End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +2365 -0
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 100.0,
|
3 |
+
"eval_loss": 0.22239123284816742,
|
4 |
+
"eval_runtime": 238.7764,
|
5 |
+
"eval_samples": 8339,
|
6 |
+
"eval_samples_per_second": 34.924,
|
7 |
+
"eval_steps_per_second": 1.093,
|
8 |
+
"eval_wer": 0.2869353988587666,
|
9 |
+
"train_loss": 1.2247810986577248,
|
10 |
+
"train_runtime": 113701.5353,
|
11 |
+
"train_samples": 25058,
|
12 |
+
"train_samples_per_second": 22.038,
|
13 |
+
"train_steps_per_second": 0.69
|
14 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 100.0,
|
3 |
+
"eval_loss": 0.22239123284816742,
|
4 |
+
"eval_runtime": 238.7764,
|
5 |
+
"eval_samples": 8339,
|
6 |
+
"eval_samples_per_second": 34.924,
|
7 |
+
"eval_steps_per_second": 1.093,
|
8 |
+
"eval_wer": 0.2869353988587666
|
9 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 100.0,
|
3 |
+
"train_loss": 1.2247810986577248,
|
4 |
+
"train_runtime": 113701.5353,
|
5 |
+
"train_samples": 25058,
|
6 |
+
"train_samples_per_second": 22.038,
|
7 |
+
"train_steps_per_second": 0.69
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,2365 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 100.0,
|
5 |
+
"global_step": 78400,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.64,
|
12 |
+
"learning_rate": 1.9880000000000003e-05,
|
13 |
+
"loss": 6.8222,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.64,
|
18 |
+
"eval_loss": 3.502612352371216,
|
19 |
+
"eval_runtime": 238.4374,
|
20 |
+
"eval_samples_per_second": 34.974,
|
21 |
+
"eval_steps_per_second": 1.095,
|
22 |
+
"eval_wer": 1.0,
|
23 |
+
"step": 500
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 1.28,
|
27 |
+
"learning_rate": 1.987240051347882e-05,
|
28 |
+
"loss": 3.2136,
|
29 |
+
"step": 1000
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 1.28,
|
33 |
+
"eval_loss": 3.0592851638793945,
|
34 |
+
"eval_runtime": 239.7957,
|
35 |
+
"eval_samples_per_second": 34.775,
|
36 |
+
"eval_steps_per_second": 1.088,
|
37 |
+
"eval_wer": 0.9999770836675298,
|
38 |
+
"step": 1000
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 1.91,
|
42 |
+
"learning_rate": 1.9744030808729143e-05,
|
43 |
+
"loss": 2.8882,
|
44 |
+
"step": 1500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.91,
|
48 |
+
"eval_loss": 2.4670279026031494,
|
49 |
+
"eval_runtime": 236.3109,
|
50 |
+
"eval_samples_per_second": 35.288,
|
51 |
+
"eval_steps_per_second": 1.104,
|
52 |
+
"eval_wer": 0.9938584228979994,
|
53 |
+
"step": 1500
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 2.55,
|
57 |
+
"learning_rate": 1.9615661103979463e-05,
|
58 |
+
"loss": 2.3743,
|
59 |
+
"step": 2000
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 2.55,
|
63 |
+
"eval_loss": 1.1843628883361816,
|
64 |
+
"eval_runtime": 239.6417,
|
65 |
+
"eval_samples_per_second": 34.798,
|
66 |
+
"eval_steps_per_second": 1.089,
|
67 |
+
"eval_wer": 0.8656873753924422,
|
68 |
+
"step": 2000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 3.19,
|
72 |
+
"learning_rate": 1.9487291399229784e-05,
|
73 |
+
"loss": 1.9456,
|
74 |
+
"step": 2500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 3.19,
|
78 |
+
"eval_loss": 0.8227887153625488,
|
79 |
+
"eval_runtime": 238.3867,
|
80 |
+
"eval_samples_per_second": 34.981,
|
81 |
+
"eval_steps_per_second": 1.095,
|
82 |
+
"eval_wer": 0.7397162958040195,
|
83 |
+
"step": 2500
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 3.83,
|
87 |
+
"learning_rate": 1.9358921694480104e-05,
|
88 |
+
"loss": 1.7781,
|
89 |
+
"step": 3000
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 3.83,
|
93 |
+
"eval_loss": 0.6826250553131104,
|
94 |
+
"eval_runtime": 237.1431,
|
95 |
+
"eval_samples_per_second": 35.164,
|
96 |
+
"eval_steps_per_second": 1.101,
|
97 |
+
"eval_wer": 0.6752984852304237,
|
98 |
+
"step": 3000
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 4.46,
|
102 |
+
"learning_rate": 1.9230551989730427e-05,
|
103 |
+
"loss": 1.6848,
|
104 |
+
"step": 3500
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 4.46,
|
108 |
+
"eval_loss": 0.5885177254676819,
|
109 |
+
"eval_runtime": 235.3547,
|
110 |
+
"eval_samples_per_second": 35.432,
|
111 |
+
"eval_steps_per_second": 1.109,
|
112 |
+
"eval_wer": 0.6139743795402983,
|
113 |
+
"step": 3500
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 5.1,
|
117 |
+
"learning_rate": 1.9102182284980744e-05,
|
118 |
+
"loss": 1.6228,
|
119 |
+
"step": 4000
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 5.1,
|
123 |
+
"eval_loss": 0.527351975440979,
|
124 |
+
"eval_runtime": 237.0762,
|
125 |
+
"eval_samples_per_second": 35.174,
|
126 |
+
"eval_steps_per_second": 1.101,
|
127 |
+
"eval_wer": 0.5788894745284965,
|
128 |
+
"step": 4000
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 5.74,
|
132 |
+
"learning_rate": 1.8973812580231068e-05,
|
133 |
+
"loss": 1.5768,
|
134 |
+
"step": 4500
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 5.74,
|
138 |
+
"eval_loss": 0.48999691009521484,
|
139 |
+
"eval_runtime": 235.6055,
|
140 |
+
"eval_samples_per_second": 35.394,
|
141 |
+
"eval_steps_per_second": 1.108,
|
142 |
+
"eval_wer": 0.5518711185461879,
|
143 |
+
"step": 4500
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 6.38,
|
147 |
+
"learning_rate": 1.8845699614890885e-05,
|
148 |
+
"loss": 1.5431,
|
149 |
+
"step": 5000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 6.38,
|
153 |
+
"eval_loss": 0.45079049468040466,
|
154 |
+
"eval_runtime": 235.5935,
|
155 |
+
"eval_samples_per_second": 35.396,
|
156 |
+
"eval_steps_per_second": 1.108,
|
157 |
+
"eval_wer": 0.5237986112702523,
|
158 |
+
"step": 5000
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 7.02,
|
162 |
+
"learning_rate": 1.871732991014121e-05,
|
163 |
+
"loss": 1.5019,
|
164 |
+
"step": 5500
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 7.02,
|
168 |
+
"eval_loss": 0.42483240365982056,
|
169 |
+
"eval_runtime": 235.975,
|
170 |
+
"eval_samples_per_second": 35.338,
|
171 |
+
"eval_steps_per_second": 1.106,
|
172 |
+
"eval_wer": 0.5021426770859592,
|
173 |
+
"step": 5500
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 7.65,
|
177 |
+
"learning_rate": 1.858921694480103e-05,
|
178 |
+
"loss": 1.4684,
|
179 |
+
"step": 6000
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 7.65,
|
183 |
+
"eval_loss": 0.4009439945220947,
|
184 |
+
"eval_runtime": 236.3695,
|
185 |
+
"eval_samples_per_second": 35.28,
|
186 |
+
"eval_steps_per_second": 1.104,
|
187 |
+
"eval_wer": 0.4827096271512707,
|
188 |
+
"step": 6000
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 8.29,
|
192 |
+
"learning_rate": 1.846084724005135e-05,
|
193 |
+
"loss": 1.4635,
|
194 |
+
"step": 6500
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 8.29,
|
198 |
+
"eval_loss": 0.3830302357673645,
|
199 |
+
"eval_runtime": 234.8449,
|
200 |
+
"eval_samples_per_second": 35.509,
|
201 |
+
"eval_steps_per_second": 1.111,
|
202 |
+
"eval_wer": 0.4700139789628068,
|
203 |
+
"step": 6500
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 8.93,
|
207 |
+
"learning_rate": 1.833247753530167e-05,
|
208 |
+
"loss": 1.4291,
|
209 |
+
"step": 7000
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 8.93,
|
213 |
+
"eval_loss": 0.3707072138786316,
|
214 |
+
"eval_runtime": 237.9649,
|
215 |
+
"eval_samples_per_second": 35.043,
|
216 |
+
"eval_steps_per_second": 1.097,
|
217 |
+
"eval_wer": 0.45947246602653713,
|
218 |
+
"step": 7000
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 9.57,
|
222 |
+
"learning_rate": 1.820410783055199e-05,
|
223 |
+
"loss": 1.4271,
|
224 |
+
"step": 7500
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 9.57,
|
228 |
+
"eval_loss": 0.3570108115673065,
|
229 |
+
"eval_runtime": 236.9986,
|
230 |
+
"eval_samples_per_second": 35.186,
|
231 |
+
"eval_steps_per_second": 1.101,
|
232 |
+
"eval_wer": 0.4514059169970438,
|
233 |
+
"step": 7500
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 10.2,
|
237 |
+
"learning_rate": 1.807573812580231e-05,
|
238 |
+
"loss": 1.3938,
|
239 |
+
"step": 8000
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 10.2,
|
243 |
+
"eval_loss": 0.34792855381965637,
|
244 |
+
"eval_runtime": 234.9447,
|
245 |
+
"eval_samples_per_second": 35.493,
|
246 |
+
"eval_steps_per_second": 1.111,
|
247 |
+
"eval_wer": 0.4378394481747141,
|
248 |
+
"step": 8000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 10.84,
|
252 |
+
"learning_rate": 1.794762516046213e-05,
|
253 |
+
"loss": 1.3914,
|
254 |
+
"step": 8500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 10.84,
|
258 |
+
"eval_loss": 0.3395732045173645,
|
259 |
+
"eval_runtime": 235.1285,
|
260 |
+
"eval_samples_per_second": 35.466,
|
261 |
+
"eval_steps_per_second": 1.11,
|
262 |
+
"eval_wer": 0.4368082132135573,
|
263 |
+
"step": 8500
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 11.48,
|
267 |
+
"learning_rate": 1.7819255455712454e-05,
|
268 |
+
"loss": 1.3767,
|
269 |
+
"step": 9000
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 11.48,
|
273 |
+
"eval_loss": 0.3253381848335266,
