Commit
•
2edb7e4
1
Parent(s):
a235b68
End of training
Browse files- all_results.json +8 -0
- test_results.json +8 -0
- trainer_state.json +2181 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 48.02,
|
| 3 |
+
"eval_accuracy": 0.5,
|
| 4 |
+
"eval_loss": 1.6418968439102173,
|
| 5 |
+
"eval_runtime": 4.5735,
|
| 6 |
+
"eval_samples_per_second": 5.248,
|
| 7 |
+
"eval_steps_per_second": 1.312
|
| 8 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 48.02,
|
| 3 |
+
"eval_accuracy": 0.5,
|
| 4 |
+
"eval_loss": 1.6418968439102173,
|
| 5 |
+
"eval_runtime": 4.5735,
|
| 6 |
+
"eval_samples_per_second": 5.248,
|
| 7 |
+
"eval_steps_per_second": 1.312
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2181 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7368421052631579,
|
| 3 |
+
"best_model_checkpoint": "MAE-CT-CPC-Dicotomized-n0-m1-d1-v9/checkpoint-735",
|
| 4 |
+
"epoch": 48.02,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2400,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.004166666666666667,
|
| 13 |
+
"grad_norm": 2.6540045738220215,
|
| 14 |
+
"learning_rate": 4.1666666666666667e-07,
|
| 15 |
+
"loss": 0.6936,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.008333333333333333,
|
| 20 |
+
"grad_norm": 3.005134344100952,
|
| 21 |
+
"learning_rate": 8.333333333333333e-07,
|
| 22 |
+
"loss": 0.6954,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0125,
|
| 27 |
+
"grad_norm": 2.9051997661590576,
|
| 28 |
+
"learning_rate": 1.25e-06,
|
| 29 |
+
"loss": 0.6957,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.016666666666666666,
|
| 34 |
+
"grad_norm": 5.865764617919922,
|
| 35 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 36 |
+
"loss": 0.6819,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.020416666666666666,
|
| 41 |
+
"eval_accuracy": 0.5789473684210527,
|
| 42 |
+
"eval_loss": 0.6640610694885254,
|
| 43 |
+
"eval_runtime": 4.1603,
|
| 44 |
+
"eval_samples_per_second": 4.567,
|
| 45 |
+
"eval_steps_per_second": 1.202,
|
| 46 |
+
"step": 49
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.0004166666666667,
|
| 50 |
+
"grad_norm": 2.7976856231689453,
|
| 51 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 52 |
+
"loss": 0.6419,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.0045833333333334,
|
| 57 |
+
"grad_norm": 4.449363708496094,
|
| 58 |
+
"learning_rate": 2.5e-06,
|
| 59 |
+
"loss": 0.6596,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.00875,
|
| 64 |
+
"grad_norm": 3.9453043937683105,
|
| 65 |
+
"learning_rate": 2.916666666666667e-06,
|
| 66 |
+
"loss": 0.7489,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.0129166666666667,
|
| 71 |
+
"grad_norm": 4.261590480804443,
|
| 72 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 73 |
+
"loss": 0.6614,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.0170833333333333,
|
| 78 |
+
"grad_norm": 8.211638450622559,
|
| 79 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 80 |
+
"loss": 0.6201,
|
| 81 |
+
"step": 90
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.0204166666666667,
|
| 85 |
+
"eval_accuracy": 0.5789473684210527,
|
| 86 |
+
"eval_loss": 0.6367576718330383,
|
| 87 |
+
"eval_runtime": 3.6676,
|
| 88 |
+
"eval_samples_per_second": 5.181,
|
| 89 |
+
"eval_steps_per_second": 1.363,
|
| 90 |
+
"step": 98
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.0008333333333335,
|
| 94 |
+
"grad_norm": 5.028270244598389,
|
| 95 |
+
"learning_rate": 4.166666666666667e-06,
|
| 96 |
+
"loss": 0.6964,
|
| 97 |
+
"step": 100
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.005,
|
| 101 |
+
"grad_norm": 4.883535861968994,
|
| 102 |
+
"learning_rate": 4.583333333333333e-06,
|
| 103 |
+
"loss": 0.6823,
|
| 104 |
+
"step": 110
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.0091666666666668,
|
| 108 |
+
"grad_norm": 13.357542037963867,
|
| 109 |
+
"learning_rate": 5e-06,
|
| 110 |
+
"loss": 0.6742,
|
| 111 |
+
"step": 120
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.013333333333333,
|
| 115 |
+
"grad_norm": 6.5849480628967285,
|
| 116 |
+
"learning_rate": 5.416666666666667e-06,
|
| 117 |
+
"loss": 0.5546,
|
| 118 |
+
"step": 130
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.0175,
|
| 122 |
+
"grad_norm": 5.240979194641113,
|
| 123 |
+
"learning_rate": 5.833333333333334e-06,
|
| 124 |
+
"loss": 0.6787,
|
| 125 |
+
"step": 140
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.0204166666666667,
|
| 129 |
+
"eval_accuracy": 0.5789473684210527,
|
| 130 |
+
"eval_loss": 0.6108768582344055,
|
| 131 |
+
"eval_runtime": 3.6774,
|
| 132 |
+
"eval_samples_per_second": 5.167,
|
| 133 |
+
"eval_steps_per_second": 1.36,
|
| 134 |
+
"step": 147
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.00125,
|
| 138 |
+
"grad_norm": 7.692701816558838,
|
| 139 |
+
"learning_rate": 6.25e-06,
|
| 140 |
+
"loss": 0.7031,
|
| 141 |
+
"step": 150
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.0054166666666666,
|
| 145 |
+
"grad_norm": 2.450507164001465,
|
| 146 |
+
"learning_rate": 6.666666666666667e-06,
|
| 147 |
+
"loss": 0.6611,
|
| 148 |
+
"step": 160
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 3.0095833333333335,
|
| 152 |
+
"grad_norm": 7.620597839355469,
|
| 153 |
+
"learning_rate": 7.083333333333335e-06,
|
| 154 |
+
"loss": 0.6429,
|
| 155 |
+
"step": 170
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.01375,
|
| 159 |
+
"grad_norm": 8.6759033203125,
|
| 160 |
+
"learning_rate": 7.500000000000001e-06,
|
| 161 |
+
"loss": 0.6269,
|
| 162 |
+
"step": 180
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 3.017916666666667,
|
| 166 |
+
"grad_norm": 8.059222221374512,
|
| 167 |
+
"learning_rate": 7.916666666666667e-06,
|
| 168 |
+
"loss": 0.6484,
|
| 169 |
+
"step": 190
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 3.0204166666666667,
|
| 173 |
+
"eval_accuracy": 0.631578947368421,
|
| 174 |
+
"eval_loss": 0.6304681897163391,
|
| 175 |
+
"eval_runtime": 3.616,
|
| 176 |
+
"eval_samples_per_second": 5.254,
|
| 177 |
+
"eval_steps_per_second": 1.383,
|
| 178 |
+
"step": 196
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.001666666666667,
|
| 182 |
+
"grad_norm": 7.98549222946167,
|
| 183 |
+
"learning_rate": 8.333333333333334e-06,
|
| 184 |
+
"loss": 0.6938,
|
| 185 |
+
"step": 200
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 4.005833333333333,
|
| 189 |
+
"grad_norm": 8.134982109069824,
|
| 190 |
+
"learning_rate": 8.750000000000001e-06,
|
| 191 |
+
"loss": 0.5986,
|
| 192 |
+
"step": 210
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 4.01,
|
| 196 |
+
"grad_norm": 15.768097877502441,
|
| 197 |
+
"learning_rate": 9.166666666666666e-06,
|
| 198 |
+
"loss": 0.3969,
|
| 199 |
+
"step": 220
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 4.014166666666667,
|
| 203 |
+
"grad_norm": 11.637251853942871,
|
| 204 |
+
"learning_rate": 9.583333333333335e-06,
|
| 205 |
+
"loss": 0.6835,
|
| 206 |
+
"step": 230
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 4.0183333333333335,
|
| 210 |
+
"grad_norm": 63.66741180419922,
|
| 211 |
+
"learning_rate": 1e-05,
|
| 212 |
+
"loss": 0.6357,
|
| 213 |
+
"step": 240
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 4.020416666666667,
|
| 217 |
+
"eval_accuracy": 0.5263157894736842,
|
| 218 |
+
"eval_loss": 0.6793826818466187,
|
| 219 |
+
"eval_runtime": 3.6154,
|
| 220 |
+
"eval_samples_per_second": 5.255,
|
| 221 |
+
"eval_steps_per_second": 1.383,
|
| 222 |
+
"step": 245
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 5.002083333333333,
|
| 226 |
+
"grad_norm": 10.716292381286621,
|
| 227 |
+
"learning_rate": 9.953703703703704e-06,
|
| 228 |
+
"loss": 0.6267,
|
| 229 |
+
"step": 250
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 5.00625,
|
| 233 |
+
"grad_norm": 10.738350868225098,
|
| 234 |
+
"learning_rate": 9.907407407407408e-06,
|
| 235 |
+
"loss": 0.6908,
|
| 236 |
+
"step": 260
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 5.010416666666667,
|
| 240 |
+
"grad_norm": 7.5224690437316895,
|
| 241 |
+
"learning_rate": 9.861111111111112e-06,
|
| 242 |
+
"loss": 0.6093,
|
| 243 |
+
"step": 270
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 5.014583333333333,
|
| 247 |
+
"grad_norm": 15.3920259475708,
|
| 248 |
+
"learning_rate": 9.814814814814815e-06,
|
| 249 |
+
"loss": 0.5489,
|
| 250 |
+
"step": 280
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 5.01875,
|
| 254 |
+
"grad_norm": 17.959043502807617,
|
| 255 |
+
"learning_rate": 9.768518518518519e-06,
|
| 256 |
+
"loss": 0.5692,
|
| 257 |
+
"step": 290
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 5.020416666666667,
|
| 261 |
+
"eval_accuracy": 0.5263157894736842,
|
| 262 |
+
"eval_loss": 0.6937870979309082,
|
| 263 |
+
"eval_runtime": 3.6363,
|
| 264 |
+
"eval_samples_per_second": 5.225,
|
| 265 |
+
"eval_steps_per_second": 1.375,
|
| 266 |
+
"step": 294
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 6.0025,
|
| 270 |
+
