End of training
Browse files
all_results.json
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 94.78,
|
3 |
+
"eval_accuracy": 0.4534356430493973,
|
4 |
+
"eval_loss": 3.1695425510406494,
|
5 |
+
"eval_runtime": 128.1811,
|
6 |
+
"eval_samples": 24055,
|
7 |
+
"eval_samples_per_second": 187.664,
|
8 |
+
"eval_steps_per_second": 5.867,
|
9 |
+
"perplexity": 23.796596138077017,
|
10 |
+
"train_loss": 3.832834072322636,
|
11 |
+
"train_runtime": 39205.3218,
|
12 |
+
"train_samples": 24910,
|
13 |
+
"train_samples_per_second": 63.537,
|
14 |
+
"train_steps_per_second": 0.122
|
15 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 94.78,
|
3 |
+
"eval_accuracy": 0.4534356430493973,
|
4 |
+
"eval_loss": 3.1695425510406494,
|
5 |
+
"eval_runtime": 128.1811,
|
6 |
+
"eval_samples": 24055,
|
7 |
+
"eval_samples_per_second": 187.664,
|
8 |
+
"eval_steps_per_second": 5.867,
|
9 |
+
"perplexity": 23.796596138077017
|
10 |
+
}
|
runs/Feb15_12-10-24_monica.sm.unibocconi.it/events.out.tfevents.1676498906.monica.sm.unibocconi.it.747021.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e2d8b00fa1c549c3efda41eec6f67bdd1f4bd9f37309a3cc712fb2fcbf5639ba
|
3 |
+
size 363
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 94.78,
|
3 |
+
"train_loss": 3.832834072322636,
|
4 |
+
"train_runtime": 39205.3218,
|
5 |
+
"train_samples": 24910,
|
6 |
+
"train_samples_per_second": 63.537,
|
7 |
+
"train_steps_per_second": 0.122
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,3574 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 3.165127754211426,
|
3 |
+
"best_model_checkpoint": "/data1/attanasiog/babylm/roberta-tiny-2l-10M/checkpoint-4400",
|
4 |
+
"epoch": 94.78098908156711,
|
5 |
+
"global_step": 4550,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.21,
|
12 |
+
"learning_rate": 8e-05,
|
13 |
+
"loss": 10.5161,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.41,
|
18 |
+
"learning_rate": 0.00016,
|
19 |
+
"loss": 9.1097,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.62,
|
24 |
+
"learning_rate": 0.00024,
|
25 |
+
"loss": 7.8514,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.82,
|
30 |
+
"learning_rate": 0.00032,
|
31 |
+
"loss": 7.3238,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.04,
|
36 |
+
"learning_rate": 0.0004,
|
37 |
+
"loss": 7.7619,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 1.04,
|
42 |
+
"eval_accuracy": 0.07476398255519703,
|
43 |
+
"eval_loss": 7.233829975128174,
|
44 |
+
"eval_runtime": 127.6677,
|
45 |
+
"eval_samples_per_second": 188.419,
|
46 |
+
"eval_steps_per_second": 5.89,
|
47 |
+
"step": 50
|
48 |
+
},
|
49 |
+
{
|
50 |
+
"epoch": 1.25,
|
51 |
+
"learning_rate": 0.000399995625676045,
|
52 |
+
"loss": 7.1368,
|
53 |
+
"step": 60
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 1.45,
|
57 |
+
"learning_rate": 0.0003999825028955268,
|
58 |
+
"loss": 6.9374,
|
59 |
+
"step": 70
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 1.66,
|
63 |
+
"learning_rate": 0.0003999606322324786,
|
64 |
+
"loss": 6.8035,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 1.86,
|
69 |
+
"learning_rate": 0.0003999300146435939,
|
70 |
+
"loss": 6.6903,
|
71 |
+
"step": 90
|
72 |
+
},
|
73 |
+
{
|
74 |
+
"epoch": 2.08,
|
75 |
+
"learning_rate": 0.00039989065146818525,
|
76 |
+
"loss": 7.0524,
|
77 |
+
"step": 100
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 2.08,
|
81 |
+
"eval_accuracy": 0.1331419040979615,
|
82 |
+
"eval_loss": 6.625187873840332,
|
83 |
+
"eval_runtime": 127.7949,
|
84 |
+
"eval_samples_per_second": 188.231,
|
85 |
+
"eval_steps_per_second": 5.884,
|
86 |
+
"step": 100
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 2.29,
|
90 |
+
"learning_rate": 0.0003998425444281255,
|
91 |
+
"loss": 6.5528,
|
92 |
+
"step": 110
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 2.49,
|
96 |
+
"learning_rate": 0.00039978569562777234,
|
97 |
+
"loss": 6.5145,
|
98 |
+
"step": 120
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 2.7,
|
102 |
+
"learning_rate": 0.0003997201075538765,
|
103 |
+
"loss": 6.4642,
|
104 |
+
"step": 130
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 2.9,
|
108 |
+
"learning_rate": 0.0003996457830754729,
|
109 |
+
"loss": 6.4561,
|
110 |
+
"step": 140
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 3.12,
|
114 |
+
"learning_rate": 0.00039956272544375493,
|
115 |
+
"loss": 6.8423,
|
116 |
+
"step": 150
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 3.12,
|
120 |
+
"eval_accuracy": 0.14625706249076864,
|
121 |
+
"eval_loss": 6.462179183959961,
|
122 |
+
"eval_runtime": 127.8455,
|
123 |
+
"eval_samples_per_second": 188.157,
|
124 |
+
"eval_steps_per_second": 5.882,
|
125 |
+
"step": 150
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 3.33,
|
129 |
+
"learning_rate": 0.00039947093829193245,
|
130 |
+
"loss": 6.3841,
|
131 |
+
"step": 160
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 3.53,
|
135 |
+
"learning_rate": 0.00039937042563507283,
|
136 |
+
"loss": 6.3875,
|
137 |
+
"step": 170
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 3.74,
|
141 |
+
"learning_rate": 0.00039926119186992537,
|
142 |
+
"loss": 6.3843,
|
143 |
+
"step": 180
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 3.95,
|
147 |
+
"learning_rate": 0.0003991432417747288,
|
148 |
+
"loss": 6.3505,
|
149 |
+
"step": 190
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 4.16,
|
153 |
+
"learning_rate": 0.0003990165805090023,
|
154 |
+
"loss": 6.7298,
|
155 |
+
"step": 200
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 4.16,
|
159 |
+
"eval_accuracy": 0.1487513607434242,
|
160 |
+
"eval_loss": 6.397099018096924,
|
161 |
+
"eval_runtime": 127.9548,
|
162 |
+
"eval_samples_per_second": 187.996,
|
163 |
+
"eval_steps_per_second": 5.877,
|
164 |
+
"step": 200
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 4.37,
|
168 |
+
"learning_rate": 0.00039888121361332003,
|
169 |
+
"loss": 6.3075,
|
170 |
+
"step": 210
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 4.58,
|
174 |
+
"learning_rate": 0.0003987371470090686,
|
175 |
+
"loss": 6.3248,
|
176 |
+
"step": 220
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 4.78,
|
180 |
+
"learning_rate": 0.00039858438699818784,
|
181 |
+
"loss": 6.2949,
|
182 |
+
"step": 230
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 4.99,
|
186 |
+
"learning_rate": 0.0003984229402628956,
|
187 |
+
"loss": 6.2768,
|
188 |
+
"step": 240
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 5.21,
|
192 |
+
"learning_rate": 0.00039825281386539503,
|
193 |
+
"loss": 6.669,
|
194 |
+
"step": 250
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 5.21,
|
198 |
+
"eval_accuracy": 0.15192705186935002,
|
199 |
+
"eval_loss": 6.362815856933594,
|
200 |
+
"eval_runtime": 127.8986,
|
201 |
+
"eval_samples_per_second": 188.079,
|
202 |
+
"eval_steps_per_second": 5.88,
|
203 |
+
"step": 250
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 5.41,
|
207 |
+
"learning_rate": 0.000398074015247566,
|
208 |
+
"loss": 6.257,
|
209 |
+
"step": 260
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 5.62,
|
213 |
+
"learning_rate": 0.0003978865522306392,
|
214 |
+
"loss": 6.2485,
|
215 |
+
"step": 270
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 5.82,
|
219 |
+
"learning_rate": 0.0003976904330148543,
|
220 |
+
"loss": 6.252,
|
221 |
+
"step": 280
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 6.04,
|
225 |
+
"learning_rate": 0.00039748566617910113,
|
226 |
+
"loss": 6.6549,
|
227 |
+
"step": 290
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 6.25,
|
231 |
+
"learning_rate": 0.0003972722606805445,
|
232 |
+
"loss": 6.2038,
|
233 |
+
"step": 300
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 6.25,
|
237 |
+
"eval_accuracy": 0.15184154910887893,
|
238 |
+
"eval_loss": 6.337147235870361,
|
239 |
+
"eval_runtime": 128.0686,
|
240 |
+
"eval_samples_per_second": 187.829,
|
241 |
+
"eval_steps_per_second": 5.872,
|
242 |
+
"step": 300
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 6.45,
|
246 |
+
"learning_rate": 0.00039705022585423216,
|
247 |
+
"loss": 6.2262,
|
248 |
+
"step": 310
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 6.66,
|
252 |
+
"learning_rate": 0.0003968195714126868,
|
253 |
+
"loss": 6.2023,
|
254 |
+
"step": 320
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 6.86,
|
258 |
+
"learning_rate": 0.00039658030744548075,
|
259 |
+
"loss": 6.2053,
|
260 |
+
"step": 330
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 7.08,
|
264 |
+
"learning_rate": 0.0003963324444187952,
|
265 |
+
"loss": 6.6064,
|
266 |
+
"step": 340
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 7.29,
|
270 |
+
"learning_rate": 0.0003960759931749619,
|
271 |
+
"loss": 6.1783,
|
272 |
+
"step": 350
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 7.29,
|
276 |
+
"eval_accuracy": 0.15316276995162847,
|
277 |
+
"eval_loss": 6.311531066894531,
|
278 |
+
"eval_runtime": 127.9978,
|
279 |
+
"eval_samples_per_second": 187.933,
|
280 |
+
"eval_steps_per_second": 5.875,
|
281 |
+
"step": 350
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 7.49,
|
285 |
+
"learning_rate": 0.00039581096493198893,
|
286 |
+
"loss": 6.178,
|
287 |
+
"step": 360
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 7.7,
|
291 |
+
"learning_rate": 0.0003955373712830703,
|
292 |
+
"loss": 6.1784,
|
293 |
+
"step": 370
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 7.9,
|
297 |
+
"learning_rate": 0.00039525522419607854,
|
298 |
+
"loss": 6.1739,
|
299 |
+
"step": 380
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 8.12,
|
303 |
+
"learning_rate": 0.0003949645360130412,
|
304 |
+
"loss": 6.5644,
|
305 |
+
"step": 390
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 8.33,
|
309 |
+
"learning_rate": 0.0003946653194496012,
|
310 |
+
"loss": 6.1459,
|
311 |
+
"step": 400
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 8.33,
|
315 |
+
"eval_accuracy": 0.15298167344807118,
|
316 |
+
"eval_loss": 6.292238712310791,
|
317 |
+
"eval_runtime": 218.0011,
|
318 |
+
"eval_samples_per_second": 110.343,
|
319 |
+
"eval_steps_per_second": 3.45,
|
320 |
+
"step": 400
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 8.53,
|
324 |
+
"learning_rate": 0.00039435758759446025,
|
325 |
+
"loss": 6.1514,
|
326 |
+
"step": 410
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 8.74,
|
330 |
+
"learning_rate": 0.00039404135390880664,
|
331 |
+
"loss": 6.1335,
|
332 |
+
"step": 420
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 8.95,
|
336 |
+
"learning_rate": 0.0003937166322257262,
|
337 |
+
"loss": 6.1613,
|
338 |
+
"step": 430
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 9.16,
|
342 |
+
"learning_rate": 0.00039338343674959745,
|
343 |
+
"loss": 6.5555,
|
344 |
+
"step": 440
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 9.37,
|
348 |
+
"learning_rate": 0.00039304178205546976,
|
349 |
+
"loss": 6.1096,
|
350 |
+
"step": 450
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 9.37,
|
354 |
+
"eval_accuracy": 0.15364433508531855,
|
355 |
+
"eval_loss": 6.269557952880859,
|
356 |
+
"eval_runtime": 245.4855,
|
357 |
+
"eval_samples_per_second": 97.989,
|
358 |
+
"eval_steps_per_second": 3.063,
|
359 |
+
"step": 450
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 9.58,
|
363 |
+
"learning_rate": 0.00039269168308842634,
|
364 |
+
"loss": 6.1131,
|
365 |
+
"step": 460
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 9.78,
|
369 |
+
"learning_rate": 0.00039233315516293006,
|
370 |
+
"loss": 6.1172,
|
371 |
+
"step": 470
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 9.99,
|
375 |
+
"learning_rate": 0.00039196621396215403,
|
376 |
+
"loss": 6.0984,
|
377 |
+
"step": 480
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 10.21,
|
381 |
+
"learning_rate": 0.000391590875537295,
|
382 |
+
"loss": 6.494,
|
383 |
+
"step": 490
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 10.41,
|
387 |
+
"learning_rate": 0.00039120715630687155,
|
388 |
+
"loss": 6.0745,
|
389 |
+
"step": 500
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 10.41,
|
393 |
+
"eval_accuracy": 0.15413396308915142,
|
394 |
+
"eval_loss": 6.25447416305542,
|
395 |
+
"eval_runtime": 243.2903,
|
396 |
+
"eval_samples_per_second": 98.874,
|
397 |
+
"eval_steps_per_second": 3.091,
|
398 |
+
"step": 500
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 10.62,
|
402 |
+
"learning_rate": 0.000390815073056006,
|
403 |
+
"loss": 6.0953,
|
404 |
+
"step": 510
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 10.82,
|
408 |
+
"learning_rate": 0.00039041464293568983,
|
409 |
+
"loss": 6.0869,