|
274 |
+
"eval_runtime": 238.3808,
|
275 |
+
"eval_samples_per_second": 34.982,
|
276 |
+
"eval_steps_per_second": 1.095,
|
277 |
+
"eval_wer": 0.42622086761234734,
|
278 |
+
"step": 9000
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 12.12,
|
282 |
+
"learning_rate": 1.7690885750962775e-05,
|
283 |
+
"loss": 1.3641,
|
284 |
+
"step": 9500
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 12.12,
|
288 |
+
"eval_loss": 0.32509171962738037,
|
289 |
+
"eval_runtime": 234.3424,
|
290 |
+
"eval_samples_per_second": 35.585,
|
291 |
+
"eval_steps_per_second": 1.114,
|
292 |
+
"eval_wer": 0.4178105735958017,
|
293 |
+
"step": 9500
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 12.76,
|
297 |
+
"learning_rate": 1.7562772785622595e-05,
|
298 |
+
"loss": 1.355,
|
299 |
+
"step": 10000
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 12.76,
|
303 |
+
"eval_loss": 0.3138255476951599,
|
304 |
+
"eval_runtime": 234.3344,
|
305 |
+
"eval_samples_per_second": 35.586,
|
306 |
+
"eval_steps_per_second": 1.114,
|
307 |
+
"eval_wer": 0.4135710520888237,
|
308 |
+
"step": 10000
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 13.39,
|
312 |
+
"learning_rate": 1.7434403080872915e-05,
|
313 |
+
"loss": 1.336,
|
314 |
+
"step": 10500
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 13.39,
|
318 |
+
"eval_loss": 0.31212833523750305,
|
319 |
+
"eval_runtime": 238.4088,
|
320 |
+
"eval_samples_per_second": 34.978,
|
321 |
+
"eval_steps_per_second": 1.095,
|
322 |
+
"eval_wer": 0.40694823200494995,
|
323 |
+
"step": 10500
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 14.03,
|
327 |
+
"learning_rate": 1.7306033376123236e-05,
|
328 |
+
"loss": 1.3292,
|
329 |
+
"step": 11000
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 14.03,
|
333 |
+
"eval_loss": 0.30410972237586975,
|
334 |
+
"eval_runtime": 233.9502,
|
335 |
+
"eval_samples_per_second": 35.644,
|
336 |
+
"eval_steps_per_second": 1.116,
|
337 |
+
"eval_wer": 0.4014483122121136,
|
338 |
+
"step": 11000
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 14.67,
|
342 |
+
"learning_rate": 1.7177663671373556e-05,
|
343 |
+
"loss": 1.3249,
|
344 |
+
"step": 11500
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 14.67,
|
348 |
+
"eval_loss": 0.3013543486595154,
|
349 |
+
"eval_runtime": 234.7846,
|
350 |
+
"eval_samples_per_second": 35.518,
|
351 |
+
"eval_steps_per_second": 1.112,
|
352 |
+
"eval_wer": 0.3931296835254486,
|
353 |
+
"step": 11500
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 15.31,
|
357 |
+
"learning_rate": 1.704929396662388e-05,
|
358 |
+
"loss": 1.3156,
|
359 |
+
"step": 12000
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 15.31,
|
363 |
+
"eval_loss": 0.301371306180954,
|
364 |
+
"eval_runtime": 237.6831,
|
365 |
+
"eval_samples_per_second": 35.085,
|
366 |
+
"eval_steps_per_second": 1.098,
|
367 |
+
"eval_wer": 0.39294635286568735,
|
368 |
+
"step": 12000
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 15.94,
|
372 |
+
"learning_rate": 1.69209242618742e-05,
|
373 |
+
"loss": 1.313,
|
374 |
+
"step": 12500
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 15.94,
|
378 |
+
"eval_loss": 0.29687899351119995,
|
379 |
+
"eval_runtime": 234.864,
|
380 |
+
"eval_samples_per_second": 35.506,
|
381 |
+
"eval_steps_per_second": 1.111,
|
382 |
+
"eval_wer": 0.39679629672067285,
|
383 |
+
"step": 12500
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 16.58,
|
387 |
+
"learning_rate": 1.679255455712452e-05,
|
388 |
+
"loss": 1.3068,
|
389 |
+
"step": 13000
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 16.58,
|
393 |
+
"eval_loss": 0.2965061664581299,
|
394 |
+
"eval_runtime": 239.4104,
|
395 |
+
"eval_samples_per_second": 34.831,
|
396 |
+
"eval_steps_per_second": 1.09,
|
397 |
+
"eval_wer": 0.39663588239338177,
|
398 |
+
"step": 13000
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 17.22,
|
402 |
+
"learning_rate": 1.666444159178434e-05,
|
403 |
+
"loss": 1.2785,
|
404 |
+
"step": 13500
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 17.22,
|
408 |
+
"eval_loss": 0.29434436559677124,
|
409 |
+
"eval_runtime": 241.1756,
|
410 |
+
"eval_samples_per_second": 34.576,
|
411 |
+
"eval_steps_per_second": 1.082,
|
412 |
+
"eval_wer": 0.38497146916607466,
|
413 |
+
"step": 13500
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 17.86,
|
417 |
+
"learning_rate": 1.653607188703466e-05,
|
418 |
+
"loss": 1.2867,
|
419 |
+
"step": 14000
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 17.86,
|
423 |
+
"eval_loss": 0.291163831949234,
|
424 |
+
"eval_runtime": 234.1726,
|
425 |
+
"eval_samples_per_second": 35.61,
|
426 |
+
"eval_steps_per_second": 1.115,
|
427 |
+
"eval_wer": 0.37821115108737996,
|
428 |
+
"step": 14000
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 18.49,
|
432 |
+
"learning_rate": 1.640770218228498e-05,
|
433 |
+
"loss": 1.2714,
|
434 |
+
"step": 14500
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 18.49,
|
438 |
+
"eval_loss": 0.281938374042511,
|
439 |
+
"eval_runtime": 233.7302,
|
440 |
+
"eval_samples_per_second": 35.678,
|
441 |
+
"eval_steps_per_second": 1.117,
|
442 |
+
"eval_wer": 0.3746820358869766,
|
443 |
+
"step": 14500
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 19.13,
|
447 |
+
"learning_rate": 1.6279332477535305e-05,
|
448 |
+
"loss": 1.2844,
|
449 |
+
"step": 15000
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 19.13,
|
453 |
+
"eval_loss": 0.28400421142578125,
|
454 |
+
"eval_runtime": 234.7428,
|
455 |
+
"eval_samples_per_second": 35.524,
|
456 |
+
"eval_steps_per_second": 1.112,
|
457 |
+
"eval_wer": 0.3740403785778124,
|
458 |
+
"step": 15000
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 19.77,
|
462 |
+
"learning_rate": 1.6150962772785625e-05,
|
463 |
+
"loss": 1.2684,
|
464 |
+
"step": 15500
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 19.77,
|
468 |
+
"eval_loss": 0.2913394272327423,
|
469 |
+
"eval_runtime": 238.9443,
|
470 |
+
"eval_samples_per_second": 34.899,
|
471 |
+
"eval_steps_per_second": 1.092,
|
472 |
+
"eval_wer": 0.38279441758141025,
|
473 |
+
"step": 15500
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 20.41,
|
477 |
+
"learning_rate": 1.6022849807445445e-05,
|
478 |
+
"loss": 1.26,
|
479 |
+
"step": 16000
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 20.41,
|
483 |
+
"eval_loss": 0.27389487624168396,
|
484 |
+
"eval_runtime": 235.3865,
|
485 |
+
"eval_samples_per_second": 35.427,
|
486 |
+
"eval_steps_per_second": 1.109,
|
487 |
+
"eval_wer": 0.3674404748264088,
|
488 |
+
"step": 16000
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 21.05,
|
492 |
+
"learning_rate": 1.5894736842105266e-05,
|
493 |
+
"loss": 1.2543,
|
494 |
+
"step": 16500
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 21.05,
|
498 |
+
"eval_loss": 0.2739649713039398,
|
499 |
+
"eval_runtime": 234.6769,
|
500 |
+
"eval_samples_per_second": 35.534,
|
501 |
+
"eval_steps_per_second": 1.112,
|
502 |
+
"eval_wer": 0.36909045076425967,
|
503 |
+
"step": 16500
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 21.68,
|
507 |
+
"learning_rate": 1.5766367137355586e-05,
|
508 |
+
"loss": 1.2532,
|
509 |
+
"step": 17000
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 21.68,
|
513 |
+
"eval_loss": 0.2709480822086334,
|
514 |
+
"eval_runtime": 236.2528,
|
515 |
+
"eval_samples_per_second": 35.297,
|
516 |
+
"eval_steps_per_second": 1.105,
|
517 |
+
"eval_wer": 0.37562160551825285,
|
518 |
+
"step": 17000
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 22.32,
|
522 |
+
"learning_rate": 1.5637997432605906e-05,
|
523 |
+
"loss": 1.2409,
|
524 |
+
"step": 17500
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 22.32,
|
528 |
+
"eval_loss": 0.266906201839447,
|
529 |
+
"eval_runtime": 233.9016,
|
530 |
+
"eval_samples_per_second": 35.652,
|
531 |
+
"eval_steps_per_second": 1.116,
|
532 |
+
"eval_wer": 0.35932809313197517,
|
533 |
+
"step": 17500
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 22.96,
|
537 |
+
"learning_rate": 1.5509627727856227e-05,
|
538 |
+
"loss": 1.2404,
|
539 |
+
"step": 18000
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 22.96,
|
543 |
+
"eval_loss": 0.2673003077507019,
|
544 |
+
"eval_runtime": 235.0288,
|
545 |
+
"eval_samples_per_second": 35.481,
|
546 |
+
"eval_steps_per_second": 1.111,
|
547 |
+
"eval_wer": 0.35763228452918394,
|
548 |
+
"step": 18000
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 23.6,
|
552 |
+
"learning_rate": 1.538125802310655e-05,
|
553 |
+
"loss": 1.2347,
|
554 |
+
"step": 18500
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 23.6,
|
558 |
+
"eval_loss": 0.2678009271621704,
|
559 |
+
"eval_runtime": 234.3074,
|
560 |
+
"eval_samples_per_second": 35.59,
|
561 |
+
"eval_steps_per_second": 1.114,
|
562 |
+
"eval_wer": 0.36427802094552786,
|
563 |
+
"step": 18500
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 24.23,
|
567 |
+
"learning_rate": 1.5252888318356869e-05,
|
568 |
+
"loss": 1.2351,
|
569 |
+
"step": 19000
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 24.23,
|