"grad_norm": 22.7791690826416,
|
| 271 |
+
"learning_rate": 9.722222222222223e-06,
|
| 272 |
+
"loss": 0.6108,
|
| 273 |
+
"step": 300
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 6.006666666666667,
|
| 277 |
+
"grad_norm": 12.141826629638672,
|
| 278 |
+
"learning_rate": 9.675925925925926e-06,
|
| 279 |
+
"loss": 0.7977,
|
| 280 |
+
"step": 310
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 6.010833333333333,
|
| 284 |
+
"grad_norm": 14.286447525024414,
|
| 285 |
+
"learning_rate": 9.62962962962963e-06,
|
| 286 |
+
"loss": 0.5559,
|
| 287 |
+
"step": 320
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 6.015,
|
| 291 |
+
"grad_norm": 12.041421890258789,
|
| 292 |
+
"learning_rate": 9.583333333333335e-06,
|
| 293 |
+
"loss": 0.5921,
|
| 294 |
+
"step": 330
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 6.019166666666667,
|
| 298 |
+
"grad_norm": 14.011693000793457,
|
| 299 |
+
"learning_rate": 9.537037037037037e-06,
|
| 300 |
+
"loss": 0.658,
|
| 301 |
+
"step": 340
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 6.020416666666667,
|
| 305 |
+
"eval_accuracy": 0.5789473684210527,
|
| 306 |
+
"eval_loss": 0.6696313619613647,
|
| 307 |
+
"eval_runtime": 3.7127,
|
| 308 |
+
"eval_samples_per_second": 5.118,
|
| 309 |
+
"eval_steps_per_second": 1.347,
|
| 310 |
+
"step": 343
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 7.002916666666667,
|
| 314 |
+
"grad_norm": 5.979766845703125,
|
| 315 |
+
"learning_rate": 9.490740740740741e-06,
|
| 316 |
+
"loss": 0.5175,
|
| 317 |
+
"step": 350
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 7.007083333333333,
|
| 321 |
+
"grad_norm": 10.06272029876709,
|
| 322 |
+
"learning_rate": 9.444444444444445e-06,
|
| 323 |
+
"loss": 0.5399,
|
| 324 |
+
"step": 360
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 7.01125,
|
| 328 |
+
"grad_norm": 6.3305983543396,
|
| 329 |
+
"learning_rate": 9.398148148148148e-06,
|
| 330 |
+
"loss": 0.4964,
|
| 331 |
+
"step": 370
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 7.015416666666667,
|
| 335 |
+
"grad_norm": 10.718533515930176,
|
| 336 |
+
"learning_rate": 9.351851851851854e-06,
|
| 337 |
+
"loss": 0.4526,
|
| 338 |
+
"step": 380
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 7.019583333333333,
|
| 342 |
+
"grad_norm": 10.034852981567383,
|
| 343 |
+
"learning_rate": 9.305555555555557e-06,
|
| 344 |
+
"loss": 0.5112,
|
| 345 |
+
"step": 390
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 7.020416666666667,
|
| 349 |
+
"eval_accuracy": 0.631578947368421,
|
| 350 |
+
"eval_loss": 0.566916823387146,
|
| 351 |
+
"eval_runtime": 3.6949,
|
| 352 |
+
"eval_samples_per_second": 5.142,
|
| 353 |
+
"eval_steps_per_second": 1.353,
|
| 354 |
+
"step": 392
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 8.003333333333334,
|
| 358 |
+
"grad_norm": 12.490917205810547,
|
| 359 |
+
"learning_rate": 9.25925925925926e-06,
|
| 360 |
+
"loss": 0.7497,
|
| 361 |
+
"step": 400
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 8.0075,
|
| 365 |
+
"grad_norm": 9.842666625976562,
|
| 366 |
+
"learning_rate": 9.212962962962963e-06,
|
| 367 |
+
"loss": 0.4709,
|
| 368 |
+
"step": 410
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 8.011666666666667,
|
| 372 |
+
"grad_norm": 18.18292236328125,
|
| 373 |
+
"learning_rate": 9.166666666666666e-06,
|
| 374 |
+
"loss": 0.4979,
|
| 375 |
+
"step": 420
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 8.015833333333333,
|
| 379 |
+
"grad_norm": 18.306493759155273,
|
| 380 |
+
"learning_rate": 9.120370370370372e-06,
|
| 381 |
+
"loss": 0.4218,
|
| 382 |
+
"step": 430
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 8.02,
|
| 386 |
+
"grad_norm": 1.3783447742462158,
|
| 387 |
+
"learning_rate": 9.074074074074075e-06,
|
| 388 |
+
"loss": 0.3719,
|
| 389 |
+
"step": 440
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 8.020416666666666,
|
| 393 |
+
"eval_accuracy": 0.631578947368421,
|
| 394 |
+
"eval_loss": 0.8023158311843872,
|
| 395 |
+
"eval_runtime": 3.6756,
|
| 396 |
+
"eval_samples_per_second": 5.169,
|
| 397 |
+
"eval_steps_per_second": 1.36,
|
| 398 |
+
"step": 441
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 9.00375,
|
| 402 |
+
"grad_norm": 43.830078125,
|
| 403 |
+
"learning_rate": 9.027777777777779e-06,
|
| 404 |
+
"loss": 0.4791,
|
| 405 |
+
"step": 450
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 9.007916666666667,
|
| 409 |
+
"grad_norm": 13.223751068115234,
|
| 410 |
+
"learning_rate": 8.981481481481483e-06,
|
| 411 |
+
"loss": 0.623,
|
| 412 |
+
"step": 460
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 9.012083333333333,
|
| 416 |
+
"grad_norm": 48.20356369018555,
|
| 417 |
+
"learning_rate": 8.935185185185186e-06,
|
| 418 |
+
"loss": 0.3433,
|
| 419 |
+
"step": 470
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 9.01625,
|
| 423 |
+
"grad_norm": 12.939654350280762,
|
| 424 |
+
"learning_rate": 8.888888888888888e-06,
|
| 425 |
+
"loss": 0.4031,
|
| 426 |
+
"step": 480
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 9.020416666666666,
|
| 430 |
+
"grad_norm": 12.470285415649414,
|
| 431 |
+
"learning_rate": 8.842592592592594e-06,
|
| 432 |
+
"loss": 0.4573,
|
| 433 |
+
"step": 490
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 9.020416666666666,
|
| 437 |
+
"eval_accuracy": 0.631578947368421,
|
| 438 |
+
"eval_loss": 0.5990714430809021,
|
| 439 |
+
"eval_runtime": 3.6148,
|
| 440 |
+
"eval_samples_per_second": 5.256,
|
| 441 |
+
"eval_steps_per_second": 1.383,
|
| 442 |
+
"step": 490
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 10.004166666666666,
|
| 446 |
+
"grad_norm": 5.166864395141602,
|
| 447 |
+
"learning_rate": 8.796296296296297e-06,
|
| 448 |
+
"loss": 0.3944,
|
| 449 |
+
"step": 500
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 10.008333333333333,
|
| 453 |
+
"grad_norm": 36.93327331542969,
|
| 454 |
+
"learning_rate": 8.750000000000001e-06,
|
| 455 |
+
"loss": 0.2934,
|
| 456 |
+
"step": 510
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 10.0125,
|
| 460 |
+
"grad_norm": 2.9084632396698,
|
| 461 |
+
"learning_rate": 8.703703703703705e-06,
|
| 462 |
+
"loss": 0.3113,
|
| 463 |
+
"step": 520
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 10.016666666666667,
|
| 467 |
+
"grad_norm": 25.439733505249023,
|
| 468 |
+
"learning_rate": 8.657407407407408e-06,
|
| 469 |
+
"loss": 0.2915,
|
| 470 |
+
"step": 530
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 10.020416666666666,
|
| 474 |
+
"eval_accuracy": 0.6842105263157895,
|
| 475 |
+
"eval_loss": 0.7898865938186646,
|
| 476 |
+
"eval_runtime": 3.6535,
|
| 477 |
+
"eval_samples_per_second": 5.2,
|
| 478 |
+
"eval_steps_per_second": 1.369,
|
| 479 |
+
"step": 539
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 11.000416666666666,
|
| 483 |
+
"grad_norm": 46.3503532409668,
|
| 484 |
+
"learning_rate": 8.611111111111112e-06,
|
| 485 |
+
"loss": 0.5269,
|
| 486 |
+
"step": 540
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 11.004583333333333,
|
| 490 |
+
"grad_norm": 0.9184814095497131,
|
| 491 |
+
"learning_rate": 8.564814814814816e-06,
|
| 492 |
+
"loss": 0.2812,
|
| 493 |
+
"step": 550
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 11.00875,
|
| 497 |
+
"grad_norm": 0.12803786993026733,
|
| 498 |
+
"learning_rate": 8.518518518518519e-06,
|
| 499 |
+
"loss": 0.6738,
|
| 500 |
+
"step": 560
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 11.012916666666667,
|
| 504 |
+
"grad_norm": 6.100522518157959,
|
| 505 |
+
"learning_rate": 8.472222222222223e-06,
|
| 506 |
+
"loss": 0.4429,
|
| 507 |
+
"step": 570
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 11.017083333333334,
|
| 511 |
+
"grad_norm": 57.438533782958984,
|
| 512 |
+
"learning_rate": 8.425925925925926e-06,
|
| 513 |
+
"loss": 0.3723,
|
| 514 |
+
"step": 580
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 11.020416666666666,
|
| 518 |
+
"eval_accuracy": 0.5263157894736842,
|
| 519 |
+
"eval_loss": 0.9848392605781555,
|
| 520 |
+
"eval_runtime": 3.676,
|
| 521 |
+
"eval_samples_per_second": 5.169,
|
| 522 |
+
"eval_steps_per_second": 1.36,
|
| 523 |
+
"step": 588
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 12.000833333333333,
|
| 527 |
+
"grad_norm": 39.162044525146484,
|
| 528 |
+
"learning_rate": 8.37962962962963e-06,
|
| 529 |
+
"loss": 0.3971,
|
| 530 |
+
"step": 590
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 12.005,
|
| 534 |
+
"grad_norm": 2.6625895500183105,
|
| 535 |
+
"learning_rate": 8.333333333333334e-06,
|
| 536 |
+
"loss": 0.423,
|
| 537 |
+
"step": 600
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 12.009166666666667,
|
| 541 |
+
"grad_norm": 64.04193878173828,
|
| 542 |
+
"learning_rate": 8.287037037037037e-06,
|
| 543 |
+
"loss": 0.5011,
|
| 544 |
+
"step": 610
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 12.013333333333334,
|
| 548 |
+
"grad_norm": 3.9968278408050537,
|
| 549 |
+