|
410 |
+
"step": 520
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 11.04,
|
414 |
+
"learning_rate": 0.00039000588346203374,
|
415 |
+
"loss": 6.4846,
|
416 |
+
"step": 530
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 11.25,
|
420 |
+
"learning_rate": 0.0003895888125155014,
|
421 |
+
"loss": 6.0673,
|
422 |
+
"step": 540
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 11.45,
|
426 |
+
"learning_rate": 0.00038916344834012695,
|
427 |
+
"loss": 6.0689,
|
428 |
+
"step": 550
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 11.45,
|
432 |
+
"eval_accuracy": 0.15334541266013718,
|
433 |
+
"eval_loss": 6.24962854385376,
|
434 |
+
"eval_runtime": 238.4372,
|
435 |
+
"eval_samples_per_second": 100.886,
|
436 |
+
"eval_steps_per_second": 3.154,
|
437 |
+
"step": 550
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 11.66,
|
441 |
+
"learning_rate": 0.00038872980954271757,
|
442 |
+
"loss": 6.0805,
|
443 |
+
"step": 560
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 11.86,
|
447 |
+
"learning_rate": 0.00038828791509203895,
|
448 |
+
"loss": 6.0632,
|
449 |
+
"step": 570
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 12.08,
|
453 |
+
"learning_rate": 0.00038783778431798597,
|
454 |
+
"loss": 6.4656,
|
455 |
+
"step": 580
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 12.29,
|
459 |
+
"learning_rate": 0.0003873794369107369,
|
460 |
+
"loss": 6.0445,
|
461 |
+
"step": 590
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 12.49,
|
465 |
+
"learning_rate": 0.0003869128929198922,
|
466 |
+
"loss": 6.0562,
|
467 |
+
"step": 600
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 12.49,
|
471 |
+
"eval_accuracy": 0.15423183376396205,
|
472 |
+
"eval_loss": 6.231264114379883,
|
473 |
+
"eval_runtime": 244.3792,
|
474 |
+
"eval_samples_per_second": 98.433,
|
475 |
+
"eval_steps_per_second": 3.077,
|
476 |
+
"step": 600
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 12.7,
|
480 |
+
"learning_rate": 0.0003864381727535973,
|
481 |
+
"loss": 6.0553,
|
482 |
+
"step": 610
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 12.9,
|
486 |
+
"learning_rate": 0.00038595529717765027,
|
487 |
+
"loss": 6.0595,
|
488 |
+
"step": 620
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 13.12,
|
492 |
+
"learning_rate": 0.0003854642873145931,
|
493 |
+
"loss": 6.445,
|
494 |
+
"step": 630
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 13.33,
|
498 |
+
"learning_rate": 0.00038496516464278776,
|
499 |
+
"loss": 6.0285,
|
500 |
+
"step": 640
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 13.53,
|
504 |
+
"learning_rate": 0.00038445795099547697,
|
505 |
+
"loss": 6.0324,
|
506 |
+
"step": 650
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 13.53,
|
510 |
+
"eval_accuracy": 0.15358873091243086,
|
511 |
+
"eval_loss": 6.224751949310303,
|
512 |
+
"eval_runtime": 128.0464,
|
513 |
+
"eval_samples_per_second": 187.862,
|
514 |
+
"eval_steps_per_second": 5.873,
|
515 |
+
"step": 650
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 13.74,
|
519 |
+
"learning_rate": 0.0003839426685598287,
|
520 |
+
"loss": 6.0219,
|
521 |
+
"step": 660
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 13.95,
|
525 |
+
"learning_rate": 0.000383419339875966,
|
526 |
+
"loss": 6.0379,
|
527 |
+
"step": 670
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 14.16,
|
531 |
+
"learning_rate": 0.00038288798783598087,
|
532 |
+
"loss": 6.416,
|
533 |
+
"step": 680
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 14.37,
|
537 |
+
"learning_rate": 0.0003823486356829329,
|
538 |
+
"loss": 5.9984,
|
539 |
+
"step": 690
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 14.58,
|
543 |
+
"learning_rate": 0.0003818013070098325,
|
544 |
+
"loss": 5.9907,
|
545 |
+
"step": 700
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 14.58,
|
549 |
+
"eval_accuracy": 0.15438304013536042,
|
550 |
+
"eval_loss": 6.217936038970947,
|
551 |
+
"eval_runtime": 128.0604,
|
552 |
+
"eval_samples_per_second": 187.841,
|
553 |
+
"eval_steps_per_second": 5.872,
|
554 |
+
"step": 700
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 14.78,
|
558 |
+
"learning_rate": 0.0003812460257586089,
|
559 |
+
"loss": 6.0038,
|
560 |
+
"step": 710
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 14.99,
|
564 |
+
"learning_rate": 0.000380682816219063,
|
565 |
+
"loss": 6.0321,
|
566 |
+
"step": 720
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 15.21,
|
570 |
+
"learning_rate": 0.00038011170302780446,
|
571 |
+
"loss": 6.3685,
|
572 |
+
"step": 730
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 15.41,
|
576 |
+
"learning_rate": 0.00037953271116717444,
|
577 |
+
"loss": 5.9825,
|
578 |
+
"step": 740
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 15.62,
|
582 |
+
"learning_rate": 0.0003789458659641527,
|
583 |
+
"loss": 5.9683,
|
584 |
+
"step": 750
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 15.62,
|
588 |
+
"eval_accuracy": 0.1545422454380471,
|
589 |
+
"eval_loss": 6.183169364929199,
|
590 |
+
"eval_runtime": 128.0339,
|
591 |
+
"eval_samples_per_second": 187.88,
|
592 |
+
"eval_steps_per_second": 5.873,
|
593 |
+
"step": 750
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 15.82,
|
597 |
+
"learning_rate": 0.0003783511930892495,
|
598 |
+
"loss": 5.9712,
|
599 |
+
"step": 760
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 16.04,
|
603 |
+
"learning_rate": 0.00037774871855538275,
|
604 |
+
"loss": 6.355,
|
605 |
+
"step": 770
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 16.25,
|
609 |
+
"learning_rate": 0.00037713846871674045,
|
610 |
+
"loss": 5.9361,
|
611 |
+
"step": 780
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 16.45,
|
615 |
+
"learning_rate": 0.0003765204702676274,
|
616 |
+
"loss": 5.9281,
|
617 |
+
"step": 790
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 16.66,
|
621 |
+
"learning_rate": 0.0003758947502412978,
|
622 |
+
"loss": 5.9236,
|
623 |
+
"step": 800
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 16.66,
|
627 |
+
"eval_accuracy": 0.15502017774172816,
|
628 |
+
"eval_loss": 6.141255855560303,
|
629 |
+
"eval_runtime": 128.0268,
|
630 |
+
"eval_samples_per_second": 187.89,
|
631 |
+
"eval_steps_per_second": 5.874,
|
632 |
+
"step": 800
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 16.86,
|
636 |
+
"learning_rate": 0.0003752613360087727,
|
637 |
+
"loss": 5.9288,
|
638 |
+
"step": 810
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 17.08,
|
642 |
+
"learning_rate": 0.00037462025527764265,
|
643 |
+
"loss": 6.2842,
|
644 |
+
"step": 820
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 17.29,
|
648 |
+
"learning_rate": 0.00037397153609085553,
|
649 |
+
"loss": 5.8852,
|
650 |
+
"step": 830
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 17.49,
|
654 |
+
"learning_rate": 0.0003733152068254901,
|
655 |
+
"loss": 5.8779,
|
656 |
+
"step": 840
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 17.7,
|
660 |
+
"learning_rate": 0.00037265129619151483,
|
661 |
+
"loss": 5.8808,
|
662 |
+
"step": 850
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 17.7,
|
666 |
+
"eval_accuracy": 0.15577406052421716,
|
667 |
+
"eval_loss": 6.089950084686279,
|
668 |
+
"eval_runtime": 128.2202,
|
669 |
+
"eval_samples_per_second": 187.607,
|
670 |
+
"eval_steps_per_second": 5.865,
|
671 |
+
"step": 850
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 17.9,
|
675 |
+
"learning_rate": 0.00037197983323053143,
|
676 |
+
"loss": 5.871,
|
677 |
+
"step": 860
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 18.12,
|
681 |
+
"learning_rate": 0.00037130084731450515,
|
682 |
+
"loss": 6.2293,
|
683 |
+
"step": 870
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 18.33,
|
687 |
+
"learning_rate": 0.0003706143681444795,
|
688 |
+
"loss": 5.8282,
|
689 |
+
"step": 880
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 18.53,
|
693 |
+
"learning_rate": 0.0003699204257492774,
|
694 |
+
"loss": 5.85,
|
695 |
+
"step": 890
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 18.74,
|
699 |
+
"learning_rate": 0.0003692190504841871,
|
700 |
+
"loss": 5.8392,
|
701 |
+
"step": 900
|
702 |
+
},
|
703 |
+
{
|
704 |
+
"epoch": 18.74,
|
705 |
+
"eval_accuracy": 0.15657369591332176,
|
706 |
+
"eval_loss": 6.054327487945557,
|
707 |
+
"eval_runtime": 128.0437,
|
708 |
+
"eval_samples_per_second": 187.866,
|
709 |
+
"eval_steps_per_second": 5.873,
|
710 |
+
"step": 900
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 18.95,
|
714 |
+
"learning_rate": 0.00036851027302963493,
|
715 |
+
"loss": 5.8393,
|
716 |
+
"step": 910
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 19.16,
|
720 |
+
"learning_rate": 0.00036779412438984294,
|
721 |
+
"loss": 6.1961,
|
722 |
+
"step": 920
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 19.37,
|
726 |
+
"learning_rate": 0.0003670706358914725,
|
727 |
+
"loss": 5.8161,
|
728 |
+
"step": 930
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 19.58,
|
732 |
+
"learning_rate": 0.0003663398391822543,
|
733 |
+
"loss": 5.7886,
|
734 |
+
"step": 940
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 19.78,
|
738 |
+
"learning_rate": 0.00036560176622960403,
|
739 |
+
"loss": 5.7962,
|
740 |
+
"step": 950
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 19.78,
|
744 |
+
"eval_accuracy": 0.15750512966626293,
|
745 |
+
"eval_loss": 6.022204399108887,
|
746 |
+
"eval_runtime": 127.9436,
|
747 |
+
"eval_samples_per_second": 188.012,
|
748 |
+
"eval_steps_per_second": 5.878,
|
749 |
+
"step": 950
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 19.99,
|
753 |
+
"learning_rate": 0.00036485644931922353,
|
754 |
+
"loss": 5.7823,
|
755 |
+
"step": 960
|
756 |
+
},
|
757 |
+
{
|
758 |
+
"epoch": 20.21,
|
759 |
+
"learning_rate": 0.0003641039210536889,
|
760 |
+
"loss": 6.1533,
|
761 |
+
"step": 970
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 20.41,
|
765 |
+
"learning_rate": 0.0003633442143510245,
|
766 |
+
"loss": 5.7526,
|
767 |
+
"step": 980
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 20.62,
|
771 |
+
"learning_rate": 0.00036257736244326246,
|
772 |
+
"loss": 5.7454,
|
773 |
+
"step": 990
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 20.82,
|
777 |
+
"learning_rate": 0.0003618033988749895,
|
778 |
+
"loss": 5.7473,
|
779 |
+
"step": 1000
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 20.82,
|
783 |
+
"eval_accuracy": 0.16172566383651218,
|
784 |
+
"eval_loss": 5.947088718414307,
|
785 |
+
"eval_runtime": 128.1904,
|
786 |
+
"eval_samples_per_second": 187.65,
|
787 |
+
"eval_steps_per_second": 5.866,
|
788 |
+
"step": 1000
|
789 |
+
},
|
790 |
+
{
|
791 |
+
"epoch": 21.04,
|
792 |
+
"learning_rate": 0.0003610223575018795,
|
793 |
+
"loss": 6.0948,
|
794 |
+
"step": 1010
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 21.25,
|
798 |
+
"learning_rate": 0.00036023427248921215,
|
799 |
+
"loss": 5.6776,
|
800 |
+
"step": 1020
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 21.45,
|
804 |
+
"learning_rate": 0.0003594391783103792,
|
805 |
+
"loss": 5.6479,
|
806 |
+
"step": 1030
|
807 |
+
},
|
808 |
+
{
|
809 |
+
"epoch": 21.66,
|
810 |
+
"learning_rate": 0.00035863710974537563,
|
811 |
+
"loss": 5.6245,
|
812 |
+
"step": 1040
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 21.86,
|
816 |
+
"learning_rate": 0.00035782810187927875,
|
817 |
+
"loss": 5.5787,
|
818 |
+
"step": 1050
|
819 |
+
},
|
820 |
+
{
|
821 |
+
"epoch": 21.86,
|
822 |
+
"eval_accuracy": 0.18910485199927482,
|
823 |
+
"eval_loss": 5.7037835121154785,
|
824 |
+
"eval_runtime": 128.2758,
|
825 |
+
"eval_samples_per_second": 187.526,
|
826 |
+
"eval_steps_per_second": 5.862,
|
827 |
+
"step": 1050
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 22.08,
|
831 |
+
"learning_rate": 0.0003570121901007136,
|
832 |
+
"loss": 5.8678,
|
833 |
+
"step": 1060
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 22.29,
|
837 |
+
"learning_rate": 0.0003561894101003044,
|
838 |
+
"loss": 5.4587,
|
839 |
+
"step": 1070
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 22.49,
|
843 |
+
"learning_rate": 0.00035535979786911396,
|
844 |
+
"loss": 5.3982,
|
845 |
+
"step": 1080
|
846 |
+
},
|
847 |
+
{
|
848 |
+
"epoch": 22.7,
|
849 |
+
"learning_rate": 0.00035452338969706876,
|
850 |
+
"loss": 5.305,
|
851 |
+
"step": 1090
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 22.9,
|
855 |
+
"learning_rate": 0.00035368022217137184,
|
856 |
+
"loss": 5.2316,
|
857 |
+
"step": 1100
|
858 |
+
},
|
859 |
+
{
|
860 |
+
"epoch": 22.9,
|
861 |
+
"eval_accuracy": 0.23819718901662149,
|
862 |
+
