End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1695 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 1.9922182705152e+18,
|
| 4 |
+
"train_loss": 0.09219677979975939,
|
| 5 |
+
"train_runtime": 12464.8013,
|
| 6 |
+
"train_samples_per_second": 32.09,
|
| 7 |
+
"train_steps_per_second": 0.251
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 1.9922182705152e+18,
|
| 4 |
+
"train_loss": 0.09219677979975939,
|
| 5 |
+
"train_runtime": 12464.8013,
|
| 6 |
+
"train_samples_per_second": 32.09,
|
| 7 |
+
"train_steps_per_second": 0.251
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1695 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.08957825601100922,
|
| 3 |
+
"best_model_checkpoint": "./dinov2_Liveness_detection_v2.2/checkpoint-1792",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 128,
|
| 6 |
+
"global_step": 3125,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0256,
|
| 13 |
+
"grad_norm": 34.66463088989258,
|
| 14 |
+
"learning_rate": 4.9792e-05,
|
| 15 |
+
"loss": 1.4705,
|
| 16 |
+
"step": 16
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0512,
|
| 20 |
+
"grad_norm": 22.070404052734375,
|
| 21 |
+
"learning_rate": 4.9536e-05,
|
| 22 |
+
"loss": 0.6239,
|
| 23 |
+
"step": 32
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0768,
|
| 27 |
+
"grad_norm": 4.249419689178467,
|
| 28 |
+
"learning_rate": 4.928e-05,
|
| 29 |
+
"loss": 0.4289,
|
| 30 |
+
"step": 48
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.1024,
|
| 34 |
+
"grad_norm": 6.337409496307373,
|
| 35 |
+
"learning_rate": 4.9024000000000004e-05,
|
| 36 |
+
"loss": 0.4128,
|
| 37 |
+
"step": 64
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.128,
|
| 41 |
+
"grad_norm": 24.880699157714844,
|
| 42 |
+
"learning_rate": 4.8768000000000005e-05,
|
| 43 |
+
"loss": 0.396,
|
| 44 |
+
"step": 80
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1536,
|
| 48 |
+
"grad_norm": 10.70703125,
|
| 49 |
+
"learning_rate": 4.8512e-05,
|
| 50 |
+
"loss": 0.3921,
|
| 51 |
+
"step": 96
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1792,
|
| 55 |
+
"grad_norm": 9.0236234664917,
|
| 56 |
+
"learning_rate": 4.8256e-05,
|
| 57 |
+
"loss": 0.2849,
|
| 58 |
+
"step": 112
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.2048,
|
| 62 |
+
"grad_norm": 10.472875595092773,
|
| 63 |
+
"learning_rate": 4.8e-05,
|
| 64 |
+
"loss": 0.3279,
|
| 65 |
+
"step": 128
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.2048,
|
| 69 |
+
"eval_accuracy": 0.87485,
|
| 70 |
+
"eval_f1": 0.8772126420661694,
|
| 71 |
+
"eval_loss": 0.28580960631370544,
|
| 72 |
+
"eval_precision": 0.8773451087999092,
|
| 73 |
+
"eval_recall": 0.87485,
|
| 74 |
+
"eval_runtime": 319.5546,
|
| 75 |
+
"eval_samples_per_second": 62.587,
|
| 76 |
+
"eval_steps_per_second": 7.823,
|
| 77 |
+
"step": 128
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.2304,
|
| 81 |
+
"grad_norm": 6.47001314163208,
|
| 82 |
+
"learning_rate": 4.7744e-05,
|
| 83 |
+
"loss": 0.264,
|
| 84 |
+
"step": 144
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.256,
|
| 88 |
+
"grad_norm": 10.121650695800781,
|
| 89 |
+
"learning_rate": 4.7488000000000003e-05,
|
| 90 |
+
"loss": 0.2465,
|
| 91 |
+
"step": 160
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.2816,
|
| 95 |
+
"grad_norm": 11.107807159423828,
|
| 96 |
+
"learning_rate": 4.7232000000000005e-05,
|
| 97 |
+
"loss": 0.2455,
|
| 98 |
+
"step": 176
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.3072,
|
| 102 |
+
"grad_norm": 11.646367073059082,
|
| 103 |
+
"learning_rate": 4.6976000000000006e-05,
|
| 104 |
+
"loss": 0.1951,
|
| 105 |
+
"step": 192
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.3328,
|
| 109 |
+
"grad_norm": 17.932363510131836,
|
| 110 |
+
"learning_rate": 4.672e-05,
|
| 111 |
+
"loss": 0.2101,
|
| 112 |
+
"step": 208
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.3584,
|
| 116 |
+
"grad_norm": 7.72909688949585,
|
| 117 |
+
"learning_rate": 4.6464e-05,
|
| 118 |
+
"loss": 0.2173,
|
| 119 |
+
"step": 224
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.384,
|
| 123 |
+
"grad_norm": 9.573624610900879,
|
| 124 |
+
"learning_rate": 4.6208e-05,
|
| 125 |
+
"loss": 0.2235,
|
| 126 |
+
"step": 240
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.4096,
|
| 130 |
+
"grad_norm": 40.75331497192383,
|
| 131 |
+
"learning_rate": 4.5952e-05,
|
| 132 |
+
