End of training
Browse files
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 15.0,
|
3 |
+
"eval_loss": 0.16347353160381317,
|
4 |
+
"eval_runtime": 455.449,
|
5 |
+
"eval_samples": 5489,
|
6 |
+
"eval_samples_per_second": 12.052,
|
7 |
+
"eval_steps_per_second": 0.755,
|
8 |
+
"eval_wer": 0.10052462838822501,
|
9 |
+
"train_loss": 0.41333935497371294,
|
10 |
+
"train_runtime": 85566.3736,
|
11 |
+
"train_samples": 50312,
|
12 |
+
"train_samples_per_second": 8.82,
|
13 |
+
"train_steps_per_second": 0.276
|
14 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 15.0,
|
3 |
+
"eval_loss": 0.16347353160381317,
|
4 |
+
"eval_runtime": 455.449,
|
5 |
+
"eval_samples": 5489,
|
6 |
+
"eval_samples_per_second": 12.052,
|
7 |
+
"eval_steps_per_second": 0.755,
|
8 |
+
"eval_wer": 0.10052462838822501
|
9 |
+
}
|
runs/Feb05_04-58-47_job-c93f32d8-97c5-48e7-b5ec-c6c950f627ba/events.out.tfevents.1644123372.job-c93f32d8-97c5-48e7-b5ec-c6c950f627ba.173050.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e0cdd543b66b6773a880dd905b0369f10d02d6dfbfd795a8c971c717257700e0
|
3 |
+
size 364
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 15.0,
|
3 |
+
"train_loss": 0.41333935497371294,
|
4 |
+
"train_runtime": 85566.3736,
|
5 |
+
"train_samples": 50312,
|
6 |
+
"train_samples_per_second": 8.82,
|
7 |
+
"train_steps_per_second": 0.276
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1858 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 14.999682034976153,
|
5 |
+
"global_step": 23580,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.06,
|
12 |
+
"learning_rate": 4.950000000000001e-06,
|
13 |
+
"loss": 9.2024,
|
14 |
+
"step": 100
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.13,
|
18 |
+
"learning_rate": 9.950000000000001e-06,
|
19 |
+
"loss": 5.1397,
|
20 |
+
"step": 200
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.19,
|
24 |
+
"learning_rate": 1.4950000000000001e-05,
|
25 |
+
"loss": 3.6109,
|
26 |
+
"step": 300
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.25,
|
30 |
+
"learning_rate": 1.995e-05,
|
31 |
+
"loss": 3.2035,
|
32 |
+
"step": 400
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.32,
|
36 |
+
"learning_rate": 2.495e-05,
|
37 |
+
"loss": 3.0307,
|
38 |
+
"step": 500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.32,
|
42 |
+
"eval_loss": 3.0025861263275146,
|
43 |
+
"eval_runtime": 463.575,
|
44 |
+
"eval_samples_per_second": 11.841,
|
45 |
+
"eval_steps_per_second": 0.742,
|
46 |
+
"eval_wer": 1.0,
|
47 |
+
"step": 500
|
48 |
+
},
|
49 |
+
{
|
50 |
+
"epoch": 0.38,
|
51 |
+
"learning_rate": 2.995e-05,
|
52 |
+
"loss": 2.9725,
|
53 |
+
"step": 600
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 0.45,
|
57 |
+
"learning_rate": 3.495e-05,
|
58 |
+
"loss": 2.9483,
|
59 |
+
"step": 700
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 0.51,
|
63 |
+
"learning_rate": 3.995e-05,
|
64 |
+
"loss": 2.9349,
|
65 |
+
"step": 800
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.57,
|
69 |
+
"learning_rate": 4.495e-05,
|
70 |
+
"loss": 2.9042,
|
71 |
+
"step": 900
|
72 |
+
},
|
73 |
+
{
|
74 |
+
"epoch": 0.64,
|
75 |
+
"learning_rate": 4.995e-05,
|
76 |
+
"loss": 2.7865,
|
77 |
+
"step": 1000
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 0.64,
|
81 |
+
"eval_loss": 2.4848814010620117,
|
82 |
+
"eval_runtime": 473.8576,
|
83 |
+
"eval_samples_per_second": 11.584,
|
84 |
+
"eval_steps_per_second": 0.726,
|
85 |
+
"eval_wer": 0.9926163412027592,
|
86 |
+
"step": 1000
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.7,
|
90 |
+
"learning_rate": 5.495e-05,
|
91 |
+
"loss": 2.0237,
|
92 |
+
"step": 1100
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.76,
|
96 |
+
"learning_rate": 5.99e-05,
|
97 |
+
"loss": 1.2269,
|
98 |
+
"step": 1200
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.83,
|
102 |
+
"learning_rate": 6.49e-05,
|
103 |
+
"loss": 1.0577,
|
104 |
+
"step": 1300
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.89,
|
108 |
+
"learning_rate": 6.99e-05,
|
109 |
+
"loss": 0.8144,
|
110 |
+
"step": 1400
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.95,
|
114 |
+
"learning_rate": 7.49e-05,
|
115 |
+
"loss": 0.7522,
|
116 |
+
"step": 1500
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.95,
|
120 |
+
"eval_loss": 0.4566805362701416,
|
121 |
+
"eval_runtime": 465.7671,
|
122 |
+
"eval_samples_per_second": 11.785,
|
123 |
+
"eval_steps_per_second": 0.739,
|
124 |
+
"eval_wer": 0.3594287379772661,
|
125 |
+
"step": 1500
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 1.02,
|
129 |
+
"learning_rate": 7.99e-05,
|
130 |
+
"loss": 0.6976,
|
131 |
+
"step": 1600
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 1.08,
|
135 |
+
"learning_rate": 8.49e-05,
|
136 |
+
"loss": 0.6571,
|
137 |
+
"step": 1700
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 1.14,
|
141 |
+
"learning_rate": 8.99e-05,
|
142 |
+
"loss": 0.604,
|
143 |
+
"step": 1800
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 1.21,
|
147 |
+
"learning_rate": 9.49e-05,
|
148 |
+
"loss": 0.5644,
|
149 |
+
"step": 1900
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 1.27,
|
153 |
+
"learning_rate": 9.99e-05,
|
154 |
+
"loss": 0.5703,
|
155 |
+
"step": 2000
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 1.27,
|
159 |
+
"eval_loss": 0.34396621584892273,
|
160 |
+
"eval_runtime": 464.1797,
|
161 |
+
"eval_samples_per_second": 11.825,
|
162 |
+
"eval_steps_per_second": 0.741,
|
163 |
+
"eval_wer": 0.2586223647138832,
|
164 |
+
"step": 2000
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 1.34,
|
168 |
+
"learning_rate": 9.954587581093605e-05,
|
169 |
+
"loss": 0.5726,
|
170 |
+
"step": 2100
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 1.4,
|
174 |
+
"learning_rate": 9.908248378127896e-05,
|
175 |
+
"loss": 0.5471,
|
176 |
+
"step": 2200
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 1.46,
|
180 |
+
"learning_rate": 9.861909175162187e-05,
|
181 |
+
"loss": 0.5291,
|
182 |
+
"step": 2300
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 1.53,
|
186 |
+
"learning_rate": 9.816033364226136e-05,
|
187 |
+
"loss": 0.5082,
|
188 |
+
"step": 2400
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 1.59,
|
192 |
+
"learning_rate": 9.769694161260426e-05,
|
193 |
+
"loss": 0.4762,
|
194 |
+
"step": 2500
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 1.59,
|
198 |
+
"eval_loss": 0.2925226390361786,
|
199 |
+
"eval_runtime": 464.1083,
|
200 |
+
"eval_samples_per_second": 11.827,
|
201 |
+
"eval_steps_per_second": 0.741,
|
202 |
+
"eval_wer": 0.21781793451860487,
|
203 |
+
"step": 2500