573 |
+
"eval_loss": 0.2715151607990265,
|
574 |
+
"eval_runtime": 239.5626,
|
575 |
+
"eval_samples_per_second": 34.809,
|
576 |
+
"eval_steps_per_second": 1.089,
|
577 |
+
"eval_wer": 0.36498842725210257,
|
578 |
+
"step": 19000
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 24.87,
|
582 |
+
"learning_rate": 1.512451861360719e-05,
|
583 |
+
"loss": 1.2409,
|
584 |
+
"step": 19500
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 24.87,
|
588 |
+
"eval_loss": 0.2636788487434387,
|
589 |
+
"eval_runtime": 234.5142,
|
590 |
+
"eval_samples_per_second": 35.559,
|
591 |
+
"eval_steps_per_second": 1.113,
|
592 |
+
"eval_wer": 0.3570822925499003,
|
593 |
+
"step": 19500
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 25.51,
|
597 |
+
"learning_rate": 1.4996148908857511e-05,
|
598 |
+
"loss": 1.2152,
|
599 |
+
"step": 20000
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 25.51,
|
603 |
+
"eval_loss": 0.2785368859767914,
|
604 |
+
"eval_runtime": 234.4552,
|
605 |
+
"eval_samples_per_second": 35.568,
|
606 |
+
"eval_steps_per_second": 1.113,
|
607 |
+
"eval_wer": 0.36088640373994546,
|
608 |
+
"step": 20000
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 26.15,
|
612 |
+
"learning_rate": 1.4868035943517332e-05,
|
613 |
+
"loss": 1.2046,
|
614 |
+
"step": 20500
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 26.15,
|
618 |
+
"eval_loss": 0.2610293924808502,
|
619 |
+
"eval_runtime": 233.6859,
|
620 |
+
"eval_samples_per_second": 35.685,
|
621 |
+
"eval_steps_per_second": 1.117,
|
622 |
+
"eval_wer": 0.3508032174530788,
|
623 |
+
"step": 20500
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 26.79,
|
627 |
+
"learning_rate": 1.4739666238767652e-05,
|
628 |
+
"loss": 1.2082,
|
629 |
+
"step": 21000
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 26.79,
|
633 |
+
"eval_loss": 0.26189225912094116,
|
634 |
+
"eval_runtime": 234.8807,
|
635 |
+
"eval_samples_per_second": 35.503,
|
636 |
+
"eval_steps_per_second": 1.111,
|
637 |
+
"eval_wer": 0.3460595366317575,
|
638 |
+
"step": 21000
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 27.42,
|
642 |
+
"learning_rate": 1.4611296534017974e-05,
|
643 |
+
"loss": 1.2109,
|
644 |
+
"step": 21500
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 27.42,
|
648 |
+
"eval_loss": 0.2596839666366577,
|
649 |
+
"eval_runtime": 235.2549,
|
650 |
+
"eval_samples_per_second": 35.447,
|
651 |
+
"eval_steps_per_second": 1.109,
|
652 |
+
"eval_wer": 0.3501615601439146,
|
653 |
+
"step": 21500
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 28.06,
|
657 |
+
"learning_rate": 1.4482926829268294e-05,
|
658 |
+
"loss": 1.2014,
|
659 |
+
"step": 22000
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 28.06,
|
663 |
+
"eval_loss": 0.2608143091201782,
|
664 |
+
"eval_runtime": 234.8521,
|
665 |
+
"eval_samples_per_second": 35.507,
|
666 |
+
"eval_steps_per_second": 1.111,
|
667 |
+
"eval_wer": 0.3468386919357426,
|
668 |
+
"step": 22000
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 28.7,
|
672 |
+
"learning_rate": 1.4354813863928115e-05,
|
673 |
+
"loss": 1.1948,
|
674 |
+
"step": 22500
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 28.7,
|
678 |
+
"eval_loss": 0.2573413550853729,
|
679 |
+
"eval_runtime": 233.0487,
|
680 |
+
"eval_samples_per_second": 35.782,
|
681 |
+
"eval_steps_per_second": 1.12,
|
682 |
+
"eval_wer": 0.3456699589797649,
|
683 |
+
"step": 22500
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 29.34,
|
687 |
+
"learning_rate": 1.4226444159178435e-05,
|
688 |
+
"loss": 1.205,
|
689 |
+
"step": 23000
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 29.34,
|
693 |
+
"eval_loss": 0.2619355320930481,
|
694 |
+
"eval_runtime": 236.4546,
|
695 |
+
"eval_samples_per_second": 35.267,
|
696 |
+
"eval_steps_per_second": 1.104,
|
697 |
+
"eval_wer": 0.3463803652863396,
|
698 |
+
"step": 23000
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 29.97,
|
702 |
+
"learning_rate": 1.4098074454428757e-05,
|
703 |
+
"loss": 1.2019,
|
704 |
+
"step": 23500
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 29.97,
|
708 |
+
"eval_loss": 0.2558630406856537,
|
709 |
+
"eval_runtime": 235.5761,
|
710 |
+
"eval_samples_per_second": 35.398,
|
711 |
+
"eval_steps_per_second": 1.108,
|
712 |
+
"eval_wer": 0.34738868391502625,
|
713 |
+
"step": 23500
|
714 |
+
},
|
715 |
+
{
|
716 |
+
"epoch": 30.61,
|
717 |
+
"learning_rate": 1.3969704749679077e-05,
|
718 |
+
"loss": 1.1917,
|
719 |
+
"step": 24000
|
720 |
+
},
|
721 |
+
{
|
722 |
+
"epoch": 30.61,
|
723 |
+
"eval_loss": 0.2601245045661926,
|
724 |
+
"eval_runtime": 235.3838,
|
725 |
+
"eval_samples_per_second": 35.427,
|
726 |
+
"eval_steps_per_second": 1.109,
|
727 |
+
"eval_wer": 0.34619703462657836,
|
728 |
+
"step": 24000
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 31.25,
|
732 |
+
"learning_rate": 1.3841591784338897e-05,
|
733 |
+
"loss": 1.1939,
|
734 |
+
"step": 24500
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 31.25,
|
738 |
+
"eval_loss": 0.2575133740901947,
|
739 |
+
"eval_runtime": 235.7288,
|
740 |
+
"eval_samples_per_second": 35.375,
|
741 |
+
"eval_steps_per_second": 1.107,
|
742 |
+
"eval_wer": 0.33874922657377915,
|
743 |
+
"step": 24500
|
744 |
+
},
|
745 |
+
{
|
746 |
+
"epoch": 31.89,
|
747 |
+
"learning_rate": 1.3713478818998718e-05,
|
748 |
+
"loss": 1.1882,
|
749 |
+
"step": 25000
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 31.89,
|
753 |
+
"eval_loss": 0.2534877359867096,
|
754 |
+
"eval_runtime": 235.7724,
|
755 |
+
"eval_samples_per_second": 35.369,
|
756 |
+
"eval_steps_per_second": 1.107,
|
757 |
+
"eval_wer": 0.33675550564887596,
|
758 |
+
"step": 25000
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 32.53,
|
762 |
+
"learning_rate": 1.3585109114249038e-05,
|
763 |
+
"loss": 1.191,
|
764 |
+
"step": 25500
|
765 |
+
},
|
766 |
+
{
|
767 |
+
"epoch": 32.53,
|
768 |
+
"eval_loss": 0.24889491498470306,
|
769 |
+
"eval_runtime": 236.163,
|
770 |
+
"eval_samples_per_second": 35.31,
|
771 |
+
"eval_steps_per_second": 1.105,
|
772 |
+
"eval_wer": 0.336457593326764,
|
773 |
+
"step": 25500
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 33.16,
|
777 |
+
"learning_rate": 1.345673940949936e-05,
|
778 |
+
"loss": 1.1767,
|
779 |
+
"step": 26000
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 33.16,
|
783 |
+
"eval_loss": 0.2501368224620819,
|
784 |
+
"eval_runtime": 236.8496,
|
785 |
+
"eval_samples_per_second": 35.208,
|
786 |
+
"eval_steps_per_second": 1.102,
|
787 |
+
"eval_wer": 0.3347159520590325,
|
788 |
+
"step": 26000
|
789 |
+
},
|
790 |
+
{
|
791 |
+
"epoch": 33.8,
|
792 |
+
"learning_rate": 1.332836970474968e-05,
|
793 |
+
"loss": 1.167,
|
794 |
+
"step": 26500
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 33.8,
|
798 |
+
"eval_loss": 0.25037434697151184,
|
799 |
+
"eval_runtime": 233.4394,
|
800 |
+
"eval_samples_per_second": 35.722,
|
801 |
+
"eval_steps_per_second": 1.118,
|
802 |
+
"eval_wer": 0.33469303572656234,
|
803 |
+
"step": 26500
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 34.44,
|
807 |
+
"learning_rate": 1.3200000000000002e-05,
|
808 |
+
"loss": 1.1678,
|
809 |
+
"step": 27000
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 34.44,
|
813 |
+
"eval_loss": 0.24800854921340942,
|
814 |
+
"eval_runtime": 233.3087,
|
815 |
+
"eval_samples_per_second": 35.742,
|
816 |
+
"eval_steps_per_second": 1.119,
|
817 |
+
"eval_wer": 0.33783257327497307,
|
818 |
+
"step": 27000
|
819 |
+
},
|
820 |
+
{
|
821 |
+
"epoch": 35.08,
|
822 |
+
"learning_rate": 1.3071630295250321e-05,
|
823 |
+
"loss": 1.1803,
|
824 |
+
"step": 27500
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 35.08,
|
828 |
+
"eval_loss": 0.24871966242790222,
|
829 |
+
"eval_runtime": 235.3613,
|
830 |
+
"eval_samples_per_second": 35.431,
|
831 |
+
"eval_steps_per_second": 1.109,
|
832 |
+
"eval_wer": 0.33448678873433096,
|
833 |
+
"step": 27500
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 35.71,
|
837 |
+
"learning_rate": 1.2943260590500643e-05,
|
838 |
+
"loss": 1.167,
|
839 |
+
"step": 28000
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 35.71,
|
843 |
+
"eval_loss": 0.24418053030967712,
|
844 |
+
"eval_runtime": 232.8217,
|
845 |
+
"eval_samples_per_second": 35.817,
|
846 |
+
"eval_steps_per_second": 1.121,
|
847 |
+
"eval_wer": 0.3318514105002635,
|
848 |
+
"step": 28000
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 36.35,
|
852 |
+
"learning_rate": 1.2814890885750963e-05,
|
853 |
+
"loss": 1.1661,
|
854 |
+
"step": 28500
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 36.35,
|
858 |
+
"eval_loss": 0.24953289330005646,
|
859 |
+
"eval_runtime": 237.0041,
|
860 |
+
"eval_samples_per_second": 35.185,
|
861 |
+
"eval_steps_per_second": 1.101,
|
862 |
+
"eval_wer": 0.3333868047757637,
|
863 |
+
"step": 28500
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 36.99,
|
867 |
+
"learning_rate": 1.2686521181001285e-05,
|
868 |
+
"loss": 1.164,
|
869 |
+
"step": 29000
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 36.99,
|
873 |
+
"eval_loss": 0.24715770781040192,