"learning_rate": 8.240740740740741e-06,
|
| 550 |
+
"loss": 0.3497,
|
| 551 |
+
"step": 620
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 12.0175,
|
| 555 |
+
"grad_norm": 0.1798192709684372,
|
| 556 |
+
"learning_rate": 8.194444444444445e-06,
|
| 557 |
+
"loss": 0.1372,
|
| 558 |
+
"step": 630
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 12.020416666666666,
|
| 562 |
+
"eval_accuracy": 0.631578947368421,
|
| 563 |
+
"eval_loss": 0.9509623646736145,
|
| 564 |
+
"eval_runtime": 3.68,
|
| 565 |
+
"eval_samples_per_second": 5.163,
|
| 566 |
+
"eval_steps_per_second": 1.359,
|
| 567 |
+
"step": 637
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 13.00125,
|
| 571 |
+
"grad_norm": 0.18710507452487946,
|
| 572 |
+
"learning_rate": 8.148148148148148e-06,
|
| 573 |
+
"loss": 0.3299,
|
| 574 |
+
"step": 640
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 13.005416666666667,
|
| 578 |
+
"grad_norm": 0.11506196111440659,
|
| 579 |
+
"learning_rate": 8.101851851851854e-06,
|
| 580 |
+
"loss": 0.3919,
|
| 581 |
+
"step": 650
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 13.009583333333333,
|
| 585 |
+
"grad_norm": 27.668811798095703,
|
| 586 |
+
"learning_rate": 8.055555555555557e-06,
|
| 587 |
+
"loss": 0.1775,
|
| 588 |
+
"step": 660
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 13.01375,
|
| 592 |
+
"grad_norm": 75.68840026855469,
|
| 593 |
+
"learning_rate": 8.00925925925926e-06,
|
| 594 |
+
"loss": 0.2991,
|
| 595 |
+
"step": 670
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 13.017916666666666,
|
| 599 |
+
"grad_norm": 0.38133883476257324,
|
| 600 |
+
"learning_rate": 7.962962962962963e-06,
|
| 601 |
+
"loss": 0.0066,
|
| 602 |
+
"step": 680
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 13.020416666666666,
|
| 606 |
+
"eval_accuracy": 0.5789473684210527,
|
| 607 |
+
"eval_loss": 1.1871516704559326,
|
| 608 |
+
"eval_runtime": 3.6868,
|
| 609 |
+
"eval_samples_per_second": 5.154,
|
| 610 |
+
"eval_steps_per_second": 1.356,
|
| 611 |
+
"step": 686
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 14.001666666666667,
|
| 615 |
+
"grad_norm": 3.5586605072021484,
|
| 616 |
+
"learning_rate": 7.916666666666667e-06,
|
| 617 |
+
"loss": 0.3722,
|
| 618 |
+
"step": 690
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 14.005833333333333,
|
| 622 |
+
"grad_norm": 3.2339565753936768,
|
| 623 |
+
"learning_rate": 7.870370370370372e-06,
|
| 624 |
+
"loss": 0.1984,
|
| 625 |
+
"step": 700
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 14.01,
|
| 629 |
+
"grad_norm": 0.06678327918052673,
|
| 630 |
+
"learning_rate": 7.824074074074076e-06,
|
| 631 |
+
"loss": 0.0981,
|
| 632 |
+
"step": 710
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 14.014166666666666,
|
| 636 |
+
"grad_norm": 0.2941597104072571,
|
| 637 |
+
"learning_rate": 7.77777777777778e-06,
|
| 638 |
+
"loss": 0.5156,
|
| 639 |
+
"step": 720
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 14.018333333333333,
|
| 643 |
+
"grad_norm": 0.2059919536113739,
|
| 644 |
+
"learning_rate": 7.731481481481483e-06,
|
| 645 |
+
"loss": 0.2218,
|
| 646 |
+
"step": 730
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 14.020416666666666,
|
| 650 |
+
"eval_accuracy": 0.7368421052631579,
|
| 651 |
+
"eval_loss": 1.1691995859146118,
|
| 652 |
+
"eval_runtime": 3.7567,
|
| 653 |
+
"eval_samples_per_second": 5.058,
|
| 654 |
+
"eval_steps_per_second": 1.331,
|
| 655 |
+
"step": 735
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 15.002083333333333,
|
| 659 |
+
"grad_norm": 0.07924695312976837,
|
| 660 |
+
"learning_rate": 7.685185185185185e-06,
|
| 661 |
+
"loss": 0.2093,
|
| 662 |
+
"step": 740
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 15.00625,
|
| 666 |
+
"grad_norm": 24.16701316833496,
|
| 667 |
+
"learning_rate": 7.638888888888888e-06,
|
| 668 |
+
"loss": 0.4067,
|
| 669 |
+
"step": 750
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 15.010416666666666,
|
| 673 |
+
"grad_norm": 0.2460525929927826,
|
| 674 |
+
"learning_rate": 7.592592592592594e-06,
|
| 675 |
+
"loss": 0.3083,
|
| 676 |
+
"step": 760
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 15.014583333333333,
|
| 680 |
+
"grad_norm": 0.8729657530784607,
|
| 681 |
+
"learning_rate": 7.546296296296297e-06,
|
| 682 |
+
"loss": 0.1764,
|
| 683 |
+
"step": 770
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 15.01875,
|
| 687 |
+
"grad_norm": 107.28450775146484,
|
| 688 |
+
"learning_rate": 7.500000000000001e-06,
|
| 689 |
+
"loss": 0.7666,
|
| 690 |
+
"step": 780
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 15.020416666666666,
|
| 694 |
+
"eval_accuracy": 0.6842105263157895,
|
| 695 |
+
"eval_loss": 1.112523078918457,
|
| 696 |
+
"eval_runtime": 3.7224,
|
| 697 |
+
"eval_samples_per_second": 5.104,
|
| 698 |
+
"eval_steps_per_second": 1.343,
|
| 699 |
+
"step": 784
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 16.0025,
|
| 703 |
+
"grad_norm": 0.05592398717999458,
|
| 704 |
+
"learning_rate": 7.453703703703704e-06,
|
| 705 |
+
"loss": 0.1058,
|
| 706 |
+
"step": 790
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 16.006666666666668,
|
| 710 |
+
"grad_norm": 0.2081090658903122,
|
| 711 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 712 |
+
"loss": 0.1724,
|
| 713 |
+
"step": 800
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 16.010833333333334,
|
| 717 |
+
"grad_norm": 0.08229461312294006,
|
| 718 |
+
"learning_rate": 7.361111111111112e-06,
|
| 719 |
+
"loss": 0.0088,
|
| 720 |
+
"step": 810
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 16.015,
|
| 724 |
+
"grad_norm": 2.331199884414673,
|
| 725 |
+
"learning_rate": 7.314814814814816e-06,
|
| 726 |
+
"loss": 0.0636,
|
| 727 |
+
"step": 820
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 16.019166666666667,
|
| 731 |
+
"grad_norm": 15.751022338867188,
|
| 732 |
+
"learning_rate": 7.268518518518519e-06,
|
| 733 |
+
"loss": 0.2171,
|
| 734 |
+
"step": 830
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 16.020416666666666,
|
| 738 |
+
"eval_accuracy": 0.7368421052631579,
|
| 739 |
+
"eval_loss": 1.2614731788635254,
|
| 740 |
+
"eval_runtime": 3.6912,
|
| 741 |
+
"eval_samples_per_second": 5.147,
|
| 742 |
+
"eval_steps_per_second": 1.355,
|
| 743 |
+
"step": 833
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 17.002916666666668,
|
| 747 |
+
"grad_norm": 0.13139835000038147,
|
| 748 |
+
"learning_rate": 7.222222222222223e-06,
|
| 749 |
+
"loss": 0.0529,
|
| 750 |
+
"step": 840
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 17.007083333333334,
|
| 754 |
+
"grad_norm": 146.87222290039062,
|
| 755 |
+
"learning_rate": 7.1759259259259266e-06,
|
| 756 |
+
"loss": 0.2678,
|
| 757 |
+
"step": 850
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 17.01125,
|
| 761 |
+
"grad_norm": 350.3876953125,
|
| 762 |
+
"learning_rate": 7.129629629629629e-06,
|
| 763 |
+
"loss": 0.217,
|
| 764 |
+
"step": 860
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 17.015416666666667,
|
| 768 |
+
"grad_norm": 0.5063939094543457,
|
| 769 |
+
"learning_rate": 7.083333333333335e-06,
|
| 770 |
+
"loss": 0.0078,
|
| 771 |
+
"step": 870
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 17.019583333333333,
|
| 775 |
+
"grad_norm": 247.79161071777344,
|
| 776 |
+
"learning_rate": 7.0370370370370375e-06,
|
| 777 |
+
"loss": 0.1152,
|
| 778 |
+
"step": 880
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 17.020416666666666,
|
| 782 |
+
"eval_accuracy": 0.7368421052631579,
|
| 783 |
+
"eval_loss": 1.3682870864868164,
|
| 784 |
+
"eval_runtime": 3.6816,
|
| 785 |
+
"eval_samples_per_second": 5.161,
|
| 786 |
+
"eval_steps_per_second": 1.358,
|
| 787 |
+
"step": 882
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 18.003333333333334,
|
| 791 |
+
"grad_norm": 0.03123622015118599,
|
| 792 |
+
"learning_rate": 6.990740740740741e-06,
|
| 793 |
+
"loss": 0.3538,
|
| 794 |
+
"step": 890
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 18.0075,
|
| 798 |
+
"grad_norm": 0.6689489483833313,
|
| 799 |
+
"learning_rate": 6.944444444444445e-06,
|
| 800 |
+
"loss": 0.1337,
|
| 801 |
+
"step": 900
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 18.011666666666667,
|
| 805 |
+
"grad_norm": 120.35101318359375,
|
| 806 |
+
"learning_rate": 6.898148148148148e-06,
|
| 807 |
+
"loss": 0.2021,
|
| 808 |
+
"step": 910
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 18.015833333333333,
|
| 812 |
+
"grad_norm": 6.242348670959473,
|
| 813 |
+
"learning_rate": 6.851851851851853e-06,
|
| 814 |
+
"loss": 0.1344,
|
| 815 |
+
"step": 920
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 18.02,
|
| 819 |
+
"grad_norm": 0.015488705597817898,
|
| 820 |
+
"learning_rate": 6.8055555555555566e-06,
|
| 821 |
+
"loss": 0.1941,
|
| 822 |
+
"step": 930
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 18.020416666666666,
|
| 826 |
+
"eval_accuracy": 0.7368421052631579,
|
| 827 |
+
"eval_loss": 1.391472578048706,