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 1.65,
|
207 |
+
"learning_rate": 9.723354958294718e-05,
|
208 |
+
"loss": 0.4993,
|
209 |
+
"step": 2600
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 1.72,
|
213 |
+
"learning_rate": 9.677015755329009e-05,
|
214 |
+
"loss": 0.4715,
|
215 |
+
"step": 2700
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 1.78,
|
219 |
+
"learning_rate": 9.6306765523633e-05,
|
220 |
+
"loss": 0.4497,
|
221 |
+
"step": 2800
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 1.84,
|
225 |
+
"learning_rate": 9.584337349397591e-05,
|
226 |
+
"loss": 0.45,
|
227 |
+
"step": 2900
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 1.91,
|
231 |
+
"learning_rate": 9.537998146431882e-05,
|
232 |
+
"loss": 0.4585,
|
233 |
+
"step": 3000
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 1.91,
|
237 |
+
"eval_loss": 0.24418394267559052,
|
238 |
+
"eval_runtime": 462.5404,
|
239 |
+
"eval_samples_per_second": 11.867,
|
240 |
+
"eval_steps_per_second": 0.744,
|
241 |
+
"eval_wer": 0.19812493927912173,
|
242 |
+
"step": 3000
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 1.97,
|
246 |
+
"learning_rate": 9.491658943466173e-05,
|
247 |
+
"loss": 0.4531,
|
248 |
+
"step": 3100
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 2.04,
|
252 |
+
"learning_rate": 9.445319740500464e-05,
|
253 |
+
"loss": 0.5227,
|
254 |
+
"step": 3200
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 2.1,
|
258 |
+
"learning_rate": 9.398980537534755e-05,
|
259 |
+
"loss": 0.4052,
|
260 |
+
"step": 3300
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 2.16,
|
264 |
+
"learning_rate": 9.352641334569045e-05,
|
265 |
+
"loss": 0.3941,
|
266 |
+
"step": 3400
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 2.23,
|
270 |
+
"learning_rate": 9.306302131603337e-05,
|
271 |
+
"loss": 0.4013,
|
272 |
+
"step": 3500
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 2.23,
|
276 |
+
"eval_loss": 0.24954961240291595,
|
277 |
+
"eval_runtime": 463.9078,
|
278 |
+
"eval_samples_per_second": 11.832,
|
279 |
+
"eval_steps_per_second": 0.742,
|
280 |
+
"eval_wer": 0.18175459049839696,
|
281 |
+
"step": 3500
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 2.29,
|
285 |
+
"learning_rate": 9.259962928637628e-05,
|
286 |
+
"loss": 0.3812,
|
287 |
+
"step": 3600
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 2.35,
|
291 |
+
"learning_rate": 9.21362372567192e-05,
|
292 |
+
"loss": 0.3919,
|
293 |
+
"step": 3700
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 2.42,
|
297 |
+
"learning_rate": 9.16728452270621e-05,
|
298 |
+
"loss": 0.389,
|
299 |
+
"step": 3800
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 2.48,
|
303 |
+
"learning_rate": 9.1209453197405e-05,
|
304 |
+
"loss": 0.3597,
|
305 |
+
"step": 3900
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 2.54,
|
309 |
+
"learning_rate": 9.074606116774792e-05,
|
310 |
+
"loss": 0.449,
|
311 |
+
"step": 4000
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 2.54,
|
315 |
+
"eval_loss": 0.21517333388328552,
|
316 |
+
"eval_runtime": 464.2336,
|
317 |
+
"eval_samples_per_second": 11.824,
|
318 |
+
"eval_steps_per_second": 0.741,
|
319 |
+
"eval_wer": 0.18082191780821918,
|
320 |
+
"step": 4000
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 2.61,
|
324 |
+
"learning_rate": 9.028266913809083e-05,
|
325 |
+
"loss": 0.383,
|
326 |
+
"step": 4100
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 2.67,
|
330 |
+
"learning_rate": 8.981927710843374e-05,
|
331 |
+
"loss": 0.3625,
|
332 |
+
"step": 4200
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 2.74,
|
336 |
+
"learning_rate": 8.935588507877664e-05,
|
337 |
+
"loss": 0.3481,
|
338 |
+
"step": 4300
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 2.8,
|
342 |
+
"learning_rate": 8.889712696941613e-05,
|
343 |
+
"loss": 0.3706,
|
344 |
+
"step": 4400
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 2.86,
|
348 |
+
"learning_rate": 8.843373493975903e-05,
|
349 |
+
"loss": 0.355,
|
350 |
+
"step": 4500
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 2.86,
|
354 |
+
"eval_loss": 0.21792450547218323,
|
355 |
+
"eval_runtime": 460.2784,
|
356 |
+
"eval_samples_per_second": 11.925,
|
357 |
+
"eval_steps_per_second": 0.747,
|
358 |
+
"eval_wer": 0.16701641892548333,
|
359 |
+
"step": 4500
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 2.93,
|
363 |
+
"learning_rate": 8.797034291010195e-05,
|
364 |
+
"loss": 0.3509,
|
365 |
+
"step": 4600
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 2.99,
|
369 |
+
"learning_rate": 8.750695088044486e-05,
|
370 |
+
"loss": 0.3432,
|
371 |
+
"step": 4700
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 3.05,
|
375 |
+
"learning_rate": 8.704355885078776e-05,
|
376 |
+
"loss": 0.3184,
|
377 |
+
"step": 4800
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 3.12,
|
381 |
+
"learning_rate": 8.658016682113068e-05,
|
382 |
+
"loss": 0.3203,
|
383 |
+
"step": 4900
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 3.18,
|
387 |
+
"learning_rate": 8.611677479147359e-05,
|
388 |
+
"loss": 0.3142,
|
389 |
+
"step": 5000
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 3.18,
|
393 |
+
"eval_loss": 0.19529223442077637,
|
394 |
+
"eval_runtime": 460.5041,
|
395 |
+
"eval_samples_per_second": 11.92,
|
396 |
+
"eval_steps_per_second": 0.747,
|
397 |
+
"eval_wer": 0.15418245409501602,
|
398 |
+
"step": 5000
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 3.24,
|
402 |
+
"learning_rate": 8.56533827618165e-05,
|
403 |
+
"loss": 0.3081,
|
404 |
+
"step": 5100
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 3.31,
|
408 |
+
"learning_rate": 8.518999073215941e-05,
|
409 |
+
"loss": 0.3916,
|
410 |
+
"step": 5200
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 3.37,
|
414 |
+
"learning_rate": 8.472659870250233e-05,
|
415 |
+
"loss": 0.3128,
|
416 |
+
"step": 5300
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 3.43,
|
420 |
+
"learning_rate": 8.426320667284524e-05,
|
421 |
+
"loss": 0.3095,
|
422 |
+
"step": 5400
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 3.5,
|
426 |
+
"learning_rate": 8.379981464318814e-05,
|
427 |
+
"loss": 0.3242,
|
428 |
+
"step": 5500
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 3.5,
|
432 |
+
"eval_loss": 0.21034082770347595,
|
433 |
+
"eval_runtime": 458.718,
|
434 |
+
"eval_samples_per_second": 11.966,
|
435 |
+
"eval_steps_per_second": 0.75,
|
436 |
+
"eval_wer": 0.15256970756825028,
|
437 |
+
"step": 5500
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 3.56,
|
441 |
+
"learning_rate": 8.333642261353105e-05,