"eval_loss": 5.270751476287842,
|
863 |
+
"eval_runtime": 128.1695,
|
864 |
+
"eval_samples_per_second": 187.681,
|
865 |
+
"eval_steps_per_second": 5.867,
|
866 |
+
"step": 1100
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 23.12,
|
870 |
+
"learning_rate": 0.00035283033217490227,
|
871 |
+
"loss": 5.4202,
|
872 |
+
"step": 1110
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 23.33,
|
876 |
+
"learning_rate": 0.00035197375688460176,
|
877 |
+
"loss": 4.9911,
|
878 |
+
"step": 1120
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 23.53,
|
882 |
+
"learning_rate": 0.0003511105337698484,
|
883 |
+
"loss": 4.8741,
|
884 |
+
"step": 1130
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 23.74,
|
888 |
+
"learning_rate": 0.0003502407005908177,
|
889 |
+
"loss": 4.7582,
|
890 |
+
"step": 1140
|
891 |
+
},
|
892 |
+
{
|
893 |
+
"epoch": 23.95,
|
894 |
+
"learning_rate": 0.0003493642953968308,
|
895 |
+
"loss": 4.6613,
|
896 |
+
"step": 1150
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 23.95,
|
900 |
+
"eval_accuracy": 0.29748286605712254,
|
901 |
+
"eval_loss": 4.707459926605225,
|
902 |
+
"eval_runtime": 131.3634,
|
903 |
+
"eval_samples_per_second": 183.118,
|
904 |
+
"eval_steps_per_second": 5.725,
|
905 |
+
"step": 1150
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 24.16,
|
909 |
+
"learning_rate": 0.00034848135652469,
|
910 |
+
"loss": 4.8536,
|
911 |
+
"step": 1160
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 24.37,
|
915 |
+
"learning_rate": 0.00034759192259700196,
|
916 |
+
"loss": 4.4822,
|
917 |
+
"step": 1170
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 24.58,
|
921 |
+
"learning_rate": 0.000346696032520488,
|
922 |
+
"loss": 4.4126,
|
923 |
+
"step": 1180
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 24.78,
|
927 |
+
"learning_rate": 0.00034579372548428235,
|
928 |
+
"loss": 4.3707,
|
929 |
+
"step": 1190
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 24.99,
|
933 |
+
"learning_rate": 0.00034488504095821784,
|
934 |
+
"loss": 4.3006,
|
935 |
+
"step": 1200
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 24.99,
|
939 |
+
"eval_accuracy": 0.3221731878424314,
|
940 |
+
"eval_loss": 4.417978763580322,
|
941 |
+
"eval_runtime": 144.7447,
|
942 |
+
"eval_samples_per_second": 166.189,
|
943 |
+
"eval_steps_per_second": 5.195,
|
944 |
+
"step": 1200
|
945 |
+
},
|
946 |
+
{
|
947 |
+
"epoch": 25.21,
|
948 |
+
"learning_rate": 0.0003439700186910993,
|
949 |
+
"loss": 4.5185,
|
950 |
+
"step": 1210
|
951 |
+
},
|
952 |
+
{
|
953 |
+
"epoch": 25.41,
|
954 |
+
"learning_rate": 0.00034304869870896513,
|
955 |
+
"loss": 4.2011,
|
956 |
+
"step": 1220
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 25.62,
|
960 |
+
"learning_rate": 0.00034212112131333587,
|
961 |
+
"loss": 4.1513,
|
962 |
+
"step": 1230
|
963 |
+
},
|
964 |
+
{
|
965 |
+
"epoch": 25.82,
|
966 |
+
"learning_rate": 0.0003411873270794518,
|
967 |
+
"loss": 4.1584,
|
968 |
+
"step": 1240
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 26.04,
|
972 |
+
"learning_rate": 0.00034024735685449773,
|
973 |
+
"loss": 4.3754,
|
974 |
+
"step": 1250
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 26.04,
|
978 |
+
"eval_accuracy": 0.33853883405739793,
|
979 |
+
"eval_loss": 4.238345146179199,
|
980 |
+
"eval_runtime": 144.4875,
|
981 |
+
"eval_samples_per_second": 166.485,
|
982 |
+
"eval_steps_per_second": 5.205,
|
983 |
+
"step": 1250
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 26.25,
|
987 |
+
"learning_rate": 0.00033930125175581647,
|
988 |
+
"loss": 4.0477,
|
989 |
+
"step": 1260
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 26.45,
|
993 |
+
"learning_rate": 0.0003383490531691099,
|
994 |
+
"loss": 4.0339,
|
995 |
+
"step": 1270
|
996 |
+
},
|
997 |
+
{
|
998 |
+
"epoch": 26.66,
|
999 |
+
"learning_rate": 0.0003373908027466289,
|
1000 |
+
"loss": 4.0184,
|
1001 |
+
"step": 1280
|
1002 |
+
},
|
1003 |
+
{
|
1004 |
+
"epoch": 26.86,
|
1005 |
+
"learning_rate": 0.00033642654240535134,
|
1006 |
+
"loss": 3.9835,
|
1007 |
+
"step": 1290
|
1008 |
+
},
|
1009 |
+
{
|
1010 |
+
"epoch": 27.08,
|
1011 |
+
"learning_rate": 0.00033545631432514825,
|
1012 |
+
"loss": 4.2531,
|
1013 |
+
"step": 1300
|
1014 |
+
},
|
1015 |
+
{
|
1016 |
+
"epoch": 27.08,
|
1017 |
+
"eval_accuracy": 0.34910193403738843,
|
1018 |
+
"eval_loss": 4.1157379150390625,
|
1019 |
+
"eval_runtime": 144.9052,
|
1020 |
+
"eval_samples_per_second": 166.005,
|
1021 |
+
"eval_steps_per_second": 5.19,
|
1022 |
+
"step": 1300
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 27.29,
|
1026 |
+
"learning_rate": 0.00033448016094693895,
|
1027 |
+
"loss": 3.9085,
|
1028 |
+
"step": 1310
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 27.49,
|
1032 |
+
"learning_rate": 0.0003334981249708345,
|
1033 |
+
"loss": 3.9205,
|
1034 |
+
"step": 1320
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 27.7,
|
1038 |
+
"learning_rate": 0.00033251024935427,
|
1039 |
+
"loss": 3.8786,
|
1040 |
+
"step": 1330
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 27.9,
|
1044 |
+
"learning_rate": 0.0003315165773101249,
|
1045 |
+
"loss": 3.8839,
|
1046 |
+
"step": 1340
|
1047 |
+
},
|
1048 |
+
{
|
1049 |
+
"epoch": 28.12,
|
1050 |
+
"learning_rate": 0.00033051715230483374,
|
1051 |
+
"loss": 4.0987,
|
1052 |
+
"step": 1350
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 28.12,
|
1056 |
+
"eval_accuracy": 0.3577664889956034,
|
1057 |
+
"eval_loss": 4.0197062492370605,
|
1058 |
+
"eval_runtime": 128.1918,
|
1059 |
+
"eval_samples_per_second": 187.649,
|
1060 |
+
"eval_steps_per_second": 5.866,
|
1061 |
+
"step": 1350
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 28.33,
|
1065 |
+
"learning_rate": 0.0003295120180564838,
|
1066 |
+
"loss": 3.8043,
|
1067 |
+
"step": 1360
|
1068 |
+
},
|
1069 |
+
{
|
1070 |
+
"epoch": 28.53,
|
1071 |
+
"learning_rate": 0.00032850121853290334,
|
1072 |
+
"loss": 3.8048,
|
1073 |
+
"step": 1370
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 28.74,
|
1077 |
+
"learning_rate": 0.000327484797949738,
|
1078 |
+
"loss": 3.8127,
|
1079 |
+
"step": 1380
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 28.95,
|
1083 |
+
"learning_rate": 0.00032646280076851684,
|
1084 |
+
"loss": 3.7773,
|
1085 |
+
"step": 1390
|
1086 |
+
},
|
1087 |
+
{
|
1088 |
+
"epoch": 29.16,
|
1089 |
+
"learning_rate": 0.0003254352716947074,
|
1090 |
+
"loss": 4.0045,
|
1091 |
+
"step": 1400
|
1092 |
+
},
|
1093 |
+
{
|
1094 |
+
"epoch": 29.16,
|
1095 |
+
"eval_accuracy": 0.3656404849021274,
|
1096 |
+
"eval_loss": 3.9503591060638428,
|
1097 |
+
"eval_runtime": 128.0514,
|
1098 |
+
"eval_samples_per_second": 187.854,
|
1099 |
+
"eval_steps_per_second": 5.873,
|
1100 |
+
"step": 1400
|
1101 |
+
},
|
1102 |
+
{
|
1103 |
+
"epoch": 29.37,
|
1104 |
+
"learning_rate": 0.0003244022556757602,
|
1105 |
+
"loss": 3.7454,
|
1106 |
+
"step": 1410
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 29.58,
|
1110 |
+
"learning_rate": 0.0003233637978991422,
|
1111 |
+
"loss": 3.7151,
|
1112 |
+
"step": 1420
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 29.78,
|
1116 |
+
"learning_rate": 0.00032231994379036086,
|
1117 |
+
"loss": 3.7243,
|
1118 |
+
"step": 1430
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 29.99,
|
1122 |
+
"learning_rate": 0.0003212707390109765,
|
1123 |
+
"loss": 3.691,
|
1124 |
+
"step": 1440
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 30.21,
|
1128 |
+
"learning_rate": 0.00032021622945660504,
|
1129 |
+
"loss": 3.9145,
|
1130 |
+
"step": 1450
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 30.21,
|
1134 |
+
"eval_accuracy": 0.37178706581714294,
|
1135 |
+
"eval_loss": 3.8818981647491455,
|
1136 |
+
"eval_runtime": 128.0667,
|
1137 |
+
"eval_samples_per_second": 187.832,
|
1138 |
+
"eval_steps_per_second": 5.872,
|
1139 |
+
"step": 1450
|
1140 |
+
},
|
1141 |
+
{
|
1142 |
+
"epoch": 30.41,
|
1143 |
+
"learning_rate": 0.0003191564612549106,
|
1144 |
+
"loss": 3.6485,
|
1145 |
+
"step": 1460
|
1146 |
+
},
|
1147 |
+
{
|
1148 |
+
"epoch": 30.62,
|
1149 |
+
"learning_rate": 0.0003180914807635874,
|
1150 |
+
"loss": 3.6517,
|
1151 |
+
"step": 1470
|
1152 |
+
},
|
1153 |
+
{
|
1154 |
+
"epoch": 30.82,
|
1155 |
+
"learning_rate": 0.00031702133456833236,
|
1156 |
+
"loss": 3.6453,
|
1157 |
+
"step": 1480
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 31.04,
|
1161 |
+
"learning_rate": 0.00031594606948080663,
|
1162 |
+
"loss": 3.8657,
|
1163 |
+
"step": 1490
|
1164 |
+
},
|
1165 |
+
{
|
1166 |
+
"epoch": 31.25,
|
1167 |
+
"learning_rate": 0.00031486573253658874,
|
1168 |
+
"loss": 3.5808,
|
1169 |
+
"step": 1500
|
1170 |
+
},
|
1171 |
+
{
|
1172 |
+
"epoch": 31.25,
|
1173 |
+
"eval_accuracy": 0.3780561563488311,
|
1174 |
+
"eval_loss": 3.827902317047119,
|
1175 |
+
"eval_runtime": 128.1177,
|
1176 |
+
"eval_samples_per_second": 187.757,
|
1177 |
+
"eval_steps_per_second": 5.87,
|
1178 |
+
"step": 1500
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 31.45,
|
1182 |
+
"learning_rate": 0.00031378037099311627,
|
1183 |
+
"loss": 3.5983,
|
1184 |
+
"step": 1510
|
1185 |
+
},
|
1186 |
+
{
|
1187 |
+
"epoch": 31.66,
|
1188 |
+
"learning_rate": 0.00031269003232761933,
|
1189 |
+
"loss": 3.5947,
|
1190 |
+
"step": 1520
|
1191 |
+
},
|
1192 |
+
{
|
1193 |
+
"epoch": 31.86,
|
1194 |
+
"learning_rate": 0.0003115947642350433,
|
1195 |
+
"loss": 3.5773,
|
1196 |
+
"step": 1530
|
1197 |
+
},
|
1198 |
+
{
|
1199 |
+
"epoch": 32.08,
|
1200 |
+
"learning_rate": 0.00031049461462596267,
|
1201 |
+
"loss": 3.7941,
|
1202 |
+
"step": 1540
|
1203 |
+
},
|
1204 |
+
{
|
1205 |
+
"epoch": 32.29,
|
1206 |
+
"learning_rate": 0.00030938963162448544,
|
1207 |
+
"loss": 3.5354,
|
1208 |
+
"step": 1550
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 32.29,
|
1212 |
+
"eval_accuracy": 0.3825753450782098,
|
1213 |
+
"eval_loss": 3.7829582691192627,
|
1214 |
+
"eval_runtime": 128.1881,
|
1215 |
+
"eval_samples_per_second": 187.654,
|
1216 |
+
"eval_steps_per_second": 5.866,
|
1217 |
+
"step": 1550
|
1218 |
+
},
|
1219 |
+
{
|
1220 |
+
"epoch": 32.49,
|
1221 |
+
"learning_rate": 0.0003082798635661476,
|
1222 |
+
"loss": 3.5355,
|
1223 |
+
"step": 1560
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 32.7,
|
1227 |
+
"learning_rate": 0.0003071653589957993,
|
1228 |
+
"loss": 3.5166,
|
1229 |
+
"step": 1570
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 32.9,
|
1233 |
+
"learning_rate": 0.000306046166665481,
|
1234 |
+
"loss": 3.5295,
|
1235 |
+
"step": 1580
|
1236 |
+
},
|
1237 |
+
{
|
1238 |
+
"epoch": 33.12,
|
1239 |
+
"learning_rate": 0.00030492233553229076,
|
1240 |
+
"loss": 3.7281,
|
1241 |
+
"step": 1590
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 33.33,
|
1245 |
+
"learning_rate": 0.00030379391475624304,
|
1246 |
+
"loss": 3.4788,
|
1247 |
+
"step": 1600
|
1248 |
+
},
|
1249 |
+
{
|
1250 |
+
"epoch": 33.33,
|
1251 |
+
"eval_accuracy": 0.3872195585541312,
|
1252 |
+
"eval_loss": 3.7400357723236084,
|
1253 |
+
"eval_runtime": 128.395,
|
1254 |
+
"eval_samples_per_second": 187.352,
|
1255 |
+
"eval_steps_per_second": 5.857,
|
1256 |
+
"step": 1600
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 33.53,
|
1260 |
+
"learning_rate": 0.0003026609536981183,
|
1261 |
+
"loss": 3.4722,
|
1262 |
+
"step": 1610
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 33.74,
|
1266 |
+
"learning_rate": 0.0003015235019173034,
|
1267 |
+
"loss": 3.4888,
|
1268 |
+
"step": 1620
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 33.95,
|
1272 |
+
"learning_rate": 0.00030038160916962404,
|
1273 |
+
"loss": 3.472,
|
1274 |
+
"step": 1630
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 34.16,
|
1278 |
+
"learning_rate": 0.00029923532540516843,
|
1279 |
+
"loss": 3.6802,
|
1280 |
+
"step": 1640
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 34.37,
|
1284 |
+
"learning_rate": 0.00029808470076610167,
|
1285 |
+
"loss": 3.4315,
|
1286 |
+
"step": 1650
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 34.37,
|
1290 |
+
"eval_accuracy": 0.39113526163842305,
|
1291 |
+
"eval_loss": 3.702760696411133,
|
1292 |
+
"eval_runtime": 147.8935,
|
1293 |
+
"eval_samples_per_second": 162.651,
|
1294 |
+
"eval_steps_per_second": 5.085,
|
1295 |
+
"step": 1650
|
1296 |
+
},
|
1297 |
+
{
|
1298 |
+
"epoch": 34.58,
|
1299 |
+
"learning_rate": 0.00029692978558447305,
|
1300 |
+
"loss": 3.4258,
|
1301 |
+
"step": 1660
|
1302 |
+
},
|
1303 |
+
{
|
1304 |
+
"epoch": 34.78,
|
1305 |
+
"learning_rate": 0.0002957706303800139,
|
1306 |
+
"loss": 3.4325,
|
1307 |
+
"step": 1670
|
1308 |
+
},
|
1309 |
+
{
|
1310 |
+
"epoch": 34.99,