|
874 |
+
"eval_runtime": 233.2505,
|
875 |
+
"eval_samples_per_second": 35.751,
|
876 |
+
"eval_steps_per_second": 1.119,
|
877 |
+
"eval_wer": 0.32923894859866626,
|
878 |
+
"step": 29000
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 37.63,
|
882 |
+
"learning_rate": 1.2558151476251605e-05,
|
883 |
+
"loss": 1.1578,
|
884 |
+
"step": 29500
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 37.63,
|
888 |
+
"eval_loss": 0.24420544505119324,
|
889 |
+
"eval_runtime": 239.1685,
|
890 |
+
"eval_samples_per_second": 34.867,
|
891 |
+
"eval_steps_per_second": 1.091,
|
892 |
+
"eval_wer": 0.3242431881201732,
|
893 |
+
"step": 29500
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 38.27,
|
897 |
+
"learning_rate": 1.2430038510911426e-05,
|
898 |
+
"loss": 1.1584,
|
899 |
+
"step": 30000
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 38.27,
|
903 |
+
"eval_loss": 0.243063285946846,
|
904 |
+
"eval_runtime": 235.3653,
|
905 |
+
"eval_samples_per_second": 35.43,
|
906 |
+
"eval_steps_per_second": 1.109,
|
907 |
+
"eval_wer": 0.3314389165158008,
|
908 |
+
"step": 30000
|
909 |
+
},
|
910 |
+
{
|
911 |
+
"epoch": 38.9,
|
912 |
+
"learning_rate": 1.2301668806161746e-05,
|
913 |
+
"loss": 1.1526,
|
914 |
+
"step": 30500
|
915 |
+
},
|
916 |
+
{
|
917 |
+
"epoch": 38.9,
|
918 |
+
"eval_loss": 0.2440960705280304,
|
919 |
+
"eval_runtime": 238.3272,
|
920 |
+
"eval_samples_per_second": 34.99,
|
921 |
+
"eval_steps_per_second": 1.095,
|
922 |
+
"eval_wer": 0.3347159520590325,
|
923 |
+
"step": 30500
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 39.54,
|
927 |
+
"learning_rate": 1.2173299101412068e-05,
|
928 |
+
"loss": 1.1542,
|
929 |
+
"step": 31000
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 39.54,
|
933 |
+
"eval_loss": 0.2436649352312088,
|
934 |
+
"eval_runtime": 236.1705,
|
935 |
+
"eval_samples_per_second": 35.309,
|
936 |
+
"eval_steps_per_second": 1.105,
|
937 |
+
"eval_wer": 0.3330430597887114,
|
938 |
+
"step": 31000
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 40.18,
|
942 |
+
"learning_rate": 1.2044929396662388e-05,
|
943 |
+
"loss": 1.1508,
|
944 |
+
"step": 31500
|
945 |
+
},
|
946 |
+
{
|
947 |
+
"epoch": 40.18,
|
948 |
+
"eval_loss": 0.24334189295768738,
|
949 |
+
"eval_runtime": 234.2613,
|
950 |
+
"eval_samples_per_second": 35.597,
|
951 |
+
"eval_steps_per_second": 1.114,
|
952 |
+
"eval_wer": 0.3294222792584275,
|
953 |
+
"step": 31500
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 40.82,
|
957 |
+
"learning_rate": 1.191655969191271e-05,
|
958 |
+
"loss": 1.1406,
|
959 |
+
"step": 32000
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 40.82,
|
963 |
+
"eval_loss": 0.24340486526489258,
|
964 |
+
"eval_runtime": 233.0368,
|
965 |
+
"eval_samples_per_second": 35.784,
|
966 |
+
"eval_steps_per_second": 1.12,
|
967 |
+
"eval_wer": 0.3270618970140019,
|
968 |
+
"step": 32000
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 41.45,
|
972 |
+
"learning_rate": 1.178870346598203e-05,
|
973 |
+
"loss": 1.1514,
|
974 |
+
"step": 32500
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 41.45,
|
978 |
+
"eval_loss": 0.24264076352119446,
|
979 |
+
"eval_runtime": 238.874,
|
980 |
+
"eval_samples_per_second": 34.91,
|
981 |
+
"eval_steps_per_second": 1.093,
|
982 |
+
"eval_wer": 0.32552650273850176,
|
983 |
+
"step": 32500
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 42.09,
|
987 |
+
"learning_rate": 1.166033376123235e-05,
|
988 |
+
"loss": 1.1418,
|
989 |
+
"step": 33000
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 42.09,
|
993 |
+
"eval_loss": 0.2431638389825821,
|
994 |
+
"eval_runtime": 238.255,
|
995 |
+
"eval_samples_per_second": 35.0,
|
996 |
+
"eval_steps_per_second": 1.095,
|
997 |
+
"eval_wer": 0.3232807021564269,
|
998 |
+
"step": 33000
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 42.73,
|
1002 |
+
"learning_rate": 1.1531964056482671e-05,
|
1003 |
+
"loss": 1.1365,
|
1004 |
+
"step": 33500
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 42.73,
|
1008 |
+
"eval_loss": 0.2436181753873825,
|
1009 |
+
"eval_runtime": 234.1309,
|
1010 |
+
"eval_samples_per_second": 35.617,
|
1011 |
+
"eval_steps_per_second": 1.115,
|
1012 |
+
"eval_wer": 0.32399110846300155,
|
1013 |
+
"step": 33500
|
1014 |
+
},
|
1015 |
+
{
|
1016 |
+
"epoch": 43.37,
|
1017 |
+
"learning_rate": 1.1403594351732992e-05,
|
1018 |
+
"loss": 1.1348,
|
1019 |
+
"step": 34000
|
1020 |
+
},
|
1021 |
+
{
|
1022 |
+
"epoch": 43.37,
|
1023 |
+
"eval_loss": 0.2482648491859436,
|
1024 |
+
"eval_runtime": 238.6079,
|
1025 |
+
"eval_samples_per_second": 34.949,
|
1026 |
+
"eval_steps_per_second": 1.094,
|
1027 |
+
"eval_wer": 0.3257327497307331,
|
1028 |
+
"step": 34000
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 44.01,
|
1032 |
+
"learning_rate": 1.1275224646983314e-05,
|
1033 |
+
"loss": 1.1301,
|
1034 |
+
"step": 34500
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 44.01,
|
1038 |
+
"eval_loss": 0.24202857911586761,
|
1039 |
+
"eval_runtime": 232.8764,
|
1040 |
+
"eval_samples_per_second": 35.809,
|
1041 |
+
"eval_steps_per_second": 1.121,
|
1042 |
+
"eval_wer": 0.3271077296789422,
|
1043 |
+
"step": 34500
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 44.64,
|
1047 |
+
"learning_rate": 1.1146854942233634e-05,
|
1048 |
+
"loss": 1.1268,
|
1049 |
+
"step": 35000
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 44.64,
|
1053 |
+
"eval_loss": 0.24715635180473328,
|
1054 |
+
"eval_runtime": 233.1436,
|
1055 |
+
"eval_samples_per_second": 35.768,
|
1056 |
+
"eval_steps_per_second": 1.119,
|
1057 |
+
"eval_wer": 0.3224786305199716,
|
1058 |
+
"step": 35000
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 45.28,
|
1062 |
+
"learning_rate": 1.1018485237483956e-05,
|
1063 |
+
"loss": 1.1224,
|
1064 |
+
"step": 35500
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 45.28,
|
1068 |
+
"eval_loss": 0.2381853461265564,
|
1069 |
+
"eval_runtime": 238.9288,
|
1070 |
+
"eval_samples_per_second": 34.902,
|
1071 |
+
"eval_steps_per_second": 1.092,
|
1072 |
+
"eval_wer": 0.3204849095950684,
|
1073 |
+
"step": 35500
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 45.92,
|
1077 |
+
"learning_rate": 1.0890115532734276e-05,
|
1078 |
+
"loss": 1.1224,
|
1079 |
+
"step": 36000
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 45.92,
|
1083 |
+
"eval_loss": 0.23876558244228363,
|
1084 |
+
"eval_runtime": 243.8522,
|
1085 |
+
"eval_samples_per_second": 34.197,
|
1086 |
+
"eval_steps_per_second": 1.07,
|
1087 |
+
"eval_wer": 0.3184224396727548,
|
1088 |
+
"step": 36000
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 46.56,
|
1092 |
+
"learning_rate": 1.0762002567394097e-05,
|
1093 |
+
"loss": 1.1198,
|
1094 |
+
"step": 36500
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 46.56,
|
1098 |
+
"eval_loss": 0.23819147050380707,
|
1099 |
+
"eval_runtime": 238.5089,
|
1100 |
+
"eval_samples_per_second": 34.963,
|
1101 |
+
"eval_steps_per_second": 1.094,
|
1102 |
+
"eval_wer": 0.32016408094048626,
|
1103 |
+
"step": 36500
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 47.19,
|
1107 |
+
"learning_rate": 1.0633632862644417e-05,
|
1108 |
+
"loss": 1.1274,
|
1109 |
+
"step": 37000
|
1110 |
+
},
|
1111 |
+
{
|
1112 |
+
"epoch": 47.19,
|
1113 |
+
"eval_loss": 0.24040183424949646,
|
1114 |
+
"eval_runtime": 237.5269,
|
1115 |
+
"eval_samples_per_second": 35.108,
|
1116 |
+
"eval_steps_per_second": 1.099,
|
1117 |
+
"eval_wer": 0.31720787405183676,
|
1118 |
+
"step": 37000
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 47.83,
|
1122 |
+
"learning_rate": 1.0505263157894739e-05,
|
1123 |
+
"loss": 1.1147,
|
1124 |
+
"step": 37500
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 47.83,
|
1128 |
+
"eval_loss": 0.23944656550884247,
|
1129 |
+
"eval_runtime": 235.5135,
|
1130 |
+
"eval_samples_per_second": 35.408,
|
1131 |
+
"eval_steps_per_second": 1.108,
|
1132 |
+
"eval_wer": 0.3164287187478516,
|
1133 |
+
"step": 37500
|
1134 |
+
},
|
1135 |
+
{
|
1136 |
+
"epoch": 48.47,
|
1137 |
+
"learning_rate": 1.0377150192554558e-05,
|
1138 |
+
"loss": 1.121,
|
1139 |
+
"step": 38000
|
1140 |
+
},
|
1141 |
+
{
|
1142 |
+
"epoch": 48.47,
|
1143 |
+
"eval_loss": 0.2405969202518463,
|
1144 |
+
"eval_runtime": 233.9426,
|
1145 |
+
"eval_samples_per_second": 35.645,
|
1146 |
+
"eval_steps_per_second": 1.116,
|
1147 |
+
"eval_wer": 0.3202099136054266,
|
1148 |
+
"step": 38000
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 49.11,
|
1152 |
+
"learning_rate": 1.024878048780488e-05,
|
1153 |
+
"loss": 1.1109,
|
1154 |
+
"step": 38500
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 49.11,
|
1158 |
+
"eval_loss": 0.23839789628982544,
|
1159 |
+
"eval_runtime": 240.6944,
|
1160 |
+
"eval_samples_per_second": 34.646,
|
1161 |
+
"eval_steps_per_second": 1.084,
|
1162 |
+
"eval_wer": 0.31537456745422465,
|
1163 |
+
"step": 38500
|
1164 |
+
},
|
1165 |
+
{
|
1166 |
+
"epoch": 49.74,
|
1167 |
+
"learning_rate": 1.0120410783055198e-05,
|
1168 |
+
"loss": 1.1164,
|
1169 |
+
"step": 39000
|
1170 |
+
},
|
1171 |