|
| 828 |
+
"eval_runtime": 3.6708,
|
| 829 |
+
"eval_samples_per_second": 5.176,
|
| 830 |
+
"eval_steps_per_second": 1.362,
|
| 831 |
+
"step": 931
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 19.00375,
|
| 835 |
+
"grad_norm": 107.10614013671875,
|
| 836 |
+
"learning_rate": 6.75925925925926e-06,
|
| 837 |
+
"loss": 0.0318,
|
| 838 |
+
"step": 940
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 19.007916666666667,
|
| 842 |
+
"grad_norm": 0.07496990263462067,
|
| 843 |
+
"learning_rate": 6.712962962962963e-06,
|
| 844 |
+
"loss": 0.0013,
|
| 845 |
+
"step": 950
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 19.012083333333333,
|
| 849 |
+
"grad_norm": 0.028873112052679062,
|
| 850 |
+
"learning_rate": 6.666666666666667e-06,
|
| 851 |
+
"loss": 0.2196,
|
| 852 |
+
"step": 960
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 19.01625,
|
| 856 |
+
"grad_norm": 0.06660551577806473,
|
| 857 |
+
"learning_rate": 6.620370370370371e-06,
|
| 858 |
+
"loss": 0.0014,
|
| 859 |
+
"step": 970
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 19.020416666666666,
|
| 863 |
+
"grad_norm": 89.06936645507812,
|
| 864 |
+
"learning_rate": 6.574074074074075e-06,
|
| 865 |
+
"loss": 0.0233,
|
| 866 |
+
"step": 980
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 19.020416666666666,
|
| 870 |
+
"eval_accuracy": 0.5789473684210527,
|
| 871 |
+
"eval_loss": 1.5971713066101074,
|
| 872 |
+
"eval_runtime": 3.6747,
|
| 873 |
+
"eval_samples_per_second": 5.171,
|
| 874 |
+
"eval_steps_per_second": 1.361,
|
| 875 |
+
"step": 980
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 20.004166666666666,
|
| 879 |
+
"grad_norm": 12.551525115966797,
|
| 880 |
+
"learning_rate": 6.5277777777777784e-06,
|
| 881 |
+
"loss": 0.1686,
|
| 882 |
+
"step": 990
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 20.008333333333333,
|
| 886 |
+
"grad_norm": 1.433871865272522,
|
| 887 |
+
"learning_rate": 6.481481481481482e-06,
|
| 888 |
+
"loss": 0.1168,
|
| 889 |
+
"step": 1000
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 20.0125,
|
| 893 |
+
"grad_norm": 0.3970452547073364,
|
| 894 |
+
"learning_rate": 6.435185185185186e-06,
|
| 895 |
+
"loss": 0.2299,
|
| 896 |
+
"step": 1010
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 20.016666666666666,
|
| 900 |
+
"grad_norm": 198.20985412597656,
|
| 901 |
+
"learning_rate": 6.3888888888888885e-06,
|
| 902 |
+
"loss": 0.2367,
|
| 903 |
+
"step": 1020
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 20.020416666666666,
|
| 907 |
+
"eval_accuracy": 0.6842105263157895,
|
| 908 |
+
"eval_loss": 1.7444992065429688,
|
| 909 |
+
"eval_runtime": 3.6107,
|
| 910 |
+
"eval_samples_per_second": 5.262,
|
| 911 |
+
"eval_steps_per_second": 1.385,
|
| 912 |
+
"step": 1029
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 21.000416666666666,
|
| 916 |
+
"grad_norm": 11.477893829345703,
|
| 917 |
+
"learning_rate": 6.342592592592594e-06,
|
| 918 |
+
"loss": 0.0874,
|
| 919 |
+
"step": 1030
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 21.004583333333333,
|
| 923 |
+
"grad_norm": 0.04803403466939926,
|
| 924 |
+
"learning_rate": 6.296296296296297e-06,
|
| 925 |
+
"loss": 0.0026,
|
| 926 |
+
"step": 1040
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 21.00875,
|
| 930 |
+
"grad_norm": 235.3900604248047,
|
| 931 |
+
"learning_rate": 6.25e-06,
|
| 932 |
+
"loss": 0.0272,
|
| 933 |
+
"step": 1050
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 21.012916666666666,
|
| 937 |
+
"grad_norm": 0.027185214683413506,
|
| 938 |
+
"learning_rate": 6.203703703703704e-06,
|
| 939 |
+
"loss": 0.1929,
|
| 940 |
+
"step": 1060
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 21.017083333333332,
|
| 944 |
+
"grad_norm": 1.3234747648239136,
|
| 945 |
+
"learning_rate": 6.157407407407408e-06,
|
| 946 |
+
"loss": 0.1575,
|
| 947 |
+
"step": 1070
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 21.020416666666666,
|
| 951 |
+
"eval_accuracy": 0.631578947368421,
|
| 952 |
+
"eval_loss": 2.403568744659424,
|
| 953 |
+
"eval_runtime": 3.6124,
|
| 954 |
+
"eval_samples_per_second": 5.26,
|
| 955 |
+
"eval_steps_per_second": 1.384,
|
| 956 |
+
"step": 1078
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 22.000833333333333,
|
| 960 |
+
"grad_norm": 0.030825955793261528,
|
| 961 |
+
"learning_rate": 6.111111111111112e-06,
|
| 962 |
+
"loss": 0.0207,
|
| 963 |
+
"step": 1080
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 22.005,
|
| 967 |
+
"grad_norm": 0.044659119099378586,
|
| 968 |
+
"learning_rate": 6.064814814814816e-06,
|
| 969 |
+
"loss": 0.1655,
|
| 970 |
+
"step": 1090
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 22.009166666666665,
|
| 974 |
+
"grad_norm": 0.07954849302768707,
|
| 975 |
+
"learning_rate": 6.018518518518519e-06,
|
| 976 |
+
"loss": 0.0036,
|
| 977 |
+
"step": 1100
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 22.013333333333332,
|
| 981 |
+
"grad_norm": 0.3902687132358551,
|
| 982 |
+
"learning_rate": 5.972222222222222e-06,
|
| 983 |
+
"loss": 0.0029,
|
| 984 |
+
"step": 1110
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 22.0175,
|
| 988 |
+
"grad_norm": 0.1071692705154419,
|
| 989 |
+
"learning_rate": 5.925925925925926e-06,
|
| 990 |
+
"loss": 0.0008,
|
| 991 |
+
"step": 1120
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 22.020416666666666,
|
| 995 |
+
"eval_accuracy": 0.631578947368421,
|
| 996 |
+
"eval_loss": 2.263265609741211,
|
| 997 |
+
"eval_runtime": 3.6163,
|
| 998 |
+
"eval_samples_per_second": 5.254,
|
| 999 |
+
"eval_steps_per_second": 1.383,
|
| 1000 |
+
"step": 1127
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 23.00125,
|
| 1004 |
+
"grad_norm": 0.015230727382004261,
|
| 1005 |
+
"learning_rate": 5.8796296296296295e-06,
|
| 1006 |
+
"loss": 0.0928,
|
| 1007 |
+
"step": 1130
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 23.005416666666665,
|
| 1011 |
+
"grad_norm": 42.0711669921875,
|
| 1012 |
+
"learning_rate": 5.833333333333334e-06,
|
| 1013 |
+
"loss": 0.3605,
|
| 1014 |
+
"step": 1140
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 23.00958333333333,
|
| 1018 |
+
"grad_norm": 0.05277286097407341,
|
| 1019 |
+
"learning_rate": 5.787037037037038e-06,
|
| 1020 |
+
"loss": 0.0005,
|
| 1021 |
+
"step": 1150
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 23.01375,
|
| 1025 |
+
"grad_norm": 0.020604358986020088,
|
| 1026 |
+
"learning_rate": 5.740740740740741e-06,
|
| 1027 |
+
"loss": 0.0006,
|
| 1028 |
+
"step": 1160
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 23.017916666666668,
|
| 1032 |
+
"grad_norm": 0.007125093135982752,
|
| 1033 |
+
"learning_rate": 5.694444444444445e-06,
|
| 1034 |
+
"loss": 0.019,
|
| 1035 |
+
"step": 1170
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 23.020416666666666,
|
| 1039 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1040 |
+
"eval_loss": 2.2078418731689453,
|
| 1041 |
+
"eval_runtime": 3.6116,
|
| 1042 |
+
"eval_samples_per_second": 5.261,
|
| 1043 |
+
"eval_steps_per_second": 1.384,
|
| 1044 |
+
"step": 1176
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 24.001666666666665,
|
| 1048 |
+
"grad_norm": 0.04864019155502319,
|
| 1049 |
+
"learning_rate": 5.6481481481481485e-06,
|
| 1050 |
+
"loss": 0.0005,
|
| 1051 |
+
"step": 1180
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 24.00583333333333,
|
| 1055 |
+
"grad_norm": 0.06986731290817261,
|
| 1056 |
+
"learning_rate": 5.601851851851853e-06,
|
| 1057 |
+
"loss": 0.1865,
|
| 1058 |
+
"step": 1190
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 24.01,
|
| 1062 |
+
"grad_norm": 0.045288749039173126,
|
| 1063 |
+
"learning_rate": 5.555555555555557e-06,
|
| 1064 |
+
"loss": 0.0021,
|
| 1065 |
+
"step": 1200
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 24.014166666666668,
|
| 1069 |
+
"grad_norm": 0.12277600169181824,
|
| 1070 |
+
"learning_rate": 5.5092592592592595e-06,
|
| 1071 |
+
"loss": 0.0004,
|
| 1072 |
+
"step": 1210
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 24.018333333333334,
|
| 1076 |
+
"grad_norm": 0.006932346150279045,
|
| 1077 |
+
"learning_rate": 5.462962962962963e-06,
|
| 1078 |
+
"loss": 0.0006,
|
| 1079 |
+
"step": 1220
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 24.020416666666666,
|
| 1083 |
+
"eval_accuracy": 0.631578947368421,
|
| 1084 |
+
"eval_loss": 2.1280739307403564,
|
| 1085 |
+
"eval_runtime": 3.6117,
|
| 1086 |
+
"eval_samples_per_second": 5.261,
|
| 1087 |
+
"eval_steps_per_second": 1.384,
|
| 1088 |
+
"step": 1225
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 25.002083333333335,
|
| 1092 |
+
"grad_norm": 0.016528107225894928,
|
| 1093 |
+
"learning_rate": 5.416666666666667e-06,
|
| 1094 |
+
"loss": 0.0003,
|
| 1095 |
+
"step": 1230
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 25.00625,
|
| 1099 |
+
"grad_norm": 0.013506701216101646,
|
| 1100 |
+
"learning_rate": 5.370370370370371e-06,
|
| 1101 |
+