|
1311 |
+
"learning_rate": 0.0002946072858579282,
|
1312 |
+
"loss": 3.418,
|
1313 |
+
"step": 1680
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 35.21,
|
1317 |
+
"learning_rate": 0.0002934398029066739,
|
1318 |
+
"loss": 3.6162,
|
1319 |
+
"step": 1690
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 35.41,
|
1323 |
+
"learning_rate": 0.0002922682325957376,
|
1324 |
+
"loss": 3.3906,
|
1325 |
+
"step": 1700
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 35.41,
|
1329 |
+
"eval_accuracy": 0.3955544302044244,
|
1330 |
+
"eval_loss": 3.6628527641296387,
|
1331 |
+
"eval_runtime": 147.4398,
|
1332 |
+
"eval_samples_per_second": 163.151,
|
1333 |
+
"eval_steps_per_second": 5.1,
|
1334 |
+
"step": 1700
|
1335 |
+
},
|
1336 |
+
{
|
1337 |
+
"epoch": 35.62,
|
1338 |
+
"learning_rate": 0.00029109262617339987,
|
1339 |
+
"loss": 3.3731,
|
1340 |
+
"step": 1710
|
1341 |
+
},
|
1342 |
+
{
|
1343 |
+
"epoch": 35.82,
|
1344 |
+
"learning_rate": 0.0002899130350644941,
|
1345 |
+
"loss": 3.381,
|
1346 |
+
"step": 1720
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 36.04,
|
1350 |
+
"learning_rate": 0.00028872951086815685,
|
1351 |
+
"loss": 3.608,
|
1352 |
+
"step": 1730
|
1353 |
+
},
|
1354 |
+
{
|
1355 |
+
"epoch": 36.25,
|
1356 |
+
"learning_rate": 0.00028754210535557036,
|
1357 |
+
"loss": 3.3345,
|
1358 |
+
"step": 1740
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 36.45,
|
1362 |
+
"learning_rate": 0.00028635087046769857,
|
1363 |
+
"loss": 3.3508,
|
1364 |
+
"step": 1750
|
1365 |
+
},
|
1366 |
+
{
|
1367 |
+
"epoch": 36.45,
|
1368 |
+
"eval_accuracy": 0.3984451691466028,
|
1369 |
+
"eval_loss": 3.6344361305236816,
|
1370 |
+
"eval_runtime": 147.716,
|
1371 |
+
"eval_samples_per_second": 162.846,
|
1372 |
+
"eval_steps_per_second": 5.091,
|
1373 |
+
"step": 1750
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 36.66,
|
1377 |
+
"learning_rate": 0.00028515585831301456,
|
1378 |
+
"loss": 3.3503,
|
1379 |
+
"step": 1760
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 36.86,
|
1383 |
+
"learning_rate": 0.0002839571211652212,
|
1384 |
+
"loss": 3.3494,
|
1385 |
+
"step": 1770
|
1386 |
+
},
|
1387 |
+
{
|
1388 |
+
"epoch": 37.08,
|
1389 |
+
"learning_rate": 0.00028275471146096466,
|
1390 |
+
"loss": 3.539,
|
1391 |
+
"step": 1780
|
1392 |
+
},
|
1393 |
+
{
|
1394 |
+
"epoch": 37.29,
|
1395 |
+
"learning_rate": 0.00028154868179754074,
|
1396 |
+
"loss": 3.3145,
|
1397 |
+
"step": 1790
|
1398 |
+
},
|
1399 |
+
{
|
1400 |
+
"epoch": 37.49,
|
1401 |
+
"learning_rate": 0.0002803390849305939,
|
1402 |
+
"loss": 3.288,
|
1403 |
+
"step": 1800
|
1404 |
+
},
|
1405 |
+
{
|
1406 |
+
"epoch": 37.49,
|
1407 |
+
"eval_accuracy": 0.4019129197606658,
|
1408 |
+
"eval_loss": 3.6045737266540527,
|
1409 |
+
"eval_runtime": 128.0819,
|
1410 |
+
"eval_samples_per_second": 187.81,
|
1411 |
+
"eval_steps_per_second": 5.871,
|
1412 |
+
"step": 1800
|
1413 |
+
},
|
1414 |
+
{
|
1415 |
+
"epoch": 37.7,
|
1416 |
+
"learning_rate": 0.0002791259737718097,
|
1417 |
+
"loss": 3.318,
|
1418 |
+
"step": 1810
|
1419 |
+
},
|
1420 |
+
{
|
1421 |
+
"epoch": 37.9,
|
1422 |
+
"learning_rate": 0.0002779094013866001,
|
1423 |
+
"loss": 3.3005,
|
1424 |
+
"step": 1820
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 38.12,
|
1428 |
+
"learning_rate": 0.00027668942099178234,
|
1429 |
+
"loss": 3.4959,
|
1430 |
+
"step": 1830
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 38.33,
|
1434 |
+
"learning_rate": 0.00027546608595325117,
|
1435 |
+
"loss": 3.2771,
|
1436 |
+
"step": 1840
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 38.53,
|
1440 |
+
"learning_rate": 0.00027423944978364416,
|
1441 |
+
"loss": 3.2678,
|
1442 |
+
"step": 1850
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 38.53,
|
1446 |
+
"eval_accuracy": 0.40528409278500616,
|
1447 |
+
"eval_loss": 3.5798938274383545,
|
1448 |
+
"eval_runtime": 128.148,
|
1449 |
+
"eval_samples_per_second": 187.713,
|
1450 |
+
"eval_steps_per_second": 5.868,
|
1451 |
+
"step": 1850
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 38.74,
|
1455 |
+
"learning_rate": 0.00027300956614000115,
|
1456 |
+
"loss": 3.2707,
|
1457 |
+
"step": 1860
|
1458 |
+
},
|
1459 |
+
{
|
1460 |
+
"epoch": 38.95,
|
1461 |
+
"learning_rate": 0.00027177648882141704,
|
1462 |
+
"loss": 3.276,
|
1463 |
+
"step": 1870
|
1464 |
+
},
|
1465 |
+
{
|
1466 |
+
"epoch": 39.16,
|
1467 |
+
"learning_rate": 0.0002705402717666883,
|
1468 |
+
"loss": 3.4633,
|
1469 |
+
"step": 1880
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 39.37,
|
1473 |
+
"learning_rate": 0.00026930096905195363,
|
1474 |
+
"loss": 3.2392,
|
1475 |
+
"step": 1890
|
1476 |
+
},
|
1477 |
+
{
|
1478 |
+
"epoch": 39.58,
|
1479 |
+
"learning_rate": 0.00026805863488832865,
|
1480 |
+
"loss": 3.2382,
|
1481 |
+
"step": 1900
|
1482 |
+
},
|
1483 |
+
{
|
1484 |
+
"epoch": 39.58,
|
1485 |
+
"eval_accuracy": 0.40741369917263087,
|
1486 |
+
"eval_loss": 3.5548534393310547,
|
1487 |
+
"eval_runtime": 128.15,
|
1488 |
+
"eval_samples_per_second": 187.71,
|
1489 |
+
"eval_steps_per_second": 5.868,
|
1490 |
+
"step": 1900
|
1491 |
+
},
|
1492 |
+
{
|
1493 |
+
"epoch": 39.78,
|
1494 |
+
"learning_rate": 0.00026681332361953424,
|
1495 |
+
"loss": 3.2304,
|
1496 |
+
"step": 1910
|
1497 |
+
},
|
1498 |
+
{
|
1499 |
+
"epoch": 39.99,
|
1500 |
+
"learning_rate": 0.0002655650897195195,
|
1501 |
+
"loss": 3.24,
|
1502 |
+
"step": 1920
|
1503 |
+
},
|
1504 |
+
{
|
1505 |
+
"epoch": 40.21,
|
1506 |
+
"learning_rate": 0.0002643139877900791,
|
1507 |
+
"loss": 3.4143,
|
1508 |
+
"step": 1930
|
1509 |
+
},
|
1510 |
+
{
|
1511 |
+
"epoch": 40.41,
|
1512 |
+
"learning_rate": 0.00026306007255846436,
|
1513 |
+
"loss": 3.203,
|
1514 |
+
"step": 1940
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 40.62,
|
1518 |
+
"learning_rate": 0.00026180339887498953,
|
1519 |
+
"loss": 3.2151,
|
1520 |
+
"step": 1950
|
1521 |
+
},
|
1522 |
+
{
|
1523 |
+
"epoch": 40.62,
|
1524 |
+
"eval_accuracy": 0.41034103588846577,
|
1525 |
+
"eval_loss": 3.5284957885742188,
|
1526 |
+
"eval_runtime": 128.1661,
|
1527 |
+
"eval_samples_per_second": 187.686,
|
1528 |
+
"eval_steps_per_second": 5.867,
|
1529 |
+
"step": 1950
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 40.82,
|
1533 |
+
"learning_rate": 0.00026054402171063267,
|
1534 |
+
"loss": 3.2063,
|
1535 |
+
"step": 1960
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 41.04,
|
1539 |
+
"learning_rate": 0.0002592819961546308,
|
1540 |
+
"loss": 3.4173,
|
1541 |
+
"step": 1970
|
1542 |
+
},
|
1543 |
+
{
|
1544 |
+
"epoch": 41.25,
|
1545 |
+
"learning_rate": 0.00025801737741207005,
|
1546 |
+
"loss": 3.1796,
|
1547 |
+
"step": 1980
|
1548 |
+
},
|
1549 |
+
{
|
1550 |
+
"epoch": 41.45,
|
1551 |
+
"learning_rate": 0.000256750220801471,
|
1552 |
+
"loss": 3.1799,
|
1553 |
+
"step": 1990
|
1554 |
+
},
|
1555 |
+
{
|
1556 |
+
"epoch": 41.66,
|
1557 |
+
"learning_rate": 0.0002554805817523689,
|
1558 |
+
"loss": 3.1777,
|
1559 |
+
"step": 2000
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 41.66,
|
1563 |
+
"eval_accuracy": 0.41320715004942615,
|
1564 |
+
"eval_loss": 3.506920337677002,
|
1565 |
+
"eval_runtime": 128.1475,
|
1566 |
+
"eval_samples_per_second": 187.713,
|
1567 |
+
"eval_steps_per_second": 5.868,
|
1568 |
+
"step": 2000
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 41.86,
|
1572 |
+
"learning_rate": 0.0002542085158028889,
|
1573 |
+
"loss": 3.1791,
|
1574 |
+
"step": 2010
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 42.08,
|
1578 |
+
"learning_rate": 0.00025293407859731633,
|
1579 |
+
"loss": 3.363,
|
1580 |
+
"step": 2020
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 42.29,
|
1584 |
+
"learning_rate": 0.00025165732588366334,
|
1585 |
+
"loss": 3.1381,
|
1586 |
+
"step": 2030
|
1587 |
+
},
|
1588 |
+
{
|
1589 |
+
"epoch": 42.49,
|
1590 |
+
"learning_rate": 0.00025037831351122967,
|
1591 |
+
"loss": 3.1556,
|
1592 |
+
"step": 2040
|
1593 |
+
},
|
1594 |
+
{
|
1595 |
+
"epoch": 42.7,
|
1596 |
+
"learning_rate": 0.0002490970974281599,
|
1597 |
+
"loss": 3.1499,
|
1598 |
+
"step": 2050
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 42.7,
|
1602 |
+
"eval_accuracy": 0.4150287828947368,
|
1603 |
+
"eval_loss": 3.491703987121582,
|
1604 |
+
"eval_runtime": 128.345,
|
1605 |
+
"eval_samples_per_second": 187.425,
|
1606 |
+
"eval_steps_per_second": 5.859,
|
1607 |
+
"step": 2050
|
1608 |
+
},
|
1609 |
+
{
|
1610 |
+
"epoch": 42.9,
|
1611 |
+
"learning_rate": 0.00024781373367899597,
|
1612 |
+
"loss": 3.1374,
|
1613 |
+
"step": 2060
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 43.12,
|
1617 |
+
"learning_rate": 0.00024652827840222606,
|
1618 |
+
"loss": 3.3333,
|
1619 |
+
"step": 2070
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 43.33,
|
1623 |
+
"learning_rate": 0.00024524078782782807,
|
1624 |
+
"loss": 3.1338,
|
1625 |
+
"step": 2080
|
1626 |
+
},
|
1627 |
+
{
|
1628 |
+
"epoch": 43.53,
|
1629 |
+
"learning_rate": 0.00024395131827481062,
|
1630 |
+
"loss": 3.1092,
|
1631 |
+
"step": 2090
|
1632 |
+
},
|
1633 |
+
{
|
1634 |
+
"epoch": 43.74,
|
1635 |
+
"learning_rate": 0.0002426599261487494,
|
1636 |
+
"loss": 3.131,
|
1637 |
+
"step": 2100
|
1638 |
+
},
|
1639 |
+
{
|
1640 |
+
"epoch": 43.74,
|
1641 |
+
"eval_accuracy": 0.4168157298218596,
|
1642 |
+
"eval_loss": 3.4700751304626465,
|
1643 |
+
"eval_runtime": 128.1928,
|
1644 |
+
"eval_samples_per_second": 187.647,
|
1645 |
+
"eval_steps_per_second": 5.866,
|
1646 |
+
"step": 2100
|
1647 |
+
},
|
1648 |
+
{
|
1649 |
+
"epoch": 43.95,
|
1650 |
+
"learning_rate": 0.00024136666793931935,
|
1651 |
+
"loss": 3.1197,
|
1652 |
+
"step": 2110
|
1653 |
+
},
|
1654 |
+
{
|
1655 |
+
"epoch": 44.16,
|
1656 |
+
"learning_rate": 0.00024007160021782427,
|
1657 |
+
"loss": 3.3001,
|
1658 |
+
"step": 2120
|
1659 |
+
},
|
1660 |
+
{
|
1661 |
+
"epoch": 44.37,
|
1662 |
+
"learning_rate": 0.0002387747796347217,
|
1663 |
+
"loss": 3.0929,
|
1664 |
+
"step": 2130
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 44.58,
|
1668 |
+
"learning_rate": 0.00023747626291714498,
|
1669 |
+
"loss": 3.0968,
|
1670 |
+
"step": 2140
|
1671 |
+
},
|
1672 |
+
{
|
1673 |
+
"epoch": 44.78,
|
1674 |
+
"learning_rate": 0.000236176106866422,
|
1675 |
+
"loss": 3.0942,
|
1676 |
+
"step": 2150
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 44.78,
|
1680 |
+
"eval_accuracy": 0.4189149215354626,
|
1681 |
+
"eval_loss": 3.4530041217803955,
|
1682 |
+
"eval_runtime": 128.3801,
|
1683 |
+
"eval_samples_per_second": 187.373,
|
1684 |
+
"eval_steps_per_second": 5.858,
|
1685 |
+
"step": 2150
|
1686 |
+
},
|
1687 |
+
{
|
1688 |
+
"epoch": 44.99,
|
1689 |
+
"learning_rate": 0.00023487436835559035,
|
1690 |
+
"loss": 3.1072,
|
1691 |
+
"step": 2160
|
1692 |
+
},
|
1693 |
+
{
|
1694 |
+
"epoch": 45.21,
|
1695 |
+
"learning_rate": 0.00023357110432690954,
|
1696 |
+
"loss": 3.268,
|
1697 |
+
"step": 2170
|
1698 |
+
},
|
1699 |
+
{
|
1700 |
+
"epoch": 45.41,
|
1701 |
+
"learning_rate": 0.00023226637178937022,
|
1702 |
+
"loss": 3.0772,
|
1703 |
+
"step": 2180
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 45.62,
|
1707 |
+
"learning_rate": 0.00023096022781620034,
|
1708 |
+
"loss": 3.071,
|
1709 |
+
"step": 2190
|
1710 |
+
},
|
1711 |
+
{
|
1712 |
+
"epoch": 45.82,
|
1713 |
+
"learning_rate": 0.0002296527295423684,
|
1714 |
+
"loss": 3.0683,
|
1715 |
+
"step": 2200
|
1716 |
+
},
|
1717 |
+
{
|
1718 |
+
"epoch": 45.82,
|
1719 |
+
"eval_accuracy": 0.42115190069686975,
|
1720 |
+
"eval_loss": 3.4319911003112793,
|
1721 |
+
"eval_runtime": 128.2347,
|
1722 |
+
"eval_samples_per_second": 187.586,
|
1723 |
+
"eval_steps_per_second": 5.864,
|
1724 |
+
"step": 2200
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 46.04,
|
1728 |
+
"learning_rate": 0.00022834393416208486,
|
1729 |
+
"loss": 3.2606,
|
1730 |
+
"step": 2210
|
1731 |
+
},
|
1732 |
+
{
|
1733 |
+
"epoch": 46.25,
|
1734 |
+
"learning_rate": 0.0002270338989262994,
|
1735 |
+
"loss": 3.0464,
|
1736 |
+
"step": 2220
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 46.45,
|
1740 |
+
"learning_rate": 0.00022572268114019726,
|
1741 |
+
"loss": 3.0424,
|
1742 |
+
"step": 2230
|
1743 |
+
},
|
1744 |
+
{
|
1745 |
+
"epoch": 46.66,
|
1746 |
+
"learning_rate": 0.00022441033816069202,
|
1747 |
+
"loss": 3.0469,
|
1748 |
+
"step": 2240
|
1749 |
+
},
|
1750 |
+
{
|
1751 |
+
"epoch": 46.86,
|
1752 |
+
"learning_rate": 0.00022309692739391727,
|
1753 |
+
"loss": 3.0363,
|
1754 |
+
"step": 2250
|
1755 |
+
},
|
1756 |
+
{
|
1757 |
+
"epoch": 46.86,
|
1758 |
+
"eval_accuracy": 0.42269065003604217,
|
1759 |
+