|
442 |
+
"loss": 0.3029,
|
443 |
+
"step": 5600
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 3.63,
|
447 |
+
"learning_rate": 8.287303058387395e-05,
|
448 |
+
"loss": 0.3136,
|
449 |
+
"step": 5700
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 3.69,
|
453 |
+
"learning_rate": 8.240963855421687e-05,
|
454 |
+
"loss": 0.3069,
|
455 |
+
"step": 5800
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 3.75,
|
459 |
+
"learning_rate": 8.194624652455978e-05,
|
460 |
+
"loss": 0.3158,
|
461 |
+
"step": 5900
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 3.82,
|
465 |
+
"learning_rate": 8.14828544949027e-05,
|
466 |
+
"loss": 0.3016,
|
467 |
+
"step": 6000
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 3.82,
|
471 |
+
"eval_loss": 0.19107381999492645,
|
472 |
+
"eval_runtime": 460.0828,
|
473 |
+
"eval_samples_per_second": 11.93,
|
474 |
+
"eval_steps_per_second": 0.748,
|
475 |
+
"eval_wer": 0.1477217526474303,
|
476 |
+
"step": 6000
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 3.88,
|
480 |
+
"learning_rate": 8.10194624652456e-05,
|
481 |
+
"loss": 0.3116,
|
482 |
+
"step": 6100
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 3.94,
|
486 |
+
"learning_rate": 8.055607043558852e-05,
|
487 |
+
"loss": 0.302,
|
488 |
+
"step": 6200
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 4.01,
|
492 |
+
"learning_rate": 8.009731232622799e-05,
|
493 |
+
"loss": 0.4194,
|
494 |
+
"step": 6300
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 4.07,
|
498 |
+
"learning_rate": 7.963392029657091e-05,
|
499 |
+
"loss": 0.2762,
|
500 |
+
"step": 6400
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 4.13,
|
504 |
+
"learning_rate": 7.917052826691382e-05,
|
505 |
+
"loss": 0.2713,
|
506 |
+
"step": 6500
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 4.13,
|
510 |
+
"eval_loss": 0.18356408178806305,
|
511 |
+
"eval_runtime": 459.9096,
|
512 |
+
"eval_samples_per_second": 11.935,
|
513 |
+
"eval_steps_per_second": 0.748,
|
514 |
+
"eval_wer": 0.14217429320897698,
|
515 |
+
"step": 6500
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 4.2,
|
519 |
+
"learning_rate": 7.870713623725672e-05,
|
520 |
+
"loss": 0.2812,
|
521 |
+
"step": 6600
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 4.26,
|
525 |
+
"learning_rate": 7.824374420759963e-05,
|
526 |
+
"loss": 0.2775,
|
527 |
+
"step": 6700
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 4.33,
|
531 |
+
"learning_rate": 7.778035217794253e-05,
|
532 |
+
"loss": 0.263,
|
533 |
+
"step": 6800
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 4.39,
|
537 |
+
"learning_rate": 7.731696014828545e-05,
|
538 |
+
"loss": 0.2734,
|
539 |
+
"step": 6900
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 4.45,
|
543 |
+
"learning_rate": 7.685820203892493e-05,
|
544 |
+
"loss": 0.2807,
|
545 |
+
"step": 7000
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 4.45,
|
549 |
+
"eval_loss": 0.1924305111169815,
|
550 |
+
"eval_runtime": 455.2841,
|
551 |
+
"eval_samples_per_second": 12.056,
|
552 |
+
"eval_steps_per_second": 0.756,
|
553 |
+
"eval_wer": 0.14470999708539783,
|
554 |
+
"step": 7000
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 4.52,
|
558 |
+
"learning_rate": 7.639481000926784e-05,
|
559 |
+
"loss": 0.2794,
|
560 |
+
"step": 7100
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 4.58,
|
564 |
+
"learning_rate": 7.593141797961075e-05,
|
565 |
+
"loss": 0.2705,
|
566 |
+
"step": 7200
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 4.64,
|
570 |
+
"learning_rate": 7.546802594995367e-05,
|
571 |
+
"loss": 0.2698,
|
572 |
+
"step": 7300
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 4.71,
|
576 |
+
"learning_rate": 7.500463392029657e-05,
|
577 |
+
"loss": 0.2761,
|
578 |
+
"step": 7400
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 4.77,
|
582 |
+
"learning_rate": 7.454124189063948e-05,
|
583 |
+
"loss": 0.2929,
|
584 |
+
"step": 7500
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 4.77,
|
588 |
+
"eval_loss": 0.1848202645778656,
|
589 |
+
"eval_runtime": 454.4795,
|
590 |
+
"eval_samples_per_second": 12.078,
|
591 |
+
"eval_steps_per_second": 0.757,
|
592 |
+
"eval_wer": 0.14020207908287186,
|
593 |
+
"step": 7500
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 4.83,
|
597 |
+
"learning_rate": 7.40778498609824e-05,
|
598 |
+
"loss": 0.2729,
|
599 |
+
"step": 7600
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 4.9,
|
603 |
+
"learning_rate": 7.36144578313253e-05,
|
604 |
+
"loss": 0.2654,
|
605 |
+
"step": 7700
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 4.96,
|
609 |
+
"learning_rate": 7.315106580166822e-05,
|
610 |
+
"loss": 0.2627,
|
611 |
+
"step": 7800
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 5.03,
|
615 |
+
"learning_rate": 7.268767377201113e-05,
|
616 |
+
"loss": 0.255,
|
617 |
+
"step": 7900
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 5.09,
|
621 |
+
"learning_rate": 7.222428174235403e-05,
|
622 |
+
"loss": 0.2595,
|
623 |
+
"step": 8000
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 5.09,
|
627 |
+
"eval_loss": 0.17834417521953583,
|
628 |
+
"eval_runtime": 456.3292,
|
629 |
+
"eval_samples_per_second": 12.029,
|
630 |
+
"eval_steps_per_second": 0.754,
|
631 |
+
"eval_wer": 0.13299329641503935,
|
632 |
+
"step": 8000
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 5.15,
|
636 |
+
"learning_rate": 7.176088971269694e-05,
|
637 |
+
"loss": 0.2411,
|
638 |
+
"step": 8100
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 5.22,
|
642 |
+
"learning_rate": 7.129749768303986e-05,
|
643 |
+
"loss": 0.256,
|
644 |
+
"step": 8200
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 5.28,
|
648 |
+
"learning_rate": 7.083410565338276e-05,
|
649 |
+
"loss": 0.2605,
|
650 |
+
"step": 8300
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 5.34,
|
654 |
+
"learning_rate": 7.037071362372567e-05,
|
655 |
+
"loss": 0.2468,
|
656 |
+
"step": 8400
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 5.41,
|
660 |
+
"learning_rate": 6.990732159406859e-05,
|
661 |
+
"loss": 0.2289,
|
662 |
+
"step": 8500
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 5.41,
|
666 |
+
"eval_loss": 0.19014818966388702,
|
667 |
+
"eval_runtime": 455.481,
|
668 |
+
"eval_samples_per_second": 12.051,
|
669 |
+
"eval_steps_per_second": 0.755,
|
670 |
+
"eval_wer": 0.13133197318566014,
|
671 |
+
"step": 8500
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 5.47,
|
675 |
+
"learning_rate": 6.94439295644115e-05,
|
676 |
+
"loss": 0.2556,
|
677 |
+
"step": 8600
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 5.53,
