End of training
Browse files- all_results.json +8 -0
- test_results.json +8 -0
- trainer_state.json +2160 -0
all_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 49.01,
|
3 |
+
"eval_accuracy": 0.7824074074074074,
|
4 |
+
"eval_loss": 0.6378525495529175,
|
5 |
+
"eval_runtime": 181.9628,
|
6 |
+
"eval_samples_per_second": 1.187,
|
7 |
+
"eval_steps_per_second": 0.148
|
8 |
+
}
|
test_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 49.01,
|
3 |
+
"eval_accuracy": 0.7824074074074074,
|
4 |
+
"eval_loss": 0.6378525495529175,
|
5 |
+
"eval_runtime": 181.9628,
|
6 |
+
"eval_samples_per_second": 1.187,
|
7 |
+
"eval_steps_per_second": 0.148
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,2160 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.7926267281105991,
|
3 |
+
"best_model_checkpoint": "videomae-base-finetuned-subset-0401/checkpoint-896",
|
4 |
+
"epoch": 49.01117117117117,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 2775,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0,
|
13 |
+
"learning_rate": 1.7985611510791366e-06,
|
14 |
+
"loss": 1.6857,
|
15 |
+
"step": 10
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.01,
|
19 |
+
"learning_rate": 3.5971223021582732e-06,
|
20 |
+
"loss": 1.6911,
|
21 |
+
"step": 20
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.01,
|
25 |
+
"learning_rate": 5.3956834532374105e-06,
|
26 |
+
"loss": 1.6168,
|
27 |
+
"step": 30
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.01,
|
31 |
+
"learning_rate": 7.1942446043165465e-06,
|
32 |
+
"loss": 1.6689,
|
33 |
+
"step": 40
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.02,
|
37 |
+
"learning_rate": 8.992805755395683e-06,
|
38 |
+
"loss": 1.6048,
|
39 |
+
"step": 50
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.02,
|
43 |
+
"eval_accuracy": 0.08294930875576037,
|
44 |
+
"eval_loss": 1.6212950944900513,
|
45 |
+
"eval_runtime": 176.5128,
|
46 |
+
"eval_samples_per_second": 1.229,
|
47 |
+
"eval_steps_per_second": 0.159,
|
48 |
+
"step": 56
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 1.0,
|
52 |
+
"learning_rate": 1.0791366906474821e-05,
|
53 |
+
"loss": 1.6118,
|
54 |
+
"step": 60
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 1.01,
|
58 |
+
"learning_rate": 1.2589928057553957e-05,
|
59 |
+
"loss": 1.6008,
|
60 |
+
"step": 70
|
61 |
+
},
|
62 |
+
{
|
63 |
+
"epoch": 1.01,
|
64 |
+
"learning_rate": 1.4388489208633093e-05,
|
65 |
+
"loss": 1.5996,
|
66 |
+
"step": 80
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 1.01,
|
70 |
+
"learning_rate": 1.618705035971223e-05,
|
71 |
+
"loss": 1.6264,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 1.02,
|
76 |
+
"learning_rate": 1.7985611510791367e-05,
|
77 |
+
"loss": 1.6071,
|
78 |
+
"step": 100
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 1.02,
|
82 |
+
"learning_rate": 1.9784172661870504e-05,
|
83 |
+
"loss": 1.5891,
|
84 |
+
"step": 110
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 1.02,
|
88 |
+
"eval_accuracy": 0.28110599078341014,
|
89 |
+
"eval_loss": 1.5230350494384766,
|
90 |
+
"eval_runtime": 169.5488,
|
91 |
+
"eval_samples_per_second": 1.28,
|
92 |
+
"eval_steps_per_second": 0.165,
|
93 |
+
"step": 112
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 2.0,
|
97 |
+
"learning_rate": 2.1582733812949642e-05,
|
98 |
+
"loss": 1.5625,
|
99 |
+
"step": 120
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 2.01,
|
103 |
+
"learning_rate": 2.3381294964028776e-05,
|
104 |
+
"loss": 1.5854,
|
105 |
+
"step": 130
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 2.01,
|
109 |
+
"learning_rate": 2.5179856115107914e-05,
|
110 |
+
"loss": 1.6007,
|
111 |
+
"step": 140
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 2.01,
|
115 |
+
"learning_rate": 2.697841726618705e-05,
|
116 |
+
"loss": 1.5949,
|
117 |
+
"step": 150
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 2.02,
|
121 |
+
"learning_rate": 2.8776978417266186e-05,
|
122 |
+
"loss": 1.4797,
|
123 |
+
"step": 160
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 2.02,
|
127 |
+
"eval_accuracy": 0.19815668202764977,
|
128 |
+
"eval_loss": 1.6437386274337769,
|
129 |
+
"eval_runtime": 169.6444,
|
130 |
+
"eval_samples_per_second": 1.279,
|
131 |
+
"eval_steps_per_second": 0.165,
|
132 |
+
"step": 168
|
133 |
+
},
|
134 |
+
{
|
135 |
+
"epoch": 3.0,
|
136 |
+
"learning_rate": 3.0575539568345324e-05,
|
137 |
+
"loss": 1.5867,
|
138 |
+
"step": 170
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 3.0,
|
142 |
+
"learning_rate": 3.237410071942446e-05,
|
143 |
+
"loss": 1.4592,
|
144 |
+
"step": 180
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 3.01,
|
148 |
+
"learning_rate": 3.41726618705036e-05,
|
149 |
+
"loss": 1.3124,
|
150 |
+
"step": 190
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 3.01,
|
154 |
+
"learning_rate": 3.597122302158273e-05,
|
155 |
+
"loss": 1.3075,
|
156 |
+
"step": 200
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 3.02,
|
160 |
+
"learning_rate": 3.776978417266187e-05,
|
161 |
+
"loss": 1.1932,
|
162 |
+
"step": 210
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 3.02,
|
166 |
+
"learning_rate": 3.956834532374101e-05,
|
167 |
+
"loss": 1.3999,
|
168 |
+
"step": 220
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 3.02,
|
172 |
+
"eval_accuracy": 0.7465437788018433,
|
173 |
+
"eval_loss": 0.9262827634811401,
|
174 |
+
"eval_runtime": 168.7616,
|
175 |
+
"eval_samples_per_second": 1.286,
|
176 |
+
"eval_steps_per_second": 0.166,
|
177 |
+
"step": 224
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 4.0,
|
181 |
+
"learning_rate": 4.136690647482014e-05,
|
182 |
+
"loss": 1.2996,
|
183 |
+
"step": 230
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 4.01,
|
187 |
+
"learning_rate": 4.3165467625899284e-05,
|
188 |
+
"loss": 1.313,
|
189 |
+
"step": 240
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 4.01,
|
193 |
+
"learning_rate": 4.496402877697842e-05,
|
194 |
+
"loss": 1.2644,
|
195 |
+
"step": 250
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 4.01,
|
199 |
+
"learning_rate": 4.676258992805755e-05,
|
200 |
+
"loss": 1.1732,
|
201 |
+
"step": 260
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 4.02,
|
205 |
+
"learning_rate": 4.8561151079136694e-05,
|
206 |
+
"loss": 1.125,
|
207 |
+
"step": 270
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 4.02,
|
211 |
+
"learning_rate": 4.9959951942330803e-05,
|
212 |
+
"loss": 1.0917,
|
213 |
+
"step": 280
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 4.02,
|
217 |
+
"eval_accuracy": 0.4930875576036866,
|
218 |
+
"eval_loss": 1.2307826280593872,
|
219 |
+
"eval_runtime": 169.9213,
|
220 |
+
"eval_samples_per_second": 1.277,
|
221 |
+
"eval_steps_per_second": 0.165,
|
222 |
+
"step": 280
|
223 |
+
},
|
224 |
+
{
|
225 |
+
"epoch": 5.0,
|
226 |
+
"learning_rate": 4.975971165398479e-05,
|
227 |
+
"loss": 1.2046,
|
228 |
+
"step": 290
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 5.01,
|
232 |
+
"learning_rate": 4.955947136563877e-05,
|
233 |
+
"loss": 1.2355,
|
234 |
+
"step": 300
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 5.01,
|
238 |
+
"learning_rate": 4.935923107729276e-05,
|
239 |
+
"loss": 1.3358,
|
240 |
+
"step": 310
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 5.01,
|
244 |
+
"learning_rate": 4.915899078894674e-05,
|
245 |
+
"loss": 1.1835,
|
246 |
+
"step": 320
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 5.02,
|
250 |
+
"learning_rate": 4.895875050060073e-05,
|
251 |
+
"loss": 1.238,
|
252 |
+