+
{
|
1172 |
+
"epoch": 49.74,
|
1173 |
+
"eval_loss": 0.2374895066022873,
|
1174 |
+
"eval_runtime": 237.3857,
|
1175 |
+
"eval_samples_per_second": 35.128,
|
1176 |
+
"eval_steps_per_second": 1.099,
|
1177 |
+
"eval_wer": 0.31686412906478445,
|
1178 |
+
"step": 39000
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 50.38,
|
1182 |
+
"learning_rate": 9.992041078305522e-06,
|
1183 |
+
"loss": 1.1105,
|
1184 |
+
"step": 39500
|
1185 |
+
},
|
1186 |
+
{
|
1187 |
+
"epoch": 50.38,
|
1188 |
+
"eval_loss": 0.23865096271038055,
|
1189 |
+
"eval_runtime": 232.9433,
|
1190 |
+
"eval_samples_per_second": 35.798,
|
1191 |
+
"eval_steps_per_second": 1.12,
|
1192 |
+
"eval_wer": 0.31729953938171734,
|
1193 |
+
"step": 39500
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 51.02,
|
1197 |
+
"learning_rate": 9.86392811296534e-06,
|
1198 |
+
"loss": 1.1054,
|
1199 |
+
"step": 40000
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 51.02,
|
1203 |
+
"eval_loss": 0.23617714643478394,
|
1204 |
+
"eval_runtime": 233.5533,
|
1205 |
+
"eval_samples_per_second": 35.705,
|
1206 |
+
"eval_steps_per_second": 1.118,
|
1207 |
+
"eval_wer": 0.31200586658111235,
|
1208 |
+
"step": 40000
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 51.66,
|
1212 |
+
"learning_rate": 9.735558408215663e-06,
|
1213 |
+
"loss": 1.0893,
|
1214 |
+
"step": 40500
|
1215 |
+
},
|
1216 |
+
{
|
1217 |
+
"epoch": 51.66,
|
1218 |
+
"eval_loss": 0.23990453779697418,
|
1219 |
+
"eval_runtime": 234.1326,
|
1220 |
+
"eval_samples_per_second": 35.617,
|
1221 |
+
"eval_steps_per_second": 1.115,
|
1222 |
+
"eval_wer": 0.31299126887732887,
|
1223 |
+
"step": 40500
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 52.3,
|
1227 |
+
"learning_rate": 9.607188703465983e-06,
|
1228 |
+
"loss": 1.0913,
|
1229 |
+
"step": 41000
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 52.3,
|
1233 |
+
"eval_loss": 0.2356649786233902,
|
1234 |
+
"eval_runtime": 238.2462,
|
1235 |
+
"eval_samples_per_second": 35.002,
|
1236 |
+
"eval_steps_per_second": 1.096,
|
1237 |
+
"eval_wer": 0.3087517473703508,
|
1238 |
+
"step": 41000
|
1239 |
+
},
|
1240 |
+
{
|
1241 |
+
"epoch": 52.93,
|
1242 |
+
"learning_rate": 9.478818998716303e-06,
|
1243 |
+
"loss": 1.1017,
|
1244 |
+
"step": 41500
|
1245 |
+
},
|
1246 |
+
{
|
1247 |
+
"epoch": 52.93,
|
1248 |
+
"eval_loss": 0.23450981080532074,
|
1249 |
+
"eval_runtime": 233.807,
|
1250 |
+
"eval_samples_per_second": 35.666,
|
1251 |
+
"eval_steps_per_second": 1.116,
|
1252 |
+
"eval_wer": 0.3083621697183583,
|
1253 |
+
"step": 41500
|
1254 |
+
},
|
1255 |
+
{
|
1256 |
+
"epoch": 53.57,
|
1257 |
+
"learning_rate": 9.350449293966625e-06,
|
1258 |
+
"loss": 1.0937,
|
1259 |
+
"step": 42000
|
1260 |
+
},
|
1261 |
+
{
|
1262 |
+
"epoch": 53.57,
|
1263 |
+
"eval_loss": 0.23296880722045898,
|
1264 |
+
"eval_runtime": 238.7521,
|
1265 |
+
"eval_samples_per_second": 34.927,
|
1266 |
+
"eval_steps_per_second": 1.093,
|
1267 |
+
"eval_wer": 0.31395375484107524,
|
1268 |
+
"step": 42000
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 54.21,
|
1272 |
+
"learning_rate": 9.222079589216945e-06,
|
1273 |
+
"loss": 1.0945,
|
1274 |
+
"step": 42500
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 54.21,
|
1278 |
+
"eval_loss": 0.23989614844322205,
|
1279 |
+
"eval_runtime": 233.2848,
|
1280 |
+
"eval_samples_per_second": 35.746,
|
1281 |
+
"eval_steps_per_second": 1.119,
|
1282 |
+
"eval_wer": 0.3106767192978436,
|
1283 |
+
"step": 42500
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 54.85,
|
1287 |
+
"learning_rate": 9.093966623876766e-06,
|
1288 |
+
"loss": 1.0933,
|
1289 |
+
"step": 43000
|
1290 |
+
},
|
1291 |
+
{
|
1292 |
+
"epoch": 54.85,
|
1293 |
+
"eval_loss": 0.23831374943256378,
|
1294 |
+
"eval_runtime": 234.0959,
|
1295 |
+
"eval_samples_per_second": 35.622,
|
1296 |
+
"eval_steps_per_second": 1.115,
|
1297 |
+
"eval_wer": 0.31338084652932147,
|
1298 |
+
"step": 43000
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 55.48,
|
1302 |
+
"learning_rate": 8.965596919127086e-06,
|
1303 |
+
"loss": 1.0912,
|
1304 |
+
"step": 43500
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 55.48,
|
1308 |
+
"eval_loss": 0.2371690571308136,
|
1309 |
+
"eval_runtime": 234.0624,
|
1310 |
+
"eval_samples_per_second": 35.627,
|
1311 |
+
"eval_steps_per_second": 1.115,
|
1312 |
+
"eval_wer": 0.30769759607672387,
|
1313 |
+
"step": 43500
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 56.12,
|
1317 |
+
"learning_rate": 8.837227214377408e-06,
|
1318 |
+
"loss": 1.0898,
|
1319 |
+
"step": 44000
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 56.12,
|
1323 |
+
"eval_loss": 0.23387998342514038,
|
1324 |
+
"eval_runtime": 233.2957,
|
1325 |
+
"eval_samples_per_second": 35.744,
|
1326 |
+
"eval_steps_per_second": 1.119,
|
1327 |
+
"eval_wer": 0.30829342072094784,
|
1328 |
+
"step": 44000
|
1329 |
+
},
|
1330 |
+
{
|
1331 |
+
"epoch": 56.76,
|
1332 |
+
"learning_rate": 8.708857509627728e-06,
|
1333 |
+
"loss": 1.0903,
|
1334 |
+
"step": 44500
|
1335 |
+
},
|
1336 |
+
{
|
1337 |
+
"epoch": 56.76,
|
1338 |
+
"eval_loss": 0.23673543334007263,
|
1339 |
+
"eval_runtime": 236.1402,
|
1340 |
+
"eval_samples_per_second": 35.314,
|
1341 |
+
"eval_steps_per_second": 1.105,
|
1342 |
+
"eval_wer": 0.3065288631207462,
|
1343 |
+
"step": 44500
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 57.4,
|
1347 |
+
"learning_rate": 8.58048780487805e-06,
|
1348 |
+
"loss": 1.0947,
|
1349 |
+
"step": 45000
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 57.4,
|
1353 |
+
"eval_loss": 0.23515641689300537,
|
1354 |
+
"eval_runtime": 235.6322,
|
1355 |
+
"eval_samples_per_second": 35.39,
|
1356 |
+
"eval_steps_per_second": 1.108,
|
1357 |
+
"eval_wer": 0.31035589064326147,
|
1358 |
+
"step": 45000
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 58.04,
|
1362 |
+
"learning_rate": 8.45211810012837e-06,
|
1363 |
+
"loss": 1.0751,
|
1364 |
+
"step": 45500
|
1365 |
+
},
|
1366 |
+
{
|
1367 |
+
"epoch": 58.04,
|
1368 |
+
"eval_loss": 0.23344124853610992,
|
1369 |
+
"eval_runtime": 236.1866,
|
1370 |
+
"eval_samples_per_second": 35.307,
|
1371 |
+
"eval_steps_per_second": 1.105,
|
1372 |
+
"eval_wer": 0.3084309187157687,
|
1373 |
+
"step": 45500
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 58.67,
|
1377 |
+
"learning_rate": 8.32374839537869e-06,
|
1378 |
+
"loss": 1.09,
|
1379 |
+
"step": 46000
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 58.67,
|
1383 |
+
"eval_loss": 0.2327876091003418,
|
1384 |
+
"eval_runtime": 233.0944,
|
1385 |
+
"eval_samples_per_second": 35.775,
|
1386 |
+
"eval_steps_per_second": 1.12,
|
1387 |
+
"eval_wer": 0.3100350619886793,
|
1388 |
+
"step": 46000
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 59.31,
|
1392 |
+
"learning_rate": 8.195378690629013e-06,
|
1393 |
+
"loss": 1.0876,
|
1394 |
+
"step": 46500
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 59.31,
|
1398 |
+
"eval_loss": 0.22762973606586456,
|
1399 |
+
"eval_runtime": 232.6706,
|
1400 |
+
"eval_samples_per_second": 35.84,
|
1401 |
+
"eval_steps_per_second": 1.122,
|
1402 |
+
"eval_wer": 0.3049705525127758,
|
1403 |
+
"step": 46500
|
1404 |
+
},
|
1405 |
+
{
|
1406 |
+
"epoch": 59.95,
|
1407 |
+
"learning_rate": 8.067008985879333e-06,
|
1408 |
+
"loss": 1.076,
|
1409 |
+
"step": 47000
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 59.95,
|
1413 |
+
"eval_loss": 0.23087561130523682,
|
1414 |
+
"eval_runtime": 233.5675,
|
1415 |
+
"eval_samples_per_second": 35.703,
|
1416 |
+
"eval_steps_per_second": 1.117,
|
1417 |
+
"eval_wer": 0.304695556523134,
|
1418 |
+
"step": 47000
|
1419 |
+
},
|
1420 |
+
{
|
1421 |
+
"epoch": 60.59,
|
1422 |
+
"learning_rate": 7.938896020539153e-06,
|
1423 |
+
"loss": 1.086,
|
1424 |
+
"step": 47500
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 60.59,
|
1428 |
+
"eval_loss": 0.229255810379982,
|
1429 |
+
"eval_runtime": 234.5783,
|
1430 |
+
"eval_samples_per_second": 35.549,
|
1431 |
+
"eval_steps_per_second": 1.113,
|
1432 |
+
"eval_wer": 0.304695556523134,
|
1433 |
+
"step": 47500
|
1434 |
+
},
|
1435 |
+
{
|
1436 |
+
"epoch": 61.22,
|
1437 |
+
"learning_rate": 7.810526315789474e-06,
|
1438 |
+
"loss": 1.082,
|
1439 |
+
"step": 48000
|
1440 |
+
},
|
1441 |
+
{
|
1442 |
+
"epoch": 61.22,
|
1443 |
+
"eval_loss": 0.232817605137825,
|
1444 |
+
"eval_runtime": 239.1878,
|
1445 |
+
"eval_samples_per_second": 34.864,
|
1446 |
+
"eval_steps_per_second": 1.091,
|
1447 |
+
"eval_wer": 0.3027476682631712,
|
1448 |
+
"step": 48000
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 61.86,
|
1452 |
+
"learning_rate": 7.682413350449294e-06,
|
1453 |
+
"loss": 1.0714,
|
1454 |
+
"step": 48500
|
1455 |
+
},
|
1456 |
+
{
|
1457 |
+
"epoch": 61.86,
|
1458 |
+
"eval_loss": 0.22896084189414978,
|
1459 |
+
"eval_runtime": 234.6636,
|
1460 |
+
"eval_samples_per_second": 35.536,
|
1461 |
+
"eval_steps_per_second": 1.112,
|
1462 |
+
"eval_wer": 0.3019914292916562,
|
1463 |
+
"step": 48500
|
1464 |
+
},
|
1465 |
+
{
|
1466 |
+
"epoch": 62.5,
|
1467 |
+
"learning_rate": 7.554043645699616e-06,