|
681 |
+
"learning_rate": 6.898517145505098e-05,
|
682 |
+
"loss": 0.2458,
|
683 |
+
"step": 8700
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 5.6,
|
687 |
+
"learning_rate": 6.852177942539389e-05,
|
688 |
+
"loss": 0.2389,
|
689 |
+
"step": 8800
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 5.66,
|
693 |
+
"learning_rate": 6.80583873957368e-05,
|
694 |
+
"loss": 0.251,
|
695 |
+
"step": 8900
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 5.72,
|
699 |
+
"learning_rate": 6.759499536607971e-05,
|
700 |
+
"loss": 0.2567,
|
701 |
+
"step": 9000
|
702 |
+
},
|
703 |
+
{
|
704 |
+
"epoch": 5.72,
|
705 |
+
"eval_loss": 0.17836444079875946,
|
706 |
+
"eval_runtime": 456.7005,
|
707 |
+
"eval_samples_per_second": 12.019,
|
708 |
+
"eval_steps_per_second": 0.753,
|
709 |
+
"eval_wer": 0.12981638006412125,
|
710 |
+
"step": 9000
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 5.79,
|
714 |
+
"learning_rate": 6.713160333642262e-05,
|
715 |
+
"loss": 0.2503,
|
716 |
+
"step": 9100
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 5.85,
|
720 |
+
"learning_rate": 6.666821130676552e-05,
|
721 |
+
"loss": 0.2525,
|
722 |
+
"step": 9200
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 5.92,
|
726 |
+
"learning_rate": 6.620481927710843e-05,
|
727 |
+
"loss": 0.2416,
|
728 |
+
"step": 9300
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 5.98,
|
732 |
+
"learning_rate": 6.574142724745135e-05,
|
733 |
+
"loss": 0.3107,
|
734 |
+
"step": 9400
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 6.04,
|
738 |
+
"learning_rate": 6.527803521779425e-05,
|
739 |
+
"loss": 0.2401,
|
740 |
+
"step": 9500
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 6.04,
|
744 |
+
"eval_loss": 0.19564008712768555,
|
745 |
+
"eval_runtime": 455.4461,
|
746 |
+
"eval_samples_per_second": 12.052,
|
747 |
+
"eval_steps_per_second": 0.755,
|
748 |
+
"eval_wer": 0.1297775187020305,
|
749 |
+
"step": 9500
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 6.11,
|
753 |
+
"learning_rate": 6.481464318813717e-05,
|
754 |
+
"loss": 0.2336,
|
755 |
+
"step": 9600
|
756 |
+
},
|
757 |
+
{
|
758 |
+
"epoch": 6.17,
|
759 |
+
"learning_rate": 6.435125115848008e-05,
|
760 |
+
"loss": 0.2305,
|
761 |
+
"step": 9700
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 6.23,
|
765 |
+
"learning_rate": 6.3887859128823e-05,
|
766 |
+
"loss": 0.2109,
|
767 |
+
"step": 9800
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 6.3,
|
771 |
+
"learning_rate": 6.34244670991659e-05,
|
772 |
+
"loss": 0.2321,
|
773 |
+
"step": 9900
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 6.36,
|
777 |
+
"learning_rate": 6.296570898980539e-05,
|
778 |
+
"loss": 0.2098,
|
779 |
+
"step": 10000
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 6.36,
|
783 |
+
"eval_loss": 0.17475813627243042,
|
784 |
+
"eval_runtime": 454.9716,
|
785 |
+
"eval_samples_per_second": 12.064,
|
786 |
+
"eval_steps_per_second": 0.756,
|
787 |
+
"eval_wer": 0.12771786651122122,
|
788 |
+
"step": 10000
|
789 |
+
},
|
790 |
+
{
|
791 |
+
"epoch": 6.42,
|
792 |
+
"learning_rate": 6.250231696014829e-05,
|
793 |
+
"loss": 0.2226,
|
794 |
+
"step": 10100
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 6.49,
|
798 |
+
"learning_rate": 6.20389249304912e-05,
|
799 |
+
"loss": 0.2165,
|
800 |
+
"step": 10200
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 6.55,
|
804 |
+
"learning_rate": 6.157553290083412e-05,
|
805 |
+
"loss": 0.2228,
|
806 |
+
"step": 10300
|
807 |
+
},
|
808 |
+
{
|
809 |
+
"epoch": 6.62,
|
810 |
+
"learning_rate": 6.111214087117701e-05,
|
811 |
+
"loss": 0.224,
|
812 |
+
"step": 10400
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 6.68,
|
816 |
+
"learning_rate": 6.0648748841519934e-05,
|
817 |
+
"loss": 0.2246,
|
818 |
+
"step": 10500
|
819 |
+
},
|
820 |
+
{
|
821 |
+
"epoch": 6.68,
|
822 |
+
"eval_loss": 0.1776755452156067,
|
823 |
+
"eval_runtime": 454.2959,
|
824 |
+
"eval_samples_per_second": 12.082,
|
825 |
+
"eval_steps_per_second": 0.757,
|
826 |
+
"eval_wer": 0.12535703876420867,
|
827 |
+
"step": 10500
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 6.74,
|
831 |
+
"learning_rate": 6.018535681186284e-05,
|
832 |
+
"loss": 0.2872,
|
833 |
+
"step": 10600
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 6.81,
|
837 |
+
"learning_rate": 5.972196478220575e-05,
|
838 |
+
"loss": 0.2305,
|
839 |
+
"step": 10700
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 6.87,
|
843 |
+
"learning_rate": 5.925857275254866e-05,
|
844 |
+
"loss": 0.2252,
|
845 |
+
"step": 10800
|
846 |
+
},
|
847 |
+
{
|
848 |
+
"epoch": 6.93,
|
849 |
+
"learning_rate": 5.8795180722891576e-05,
|
850 |
+
"loss": 0.2218,
|
851 |
+
"step": 10900
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 7.0,
|
855 |
+
"learning_rate": 5.833178869323448e-05,
|
856 |
+
"loss": 0.2197,
|
857 |
+
"step": 11000
|
858 |
+
},
|
859 |
+
{
|
860 |
+
"epoch": 7.0,
|
861 |
+
"eval_loss": 0.17029327154159546,
|
862 |
+
"eval_runtime": 456.3521,
|
863 |
+
"eval_samples_per_second": 12.028,
|
864 |
+
"eval_steps_per_second": 0.754,
|
865 |
+
"eval_wer": 0.12218983775381327,
|
866 |
+
"step": 11000
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 7.06,
|
870 |
+
"learning_rate": 5.786839666357739e-05,
|
871 |
+
"loss": 0.1959,
|
872 |
+
"step": 11100
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 7.12,
|
876 |
+
"learning_rate": 5.74050046339203e-05,
|
877 |
+
"loss": 0.2026,
|
878 |
+
"step": 11200
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 7.19,
|
882 |
+
"learning_rate": 5.6941612604263204e-05,
|
883 |
+
"loss": 0.1998,
|
884 |
+
"step": 11300
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 7.25,
|
888 |
+
"learning_rate": 5.6478220574606123e-05,
|
889 |
+
"loss": 0.2016,
|
890 |
+
"step": 11400
|
891 |
+
},
|
892 |
+
{
|
893 |
+
"epoch": 7.32,
|
894 |
+
"learning_rate": 5.601482854494903e-05,
|
895 |
+
"loss": 0.2122,
|
896 |
+
"step": 11500
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 7.32,
|
900 |
+
"eval_loss": 0.1917443722486496,
|
901 |
+
"eval_runtime": 453.8398,
|
902 |
+
"eval_samples_per_second": 12.095,
|
903 |
+
"eval_steps_per_second": 0.758,
|
904 |
+
"eval_wer": 0.12212183037015448,
|
905 |
+
"step": 11500
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 7.38,
|
909 |
+
"learning_rate": 5.555143651529194e-05,
|
910 |
+
"loss": 0.2096,
|
911 |
+
"step": 11600
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 7.44,
|
915 |
+
"learning_rate": 5.5088044485634847e-05,
|
916 |
+
"loss": 0.2012,
|
917 |
+
"step": 11700
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 7.51,