"eval_loss": 3.419463872909546,
|
1760 |
+
"eval_runtime": 128.2052,
|
1761 |
+
"eval_samples_per_second": 187.629,
|
1762 |
+
"eval_steps_per_second": 5.866,
|
1763 |
+
"step": 2250
|
1764 |
+
},
|
1765 |
+
{
|
1766 |
+
"epoch": 47.08,
|
1767 |
+
"learning_rate": 0.00022178250629271452,
|
1768 |
+
"loss": 3.2579,
|
1769 |
+
"step": 2260
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 47.29,
|
1773 |
+
"learning_rate": 0.00022046713235412103,
|
1774 |
+
"loss": 3.0223,
|
1775 |
+
"step": 2270
|
1776 |
+
},
|
1777 |
+
{
|
1778 |
+
"epoch": 47.49,
|
1779 |
+
"learning_rate": 0.00021915086311685404,
|
1780 |
+
"loss": 3.0431,
|
1781 |
+
"step": 2280
|
1782 |
+
},
|
1783 |
+
{
|
1784 |
+
"epoch": 47.7,
|
1785 |
+
"learning_rate": 0.00021783375615879415,
|
1786 |
+
"loss": 3.0339,
|
1787 |
+
"step": 2290
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 47.9,
|
1791 |
+
"learning_rate": 0.0002165158690944665,
|
1792 |
+
"loss": 3.0264,
|
1793 |
+
"step": 2300
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 47.9,
|
1797 |
+
"eval_accuracy": 0.4248503269779865,
|
1798 |
+
"eval_loss": 3.4046127796173096,
|
1799 |
+
"eval_runtime": 128.0991,
|
1800 |
+
"eval_samples_per_second": 187.784,
|
1801 |
+
"eval_steps_per_second": 5.87,
|
1802 |
+
"step": 2300
|
1803 |
+
},
|
1804 |
+
{
|
1805 |
+
"epoch": 48.12,
|
1806 |
+
"learning_rate": 0.00021519725957252063,
|
1807 |
+
"loss": 3.2189,
|
1808 |
+
"step": 2310
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 48.33,
|
1812 |
+
"learning_rate": 0.00021387798527320882,
|
1813 |
+
"loss": 3.0121,
|
1814 |
+
"step": 2320
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 48.53,
|
1818 |
+
"learning_rate": 0.0002125581039058627,
|
1819 |
+
"loss": 3.0031,
|
1820 |
+
"step": 2330
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 48.74,
|
1824 |
+
"learning_rate": 0.0002112376732063691,
|
1825 |
+
"loss": 2.9933,
|
1826 |
+
"step": 2340
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 48.95,
|
1830 |
+
"learning_rate": 0.00020991675093464448,
|
1831 |
+
"loss": 3.0079,
|
1832 |
+
"step": 2350
|
1833 |
+
},
|
1834 |
+
{
|
1835 |
+
"epoch": 48.95,
|
1836 |
+
"eval_accuracy": 0.4266670495134685,
|
1837 |
+
"eval_loss": 3.3874006271362305,
|
1838 |
+
"eval_runtime": 128.1597,
|
1839 |
+
"eval_samples_per_second": 187.696,
|
1840 |
+
"eval_steps_per_second": 5.868,
|
1841 |
+
"step": 2350
|
1842 |
+
},
|
1843 |
+
{
|
1844 |
+
"epoch": 49.16,
|
1845 |
+
"learning_rate": 0.00020859539487210813,
|
1846 |
+
"loss": 3.2002,
|
1847 |
+
"step": 2360
|
1848 |
+
},
|
1849 |
+
{
|
1850 |
+
"epoch": 49.37,
|
1851 |
+
"learning_rate": 0.0002072736628191549,
|
1852 |
+
"loss": 2.9868,
|
1853 |
+
"step": 2370
|
1854 |
+
},
|
1855 |
+
{
|
1856 |
+
"epoch": 49.58,
|
1857 |
+
"learning_rate": 0.0002059516125926265,
|
1858 |
+
"loss": 2.9996,
|
1859 |
+
"step": 2380
|
1860 |
+
},
|
1861 |
+
{
|
1862 |
+
"epoch": 49.78,
|
1863 |
+
"learning_rate": 0.00020462930202328278,
|
1864 |
+
"loss": 2.9719,
|
1865 |
+
"step": 2390
|
1866 |
+
},
|
1867 |
+
{
|
1868 |
+
"epoch": 49.99,
|
1869 |
+
"learning_rate": 0.00020330678895327174,
|
1870 |
+
"loss": 2.9869,
|
1871 |
+
"step": 2400
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 49.99,
|
1875 |
+
"eval_accuracy": 0.42770797209455824,
|
1876 |
+
"eval_loss": 3.3792383670806885,
|
1877 |
+
"eval_runtime": 128.195,
|
1878 |
+
"eval_samples_per_second": 187.644,
|
1879 |
+
"eval_steps_per_second": 5.866,
|
1880 |
+
"step": 2400
|
1881 |
+
},
|
1882 |
+
{
|
1883 |
+
"epoch": 50.21,
|
1884 |
+
"learning_rate": 0.00020198413123359926,
|
1885 |
+
"loss": 3.1735,
|
1886 |
+
"step": 2410
|
1887 |
+
},
|
1888 |
+
{
|
1889 |
+
"epoch": 50.41,
|
1890 |
+
"learning_rate": 0.00020066138672159903,
|
1891 |
+
"loss": 2.9707,
|
1892 |
+
"step": 2420
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 50.62,
|
1896 |
+
"learning_rate": 0.00019933861327840098,
|
1897 |
+
"loss": 2.9682,
|
1898 |
+
"step": 2430
|
1899 |
+
},
|
1900 |
+
{
|
1901 |
+
"epoch": 50.82,
|
1902 |
+
"learning_rate": 0.00019801586876640073,
|
1903 |
+
"loss": 2.9752,
|
1904 |
+
"step": 2440
|
1905 |
+
},
|
1906 |
+
{
|
1907 |
+
"epoch": 51.04,
|
1908 |
+
"learning_rate": 0.0001966932110467283,
|
1909 |
+
"loss": 3.1592,
|
1910 |
+
"step": 2450
|
1911 |
+
},
|
1912 |
+
{
|
1913 |
+
"epoch": 51.04,
|
1914 |
+
"eval_accuracy": 0.4289155778595229,
|
1915 |
+
"eval_loss": 3.3654892444610596,
|
1916 |
+
"eval_runtime": 128.18,
|
1917 |
+
"eval_samples_per_second": 187.666,
|
1918 |
+
"eval_steps_per_second": 5.867,
|
1919 |
+
"step": 2450
|
1920 |
+
},
|
1921 |
+
{
|
1922 |
+
"epoch": 51.25,
|
1923 |
+
"learning_rate": 0.00019537069797671724,
|
1924 |
+
"loss": 2.948,
|
1925 |
+
"step": 2460
|
1926 |
+
},
|
1927 |
+
{
|
1928 |
+
"epoch": 51.45,
|
1929 |
+
"learning_rate": 0.0001940483874073735,
|
1930 |
+
"loss": 2.9368,
|
1931 |
+
"step": 2470
|
1932 |
+
},
|
1933 |
+
{
|
1934 |
+
"epoch": 51.66,
|
1935 |
+
"learning_rate": 0.00019272633718084517,
|
1936 |
+
"loss": 2.9466,
|
1937 |
+
"step": 2480
|
1938 |
+
},
|
1939 |
+
{
|
1940 |
+
"epoch": 51.86,
|
1941 |
+
"learning_rate": 0.0001914046051278919,
|
1942 |
+
"loss": 2.9541,
|
1943 |
+
"step": 2490
|
1944 |
+
},
|
1945 |
+
{
|
1946 |
+
"epoch": 52.08,
|
1947 |
+
"learning_rate": 0.00019008324906535554,
|
1948 |
+
"loss": 3.1353,
|
1949 |
+
"step": 2500
|
1950 |
+
},
|
1951 |
+
{
|
1952 |
+
"epoch": 52.08,
|
1953 |
+
"eval_accuracy": 0.43104300088517533,
|
1954 |
+
"eval_loss": 3.3548085689544678,
|
1955 |
+
"eval_runtime": 128.179,
|
1956 |
+
"eval_samples_per_second": 187.667,
|
1957 |
+
"eval_steps_per_second": 5.867,
|
1958 |
+
"step": 2500
|
1959 |
+
},
|
1960 |
+
{
|
1961 |
+
"epoch": 52.29,
|
1962 |
+
"learning_rate": 0.0001887623267936309,
|
1963 |
+
"loss": 2.9264,
|
1964 |
+
"step": 2510
|
1965 |
+
},
|
1966 |
+
{
|
1967 |
+
"epoch": 52.49,
|
1968 |
+
"learning_rate": 0.00018744189609413734,
|
1969 |
+
"loss": 2.9325,
|
1970 |
+
"step": 2520
|
1971 |
+
},
|
1972 |
+
{
|
1973 |
+
"epoch": 52.7,
|
1974 |
+
"learning_rate": 0.0001861220147267912,
|
1975 |
+
"loss": 2.9263,
|
1976 |
+
"step": 2530
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 52.9,
|
1980 |
+
"learning_rate": 0.0001848027404274794,
|
1981 |
+
"loss": 2.9275,
|
1982 |
+
"step": 2540
|
1983 |
+
},
|
1984 |
+
{
|
1985 |
+
"epoch": 53.12,
|
1986 |
+
"learning_rate": 0.00018348413090553354,
|
1987 |
+
"loss": 3.1257,
|
1988 |
+
"step": 2550
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 53.12,
|
1992 |
+
"eval_accuracy": 0.43083924373522625,
|
1993 |
+
"eval_loss": 3.348921775817871,
|
1994 |
+
"eval_runtime": 128.2056,
|
1995 |
+
"eval_samples_per_second": 187.628,
|
1996 |
+
"eval_steps_per_second": 5.866,
|
1997 |
+
"step": 2550
|
1998 |
+
},
|
1999 |
+
{
|
2000 |
+
"epoch": 53.33,
|
2001 |
+
"learning_rate": 0.00018216624384120595,
|
2002 |
+
"loss": 2.9018,
|
2003 |
+
"step": 2560
|
2004 |
+
},
|
2005 |
+
{
|
2006 |
+
"epoch": 53.53,
|
2007 |
+
"learning_rate": 0.00018084913688314597,
|
2008 |
+
"loss": 2.9135,
|
2009 |
+
"step": 2570
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 53.74,
|
2013 |
+
"learning_rate": 0.000179532867645879,
|
2014 |
+
"loss": 2.9067,
|
2015 |
+
"step": 2580
|
2016 |
+
},
|
2017 |
+
{
|
2018 |
+
"epoch": 53.95,
|
2019 |
+
"learning_rate": 0.0001782174937072855,
|
2020 |
+
"loss": 2.9146,
|
2021 |
+
"step": 2590
|
2022 |
+
},
|
2023 |
+
{
|
2024 |
+
"epoch": 54.16,
|
2025 |
+
"learning_rate": 0.00017690307260608278,
|
2026 |
+
"loss": 3.0822,
|
2027 |
+
"step": 2600
|
2028 |
+
},
|
2029 |
+
{
|
2030 |
+
"epoch": 54.16,
|
2031 |
+
"eval_accuracy": 0.4326622137249495,
|
2032 |
+
"eval_loss": 3.3352506160736084,
|
2033 |
+
"eval_runtime": 128.2781,
|
2034 |
+
"eval_samples_per_second": 187.522,
|
2035 |
+
"eval_steps_per_second": 5.862,
|
2036 |
+
"step": 2600
|
2037 |
+
},
|
2038 |
+
{
|
2039 |
+
"epoch": 54.37,
|
2040 |
+
"learning_rate": 0.000175589661839308,
|
2041 |
+
"loss": 2.8995,
|
2042 |
+
"step": 2610
|
2043 |
+
},
|
2044 |
+
{
|
2045 |
+
"epoch": 54.58,
|
2046 |
+
"learning_rate": 0.00017427731885980282,
|
2047 |
+
"loss": 2.8945,
|
2048 |
+
"step": 2620
|
2049 |
+
},
|
2050 |
+
{
|
2051 |
+
"epoch": 54.78,
|
2052 |
+
"learning_rate": 0.0001729661010737007,
|
2053 |
+
"loss": 2.905,
|
2054 |
+
"step": 2630
|
2055 |
+
},
|
2056 |
+
{
|
2057 |
+
"epoch": 54.99,
|
2058 |
+
"learning_rate": 0.00017165606583791515,
|
2059 |
+
"loss": 2.9128,
|
2060 |
+
"step": 2640
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 55.21,
|
2064 |
+
"learning_rate": 0.00017034727045763158,
|
2065 |
+
"loss": 3.0771,
|
2066 |
+
"step": 2650
|
2067 |
+
},
|
2068 |
+
{
|
2069 |
+
"epoch": 55.21,
|
2070 |
+
"eval_accuracy": 0.434098312415683,
|
2071 |
+
"eval_loss": 3.3219847679138184,
|
2072 |
+
"eval_runtime": 128.3146,
|
2073 |
+
"eval_samples_per_second": 187.469,
|
2074 |
+
"eval_steps_per_second": 5.861,
|
2075 |
+
"step": 2650
|
2076 |
+
},
|
2077 |
+
{
|
2078 |
+
"epoch": 55.41,
|
2079 |
+
"learning_rate": 0.00016903977218379974,
|
2080 |
+
"loss": 2.8695,
|
2081 |
+
"step": 2660
|
2082 |
+
},
|
2083 |
+
{
|
2084 |
+
"epoch": 55.62,
|
2085 |
+
"learning_rate": 0.00016773362821062983,
|
2086 |
+
"loss": 2.8839,
|
2087 |
+
"step": 2670
|
2088 |
+
},
|
2089 |
+
{
|
2090 |
+
"epoch": 55.82,
|
2091 |
+
"learning_rate": 0.00016642889567309048,
|
2092 |
+
"loss": 2.8887,
|
2093 |
+
"step": 2680
|
2094 |
+
},
|
2095 |
+
{
|
2096 |
+
"epoch": 56.04,
|
2097 |
+
"learning_rate": 0.0001651256316444097,
|
2098 |
+
"loss": 3.0754,
|
2099 |
+
"step": 2690
|
2100 |
+
},
|
2101 |
+
{
|
2102 |
+
"epoch": 56.25,
|
2103 |
+
"learning_rate": 0.0001638238931335781,
|
2104 |
+
"loss": 2.8639,
|
2105 |
+
"step": 2700
|
2106 |
+
},
|
2107 |
+
{
|
2108 |
+
"epoch": 56.25,
|
2109 |
+
"eval_accuracy": 0.4353990105725288,
|
2110 |
+
"eval_loss": 3.3119492530822754,
|
2111 |
+
"eval_runtime": 128.0745,
|
2112 |
+
"eval_samples_per_second": 187.82,
|
2113 |
+
"eval_steps_per_second": 5.872,
|
2114 |
+
"step": 2700
|
2115 |
+
},
|
2116 |
+
{
|
2117 |
+
"epoch": 56.45,
|
2118 |
+
"learning_rate": 0.00016252373708285504,
|
2119 |
+
"loss": 2.8653,
|
2120 |
+
"step": 2710
|
2121 |
+
},
|
2122 |
+
{
|
2123 |
+
"epoch": 56.66,
|
2124 |
+
"learning_rate": 0.00016122522036527838,
|
2125 |
+
"loss": 2.8696,
|
2126 |
+
"step": 2720
|
2127 |
+
},
|
2128 |
+
{
|
2129 |
+
"epoch": 56.86,
|
2130 |
+
"learning_rate": 0.00015992839978217578,
|
2131 |
+
"loss": 2.8665,
|
2132 |
+
"step": 2730
|
2133 |
+
},
|
2134 |
+
{
|
2135 |
+
"epoch": 57.08,
|
2136 |
+
"learning_rate": 0.00015863333206068067,
|
2137 |
+
"loss": 3.0651,
|
2138 |
+
"step": 2740
|
2139 |
+
},
|
2140 |
+
{
|
2141 |
+
"epoch": 57.29,
|
2142 |
+
"learning_rate": 0.00015734007385125067,
|
2143 |
+
"loss": 2.8477,
|
2144 |
+
"step": 2750
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 57.29,
|
2148 |
+
"eval_accuracy": 0.4360402472560164,
|
2149 |
+
"eval_loss": 3.310389280319214,
|
2150 |
+
"eval_runtime": 128.2649,
|
2151 |
+
"eval_samples_per_second": 187.542,
|
2152 |
+
"eval_steps_per_second": 5.863,
|
2153 |
+
"step": 2750
|
2154 |
+
},
|
2155 |
+
{
|
2156 |
+
"epoch": 57.49,
|
2157 |
+
"learning_rate": 0.0001560486817251894,
|
2158 |
+
"loss": 2.8511,
|
2159 |
+
"step": 2760
|
2160 |
+
},
|
2161 |
+
{
|
2162 |
+
"epoch": 57.7,
|
2163 |
+
"learning_rate": 0.000154759212172172,
|
2164 |
+
"loss": 2.8615,
|
2165 |
+
"step": 2770
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 57.9,
|
2169 |
+
"learning_rate": 0.00015347172159777396,
|
2170 |
+
"loss": 2.8619,
|
2171 |
+
"step": 2780
|
2172 |
+
},
|
2173 |
+
{
|
2174 |
+
"epoch": 58.12,
|
2175 |
+
"learning_rate": 0.000152186266321004,
|
2176 |
+
"loss": 3.0316,
|
2177 |
+
"step": 2790
|
2178 |
+
},
|
2179 |
+
{
|
2180 |
+
"epoch": 58.33,
|
2181 |
+
"learning_rate": 0.0001509029025718402,
|
2182 |
+
"loss": 2.8373,
|
2183 |
+
"step": 2800
|
2184 |
+
},
|
2185 |
+
{
|
2186 |
+
"epoch": 58.33,
|
2187 |
+
"eval_accuracy": 0.4378144877232535,
|
2188 |
+
"eval_loss": 3.295414686203003,
|
2189 |
+
"eval_runtime": 128.0673,
|
2190 |
+
"eval_samples_per_second": 187.831,
|
2191 |
+
"eval_steps_per_second": 5.872,
|
2192 |
+
"step": 2800
|
2193 |
+
},
|
2194 |
+
{
|
2195 |
+
"epoch": 58.53,
|
2196 |
+
"learning_rate": 0.0001496216864887704,
|
2197 |
+
"loss": 2.8292,
|
2198 |
+
"step": 2810
|
2199 |
+
},
|
2200 |
+
{
|
2201 |
+
"epoch": 58.74,
|
2202 |
+
"learning_rate": 0.00014834267411633674,
|
2203 |
+
"loss": 2.8361,
|
2204 |
+
"step": 2820
|
2205 |
+
},
|
2206 |
+
{
|
2207 |
+
"epoch": 58.95,
|
2208 |
+
"learning_rate": 0.0001470659214026837,
|
2209 |
+
"loss": 2.8417,
|
2210 |