|
1468 |
+
"loss": 1.0746,
|
1469 |
+
"step": 49000
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 62.5,
|
1473 |
+
"eval_loss": 0.2313106656074524,
|
1474 |
+
"eval_runtime": 234.936,
|
1475 |
+
"eval_samples_per_second": 35.495,
|
1476 |
+
"eval_steps_per_second": 1.111,
|
1477 |
+
"eval_wer": 0.30586428947911176,
|
1478 |
+
"step": 49000
|
1479 |
+
},
|
1480 |
+
{
|
1481 |
+
"epoch": 63.14,
|
1482 |
+
"learning_rate": 7.425673940949937e-06,
|
1483 |
+
"loss": 1.076,
|
1484 |
+
"step": 49500
|
1485 |
+
},
|
1486 |
+
{
|
1487 |
+
"epoch": 63.14,
|
1488 |
+
"eval_loss": 0.23419487476348877,
|
1489 |
+
"eval_runtime": 235.1421,
|
1490 |
+
"eval_samples_per_second": 35.464,
|
1491 |
+
"eval_steps_per_second": 1.11,
|
1492 |
+
"eval_wer": 0.3050393015101863,
|
1493 |
+
"step": 49500
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 63.78,
|
1497 |
+
"learning_rate": 7.297304236200257e-06,
|
1498 |
+
"loss": 1.0648,
|
1499 |
+
"step": 50000
|
1500 |
+
},
|
1501 |
+
{
|
1502 |
+
"epoch": 63.78,
|
1503 |
+
"eval_loss": 0.22855186462402344,
|
1504 |
+
"eval_runtime": 233.6212,
|
1505 |
+
"eval_samples_per_second": 35.695,
|
1506 |
+
"eval_steps_per_second": 1.117,
|
1507 |
+
"eval_wer": 0.30247267227352936,
|
1508 |
+
"step": 50000
|
1509 |
+
},
|
1510 |
+
{
|
1511 |
+
"epoch": 64.41,
|
1512 |
+
"learning_rate": 7.169191270860078e-06,
|
1513 |
+
"loss": 1.0586,
|
1514 |
+
"step": 50500
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 64.41,
|
1518 |
+
"eval_loss": 0.23375432193279266,
|
1519 |
+
"eval_runtime": 233.8132,
|
1520 |
+
"eval_samples_per_second": 35.665,
|
1521 |
+
"eval_steps_per_second": 1.116,
|
1522 |
+
"eval_wer": 0.30435181153608176,
|
1523 |
+
"step": 50500
|
1524 |
+
},
|
1525 |
+
{
|
1526 |
+
"epoch": 65.05,
|
1527 |
+
"learning_rate": 7.040821566110398e-06,
|
1528 |
+
"loss": 1.0753,
|
1529 |
+
"step": 51000
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 65.05,
|
1533 |
+
"eval_loss": 0.23079071938991547,
|
1534 |
+
"eval_runtime": 238.4184,
|
1535 |
+
"eval_samples_per_second": 34.976,
|
1536 |
+
"eval_steps_per_second": 1.095,
|
1537 |
+
"eval_wer": 0.30446639319843255,
|
1538 |
+
"step": 51000
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 65.69,
|
1542 |
+
"learning_rate": 6.912451861360719e-06,
|
1543 |
+
"loss": 1.0664,
|
1544 |
+
"step": 51500
|
1545 |
+
},
|
1546 |
+
{
|
1547 |
+
"epoch": 65.69,
|
1548 |
+
"eval_loss": 0.22733470797538757,
|
1549 |
+
"eval_runtime": 233.5628,
|
1550 |
+
"eval_samples_per_second": 35.703,
|
1551 |
+
"eval_steps_per_second": 1.117,
|
1552 |
+
"eval_wer": 0.3009372779980292,
|
1553 |
+
"step": 51500
|
1554 |
+
},
|
1555 |
+
{
|
1556 |
+
"epoch": 66.33,
|
1557 |
+
"learning_rate": 6.78408215661104e-06,
|
1558 |
+
"loss": 1.0739,
|
1559 |
+
"step": 52000
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 66.33,
|
1563 |
+
"eval_loss": 0.2298172414302826,
|
1564 |
+
"eval_runtime": 234.5885,
|
1565 |
+
"eval_samples_per_second": 35.547,
|
1566 |
+
"eval_steps_per_second": 1.113,
|
1567 |
+
"eval_wer": 0.30272475193070103,
|
1568 |
+
"step": 52000
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 66.96,
|
1572 |
+
"learning_rate": 6.655712451861361e-06,
|
1573 |
+
"loss": 1.0695,
|
1574 |
+
"step": 52500
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 66.96,
|
1578 |
+
"eval_loss": 0.2247009128332138,
|
1579 |
+
"eval_runtime": 234.2623,
|
1580 |
+
"eval_samples_per_second": 35.597,
|
1581 |
+
"eval_steps_per_second": 1.114,
|
1582 |
+
"eval_wer": 0.29958521438229024,
|
1583 |
+
"step": 52500
|
1584 |
+
},
|
1585 |
+
{
|
1586 |
+
"epoch": 67.6,
|
1587 |
+
"learning_rate": 6.5273427471116825e-06,
|
1588 |
+
"loss": 1.06,
|
1589 |
+
"step": 53000
|
1590 |
+
},
|
1591 |
+
{
|
1592 |
+
"epoch": 67.6,
|
1593 |
+
"eval_loss": 0.22762586176395416,
|
1594 |
+
"eval_runtime": 238.2043,
|
1595 |
+
"eval_samples_per_second": 35.008,
|
1596 |
+
"eval_steps_per_second": 1.096,
|
1597 |
+
"eval_wer": 0.30148726997731284,
|
1598 |
+
"step": 53000
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 68.24,
|
1602 |
+
"learning_rate": 6.399229781771502e-06,
|
1603 |
+
"loss": 1.0742,
|
1604 |
+
"step": 53500
|
1605 |
+
},
|
1606 |
+
{
|
1607 |
+
"epoch": 68.24,
|
1608 |
+
"eval_loss": 0.22798208892345428,
|
1609 |
+
"eval_runtime": 234.5841,
|
1610 |
+
"eval_samples_per_second": 35.548,
|
1611 |
+
"eval_steps_per_second": 1.113,
|
1612 |
+
"eval_wer": 0.2974081627976259,
|
1613 |
+
"step": 53500
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 68.88,
|
1617 |
+
"learning_rate": 6.270860077021823e-06,
|
1618 |
+
"loss": 1.0618,
|
1619 |
+
"step": 54000
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 68.88,
|
1623 |
+
"eval_loss": 0.2291172742843628,
|
1624 |
+
"eval_runtime": 233.7512,
|
1625 |
+
"eval_samples_per_second": 35.675,
|
1626 |
+
"eval_steps_per_second": 1.117,
|
1627 |
+
"eval_wer": 0.2988518917432454,
|
1628 |
+
"step": 54000
|
1629 |
+
},
|
1630 |
+
{
|
1631 |
+
"epoch": 69.52,
|
1632 |
+
"learning_rate": 6.142490372272144e-06,
|
1633 |
+
"loss": 1.062,
|
1634 |
+
"step": 54500
|
1635 |
+
},
|
1636 |
+
{
|
1637 |
+
"epoch": 69.52,
|
1638 |
+
"eval_loss": 0.23021718859672546,
|
1639 |
+
"eval_runtime": 236.2703,
|
1640 |
+
"eval_samples_per_second": 35.294,
|
1641 |
+
"eval_steps_per_second": 1.105,
|
1642 |
+
"eval_wer": 0.2970873341430437,
|
1643 |
+
"step": 54500
|
1644 |
+
},
|
1645 |
+
{
|
1646 |
+
"epoch": 70.15,
|
1647 |
+
"learning_rate": 6.0141206675224654e-06,
|
1648 |
+
"loss": 1.0572,
|
1649 |
+
"step": 55000
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 70.15,
|
1653 |
+
"eval_loss": 0.22801047563552856,
|
1654 |
+
"eval_runtime": 232.8765,
|
1655 |
+
"eval_samples_per_second": 35.809,
|
1656 |
+
"eval_steps_per_second": 1.121,
|
1657 |
+
"eval_wer": 0.2989893897380663,
|
1658 |
+
"step": 55000
|
1659 |
+
},
|
1660 |
+
{
|
1661 |
+
"epoch": 70.79,
|
1662 |
+
"learning_rate": 5.885750962772786e-06,
|
1663 |
+
"loss": 1.055,
|
1664 |
+
"step": 55500
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 70.79,
|
1668 |
+
"eval_loss": 0.22775860130786896,
|
1669 |
+
"eval_runtime": 234.9515,
|
1670 |
+
"eval_samples_per_second": 35.492,
|
1671 |
+
"eval_steps_per_second": 1.111,
|
1672 |
+
"eval_wer": 0.29834773242890206,
|
1673 |
+
"step": 55500
|
1674 |
+
},
|
1675 |
+
{
|
1676 |
+
"epoch": 71.43,
|
1677 |
+
"learning_rate": 5.757381258023107e-06,
|
1678 |
+
"loss": 1.0553,
|
1679 |
+
"step": 56000
|
1680 |
+
},
|
1681 |
+
{
|
1682 |
+
"epoch": 71.43,
|
1683 |
+
"eval_loss": 0.22817462682724,
|
1684 |
+
"eval_runtime": 233.8475,
|
1685 |
+
"eval_samples_per_second": 35.66,
|
1686 |
+
"eval_steps_per_second": 1.116,
|
1687 |
+
"eval_wer": 0.29905813873547676,
|
1688 |
+
"step": 56000
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 72.07,
|
1692 |
+
"learning_rate": 5.629011553273428e-06,
|
1693 |
+
"loss": 1.0509,
|
1694 |
+
"step": 56500
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 72.07,
|
1698 |
+
"eval_loss": 0.2261197566986084,
|
1699 |
+
"eval_runtime": 235.4495,
|
1700 |
+
"eval_samples_per_second": 35.417,
|
1701 |
+
"eval_steps_per_second": 1.109,
|
1702 |
+
"eval_wer": 0.29591860118706603,
|
1703 |
+
"step": 56500
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 72.7,
|
1707 |
+
"learning_rate": 5.500641848523749e-06,
|
1708 |
+
"loss": 1.0469,
|
1709 |
+
"step": 57000
|
1710 |
+
},
|
1711 |
+
{
|
1712 |
+
"epoch": 72.7,
|
1713 |
+
"eval_loss": 0.22159287333488464,
|
1714 |
+
"eval_runtime": 233.6081,
|
1715 |
+
"eval_samples_per_second": 35.697,
|
1716 |
+
"eval_steps_per_second": 1.117,
|
1717 |
+
"eval_wer": 0.29188532667231937,
|
1718 |
+
"step": 57000
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 73.34,
|
1722 |
+
"learning_rate": 5.37227214377407e-06,
|
1723 |
+
"loss": 1.0476,
|
1724 |
+
"step": 57500
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 73.34,
|
1728 |
+
"eval_loss": 0.22673606872558594,
|
1729 |
+
"eval_runtime": 233.9171,
|
1730 |
+
"eval_samples_per_second": 35.649,
|
1731 |
+
"eval_steps_per_second": 1.116,
|
1732 |
+
"eval_wer": 0.2988518917432454,
|
1733 |
+
"step": 57500
|
1734 |
+
},
|
1735 |
+
{
|
1736 |
+
"epoch": 73.98,
|
1737 |
+
"learning_rate": 5.24415917843389e-06,
|
1738 |
+
"loss": 1.0494,
|
1739 |
+
"step": 58000
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 73.98,
|
1743 |
+
"eval_loss": 0.22603431344032288,
|
1744 |
+
"eval_runtime": 234.549,
|
1745 |
+
"eval_samples_per_second": 35.553,
|
1746 |
+
"eval_steps_per_second": 1.113,
|
1747 |
+
"eval_wer": 0.2960102665169466,
|
1748 |
+
"step": 58000
|
1749 |
+
},
|
1750 |
+
{
|
1751 |
+
"epoch": 74.62,
|
1752 |
+
"learning_rate": 5.115789473684211e-06,
|
1753 |
+
"loss": 1.0517,
|
1754 |
+
"step": 58500
|
1755 |
+
},
|
1756 |
+
{
|
1757 |
+
"epoch": 74.62,
|
1758 |
+
"eval_loss": 0.2297457605600357,
|
1759 |
+
"eval_runtime": 234.3971,
|
1760 |
+
"eval_samples_per_second": 35.576,
|
1761 |
+
"eval_steps_per_second": 1.113,
|
1762 |
+
"eval_wer": 0.29894355707312603,
|
1763 |
+
"step": 58500
|
1764 |
+
},
|
1765 |
+
{
|
1766 |
+
"epoch": 75.26,
