+
"loss": 0.0027,
|
2057 |
+
"step": 2980
|
2058 |
+
},
|
2059 |
+
{
|
2060 |
+
"epoch": 3.68,
|
2061 |
+
"learning_rate": 1.617466174661747e-05,
|
2062 |
+
"loss": 0.0037,
|
2063 |
+
"step": 2990
|
2064 |
+
},
|
2065 |
+
{
|
2066 |
+
"epoch": 3.69,
|
2067 |
+
"learning_rate": 1.5559655596555967e-05,
|
2068 |
+
"loss": 0.0219,
|
2069 |
+
"step": 3000
|
2070 |
+
},
|
2071 |
+
{
|
2072 |
+
"epoch": 3.69,
|
2073 |
+
"eval_accuracy": 0.9880769230769231,
|
2074 |
+
"eval_loss": 0.05274398624897003,
|
2075 |
+
"eval_runtime": 20.6528,
|
2076 |
+
"eval_samples_per_second": 125.891,
|
2077 |
+
"eval_steps_per_second": 7.892,
|
2078 |
+
"step": 3000
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 3.7,
|
2082 |
+
"learning_rate": 1.4944649446494467e-05,
|
2083 |
+
"loss": 0.0038,
|
2084 |
+
"step": 3010
|
2085 |
+
},
|
2086 |
+
{
|
2087 |
+
"epoch": 3.71,
|
2088 |
+
"learning_rate": 1.4329643296432965e-05,
|
2089 |
+
"loss": 0.0445,
|
2090 |
+
"step": 3020
|
2091 |
+
},
|
2092 |
+
{
|
2093 |
+
"epoch": 3.73,
|
2094 |
+
"learning_rate": 1.3714637146371464e-05,
|
2095 |
+
"loss": 0.004,
|
2096 |
+
"step": 3030
|
2097 |
+
},
|
2098 |
+
{
|
2099 |
+
"epoch": 3.74,
|
2100 |
+
"learning_rate": 1.3099630996309964e-05,
|
2101 |
+
"loss": 0.0065,
|
2102 |
+
"step": 3040
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 3.75,
|
2106 |
+
"learning_rate": 1.2484624846248464e-05,
|
2107 |
+
"loss": 0.0029,
|
2108 |
+
"step": 3050
|
2109 |
+
},
|
2110 |
+
{
|
2111 |
+
"epoch": 3.76,
|
2112 |
+
"learning_rate": 1.1869618696186962e-05,
|
2113 |
+
"loss": 0.0055,
|
2114 |
+
"step": 3060
|
2115 |
+
},
|
2116 |
+
{
|
2117 |
+
"epoch": 3.78,
|
2118 |
+
"learning_rate": 1.1254612546125462e-05,
|
2119 |
+
"loss": 0.007,
|
2120 |
+
"step": 3070
|
2121 |
+
},
|
2122 |
+
{
|
2123 |
+
"epoch": 3.79,
|
2124 |
+
"learning_rate": 1.0639606396063962e-05,
|
2125 |
+
"loss": 0.0099,
|
2126 |
+
"step": 3080
|
2127 |
+
},
|
2128 |
+
{
|
2129 |
+
"epoch": 3.8,
|
2130 |
+
"learning_rate": 1.0024600246002461e-05,
|
2131 |
+
"loss": 0.0067,
|
2132 |
+
"step": 3090
|
2133 |
+
},
|
2134 |
+
{
|
2135 |
+
"epoch": 3.81,
|
2136 |
+
"learning_rate": 9.40959409594096e-06,
|
2137 |
+
"loss": 0.0117,
|
2138 |
+
"step": 3100
|
2139 |
+
},
|
2140 |
+
{
|
2141 |
+
"epoch": 3.81,
|
2142 |
+
"eval_accuracy": 0.9888461538461538,
|
2143 |
+
"eval_loss": 0.05230804532766342,
|
2144 |
+
"eval_runtime": 20.2347,
|
2145 |
+
"eval_samples_per_second": 128.492,
|
2146 |
+
"eval_steps_per_second": 8.055,
|
2147 |
+
"step": 3100
|
2148 |
+
},
|
2149 |
+
{
|
2150 |
+
"epoch": 3.83,
|
2151 |
+
"learning_rate": 8.79458794587946e-06,
|
2152 |
+
"loss": 0.0271,
|
2153 |
+
"step": 3110
|
2154 |
+
},
|
2155 |
+
{
|
2156 |
+
"epoch": 3.84,
|
2157 |
+
"learning_rate": 8.179581795817959e-06,
|
2158 |
+
"loss": 0.0219,
|
2159 |
+
"step": 3120
|
2160 |
+