+
"step": 2830
|
2211 |
+
},
|
2212 |
+
{
|
2213 |
+
"epoch": 59.16,
|
2214 |
+
"learning_rate": 0.00014579148419711119,
|
2215 |
+
"loss": 3.0263,
|
2216 |
+
"step": 2840
|
2217 |
+
},
|
2218 |
+
{
|
2219 |
+
"epoch": 59.37,
|
2220 |
+
"learning_rate": 0.00014451941824763113,
|
2221 |
+
"loss": 2.818,
|
2222 |
+
"step": 2850
|
2223 |
+
},
|
2224 |
+
{
|
2225 |
+
"epoch": 59.37,
|
2226 |
+
"eval_accuracy": 0.43805501144654146,
|
2227 |
+
"eval_loss": 3.2935194969177246,
|
2228 |
+
"eval_runtime": 128.0232,
|
2229 |
+
"eval_samples_per_second": 187.896,
|
2230 |
+
"eval_steps_per_second": 5.874,
|
2231 |
+
"step": 2850
|
2232 |
+
},
|
2233 |
+
{
|
2234 |
+
"epoch": 59.58,
|
2235 |
+
"learning_rate": 0.000143249779198529,
|
2236 |
+
"loss": 2.8253,
|
2237 |
+
"step": 2860
|
2238 |
+
},
|
2239 |
+
{
|
2240 |
+
"epoch": 59.78,
|
2241 |
+
"learning_rate": 0.00014198262258793002,
|
2242 |
+
"loss": 2.8424,
|
2243 |
+
"step": 2870
|
2244 |
+
},
|
2245 |
+
{
|
2246 |
+
"epoch": 59.99,
|
2247 |
+
"learning_rate": 0.00014071800384536927,
|
2248 |
+
"loss": 2.8335,
|
2249 |
+
"step": 2880
|
2250 |
+
},
|
2251 |
+
{
|
2252 |
+
"epoch": 60.21,
|
2253 |
+
"learning_rate": 0.00013945597828936737,
|
2254 |
+
"loss": 2.9887,
|
2255 |
+
"step": 2890
|
2256 |
+
},
|
2257 |
+
{
|
2258 |
+
"epoch": 60.41,
|
2259 |
+
"learning_rate": 0.00013819660112501054,
|
2260 |
+
"loss": 2.8137,
|
2261 |
+
"step": 2900
|
2262 |
+
},
|
2263 |
+
{
|
2264 |
+
"epoch": 60.41,
|
2265 |
+
"eval_accuracy": 0.4394361605808428,
|
2266 |
+
"eval_loss": 3.278566598892212,
|
2267 |
+
"eval_runtime": 128.078,
|
2268 |
+
"eval_samples_per_second": 187.815,
|
2269 |
+
"eval_steps_per_second": 5.871,
|
2270 |
+
"step": 2900
|
2271 |
+
},
|
2272 |
+
{
|
2273 |
+
"epoch": 60.62,
|
2274 |
+
"learning_rate": 0.00013693992744153572,
|
2275 |
+
"loss": 2.8271,
|
2276 |
+
"step": 2910
|
2277 |
+
},
|
2278 |
+
{
|
2279 |
+
"epoch": 60.82,
|
2280 |
+
"learning_rate": 0.00013568601220992097,
|
2281 |
+
"loss": 2.8286,
|
2282 |
+
"step": 2920
|
2283 |
+
},
|
2284 |
+
{
|
2285 |
+
"epoch": 61.04,
|
2286 |
+
"learning_rate": 0.00013443491028048045,
|
2287 |
+
"loss": 3.0135,
|
2288 |
+
"step": 2930
|
2289 |
+
},
|
2290 |
+
{
|
2291 |
+
"epoch": 61.25,
|
2292 |
+
"learning_rate": 0.0001331866763804658,
|
2293 |
+
"loss": 2.8038,
|
2294 |
+
"step": 2940
|
2295 |
+
},
|
2296 |
+
{
|
2297 |
+
"epoch": 61.45,
|
2298 |
+
"learning_rate": 0.0001319413651116714,
|
2299 |
+
"loss": 2.7985,
|
2300 |
+
"step": 2950
|
2301 |
+
},
|
2302 |
+
{
|
2303 |
+
"epoch": 61.45,
|
2304 |
+
"eval_accuracy": 0.4401244630436134,
|
2305 |
+
"eval_loss": 3.2746615409851074,
|
2306 |
+
"eval_runtime": 128.0922,
|
2307 |
+
"eval_samples_per_second": 187.794,
|
2308 |
+
"eval_steps_per_second": 5.871,
|
2309 |
+
"step": 2950
|
2310 |
+
},
|
2311 |
+
{
|
2312 |
+
"epoch": 61.66,
|
2313 |
+
"learning_rate": 0.00013069903094804644,
|
2314 |
+
"loss": 2.7993,
|
2315 |
+
"step": 2960
|
2316 |
+
},
|
2317 |
+
{
|
2318 |
+
"epoch": 61.86,
|
2319 |
+
"learning_rate": 0.0001294597282333118,
|
2320 |
+
"loss": 2.8132,
|
2321 |
+
"step": 2970
|
2322 |
+
},
|
2323 |
+
{
|
2324 |
+
"epoch": 62.08,
|
2325 |
+
"learning_rate": 0.00012822351117858303,
|
2326 |
+
"loss": 2.9785,
|
2327 |
+
"step": 2980
|
2328 |
+
},
|
2329 |
+
{
|
2330 |
+
"epoch": 62.29,
|
2331 |
+
"learning_rate": 0.0001269904338599989,
|
2332 |
+
"loss": 2.7959,
|
2333 |
+
"step": 2990
|
2334 |
+
},
|
2335 |
+
{
|
2336 |
+
"epoch": 62.49,
|
2337 |
+
"learning_rate": 0.0001257605502163558,
|
2338 |
+
"loss": 2.7936,
|
2339 |
+
"step": 3000
|
2340 |
+
},
|
2341 |
+
{
|
2342 |
+
"epoch": 62.49,
|
2343 |
+
"eval_accuracy": 0.44108544914689357,
|
2344 |
+
"eval_loss": 3.266845941543579,
|
2345 |
+
"eval_runtime": 128.1403,
|
2346 |
+
"eval_samples_per_second": 187.724,
|
2347 |
+
"eval_steps_per_second": 5.869,
|
2348 |
+
"step": 3000
|
2349 |
+
},
|
2350 |
+
{
|
2351 |
+
"epoch": 62.7,
|
2352 |
+
"learning_rate": 0.00012453391404674885,
|
2353 |
+
"loss": 2.7904,
|
2354 |
+
"step": 3010
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 62.9,
|
2358 |
+
"learning_rate": 0.00012331057900821768,
|
2359 |
+
"loss": 2.7934,
|
2360 |
+
"step": 3020
|
2361 |
+
},
|
2362 |
+
{
|
2363 |
+
"epoch": 63.12,
|
2364 |
+
"learning_rate": 0.0001220905986134,
|
2365 |
+
"loss": 2.9571,
|
2366 |
+
"step": 3030
|
2367 |
+
},
|
2368 |
+
{
|
2369 |
+
"epoch": 63.33,
|
2370 |
+
"learning_rate": 0.00012087402622819039,
|
2371 |
+
"loss": 2.7925,
|
2372 |
+
"step": 3040
|
2373 |
+
},
|
2374 |
+
{
|
2375 |
+
"epoch": 63.53,
|
2376 |
+
"learning_rate": 0.00011966091506940616,
|
2377 |
+
"loss": 2.7764,
|
2378 |
+
"step": 3050
|
2379 |
+
},
|
2380 |
+
{
|
2381 |
+
"epoch": 63.53,
|
2382 |
+
"eval_accuracy": 0.441903341927903,
|
2383 |
+
"eval_loss": 3.256887197494507,
|
2384 |
+
"eval_runtime": 128.1402,
|
2385 |
+
"eval_samples_per_second": 187.724,
|
2386 |
+
"eval_steps_per_second": 5.869,
|
2387 |
+
"step": 3050
|
2388 |
+
},
|
2389 |
+
{
|
2390 |
+
"epoch": 63.74,
|
2391 |
+
"learning_rate": 0.00011845131820245934,
|
2392 |
+
"loss": 2.7851,
|
2393 |
+
"step": 3060
|
2394 |
+
},
|
2395 |
+
{
|
2396 |
+
"epoch": 63.95,
|
2397 |
+
"learning_rate": 0.00011724528853903536,
|
2398 |
+
"loss": 2.7837,
|
2399 |
+
"step": 3070
|
2400 |
+
},
|
2401 |
+
{
|
2402 |
+
"epoch": 64.16,
|
2403 |
+
"learning_rate": 0.00011604287883477889,
|
2404 |
+
"loss": 2.9344,
|
2405 |
+
"step": 3080
|
2406 |
+
},
|
2407 |
+
{
|
2408 |
+
"epoch": 64.37,
|
2409 |
+
"learning_rate": 0.00011484414168698547,
|
2410 |
+
"loss": 2.7703,
|
2411 |
+
"step": 3090
|
2412 |
+
},
|
2413 |
+
{
|
2414 |
+
"epoch": 64.58,
|
2415 |
+
"learning_rate": 0.00011364912953230145,
|
2416 |
+
"loss": 2.7819,
|
2417 |
+
"step": 3100
|
2418 |
+
},
|
2419 |
+
{
|
2420 |
+
"epoch": 64.58,
|
2421 |
+
"eval_accuracy": 0.44339571520227505,
|
2422 |
+
"eval_loss": 3.2492308616638184,
|
2423 |
+
"eval_runtime": 128.0362,
|
2424 |
+
"eval_samples_per_second": 187.877,
|
2425 |
+
"eval_steps_per_second": 5.873,
|
2426 |
+
"step": 3100
|
2427 |
+
},
|
2428 |
+
{
|
2429 |
+
"epoch": 64.78,
|
2430 |
+
"learning_rate": 0.00011245789464442964,
|
2431 |
+
"loss": 2.7841,
|
2432 |
+
"step": 3110
|
2433 |
+
},
|
2434 |
+
{
|
2435 |
+
"epoch": 64.99,
|
2436 |
+
"learning_rate": 0.00011127048913184326,
|
2437 |
+
"loss": 2.7794,
|
2438 |
+
"step": 3120
|
2439 |
+
},
|
2440 |
+
{
|
2441 |
+
"epoch": 65.21,
|
2442 |
+
"learning_rate": 0.00011008696493550599,
|
2443 |
+
"loss": 2.9422,
|
2444 |
+
"step": 3130
|
2445 |
+
},
|
2446 |
+
{
|
2447 |
+
"epoch": 65.41,
|
2448 |
+
"learning_rate": 0.00010890737382660015,
|
2449 |
+
"loss": 2.7573,
|
2450 |
+
"step": 3140
|
2451 |
+
},
|
2452 |
+
{
|
2453 |
+
"epoch": 65.62,
|
2454 |
+
"learning_rate": 0.00010773176740426248,
|
2455 |
+
"loss": 2.7672,
|
2456 |
+
"step": 3150
|
2457 |
+
},
|
2458 |
+
{
|
2459 |
+
"epoch": 65.62,
|
2460 |
+
"eval_accuracy": 0.4433201371393935,
|
2461 |
+
"eval_loss": 3.2493698596954346,
|
2462 |
+
"eval_runtime": 128.0217,
|
2463 |
+
"eval_samples_per_second": 187.898,
|
2464 |
+
"eval_steps_per_second": 5.874,
|
2465 |
+
"step": 3150
|
2466 |
+
},
|
2467 |
+
{
|
2468 |
+
"epoch": 65.82,
|
2469 |
+
"learning_rate": 0.00010656019709332606,
|
2470 |
+
"loss": 2.7557,
|
2471 |
+
"step": 3160
|
2472 |
+
},
|
2473 |
+
{
|
2474 |
+
"epoch": 66.04,
|
2475 |
+
"learning_rate": 0.00010539271414207186,
|
2476 |
+
"loss": 2.9353,
|
2477 |
+
"step": 3170
|
2478 |
+
},
|
2479 |
+
{
|
2480 |
+
"epoch": 66.25,
|
2481 |
+
"learning_rate": 0.00010422936961998609,
|
2482 |
+
"loss": 2.7494,
|
2483 |
+
"step": 3180
|
2484 |
+
},
|
2485 |
+
{
|
2486 |
+
"epoch": 66.45,
|
2487 |
+
"learning_rate": 0.00010307021441552707,
|
2488 |
+
"loss": 2.7401,
|
2489 |
+
"step": 3190
|
2490 |
+
},
|
2491 |
+
{
|
2492 |
+
"epoch": 66.66,
|
2493 |
+
"learning_rate": 0.00010191529923389845,
|
2494 |
+
"loss": 2.7629,
|
2495 |
+
"step": 3200
|
2496 |
+
},
|
2497 |
+
{
|
2498 |
+
"epoch": 66.66,
|
2499 |
+
"eval_accuracy": 0.44430680533611233,
|
2500 |
+
"eval_loss": 3.240968704223633,
|
2501 |
+
"eval_runtime": 128.0927,
|
2502 |
+
"eval_samples_per_second": 187.794,
|
2503 |
+
"eval_steps_per_second": 5.871,
|
2504 |
+
"step": 3200
|
2505 |
+
},
|
2506 |
+
{
|
2507 |
+
"epoch": 66.86,
|
2508 |
+
"learning_rate": 0.00010076467459483155,
|
2509 |
+
"loss": 2.7537,
|
2510 |
+
"step": 3210
|
2511 |
+
},
|
2512 |
+
{
|
2513 |
+
"epoch": 67.08,
|
2514 |
+
"learning_rate": 9.961839083037592e-05,
|
2515 |
+
"loss": 2.9359,
|
2516 |
+
"step": 3220
|
2517 |
+
},
|
2518 |
+
{
|
2519 |
+
"epoch": 67.29,
|
2520 |
+
"learning_rate": 9.847649808269658e-05,
|
2521 |
+
"loss": 2.7575,
|
2522 |
+
"step": 3230
|
2523 |
+
},
|
2524 |
+
{
|
2525 |
+
"epoch": 67.49,
|
2526 |
+
"learning_rate": 9.733904630188176e-05,
|
2527 |
+
"loss": 2.7294,
|
2528 |
+
"step": 3240
|
2529 |
+
},
|
2530 |
+
{
|
2531 |
+
"epoch": 67.7,
|
2532 |
+
"learning_rate": 9.620608524375703e-05,
|
2533 |
+
"loss": 2.747,
|
2534 |
+
"step": 3250
|
2535 |
+
},
|
2536 |
+
{
|
2537 |
+
"epoch": 67.7,
|
2538 |
+
"eval_accuracy": 0.4446199448310505,
|
2539 |
+
"eval_loss": 3.236819267272949,
|
2540 |
+
"eval_runtime": 127.9991,
|
2541 |
+
"eval_samples_per_second": 187.931,
|
2542 |
+
"eval_steps_per_second": 5.875,
|
2543 |
+
"step": 3250
|
2544 |
+
},
|
2545 |
+
{
|
2546 |
+
"epoch": 67.9,
|
2547 |
+
"learning_rate": 9.507766446770934e-05,
|
2548 |
+
"loss": 2.7458,
|
2549 |
+
"step": 3260
|
2550 |
+
},
|
2551 |
+
{
|
2552 |
+
"epoch": 68.12,
|
2553 |
+
"learning_rate": 9.39538333345191e-05,
|
2554 |
+
"loss": 2.9246,
|
2555 |
+
"step": 3270
|
2556 |
+
},
|
2557 |
+
{
|
2558 |
+
"epoch": 68.33,
|
2559 |
+
"learning_rate": 9.283464100420063e-05,
|
2560 |
+
"loss": 2.741,
|
2561 |
+
"step": 3280
|
2562 |
+
},
|
2563 |
+
{
|
2564 |
+
"epoch": 68.53,
|
2565 |
+
"learning_rate": 9.17201364338524e-05,
|
2566 |
+
"loss": 2.7421,
|
2567 |
+
"step": 3290
|
2568 |
+
},
|
2569 |
+
{
|
2570 |
+
"epoch": 68.74,
|
2571 |
+
"learning_rate": 9.061036837551466e-05,
|
2572 |
+
"loss": 2.7303,
|
2573 |
+
"step": 3300
|
2574 |
+
},
|
2575 |
+
{
|
2576 |
+
"epoch": 68.74,
|
2577 |
+
"eval_accuracy": 0.44596990309042184,
|
2578 |
+
"eval_loss": 3.224606990814209,
|
2579 |
+
"eval_runtime": 128.0175,
|
2580 |
+
"eval_samples_per_second": 187.904,
|
2581 |
+
"eval_steps_per_second": 5.874,
|
2582 |
+
"step": 3300
|
2583 |
+
},
|
2584 |
+
{
|
2585 |
+
"epoch": 68.95,
|
2586 |
+
"learning_rate": 8.950538537403736e-05,
|
2587 |
+
"loss": 2.7291,
|
2588 |
+
"step": 3310
|
2589 |
+
},
|
2590 |
+
{
|
2591 |
+
"epoch": 69.16,
|
2592 |
+
"learning_rate": 8.840523576495681e-05,
|
2593 |
+
"loss": 2.903,
|
2594 |
+
"step": 3320
|
2595 |
+
},
|
2596 |
+
{
|
2597 |
+
"epoch": 69.37,
|
2598 |
+
"learning_rate": 8.730996767238072e-05,
|
2599 |
+
"loss": 2.7319,
|
2600 |
+
"step": 3330
|
2601 |
+
},
|
2602 |
+
{
|
2603 |
+
"epoch": 69.58,
|
2604 |
+
"learning_rate": 8.621962900688378e-05,
|
2605 |
+
"loss": 2.7166,
|
2606 |
+
"step": 3340
|
2607 |
+
},
|
2608 |
+
{
|
2609 |
+
"epoch": 69.78,
|
2610 |
+
"learning_rate": 8.513426746341128e-05,
|
2611 |
+
"loss": 2.7461,
|
2612 |
+
"step": 3350
|
2613 |
+
},
|
2614 |
+
{
|
2615 |
+
"epoch": 69.78,
|
2616 |
+
"eval_accuracy": 0.44624793300809595,
|
2617 |
+
"eval_loss": 3.2212436199188232,
|
2618 |
+
"eval_runtime": 128.0443,
|
2619 |
+
"eval_samples_per_second": 187.865,
|
2620 |
+
"eval_steps_per_second": 5.873,
|
2621 |
+
"step": 3350
|
2622 |
+
},
|
2623 |
+
{
|
2624 |
+
"epoch": 69.99,
|
2625 |
+
"learning_rate": 8.405393051919333e-05,
|
2626 |
+
"loss": 2.7214,
|
2627 |
+
"step": 3360
|
2628 |
+
},
|
2629 |
+
{
|
2630 |
+
"epoch": 70.21,
|
2631 |
+
"learning_rate": 8.29786654316677e-05,
|
2632 |
+
"loss": 2.8969,
|
2633 |
+
"step": 3370
|
2634 |
+
},
|
2635 |
+
{
|
2636 |
+
"epoch": 70.41,
|
2637 |
+
"learning_rate": 8.190851923641259e-05,
|
2638 |
+
"loss": 2.6964,
|
2639 |
+
"step": 3380
|
2640 |
+
},
|
2641 |
+
{
|
2642 |
+
"epoch": 70.62,
|
2643 |
+
"learning_rate": 8.084353874508947e-05,
|
2644 |
+
"loss": 2.7295,
|
2645 |
+
"step": 3390
|
2646 |
+
},
|
2647 |
+
{
|
2648 |
+
"epoch": 70.82,
|
2649 |
+
"learning_rate": 7.978377054339499e-05,
|
2650 |
+
"loss": 2.7179,
|
2651 |
+
"step": 3400
|
2652 |
+
},
|
2653 |
+
{
|
2654 |
+
"epoch": 70.82,
|
2655 |
+
"eval_accuracy": 0.4470331759822518,
|
2656 |
+
"eval_loss": 3.221658706665039,
|
2657 |
+
"eval_runtime": 127.9471,
|
2658 |
+
"eval_samples_per_second": 188.007,
|
2659 |
+
"eval_steps_per_second": 5.877,
|
2660 |
+
"step": 3400
|
2661 |
+
},
|
2662 |
+
{
|
2663 |
+
"epoch": 71.04,
|
2664 |
+
"learning_rate": 7.872926098902358e-05,