"step": 330
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 5.02,
|
256 |
+
"eval_accuracy": 0.6589861751152074,
|
257 |
+
"eval_loss": 0.940632700920105,
|
258 |
+
"eval_runtime": 168.194,
|
259 |
+
"eval_samples_per_second": 1.29,
|
260 |
+
"eval_steps_per_second": 0.166,
|
261 |
+
"step": 336
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 6.0,
|
265 |
+
"learning_rate": 4.875851021225471e-05,
|
266 |
+
"loss": 1.3625,
|
267 |
+
"step": 340
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 6.01,
|
271 |
+
"learning_rate": 4.8558269923908696e-05,
|
272 |
+
"loss": 1.2342,
|
273 |
+
"step": 350
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 6.01,
|
277 |
+
"learning_rate": 4.835802963556268e-05,
|
278 |
+
"loss": 1.0534,
|
279 |
+
"step": 360
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 6.01,
|
283 |
+
"learning_rate": 4.8157789347216665e-05,
|
284 |
+
"loss": 1.1462,
|
285 |
+
"step": 370
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 6.02,
|
289 |
+
"learning_rate": 4.795754905887065e-05,
|
290 |
+
"loss": 1.0197,
|
291 |
+
"step": 380
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 6.02,
|
295 |
+
"learning_rate": 4.7757308770524635e-05,
|
296 |
+
"loss": 1.1525,
|
297 |
+
"step": 390
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 6.02,
|
301 |
+
"eval_accuracy": 0.7050691244239631,
|
302 |
+
"eval_loss": 0.8809296488761902,
|
303 |
+
"eval_runtime": 168.4512,
|
304 |
+
"eval_samples_per_second": 1.288,
|
305 |
+
"eval_steps_per_second": 0.166,
|
306 |
+
"step": 392
|
307 |
+
},
|
308 |
+
{
|
309 |
+
"epoch": 7.0,
|
310 |
+
"learning_rate": 4.755706848217862e-05,
|
311 |
+
"loss": 1.0159,
|
312 |
+
"step": 400
|
313 |
+
},
|
314 |
+
{
|
315 |
+
"epoch": 7.01,
|
316 |
+
"learning_rate": 4.7356828193832604e-05,
|
317 |
+
"loss": 0.9716,
|
318 |
+
"step": 410
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 7.01,
|
322 |
+
"learning_rate": 4.715658790548659e-05,
|
323 |
+
"loss": 0.9998,
|
324 |
+
"step": 420
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 7.01,
|
328 |
+
"learning_rate": 4.6956347617140574e-05,
|
329 |
+
"loss": 1.1153,
|
330 |
+
"step": 430
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 7.02,
|
334 |
+
"learning_rate": 4.675610732879456e-05,
|
335 |
+
"loss": 1.0806,
|
336 |
+
"step": 440
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 7.02,
|
340 |
+
"eval_accuracy": 0.5944700460829493,
|
341 |
+
"eval_loss": 1.0089017152786255,
|
342 |
+
"eval_runtime": 171.2435,
|
343 |
+
"eval_samples_per_second": 1.267,
|
344 |
+
"eval_steps_per_second": 0.164,
|
345 |
+
"step": 448
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 8.0,
|
349 |
+
"learning_rate": 4.655586704044854e-05,
|
350 |
+
"loss": 1.1055,
|
351 |
+
"step": 450
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 8.0,
|
355 |
+
"learning_rate": 4.635562675210253e-05,
|
356 |
+
"loss": 0.9111,
|
357 |
+
"step": 460
|
358 |
+
},
|
359 |
+
{
|
360 |
+
"epoch": 8.01,
|
361 |
+
"learning_rate": 4.615538646375651e-05,
|
362 |
+
"loss": 1.0937,
|
363 |
+
"step": 470
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 8.01,
|
367 |
+
"learning_rate": 4.59551461754105e-05,
|
368 |
+
"loss": 1.1375,
|
369 |
+
"step": 480
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 8.02,
|
373 |
+
"learning_rate": 4.575490588706448e-05,
|
374 |
+
"loss": 0.8357,
|
375 |
+
"step": 490
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"epoch": 8.02,
|
379 |
+
"learning_rate": 4.5554665598718466e-05,
|
380 |
+
"loss": 0.8483,
|
381 |
+
"step": 500
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 8.02,
|
385 |
+
"eval_accuracy": 0.5852534562211982,
|
386 |
+
"eval_loss": 0.9700098037719727,
|
387 |
+
"eval_runtime": 171.1205,
|
388 |
+
"eval_samples_per_second": 1.268,
|
389 |
+
"eval_steps_per_second": 0.164,
|
390 |
+
"step": 504
|
391 |
+
},
|
392 |
+
{
|
393 |
+
"epoch": 9.0,
|
394 |
+
"learning_rate": 4.535442531037245e-05,
|
395 |
+
"loss": 0.9293,
|
396 |
+
"step": 510
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 9.01,
|
400 |
+
"learning_rate": 4.5154185022026436e-05,
|
401 |
+
"loss": 1.1123,
|
402 |
+
"step": 520
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 9.01,
|
406 |
+
"learning_rate": 4.495394473368042e-05,
|
407 |
+
"loss": 1.0562,
|
408 |
+
"step": 530
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 9.01,
|
412 |
+
"learning_rate": 4.4753704445334405e-05,
|
413 |
+
"loss": 0.8454,
|
414 |
+
"step": 540
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 9.02,
|
418 |
+
"learning_rate": 4.455346415698839e-05,
|
419 |
+
"loss": 0.8096,
|
420 |
+
"step": 550
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 9.02,
|
424 |
+
"learning_rate": 4.4353223868642374e-05,
|
425 |
+
"loss": 0.992,
|
426 |
+
"step": 560
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 9.02,
|
430 |
+
"eval_accuracy": 0.48847926267281105,
|
431 |
+
"eval_loss": 1.1879829168319702,
|
432 |
+
"eval_runtime": 168.7686,
|
433 |
+
"eval_samples_per_second": 1.286,
|
434 |
+
"eval_steps_per_second": 0.166,
|
435 |
+
"step": 560
|
436 |
+
},
|
437 |
+
{
|
438 |
+
"epoch": 10.0,
|
439 |
+
"learning_rate": 4.415298358029636e-05,
|
440 |
+
"loss": 1.0682,
|
441 |
+
"step": 570
|
442 |
+
},
|
443 |
+
{
|
444 |
+
"epoch": 10.01,
|
445 |
+
"learning_rate": 4.3952743291950344e-05,
|
446 |
+
"loss": 0.954,
|
447 |
+
"step": 580
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 10.01,
|
451 |
+
"learning_rate": 4.375250300360433e-05,
|
452 |
+
"loss": 0.9788,
|
453 |
+
"step": 590
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 10.01,
|
457 |
+
"learning_rate": 4.355226271525831e-05,
|
458 |
+
"loss": 0.975,
|
459 |
+
"step": 600
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 10.02,
|
463 |
+
"learning_rate": 4.33520224269123e-05,
|
464 |
+
"loss": 0.862,
|
465 |
+
"step": 610
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 10.02,
|
469 |
+
"eval_accuracy": 0.7511520737327189,
|
470 |
+
"eval_loss": 0.7174272537231445,
|
471 |
+
"eval_runtime": 168.7207,
|
472 |
+
"eval_samples_per_second": 1.286,
|
473 |
+
"eval_steps_per_second": 0.166,
|
474 |
+
"step": 616
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 11.0,
|
478 |
+
"learning_rate": 4.315178213856628e-05,
|
479 |
+
"loss": 0.7987,
|
480 |
+
"step": 620
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 11.01,
|
484 |
+
"learning_rate": 4.295154185022027e-05,
|
485 |
+
"loss": 0.7351,
|
486 |
+
"step": 630
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 11.01,
|
490 |
+
"learning_rate": 4.275130156187425e-05,
|
491 |
+
"loss": 0.8315,
|
492 |
+
"step": 640
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 11.01,
|
496 |
+
"learning_rate": 4.2551061273528236e-05,
|
497 |
+
"loss": 0.92,
|
498 |
+
"step": 650
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 11.02,
|
502 |
+
"learning_rate": 4.235082098518222e-05,
|
503 |
+
"loss": 0.6532,
|
504 |
+
"step": 660
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 11.02,
|
508 |
+
"learning_rate": 4.2150580696836206e-05,
|
509 |
+
"loss": 1.0694,
|
510 |
+
"step": 670
|
511 |
+
},
|
512 |
+
{
|
513 |
+
"epoch": 11.02,
|
514 |
+
"eval_accuracy": 0.7142857142857143,
|
515 |
+
"eval_loss": 0.8598370552062988,
|
516 |
+
"eval_runtime": 168.9431,
|
517 |
+
"eval_samples_per_second": 1.284,
|
518 |
+
"eval_steps_per_second": 0.166,
|
519 |
+
"step": 672
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 12.0,
|
523 |
+
"learning_rate": 4.195034040849019e-05,
|
524 |
+
"loss": 0.7898,
|