|
921 |
+
"learning_rate": 5.4624652455977766e-05,
|
922 |
+
"loss": 0.2071,
|
923 |
+
"step": 11800
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 7.57,
|
927 |
+
"learning_rate": 5.416126042632067e-05,
|
928 |
+
"loss": 0.1978,
|
929 |
+
"step": 11900
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 7.63,
|
933 |
+
"learning_rate": 5.3697868396663576e-05,
|
934 |
+
"loss": 0.2746,
|
935 |
+
"step": 12000
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 7.63,
|
939 |
+
"eval_loss": 0.1768909990787506,
|
940 |
+
"eval_runtime": 454.1012,
|
941 |
+
"eval_samples_per_second": 12.088,
|
942 |
+
"eval_steps_per_second": 0.758,
|
943 |
+
"eval_wer": 0.12147090255513456,
|
944 |
+
"step": 12000
|
945 |
+
},
|
946 |
+
{
|
947 |
+
"epoch": 7.7,
|
948 |
+
"learning_rate": 5.323447636700649e-05,
|
949 |
+
"loss": 0.2089,
|
950 |
+
"step": 12100
|
951 |
+
},
|
952 |
+
{
|
953 |
+
"epoch": 7.76,
|
954 |
+
"learning_rate": 5.277571825764597e-05,
|
955 |
+
"loss": 0.2055,
|
956 |
+
"step": 12200
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 7.82,
|
960 |
+
"learning_rate": 5.231232622798888e-05,
|
961 |
+
"loss": 0.197,
|
962 |
+
"step": 12300
|
963 |
+
},
|
964 |
+
{
|
965 |
+
"epoch": 7.89,
|
966 |
+
"learning_rate": 5.1848934198331786e-05,
|
967 |
+
"loss": 0.1963,
|
968 |
+
"step": 12400
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 7.95,
|
972 |
+
"learning_rate": 5.1385542168674705e-05,
|
973 |
+
"loss": 0.2148,
|
974 |
+
"step": 12500
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 7.95,
|
978 |
+
"eval_loss": 0.1736288219690323,
|
979 |
+
"eval_runtime": 454.4442,
|
980 |
+
"eval_samples_per_second": 12.078,
|
981 |
+
"eval_steps_per_second": 0.757,
|
982 |
+
"eval_wer": 0.11932381229962111,
|
983 |
+
"step": 12500
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 8.02,
|
987 |
+
"learning_rate": 5.092215013901761e-05,
|
988 |
+
"loss": 0.2049,
|
989 |
+
"step": 12600
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 8.08,
|
993 |
+
"learning_rate": 5.045875810936053e-05,
|
994 |
+
"loss": 0.1886,
|
995 |
+
"step": 12700
|
996 |
+
},
|
997 |
+
{
|
998 |
+
"epoch": 8.14,
|
999 |
+
"learning_rate": 4.9995366079703435e-05,
|
1000 |
+
"loss": 0.1955,
|
1001 |
+
"step": 12800
|
1002 |
+
},
|
1003 |
+
{
|
1004 |
+
"epoch": 8.21,
|
1005 |
+
"learning_rate": 4.953197405004634e-05,
|
1006 |
+
"loss": 0.1903,
|
1007 |
+
"step": 12900
|
1008 |
+
},
|
1009 |
+
{
|
1010 |
+
"epoch": 8.27,
|
1011 |
+
"learning_rate": 4.906858202038925e-05,
|
1012 |
+
"loss": 0.1915,
|
1013 |
+
"step": 13000
|
1014 |
+
},
|
1015 |
+
{
|
1016 |
+
"epoch": 8.27,
|
1017 |
+
"eval_loss": 0.18138590455055237,
|
1018 |
+
"eval_runtime": 453.4078,
|
1019 |
+
"eval_samples_per_second": 12.106,
|
1020 |
+
"eval_steps_per_second": 0.759,
|
1021 |
+
"eval_wer": 0.11608860390556688,
|
1022 |
+
"step": 13000
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 8.33,
|
1026 |
+
"learning_rate": 4.860982391102873e-05,
|
1027 |
+
"loss": 0.1925,
|
1028 |
+
"step": 13100
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 8.4,
|
1032 |
+
"learning_rate": 4.8146431881371644e-05,
|
1033 |
+
"loss": 0.201,
|
1034 |
+
"step": 13200
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 8.46,
|
1038 |
+
"learning_rate": 4.768303985171455e-05,
|
1039 |
+
"loss": 0.2033,
|
1040 |
+
"step": 13300
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 8.52,
|
1044 |
+
"learning_rate": 4.721964782205746e-05,
|
1045 |
+
"loss": 0.1893,
|
1046 |
+
"step": 13400
|
1047 |
+
},
|
1048 |
+
{
|
1049 |
+
"epoch": 8.59,
|
1050 |
+
"learning_rate": 4.6756255792400374e-05,
|
1051 |
+
"loss": 0.2462,
|
1052 |
+
"step": 13500
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 8.59,
|
1056 |
+
"eval_loss": 0.1748134046792984,
|
1057 |
+
"eval_runtime": 459.3573,
|
1058 |
+
"eval_samples_per_second": 11.949,
|
1059 |
+
"eval_steps_per_second": 0.749,
|
1060 |
+
"eval_wer": 0.11656465559117847,
|
1061 |
+
"step": 13500
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 8.65,
|
1065 |
+
"learning_rate": 4.629286376274328e-05,
|
1066 |
+
"loss": 0.1884,
|
1067 |
+
"step": 13600
|
1068 |
+
},
|
1069 |
+
{
|
1070 |
+
"epoch": 8.71,
|
1071 |
+
"learning_rate": 4.582947173308619e-05,
|
1072 |
+
"loss": 0.1893,
|
1073 |
+
"step": 13700
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 8.78,
|
1077 |
+
"learning_rate": 4.5366079703429104e-05,
|
1078 |
+
"loss": 0.1927,
|
1079 |
+
"step": 13800
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 8.84,
|
1083 |
+
"learning_rate": 4.4902687673772016e-05,
|
1084 |
+
"loss": 0.1889,
|
1085 |
+
"step": 13900
|
1086 |
+
},
|
1087 |
+
{
|
1088 |
+
"epoch": 8.91,
|
1089 |
+
"learning_rate": 4.443929564411493e-05,
|
1090 |
+
"loss": 0.1872,
|
1091 |
+
"step": 14000
|
1092 |
+
},
|
1093 |
+
{
|
1094 |
+
"epoch": 8.91,
|
1095 |
+
"eval_loss": 0.1768641322851181,
|
1096 |
+
"eval_runtime": 453.6375,
|
1097 |
+
"eval_samples_per_second": 12.1,
|
1098 |
+
"eval_steps_per_second": 0.758,
|
1099 |
+
"eval_wer": 0.11334887787816963,
|
1100 |
+
"step": 14000
|
1101 |
+
},
|
1102 |
+
{
|
1103 |
+
"epoch": 8.97,
|
1104 |
+
"learning_rate": 4.3975903614457834e-05,
|
1105 |
+
"loss": 0.1842,
|
1106 |
+
"step": 14100
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 9.03,
|
1110 |
+
"learning_rate": 4.351251158480074e-05,
|
1111 |
+
"loss": 0.1824,
|
1112 |
+
"step": 14200
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 9.1,
|
1116 |
+
"learning_rate": 4.304911955514365e-05,
|
1117 |
+
"loss": 0.1811,
|
1118 |
+
"step": 14300
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 9.16,
|
1122 |
+
"learning_rate": 4.2585727525486564e-05,
|
1123 |
+
"loss": 0.1686,
|
1124 |
+
"step": 14400
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 9.22,
|
1128 |
+
"learning_rate": 4.2122335495829476e-05,
|
1129 |
+
"loss": 0.1886,
|
1130 |
+
"step": 14500
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 9.22,
|
1134 |
+
"eval_loss": 0.18517228960990906,
|
1135 |
+
"eval_runtime": 454.3062,
|
1136 |
+
"eval_samples_per_second": 12.082,
|
1137 |
+
"eval_steps_per_second": 0.757,
|
1138 |
+
"eval_wer": 0.11428155056834742,
|
1139 |
+
"step": 14500
|
1140 |
+
},
|
1141 |
+
{
|
1142 |
+
"epoch": 9.29,
|
1143 |
+
"learning_rate": 4.1663577386468956e-05,
|
1144 |
+
"loss": 0.166,
|
1145 |
+
"step": 14600
|
1146 |
+
},
|
1147 |
+
{
|
1148 |
+
"epoch": 9.35,
|
1149 |
+
"learning_rate": 4.120018535681187e-05,
|
1150 |
+
"loss": 0.1796,
|
1151 |
+
"step": 14700
|
1152 |
+
},
|
1153 |
+
{
|
1154 |
+
"epoch": 9.41,
|
1155 |
+
"learning_rate": 4.0736793327154773e-05,
|