|
1767 |
+
"learning_rate": 4.987419768934532e-06,
|
1768 |
+
"loss": 1.0458,
|
1769 |
+
"step": 59000
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 75.26,
|
1773 |
+
"eval_loss": 0.2246260643005371,
|
1774 |
+
"eval_runtime": 233.0094,
|
1775 |
+
"eval_samples_per_second": 35.788,
|
1776 |
+
"eval_steps_per_second": 1.12,
|
1777 |
+
"eval_wer": 0.2923436533217224,
|
1778 |
+
"step": 59000
|
1779 |
+
},
|
1780 |
+
{
|
1781 |
+
"epoch": 75.89,
|
1782 |
+
"learning_rate": 4.859306803594352e-06,
|
1783 |
+
"loss": 1.0382,
|
1784 |
+
"step": 59500
|
1785 |
+
},
|
1786 |
+
{
|
1787 |
+
"epoch": 75.89,
|
1788 |
+
"eval_loss": 0.22546544671058655,
|
1789 |
+
"eval_runtime": 233.7397,
|
1790 |
+
"eval_samples_per_second": 35.676,
|
1791 |
+
"eval_steps_per_second": 1.117,
|
1792 |
+
"eval_wer": 0.2921603226619612,
|
1793 |
+
"step": 59500
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 76.53,
|
1797 |
+
"learning_rate": 4.730937098844673e-06,
|
1798 |
+
"loss": 1.0462,
|
1799 |
+
"step": 60000
|
1800 |
+
},
|
1801 |
+
{
|
1802 |
+
"epoch": 76.53,
|
1803 |
+
"eval_loss": 0.22583839297294617,
|
1804 |
+
"eval_runtime": 233.483,
|
1805 |
+
"eval_samples_per_second": 35.716,
|
1806 |
+
"eval_steps_per_second": 1.118,
|
1807 |
+
"eval_wer": 0.29539152554025255,
|
1808 |
+
"step": 60000
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 77.17,
|
1812 |
+
"learning_rate": 4.602567394094994e-06,
|
1813 |
+
"loss": 1.0375,
|
1814 |
+
"step": 60500
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 77.17,
|
1818 |
+
"eval_loss": 0.22505535185337067,
|
1819 |
+
"eval_runtime": 239.1762,
|
1820 |
+
"eval_samples_per_second": 34.866,
|
1821 |
+
"eval_steps_per_second": 1.091,
|
1822 |
+
"eval_wer": 0.2928707289685359,
|
1823 |
+
"step": 60500
|
1824 |
+
},
|
1825 |
+
{
|
1826 |
+
"epoch": 77.81,
|
1827 |
+
"learning_rate": 4.474197689345315e-06,
|
1828 |
+
"loss": 1.0332,
|
1829 |
+
"step": 61000
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 77.81,
|
1833 |
+
"eval_loss": 0.22773370146751404,
|
1834 |
+
"eval_runtime": 237.0288,
|
1835 |
+
"eval_samples_per_second": 35.181,
|
1836 |
+
"eval_steps_per_second": 1.101,
|
1837 |
+
"eval_wer": 0.2939936292595733,
|
1838 |
+
"step": 61000
|
1839 |
+
},
|
1840 |
+
{
|
1841 |
+
"epoch": 78.44,
|
1842 |
+
"learning_rate": 4.345827984595636e-06,
|
1843 |
+
"loss": 1.0423,
|
1844 |
+
"step": 61500
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 78.44,
|
1848 |
+
"eval_loss": 0.22426965832710266,
|
1849 |
+
"eval_runtime": 233.5981,
|
1850 |
+
"eval_samples_per_second": 35.698,
|
1851 |
+
"eval_steps_per_second": 1.117,
|
1852 |
+
"eval_wer": 0.28961660975777437,
|
1853 |
+
"step": 61500
|
1854 |
+
},
|
1855 |
+
{
|
1856 |
+
"epoch": 79.08,
|
1857 |
+
"learning_rate": 4.217458279845957e-06,
|
1858 |
+
"loss": 1.0379,
|
1859 |
+
"step": 62000
|
1860 |
+
},
|
1861 |
+
{
|
1862 |
+
"epoch": 79.08,
|
1863 |
+
"eval_loss": 0.2274049073457718,
|
1864 |
+
"eval_runtime": 234.6064,
|
1865 |
+
"eval_samples_per_second": 35.545,
|
1866 |
+
"eval_steps_per_second": 1.113,
|
1867 |
+
"eval_wer": 0.29280197997112545,
|
1868 |
+
"step": 62000
|
1869 |
+
},
|
1870 |
+
{
|
1871 |
+
"epoch": 79.72,
|
1872 |
+
"learning_rate": 4.089345314505777e-06,
|
1873 |
+
"loss": 1.0398,
|
1874 |
+
"step": 62500
|
1875 |
+
},
|
1876 |
+
{
|
1877 |
+
"epoch": 79.72,
|
1878 |
+
"eval_loss": 0.22373217344284058,
|
1879 |
+
"eval_runtime": 234.7196,
|
1880 |
+
"eval_samples_per_second": 35.527,
|
1881 |
+
"eval_steps_per_second": 1.112,
|
1882 |
+
"eval_wer": 0.29280197997112545,
|
1883 |
+
"step": 62500
|
1884 |
+
},
|
1885 |
+
{
|
1886 |
+
"epoch": 80.36,
|
1887 |
+
"learning_rate": 3.960975609756098e-06,
|
1888 |
+
"loss": 1.0395,
|
1889 |
+
"step": 63000
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 80.36,
|
1893 |
+
"eval_loss": 0.22652971744537354,
|
1894 |
+
"eval_runtime": 233.8582,
|
1895 |
+
"eval_samples_per_second": 35.658,
|
1896 |
+
"eval_steps_per_second": 1.116,
|
1897 |
+
"eval_wer": 0.2956206888649541,
|
1898 |
+
"step": 63000
|
1899 |
+
},
|
1900 |
+
{
|
1901 |
+
"epoch": 80.99,
|
1902 |
+
"learning_rate": 3.832605905006419e-06,
|
1903 |
+
"loss": 1.0397,
|
1904 |
+
"step": 63500
|
1905 |
+
},
|
1906 |
+
{
|
1907 |
+
"epoch": 80.99,
|
1908 |
+
"eval_loss": 0.22396665811538696,
|
1909 |
+
"eval_runtime": 235.3841,
|
1910 |
+
"eval_samples_per_second": 35.427,
|
1911 |
+
"eval_steps_per_second": 1.109,
|
1912 |
+
"eval_wer": 0.29202282466714025,
|
1913 |
+
"step": 63500
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 81.63,
|
1917 |
+
"learning_rate": 3.7042362002567396e-06,
|
1918 |
+
"loss": 1.0262,
|
1919 |
+
"step": 64000
|
1920 |
+
},
|
1921 |
+
{
|
1922 |
+
"epoch": 81.63,
|
1923 |
+
"eval_loss": 0.22437626123428345,
|
1924 |
+
"eval_runtime": 239.3117,
|
1925 |
+
"eval_samples_per_second": 34.846,
|
1926 |
+
"eval_steps_per_second": 1.091,
|
1927 |
+
"eval_wer": 0.2933748882828792,
|
1928 |
+
"step": 64000
|
1929 |
+
},
|
1930 |
+
{
|
1931 |
+
"epoch": 82.27,
|
1932 |
+
"learning_rate": 3.5758664955070607e-06,
|
1933 |
+
"loss": 1.0335,
|
1934 |
+
"step": 64500
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 82.27,
|
1938 |
+
"eval_loss": 0.22651708126068115,
|
1939 |
+
"eval_runtime": 234.0011,
|
1940 |
+
"eval_samples_per_second": 35.637,
|
1941 |
+
"eval_steps_per_second": 1.115,
|
1942 |
+
"eval_wer": 0.2936269679400509,
|
1943 |
+
"step": 64500
|
1944 |
+
},
|
1945 |
+
{
|
1946 |
+
"epoch": 82.91,
|
1947 |
+
"learning_rate": 3.4474967907573813e-06,
|
1948 |
+
"loss": 1.0385,
|
1949 |
+
"step": 65000
|
1950 |
+
},
|
1951 |
+
{
|
1952 |
+
"epoch": 82.91,
|
1953 |
+
"eval_loss": 0.22380606830120087,
|
1954 |
+
"eval_runtime": 236.2381,
|
1955 |
+
"eval_samples_per_second": 35.299,
|
1956 |
+
"eval_steps_per_second": 1.105,
|
1957 |
+
"eval_wer": 0.29284781263606574,
|
1958 |
+
"step": 65000
|
1959 |
+
},
|
1960 |
+
{
|
1961 |
+
"epoch": 83.55,
|
1962 |
+
"learning_rate": 3.3191270860077024e-06,
|
1963 |
+
"loss": 1.0289,
|
1964 |
+
"step": 65500
|
1965 |
+
},
|
1966 |
+
{
|
1967 |
+
"epoch": 83.55,
|
1968 |
+
"eval_loss": 0.221877783536911,
|
1969 |
+
"eval_runtime": 233.6081,
|
1970 |
+
"eval_samples_per_second": 35.697,
|
1971 |
+
"eval_steps_per_second": 1.117,
|
1972 |
+
"eval_wer": 0.2911520040332745,
|
1973 |
+
"step": 65500
|
1974 |
+
},
|
1975 |
+
{
|
1976 |
+
"epoch": 84.18,
|
1977 |
+
"learning_rate": 3.1910141206675225e-06,
|
1978 |
+
"loss": 1.0372,
|
1979 |
+
"step": 66000
|
1980 |
+
},
|
1981 |
+
{
|
1982 |
+
"epoch": 84.18,
|
1983 |
+
"eval_loss": 0.22357773780822754,
|
1984 |
+
"eval_runtime": 234.3938,
|
1985 |
+
"eval_samples_per_second": 35.577,
|
1986 |
+
"eval_steps_per_second": 1.114,
|
1987 |
+
"eval_wer": 0.2898457730824759,
|
1988 |
+
"step": 66000
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 84.82,
|
1992 |
+
"learning_rate": 3.0626444159178436e-06,
|
1993 |
+
"loss": 1.0279,
|
1994 |
+
"step": 66500
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 84.82,
|
1998 |
+
"eval_loss": 0.22185130417346954,
|
1999 |
+
"eval_runtime": 235.3542,
|
2000 |
+
"eval_samples_per_second": 35.432,
|
2001 |
+
"eval_steps_per_second": 1.109,
|
2002 |
+
"eval_wer": 0.2902124344019983,
|
2003 |
+
"step": 66500
|
2004 |
+
},
|
2005 |
+
{
|
2006 |
+
"epoch": 85.46,
|
2007 |
+
"learning_rate": 2.9342747111681647e-06,
|
2008 |
+
"loss": 1.0325,
|
2009 |
+
"step": 67000
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 85.46,
|
2013 |
+
"eval_loss": 0.22399279475212097,
|
2014 |
+
"eval_runtime": 232.7577,
|
2015 |
+
"eval_samples_per_second": 35.827,
|
2016 |
+
"eval_steps_per_second": 1.121,
|
2017 |
+
"eval_wer": 0.2907853427137521,
|
2018 |
+
"step": 67000
|
2019 |
+
},
|
2020 |
+
{
|
2021 |
+
"epoch": 86.1,
|
2022 |
+
"learning_rate": 2.8059050064184858e-06,
|
2023 |
+
"loss": 1.0202,
|
2024 |
+
"step": 67500
|
2025 |
+
},
|
2026 |
+
{
|
2027 |
+
"epoch": 86.1,
|
2028 |
+
"eval_loss": 0.22061078250408173,
|
2029 |
+
"eval_runtime": 236.8657,
|
2030 |
+
"eval_samples_per_second": 35.206,
|
2031 |
+
"eval_steps_per_second": 1.102,
|
2032 |
+
"eval_wer": 0.28863120746155785,
|
2033 |
+
"step": 67500
|
2034 |
+
},
|
2035 |
+
{
|
2036 |
+
"epoch": 86.73,
|
2037 |
+
"learning_rate": 2.6775353016688064e-06,
|
2038 |
+
"loss": 1.0166,
|
2039 |
+
"step": 68000
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 86.73,
|
2043 |
+
"eval_loss": 0.2218720018863678,
|
2044 |
+
"eval_runtime": 233.7184,
|
2045 |
+
"eval_samples_per_second": 35.68,
|
2046 |
+
"eval_steps_per_second": 1.117,
|
2047 |
+
"eval_wer": 0.2885624584641474,
|
2048 |
+
"step": 68000
|
2049 |
+
},
|
2050 |
+
{
|
2051 |
+
"epoch": 87.37,
|
2052 |
+
"learning_rate": 2.5491655969191275e-06,
|
2053 |
+
"loss": 1.0259,
|
2054 |
+
"step": 68500
|
2055 |
+
},
|
2056 |
+
{
|
2057 |
+
"epoch": 87.37,
|
2058 |
+
"eval_loss": 0.2234880030155182,
|
2059 |
+
"eval_runtime": 235.0768,
|
2060 |
+
"eval_samples_per_second": 35.474,
|
2061 |
+
"eval_steps_per_second": 1.11,
|
2062 |
+
"eval_wer": 0.28970827508765495,
|
2063 |
+
"step": 68500
|
2064 |
+
},
|