525 |
+
"step": 680
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 12.01,
|
529 |
+
"learning_rate": 4.1750100120144175e-05,
|
530 |
+
"loss": 0.8624,
|
531 |
+
"step": 690
|
532 |
+
},
|
533 |
+
{
|
534 |
+
"epoch": 12.01,
|
535 |
+
"learning_rate": 4.154985983179816e-05,
|
536 |
+
"loss": 0.9415,
|
537 |
+
"step": 700
|
538 |
+
},
|
539 |
+
{
|
540 |
+
"epoch": 12.01,
|
541 |
+
"learning_rate": 4.1349619543452144e-05,
|
542 |
+
"loss": 0.806,
|
543 |
+
"step": 710
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 12.02,
|
547 |
+
"learning_rate": 4.114937925510613e-05,
|
548 |
+
"loss": 0.8885,
|
549 |
+
"step": 720
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 12.02,
|
553 |
+
"eval_accuracy": 0.7096774193548387,
|
554 |
+
"eval_loss": 0.8289774656295776,
|
555 |
+
"eval_runtime": 169.3457,
|
556 |
+
"eval_samples_per_second": 1.281,
|
557 |
+
"eval_steps_per_second": 0.165,
|
558 |
+
"step": 728
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 13.0,
|
562 |
+
"learning_rate": 4.0949138966760114e-05,
|
563 |
+
"loss": 0.8971,
|
564 |
+
"step": 730
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 13.0,
|
568 |
+
"learning_rate": 4.07488986784141e-05,
|
569 |
+
"loss": 0.752,
|
570 |
+
"step": 740
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 13.01,
|
574 |
+
"learning_rate": 4.054865839006808e-05,
|
575 |
+
"loss": 0.9724,
|
576 |
+
"step": 750
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 13.01,
|
580 |
+
"learning_rate": 4.034841810172207e-05,
|
581 |
+
"loss": 0.9637,
|
582 |
+
"step": 760
|
583 |
+
},
|
584 |
+
{
|
585 |
+
"epoch": 13.02,
|
586 |
+
"learning_rate": 4.014817781337605e-05,
|
587 |
+
"loss": 1.0046,
|
588 |
+
"step": 770
|
589 |
+
},
|
590 |
+
{
|
591 |
+
"epoch": 13.02,
|
592 |
+
"learning_rate": 3.994793752503004e-05,
|
593 |
+
"loss": 0.8965,
|
594 |
+
"step": 780
|
595 |
+
},
|
596 |
+
{
|
597 |
+
"epoch": 13.02,
|
598 |
+
"eval_accuracy": 0.7142857142857143,
|
599 |
+
"eval_loss": 0.8304111361503601,
|
600 |
+
"eval_runtime": 171.3216,
|
601 |
+
"eval_samples_per_second": 1.267,
|
602 |
+
"eval_steps_per_second": 0.163,
|
603 |
+
"step": 784
|
604 |
+
},
|
605 |
+
{
|
606 |
+
"epoch": 14.0,
|
607 |
+
"learning_rate": 3.974769723668402e-05,
|
608 |
+
"loss": 0.9904,
|
609 |
+
"step": 790
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 14.01,
|
613 |
+
"learning_rate": 3.9547456948338006e-05,
|
614 |
+
"loss": 0.9846,
|
615 |
+
"step": 800
|
616 |
+
},
|
617 |
+
{
|
618 |
+
"epoch": 14.01,
|
619 |
+
"learning_rate": 3.934721665999199e-05,
|
620 |
+
"loss": 0.9145,
|
621 |
+
"step": 810
|
622 |
+
},
|
623 |
+
{
|
624 |
+
"epoch": 14.01,
|
625 |
+
"learning_rate": 3.9146976371645976e-05,
|
626 |
+
"loss": 0.7486,
|
627 |
+
"step": 820
|
628 |
+
},
|
629 |
+
{
|
630 |
+
"epoch": 14.02,
|
631 |
+
"learning_rate": 3.894673608329996e-05,
|
632 |
+
"loss": 0.7721,
|
633 |
+
"step": 830
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 14.02,
|
637 |
+
"learning_rate": 3.8746495794953945e-05,
|
638 |
+
"loss": 0.7371,
|
639 |
+
"step": 840
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 14.02,
|
643 |
+
"eval_accuracy": 0.7695852534562212,
|
644 |
+
"eval_loss": 0.7009211778640747,
|
645 |
+
"eval_runtime": 168.9295,
|
646 |
+
"eval_samples_per_second": 1.285,
|
647 |
+
"eval_steps_per_second": 0.166,
|
648 |
+
"step": 840
|
649 |
+
},
|
650 |
+
{
|
651 |
+
"epoch": 15.0,
|
652 |
+
"learning_rate": 3.854625550660793e-05,
|
653 |
+
"loss": 0.923,
|
654 |
+
"step": 850
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 15.01,
|
658 |
+
"learning_rate": 3.8346015218261915e-05,
|
659 |
+
"loss": 0.9008,
|
660 |
+
"step": 860
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 15.01,
|
664 |
+
"learning_rate": 3.81457749299159e-05,
|
665 |
+
"loss": 0.7429,
|
666 |
+
"step": 870
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 15.01,
|
670 |
+
"learning_rate": 3.7945534641569884e-05,
|
671 |
+
"loss": 0.6912,
|
672 |
+
"step": 880
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 15.02,
|
676 |
+
"learning_rate": 3.774529435322387e-05,
|
677 |
+
"loss": 0.6872,
|
678 |
+
"step": 890
|
679 |
+
},
|
680 |
+
{
|
681 |
+
"epoch": 15.02,
|
682 |
+
"eval_accuracy": 0.7926267281105991,
|
683 |
+
"eval_loss": 0.6768301129341125,
|
684 |
+
"eval_runtime": 170.318,
|
685 |
+
"eval_samples_per_second": 1.274,
|
686 |
+
"eval_steps_per_second": 0.164,
|
687 |
+
"step": 896
|
688 |
+
},
|
689 |
+
{
|
690 |
+
"epoch": 16.0,
|
691 |
+
"learning_rate": 3.754505406487785e-05,
|
692 |
+
"loss": 0.636,
|
693 |
+
"step": 900
|
694 |
+
},
|
695 |
+
{
|
696 |
+
"epoch": 16.01,
|
697 |
+
"learning_rate": 3.734481377653184e-05,
|
698 |
+
"loss": 0.7347,
|
699 |
+
"step": 910
|
700 |
+
},
|
701 |
+
{
|
702 |
+
"epoch": 16.01,
|
703 |
+
"learning_rate": 3.714457348818582e-05,
|
704 |
+
"loss": 0.7698,
|
705 |
+
"step": 920
|
706 |
+
},
|
707 |
+
{
|
708 |
+
"epoch": 16.01,
|
709 |
+
"learning_rate": 3.694433319983981e-05,
|
710 |
+
"loss": 0.7495,
|
711 |
+
"step": 930
|
712 |
+
},
|
713 |
+
{
|
714 |
+
"epoch": 16.02,
|
715 |
+
"learning_rate": 3.674409291149379e-05,
|
716 |
+
"loss": 0.725,
|
717 |
+
"step": 940
|
718 |
+
},
|
719 |
+
{
|
720 |
+
"epoch": 16.02,
|
721 |
+
"learning_rate": 3.6543852623147777e-05,
|
722 |
+
"loss": 0.6022,
|
723 |
+
"step": 950
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 16.02,
|
727 |
+
"eval_accuracy": 0.7373271889400922,
|
728 |
+
"eval_loss": 0.7512706518173218,
|
729 |
+
"eval_runtime": 168.81,
|
730 |
+
"eval_samples_per_second": 1.285,
|
731 |
+
"eval_steps_per_second": 0.166,
|
732 |
+
"step": 952
|
733 |
+
},
|
734 |
+
{
|
735 |
+
"epoch": 17.0,
|
736 |
+
"learning_rate": 3.634361233480176e-05,
|
737 |
+
"loss": 0.8774,
|
738 |
+
"step": 960
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 17.01,
|
742 |
+
"learning_rate": 3.6143372046455746e-05,
|
743 |
+
"loss": 0.6907,
|
744 |
+
"step": 970
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 17.01,
|
748 |
+
"learning_rate": 3.594313175810973e-05,
|
749 |
+
"loss": 0.6035,
|
750 |
+
"step": 980
|
751 |
+
},
|
752 |
+
{
|
753 |
+
"epoch": 17.01,
|
754 |
+
"learning_rate": 3.5742891469763715e-05,
|
755 |
+
"loss": 0.7363,
|
756 |
+
"step": 990
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 17.02,
|
760 |
+
"learning_rate": 3.55426511814177e-05,
|
761 |
+
"loss": 0.9308,
|
762 |
+
"step": 1000
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 17.02,
|
766 |
+
"eval_accuracy": 0.7096774193548387,
|
767 |
+
"eval_loss": 0.8054620623588562,
|
768 |
+
"eval_runtime": 169.493,
|
769 |
+
"eval_samples_per_second": 1.28,
|
770 |
+
"eval_steps_per_second": 0.165,
|
771 |
+
"step": 1008
|
772 |
+
},
|
773 |
+
{
|
774 |
+
"epoch": 18.0,
|
775 |
+
"learning_rate": 3.5342410893071685e-05,
|
776 |
+
"loss": 0.73,
|
777 |
+
"step": 1010
|
778 |
+
},
|
779 |
+
{
|
780 |
+
"epoch": 18.0,
|
781 |
+
"learning_rate": 3.514217060472567e-05,
|
782 |
+
"loss": 0.9107,
|
783 |
+
"step": 1020
|
784 |
+
},
|
785 |
+
{
|
786 |
+
"epoch": 18.01,
|
787 |
+
"learning_rate": 3.4941930316379654e-05,
|
788 |
+
"loss": 0.739,
|
789 |
+
"step": 1030
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 18.01,
|
793 |
+
"learning_rate": 3.474169002803364e-05,
|
794 |
+
"loss": 0.8472,
|
795 |
+
"step": 1040
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 18.02,
|
799 |
+
"learning_rate": 3.454144973968762e-05,
|
800 |
+
"loss": 0.7395,
|
801 |
+
"step": 1050
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 18.02,
|
805 |
+