1156 |
+
"loss": 0.1744,
|
1157 |
+
"step": 14800
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 9.48,
|
1161 |
+
"learning_rate": 4.0273401297497686e-05,
|
1162 |
+
"loss": 0.1798,
|
1163 |
+
"step": 14900
|
1164 |
+
},
|
1165 |
+
{
|
1166 |
+
"epoch": 9.54,
|
1167 |
+
"learning_rate": 3.98100092678406e-05,
|
1168 |
+
"loss": 0.1789,
|
1169 |
+
"step": 15000
|
1170 |
+
},
|
1171 |
+
{
|
1172 |
+
"epoch": 9.54,
|
1173 |
+
"eval_loss": 0.1695661097764969,
|
1174 |
+
"eval_runtime": 453.7028,
|
1175 |
+
"eval_samples_per_second": 12.098,
|
1176 |
+
"eval_steps_per_second": 0.758,
|
1177 |
+
"eval_wer": 0.11260079665792286,
|
1178 |
+
"step": 15000
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 9.61,
|
1182 |
+
"learning_rate": 3.93466172381835e-05,
|
1183 |
+
"loss": 0.2102,
|
1184 |
+
"step": 15100
|
1185 |
+
},
|
1186 |
+
{
|
1187 |
+
"epoch": 9.67,
|
1188 |
+
"learning_rate": 3.8883225208526416e-05,
|
1189 |
+
"loss": 0.1897,
|
1190 |
+
"step": 15200
|
1191 |
+
},
|
1192 |
+
{
|
1193 |
+
"epoch": 9.73,
|
1194 |
+
"learning_rate": 3.841983317886933e-05,
|
1195 |
+
"loss": 0.1768,
|
1196 |
+
"step": 15300
|
1197 |
+
},
|
1198 |
+
{
|
1199 |
+
"epoch": 9.8,
|
1200 |
+
"learning_rate": 3.795644114921223e-05,
|
1201 |
+
"loss": 0.1751,
|
1202 |
+
"step": 15400
|
1203 |
+
},
|
1204 |
+
{
|
1205 |
+
"epoch": 9.86,
|
1206 |
+
"learning_rate": 3.7493049119555145e-05,
|
1207 |
+
"loss": 0.1692,
|
1208 |
+
"step": 15500
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 9.86,
|
1212 |
+
"eval_loss": 0.18170228600502014,
|
1213 |
+
"eval_runtime": 455.8222,
|
1214 |
+
"eval_samples_per_second": 12.042,
|
1215 |
+
"eval_steps_per_second": 0.755,
|
1216 |
+
"eval_wer": 0.11220246769649277,
|
1217 |
+
"step": 15500
|
1218 |
+
},
|
1219 |
+
{
|
1220 |
+
"epoch": 9.92,
|
1221 |
+
"learning_rate": 3.702965708989806e-05,
|
1222 |
+
"loss": 0.1833,
|
1223 |
+
"step": 15600
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 9.99,
|
1227 |
+
"learning_rate": 3.656626506024097e-05,
|
1228 |
+
"loss": 0.179,
|
1229 |
+
"step": 15700
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 10.05,
|
1233 |
+
"learning_rate": 3.6102873030583875e-05,
|
1234 |
+
"loss": 0.1645,
|
1235 |
+
"step": 15800
|
1236 |
+
},
|
1237 |
+
{
|
1238 |
+
"epoch": 10.11,
|
1239 |
+
"learning_rate": 3.563948100092678e-05,
|
1240 |
+
"loss": 0.1699,
|
1241 |
+
"step": 15900
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 10.18,
|
1245 |
+
"learning_rate": 3.517608897126969e-05,
|
1246 |
+
"loss": 0.1765,
|
1247 |
+
"step": 16000
|
1248 |
+
},
|
1249 |
+
{
|
1250 |
+
"epoch": 10.18,
|
1251 |
+
"eval_loss": 0.17690393328666687,
|
1252 |
+
"eval_runtime": 456.2471,
|
1253 |
+
"eval_samples_per_second": 12.031,
|
1254 |
+
"eval_steps_per_second": 0.754,
|
1255 |
+
"eval_wer": 0.10927815019916448,
|
1256 |
+
"step": 16000
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 10.24,
|
1260 |
+
"learning_rate": 3.4712696941612605e-05,
|
1261 |
+
"loss": 0.1586,
|
1262 |
+
"step": 16100
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 10.31,
|
1266 |
+
"learning_rate": 3.424930491195552e-05,
|
1267 |
+
"loss": 0.1762,
|
1268 |
+
"step": 16200
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 10.37,
|
1272 |
+
"learning_rate": 3.378591288229842e-05,
|
1273 |
+
"loss": 0.1766,
|
1274 |
+
"step": 16300
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 10.43,
|
1278 |
+
"learning_rate": 3.3322520852641335e-05,
|
1279 |
+
"loss": 0.1698,
|
1280 |
+
"step": 16400
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 10.5,
|
1284 |
+
"learning_rate": 3.285912882298425e-05,
|
1285 |
+
"loss": 0.1699,
|
1286 |
+
"step": 16500
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 10.5,
|
1290 |
+
"eval_loss": 0.16039559245109558,
|
1291 |
+
"eval_runtime": 453.5742,
|
1292 |
+
"eval_samples_per_second": 12.102,
|
1293 |
+
"eval_steps_per_second": 0.758,
|
1294 |
+
"eval_wer": 0.10837462353055474,
|
1295 |
+
"step": 16500
|
1296 |
+
},
|
1297 |
+
{
|
1298 |
+
"epoch": 10.56,
|
1299 |
+
"learning_rate": 3.239573679332716e-05,
|
1300 |
+
"loss": 0.1649,
|
1301 |
+
"step": 16600
|
1302 |
+
},
|
1303 |
+
{
|
1304 |
+
"epoch": 10.62,
|
1305 |
+
"learning_rate": 3.193234476367007e-05,
|
1306 |
+
"loss": 0.1714,
|
1307 |
+
"step": 16700
|
1308 |
+
},
|
1309 |
+
{
|
1310 |
+
"epoch": 10.69,
|
1311 |
+
"learning_rate": 3.146895273401298e-05,
|
1312 |
+
"loss": 0.1609,
|
1313 |
+
"step": 16800
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 10.75,
|
1317 |
+
"learning_rate": 3.100556070435588e-05,
|
1318 |
+
"loss": 0.1678,
|
1319 |
+
"step": 16900
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 10.81,
|
1323 |
+
"learning_rate": 3.054680259499537e-05,
|
1324 |
+
"loss": 0.1591,
|
1325 |
+
"step": 17000
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 10.81,
|
1329 |
+
"eval_loss": 0.17774704098701477,
|
1330 |
+
"eval_runtime": 452.7751,
|
1331 |
+
"eval_samples_per_second": 12.123,
|
1332 |
+
"eval_steps_per_second": 0.76,
|
1333 |
+
"eval_wer": 0.10795686388807928,
|
1334 |
+
"step": 17000
|
1335 |
+
},
|
1336 |
+
{
|
1337 |
+
"epoch": 10.88,
|
1338 |
+
"learning_rate": 3.0083410565338278e-05,
|
1339 |
+
"loss": 0.1589,
|
1340 |
+
"step": 17100
|
1341 |
+
},
|
1342 |
+
{
|
1343 |
+
"epoch": 10.94,
|
1344 |
+
"learning_rate": 2.9620018535681187e-05,
|
1345 |
+
"loss": 0.2177,
|
1346 |
+
"step": 17200
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 11.01,
|
1350 |
+
"learning_rate": 2.91566265060241e-05,
|
1351 |
+
"loss": 0.171,
|
1352 |
+
"step": 17300
|
1353 |
+
},
|
1354 |
+
{
|
1355 |
+
"epoch": 11.07,
|
1356 |
+
"learning_rate": 2.8693234476367008e-05,
|
1357 |
+
"loss": 0.1625,
|
1358 |
+
"step": 17400
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 11.13,
|
1362 |
+
"learning_rate": 2.822984244670992e-05,
|
1363 |
+
"loss": 0.1499,
|
1364 |
+
"step": 17500
|
1365 |
+
},
|
1366 |
+
{
|
1367 |
+
"epoch": 11.13,
|
1368 |
+
"eval_loss": 0.16445568203926086,
|
1369 |
+
"eval_runtime": 453.5993,
|
1370 |
+
"eval_samples_per_second": 12.101,
|
1371 |
+
"eval_steps_per_second": 0.758,
|
1372 |
+
"eval_wer": 0.10737394345671816,
|
1373 |
+
"step": 17500
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 11.2,
|
1377 |
+
"learning_rate": 2.7766450417052832e-05,
|
1378 |
+
"loss": 0.1796,
|
1379 |
+
"step": 17600
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 11.26,
|
1383 |
+
"learning_rate": 2.7303058387395734e-05,
|
1384 |
+
"loss": 0.1524,
|
1385 |
+
"step": 17700
|
1386 |
+
},
|
1387 |
+
{
|
1388 |
+
"epoch": 11.32,
|
1389 |
+
"learning_rate": 2.6839666357738646e-05,
|
1390 |
+