2065 |
+
{
|
2066 |
+
"epoch": 88.01,
|
2067 |
+
"learning_rate": 2.420795892169448e-06,
|
2068 |
+
"loss": 1.0337,
|
2069 |
+
"step": 69000
|
2070 |
+
},
|
2071 |
+
{
|
2072 |
+
"epoch": 88.01,
|
2073 |
+
"eval_loss": 0.220963254570961,
|
2074 |
+
"eval_runtime": 233.4278,
|
2075 |
+
"eval_samples_per_second": 35.724,
|
2076 |
+
"eval_steps_per_second": 1.118,
|
2077 |
+
"eval_wer": 0.2873249765107592,
|
2078 |
+
"step": 69000
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 88.65,
|
2082 |
+
"learning_rate": 2.2926829268292687e-06,
|
2083 |
+
"loss": 1.0264,
|
2084 |
+
"step": 69500
|
2085 |
+
},
|
2086 |
+
{
|
2087 |
+
"epoch": 88.65,
|
2088 |
+
"eval_loss": 0.22157110273838043,
|
2089 |
+
"eval_runtime": 233.5356,
|
2090 |
+
"eval_samples_per_second": 35.708,
|
2091 |
+
"eval_steps_per_second": 1.118,
|
2092 |
+
"eval_wer": 0.2882187134770951,
|
2093 |
+
"step": 69500
|
2094 |
+
},
|
2095 |
+
{
|
2096 |
+
"epoch": 89.29,
|
2097 |
+
"learning_rate": 2.1643132220795894e-06,
|
2098 |
+
"loss": 1.0231,
|
2099 |
+
"step": 70000
|
2100 |
+
},
|
2101 |
+
{
|
2102 |
+
"epoch": 89.29,
|
2103 |
+
"eval_loss": 0.2222645878791809,
|
2104 |
+
"eval_runtime": 238.1657,
|
2105 |
+
"eval_samples_per_second": 35.013,
|
2106 |
+
"eval_steps_per_second": 1.096,
|
2107 |
+
"eval_wer": 0.2899145220798863,
|
2108 |
+
"step": 70000
|
2109 |
+
},
|
2110 |
+
{
|
2111 |
+
"epoch": 89.92,
|
2112 |
+
"learning_rate": 2.03594351732991e-06,
|
2113 |
+
"loss": 1.0281,
|
2114 |
+
"step": 70500
|
2115 |
+
},
|
2116 |
+
{
|
2117 |
+
"epoch": 89.92,
|
2118 |
+
"eval_loss": 0.22140374779701233,
|
2119 |
+
"eval_runtime": 233.9034,
|
2120 |
+
"eval_samples_per_second": 35.651,
|
2121 |
+
"eval_steps_per_second": 1.116,
|
2122 |
+
"eval_wer": 0.28723331118087864,
|
2123 |
+
"step": 70500
|
2124 |
+
},
|
2125 |
+
{
|
2126 |
+
"epoch": 90.56,
|
2127 |
+
"learning_rate": 1.907573812580231e-06,
|
2128 |
+
"loss": 1.0135,
|
2129 |
+
"step": 71000
|
2130 |
+
},
|
2131 |
+
{
|
2132 |
+
"epoch": 90.56,
|
2133 |
+
"eval_loss": 0.22179700434207916,
|
2134 |
+
"eval_runtime": 235.221,
|
2135 |
+
"eval_samples_per_second": 35.452,
|
2136 |
+
"eval_steps_per_second": 1.11,
|
2137 |
+
"eval_wer": 0.28684373352888604,
|
2138 |
+
"step": 71000
|
2139 |
+
},
|
2140 |
+
{
|
2141 |
+
"epoch": 91.2,
|
2142 |
+
"learning_rate": 1.7792041078305522e-06,
|
2143 |
+
"loss": 1.0291,
|
2144 |
+
"step": 71500
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 91.2,
|
2148 |
+
"eval_loss": 0.22093360126018524,
|
2149 |
+
"eval_runtime": 236.8894,
|
2150 |
+
"eval_samples_per_second": 35.202,
|
2151 |
+
"eval_steps_per_second": 1.102,
|
2152 |
+
"eval_wer": 0.2862937415496024,
|
2153 |
+
"step": 71500
|
2154 |
+
},
|
2155 |
+
{
|
2156 |
+
"epoch": 91.84,
|
2157 |
+
"learning_rate": 1.6508344030808731e-06,
|
2158 |
+
"loss": 1.0321,
|
2159 |
+
"step": 72000
|
2160 |
+
},
|
2161 |
+
{
|
2162 |
+
"epoch": 91.84,
|
2163 |
+
"eval_loss": 0.219914972782135,
|
2164 |
+
"eval_runtime": 233.825,
|
2165 |
+
"eval_samples_per_second": 35.663,
|
2166 |
+
"eval_steps_per_second": 1.116,
|
2167 |
+
"eval_wer": 0.2875770561679309,
|
2168 |
+
"step": 72000
|
2169 |
+
},
|
2170 |
+
{
|
2171 |
+
"epoch": 92.47,
|
2172 |
+
"learning_rate": 1.5227214377406932e-06,
|
2173 |
+
"loss": 1.028,
|
2174 |
+
"step": 72500
|
2175 |
+
},
|
2176 |
+
{
|
2177 |
+
"epoch": 92.47,
|
2178 |
+
"eval_loss": 0.22144030034542084,
|
2179 |
+
"eval_runtime": 234.5883,
|
2180 |
+
"eval_samples_per_second": 35.547,
|
2181 |
+
"eval_steps_per_second": 1.113,
|
2182 |
+
"eval_wer": 0.28576666590278893,
|
2183 |
+
"step": 72500
|
2184 |
+
},
|
2185 |
+
{
|
2186 |
+
"epoch": 93.11,
|
2187 |
+
"learning_rate": 1.3943517329910141e-06,
|
2188 |
+
"loss": 1.0213,
|
2189 |
+
"step": 73000
|
2190 |
+
},
|
2191 |
+
{
|
2192 |
+
"epoch": 93.11,
|
2193 |
+
"eval_loss": 0.2219470739364624,
|
2194 |
+
"eval_runtime": 234.1897,
|
2195 |
+
"eval_samples_per_second": 35.608,
|
2196 |
+
"eval_steps_per_second": 1.114,
|
2197 |
+
"eval_wer": 0.2874624745055801,
|
2198 |
+
"step": 73000
|
2199 |
+
},
|
2200 |
+
{
|
2201 |
+
"epoch": 93.75,
|
2202 |
+
"learning_rate": 1.2659820282413352e-06,
|
2203 |
+
"loss": 1.0261,
|
2204 |
+
"step": 73500
|
2205 |
+
},
|
2206 |
+
{
|
2207 |
+
"epoch": 93.75,
|
2208 |
+
"eval_loss": 0.22315841913223267,
|
2209 |
+
"eval_runtime": 234.7799,
|
2210 |
+
"eval_samples_per_second": 35.518,
|
2211 |
+
"eval_steps_per_second": 1.112,
|
2212 |
+
"eval_wer": 0.28688956619382633,
|
2213 |
+
"step": 73500
|
2214 |
+
},
|
2215 |
+
{
|
2216 |
+
"epoch": 94.39,
|
2217 |
+
"learning_rate": 1.137612323491656e-06,
|
2218 |
+
"loss": 1.0197,
|
2219 |
+
"step": 74000
|
2220 |
+
},
|
2221 |
+
{
|
2222 |
+
"epoch": 94.39,
|
2223 |
+
"eval_loss": 0.22271272540092468,
|
2224 |
+
"eval_runtime": 233.7941,
|
2225 |
+
"eval_samples_per_second": 35.668,
|
2226 |
+
"eval_steps_per_second": 1.116,
|
2227 |
+
"eval_wer": 0.2866145702041845,
|
2228 |
+
"step": 74000
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 95.03,
|
2232 |
+
"learning_rate": 1.009242618741977e-06,
|
2233 |
+
"loss": 1.0298,
|
2234 |
+
"step": 74500
|
2235 |
+
},
|
2236 |
+
{
|
2237 |
+
"epoch": 95.03,
|
2238 |
+
"eval_loss": 0.22284960746765137,
|
2239 |
+
"eval_runtime": 236.8784,
|
2240 |
+
"eval_samples_per_second": 35.204,
|
2241 |
+
"eval_steps_per_second": 1.102,
|
2242 |
+
"eval_wer": 0.28675206819900545,
|
2243 |
+
"step": 74500
|
2244 |
+
},
|
2245 |
+
{
|
2246 |
+
"epoch": 95.66,
|
2247 |
+
"learning_rate": 8.808729139922979e-07,
|
2248 |
+
"loss": 1.0192,
|
2249 |
+
"step": 75000
|
2250 |
+
},
|
2251 |
+
{
|
2252 |
+
"epoch": 95.66,
|
2253 |
+
"eval_loss": 0.22300392389297485,
|
2254 |
+
"eval_runtime": 233.5836,
|
2255 |
+
"eval_samples_per_second": 35.7,
|
2256 |
+
"eval_steps_per_second": 1.117,
|
2257 |
+
"eval_wer": 0.28652290487430393,
|
2258 |
+
"step": 75000
|
2259 |
+
},
|
2260 |
+
{
|
2261 |
+
"epoch": 96.3,
|
2262 |
+
"learning_rate": 7.525032092426188e-07,
|
2263 |
+
"loss": 1.0156,
|
2264 |
+
"step": 75500
|
2265 |
+
},
|
2266 |
+
{
|
2267 |
+
"epoch": 96.3,
|
2268 |
+
"eval_loss": 0.2220136970281601,
|
2269 |
+
"eval_runtime": 231.7898,
|
2270 |
+
"eval_samples_per_second": 35.977,
|
2271 |
+
"eval_steps_per_second": 1.126,
|
2272 |
+
"eval_wer": 0.2869353988587666,
|
2273 |
+
"step": 75500
|
2274 |
+
},
|
2275 |
+
{
|
2276 |
+
"epoch": 96.94,
|
2277 |
+
"learning_rate": 6.241335044929398e-07,
|
2278 |
+
"loss": 1.0075,
|
2279 |
+
"step": 76000
|
2280 |
+
},
|
2281 |
+
{
|
2282 |
+
"epoch": 96.94,
|
2283 |
+
"eval_loss": 0.2222670316696167,
|
2284 |
+
"eval_runtime": 232.6992,
|
2285 |
+
"eval_samples_per_second": 35.836,
|
2286 |
+
"eval_steps_per_second": 1.122,
|
2287 |
+
"eval_wer": 0.2866145702041845,
|
2288 |
+
"step": 76000
|
2289 |
+
},
|
2290 |
+
{
|
2291 |
+
"epoch": 97.58,
|
2292 |
+
"learning_rate": 4.9602053915276e-07,
|
2293 |
+
"loss": 1.0201,
|
2294 |
+
"step": 76500
|
2295 |
+
},
|
2296 |
+
{
|
2297 |
+
"epoch": 97.58,
|
2298 |
+
"eval_loss": 0.2218976616859436,
|
2299 |
+
"eval_runtime": 233.5288,
|
2300 |
+
"eval_samples_per_second": 35.709,
|
2301 |
+
"eval_steps_per_second": 1.118,
|
2302 |
+
"eval_wer": 0.28659165387171437,
|
2303 |
+
"step": 76500
|
2304 |
+
},
|
2305 |
+
{
|
2306 |
+
"epoch": 98.21,
|
2307 |
+
"learning_rate": 3.676508344030809e-07,
|
2308 |
+
"loss": 1.0159,
|
2309 |
+
"step": 77000
|
2310 |
+
},
|
2311 |
+
{
|
2312 |
+
"epoch": 98.21,
|
2313 |
+
"eval_loss": 0.22194130718708038,
|
2314 |
+
"eval_runtime": 233.434,
|
2315 |
+
"eval_samples_per_second": 35.723,
|
2316 |
+
"eval_steps_per_second": 1.118,
|
2317 |
+
"eval_wer": 0.28764580516534133,
|
2318 |
+
"step": 77000
|
2319 |
+
},
|
2320 |
+
{
|
2321 |
+
"epoch": 98.85,
|
2322 |
+
"learning_rate": 2.395378690629012e-07,
|
2323 |
+
"loss": 1.0087,
|
2324 |
+
"step": 77500
|
2325 |
+
},
|
2326 |
+
{
|
2327 |
+
"epoch": 98.85,
|
2328 |
+
"eval_loss": 0.2219318002462387,
|
2329 |
+
"eval_runtime": 235.7197,
|
2330 |
+
"eval_samples_per_second": 35.377,
|
2331 |
+
"eval_steps_per_second": 1.107,
|
2332 |
+
"eval_wer": 0.2873249765107592,
|
2333 |
+
"step": 77500
|
2334 |
+
},
|
2335 |
+
{
|
2336 |
+
"epoch": 99.49,
|
2337 |
+
"learning_rate": 1.111681643132221e-07,
|
2338 |
+
"loss": 1.0159,
|
2339 |
+
"step": 78000
|
2340 |
+
},
|
2341 |
+
{
|
2342 |
+
"epoch": 99.49,
|
2343 |
+
"eval_loss": 0.2223382443189621,
|
2344 |
+
"eval_runtime": 234.8374,
|
2345 |
+
"eval_samples_per_second": 35.51,
|
2346 |
+
"eval_steps_per_second": 1.111,
|
2347 |
+
"eval_wer": 0.2867291518665353,
|
2348 |
+
"step": 78000
|
2349 |
+
},
|
2350 |
+
{
|
2351 |
+
"epoch": 100.0,
|
2352 |
+
"step": 78400,
|
2353 |
+
"total_flos": 1.850628803077507e+20,
|
2354 |
+
"train_loss": 1.2247810986577248,
|
2355 |
+
"train_runtime": 113701.5353,
|
2356 |
+
"train_samples_per_second": 22.038,
|
2357 |
+
"train_steps_per_second": 0.69
|
2358 |
+
}
|
2359 |
+
],
|
2360 |
+
"max_steps": 78400,
|
2361 |
+
"num_train_epochs": 100,
|
2362 |
+
"total_flos": 1.850628803077507e+20,
|
2363 |
+
"trial_name": null,
|
2364 |
+
"trial_params": null
|
2365 |
+
}
|