"learning_rate": 3.434120945134161e-05,
|
806 |
+
"loss": 0.4456,
|
807 |
+
"step": 1060
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 18.02,
|
811 |
+
"eval_accuracy": 0.6728110599078341,
|
812 |
+
"eval_loss": 0.7876001000404358,
|
813 |
+
"eval_runtime": 172.0864,
|
814 |
+
"eval_samples_per_second": 1.261,
|
815 |
+
"eval_steps_per_second": 0.163,
|
816 |
+
"step": 1064
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 19.0,
|
820 |
+
"learning_rate": 3.414096916299559e-05,
|
821 |
+
"loss": 0.581,
|
822 |
+
"step": 1070
|
823 |
+
},
|
824 |
+
{
|
825 |
+
"epoch": 19.01,
|
826 |
+
"learning_rate": 3.394072887464958e-05,
|
827 |
+
"loss": 1.0824,
|
828 |
+
"step": 1080
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 19.01,
|
832 |
+
"learning_rate": 3.374048858630356e-05,
|
833 |
+
"loss": 0.6623,
|
834 |
+
"step": 1090
|
835 |
+
},
|
836 |
+
{
|
837 |
+
"epoch": 19.01,
|
838 |
+
"learning_rate": 3.354024829795755e-05,
|
839 |
+
"loss": 0.6078,
|
840 |
+
"step": 1100
|
841 |
+
},
|
842 |
+
{
|
843 |
+
"epoch": 19.02,
|
844 |
+
"learning_rate": 3.334000800961153e-05,
|
845 |
+
"loss": 0.6867,
|
846 |
+
"step": 1110
|
847 |
+
},
|
848 |
+
{
|
849 |
+
"epoch": 19.02,
|
850 |
+
"learning_rate": 3.3139767721265516e-05,
|
851 |
+
"loss": 0.6802,
|
852 |
+
"step": 1120
|
853 |
+
},
|
854 |
+
{
|
855 |
+
"epoch": 19.02,
|
856 |
+
"eval_accuracy": 0.7235023041474654,
|
857 |
+
"eval_loss": 0.722358763217926,
|
858 |
+
"eval_runtime": 170.3701,
|
859 |
+
"eval_samples_per_second": 1.274,
|
860 |
+
"eval_steps_per_second": 0.164,
|
861 |
+
"step": 1120
|
862 |
+
},
|
863 |
+
{
|
864 |
+
"epoch": 20.0,
|
865 |
+
"learning_rate": 3.29395274329195e-05,
|
866 |
+
"loss": 0.6218,
|
867 |
+
"step": 1130
|
868 |
+
},
|
869 |
+
{
|
870 |
+
"epoch": 20.01,
|
871 |
+
"learning_rate": 3.2739287144573485e-05,
|
872 |
+
"loss": 0.6368,
|
873 |
+
"step": 1140
|
874 |
+
},
|
875 |
+
{
|
876 |
+
"epoch": 20.01,
|
877 |
+
"learning_rate": 3.253904685622747e-05,
|
878 |
+
"loss": 0.7407,
|
879 |
+
"step": 1150
|
880 |
+
},
|
881 |
+
{
|
882 |
+
"epoch": 20.01,
|
883 |
+
"learning_rate": 3.2338806567881455e-05,
|
884 |
+
"loss": 0.6189,
|
885 |
+
"step": 1160
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 20.02,
|
889 |
+
"learning_rate": 3.213856627953544e-05,
|
890 |
+
"loss": 0.7154,
|
891 |
+
"step": 1170
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 20.02,
|
895 |
+
"eval_accuracy": 0.7050691244239631,
|
896 |
+
"eval_loss": 0.7434002161026001,
|
897 |
+
"eval_runtime": 168.861,
|
898 |
+
"eval_samples_per_second": 1.285,
|
899 |
+
"eval_steps_per_second": 0.166,
|
900 |
+
"step": 1176
|
901 |
+
},
|
902 |
+
{
|
903 |
+
"epoch": 21.0,
|
904 |
+
"learning_rate": 3.1938325991189424e-05,
|
905 |
+
"loss": 0.6084,
|
906 |
+
"step": 1180
|
907 |
+
},
|
908 |
+
{
|
909 |
+
"epoch": 21.01,
|
910 |
+
"learning_rate": 3.173808570284341e-05,
|
911 |
+
"loss": 0.656,
|
912 |
+
"step": 1190
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 21.01,
|
916 |
+
"learning_rate": 3.1537845414497393e-05,
|
917 |
+
"loss": 0.7274,
|
918 |
+
"step": 1200
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 21.01,
|
922 |
+
"learning_rate": 3.133760512615138e-05,
|
923 |
+
"loss": 0.5818,
|
924 |
+
"step": 1210
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 21.02,
|
928 |
+
"learning_rate": 3.113736483780536e-05,
|
929 |
+
"loss": 0.8351,
|
930 |
+
"step": 1220
|
931 |
+
},
|
932 |
+
{
|
933 |
+
"epoch": 21.02,
|
934 |
+
"learning_rate": 3.0937124549459354e-05,
|
935 |
+
"loss": 0.503,
|
936 |
+
"step": 1230
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 21.02,
|
940 |
+
"eval_accuracy": 0.695852534562212,
|
941 |
+
"eval_loss": 0.8345744609832764,
|
942 |
+
"eval_runtime": 168.411,
|
943 |
+
"eval_samples_per_second": 1.289,
|
944 |
+
"eval_steps_per_second": 0.166,
|
945 |
+
"step": 1232
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 22.0,
|
949 |
+
"learning_rate": 3.073688426111334e-05,
|
950 |
+
"loss": 0.5257,
|
951 |
+
"step": 1240
|
952 |
+
},
|
953 |
+
{
|
954 |
+
"epoch": 22.01,
|
955 |
+
"learning_rate": 3.0536643972767324e-05,
|
956 |
+
"loss": 0.511,
|
957 |
+
"step": 1250
|
958 |
+
},
|
959 |
+
{
|
960 |
+
"epoch": 22.01,
|
961 |
+
"learning_rate": 3.0336403684421305e-05,
|
962 |
+
"loss": 0.6193,
|
963 |
+
"step": 1260
|
964 |
+
},
|
965 |
+
{
|
966 |
+
"epoch": 22.01,
|
967 |
+
"learning_rate": 3.013616339607529e-05,
|
968 |
+
"loss": 0.6704,
|
969 |
+
"step": 1270
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 22.02,
|
973 |
+
"learning_rate": 2.9935923107729274e-05,
|
974 |
+
"loss": 0.7203,
|
975 |
+
"step": 1280
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 22.02,
|
979 |
+
"eval_accuracy": 0.5990783410138248,
|
980 |
+
"eval_loss": 0.9694386720657349,
|
981 |
+
"eval_runtime": 169.6323,
|
982 |
+
"eval_samples_per_second": 1.279,
|
983 |
+
"eval_steps_per_second": 0.165,
|
984 |
+
"step": 1288
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 23.0,
|
988 |
+
"learning_rate": 2.973568281938326e-05,
|
989 |
+
"loss": 0.605,
|
990 |
+
"step": 1290
|
991 |
+
},
|
992 |
+
{
|
993 |
+
"epoch": 23.0,
|
994 |
+
"learning_rate": 2.9535442531037244e-05,
|
995 |
+
"loss": 0.6581,
|
996 |
+
"step": 1300
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 23.01,
|
1000 |
+
"learning_rate": 2.9335202242691228e-05,
|
1001 |
+
"loss": 0.5423,
|
1002 |
+
"step": 1310
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 23.01,
|
1006 |
+
"learning_rate": 2.9134961954345213e-05,
|
1007 |
+
"loss": 0.6137,
|
1008 |
+
"step": 1320
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 23.02,
|
1012 |
+
"learning_rate": 2.8934721665999198e-05,
|
1013 |
+
"loss": 0.562,
|
1014 |
+
"step": 1330
|
1015 |
+
},
|
1016 |
+
{
|
1017 |
+
"epoch": 23.02,
|
1018 |
+
"learning_rate": 2.8734481377653182e-05,
|
1019 |
+
"loss": 0.6799,
|
1020 |
+
"step": 1340
|
1021 |
+
},
|
1022 |
+
{
|
1023 |
+
"epoch": 23.02,
|
1024 |
+
"eval_accuracy": 0.7695852534562212,
|
1025 |
+
"eval_loss": 0.647365152835846,
|
1026 |
+
"eval_runtime": 168.7434,
|
1027 |
+
"eval_samples_per_second": 1.286,
|
1028 |
+
"eval_steps_per_second": 0.166,
|
1029 |
+
"step": 1344
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 24.0,
|
1033 |
+
"learning_rate": 2.8534241089307167e-05,
|
1034 |
+
"loss": 0.5556,
|
1035 |
+
"step": 1350
|
1036 |
+
},
|
1037 |
+
{
|
1038 |
+
"epoch": 24.01,
|
1039 |
+
"learning_rate": 2.833400080096115e-05,
|
1040 |
+
"loss": 0.7249,
|
1041 |
+
"step": 1360
|
1042 |
+
},
|
1043 |
+
{
|
1044 |
+
"epoch": 24.01,
|
1045 |
+
"learning_rate": 2.8133760512615136e-05,
|
1046 |
+
"loss": 0.563,
|
1047 |
+
"step": 1370
|
1048 |
+
},
|
1049 |
+
{
|
1050 |
+
"epoch": 24.01,
|
1051 |
+
"learning_rate": 2.793352022426912e-05,
|
1052 |
+
"loss": 0.6947,
|
1053 |
+
"step": 1380
|
1054 |
+
},
|
1055 |
+
{
|
1056 |
+
"epoch": 24.02,
|
1057 |
+
"learning_rate": 2.7733279935923106e-05,
|
1058 |
+
"loss": 0.8462,
|
1059 |
+
"step": 1390
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 24.02,
|
1063 |
+
"learning_rate": 2.753303964757709e-05,
|
1064 |
+
"loss": 0.5802,
|
1065 |
+
"step": 1400
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 24.02,
|
1069 |
+
"eval_accuracy": 0.6359447004608295,
|
1070 |
+
"eval_loss": 0.957251787185669,
|
1071 |
+
"eval_runtime": 168.4758,
|
1072 |
+
"eval_samples_per_second": 1.288,
|
1073 |
+
"eval_steps_per_second": 0.166,
|
1074 |
+
"step": 1400
|
1075 |
+
},
|
1076 |
+
{
|
1077 |
+
"epoch": 25.0,
|
1078 |
+
"learning_rate": 2.7332799359231075e-05,