"loss": 0.148,
|
1391 |
+
"step": 17800
|
1392 |
+
},
|
1393 |
+
{
|
1394 |
+
"epoch": 11.39,
|
1395 |
+
"learning_rate": 2.637627432808156e-05,
|
1396 |
+
"loss": 0.1566,
|
1397 |
+
"step": 17900
|
1398 |
+
},
|
1399 |
+
{
|
1400 |
+
"epoch": 11.45,
|
1401 |
+
"learning_rate": 2.5912882298424467e-05,
|
1402 |
+
"loss": 0.163,
|
1403 |
+
"step": 18000
|
1404 |
+
},
|
1405 |
+
{
|
1406 |
+
"epoch": 11.45,
|
1407 |
+
"eval_loss": 0.17037604749202728,
|
1408 |
+
"eval_runtime": 453.4998,
|
1409 |
+
"eval_samples_per_second": 12.104,
|
1410 |
+
"eval_steps_per_second": 0.759,
|
1411 |
+
"eval_wer": 0.10646070144758574,
|
1412 |
+
"step": 18000
|
1413 |
+
},
|
1414 |
+
{
|
1415 |
+
"epoch": 11.51,
|
1416 |
+
"learning_rate": 2.544949026876738e-05,
|
1417 |
+
"loss": 0.1622,
|
1418 |
+
"step": 18100
|
1419 |
+
},
|
1420 |
+
{
|
1421 |
+
"epoch": 11.58,
|
1422 |
+
"learning_rate": 2.498609823911029e-05,
|
1423 |
+
"loss": 0.1528,
|
1424 |
+
"step": 18200
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 11.64,
|
1428 |
+
"learning_rate": 2.4522706209453197e-05,
|
1429 |
+
"loss": 0.1699,
|
1430 |
+
"step": 18300
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 11.7,
|
1434 |
+
"learning_rate": 2.405931417979611e-05,
|
1435 |
+
"loss": 0.1587,
|
1436 |
+
"step": 18400
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 11.77,
|
1440 |
+
"learning_rate": 2.3595922150139018e-05,
|
1441 |
+
"loss": 0.1597,
|
1442 |
+
"step": 18500
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 11.77,
|
1446 |
+
"eval_loss": 0.15760080516338348,
|
1447 |
+
"eval_runtime": 452.5827,
|
1448 |
+
"eval_samples_per_second": 12.128,
|
1449 |
+
"eval_steps_per_second": 0.76,
|
1450 |
+
"eval_wer": 0.10640240940444963,
|
1451 |
+
"step": 18500
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 11.83,
|
1455 |
+
"learning_rate": 2.31371640407785e-05,
|
1456 |
+
"loss": 0.1507,
|
1457 |
+
"step": 18600
|
1458 |
+
},
|
1459 |
+
{
|
1460 |
+
"epoch": 11.9,
|
1461 |
+
"learning_rate": 2.267377201112141e-05,
|
1462 |
+
"loss": 0.1675,
|
1463 |
+
"step": 18700
|
1464 |
+
},
|
1465 |
+
{
|
1466 |
+
"epoch": 11.96,
|
1467 |
+
"learning_rate": 2.221037998146432e-05,
|
1468 |
+
"loss": 0.1515,
|
1469 |
+
"step": 18800
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 12.02,
|
1473 |
+
"learning_rate": 2.1746987951807228e-05,
|
1474 |
+
"loss": 0.1398,
|
1475 |
+
"step": 18900
|
1476 |
+
},
|
1477 |
+
{
|
1478 |
+
"epoch": 12.09,
|
1479 |
+
"learning_rate": 2.128359592215014e-05,
|
1480 |
+
"loss": 0.1484,
|
1481 |
+
"step": 19000
|
1482 |
+
},
|
1483 |
+
{
|
1484 |
+
"epoch": 12.09,
|
1485 |
+
"eval_loss": 0.16369372606277466,
|
1486 |
+
"eval_runtime": 452.8951,
|
1487 |
+
"eval_samples_per_second": 12.12,
|
1488 |
+
"eval_steps_per_second": 0.76,
|
1489 |
+
"eval_wer": 0.10407072767900515,
|
1490 |
+
"step": 19000
|
1491 |
+
},
|
1492 |
+
{
|
1493 |
+
"epoch": 12.15,
|
1494 |
+
"learning_rate": 2.0820203892493052e-05,
|
1495 |
+
"loss": 0.1459,
|
1496 |
+
"step": 19100
|
1497 |
+
},
|
1498 |
+
{
|
1499 |
+
"epoch": 12.21,
|
1500 |
+
"learning_rate": 2.0361445783132532e-05,
|
1501 |
+
"loss": 0.155,
|
1502 |
+
"step": 19200
|
1503 |
+
},
|
1504 |
+
{
|
1505 |
+
"epoch": 12.28,
|
1506 |
+
"learning_rate": 1.989805375347544e-05,
|
1507 |
+
"loss": 0.1655,
|
1508 |
+
"step": 19300
|
1509 |
+
},
|
1510 |
+
{
|
1511 |
+
"epoch": 12.34,
|
1512 |
+
"learning_rate": 1.943466172381835e-05,
|
1513 |
+
"loss": 0.137,
|
1514 |
+
"step": 19400
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 12.4,
|
1518 |
+
"learning_rate": 1.8971269694161262e-05,
|
1519 |
+
"loss": 0.1464,
|
1520 |
+
"step": 19500
|
1521 |
+
},
|
1522 |
+
{
|
1523 |
+
"epoch": 12.4,
|
1524 |
+
"eval_loss": 0.16305148601531982,
|
1525 |
+
"eval_runtime": 454.6084,
|
1526 |
+
"eval_samples_per_second": 12.074,
|
1527 |
+
"eval_steps_per_second": 0.757,
|
1528 |
+
"eval_wer": 0.10473137083454775,
|
1529 |
+
"step": 19500
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 12.47,
|
1533 |
+
"learning_rate": 1.8507877664504174e-05,
|
1534 |
+
"loss": 0.1514,
|
1535 |
+
"step": 19600
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 12.53,
|
1539 |
+
"learning_rate": 1.804448563484708e-05,
|
1540 |
+
"loss": 0.1485,
|
1541 |
+
"step": 19700
|
1542 |
+
},
|
1543 |
+
{
|
1544 |
+
"epoch": 12.6,
|
1545 |
+
"learning_rate": 1.7581093605189992e-05,
|
1546 |
+
"loss": 0.1524,
|
1547 |
+
"step": 19800
|
1548 |
+
},
|
1549 |
+
{
|
1550 |
+
"epoch": 12.66,
|
1551 |
+
"learning_rate": 1.71177015755329e-05,
|
1552 |
+
"loss": 0.1488,
|
1553 |
+
"step": 19900
|
1554 |
+
},
|
1555 |
+
{
|
1556 |
+
"epoch": 12.72,
|
1557 |
+
"learning_rate": 1.6654309545875813e-05,
|
1558 |
+
"loss": 0.156,
|
1559 |
+
"step": 20000
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 12.72,
|
1563 |
+
"eval_loss": 0.16862976551055908,
|
1564 |
+
"eval_runtime": 454.3682,
|
1565 |
+
"eval_samples_per_second": 12.081,
|
1566 |
+
"eval_steps_per_second": 0.757,
|
1567 |
+
"eval_wer": 0.10287574079471486,
|
1568 |
+
"step": 20000
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 12.79,
|
1572 |
+
"learning_rate": 1.619091751621872e-05,
|
1573 |
+
"loss": 0.1517,
|
1574 |
+
"step": 20100
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 12.85,
|
1578 |
+
"learning_rate": 1.572752548656163e-05,
|
1579 |
+
"loss": 0.1482,
|
1580 |
+
"step": 20200
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 12.91,
|
1584 |
+
"learning_rate": 1.5264133456904543e-05,
|
1585 |
+
"loss": 0.1451,
|
1586 |
+
"step": 20300
|
1587 |
+
},
|
1588 |
+
{
|
1589 |
+
"epoch": 12.98,
|
1590 |
+
"learning_rate": 1.4800741427247453e-05,
|
1591 |
+
"loss": 0.1506,
|
1592 |
+
"step": 20400
|
1593 |
+
},
|
1594 |
+
{
|
1595 |
+
"epoch": 13.04,
|
1596 |
+
"learning_rate": 1.4337349397590364e-05,
|
1597 |
+
"loss": 0.1625,
|
1598 |
+
"step": 20500
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 13.04,
|
1602 |
+
"eval_loss": 0.16482920944690704,
|
1603 |
+
"eval_runtime": 453.7585,
|
1604 |
+
"eval_samples_per_second": 12.097,
|
1605 |
+
"eval_steps_per_second": 0.758,
|
1606 |
+
"eval_wer": 0.10226367434178568,
|
1607 |
+
"step": 20500
|
1608 |
+
},
|
1609 |
+
{
|
1610 |
+
"epoch": 13.1,
|
1611 |
+
"learning_rate": 1.387395736793327e-05,
|
1612 |
+
"loss": 0.1317,
|
1613 |
+
"step": 20600
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 13.17,
|
1617 |
+
"learning_rate": 1.3410565338276181e-05,
|
1618 |
+
"loss": 0.1543,
|
1619 |
+
"step": 20700
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 13.23,
|
1623 |
+
"learning_rate": 1.2947173308619093e-05,
|
1624 |
+
"loss": 0.1401,
|
1625 |
+