"loss": 0.0024,
|
| 1102 |
+
"step": 1240
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 25.010416666666668,
|
| 1106 |
+
"grad_norm": 0.5438185334205627,
|
| 1107 |
+
"learning_rate": 5.324074074074075e-06,
|
| 1108 |
+
"loss": 0.0005,
|
| 1109 |
+
"step": 1250
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 25.014583333333334,
|
| 1113 |
+
"grad_norm": 0.03781206160783768,
|
| 1114 |
+
"learning_rate": 5.2777777777777785e-06,
|
| 1115 |
+
"loss": 0.0003,
|
| 1116 |
+
"step": 1260
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 25.01875,
|
| 1120 |
+
"grad_norm": 0.0206539798527956,
|
| 1121 |
+
"learning_rate": 5.231481481481482e-06,
|
| 1122 |
+
"loss": 0.0004,
|
| 1123 |
+
"step": 1270
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 25.020416666666666,
|
| 1127 |
+
"eval_accuracy": 0.7368421052631579,
|
| 1128 |
+
"eval_loss": 2.010007381439209,
|
| 1129 |
+
"eval_runtime": 3.6278,
|
| 1130 |
+
"eval_samples_per_second": 5.237,
|
| 1131 |
+
"eval_steps_per_second": 1.378,
|
| 1132 |
+
"step": 1274
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 26.0025,
|
| 1136 |
+
"grad_norm": 0.023114336654543877,
|
| 1137 |
+
"learning_rate": 5.185185185185185e-06,
|
| 1138 |
+
"loss": 0.0002,
|
| 1139 |
+
"step": 1280
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 26.006666666666668,
|
| 1143 |
+
"grad_norm": 0.013670257292687893,
|
| 1144 |
+
"learning_rate": 5.138888888888889e-06,
|
| 1145 |
+
"loss": 0.002,
|
| 1146 |
+
"step": 1290
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 26.010833333333334,
|
| 1150 |
+
"grad_norm": 0.0056487866677343845,
|
| 1151 |
+
"learning_rate": 5.092592592592593e-06,
|
| 1152 |
+
"loss": 0.0004,
|
| 1153 |
+
"step": 1300
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 26.015,
|
| 1157 |
+
"grad_norm": 0.004165478516370058,
|
| 1158 |
+
"learning_rate": 5.046296296296297e-06,
|
| 1159 |
+
"loss": 0.0003,
|
| 1160 |
+
"step": 1310
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 26.019166666666667,
|
| 1164 |
+
"grad_norm": 0.004816366359591484,
|
| 1165 |
+
"learning_rate": 5e-06,
|
| 1166 |
+
"loss": 0.0002,
|
| 1167 |
+
"step": 1320
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 26.020416666666666,
|
| 1171 |
+
"eval_accuracy": 0.631578947368421,
|
| 1172 |
+
"eval_loss": 2.3861982822418213,
|
| 1173 |
+
"eval_runtime": 3.6074,
|
| 1174 |
+
"eval_samples_per_second": 5.267,
|
| 1175 |
+
"eval_steps_per_second": 1.386,
|
| 1176 |
+
"step": 1323
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 27.002916666666668,
|
| 1180 |
+
"grad_norm": 0.006848832592368126,
|
| 1181 |
+
"learning_rate": 4.953703703703704e-06,
|
| 1182 |
+
"loss": 0.0062,
|
| 1183 |
+
"step": 1330
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 27.007083333333334,
|
| 1187 |
+
"grad_norm": 530.533935546875,
|
| 1188 |
+
"learning_rate": 4.907407407407408e-06,
|
| 1189 |
+
"loss": 0.0309,
|
| 1190 |
+
"step": 1340
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 27.01125,
|
| 1194 |
+
"grad_norm": 0.007108121179044247,
|
| 1195 |
+
"learning_rate": 4.861111111111111e-06,
|
| 1196 |
+
"loss": 0.0002,
|
| 1197 |
+
"step": 1350
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 27.015416666666667,
|
| 1201 |
+
"grad_norm": 0.0033726587425917387,
|
| 1202 |
+
"learning_rate": 4.814814814814815e-06,
|
| 1203 |
+
"loss": 0.116,
|
| 1204 |
+
"step": 1360
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 27.019583333333333,
|
| 1208 |
+
"grad_norm": 0.020084122195839882,
|
| 1209 |
+
"learning_rate": 4.768518518518519e-06,
|
| 1210 |
+
"loss": 0.0003,
|
| 1211 |
+
"step": 1370
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 27.020416666666666,
|
| 1215 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1216 |
+
"eval_loss": 2.1578803062438965,
|
| 1217 |
+
"eval_runtime": 3.6309,
|
| 1218 |
+
"eval_samples_per_second": 5.233,
|
| 1219 |
+
"eval_steps_per_second": 1.377,
|
| 1220 |
+
"step": 1372
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 28.003333333333334,
|
| 1224 |
+
"grad_norm": 0.05458810552954674,
|
| 1225 |
+
"learning_rate": 4.722222222222222e-06,
|
| 1226 |
+
"loss": 0.0002,
|
| 1227 |
+
"step": 1380
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 28.0075,
|
| 1231 |
+
"grad_norm": 0.005413507577031851,
|
| 1232 |
+
"learning_rate": 4.675925925925927e-06,
|
| 1233 |
+
"loss": 0.0002,
|
| 1234 |
+
"step": 1390
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 28.011666666666667,
|
| 1238 |
+
"grad_norm": 0.006965883541852236,
|
| 1239 |
+
"learning_rate": 4.62962962962963e-06,
|
| 1240 |
+
"loss": 0.0002,
|
| 1241 |
+
"step": 1400
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 28.015833333333333,
|
| 1245 |
+
"grad_norm": 0.06325055658817291,
|
| 1246 |
+
"learning_rate": 4.583333333333333e-06,
|
| 1247 |
+
"loss": 0.0002,
|
| 1248 |
+
"step": 1410
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 28.02,
|
| 1252 |
+
"grad_norm": 0.031983036547899246,
|
| 1253 |
+
"learning_rate": 4.537037037037038e-06,
|
| 1254 |
+
"loss": 0.0372,
|
| 1255 |
+
"step": 1420
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 28.020416666666666,
|
| 1259 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1260 |
+
"eval_loss": 2.122598648071289,
|
| 1261 |
+
"eval_runtime": 3.5383,
|
| 1262 |
+
"eval_samples_per_second": 5.37,
|
| 1263 |
+
"eval_steps_per_second": 1.413,
|
| 1264 |
+
"step": 1421
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 29.00375,
|
| 1268 |
+
"grad_norm": 0.07857130467891693,
|
| 1269 |
+
"learning_rate": 4.490740740740741e-06,
|
| 1270 |
+
"loss": 0.0006,
|
| 1271 |
+
"step": 1430
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 29.007916666666667,
|
| 1275 |
+
"grad_norm": 0.016926715150475502,
|
| 1276 |
+
"learning_rate": 4.444444444444444e-06,
|
| 1277 |
+
"loss": 0.001,
|
| 1278 |
+
"step": 1440
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 29.012083333333333,
|
| 1282 |
+
"grad_norm": 0.009676299057900906,
|
| 1283 |
+
"learning_rate": 4.398148148148149e-06,
|
| 1284 |
+
"loss": 0.0705,
|
| 1285 |
+
"step": 1450
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 29.01625,
|
| 1289 |
+
"grad_norm": 0.006083638872951269,
|
| 1290 |
+
"learning_rate": 4.351851851851852e-06,
|
| 1291 |
+
"loss": 0.0002,
|
| 1292 |
+
"step": 1460
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 29.020416666666666,
|
| 1296 |
+
"grad_norm": 0.006985980086028576,
|
| 1297 |
+
"learning_rate": 4.305555555555556e-06,
|
| 1298 |
+
"loss": 0.1547,
|
| 1299 |
+
"step": 1470
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 29.020416666666666,
|
| 1303 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1304 |
+
"eval_loss": 2.4948902130126953,
|
| 1305 |
+
"eval_runtime": 3.5201,
|
| 1306 |
+
"eval_samples_per_second": 5.398,
|
| 1307 |
+
"eval_steps_per_second": 1.42,
|
| 1308 |
+
"step": 1470
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 30.004166666666666,
|
| 1312 |
+
"grad_norm": 0.009625507518649101,
|
| 1313 |
+
"learning_rate": 4.2592592592592596e-06,
|
| 1314 |
+
"loss": 0.0104,
|
| 1315 |
+
"step": 1480
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 30.008333333333333,
|
| 1319 |
+
"grad_norm": 0.006220038048923016,
|
| 1320 |
+
"learning_rate": 4.212962962962963e-06,
|
| 1321 |
+
"loss": 0.0002,
|
| 1322 |
+
"step": 1490
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 30.0125,
|
| 1326 |
+
"grad_norm": 0.0076056793332099915,
|
| 1327 |
+
"learning_rate": 4.166666666666667e-06,
|
| 1328 |
+
"loss": 0.0006,
|
| 1329 |
+
"step": 1500
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 30.016666666666666,
|
| 1333 |
+
"grad_norm": 16.74993133544922,
|
| 1334 |
+
"learning_rate": 4.1203703703703705e-06,
|
| 1335 |
+
"loss": 0.0015,
|
| 1336 |
+
"step": 1510
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 30.020416666666666,
|
| 1340 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1341 |
+
"eval_loss": 1.881516933441162,
|
| 1342 |
+
"eval_runtime": 3.5962,
|
| 1343 |
+
"eval_samples_per_second": 5.283,
|
| 1344 |
+
"eval_steps_per_second": 1.39,
|
| 1345 |
+
"step": 1519
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 31.000416666666666,
|
| 1349 |
+
"grad_norm": 0.010737285017967224,
|
| 1350 |
+
"learning_rate": 4.074074074074074e-06,
|
| 1351 |
+
"loss": 0.0596,
|
| 1352 |
+
"step": 1520
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 31.004583333333333,
|
| 1356 |
+
"grad_norm": 0.004095606505870819,
|
| 1357 |
+
"learning_rate": 4.027777777777779e-06,
|
| 1358 |
+
"loss": 0.0001,
|
| 1359 |
+
"step": 1530
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 31.00875,
|
| 1363 |
+
"grad_norm": 0.38500717282295227,
|
| 1364 |
+
"learning_rate": 3.9814814814814814e-06,
|
| 1365 |
+
"loss": 0.0003,
|
| 1366 |
+
"step": 1540
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 31.012916666666666,
|
| 1370 |
+
"grad_norm": 35.35988235473633,
|
| 1371 |
+
"learning_rate": 3.935185185185186e-06,