|
1079 |
+
"loss": 0.5186,
|
1080 |
+
"step": 1410
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 25.01,
|
1084 |
+
"learning_rate": 2.713255907088506e-05,
|
1085 |
+
"loss": 0.5462,
|
1086 |
+
"step": 1420
|
1087 |
+
},
|
1088 |
+
{
|
1089 |
+
"epoch": 25.01,
|
1090 |
+
"learning_rate": 2.6932318782539044e-05,
|
1091 |
+
"loss": 0.6317,
|
1092 |
+
"step": 1430
|
1093 |
+
},
|
1094 |
+
{
|
1095 |
+
"epoch": 25.01,
|
1096 |
+
"learning_rate": 2.673207849419303e-05,
|
1097 |
+
"loss": 0.5265,
|
1098 |
+
"step": 1440
|
1099 |
+
},
|
1100 |
+
{
|
1101 |
+
"epoch": 25.02,
|
1102 |
+
"learning_rate": 2.6531838205847014e-05,
|
1103 |
+
"loss": 0.7047,
|
1104 |
+
"step": 1450
|
1105 |
+
},
|
1106 |
+
{
|
1107 |
+
"epoch": 25.02,
|
1108 |
+
"eval_accuracy": 0.695852534562212,
|
1109 |
+
"eval_loss": 0.9120410680770874,
|
1110 |
+
"eval_runtime": 168.5578,
|
1111 |
+
"eval_samples_per_second": 1.287,
|
1112 |
+
"eval_steps_per_second": 0.166,
|
1113 |
+
"step": 1456
|
1114 |
+
},
|
1115 |
+
{
|
1116 |
+
"epoch": 26.0,
|
1117 |
+
"learning_rate": 2.6331597917501e-05,
|
1118 |
+
"loss": 0.5364,
|
1119 |
+
"step": 1460
|
1120 |
+
},
|
1121 |
+
{
|
1122 |
+
"epoch": 26.01,
|
1123 |
+
"learning_rate": 2.613135762915499e-05,
|
1124 |
+
"loss": 0.5322,
|
1125 |
+
"step": 1470
|
1126 |
+
},
|
1127 |
+
{
|
1128 |
+
"epoch": 26.01,
|
1129 |
+
"learning_rate": 2.5931117340808974e-05,
|
1130 |
+
"loss": 0.897,
|
1131 |
+
"step": 1480
|
1132 |
+
},
|
1133 |
+
{
|
1134 |
+
"epoch": 26.01,
|
1135 |
+
"learning_rate": 2.573087705246296e-05,
|
1136 |
+
"loss": 0.4983,
|
1137 |
+
"step": 1490
|
1138 |
+
},
|
1139 |
+
{
|
1140 |
+
"epoch": 26.02,
|
1141 |
+
"learning_rate": 2.5530636764116944e-05,
|
1142 |
+
"loss": 0.7582,
|
1143 |
+
"step": 1500
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 26.02,
|
1147 |
+
"learning_rate": 2.533039647577093e-05,
|
1148 |
+
"loss": 0.6701,
|
1149 |
+
"step": 1510
|
1150 |
+
},
|
1151 |
+
{
|
1152 |
+
"epoch": 26.02,
|
1153 |
+
"eval_accuracy": 0.5852534562211982,
|
1154 |
+
"eval_loss": 1.1690046787261963,
|
1155 |
+
"eval_runtime": 169.2732,
|
1156 |
+
"eval_samples_per_second": 1.282,
|
1157 |
+
"eval_steps_per_second": 0.165,
|
1158 |
+
"step": 1512
|
1159 |
+
},
|
1160 |
+
{
|
1161 |
+
"epoch": 27.0,
|
1162 |
+
"learning_rate": 2.5130156187424913e-05,
|
1163 |
+
"loss": 0.4735,
|
1164 |
+
"step": 1520
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 27.01,
|
1168 |
+
"learning_rate": 2.4929915899078894e-05,
|
1169 |
+
"loss": 0.6383,
|
1170 |
+
"step": 1530
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 27.01,
|
1174 |
+
"learning_rate": 2.472967561073288e-05,
|
1175 |
+
"loss": 0.7644,
|
1176 |
+
"step": 1540
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 27.01,
|
1180 |
+
"learning_rate": 2.4529435322386864e-05,
|
1181 |
+
"loss": 0.6395,
|
1182 |
+
"step": 1550
|
1183 |
+
},
|
1184 |
+
{
|
1185 |
+
"epoch": 27.02,
|
1186 |
+
"learning_rate": 2.432919503404085e-05,
|
1187 |
+
"loss": 0.5514,
|
1188 |
+
"step": 1560
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"epoch": 27.02,
|
1192 |
+
"eval_accuracy": 0.6866359447004609,
|
1193 |
+
"eval_loss": 0.9173929691314697,
|
1194 |
+
"eval_runtime": 169.8137,
|
1195 |
+
"eval_samples_per_second": 1.278,
|
1196 |
+
"eval_steps_per_second": 0.165,
|
1197 |
+
"step": 1568
|
1198 |
+
},
|
1199 |
+
{
|
1200 |
+
"epoch": 28.0,
|
1201 |
+
"learning_rate": 2.4128954745694833e-05,
|
1202 |
+
"loss": 0.5936,
|
1203 |
+
"step": 1570
|
1204 |
+
},
|
1205 |
+
{
|
1206 |
+
"epoch": 28.0,
|
1207 |
+
"learning_rate": 2.3928714457348818e-05,
|
1208 |
+
"loss": 0.4936,
|
1209 |
+
"step": 1580
|
1210 |
+
},
|
1211 |
+
{
|
1212 |
+
"epoch": 28.01,
|
1213 |
+
"learning_rate": 2.3728474169002803e-05,
|
1214 |
+
"loss": 0.6134,
|
1215 |
+
"step": 1590
|
1216 |
+
},
|
1217 |
+
{
|
1218 |
+
"epoch": 28.01,
|
1219 |
+
"learning_rate": 2.3528233880656787e-05,
|
1220 |
+
"loss": 0.654,
|
1221 |
+
"step": 1600
|
1222 |
+
},
|
1223 |
+
{
|
1224 |
+
"epoch": 28.02,
|
1225 |
+
"learning_rate": 2.3327993592310772e-05,
|
1226 |
+
"loss": 0.6997,
|
1227 |
+
"step": 1610
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 28.02,
|
1231 |
+
"learning_rate": 2.3127753303964757e-05,
|
1232 |
+
"loss": 0.538,
|
1233 |
+
"step": 1620
|
1234 |
+
},
|
1235 |
+
{
|
1236 |
+
"epoch": 28.02,
|
1237 |
+
"eval_accuracy": 0.6866359447004609,
|
1238 |
+
"eval_loss": 0.854290246963501,
|
1239 |
+
"eval_runtime": 171.1701,
|
1240 |
+
"eval_samples_per_second": 1.268,
|
1241 |
+
"eval_steps_per_second": 0.164,
|
1242 |
+
"step": 1624
|
1243 |
+
},
|
1244 |
+
{
|
1245 |
+
"epoch": 29.0,
|
1246 |
+
"learning_rate": 2.292751301561874e-05,
|
1247 |
+
"loss": 0.6431,
|
1248 |
+
"step": 1630
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 29.01,
|
1252 |
+
"learning_rate": 2.272727272727273e-05,
|
1253 |
+
"loss": 0.6269,
|
1254 |
+
"step": 1640
|
1255 |
+
},
|
1256 |
+
{
|
1257 |
+
"epoch": 29.01,
|
1258 |
+
"learning_rate": 2.2527032438926714e-05,
|
1259 |
+
"loss": 0.6239,
|
1260 |
+
"step": 1650
|
1261 |
+
},
|
1262 |
+
{
|
1263 |
+
"epoch": 29.01,
|
1264 |
+
"learning_rate": 2.23267921505807e-05,
|
1265 |
+
"loss": 0.5901,
|
1266 |
+
"step": 1660
|
1267 |
+
},
|
1268 |
+
{
|
1269 |
+
"epoch": 29.02,
|
1270 |
+
"learning_rate": 2.2126551862234683e-05,
|
1271 |
+
"loss": 0.5673,
|
1272 |
+
"step": 1670
|
1273 |
+
},
|
1274 |
+
{
|
1275 |
+
"epoch": 29.02,
|
1276 |
+
"learning_rate": 2.1926311573888668e-05,
|
1277 |
+
"loss": 0.7226,
|
1278 |
+
"step": 1680
|
1279 |
+
},
|
1280 |
+
{
|
1281 |
+
"epoch": 29.02,
|
1282 |
+
"eval_accuracy": 0.7465437788018433,
|
1283 |
+
"eval_loss": 0.7773527503013611,
|
1284 |
+
"eval_runtime": 170.9184,
|
1285 |
+
"eval_samples_per_second": 1.27,
|
1286 |
+
"eval_steps_per_second": 0.164,
|
1287 |
+
"step": 1680
|
1288 |
+
},
|
1289 |
+
{
|
1290 |
+
"epoch": 30.0,
|
1291 |
+
"learning_rate": 2.1726071285542653e-05,
|
1292 |
+
"loss": 0.5368,
|
1293 |
+
"step": 1690
|
1294 |
+
},
|
1295 |
+
{
|
1296 |
+
"epoch": 30.01,
|
1297 |
+
"learning_rate": 2.1525830997196637e-05,
|
1298 |
+
"loss": 0.6407,
|
1299 |
+
"step": 1700
|
1300 |
+
},
|
1301 |
+
{
|
1302 |
+
"epoch": 30.01,
|
1303 |
+
"learning_rate": 2.1325590708850622e-05,
|
1304 |
+
"loss": 0.6071,
|
1305 |
+
"step": 1710
|
1306 |
+
},
|
1307 |
+
{
|
1308 |
+
"epoch": 30.01,
|
1309 |
+
"learning_rate": 2.1125350420504607e-05,
|
1310 |
+
"loss": 0.587,
|
1311 |
+
"step": 1720
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 30.02,
|
1315 |
+
"learning_rate": 2.092511013215859e-05,
|
1316 |
+
"loss": 0.4459,
|
1317 |
+
"step": 1730
|
1318 |
+
},
|
1319 |
+
{
|
1320 |
+
"epoch": 30.02,
|
1321 |
+
"eval_accuracy": 0.6359447004608295,
|
1322 |
+
"eval_loss": 0.9134915471076965,
|
1323 |
+
"eval_runtime": 169.6863,
|
1324 |
+
"eval_samples_per_second": 1.279,
|
1325 |
+
"eval_steps_per_second": 0.165,
|
1326 |
+
"step": 1736
|
1327 |
+
},
|
1328 |
+
{
|
1329 |
+
"epoch": 31.0,
|
1330 |
+
"learning_rate": 2.0724869843812576e-05,
|
1331 |
+
"loss": 0.6281,
|
1332 |
+
"step": 1740
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 31.01,
|
1336 |
+
"learning_rate": 2.052462955546656e-05,
|
1337 |
+
"loss": 0.5999,
|
1338 |
+
"step": 1750
|
1339 |
+
},
|
1340 |
+
{
|
1341 |
+
"epoch": 31.01,
|
1342 |
+
"learning_rate": 2.0324389267120545e-05,
|
1343 |
+
"loss": 0.4052,
|
1344 |
+
"step": 1760
|
1345 |
+
},
|
1346 |
+
{
|
1347 |
+
"epoch": 31.01,
|
1348 |
+
"learning_rate": 2.012414897877453e-05,
|
1349 |
+
"loss": 0.7613,
|
1350 |
+