"step": 20800
|
1626 |
+
},
|
1627 |
+
{
|
1628 |
+
"epoch": 13.3,
|
1629 |
+
"learning_rate": 1.2483781278962002e-05,
|
1630 |
+
"loss": 0.1445,
|
1631 |
+
"step": 20900
|
1632 |
+
},
|
1633 |
+
{
|
1634 |
+
"epoch": 13.36,
|
1635 |
+
"learning_rate": 1.2020389249304913e-05,
|
1636 |
+
"loss": 0.1395,
|
1637 |
+
"step": 21000
|
1638 |
+
},
|
1639 |
+
{
|
1640 |
+
"epoch": 13.36,
|
1641 |
+
"eval_loss": 0.16876418888568878,
|
1642 |
+
"eval_runtime": 457.4225,
|
1643 |
+
"eval_samples_per_second": 12.0,
|
1644 |
+
"eval_steps_per_second": 0.752,
|
1645 |
+
"eval_wer": 0.10274944136791994,
|
1646 |
+
"step": 21000
|
1647 |
+
},
|
1648 |
+
{
|
1649 |
+
"epoch": 13.42,
|
1650 |
+
"learning_rate": 1.1556997219647822e-05,
|
1651 |
+
"loss": 0.1385,
|
1652 |
+
"step": 21100
|
1653 |
+
},
|
1654 |
+
{
|
1655 |
+
"epoch": 13.49,
|
1656 |
+
"learning_rate": 1.1093605189990732e-05,
|
1657 |
+
"loss": 0.1528,
|
1658 |
+
"step": 21200
|
1659 |
+
},
|
1660 |
+
{
|
1661 |
+
"epoch": 13.55,
|
1662 |
+
"learning_rate": 1.0634847080630214e-05,
|
1663 |
+
"loss": 0.14,
|
1664 |
+
"step": 21300
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 13.61,
|
1668 |
+
"learning_rate": 1.0171455050973124e-05,
|
1669 |
+
"loss": 0.1438,
|
1670 |
+
"step": 21400
|
1671 |
+
},
|
1672 |
+
{
|
1673 |
+
"epoch": 13.68,
|
1674 |
+
"learning_rate": 9.708063021316035e-06,
|
1675 |
+
"loss": 0.1387,
|
1676 |
+
"step": 21500
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 13.68,
|
1680 |
+
"eval_loss": 0.16695508360862732,
|
1681 |
+
"eval_runtime": 453.4195,
|
1682 |
+
"eval_samples_per_second": 12.106,
|
1683 |
+
"eval_steps_per_second": 0.759,
|
1684 |
+
"eval_wer": 0.10130185563003984,
|
1685 |
+
"step": 21500
|
1686 |
+
},
|
1687 |
+
{
|
1688 |
+
"epoch": 13.74,
|
1689 |
+
"learning_rate": 9.244670991658943e-06,
|
1690 |
+
"loss": 0.151,
|
1691 |
+
"step": 21600
|
1692 |
+
},
|
1693 |
+
{
|
1694 |
+
"epoch": 13.8,
|
1695 |
+
"learning_rate": 8.781278962001854e-06,
|
1696 |
+
"loss": 0.1425,
|
1697 |
+
"step": 21700
|
1698 |
+
},
|
1699 |
+
{
|
1700 |
+
"epoch": 13.87,
|
1701 |
+
"learning_rate": 8.317886932344763e-06,
|
1702 |
+
"loss": 0.1429,
|
1703 |
+
"step": 21800
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 13.93,
|
1707 |
+
"learning_rate": 7.854494902687675e-06,
|
1708 |
+
"loss": 0.139,
|
1709 |
+
"step": 21900
|
1710 |
+
},
|
1711 |
+
{
|
1712 |
+
"epoch": 13.99,
|
1713 |
+
"learning_rate": 7.391102873030584e-06,
|
1714 |
+
"loss": 0.1434,
|
1715 |
+
"step": 22000
|
1716 |
+
},
|
1717 |
+
{
|
1718 |
+
"epoch": 13.99,
|
1719 |
+
"eval_loss": 0.16770704090595245,
|
1720 |
+
"eval_runtime": 452.4411,
|
1721 |
+
"eval_samples_per_second": 12.132,
|
1722 |
+
"eval_steps_per_second": 0.76,
|
1723 |
+
"eval_wer": 0.10168075391042455,
|
1724 |
+
"step": 22000
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 14.06,
|
1728 |
+
"learning_rate": 6.927710843373494e-06,
|
1729 |
+
"loss": 0.1225,
|
1730 |
+
"step": 22100
|
1731 |
+
},
|
1732 |
+
{
|
1733 |
+
"epoch": 14.12,
|
1734 |
+
"learning_rate": 6.464318813716404e-06,
|
1735 |
+
"loss": 0.1387,
|
1736 |
+
"step": 22200
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 14.19,
|
1740 |
+
"learning_rate": 6.0009267840593145e-06,
|
1741 |
+
"loss": 0.135,
|
1742 |
+
"step": 22300
|
1743 |
+
},
|
1744 |
+
{
|
1745 |
+
"epoch": 14.25,
|
1746 |
+
"learning_rate": 5.537534754402224e-06,
|
1747 |
+
"loss": 0.1362,
|
1748 |
+
"step": 22400
|
1749 |
+
},
|
1750 |
+
{
|
1751 |
+
"epoch": 14.31,
|
1752 |
+
"learning_rate": 5.074142724745135e-06,
|
1753 |
+
"loss": 0.1442,
|
1754 |
+
"step": 22500
|
1755 |
+
},
|
1756 |
+
{
|
1757 |
+
"epoch": 14.31,
|
1758 |
+
"eval_loss": 0.16875232756137848,
|
1759 |
+
"eval_runtime": 455.5341,
|
1760 |
+
"eval_samples_per_second": 12.05,
|
1761 |
+
"eval_steps_per_second": 0.755,
|
1762 |
+
"eval_wer": 0.10080637326338288,
|
1763 |
+
"step": 22500
|
1764 |
+
},
|
1765 |
+
{
|
1766 |
+
"epoch": 14.38,
|
1767 |
+
"learning_rate": 4.615384615384616e-06,
|
1768 |
+
"loss": 0.1383,
|
1769 |
+
"step": 22600
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 14.44,
|
1773 |
+
"learning_rate": 4.151992585727526e-06,
|
1774 |
+
"loss": 0.1407,
|
1775 |
+
"step": 22700
|
1776 |
+
},
|
1777 |
+
{
|
1778 |
+
"epoch": 14.5,
|
1779 |
+
"learning_rate": 3.688600556070436e-06,
|
1780 |
+
"loss": 0.1412,
|
1781 |
+
"step": 22800
|
1782 |
+
},
|
1783 |
+
{
|
1784 |
+
"epoch": 14.57,
|
1785 |
+
"learning_rate": 3.225208526413346e-06,
|
1786 |
+
"loss": 0.1467,
|
1787 |
+
"step": 22900
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 14.63,
|
1791 |
+
"learning_rate": 2.761816496756256e-06,
|
1792 |
+
"loss": 0.1439,
|
1793 |
+
"step": 23000
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 14.63,
|
1797 |
+
"eval_loss": 0.16467291116714478,
|
1798 |
+
"eval_runtime": 451.7915,
|
1799 |
+
"eval_samples_per_second": 12.149,
|
1800 |
+
"eval_steps_per_second": 0.761,
|
1801 |
+
"eval_wer": 0.10040804430195278,
|
1802 |
+
"step": 23000
|
1803 |
+
},
|
1804 |
+
{
|
1805 |
+
"epoch": 14.69,
|
1806 |
+
"learning_rate": 2.298424467099166e-06,
|
1807 |
+
"loss": 0.1439,
|
1808 |
+
"step": 23100
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 14.76,
|
1812 |
+
"learning_rate": 1.835032437442076e-06,
|
1813 |
+
"loss": 0.1346,
|
1814 |
+
"step": 23200
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 14.82,
|
1818 |
+
"learning_rate": 1.3716404077849862e-06,
|
1819 |
+
"loss": 0.1441,
|
1820 |
+
"step": 23300
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 14.89,
|
1824 |
+
"learning_rate": 9.082483781278962e-07,
|
1825 |
+
"loss": 0.1386,
|
1826 |
+
"step": 23400
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 14.95,
|
1830 |
+
"learning_rate": 4.448563484708063e-07,
|
1831 |
+
"loss": 0.137,
|
1832 |
+
"step": 23500
|
1833 |
+
},
|
1834 |
+
{
|
1835 |
+
"epoch": 14.95,
|
1836 |
+
"eval_loss": 0.1636081039905548,
|
1837 |
+
"eval_runtime": 454.2899,
|
1838 |
+
"eval_samples_per_second": 12.083,
|
1839 |
+
"eval_steps_per_second": 0.757,
|
1840 |
+
"eval_wer": 0.10055377440979306,
|
1841 |
+
"step": 23500
|
1842 |
+
},
|
1843 |
+
{
|
1844 |
+
"epoch": 15.0,
|
1845 |
+
"step": 23580,
|
1846 |
+
"total_flos": 1.645470128774266e+20,
|
1847 |
+
"train_loss": 0.41333935497371294,
|
1848 |
+
"train_runtime": 85566.3736,
|
1849 |
+
"train_samples_per_second": 8.82,
|
1850 |
+
"train_steps_per_second": 0.276
|
1851 |
+
}
|
1852 |
+
],
|
1853 |
+
"max_steps": 23580,
|
1854 |
+
"num_train_epochs": 15,
|
1855 |
+
"total_flos": 1.645470128774266e+20,
|
1856 |
+
"trial_name": null,
|
1857 |
+
"trial_params": null
|
1858 |
+
}
|