|
| 1372 |
+
"loss": 0.0024,
|
| 1373 |
+
"step": 1550
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 31.017083333333332,
|
| 1377 |
+
"grad_norm": 0.006391994189471006,
|
| 1378 |
+
"learning_rate": 3.88888888888889e-06,
|
| 1379 |
+
"loss": 0.0005,
|
| 1380 |
+
"step": 1560
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 31.020416666666666,
|
| 1384 |
+
"eval_accuracy": 0.631578947368421,
|
| 1385 |
+
"eval_loss": 2.258211135864258,
|
| 1386 |
+
"eval_runtime": 3.5634,
|
| 1387 |
+
"eval_samples_per_second": 5.332,
|
| 1388 |
+
"eval_steps_per_second": 1.403,
|
| 1389 |
+
"step": 1568
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 32.00083333333333,
|
| 1393 |
+
"grad_norm": 0.0057472470216453075,
|
| 1394 |
+
"learning_rate": 3.842592592592592e-06,
|
| 1395 |
+
"loss": 0.0005,
|
| 1396 |
+
"step": 1570
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 32.005,
|
| 1400 |
+
"grad_norm": 0.011224722489714622,
|
| 1401 |
+
"learning_rate": 3.796296296296297e-06,
|
| 1402 |
+
"loss": 0.1187,
|
| 1403 |
+
"step": 1580
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 32.009166666666665,
|
| 1407 |
+
"grad_norm": 0.0647362545132637,
|
| 1408 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 1409 |
+
"loss": 0.0003,
|
| 1410 |
+
"step": 1590
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 32.013333333333335,
|
| 1414 |
+
"grad_norm": 0.15853719413280487,
|
| 1415 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 1416 |
+
"loss": 0.0002,
|
| 1417 |
+
"step": 1600
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 32.0175,
|
| 1421 |
+
"grad_norm": 0.00835301075130701,
|
| 1422 |
+
"learning_rate": 3.657407407407408e-06,
|
| 1423 |
+
"loss": 0.0002,
|
| 1424 |
+
"step": 1610
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 32.02041666666667,
|
| 1428 |
+
"eval_accuracy": 0.631578947368421,
|
| 1429 |
+
"eval_loss": 2.1774301528930664,
|
| 1430 |
+
"eval_runtime": 3.4326,
|
| 1431 |
+
"eval_samples_per_second": 5.535,
|
| 1432 |
+
"eval_steps_per_second": 1.457,
|
| 1433 |
+
"step": 1617
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 33.00125,
|
| 1437 |
+
"grad_norm": 0.04617740586400032,
|
| 1438 |
+
"learning_rate": 3.6111111111111115e-06,
|
| 1439 |
+
"loss": 0.0002,
|
| 1440 |
+
"step": 1620
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 33.00541666666667,
|
| 1444 |
+
"grad_norm": 0.009918724186718464,
|
| 1445 |
+
"learning_rate": 3.5648148148148147e-06,
|
| 1446 |
+
"loss": 0.2072,
|
| 1447 |
+
"step": 1630
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"epoch": 33.00958333333333,
|
| 1451 |
+
"grad_norm": 0.020935775712132454,
|
| 1452 |
+
"learning_rate": 3.5185185185185187e-06,
|
| 1453 |
+
"loss": 0.2234,
|
| 1454 |
+
"step": 1640
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 33.01375,
|
| 1458 |
+
"grad_norm": 0.12407401949167252,
|
| 1459 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 1460 |
+
"loss": 0.0177,
|
| 1461 |
+
"step": 1650
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 33.017916666666665,
|
| 1465 |
+
"grad_norm": 0.007428634911775589,
|
| 1466 |
+
"learning_rate": 3.4259259259259265e-06,
|
| 1467 |
+
"loss": 0.0001,
|
| 1468 |
+
"step": 1660
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 33.02041666666667,
|
| 1472 |
+
"eval_accuracy": 0.631578947368421,
|
| 1473 |
+
"eval_loss": 2.2221832275390625,
|
| 1474 |
+
"eval_runtime": 3.4343,
|
| 1475 |
+
"eval_samples_per_second": 5.532,
|
| 1476 |
+
"eval_steps_per_second": 1.456,
|
| 1477 |
+
"step": 1666
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 34.001666666666665,
|
| 1481 |
+
"grad_norm": 0.0065287998877465725,
|
| 1482 |
+
"learning_rate": 3.37962962962963e-06,
|
| 1483 |
+
"loss": 0.0002,
|
| 1484 |
+
"step": 1670
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 34.005833333333335,
|
| 1488 |
+
"grad_norm": 0.11280619353055954,
|
| 1489 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 1490 |
+
"loss": 0.0003,
|
| 1491 |
+
"step": 1680
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 34.01,
|
| 1495 |
+
"grad_norm": 0.019788485020399094,
|
| 1496 |
+
"learning_rate": 3.2870370370370374e-06,
|
| 1497 |
+
"loss": 0.0003,
|
| 1498 |
+
"step": 1690
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 34.01416666666667,
|
| 1502 |
+
"grad_norm": 0.003129773773252964,
|
| 1503 |
+
"learning_rate": 3.240740740740741e-06,
|
| 1504 |
+
"loss": 0.0006,
|
| 1505 |
+
"step": 1700
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 34.01833333333333,
|
| 1509 |
+
"grad_norm": 0.004923813510686159,
|
| 1510 |
+
"learning_rate": 3.1944444444444443e-06,
|
| 1511 |
+
"loss": 0.0001,
|
| 1512 |
+
"step": 1710
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 34.02041666666667,
|
| 1516 |
+
"eval_accuracy": 0.631578947368421,
|
| 1517 |
+
"eval_loss": 2.356006383895874,
|
| 1518 |
+
"eval_runtime": 3.4358,
|
| 1519 |
+
"eval_samples_per_second": 5.53,
|
| 1520 |
+
"eval_steps_per_second": 1.455,
|
| 1521 |
+
"step": 1715
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 35.00208333333333,
|
| 1525 |
+
"grad_norm": 0.04086696729063988,
|
| 1526 |
+
"learning_rate": 3.1481481481481483e-06,
|
| 1527 |
+
"loss": 0.0002,
|
| 1528 |
+
"step": 1720
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 35.00625,
|
| 1532 |
+
"grad_norm": 0.0031430383678525686,
|
| 1533 |
+
"learning_rate": 3.101851851851852e-06,
|
| 1534 |
+
"loss": 0.0002,
|
| 1535 |
+
"step": 1730
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 35.010416666666664,
|
| 1539 |
+
"grad_norm": 0.004690348170697689,
|
| 1540 |
+
"learning_rate": 3.055555555555556e-06,
|
| 1541 |
+
"loss": 0.0001,
|
| 1542 |
+
"step": 1740
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 35.014583333333334,
|
| 1546 |
+
"grad_norm": 0.005091555416584015,
|
| 1547 |
+
"learning_rate": 3.0092592592592597e-06,
|
| 1548 |
+
"loss": 0.0002,
|
| 1549 |
+
"step": 1750
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 35.01875,
|
| 1553 |
+
"grad_norm": 0.07692966610193253,
|
| 1554 |
+
"learning_rate": 2.962962962962963e-06,
|
| 1555 |
+
"loss": 0.0001,
|
| 1556 |
+
"step": 1760
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 35.02041666666667,
|
| 1560 |
+
"eval_accuracy": 0.631578947368421,
|
| 1561 |
+
"eval_loss": 2.390404462814331,
|
| 1562 |
+
"eval_runtime": 3.4314,
|
| 1563 |
+
"eval_samples_per_second": 5.537,
|
| 1564 |
+
"eval_steps_per_second": 1.457,
|
| 1565 |
+
"step": 1764
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 36.0025,
|
| 1569 |
+
"grad_norm": 0.003334124805405736,
|
| 1570 |
+
"learning_rate": 2.916666666666667e-06,
|
| 1571 |
+
"loss": 0.0002,
|
| 1572 |
+
"step": 1770
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 36.00666666666667,
|
| 1576 |
+
"grad_norm": 258.331298828125,
|
| 1577 |
+
"learning_rate": 2.8703703703703706e-06,
|
| 1578 |
+
"loss": 0.0424,
|
| 1579 |
+
"step": 1780
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 36.01083333333333,
|
| 1583 |
+
"grad_norm": 0.010029267519712448,
|
| 1584 |
+
"learning_rate": 2.8240740740740743e-06,
|
| 1585 |
+
"loss": 0.0001,
|
| 1586 |
+
"step": 1790
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 36.015,
|
| 1590 |
+
"grad_norm": 0.0017382479272782803,
|
| 1591 |
+
"learning_rate": 2.7777777777777783e-06,
|
| 1592 |
+
"loss": 0.0001,
|
| 1593 |
+
"step": 1800
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 36.01916666666666,
|
| 1597 |
+
"grad_norm": 0.0759102925658226,
|
| 1598 |
+
"learning_rate": 2.7314814814814816e-06,
|
| 1599 |
+
"loss": 0.0001,
|
| 1600 |
+
"step": 1810
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 36.02041666666667,
|
| 1604 |
+
"eval_accuracy": 0.7368421052631579,
|
| 1605 |
+
"eval_loss": 2.2239937782287598,
|
| 1606 |
+
"eval_runtime": 3.6381,
|
| 1607 |
+
"eval_samples_per_second": 5.223,
|
| 1608 |
+
"eval_steps_per_second": 1.374,
|
| 1609 |
+
"step": 1813
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 37.002916666666664,
|
| 1613 |
+
"grad_norm": 0.003970419522374868,
|
| 1614 |
+
"learning_rate": 2.6851851851851856e-06,
|
| 1615 |
+
"loss": 0.0001,
|
| 1616 |
+
"step": 1820
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 37.007083333333334,
|
| 1620 |
+
"grad_norm": 0.005589651875197887,
|
| 1621 |
+
"learning_rate": 2.6388888888888893e-06,
|
| 1622 |
+
"loss": 0.0001,
|
| 1623 |
+
"step": 1830
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 37.01125,
|
| 1627 |
+
"grad_norm": 0.00631442666053772,
|
| 1628 |
+
"learning_rate": 2.5925925925925925e-06,
|
| 1629 |
+
"loss": 0.0007,
|
| 1630 |
+
"step": 1840
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 37.01541666666667,
|
| 1634 |
+
"grad_norm": 0.0022680216934531927,
|
| 1635 |
+
"learning_rate": 2.5462962962962966e-06,
|
| 1636 |
+
"loss": 0.0001,
|
| 1637 |
+
"step": 1850
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 37.01958333333334,
|
| 1641 |
+
"grad_norm": 0.630527913570404,
|
| 1642 |