"step": 1770
|
1351 |
+
},
|
1352 |
+
{
|
1353 |
+
"epoch": 31.02,
|
1354 |
+
"learning_rate": 1.9923908690428515e-05,
|
1355 |
+
"loss": 0.6318,
|
1356 |
+
"step": 1780
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 31.02,
|
1360 |
+
"learning_rate": 1.9723668402082503e-05,
|
1361 |
+
"loss": 0.3905,
|
1362 |
+
"step": 1790
|
1363 |
+
},
|
1364 |
+
{
|
1365 |
+
"epoch": 31.02,
|
1366 |
+
"eval_accuracy": 0.6728110599078341,
|
1367 |
+
"eval_loss": 0.8585994243621826,
|
1368 |
+
"eval_runtime": 165.9142,
|
1369 |
+
"eval_samples_per_second": 1.308,
|
1370 |
+
"eval_steps_per_second": 0.169,
|
1371 |
+
"step": 1792
|
1372 |
+
},
|
1373 |
+
{
|
1374 |
+
"epoch": 32.0,
|
1375 |
+
"learning_rate": 1.9523428113736487e-05,
|
1376 |
+
"loss": 0.5,
|
1377 |
+
"step": 1800
|
1378 |
+
},
|
1379 |
+
{
|
1380 |
+
"epoch": 32.01,
|
1381 |
+
"learning_rate": 1.9323187825390472e-05,
|
1382 |
+
"loss": 0.4816,
|
1383 |
+
"step": 1810
|
1384 |
+
},
|
1385 |
+
{
|
1386 |
+
"epoch": 32.01,
|
1387 |
+
"learning_rate": 1.9122947537044457e-05,
|
1388 |
+
"loss": 0.4979,
|
1389 |
+
"step": 1820
|
1390 |
+
},
|
1391 |
+
{
|
1392 |
+
"epoch": 32.01,
|
1393 |
+
"learning_rate": 1.892270724869844e-05,
|
1394 |
+
"loss": 0.6203,
|
1395 |
+
"step": 1830
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 32.02,
|
1399 |
+
"learning_rate": 1.8722466960352426e-05,
|
1400 |
+
"loss": 0.7071,
|
1401 |
+
"step": 1840
|
1402 |
+
},
|
1403 |
+
{
|
1404 |
+
"epoch": 32.02,
|
1405 |
+
"eval_accuracy": 0.7327188940092166,
|
1406 |
+
"eval_loss": 0.7919384837150574,
|
1407 |
+
"eval_runtime": 171.3033,
|
1408 |
+
"eval_samples_per_second": 1.267,
|
1409 |
+
"eval_steps_per_second": 0.163,
|
1410 |
+
"step": 1848
|
1411 |
+
},
|
1412 |
+
{
|
1413 |
+
"epoch": 33.0,
|
1414 |
+
"learning_rate": 1.852222667200641e-05,
|
1415 |
+
"loss": 0.5632,
|
1416 |
+
"step": 1850
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 33.0,
|
1420 |
+
"learning_rate": 1.8321986383660396e-05,
|
1421 |
+
"loss": 0.6323,
|
1422 |
+
"step": 1860
|
1423 |
+
},
|
1424 |
+
{
|
1425 |
+
"epoch": 33.01,
|
1426 |
+
"learning_rate": 1.812174609531438e-05,
|
1427 |
+
"loss": 0.6151,
|
1428 |
+
"step": 1870
|
1429 |
+
},
|
1430 |
+
{
|
1431 |
+
"epoch": 33.01,
|
1432 |
+
"learning_rate": 1.7921505806968365e-05,
|
1433 |
+
"loss": 0.4412,
|
1434 |
+
"step": 1880
|
1435 |
+
},
|
1436 |
+
{
|
1437 |
+
"epoch": 33.02,
|
1438 |
+
"learning_rate": 1.772126551862235e-05,
|
1439 |
+
"loss": 0.4863,
|
1440 |
+
"step": 1890
|
1441 |
+
},
|
1442 |
+
{
|
1443 |
+
"epoch": 33.02,
|
1444 |
+
"learning_rate": 1.7521025230276334e-05,
|
1445 |
+
"loss": 0.4983,
|
1446 |
+
"step": 1900
|
1447 |
+
},
|
1448 |
+
{
|
1449 |
+
"epoch": 33.02,
|
1450 |
+
"eval_accuracy": 0.7511520737327189,
|
1451 |
+
"eval_loss": 0.7506622672080994,
|
1452 |
+
"eval_runtime": 169.3822,
|
1453 |
+
"eval_samples_per_second": 1.281,
|
1454 |
+
"eval_steps_per_second": 0.165,
|
1455 |
+
"step": 1904
|
1456 |
+
},
|
1457 |
+
{
|
1458 |
+
"epoch": 34.0,
|
1459 |
+
"learning_rate": 1.732078494193032e-05,
|
1460 |
+
"loss": 0.4416,
|
1461 |
+
"step": 1910
|
1462 |
+
},
|
1463 |
+
{
|
1464 |
+
"epoch": 34.01,
|
1465 |
+
"learning_rate": 1.7120544653584304e-05,
|
1466 |
+
"loss": 0.5567,
|
1467 |
+
"step": 1920
|
1468 |
+
},
|
1469 |
+
{
|
1470 |
+
"epoch": 34.01,
|
1471 |
+
"learning_rate": 1.6920304365238288e-05,
|
1472 |
+
"loss": 0.3794,
|
1473 |
+
"step": 1930
|
1474 |
+
},
|
1475 |
+
{
|
1476 |
+
"epoch": 34.01,
|
1477 |
+
"learning_rate": 1.6720064076892273e-05,
|
1478 |
+
"loss": 0.5672,
|
1479 |
+
"step": 1940
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 34.02,
|
1483 |
+
"learning_rate": 1.6519823788546258e-05,
|
1484 |
+
"loss": 0.4093,
|
1485 |
+
"step": 1950
|
1486 |
+
},
|
1487 |
+
{
|
1488 |
+
"epoch": 34.02,
|
1489 |
+
"learning_rate": 1.6319583500200242e-05,
|
1490 |
+
"loss": 0.5654,
|
1491 |
+
"step": 1960
|
1492 |
+
},
|
1493 |
+
{
|
1494 |
+
"epoch": 34.02,
|
1495 |
+
"eval_accuracy": 0.7142857142857143,
|
1496 |
+
"eval_loss": 0.767917275428772,
|
1497 |
+
"eval_runtime": 173.7976,
|
1498 |
+
"eval_samples_per_second": 1.249,
|
1499 |
+
"eval_steps_per_second": 0.161,
|
1500 |
+
"step": 1960
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 35.0,
|
1504 |
+
"learning_rate": 1.6119343211854227e-05,
|
1505 |
+
"loss": 0.3909,
|
1506 |
+
"step": 1970
|
1507 |
+
},
|
1508 |
+
{
|
1509 |
+
"epoch": 35.01,
|
1510 |
+
"learning_rate": 1.591910292350821e-05,
|
1511 |
+
"loss": 0.5576,
|
1512 |
+
"step": 1980
|
1513 |
+
},
|
1514 |
+
{
|
1515 |
+
"epoch": 35.01,
|
1516 |
+
"learning_rate": 1.5718862635162196e-05,
|
1517 |
+
"loss": 0.4247,
|
1518 |
+
"step": 1990
|
1519 |
+
},
|
1520 |
+
{
|
1521 |
+
"epoch": 35.01,
|
1522 |
+
"learning_rate": 1.551862234681618e-05,
|
1523 |
+
"loss": 0.747,
|
1524 |
+
"step": 2000
|
1525 |
+
},
|
1526 |
+
{
|
1527 |
+
"epoch": 35.02,
|
1528 |
+
"learning_rate": 1.5318382058470166e-05,
|
1529 |
+
"loss": 0.5569,
|
1530 |
+
"step": 2010
|
1531 |
+
},
|
1532 |
+
{
|
1533 |
+
"epoch": 35.02,
|
1534 |
+
"eval_accuracy": 0.7096774193548387,
|
1535 |
+
"eval_loss": 0.8438030481338501,
|
1536 |
+
"eval_runtime": 168.1341,
|
1537 |
+
"eval_samples_per_second": 1.291,
|
1538 |
+
"eval_steps_per_second": 0.167,
|
1539 |
+
"step": 2016
|
1540 |
+
},
|
1541 |
+
{
|
1542 |
+
"epoch": 36.0,
|
1543 |
+
"learning_rate": 1.511814177012415e-05,
|
1544 |
+
"loss": 0.5796,
|
1545 |
+
"step": 2020
|
1546 |
+
},
|
1547 |
+
{
|
1548 |
+
"epoch": 36.01,
|
1549 |
+
"learning_rate": 1.4917901481778135e-05,
|
1550 |
+
"loss": 0.6004,
|
1551 |
+
"step": 2030
|
1552 |
+
},
|
1553 |
+
{
|
1554 |
+
"epoch": 36.01,
|
1555 |
+
"learning_rate": 1.471766119343212e-05,
|
1556 |
+
"loss": 0.4482,
|
1557 |
+
"step": 2040
|
1558 |
+
},
|
1559 |
+
{
|
1560 |
+
"epoch": 36.01,
|
1561 |
+
"learning_rate": 1.4517420905086104e-05,
|
1562 |
+
"loss": 0.593,
|
1563 |
+
"step": 2050
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 36.02,
|
1567 |
+
"learning_rate": 1.4317180616740089e-05,
|
1568 |
+
"loss": 0.6642,
|
1569 |
+
"step": 2060
|
1570 |
+
},
|
1571 |
+
{
|
1572 |
+
"epoch": 36.02,
|
1573 |
+
"learning_rate": 1.4116940328394074e-05,
|
1574 |
+
"loss": 0.3998,
|
1575 |
+
"step": 2070
|
1576 |
+
},
|
1577 |
+
{
|
1578 |
+
"epoch": 36.02,
|
1579 |
+
"eval_accuracy": 0.7188940092165899,
|
1580 |
+
"eval_loss": 0.8691464066505432,
|
1581 |
+
"eval_runtime": 171.2911,
|
1582 |
+
"eval_samples_per_second": 1.267,
|
1583 |
+
"eval_steps_per_second": 0.163,
|
1584 |
+
"step": 2072
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 37.0,
|
1588 |
+
"learning_rate": 1.3916700040048058e-05,
|
1589 |
+
"loss": 0.3557,
|
1590 |
+
"step": 2080
|
1591 |
+
},
|
1592 |
+
{
|
1593 |
+
"epoch": 37.01,
|
1594 |
+
"learning_rate": 1.3716459751702043e-05,
|
1595 |
+
"loss": 0.3628,
|
1596 |
+
"step": 2090
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 37.01,
|
1600 |
+
"learning_rate": 1.3516219463356028e-05,
|
1601 |
+
"loss": 0.335,
|
1602 |
+
"step": 2100
|
1603 |
+
},
|
1604 |
+
{
|
1605 |
+
"epoch": 37.01,
|
1606 |
+
"learning_rate": 1.3315979175010012e-05,
|
1607 |
+
"loss": 0.6813,
|
1608 |
+
"step": 2110
|
1609 |
+
},
|
1610 |
+
{
|
1611 |
+
"epoch": 37.02,
|
1612 |
+
"learning_rate": 1.3115738886663997e-05,
|
1613 |
+
"loss": 0.5341,
|
1614 |
+
"step": 2120
|
1615 |
+
},
|
1616 |
+
{
|
1617 |
+
"epoch": 37.02,
|
1618 |
+
"eval_accuracy": 0.7603686635944701,
|
1619 |
+
"eval_loss": 0.8056049346923828,
|
1620 |
+
"eval_runtime": 172.8523,
|
1621 |
+
"eval_samples_per_second": 1.255,