},
|
2161 |
+
{
|
2162 |
+
"epoch": 3.85,
|
2163 |
+
"learning_rate": 7.564575645756458e-06,
|
2164 |
+
"loss": 0.0055,
|
2165 |
+
"step": 3130
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 3.86,
|
2169 |
+
"learning_rate": 6.949569495694958e-06,
|
2170 |
+
"loss": 0.0031,
|
2171 |
+
"step": 3140
|
2172 |
+
},
|
2173 |
+
{
|
2174 |
+
"epoch": 3.87,
|
2175 |
+
"learning_rate": 6.334563345633457e-06,
|
2176 |
+
"loss": 0.0115,
|
2177 |
+
"step": 3150
|
2178 |
+
},
|
2179 |
+
{
|
2180 |
+
"epoch": 3.89,
|
2181 |
+
"learning_rate": 5.7195571955719566e-06,
|
2182 |
+
"loss": 0.006,
|
2183 |
+
"step": 3160
|
2184 |
+
},
|
2185 |
+
{
|
2186 |
+
"epoch": 3.9,
|
2187 |
+
"learning_rate": 5.1045510455104555e-06,
|
2188 |
+
"loss": 0.0101,
|
2189 |
+
"step": 3170
|
2190 |
+
},
|
2191 |
+
{
|
2192 |
+
"epoch": 3.91,
|
2193 |
+
"learning_rate": 4.489544895448955e-06,
|
2194 |
+
"loss": 0.0197,
|
2195 |
+
"step": 3180
|
2196 |
+
},
|
2197 |
+
{
|
2198 |
+
"epoch": 3.92,
|
2199 |
+
"learning_rate": 3.874538745387454e-06,
|
2200 |
+
"loss": 0.0201,
|
2201 |
+
"step": 3190
|
2202 |
+
},
|
2203 |
+
{
|
2204 |
+
"epoch": 3.94,
|
2205 |
+
"learning_rate": 3.2595325953259536e-06,
|
2206 |
+
"loss": 0.0035,
|
2207 |
+
"step": 3200
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 3.94,
|
2211 |
+
"eval_accuracy": 0.9865384615384616,
|
2212 |
+
"eval_loss": 0.05585360527038574,
|
2213 |
+
"eval_runtime": 20.2068,
|
2214 |
+
"eval_samples_per_second": 128.67,
|
2215 |
+
"eval_steps_per_second": 8.067,
|
2216 |
+
"step": 3200
|
2217 |
+
},
|
2218 |
+
{
|
2219 |
+
"epoch": 3.95,
|
2220 |
+
"learning_rate": 2.6445264452644525e-06,
|
2221 |
+
"loss": 0.0056,
|
2222 |
+
"step": 3210
|
2223 |
+
},
|
2224 |
+
{
|
2225 |
+
"epoch": 3.96,
|
2226 |
+
"learning_rate": 2.029520295202952e-06,
|
2227 |
+
"loss": 0.0221,
|
2228 |
+
"step": 3220
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 3.97,
|
2232 |
+
"learning_rate": 1.4145141451414515e-06,
|
2233 |
+
"loss": 0.0238,
|
2234 |
+
"step": 3230
|
2235 |
+
},
|
2236 |
+
{
|
2237 |
+
"epoch": 3.99,
|
2238 |
+
"learning_rate": 7.995079950799507e-07,
|
2239 |
+
"loss": 0.0264,
|
2240 |
+
"step": 3240
|
2241 |
+
},
|
2242 |
+
{
|
2243 |
+
"epoch": 4.0,
|
2244 |
+
"learning_rate": 1.845018450184502e-07,
|
2245 |
+
"loss": 0.0255,
|
2246 |
+
"step": 3250
|
2247 |
+
},
|
2248 |
+
{
|
2249 |
+
"epoch": 4.0,
|
2250 |
+
"step": 3252,
|
2251 |
+
"total_flos": 8.0609004822528e+18,
|
2252 |
+
"train_loss": 0.11576663036481334,
|
2253 |
+
"train_runtime": 3043.19,
|
2254 |
+
"train_samples_per_second": 34.175,
|
2255 |
+
"train_steps_per_second": 1.069
|
2256 |
+
}
|
2257 |
+
],
|
2258 |
+
"max_steps": 3252,
|
2259 |
+
"num_train_epochs": 4,
|
2260 |
+
"total_flos": 8.0609004822528e+18,
|
2261 |
+
"trial_name": null,
|
2262 |
+
"trial_params": null
|
2263 |
+
}
|