|
2665 |
+
"loss": 2.9027,
|
2666 |
+
"step": 3410
|
2667 |
+
},
|
2668 |
+
{
|
2669 |
+
"epoch": 71.25,
|
2670 |
+
"learning_rate": 7.768005620963916e-05,
|
2671 |
+
"loss": 2.7053,
|
2672 |
+
"step": 3420
|
2673 |
+
},
|
2674 |
+
{
|
2675 |
+
"epoch": 71.45,
|
2676 |
+
"learning_rate": 7.663620210085781e-05,
|
2677 |
+
"loss": 2.709,
|
2678 |
+
"step": 3430
|
2679 |
+
},
|
2680 |
+
{
|
2681 |
+
"epoch": 71.66,
|
2682 |
+
"learning_rate": 7.55977443242399e-05,
|
2683 |
+
"loss": 2.7125,
|
2684 |
+
"step": 3440
|
2685 |
+
},
|
2686 |
+
{
|
2687 |
+
"epoch": 71.86,
|
2688 |
+
"learning_rate": 7.456472830529259e-05,
|
2689 |
+
"loss": 2.7184,
|
2690 |
+
"step": 3450
|
2691 |
+
},
|
2692 |
+
{
|
2693 |
+
"epoch": 71.86,
|
2694 |
+
"eval_accuracy": 0.44788752382659924,
|
2695 |
+
"eval_loss": 3.213238000869751,
|
2696 |
+
"eval_runtime": 127.9515,
|
2697 |
+
"eval_samples_per_second": 188.001,
|
2698 |
+
"eval_steps_per_second": 5.877,
|
2699 |
+
"step": 3450
|
2700 |
+
},
|
2701 |
+
{
|
2702 |
+
"epoch": 72.08,
|
2703 |
+
"learning_rate": 7.353719923148324e-05,
|
2704 |
+
"loss": 2.8953,
|
2705 |
+
"step": 3460
|
2706 |
+
},
|
2707 |
+
{
|
2708 |
+
"epoch": 72.29,
|
2709 |
+
"learning_rate": 7.251520205026205e-05,
|
2710 |
+
"loss": 2.6971,
|
2711 |
+
"step": 3470
|
2712 |
+
},
|
2713 |
+
{
|
2714 |
+
"epoch": 72.49,
|
2715 |
+
"learning_rate": 7.149878146709676e-05,
|
2716 |
+
"loss": 2.6983,
|
2717 |
+
"step": 3480
|
2718 |
+
},
|
2719 |
+
{
|
2720 |
+
"epoch": 72.7,
|
2721 |
+
"learning_rate": 7.048798194351625e-05,
|
2722 |
+
"loss": 2.7034,
|
2723 |
+
"step": 3490
|
2724 |
+
},
|
2725 |
+
{
|
2726 |
+
"epoch": 72.9,
|
2727 |
+
"learning_rate": 6.948284769516627e-05,
|
2728 |
+
"loss": 2.7077,
|
2729 |
+
"step": 3500
|
2730 |
+
},
|
2731 |
+
{
|
2732 |
+
"epoch": 72.9,
|
2733 |
+
"eval_accuracy": 0.44867082596467595,
|
2734 |
+
"eval_loss": 3.208606243133545,
|
2735 |
+
"eval_runtime": 128.1176,
|
2736 |
+
"eval_samples_per_second": 187.757,
|
2737 |
+
"eval_steps_per_second": 5.87,
|
2738 |
+
"step": 3500
|
2739 |
+
},
|
2740 |
+
{
|
2741 |
+
"epoch": 73.12,
|
2742 |
+
"learning_rate": 6.848342268987511e-05,
|
2743 |
+
"loss": 2.8784,
|
2744 |
+
"step": 3510
|
2745 |
+
},
|
2746 |
+
{
|
2747 |
+
"epoch": 73.33,
|
2748 |
+
"learning_rate": 6.748975064573007e-05,
|
2749 |
+
"loss": 2.694,
|
2750 |
+
"step": 3520
|
2751 |
+
},
|
2752 |
+
{
|
2753 |
+
"epoch": 73.53,
|
2754 |
+
"learning_rate": 6.650187502916552e-05,
|
2755 |
+
"loss": 2.6991,
|
2756 |
+
"step": 3530
|
2757 |
+
},
|
2758 |
+
{
|
2759 |
+
"epoch": 73.74,
|
2760 |
+
"learning_rate": 6.551983905306107e-05,
|
2761 |
+
"loss": 2.7075,
|
2762 |
+
"step": 3540
|
2763 |
+
},
|
2764 |
+
{
|
2765 |
+
"epoch": 73.95,
|
2766 |
+
"learning_rate": 6.454368567485183e-05,
|
2767 |
+
"loss": 2.6916,
|
2768 |
+
"step": 3550
|
2769 |
+
},
|
2770 |
+
{
|
2771 |
+
"epoch": 73.95,
|
2772 |
+
"eval_accuracy": 0.44818311301861347,
|
2773 |
+
"eval_loss": 3.2057085037231445,
|
2774 |
+
"eval_runtime": 128.0769,
|
2775 |
+
"eval_samples_per_second": 187.817,
|
2776 |
+
"eval_steps_per_second": 5.871,
|
2777 |
+
"step": 3550
|
2778 |
+
},
|
2779 |
+
{
|
2780 |
+
"epoch": 74.16,
|
2781 |
+
"learning_rate": 6.35734575946487e-05,
|
2782 |
+
"loss": 2.884,
|
2783 |
+
"step": 3560
|
2784 |
+
},
|
2785 |
+
{
|
2786 |
+
"epoch": 74.37,
|
2787 |
+
"learning_rate": 6.260919725337109e-05,
|
2788 |
+
"loss": 2.6885,
|
2789 |
+
"step": 3570
|
2790 |
+
},
|
2791 |
+
{
|
2792 |
+
"epoch": 74.58,
|
2793 |
+
"learning_rate": 6.165094683089015e-05,
|
2794 |
+
"loss": 2.7009,
|
2795 |
+
"step": 3580
|
2796 |
+
},
|
2797 |
+
{
|
2798 |
+
"epoch": 74.78,
|
2799 |
+
"learning_rate": 6.069874824418356e-05,
|
2800 |
+
"loss": 2.6924,
|
2801 |
+
"step": 3590
|
2802 |
+
},
|
2803 |
+
{
|
2804 |
+
"epoch": 74.99,
|
2805 |
+
"learning_rate": 5.975264314550229e-05,
|
2806 |
+
"loss": 2.6934,
|
2807 |
+
"step": 3600
|
2808 |
+
},
|
2809 |
+
{
|
2810 |
+
"epoch": 74.99,
|
2811 |
+
"eval_accuracy": 0.44951231576252726,
|
2812 |
+
"eval_loss": 3.201040506362915,
|
2813 |
+
"eval_runtime": 128.0178,
|
2814 |
+
"eval_samples_per_second": 187.904,
|
2815 |
+
"eval_steps_per_second": 5.874,
|
2816 |
+
"step": 3600
|
2817 |
+
},
|
2818 |
+
{
|
2819 |
+
"epoch": 75.21,
|
2820 |
+
"learning_rate": 5.881267292054828e-05,
|
2821 |
+
"loss": 2.8607,
|
2822 |
+
"step": 3610
|
2823 |
+
},
|
2824 |
+
{
|
2825 |
+
"epoch": 75.41,
|
2826 |
+
"learning_rate": 5.787887868666417e-05,
|
2827 |
+
"loss": 2.678,
|
2828 |
+
"step": 3620
|
2829 |
+
},
|
2830 |
+
{
|
2831 |
+
"epoch": 75.62,
|
2832 |
+
"learning_rate": 5.6951301291034945e-05,
|
2833 |
+
"loss": 2.696,
|
2834 |
+
"step": 3630
|
2835 |
+
},
|
2836 |
+
{
|
2837 |
+
"epoch": 75.82,
|
2838 |
+
"learning_rate": 5.602998130890065e-05,
|
2839 |
+
"loss": 2.6944,
|
2840 |
+
"step": 3640
|
2841 |
+
},
|
2842 |
+
{
|
2843 |
+
"epoch": 76.04,
|
2844 |
+
"learning_rate": 5.511495904178221e-05,
|
2845 |
+
"loss": 2.8585,
|
2846 |
+
"step": 3650
|
2847 |
+
},
|
2848 |
+
{
|
2849 |
+
"epoch": 76.04,
|
2850 |
+
"eval_accuracy": 0.44973287373334114,
|
2851 |
+
"eval_loss": 3.1979689598083496,
|
2852 |
+
"eval_runtime": 127.9186,
|
2853 |
+
"eval_samples_per_second": 188.049,
|
2854 |
+
"eval_steps_per_second": 5.879,
|
2855 |
+
"step": 3650
|
2856 |
+
},
|
2857 |
+
{
|
2858 |
+
"epoch": 76.25,
|
2859 |
+
"learning_rate": 5.4206274515717736e-05,
|
2860 |
+
"loss": 2.6924,
|
2861 |
+
"step": 3660
|
2862 |
+
},
|
2863 |
+
{
|
2864 |
+
"epoch": 76.45,
|
2865 |
+
"learning_rate": 5.330396747951205e-05,
|
2866 |
+
"loss": 2.6796,
|
2867 |
+
"step": 3670
|
2868 |
+
},
|
2869 |
+
{
|
2870 |
+
"epoch": 76.66,
|
2871 |
+
"learning_rate": 5.240807740299811e-05,
|
2872 |
+
"loss": 2.684,
|
2873 |
+
"step": 3680
|
2874 |
+
},
|
2875 |
+
{
|
2876 |
+
"epoch": 76.86,
|
2877 |
+
"learning_rate": 5.1518643475310034e-05,
|
2878 |
+
"loss": 2.6842,
|
2879 |
+
"step": 3690
|
2880 |
+
},
|
2881 |
+
{
|
2882 |
+
"epoch": 77.08,
|
2883 |
+
"learning_rate": 5.0635704603169287e-05,
|
2884 |
+
"loss": 2.8559,
|
2885 |
+
"step": 3700
|
2886 |
+
},
|
2887 |
+
{
|
2888 |
+
"epoch": 77.08,
|
2889 |
+
"eval_accuracy": 0.4502848474176814,
|
2890 |
+
"eval_loss": 3.1939539909362793,
|
2891 |
+
"eval_runtime": 128.1047,
|
2892 |
+
"eval_samples_per_second": 187.776,
|
2893 |
+
"eval_steps_per_second": 5.87,
|
2894 |
+
"step": 3700
|
2895 |
+
},
|
2896 |
+
{
|
2897 |
+
"epoch": 77.29,
|
2898 |
+
"learning_rate": 4.975929940918236e-05,
|
2899 |
+
"loss": 2.6777,
|
2900 |
+
"step": 3710
|
2901 |
+
},
|
2902 |
+
{
|
2903 |
+
"epoch": 77.49,
|
2904 |
+
"learning_rate": 4.8889466230151646e-05,
|
2905 |
+
"loss": 2.6673,
|
2906 |
+
"step": 3720
|
2907 |
+
},
|
2908 |
+
{
|
2909 |
+
"epoch": 77.7,
|
2910 |
+
"learning_rate": 4.8026243115398314e-05,
|
2911 |
+
"loss": 2.6694,
|
2912 |
+
"step": 3730
|
2913 |
+
},
|
2914 |
+
{
|
2915 |
+
"epoch": 77.9,
|
2916 |
+
"learning_rate": 4.7169667825097775e-05,
|
2917 |
+
"loss": 2.6734,
|
2918 |
+
"step": 3740
|
2919 |
+
},
|
2920 |
+
{
|
2921 |
+
"epoch": 78.12,
|
2922 |
+
"learning_rate": 4.631977782862824e-05,
|
2923 |
+
"loss": 2.8519,
|
2924 |
+
"step": 3750
|
2925 |
+
},
|
2926 |
+
{
|
2927 |
+
"epoch": 78.12,
|
2928 |
+
"eval_accuracy": 0.4506198615318044,
|
2929 |
+
"eval_loss": 3.1939969062805176,
|
2930 |
+
"eval_runtime": 128.0925,
|
2931 |
+
"eval_samples_per_second": 187.794,
|
2932 |
+
"eval_steps_per_second": 5.871,
|
2933 |
+
"step": 3750
|
2934 |
+
},
|
2935 |
+
{
|
2936 |
+
"epoch": 78.33,
|
2937 |
+
"learning_rate": 4.547661030293129e-05,
|
2938 |
+
"loss": 2.6742,
|
2939 |
+
"step": 3760
|
2940 |
+
},
|
2941 |
+
{
|
2942 |
+
"epoch": 78.53,
|
2943 |
+
"learning_rate": 4.464020213088611e-05,
|
2944 |
+
"loss": 2.6767,
|
2945 |
+
"step": 3770
|
2946 |
+
},
|
2947 |
+
{
|
2948 |
+
"epoch": 78.74,
|
2949 |
+
"learning_rate": 4.381058989969564e-05,
|
2950 |
+
"loss": 2.6641,
|
2951 |
+
"step": 3780
|
2952 |
+
},
|
2953 |
+
{
|
2954 |
+
"epoch": 78.95,
|
2955 |
+
"learning_rate": 4.298780989928646e-05,
|
2956 |
+
"loss": 2.6726,
|
2957 |
+
"step": 3790
|
2958 |
+
},
|
2959 |
+
{
|
2960 |
+
"epoch": 79.16,
|
2961 |
+
"learning_rate": 4.217189812072131e-05,
|
2962 |
+
"loss": 2.8391,
|
2963 |
+
"step": 3800
|
2964 |
+
},
|
2965 |
+
{
|
2966 |
+
"epoch": 79.16,
|
2967 |
+
"eval_accuracy": 0.4509423513828209,
|
2968 |
+
"eval_loss": 3.1897408962249756,
|
2969 |
+
"eval_runtime": 127.8217,
|
2970 |
+
"eval_samples_per_second": 188.192,
|
2971 |
+
"eval_steps_per_second": 5.883,
|
2972 |
+
"step": 3800
|
2973 |
+
},
|
2974 |
+
{
|
2975 |
+
"epoch": 79.37,
|
2976 |
+
"learning_rate": 4.136289025462443e-05,
|
2977 |
+
"loss": 2.6616,
|
2978 |
+
"step": 3810
|
2979 |
+
},
|
2980 |
+
{
|
2981 |
+
"epoch": 79.58,
|
2982 |
+
"learning_rate": 4.0560821689620856e-05,
|
2983 |
+
"loss": 2.6701,
|
2984 |
+
"step": 3820
|
2985 |
+
},
|
2986 |
+
{
|
2987 |
+
"epoch": 79.78,
|
2988 |
+
"learning_rate": 3.976572751078782e-05,
|
2989 |
+
"loss": 2.6546,
|
2990 |
+
"step": 3830
|
2991 |
+
},
|
2992 |
+
{
|
2993 |
+
"epoch": 79.99,
|
2994 |
+
"learning_rate": 3.8977642498120594e-05,
|
2995 |
+
"loss": 2.6719,
|
2996 |
+
"step": 3840
|
2997 |
+
},
|
2998 |
+
{
|
2999 |
+
"epoch": 80.21,
|
3000 |
+
"learning_rate": 3.819660112501053e-05,
|
3001 |
+
"loss": 2.845,
|
3002 |
+
"step": 3850
|
3003 |
+
},
|
3004 |
+
{
|
3005 |
+
"epoch": 80.21,
|
3006 |
+
"eval_accuracy": 0.45101718878618524,
|
3007 |
+
"eval_loss": 3.1857643127441406,
|
3008 |
+
"eval_runtime": 127.9876,
|
3009 |
+
"eval_samples_per_second": 187.948,
|
3010 |
+
"eval_steps_per_second": 5.876,
|
3011 |
+
"step": 3850
|
3012 |
+
},
|
3013 |
+
{
|
3014 |
+
"epoch": 80.41,
|
3015 |
+
"learning_rate": 3.742263755673758e-05,
|
3016 |
+
"loss": 2.6657,
|
3017 |
+
"step": 3860
|
3018 |
+
},
|
3019 |
+
{
|
3020 |
+
"epoch": 80.62,
|
3021 |
+
"learning_rate": 3.6655785648975585e-05,
|
3022 |
+
"loss": 2.6601,
|
3023 |
+
"step": 3870
|
3024 |
+
},
|
3025 |
+
{
|
3026 |
+
"epoch": 80.82,
|
3027 |
+
"learning_rate": 3.589607894631111e-05,
|
3028 |
+
"loss": 2.6666,
|
3029 |
+
"step": 3880
|
3030 |
+
},
|
3031 |
+
{
|
3032 |
+
"epoch": 81.04,
|
3033 |
+
"learning_rate": 3.514355068077655e-05,
|
3034 |
+
"loss": 2.8323,
|
3035 |
+
"step": 3890
|
3036 |
+
},
|
3037 |
+
{
|
3038 |
+
"epoch": 81.25,
|
3039 |
+
"learning_rate": 3.439823377039599e-05,
|
3040 |
+
"loss": 2.6636,
|
3041 |
+
"step": 3900
|
3042 |
+
},
|
3043 |
+
{
|
3044 |
+
"epoch": 81.25,
|
3045 |
+
"eval_accuracy": 0.45183219751680725,
|
3046 |
+
"eval_loss": 3.1818630695343018,
|
3047 |
+
"eval_runtime": 128.0672,
|
3048 |
+
"eval_samples_per_second": 187.831,
|
3049 |
+
"eval_steps_per_second": 5.872,
|
3050 |
+
"step": 3900
|
3051 |
+
},
|
3052 |
+
{
|
3053 |
+
"epoch": 81.45,
|
3054 |
+
"learning_rate": 3.36601608177457e-05,
|
3055 |
+
"loss": 2.6586,
|
3056 |
+
"step": 3910
|
3057 |
+
},
|
3058 |
+
{
|
3059 |
+
"epoch": 81.66,
|
3060 |
+
"learning_rate": 3.292936410852754e-05,
|
3061 |
+
"loss": 2.6674,
|
3062 |
+
"step": 3920
|
3063 |
+
},
|
3064 |
+
{
|
3065 |
+
"epoch": 81.86,
|
3066 |
+
"learning_rate": 3.220587561015709e-05,
|
3067 |
+
"loss": 2.6689,
|
3068 |
+
"step": 3930
|
3069 |
+
},
|
3070 |
+
{
|
3071 |
+
"epoch": 82.08,
|
3072 |
+
"learning_rate": 3.148972697036507e-05,
|
3073 |
+
"loss": 2.8232,
|
3074 |
+
"step": 3940
|
3075 |
+
},
|
3076 |
+
{
|
3077 |
+
"epoch": 82.29,
|
3078 |
+
"learning_rate": 3.078094951581289e-05,
|
3079 |
+
"loss": 2.6569,
|
3080 |
+
"step": 3950
|
3081 |
+
},
|
3082 |
+
{
|
3083 |
+
"epoch": 82.29,
|
3084 |
+
"eval_accuracy": 0.4517055966540888,
|
3085 |
+
"eval_loss": 3.183380603790283,
|
3086 |
+
"eval_runtime": 128.002,
|
3087 |
+
"eval_samples_per_second": 187.927,
|
3088 |
+
"eval_steps_per_second": 5.875,
|
3089 |
+
"step": 3950
|
3090 |
+
},
|
3091 |
+
{
|
3092 |
+
"epoch": 82.49,
|
3093 |
+
"learning_rate": 3.007957425072265e-05,
|
3094 |
+
"loss": 2.6544,
|
3095 |
+
"step": 3960
|
3096 |
+
},
|
3097 |
+
{
|
3098 |
+
"epoch": 82.7,
|
3099 |
+
"learning_rate": 2.9385631855520546e-05,
|
3100 |
+
"loss": 2.6622,
|
3101 |
+
"step": 3970
|
3102 |
+
},
|
3103 |
+
{
|
3104 |
+
"epoch": 82.9,
|
3105 |
+
"learning_rate": 2.8699152685494925e-05,
|
3106 |
+
"loss": 2.6505,
|
3107 |
+
"step": 3980
|
3108 |
+
},
|
3109 |
+
{
|
3110 |
+
"epoch": 83.12,
|
3111 |
+
"learning_rate": 2.8020166769468616e-05,
|
3112 |
+
"loss": 2.8267,
|
3113 |
+
"step": 3990
|
3114 |
+
},
|
3115 |
+
{
|
3116 |
+
"epoch": 83.33,
|
3117 |
+
"learning_rate": 2.7348703808485223e-05,
|
3118 |
+
"loss": 2.647,
|
3119 |
+
"step": 4000
|
3120 |
+
},
|
3121 |
+
{
|
3122 |
+
"epoch": 83.33,
|
3123 |
+