|
1622 |
+
"eval_steps_per_second": 0.162,
|
1623 |
+
"step": 2128
|
1624 |
+
},
|
1625 |
+
{
|
1626 |
+
"epoch": 38.0,
|
1627 |
+
"learning_rate": 1.2915498598317982e-05,
|
1628 |
+
"loss": 0.4075,
|
1629 |
+
"step": 2130
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 38.0,
|
1633 |
+
"learning_rate": 1.2715258309971966e-05,
|
1634 |
+
"loss": 0.5339,
|
1635 |
+
"step": 2140
|
1636 |
+
},
|
1637 |
+
{
|
1638 |
+
"epoch": 38.01,
|
1639 |
+
"learning_rate": 1.2515018021625951e-05,
|
1640 |
+
"loss": 0.4681,
|
1641 |
+
"step": 2150
|
1642 |
+
},
|
1643 |
+
{
|
1644 |
+
"epoch": 38.01,
|
1645 |
+
"learning_rate": 1.2314777733279936e-05,
|
1646 |
+
"loss": 0.4725,
|
1647 |
+
"step": 2160
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 38.02,
|
1651 |
+
"learning_rate": 1.211453744493392e-05,
|
1652 |
+
"loss": 0.5588,
|
1653 |
+
"step": 2170
|
1654 |
+
},
|
1655 |
+
{
|
1656 |
+
"epoch": 38.02,
|
1657 |
+
"learning_rate": 1.1914297156587905e-05,
|
1658 |
+
"loss": 0.4024,
|
1659 |
+
"step": 2180
|
1660 |
+
},
|
1661 |
+
{
|
1662 |
+
"epoch": 38.02,
|
1663 |
+
"eval_accuracy": 0.7880184331797235,
|
1664 |
+
"eval_loss": 0.7070650458335876,
|
1665 |
+
"eval_runtime": 170.228,
|
1666 |
+
"eval_samples_per_second": 1.275,
|
1667 |
+
"eval_steps_per_second": 0.164,
|
1668 |
+
"step": 2184
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 39.0,
|
1672 |
+
"learning_rate": 1.171405686824189e-05,
|
1673 |
+
"loss": 0.4122,
|
1674 |
+
"step": 2190
|
1675 |
+
},
|
1676 |
+
{
|
1677 |
+
"epoch": 39.01,
|
1678 |
+
"learning_rate": 1.1513816579895874e-05,
|
1679 |
+
"loss": 0.3389,
|
1680 |
+
"step": 2200
|
1681 |
+
},
|
1682 |
+
{
|
1683 |
+
"epoch": 39.01,
|
1684 |
+
"learning_rate": 1.131357629154986e-05,
|
1685 |
+
"loss": 0.4993,
|
1686 |
+
"step": 2210
|
1687 |
+
},
|
1688 |
+
{
|
1689 |
+
"epoch": 39.01,
|
1690 |
+
"learning_rate": 1.1113336003203845e-05,
|
1691 |
+
"loss": 0.5869,
|
1692 |
+
"step": 2220
|
1693 |
+
},
|
1694 |
+
{
|
1695 |
+
"epoch": 39.02,
|
1696 |
+
"learning_rate": 1.091309571485783e-05,
|
1697 |
+
"loss": 0.7077,
|
1698 |
+
"step": 2230
|
1699 |
+
},
|
1700 |
+
{
|
1701 |
+
"epoch": 39.02,
|
1702 |
+
"learning_rate": 1.0712855426511815e-05,
|
1703 |
+
"loss": 0.5011,
|
1704 |
+
"step": 2240
|
1705 |
+
},
|
1706 |
+
{
|
1707 |
+
"epoch": 39.02,
|
1708 |
+
"eval_accuracy": 0.7004608294930875,
|
1709 |
+
"eval_loss": 0.882685661315918,
|
1710 |
+
"eval_runtime": 169.5666,
|
1711 |
+
"eval_samples_per_second": 1.28,
|
1712 |
+
"eval_steps_per_second": 0.165,
|
1713 |
+
"step": 2240
|
1714 |
+
},
|
1715 |
+
{
|
1716 |
+
"epoch": 40.0,
|
1717 |
+
"learning_rate": 1.05126151381658e-05,
|
1718 |
+
"loss": 0.443,
|
1719 |
+
"step": 2250
|
1720 |
+
},
|
1721 |
+
{
|
1722 |
+
"epoch": 40.01,
|
1723 |
+
"learning_rate": 1.0312374849819784e-05,
|
1724 |
+
"loss": 0.5292,
|
1725 |
+
"step": 2260
|
1726 |
+
},
|
1727 |
+
{
|
1728 |
+
"epoch": 40.01,
|
1729 |
+
"learning_rate": 1.0112134561473769e-05,
|
1730 |
+
"loss": 0.4058,
|
1731 |
+
"step": 2270
|
1732 |
+
},
|
1733 |
+
{
|
1734 |
+
"epoch": 40.01,
|
1735 |
+
"learning_rate": 9.911894273127754e-06,
|
1736 |
+
"loss": 0.5513,
|
1737 |
+
"step": 2280
|
1738 |
+
},
|
1739 |
+
{
|
1740 |
+
"epoch": 40.02,
|
1741 |
+
"learning_rate": 9.71165398478174e-06,
|
1742 |
+
"loss": 0.5857,
|
1743 |
+
"step": 2290
|
1744 |
+
},
|
1745 |
+
{
|
1746 |
+
"epoch": 40.02,
|
1747 |
+
"eval_accuracy": 0.7096774193548387,
|
1748 |
+
"eval_loss": 0.8525260090827942,
|
1749 |
+
"eval_runtime": 169.961,
|
1750 |
+
"eval_samples_per_second": 1.277,
|
1751 |
+
"eval_steps_per_second": 0.165,
|
1752 |
+
"step": 2296
|
1753 |
+
},
|
1754 |
+
{
|
1755 |
+
"epoch": 41.0,
|
1756 |
+
"learning_rate": 9.511413696435725e-06,
|
1757 |
+
"loss": 0.6831,
|
1758 |
+
"step": 2300
|
1759 |
+
},
|
1760 |
+
{
|
1761 |
+
"epoch": 41.01,
|
1762 |
+
"learning_rate": 9.31117340808971e-06,
|
1763 |
+
"loss": 0.4212,
|
1764 |
+
"step": 2310
|
1765 |
+
},
|
1766 |
+
{
|
1767 |
+
"epoch": 41.01,
|
1768 |
+
"learning_rate": 9.110933119743694e-06,
|
1769 |
+
"loss": 0.4343,
|
1770 |
+
"step": 2320
|
1771 |
+
},
|
1772 |
+
{
|
1773 |
+
"epoch": 41.01,
|
1774 |
+
"learning_rate": 8.910692831397679e-06,
|
1775 |
+
"loss": 0.3912,
|
1776 |
+
"step": 2330
|
1777 |
+
},
|
1778 |
+
{
|
1779 |
+
"epoch": 41.02,
|
1780 |
+
"learning_rate": 8.710452543051663e-06,
|
1781 |
+
"loss": 0.5539,
|
1782 |
+
"step": 2340
|
1783 |
+
},
|
1784 |
+
{
|
1785 |
+
"epoch": 41.02,
|
1786 |
+
"learning_rate": 8.510212254705648e-06,
|
1787 |
+
"loss": 0.5619,
|
1788 |
+
"step": 2350
|
1789 |
+
},
|
1790 |
+
{
|
1791 |
+
"epoch": 41.02,
|
1792 |
+
"eval_accuracy": 0.7511520737327189,
|
1793 |
+
"eval_loss": 0.8228119611740112,
|
1794 |
+
"eval_runtime": 167.3356,
|
1795 |
+
"eval_samples_per_second": 1.297,
|
1796 |
+
"eval_steps_per_second": 0.167,
|
1797 |
+
"step": 2352
|
1798 |
+
},
|
1799 |
+
{
|
1800 |
+
"epoch": 42.0,
|
1801 |
+
"learning_rate": 8.309971966359633e-06,
|
1802 |
+
"loss": 0.3721,
|
1803 |
+
"step": 2360
|
1804 |
+
},
|
1805 |
+
{
|
1806 |
+
"epoch": 42.01,
|
1807 |
+
"learning_rate": 8.109731678013617e-06,
|
1808 |
+
"loss": 0.332,
|
1809 |
+
"step": 2370
|
1810 |
+
},
|
1811 |
+
{
|
1812 |
+
"epoch": 42.01,
|
1813 |
+
"learning_rate": 7.909491389667602e-06,
|
1814 |
+
"loss": 0.5869,
|
1815 |
+
"step": 2380
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 42.01,
|
1819 |
+
"learning_rate": 7.709251101321587e-06,
|
1820 |
+
"loss": 0.4474,
|
1821 |
+
"step": 2390
|
1822 |
+
},
|
1823 |
+
{
|
1824 |
+
"epoch": 42.02,
|
1825 |
+
"learning_rate": 7.509010812975571e-06,
|
1826 |
+
"loss": 0.6052,
|
1827 |
+
"step": 2400
|
1828 |
+
},
|
1829 |
+
{
|
1830 |
+
"epoch": 42.02,
|
1831 |
+
"eval_accuracy": 0.7373271889400922,
|
1832 |
+
"eval_loss": 0.8320425152778625,
|
1833 |
+
"eval_runtime": 168.2726,
|
1834 |
+
"eval_samples_per_second": 1.29,
|
1835 |
+
"eval_steps_per_second": 0.166,
|
1836 |
+
"step": 2408
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 43.0,
|
1840 |
+
"learning_rate": 7.308770524629556e-06,
|
1841 |
+
"loss": 0.451,
|
1842 |
+
"step": 2410
|
1843 |
+
},
|
1844 |
+
{
|
1845 |
+
"epoch": 43.0,
|
1846 |
+
"learning_rate": 7.108530236283541e-06,
|
1847 |
+
"loss": 0.282,
|
1848 |
+
"step": 2420
|
1849 |
+
},
|
1850 |
+
{
|
1851 |
+
"epoch": 43.01,
|
1852 |
+
"learning_rate": 6.908289947937525e-06,
|
1853 |
+
"loss": 0.4565,
|
1854 |
+
"step": 2430
|
1855 |
+
},
|
1856 |
+
{
|
1857 |
+
"epoch": 43.01,
|
1858 |
+
"learning_rate": 6.70804965959151e-06,
|
1859 |
+
"loss": 0.6579,
|
1860 |
+
"step": 2440
|
1861 |
+
},
|
1862 |
+
{
|
1863 |
+
"epoch": 43.02,
|
1864 |
+
"learning_rate": 6.507809371245495e-06,
|
1865 |
+
"loss": 0.3812,
|
1866 |
+
"step": 2450
|
1867 |
+
},
|
1868 |
+
{
|
1869 |
+
"epoch": 43.02,
|
1870 |
+
"learning_rate": 6.307569082899479e-06,
|
1871 |
+
"loss": 0.5124,
|
1872 |
+
"step": 2460
|
1873 |
+
},
|
1874 |
+
{
|
1875 |
+
"epoch": 43.02,
|
1876 |
+
"eval_accuracy": 0.7419354838709677,
|
1877 |
+
"eval_loss": 0.8776273131370544,
|
1878 |
+
"eval_runtime": 167.8781,
|
1879 |
+
"eval_samples_per_second": 1.293,
|
1880 |
+
"eval_steps_per_second": 0.167,
|
1881 |
+
"step": 2464
|
1882 |
+
},
|
1883 |
+
{
|
1884 |
+
"epoch": 44.0,
|
1885 |
+
"learning_rate": 6.107328794553464e-06,
|
1886 |
+
"loss": 0.2933,
|
1887 |
+
"step": 2470
|
1888 |
+
},
|
1889 |
+
{
|
1890 |
+
"epoch": 44.01,
|
1891 |
+
"learning_rate": 5.907088506207449e-06,
|
1892 |
+
"loss": 0.3212,
|
1893 |
+
"step": 2480
|
1894 |
+
},
|
1895 |
+
{
|
1896 |