+
"learning_rate": 2.5e-06,
|
| 1643 |
+
"loss": 0.0002,
|
| 1644 |
+
"step": 1860
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 37.02041666666667,
|
| 1648 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1649 |
+
"eval_loss": 2.287989854812622,
|
| 1650 |
+
"eval_runtime": 3.4557,
|
| 1651 |
+
"eval_samples_per_second": 5.498,
|
| 1652 |
+
"eval_steps_per_second": 1.447,
|
| 1653 |
+
"step": 1862
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 38.00333333333333,
|
| 1657 |
+
"grad_norm": 0.018673082813620567,
|
| 1658 |
+
"learning_rate": 2.453703703703704e-06,
|
| 1659 |
+
"loss": 0.0002,
|
| 1660 |
+
"step": 1870
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 38.0075,
|
| 1664 |
+
"grad_norm": 0.2521021068096161,
|
| 1665 |
+
"learning_rate": 2.4074074074074075e-06,
|
| 1666 |
+
"loss": 0.0011,
|
| 1667 |
+
"step": 1880
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 38.01166666666666,
|
| 1671 |
+
"grad_norm": 0.004206162411719561,
|
| 1672 |
+
"learning_rate": 2.361111111111111e-06,
|
| 1673 |
+
"loss": 0.0001,
|
| 1674 |
+
"step": 1890
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 38.01583333333333,
|
| 1678 |
+
"grad_norm": 0.003568112850189209,
|
| 1679 |
+
"learning_rate": 2.314814814814815e-06,
|
| 1680 |
+
"loss": 0.0001,
|
| 1681 |
+
"step": 1900
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 38.02,
|
| 1685 |
+
"grad_norm": 0.03349081426858902,
|
| 1686 |
+
"learning_rate": 2.268518518518519e-06,
|
| 1687 |
+
"loss": 0.0002,
|
| 1688 |
+
"step": 1910
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 38.02041666666667,
|
| 1692 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1693 |
+
"eval_loss": 2.300963878631592,
|
| 1694 |
+
"eval_runtime": 3.4283,
|
| 1695 |
+
"eval_samples_per_second": 5.542,
|
| 1696 |
+
"eval_steps_per_second": 1.458,
|
| 1697 |
+
"step": 1911
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 39.00375,
|
| 1701 |
+
"grad_norm": 0.0027008787728846073,
|
| 1702 |
+
"learning_rate": 2.222222222222222e-06,
|
| 1703 |
+
"loss": 0.0001,
|
| 1704 |
+
"step": 1920
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 39.00791666666667,
|
| 1708 |
+
"grad_norm": 0.0040114871226251125,
|
| 1709 |
+
"learning_rate": 2.175925925925926e-06,
|
| 1710 |
+
"loss": 0.0001,
|
| 1711 |
+
"step": 1930
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 39.01208333333334,
|
| 1715 |
+
"grad_norm": 0.002579754451289773,
|
| 1716 |
+
"learning_rate": 2.1296296296296298e-06,
|
| 1717 |
+
"loss": 0.0001,
|
| 1718 |
+
"step": 1940
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 39.01625,
|
| 1722 |
+
"grad_norm": 0.003918965347111225,
|
| 1723 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 1724 |
+
"loss": 0.0001,
|
| 1725 |
+
"step": 1950
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 39.02041666666667,
|
| 1729 |
+
"grad_norm": 0.002216913038864732,
|
| 1730 |
+
"learning_rate": 2.037037037037037e-06,
|
| 1731 |
+
"loss": 0.0001,
|
| 1732 |
+
"step": 1960
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 39.02041666666667,
|
| 1736 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1737 |
+
"eval_loss": 2.3123650550842285,
|
| 1738 |
+
"eval_runtime": 3.4399,
|
| 1739 |
+
"eval_samples_per_second": 5.523,
|
| 1740 |
+
"eval_steps_per_second": 1.454,
|
| 1741 |
+
"step": 1960
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 40.00416666666667,
|
| 1745 |
+
"grad_norm": 0.001679323031567037,
|
| 1746 |
+
"learning_rate": 1.9907407407407407e-06,
|
| 1747 |
+
"loss": 0.0001,
|
| 1748 |
+
"step": 1970
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 40.00833333333333,
|
| 1752 |
+
"grad_norm": 0.004570376593619585,
|
| 1753 |
+
"learning_rate": 1.944444444444445e-06,
|
| 1754 |
+
"loss": 0.0028,
|
| 1755 |
+
"step": 1980
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 40.0125,
|
| 1759 |
+
"grad_norm": 0.0024077030830085278,
|
| 1760 |
+
"learning_rate": 1.8981481481481484e-06,
|
| 1761 |
+
"loss": 0.078,
|
| 1762 |
+
"step": 1990
|
| 1763 |
+
},
|
| 1764 |
+
{
|
| 1765 |
+
"epoch": 40.016666666666666,
|
| 1766 |
+
"grad_norm": 0.0020433454774320126,
|
| 1767 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 1768 |
+
"loss": 0.0001,
|
| 1769 |
+
"step": 2000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 40.02041666666667,
|
| 1773 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1774 |
+
"eval_loss": 2.2903194427490234,
|
| 1775 |
+
"eval_runtime": 3.4213,
|
| 1776 |
+
"eval_samples_per_second": 5.553,
|
| 1777 |
+
"eval_steps_per_second": 1.461,
|
| 1778 |
+
"step": 2009
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 41.000416666666666,
|
| 1782 |
+
"grad_norm": 0.007383950520306826,
|
| 1783 |
+
"learning_rate": 1.8055555555555557e-06,
|
| 1784 |
+
"loss": 0.0001,
|
| 1785 |
+
"step": 2010
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 41.004583333333336,
|
| 1789 |
+
"grad_norm": 0.005234722048044205,
|
| 1790 |
+
"learning_rate": 1.7592592592592594e-06,
|
| 1791 |
+
"loss": 0.0002,
|
| 1792 |
+
"step": 2020
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 41.00875,
|
| 1796 |
+
"grad_norm": 0.01151933241635561,
|
| 1797 |
+
"learning_rate": 1.7129629629629632e-06,
|
| 1798 |
+
"loss": 0.0001,
|
| 1799 |
+
"step": 2030
|
| 1800 |
+
},
|
| 1801 |
+
{
|
| 1802 |
+
"epoch": 41.01291666666667,
|
| 1803 |
+
"grad_norm": 0.03747610002756119,
|
| 1804 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 1805 |
+
"loss": 0.0001,
|
| 1806 |
+
"step": 2040
|
| 1807 |
+
},
|
| 1808 |
+
{
|
| 1809 |
+
"epoch": 41.01708333333333,
|
| 1810 |
+
"grad_norm": 0.00581372668966651,
|
| 1811 |
+
"learning_rate": 1.6203703703703705e-06,
|
| 1812 |
+
"loss": 0.1744,
|
| 1813 |
+
"step": 2050
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 41.02041666666667,
|
| 1817 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1818 |
+
"eval_loss": 2.3067450523376465,
|
| 1819 |
+
"eval_runtime": 3.4324,
|
| 1820 |
+
"eval_samples_per_second": 5.535,
|
| 1821 |
+
"eval_steps_per_second": 1.457,
|
| 1822 |
+
"step": 2058
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 42.00083333333333,
|
| 1826 |
+
"grad_norm": 0.003268251661211252,
|
| 1827 |
+
"learning_rate": 1.5740740740740742e-06,
|
| 1828 |
+
"loss": 0.0001,
|
| 1829 |
+
"step": 2060
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 42.005,
|
| 1833 |
+
"grad_norm": 0.005169464275240898,
|
| 1834 |
+
"learning_rate": 1.527777777777778e-06,
|
| 1835 |
+
"loss": 0.0001,
|
| 1836 |
+
"step": 2070
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 42.009166666666665,
|
| 1840 |
+
"grad_norm": 0.0015214293962344527,
|
| 1841 |
+
"learning_rate": 1.4814814814814815e-06,
|
| 1842 |
+
"loss": 0.1229,
|
| 1843 |
+
"step": 2080
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 42.013333333333335,
|
| 1847 |
+
"grad_norm": 249.32093811035156,
|
| 1848 |
+
"learning_rate": 1.4351851851851853e-06,
|
| 1849 |
+
"loss": 0.0128,
|
| 1850 |
+
"step": 2090
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 42.0175,
|
| 1854 |
+
"grad_norm": 0.007209462579339743,
|
| 1855 |
+
"learning_rate": 1.3888888888888892e-06,
|
| 1856 |
+
"loss": 0.0001,
|
| 1857 |
+
"step": 2100
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 42.02041666666667,
|
| 1861 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1862 |
+
"eval_loss": 2.342156171798706,
|
| 1863 |
+
"eval_runtime": 3.4328,
|
| 1864 |
+
"eval_samples_per_second": 5.535,
|
| 1865 |
+
"eval_steps_per_second": 1.457,
|
| 1866 |
+
"step": 2107
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 43.00125,
|
| 1870 |
+
"grad_norm": 0.0031976569443941116,
|
| 1871 |
+
"learning_rate": 1.3425925925925928e-06,
|
| 1872 |
+
"loss": 0.0001,
|
| 1873 |
+
"step": 2110
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 43.00541666666667,
|
| 1877 |
+
"grad_norm": 0.0036477481480687857,
|
| 1878 |
+
"learning_rate": 1.2962962962962962e-06,
|
| 1879 |
+
"loss": 0.0001,
|
| 1880 |
+
"step": 2120
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 43.00958333333333,
|
| 1884 |
+
"grad_norm": 0.00543107558041811,
|
| 1885 |
+
"learning_rate": 1.25e-06,
|
| 1886 |
+
"loss": 0.0001,
|
| 1887 |
+
"step": 2130
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 43.01375,
|
| 1891 |
+
"grad_norm": 0.006350014358758926,
|
| 1892 |
+
"learning_rate": 1.2037037037037037e-06,
|
| 1893 |
+
"loss": 0.0001,
|
| 1894 |
+
"step": 2140
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 43.017916666666665,
|
| 1898 |
+
"grad_norm": 0.008976898156106472,
|
| 1899 |
+
"learning_rate": 1.1574074074074076e-06,
|
| 1900 |
+
"loss": 0.0001,
|
| 1901 |
+
"step": 2150
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 43.02041666666667,
|
| 1905 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1906 |
+
"eval_loss": 2.3476860523223877,
|
| 1907 |
+
"eval_runtime": 3.4979,
|
| 1908 |
+
"eval_samples_per_second": 5.432,
|
| 1909 |
+
"eval_steps_per_second": 1.429,