"eval_accuracy": 0.45166349665740935,
|
3124 |
+
"eval_loss": 3.1797752380371094,
|
3125 |
+
"eval_runtime": 128.0064,
|
3126 |
+
"eval_samples_per_second": 187.92,
|
3127 |
+
"eval_steps_per_second": 5.875,
|
3128 |
+
"step": 4000
|
3129 |
+
},
|
3130 |
+
{
|
3131 |
+
"epoch": 83.53,
|
3132 |
+
"learning_rate": 2.6684793174509915e-05,
|
3133 |
+
"loss": 2.6432,
|
3134 |
+
"step": 4010
|
3135 |
+
},
|
3136 |
+
{
|
3137 |
+
"epoch": 83.74,
|
3138 |
+
"learning_rate": 2.6028463909144574e-05,
|
3139 |
+
"loss": 2.6626,
|
3140 |
+
"step": 4020
|
3141 |
+
},
|
3142 |
+
{
|
3143 |
+
"epoch": 83.95,
|
3144 |
+
"learning_rate": 2.5379744722357403e-05,
|
3145 |
+
"loss": 2.6586,
|
3146 |
+
"step": 4030
|
3147 |
+
},
|
3148 |
+
{
|
3149 |
+
"epoch": 84.16,
|
3150 |
+
"learning_rate": 2.473866399122733e-05,
|
3151 |
+
"loss": 2.8349,
|
3152 |
+
"step": 4040
|
3153 |
+
},
|
3154 |
+
{
|
3155 |
+
"epoch": 84.37,
|
3156 |
+
"learning_rate": 2.410524975870221e-05,
|
3157 |
+
"loss": 2.6665,
|
3158 |
+
"step": 4050
|
3159 |
+
},
|
3160 |
+
{
|
3161 |
+
"epoch": 84.37,
|
3162 |
+
"eval_accuracy": 0.45251206379734554,
|
3163 |
+
"eval_loss": 3.178643226623535,
|
3164 |
+
"eval_runtime": 127.9924,
|
3165 |
+
"eval_samples_per_second": 187.941,
|
3166 |
+
"eval_steps_per_second": 5.875,
|
3167 |
+
"step": 4050
|
3168 |
+
},
|
3169 |
+
{
|
3170 |
+
"epoch": 84.58,
|
3171 |
+
"learning_rate": 2.347952973237262e-05,
|
3172 |
+
"loss": 2.6462,
|
3173 |
+
"step": 4060
|
3174 |
+
},
|
3175 |
+
{
|
3176 |
+
"epoch": 84.78,
|
3177 |
+
"learning_rate": 2.286153128325954e-05,
|
3178 |
+
"loss": 2.6444,
|
3179 |
+
"step": 4070
|
3180 |
+
},
|
3181 |
+
{
|
3182 |
+
"epoch": 84.99,
|
3183 |
+
"learning_rate": 2.2251281444617257e-05,
|
3184 |
+
"loss": 2.6442,
|
3185 |
+
"step": 4080
|
3186 |
+
},
|
3187 |
+
{
|
3188 |
+
"epoch": 85.21,
|
3189 |
+
"learning_rate": 2.1648806910750575e-05,
|
3190 |
+
"loss": 2.8258,
|
3191 |
+
"step": 4090
|
3192 |
+
},
|
3193 |
+
{
|
3194 |
+
"epoch": 85.41,
|
3195 |
+
"learning_rate": 2.1054134035847307e-05,
|
3196 |
+
"loss": 2.6382,
|
3197 |
+
"step": 4100
|
3198 |
+
},
|
3199 |
+
{
|
3200 |
+
"epoch": 85.41,
|
3201 |
+
"eval_accuracy": 0.4524524692406656,
|
3202 |
+
"eval_loss": 3.173250198364258,
|
3203 |
+
"eval_runtime": 127.972,
|
3204 |
+
"eval_samples_per_second": 187.971,
|
3205 |
+
"eval_steps_per_second": 5.876,
|
3206 |
+
"step": 4100
|
3207 |
+
},
|
3208 |
+
{
|
3209 |
+
"epoch": 85.62,
|
3210 |
+
"learning_rate": 2.0467288832825583e-05,
|
3211 |
+
"loss": 2.6655,
|
3212 |
+
"step": 4110
|
3213 |
+
},
|
3214 |
+
{
|
3215 |
+
"epoch": 85.82,
|
3216 |
+
"learning_rate": 1.9888296972195587e-05,
|
3217 |
+
"loss": 2.6459,
|
3218 |
+
"step": 4120
|
3219 |
+
},
|
3220 |
+
{
|
3221 |
+
"epoch": 86.04,
|
3222 |
+
"learning_rate": 1.931718378093703e-05,
|
3223 |
+
"loss": 2.8333,
|
3224 |
+
"step": 4130
|
3225 |
+
},
|
3226 |
+
{
|
3227 |
+
"epoch": 86.25,
|
3228 |
+
"learning_rate": 1.875397424139109e-05,
|
3229 |
+
"loss": 2.6533,
|
3230 |
+
"step": 4140
|
3231 |
+
},
|
3232 |
+
{
|
3233 |
+
"epoch": 86.45,
|
3234 |
+
"learning_rate": 1.81986929901675e-05,
|
3235 |
+
"loss": 2.6346,
|
3236 |
+
"step": 4150
|
3237 |
+
},
|
3238 |
+
{
|
3239 |
+
"epoch": 86.45,
|
3240 |
+
"eval_accuracy": 0.4532207710219251,
|
3241 |
+
"eval_loss": 3.1699652671813965,
|
3242 |
+
"eval_runtime": 128.0545,
|
3243 |
+
"eval_samples_per_second": 187.85,
|
3244 |
+
"eval_steps_per_second": 5.872,
|
3245 |
+
"step": 4150
|
3246 |
+
},
|
3247 |
+
{
|
3248 |
+
"epoch": 86.66,
|
3249 |
+
"learning_rate": 1.765136431706711e-05,
|
3250 |
+
"loss": 2.6558,
|
3251 |
+
"step": 4160
|
3252 |
+
},
|
3253 |
+
{
|
3254 |
+
"epoch": 86.86,
|
3255 |
+
"learning_rate": 1.711201216401912e-05,
|
3256 |
+
"loss": 2.6422,
|
3257 |
+
"step": 4170
|
3258 |
+
},
|
3259 |
+
{
|
3260 |
+
"epoch": 87.08,
|
3261 |
+
"learning_rate": 1.6580660124034032e-05,
|
3262 |
+
"loss": 2.8243,
|
3263 |
+
"step": 4180
|
3264 |
+
},
|
3265 |
+
{
|
3266 |
+
"epoch": 87.29,
|
3267 |
+
"learning_rate": 1.605733144017132e-05,
|
3268 |
+
"loss": 2.6443,
|
3269 |
+
"step": 4190
|
3270 |
+
},
|
3271 |
+
{
|
3272 |
+
"epoch": 87.49,
|
3273 |
+
"learning_rate": 1.5542049004523053e-05,
|
3274 |
+
"loss": 2.6457,
|
3275 |
+
"step": 4200
|
3276 |
+
},
|
3277 |
+
{
|
3278 |
+
"epoch": 87.49,
|
3279 |
+
"eval_accuracy": 0.45290158205139586,
|
3280 |
+
"eval_loss": 3.1713671684265137,
|
3281 |
+
"eval_runtime": 127.871,
|
3282 |
+
"eval_samples_per_second": 188.119,
|
3283 |
+
"eval_steps_per_second": 5.881,
|
3284 |
+
"step": 4200
|
3285 |
+
},
|
3286 |
+
{
|
3287 |
+
"epoch": 87.7,
|
3288 |
+
"learning_rate": 1.503483535721224e-05,
|
3289 |
+
"loss": 2.6578,
|
3290 |
+
"step": 4210
|
3291 |
+
},
|
3292 |
+
{
|
3293 |
+
"epoch": 87.9,
|
3294 |
+
"learning_rate": 1.4535712685406921e-05,
|
3295 |
+
"loss": 2.646,
|
3296 |
+
"step": 4220
|
3297 |
+
},
|
3298 |
+
{
|
3299 |
+
"epoch": 88.12,
|
3300 |
+
"learning_rate": 1.4044702822349731e-05,
|
3301 |
+
"loss": 2.8075,
|
3302 |
+
"step": 4230
|
3303 |
+
},
|
3304 |
+
{
|
3305 |
+
"epoch": 88.33,
|
3306 |
+
"learning_rate": 1.3561827246402692e-05,
|
3307 |
+
"loss": 2.6405,
|
3308 |
+
"step": 4240
|
3309 |
+
},
|
3310 |
+
{
|
3311 |
+
"epoch": 88.53,
|
3312 |
+
"learning_rate": 1.3087107080107853e-05,
|
3313 |
+
"loss": 2.6328,
|
3314 |
+
"step": 4250
|
3315 |
+
},
|
3316 |
+
{
|
3317 |
+
"epoch": 88.53,
|
3318 |
+
"eval_accuracy": 0.45369892675477164,
|
3319 |
+
"eval_loss": 3.168638229370117,
|
3320 |
+
"eval_runtime": 128.0728,
|
3321 |
+
"eval_samples_per_second": 187.823,
|
3322 |
+
"eval_steps_per_second": 5.872,
|
3323 |
+
"step": 4250
|
3324 |
+
},
|
3325 |
+
{
|
3326 |
+
"epoch": 88.74,
|
3327 |
+
"learning_rate": 1.2620563089263093e-05,
|
3328 |
+
"loss": 2.6377,
|
3329 |
+
"step": 4260
|
3330 |
+
},
|
3331 |
+
{
|
3332 |
+
"epoch": 88.95,
|
3333 |
+
"learning_rate": 1.2162215682014012e-05,
|
3334 |
+
"loss": 2.6645,
|
3335 |
+
"step": 4270
|
3336 |
+
},
|
3337 |
+
{
|
3338 |
+
"epoch": 89.16,
|
3339 |
+
"learning_rate": 1.1712084907961053e-05,
|
3340 |
+
"loss": 2.8112,
|
3341 |
+
"step": 4280
|
3342 |
+
},
|
3343 |
+
{
|
3344 |
+
"epoch": 89.37,
|
3345 |
+
"learning_rate": 1.127019045728246e-05,
|
3346 |
+
"loss": 2.6445,
|
3347 |
+
"step": 4290
|
3348 |
+
},
|
3349 |
+
{
|
3350 |
+
"epoch": 89.58,
|
3351 |
+
"learning_rate": 1.0836551659873074e-05,
|
3352 |
+
"loss": 2.6429,
|
3353 |
+
"step": 4300
|
3354 |
+
},
|
3355 |
+
{
|
3356 |
+
"epoch": 89.58,
|
3357 |
+
"eval_accuracy": 0.4534057770055075,
|
3358 |
+
"eval_loss": 3.171478033065796,
|
3359 |
+
"eval_runtime": 128.1308,
|
3360 |
+
"eval_samples_per_second": 187.738,
|
3361 |
+
"eval_steps_per_second": 5.869,
|
3362 |
+
"step": 4300
|
3363 |
+
},
|
3364 |
+
{
|
3365 |
+
"epoch": 89.78,
|
3366 |
+
"learning_rate": 1.0411187484498652e-05,
|
3367 |
+
"loss": 2.6458,
|
3368 |
+
"step": 4310
|
3369 |
+
},
|
3370 |
+
{
|
3371 |
+
"epoch": 89.99,
|
3372 |
+
"learning_rate": 9.99411653796627e-06,
|
3373 |
+
"loss": 2.6433,
|
3374 |
+
"step": 4320
|
3375 |
+
},
|
3376 |
+
{
|
3377 |
+
"epoch": 90.21,
|
3378 |
+
"learning_rate": 9.58535706431023e-06,
|
3379 |
+
"loss": 2.8107,
|
3380 |
+
"step": 4330
|
3381 |
+
},
|
3382 |
+
{
|
3383 |
+
"epoch": 90.41,
|
3384 |
+
"learning_rate": 9.184926943994044e-06,
|
3385 |
+
"loss": 2.6428,
|
3386 |
+
"step": 4340
|
3387 |
+
},
|
3388 |
+
{
|
3389 |
+
"epoch": 90.62,
|
3390 |
+
"learning_rate": 8.792843693128471e-06,
|
3391 |
+
"loss": 2.6369,
|
3392 |
+
"step": 4350
|
3393 |
+
},
|
3394 |
+
{
|
3395 |
+
"epoch": 90.62,
|
3396 |
+
"eval_accuracy": 0.4537699711828984,
|
3397 |
+
"eval_loss": 3.1687278747558594,
|
3398 |
+
"eval_runtime": 127.8724,
|
3399 |
+
"eval_samples_per_second": 188.117,
|
3400 |
+
"eval_steps_per_second": 5.881,
|
3401 |
+
"step": 4350
|
3402 |
+
},
|
3403 |
+
{
|
3404 |
+
"epoch": 90.82,
|
3405 |
+
"learning_rate": 8.409124462705032e-06,
|
3406 |
+
"loss": 2.6513,
|
3407 |
+
"step": 4360
|
3408 |
+
},
|
3409 |
+
{
|
3410 |
+
"epoch": 91.04,
|
3411 |
+
"learning_rate": 8.033786037845992e-06,
|
3412 |
+
"loss": 2.8,
|
3413 |
+
"step": 4370
|
3414 |
+
},
|
3415 |
+
{
|
3416 |
+
"epoch": 91.25,
|
3417 |
+
"learning_rate": 7.66684483706992e-06,
|
3418 |
+
"loss": 2.6456,
|
3419 |
+
"step": 4380
|
3420 |
+
},
|
3421 |
+
{
|
3422 |
+
"epoch": 91.45,
|
3423 |
+
"learning_rate": 7.308316911573721e-06,
|
3424 |
+
"loss": 2.6429,
|
3425 |
+
"step": 4390
|
3426 |
+
},
|
3427 |
+
{
|
3428 |
+
"epoch": 91.66,
|
3429 |
+
"learning_rate": 6.958217944530287e-06,
|
3430 |
+
"loss": 2.628,
|
3431 |
+
"step": 4400
|
3432 |
+
},
|
3433 |
+
{
|
3434 |
+
"epoch": 91.66,
|
3435 |
+
"eval_accuracy": 0.4539267525748088,
|
3436 |
+
"eval_loss": 3.165127754211426,
|
3437 |
+
"eval_runtime": 128.0531,
|
3438 |
+
"eval_samples_per_second": 187.852,
|
3439 |
+
"eval_steps_per_second": 5.873,
|
3440 |
+
"step": 4400
|
3441 |
+
},
|
3442 |
+
{
|
3443 |
+
"epoch": 91.86,
|
3444 |
+
"learning_rate": 6.616563250402585e-06,
|
3445 |
+
"loss": 2.6337,
|
3446 |
+
"step": 4410
|
3447 |
+
},
|
3448 |
+
{
|
3449 |
+
"epoch": 92.08,
|
3450 |
+
"learning_rate": 6.283367774273785e-06,
|
3451 |
+
"loss": 2.8133,
|
3452 |
+
"step": 4420
|
3453 |
+
},
|
3454 |
+
{
|
3455 |
+
"epoch": 92.29,
|
3456 |
+
"learning_rate": 5.958646091193387e-06,
|
3457 |
+
"loss": 2.6318,
|
3458 |
+
"step": 4430
|
3459 |
+
},
|
3460 |
+
{
|
3461 |
+
"epoch": 92.49,
|
3462 |
+
"learning_rate": 5.642412405539798e-06,
|
3463 |
+
"loss": 2.6365,
|
3464 |
+
"step": 4440
|
3465 |
+
},
|
3466 |
+
{
|
3467 |
+
"epoch": 92.7,
|
3468 |
+
"learning_rate": 5.334680550398852e-06,
|
3469 |
+
"loss": 2.6373,
|
3470 |
+
"step": 4450
|
3471 |
+
},
|
3472 |
+
{
|
3473 |
+
"epoch": 92.7,
|
3474 |
+
"eval_accuracy": 0.4538699774217477,
|
3475 |
+
"eval_loss": 3.1659765243530273,
|
3476 |
+
"eval_runtime": 128.2044,
|
3477 |
+
"eval_samples_per_second": 187.63,
|
3478 |
+
"eval_steps_per_second": 5.866,
|
3479 |
+
"step": 4450
|
3480 |
+
},
|
3481 |
+
{
|
3482 |
+
"epoch": 92.9,
|
3483 |
+
"learning_rate": 5.0354639869588e-06,
|
3484 |
+
"loss": 2.6355,
|
3485 |
+
"step": 4460
|
3486 |
+
},
|
3487 |
+
{
|
3488 |
+
"epoch": 93.12,
|
3489 |
+
"learning_rate": 4.744775803921475e-06,
|
3490 |
+
"loss": 2.8102,
|
3491 |
+
"step": 4470
|
3492 |
+
},
|
3493 |
+
{
|
3494 |
+
"epoch": 93.33,
|
3495 |
+
"learning_rate": 4.4626287169296846e-06,
|
3496 |
+
"loss": 2.6362,
|
3497 |
+
"step": 4480
|
3498 |
+
},
|
3499 |
+
{
|
3500 |
+
"epoch": 93.53,
|
3501 |
+
"learning_rate": 4.189035068011071e-06,
|
3502 |
+
"loss": 2.6226,
|
3503 |
+
"step": 4490
|
3504 |
+
},
|
3505 |
+
{
|
3506 |
+
"epoch": 93.74,
|
3507 |
+
"learning_rate": 3.924006825038129e-06,
|
3508 |
+
"loss": 2.6357,
|
3509 |
+
"step": 4500
|
3510 |
+
},
|
3511 |
+
{
|
3512 |
+
"epoch": 93.74,
|
3513 |
+
"eval_accuracy": 0.4537227251216398,
|
3514 |
+
"eval_loss": 3.1661999225616455,
|
3515 |
+
"eval_runtime": 128.0693,
|
3516 |
+
"eval_samples_per_second": 187.828,
|
3517 |
+
"eval_steps_per_second": 5.872,
|
3518 |
+
"step": 4500
|
3519 |
+
},
|
3520 |
+
{
|
3521 |
+
"epoch": 93.95,
|
3522 |
+
"learning_rate": 3.6675555812047956e-06,
|
3523 |
+
"loss": 2.6477,
|
3524 |
+
"step": 4510
|
3525 |
+
},
|
3526 |
+
{
|
3527 |
+
"epoch": 94.16,
|
3528 |
+
"learning_rate": 3.4196925545192604e-06,
|
3529 |
+
"loss": 2.808,
|
3530 |
+
"step": 4520
|
3531 |
+
},
|
3532 |
+
{
|
3533 |
+
"epoch": 94.37,
|
3534 |
+
"learning_rate": 3.1804285873132668e-06,
|
3535 |
+
"loss": 2.6339,
|
3536 |
+
"step": 4530
|
3537 |
+
},
|
3538 |
+
{
|
3539 |
+
"epoch": 94.58,
|
3540 |
+
"learning_rate": 2.9497741457678695e-06,
|
3541 |
+
"loss": 2.6228,
|
3542 |
+
"step": 4540
|
3543 |
+
},
|
3544 |
+
{
|
3545 |
+
"epoch": 94.78,
|
3546 |
+
"learning_rate": 2.7277393194555358e-06,
|
3547 |
+
"loss": 2.6302,
|
3548 |
+
"step": 4550
|
3549 |
+
},
|
3550 |
+
{
|
3551 |
+
"epoch": 94.78,
|
3552 |
+
"eval_accuracy": 0.45330473365422363,
|
3553 |
+
"eval_loss": 3.1695384979248047,
|
3554 |
+
"eval_runtime": 128.0386,
|
3555 |
+
"eval_samples_per_second": 187.873,
|
3556 |
+
"eval_steps_per_second": 5.873,
|
3557 |
+
"step": 4550
|
3558 |
+
},
|
3559 |
+
{
|
3560 |
+
"epoch": 94.78,
|
3561 |
+
"step": 4550,
|
3562 |
+
"total_flos": 1.602912650550436e+17,
|
3563 |
+
"train_loss": 3.832834072322636,
|
3564 |
+
"train_runtime": 39205.3218,
|
3565 |
+
"train_samples_per_second": 63.537,
|
3566 |
+
"train_steps_per_second": 0.122
|
3567 |
+
}
|
3568 |
+
],
|
3569 |
+
"max_steps": 4800,
|
3570 |
+
"num_train_epochs": 100,
|
3571 |
+
"total_flos": 1.602912650550436e+17,
|
3572 |
+
"trial_name": null,
|
3573 |
+
"trial_params": null
|
3574 |
+
}
|