+
"epoch": 44.01,
|
1897 |
+
"learning_rate": 5.706848217861433e-06,
|
1898 |
+
"loss": 0.47,
|
1899 |
+
"step": 2490
|
1900 |
+
},
|
1901 |
+
{
|
1902 |
+
"epoch": 44.01,
|
1903 |
+
"learning_rate": 5.506607929515419e-06,
|
1904 |
+
"loss": 0.3473,
|
1905 |
+
"step": 2500
|
1906 |
+
},
|
1907 |
+
{
|
1908 |
+
"epoch": 44.02,
|
1909 |
+
"learning_rate": 5.306367641169404e-06,
|
1910 |
+
"loss": 0.5966,
|
1911 |
+
"step": 2510
|
1912 |
+
},
|
1913 |
+
{
|
1914 |
+
"epoch": 44.02,
|
1915 |
+
"learning_rate": 5.106127352823388e-06,
|
1916 |
+
"loss": 0.3323,
|
1917 |
+
"step": 2520
|
1918 |
+
},
|
1919 |
+
{
|
1920 |
+
"epoch": 44.02,
|
1921 |
+
"eval_accuracy": 0.7465437788018433,
|
1922 |
+
"eval_loss": 0.8515135645866394,
|
1923 |
+
"eval_runtime": 169.0764,
|
1924 |
+
"eval_samples_per_second": 1.283,
|
1925 |
+
"eval_steps_per_second": 0.166,
|
1926 |
+
"step": 2520
|
1927 |
+
},
|
1928 |
+
{
|
1929 |
+
"epoch": 45.0,
|
1930 |
+
"learning_rate": 4.905887064477374e-06,
|
1931 |
+
"loss": 0.5174,
|
1932 |
+
"step": 2530
|
1933 |
+
},
|
1934 |
+
{
|
1935 |
+
"epoch": 45.01,
|
1936 |
+
"learning_rate": 4.7056467761313585e-06,
|
1937 |
+
"loss": 0.405,
|
1938 |
+
"step": 2540
|
1939 |
+
},
|
1940 |
+
{
|
1941 |
+
"epoch": 45.01,
|
1942 |
+
"learning_rate": 4.505406487785343e-06,
|
1943 |
+
"loss": 0.5241,
|
1944 |
+
"step": 2550
|
1945 |
+
},
|
1946 |
+
{
|
1947 |
+
"epoch": 45.01,
|
1948 |
+
"learning_rate": 4.305166199439328e-06,
|
1949 |
+
"loss": 0.6818,
|
1950 |
+
"step": 2560
|
1951 |
+
},
|
1952 |
+
{
|
1953 |
+
"epoch": 45.02,
|
1954 |
+
"learning_rate": 4.1049259110933125e-06,
|
1955 |
+
"loss": 0.5684,
|
1956 |
+
"step": 2570
|
1957 |
+
},
|
1958 |
+
{
|
1959 |
+
"epoch": 45.02,
|
1960 |
+
"eval_accuracy": 0.7096774193548387,
|
1961 |
+
"eval_loss": 0.9309377074241638,
|
1962 |
+
"eval_runtime": 169.732,
|
1963 |
+
"eval_samples_per_second": 1.278,
|
1964 |
+
"eval_steps_per_second": 0.165,
|
1965 |
+
"step": 2576
|
1966 |
+
},
|
1967 |
+
{
|
1968 |
+
"epoch": 46.0,
|
1969 |
+
"learning_rate": 3.904685622747297e-06,
|
1970 |
+
"loss": 0.4354,
|
1971 |
+
"step": 2580
|
1972 |
+
},
|
1973 |
+
{
|
1974 |
+
"epoch": 46.01,
|
1975 |
+
"learning_rate": 3.704445334401282e-06,
|
1976 |
+
"loss": 0.3822,
|
1977 |
+
"step": 2590
|
1978 |
+
},
|
1979 |
+
{
|
1980 |
+
"epoch": 46.01,
|
1981 |
+
"learning_rate": 3.5042050460552665e-06,
|
1982 |
+
"loss": 0.5717,
|
1983 |
+
"step": 2600
|
1984 |
+
},
|
1985 |
+
{
|
1986 |
+
"epoch": 46.01,
|
1987 |
+
"learning_rate": 3.303964757709251e-06,
|
1988 |
+
"loss": 0.2871,
|
1989 |
+
"step": 2610
|
1990 |
+
},
|
1991 |
+
{
|
1992 |
+
"epoch": 46.02,
|
1993 |
+
"learning_rate": 3.103724469363236e-06,
|
1994 |
+
"loss": 0.5375,
|
1995 |
+
"step": 2620
|
1996 |
+
},
|
1997 |
+
{
|
1998 |
+
"epoch": 46.02,
|
1999 |
+
"learning_rate": 2.9034841810172205e-06,
|
2000 |
+
"loss": 0.4406,
|
2001 |
+
"step": 2630
|
2002 |
+
},
|
2003 |
+
{
|
2004 |
+
"epoch": 46.02,
|
2005 |
+
"eval_accuracy": 0.7465437788018433,
|
2006 |
+
"eval_loss": 0.8826016783714294,
|
2007 |
+
"eval_runtime": 169.4846,
|
2008 |
+
"eval_samples_per_second": 1.28,
|
2009 |
+
"eval_steps_per_second": 0.165,
|
2010 |
+
"step": 2632
|
2011 |
+
},
|
2012 |
+
{
|
2013 |
+
"epoch": 47.0,
|
2014 |
+
"learning_rate": 2.7032438926712056e-06,
|
2015 |
+
"loss": 0.5365,
|
2016 |
+
"step": 2640
|
2017 |
+
},
|
2018 |
+
{
|
2019 |
+
"epoch": 47.01,
|
2020 |
+
"learning_rate": 2.5030036043251903e-06,
|
2021 |
+
"loss": 0.3124,
|
2022 |
+
"step": 2650
|
2023 |
+
},
|
2024 |
+
{
|
2025 |
+
"epoch": 47.01,
|
2026 |
+
"learning_rate": 2.3027633159791754e-06,
|
2027 |
+
"loss": 0.3491,
|
2028 |
+
"step": 2660
|
2029 |
+
},
|
2030 |
+
{
|
2031 |
+
"epoch": 47.01,
|
2032 |
+
"learning_rate": 2.10252302763316e-06,
|
2033 |
+
"loss": 0.4527,
|
2034 |
+
"step": 2670
|
2035 |
+
},
|
2036 |
+
{
|
2037 |
+
"epoch": 47.02,
|
2038 |
+
"learning_rate": 1.9022827392871447e-06,
|
2039 |
+
"loss": 0.6164,
|
2040 |
+
"step": 2680
|
2041 |
+
},
|
2042 |
+
{
|
2043 |
+
"epoch": 47.02,
|
2044 |
+
"eval_accuracy": 0.695852534562212,
|
2045 |
+
"eval_loss": 0.8993560075759888,
|
2046 |
+
"eval_runtime": 168.0031,
|
2047 |
+
"eval_samples_per_second": 1.292,
|
2048 |
+
"eval_steps_per_second": 0.167,
|
2049 |
+
"step": 2688
|
2050 |
+
},
|
2051 |
+
{
|
2052 |
+
"epoch": 48.0,
|
2053 |
+
"learning_rate": 1.7020424509411294e-06,
|
2054 |
+
"loss": 0.2574,
|
2055 |
+
"step": 2690
|
2056 |
+
},
|
2057 |
+
{
|
2058 |
+
"epoch": 48.0,
|
2059 |
+
"learning_rate": 1.501802162595114e-06,
|
2060 |
+
"loss": 0.2911,
|
2061 |
+
"step": 2700
|
2062 |
+
},
|
2063 |
+
{
|
2064 |
+
"epoch": 48.01,
|
2065 |
+
"learning_rate": 1.301561874249099e-06,
|
2066 |
+
"loss": 0.4747,
|
2067 |
+
"step": 2710
|
2068 |
+
},
|
2069 |
+
{
|
2070 |
+
"epoch": 48.01,
|
2071 |
+
"learning_rate": 1.1013215859030839e-06,
|
2072 |
+
"loss": 0.4525,
|
2073 |
+
"step": 2720
|
2074 |
+
},
|
2075 |
+
{
|
2076 |
+
"epoch": 48.02,
|
2077 |
+
"learning_rate": 9.010812975570685e-07,
|
2078 |
+
"loss": 0.351,
|
2079 |
+
"step": 2730
|
2080 |
+
},
|
2081 |
+
{
|
2082 |
+
"epoch": 48.02,
|
2083 |
+
"learning_rate": 7.008410092110533e-07,
|
2084 |
+
"loss": 0.4549,
|
2085 |
+
"step": 2740
|
2086 |
+
},
|
2087 |
+
{
|
2088 |
+
"epoch": 48.02,
|
2089 |
+
"eval_accuracy": 0.7188940092165899,
|
2090 |
+
"eval_loss": 0.8699988722801208,
|
2091 |
+
"eval_runtime": 168.7472,
|
2092 |
+
"eval_samples_per_second": 1.286,
|
2093 |
+
"eval_steps_per_second": 0.166,
|
2094 |
+
"step": 2744
|
2095 |
+
},
|
2096 |
+
{
|
2097 |
+
"epoch": 49.0,
|
2098 |
+
"learning_rate": 5.006007208650381e-07,
|
2099 |
+
"loss": 0.4566,
|
2100 |
+
"step": 2750
|
2101 |
+
},
|
2102 |
+
{
|
2103 |
+
"epoch": 49.01,
|
2104 |
+
"learning_rate": 3.003604325190229e-07,
|
2105 |
+
"loss": 0.4441,
|
2106 |
+
"step": 2760
|
2107 |
+
},
|
2108 |
+
{
|
2109 |
+
"epoch": 49.01,
|
2110 |
+
"learning_rate": 1.0012014417300761e-07,
|
2111 |
+
"loss": 0.3453,
|
2112 |
+
"step": 2770
|
2113 |
+
},
|
2114 |
+
{
|
2115 |
+
"epoch": 49.01,
|
2116 |
+
"eval_accuracy": 0.7188940092165899,
|
2117 |
+
"eval_loss": 0.8821613788604736,
|
2118 |
+
"eval_runtime": 168.772,
|
2119 |
+
"eval_samples_per_second": 1.286,
|
2120 |
+
"eval_steps_per_second": 0.166,
|
2121 |
+
"step": 2775
|
2122 |
+
},
|
2123 |
+
{
|
2124 |
+
"epoch": 49.01,
|
2125 |
+
"step": 2775,
|
2126 |
+
"total_flos": 2.7602315185605673e+19,
|
2127 |
+
"train_loss": 0.7408499845298561,
|
2128 |
+
"train_runtime": 29540.8546,
|
2129 |
+
"train_samples_per_second": 0.752,
|
2130 |
+
"train_steps_per_second": 0.094
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 49.01,
|
2134 |
+
"eval_accuracy": 0.7824074074074074,
|
2135 |
+
"eval_loss": 0.6378526091575623,
|
2136 |
+
"eval_runtime": 176.2386,
|
2137 |
+
"eval_samples_per_second": 1.226,
|
2138 |
+
"eval_steps_per_second": 0.153,
|
2139 |
+
"step": 2775
|
2140 |
+
},
|
2141 |
+
{
|
2142 |
+
"epoch": 49.01,
|
2143 |
+
"eval_accuracy": 0.7824074074074074,
|
2144 |
+
"eval_loss": 0.6378525495529175,
|
2145 |
+
"eval_runtime": 181.9628,
|
2146 |
+
"eval_samples_per_second": 1.187,
|
2147 |
+
"eval_steps_per_second": 0.148,
|
2148 |
+
"step": 2775
|
2149 |
+
}
|
2150 |
+
],
|
2151 |
+
"logging_steps": 10,
|
2152 |
+
"max_steps": 2775,
|
2153 |
+
"num_input_tokens_seen": 0,
|
2154 |
+
"num_train_epochs": 9223372036854775807,
|
2155 |
+
"save_steps": 500,
|
2156 |
+
"total_flos": 2.7602315185605673e+19,
|
2157 |
+
"train_batch_size": 8,
|
2158 |
+
"trial_name": null,
|
2159 |
+
"trial_params": null
|
2160 |
+
}
|