|
| 1910 |
+
"step": 2156
|
| 1911 |
+
},
|
| 1912 |
+
{
|
| 1913 |
+
"epoch": 44.001666666666665,
|
| 1914 |
+
"grad_norm": 0.006657883990556002,
|
| 1915 |
+
"learning_rate": 1.111111111111111e-06,
|
| 1916 |
+
"loss": 0.0001,
|
| 1917 |
+
"step": 2160
|
| 1918 |
+
},
|
| 1919 |
+
{
|
| 1920 |
+
"epoch": 44.005833333333335,
|
| 1921 |
+
"grad_norm": 0.005237992387264967,
|
| 1922 |
+
"learning_rate": 1.0648148148148149e-06,
|
| 1923 |
+
"loss": 0.0001,
|
| 1924 |
+
"step": 2170
|
| 1925 |
+
},
|
| 1926 |
+
{
|
| 1927 |
+
"epoch": 44.01,
|
| 1928 |
+
"grad_norm": 0.003178667975589633,
|
| 1929 |
+
"learning_rate": 1.0185185185185185e-06,
|
| 1930 |
+
"loss": 0.0001,
|
| 1931 |
+
"step": 2180
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 44.01416666666667,
|
| 1935 |
+
"grad_norm": 0.0024578028824180365,
|
| 1936 |
+
"learning_rate": 9.722222222222224e-07,
|
| 1937 |
+
"loss": 0.0003,
|
| 1938 |
+
"step": 2190
|
| 1939 |
+
},
|
| 1940 |
+
{
|
| 1941 |
+
"epoch": 44.01833333333333,
|
| 1942 |
+
"grad_norm": 0.00786686223000288,
|
| 1943 |
+
"learning_rate": 9.259259259259259e-07,
|
| 1944 |
+
"loss": 0.0001,
|
| 1945 |
+
"step": 2200
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 44.02041666666667,
|
| 1949 |
+
"eval_accuracy": 0.631578947368421,
|
| 1950 |
+
"eval_loss": 2.378264904022217,
|
| 1951 |
+
"eval_runtime": 3.5284,
|
| 1952 |
+
"eval_samples_per_second": 5.385,
|
| 1953 |
+
"eval_steps_per_second": 1.417,
|
| 1954 |
+
"step": 2205
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 45.00208333333333,
|
| 1958 |
+
"grad_norm": 0.0038455643225461245,
|
| 1959 |
+
"learning_rate": 8.796296296296297e-07,
|
| 1960 |
+
"loss": 0.0002,
|
| 1961 |
+
"step": 2210
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 45.00625,
|
| 1965 |
+
"grad_norm": 0.0019002072513103485,
|
| 1966 |
+
"learning_rate": 8.333333333333333e-07,
|
| 1967 |
+
"loss": 0.0001,
|
| 1968 |
+
"step": 2220
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 45.010416666666664,
|
| 1972 |
+
"grad_norm": 0.002483098302036524,
|
| 1973 |
+
"learning_rate": 7.870370370370371e-07,
|
| 1974 |
+
"loss": 0.0001,
|
| 1975 |
+
"step": 2230
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 45.014583333333334,
|
| 1979 |
+
"grad_norm": 0.0028833537362515926,
|
| 1980 |
+
"learning_rate": 7.407407407407407e-07,
|
| 1981 |
+
"loss": 0.0002,
|
| 1982 |
+
"step": 2240
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 45.01875,
|
| 1986 |
+
"grad_norm": 0.0019563958048820496,
|
| 1987 |
+
"learning_rate": 6.944444444444446e-07,
|
| 1988 |
+
"loss": 0.0001,
|
| 1989 |
+
"step": 2250
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 45.02041666666667,
|
| 1993 |
+
"eval_accuracy": 0.6842105263157895,
|
| 1994 |
+
"eval_loss": 2.344691038131714,
|
| 1995 |
+
"eval_runtime": 3.4954,
|
| 1996 |
+
"eval_samples_per_second": 5.436,
|
| 1997 |
+
"eval_steps_per_second": 1.43,
|
| 1998 |
+
"step": 2254
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 46.0025,
|
| 2002 |
+
"grad_norm": 0.0030405442230403423,
|
| 2003 |
+
"learning_rate": 6.481481481481481e-07,
|
| 2004 |
+
"loss": 0.0001,
|
| 2005 |
+
"step": 2260
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 46.00666666666667,
|
| 2009 |
+
"grad_norm": 0.00329338270239532,
|
| 2010 |
+
"learning_rate": 6.018518518518519e-07,
|
| 2011 |
+
"loss": 0.0001,
|
| 2012 |
+
"step": 2270
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 46.01083333333333,
|
| 2016 |
+
"grad_norm": 0.00243846676312387,
|
| 2017 |
+
"learning_rate": 5.555555555555555e-07,
|
| 2018 |
+
"loss": 0.0001,
|
| 2019 |
+
"step": 2280
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 46.015,
|
| 2023 |
+
"grad_norm": 0.0017111854394897819,
|
| 2024 |
+
"learning_rate": 5.092592592592593e-07,
|
| 2025 |
+
"loss": 0.0001,
|
| 2026 |
+
"step": 2290
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 46.01916666666666,
|
| 2030 |
+
"grad_norm": 0.010095887817442417,
|
| 2031 |
+
"learning_rate": 4.6296296296296297e-07,
|
| 2032 |
+
"loss": 0.0001,
|
| 2033 |
+
"step": 2300
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 46.02041666666667,
|
| 2037 |
+
"eval_accuracy": 0.6842105263157895,
|
| 2038 |
+
"eval_loss": 2.3414864540100098,
|
| 2039 |
+
"eval_runtime": 3.4719,
|
| 2040 |
+
"eval_samples_per_second": 5.473,
|
| 2041 |
+
"eval_steps_per_second": 1.44,
|
| 2042 |
+
"step": 2303
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 47.002916666666664,
|
| 2046 |
+
"grad_norm": 0.004163892008364201,
|
| 2047 |
+
"learning_rate": 4.1666666666666667e-07,
|
| 2048 |
+
"loss": 0.0001,
|
| 2049 |
+
"step": 2310
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 47.007083333333334,
|
| 2053 |
+
"grad_norm": 0.002049618400633335,
|
| 2054 |
+
"learning_rate": 3.7037037037037036e-07,
|
| 2055 |
+
"loss": 0.0001,
|
| 2056 |
+
"step": 2320
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 47.01125,
|
| 2060 |
+
"grad_norm": 0.003938646987080574,
|
| 2061 |
+
"learning_rate": 3.2407407407407406e-07,
|
| 2062 |
+
"loss": 0.0001,
|
| 2063 |
+
"step": 2330
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 47.01541666666667,
|
| 2067 |
+
"grad_norm": 0.004701240919530392,
|
| 2068 |
+
"learning_rate": 2.7777777777777776e-07,
|
| 2069 |
+
"loss": 0.0013,
|
| 2070 |
+
"step": 2340
|
| 2071 |
+
},
|
| 2072 |
+
{
|
| 2073 |
+
"epoch": 47.01958333333334,
|
| 2074 |
+
"grad_norm": 0.002614920726045966,
|
| 2075 |
+
"learning_rate": 2.3148148148148148e-07,
|
| 2076 |
+
"loss": 0.0001,
|
| 2077 |
+
"step": 2350
|
| 2078 |
+
},
|
| 2079 |
+
{
|
| 2080 |
+
"epoch": 47.02041666666667,
|
| 2081 |
+
"eval_accuracy": 0.6842105263157895,
|
| 2082 |
+
"eval_loss": 2.341876745223999,
|
| 2083 |
+
"eval_runtime": 3.52,
|
| 2084 |
+
"eval_samples_per_second": 5.398,
|
| 2085 |
+
"eval_steps_per_second": 1.42,
|
| 2086 |
+
"step": 2352
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"epoch": 48.00333333333333,
|
| 2090 |
+
"grad_norm": 0.0040041073225438595,
|
| 2091 |
+
"learning_rate": 1.8518518518518518e-07,
|
| 2092 |
+
"loss": 0.0014,
|
| 2093 |
+
"step": 2360
|
| 2094 |
+
},
|
| 2095 |
+
{
|
| 2096 |
+
"epoch": 48.0075,
|
| 2097 |
+
"grad_norm": 0.0024590292014181614,
|
| 2098 |
+
"learning_rate": 1.3888888888888888e-07,
|
| 2099 |
+
"loss": 0.0001,
|
| 2100 |
+
"step": 2370
|
| 2101 |
+
},
|
| 2102 |
+
{
|
| 2103 |
+
"epoch": 48.01166666666666,
|
| 2104 |
+
"grad_norm": 0.004839394707232714,
|
| 2105 |
+
"learning_rate": 9.259259259259259e-08,
|
| 2106 |
+
"loss": 0.0002,
|
| 2107 |
+
"step": 2380
|
| 2108 |
+
},
|
| 2109 |
+
{
|
| 2110 |
+
"epoch": 48.01583333333333,
|
| 2111 |
+
"grad_norm": 0.0019098540069535375,
|
| 2112 |
+
"learning_rate": 4.6296296296296295e-08,
|
| 2113 |
+
"loss": 0.0002,
|
| 2114 |
+
"step": 2390
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 48.02,
|
| 2118 |
+
"grad_norm": 0.00532747246325016,
|
| 2119 |
+
"learning_rate": 0.0,
|
| 2120 |
+
"loss": 0.0001,
|
| 2121 |
+
"step": 2400
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 48.02,
|
| 2125 |
+
"eval_accuracy": 0.6842105263157895,
|
| 2126 |
+
"eval_loss": 2.3470160961151123,
|
| 2127 |
+
"eval_runtime": 4.2841,
|
| 2128 |
+
"eval_samples_per_second": 4.435,
|
| 2129 |
+
"eval_steps_per_second": 1.167,
|
| 2130 |
+
"step": 2400
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 48.02,
|
| 2134 |
+
"step": 2400,
|
| 2135 |
+
"total_flos": 4.194308336371473e+19,
|
| 2136 |
+
"train_loss": 0.18858948054393598,
|
| 2137 |
+
"train_runtime": 4106.9782,
|
| 2138 |
+
"train_samples_per_second": 2.337,
|
| 2139 |
+
"train_steps_per_second": 0.584
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 48.02,
|
| 2143 |
+
"eval_accuracy": 0.5,
|
| 2144 |
+
"eval_loss": 1.6418968439102173,
|
| 2145 |
+
"eval_runtime": 4.996,
|
| 2146 |
+
"eval_samples_per_second": 4.804,
|
| 2147 |
+
"eval_steps_per_second": 1.201,
|
| 2148 |
+
"step": 2400
|
| 2149 |
+
},
|
| 2150 |
+
{
|
| 2151 |
+
"epoch": 48.02,
|
| 2152 |
+
"eval_accuracy": 0.5,
|
| 2153 |
+
"eval_loss": 1.6418968439102173,
|
| 2154 |
+
"eval_runtime": 4.5735,
|
| 2155 |
+
"eval_samples_per_second": 5.248,
|
| 2156 |
+
"eval_steps_per_second": 1.312,
|
| 2157 |
+
"step": 2400
|
| 2158 |
+
}
|
| 2159 |
+
],
|
| 2160 |
+
"logging_steps": 10,
|
| 2161 |
+
"max_steps": 2400,
|
| 2162 |
+
"num_input_tokens_seen": 0,
|
| 2163 |
+
"num_train_epochs": 9223372036854775807,
|
| 2164 |
+
"save_steps": 500,
|
| 2165 |
+
"stateful_callbacks": {
|
| 2166 |
+
"TrainerControl": {
|
| 2167 |
+
"args": {
|
| 2168 |
+
"should_epoch_stop": false,
|
| 2169 |
+
"should_evaluate": false,
|
| 2170 |
+
"should_log": false,
|
| 2171 |
+
"should_save": true,
|
| 2172 |
+
"should_training_stop": true
|
| 2173 |
+
},
|
| 2174 |
+
"attributes": {}
|
| 2175 |
+
}
|
| 2176 |
+
},
|
| 2177 |
+
"total_flos": 4.194308336371473e+19,
|
| 2178 |
+
"train_batch_size": 4,
|
| 2179 |
+
"trial_name": null,
|
| 2180 |
+
"trial_params": null
|
| 2181 |
+
}
|