"loss": 0.2389,
|
| 133 |
+
"step": 256
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.4096,
|
| 137 |
+
"eval_accuracy": 0.8881,
|
| 138 |
+
"eval_f1": 0.8818825953674535,
|
| 139 |
+
"eval_loss": 0.2695896029472351,
|
| 140 |
+
"eval_precision": 0.9196205778824599,
|
| 141 |
+
"eval_recall": 0.8881,
|
| 142 |
+
"eval_runtime": 283.6671,
|
| 143 |
+
"eval_samples_per_second": 70.505,
|
| 144 |
+
"eval_steps_per_second": 8.813,
|
| 145 |
+
"step": 256
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.4352,
|
| 149 |
+
"grad_norm": 11.891749382019043,
|
| 150 |
+
"learning_rate": 4.5696e-05,
|
| 151 |
+
"loss": 0.21,
|
| 152 |
+
"step": 272
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.4608,
|
| 156 |
+
"grad_norm": 9.512396812438965,
|
| 157 |
+
"learning_rate": 4.5440000000000005e-05,
|
| 158 |
+
"loss": 0.1958,
|
| 159 |
+
"step": 288
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.4864,
|
| 163 |
+
"grad_norm": 24.531909942626953,
|
| 164 |
+
"learning_rate": 4.5184000000000006e-05,
|
| 165 |
+
"loss": 0.1701,
|
| 166 |
+
"step": 304
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.512,
|
| 170 |
+
"grad_norm": 17.82181167602539,
|
| 171 |
+
"learning_rate": 4.4928e-05,
|
| 172 |
+
"loss": 0.2031,
|
| 173 |
+
"step": 320
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.5376,
|
| 177 |
+
"grad_norm": 6.475971698760986,
|
| 178 |
+
"learning_rate": 4.4672e-05,
|
| 179 |
+
"loss": 0.215,
|
| 180 |
+
"step": 336
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.5632,
|
| 184 |
+
"grad_norm": 19.699085235595703,
|
| 185 |
+
"learning_rate": 4.4416e-05,
|
| 186 |
+
"loss": 0.177,
|
| 187 |
+
"step": 352
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.5888,
|
| 191 |
+
"grad_norm": 9.223369598388672,
|
| 192 |
+
"learning_rate": 4.4160000000000004e-05,
|
| 193 |
+
"loss": 0.1452,
|
| 194 |
+
"step": 368
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.6144,
|
| 198 |
+
"grad_norm": 30.872596740722656,
|
| 199 |
+
"learning_rate": 4.3904e-05,
|
| 200 |
+
"loss": 0.186,
|
| 201 |
+
"step": 384
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.6144,
|
| 205 |
+
"eval_accuracy": 0.93825,
|
| 206 |
+
"eval_f1": 0.9386177153669228,
|
| 207 |
+
"eval_loss": 0.16135859489440918,
|
| 208 |
+
"eval_precision": 0.9380610433042933,
|
| 209 |
+
"eval_recall": 0.93825,
|
| 210 |
+
"eval_runtime": 282.8431,
|
| 211 |
+
"eval_samples_per_second": 70.711,
|
| 212 |
+
"eval_steps_per_second": 8.839,
|
| 213 |
+
"step": 384
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.64,
|
| 217 |
+
"grad_norm": 30.67617416381836,
|
| 218 |
+
"learning_rate": 4.3648e-05,
|
| 219 |
+
"loss": 0.1859,
|
| 220 |
+
"step": 400
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.6656,
|
| 224 |
+
"grad_norm": 15.600126266479492,
|
| 225 |
+
"learning_rate": 4.3392e-05,
|
| 226 |
+
"loss": 0.202,
|
| 227 |
+
"step": 416
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.6912,
|
| 231 |
+
"grad_norm": 23.811962127685547,
|
| 232 |
+
"learning_rate": 4.3136e-05,
|
| 233 |
+
"loss": 0.1852,
|
| 234 |
+
"step": 432
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.7168,
|
| 238 |
+
"grad_norm": 20.95099639892578,
|
| 239 |
+
"learning_rate": 4.288e-05,
|
| 240 |
+
"loss": 0.1413,
|
| 241 |
+
"step": 448
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.7424,
|
| 245 |
+
"grad_norm": 15.215927124023438,
|
| 246 |
+
"learning_rate": 4.2624000000000004e-05,
|
| 247 |
+
"loss": 0.179,
|
| 248 |
+
"step": 464
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.768,
|
| 252 |
+
"grad_norm": 10.095976829528809,
|
| 253 |
+
"learning_rate": 4.2368000000000005e-05,
|
| 254 |
+
"loss": 0.1521,
|
| 255 |
+
"step": 480
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.7936,
|
| 259 |
+
"grad_norm": 18.2394962310791,
|
| 260 |
+
"learning_rate": 4.2112e-05,
|
| 261 |
+
"loss": 0.1557,
|
| 262 |
+
"step": 496
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.8192,
|
| 266 |
+
"grad_norm": 7.70241641998291,
|
| 267 |
+
"learning_rate": 4.1856e-05,
|
| 268 |
+
"loss": 0.2048,
|
| 269 |
+
"step": 512
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.8192,
|
| 273 |
+
"eval_accuracy": 0.94035,
|
| 274 |
+
"eval_f1": 0.9410615921845757,
|
| 275 |
+
"eval_loss": 0.1568000316619873,
|
| 276 |
+
"eval_precision": 0.9414679285734555,
|
| 277 |
+
"eval_recall": 0.94035,
|
| 278 |
+
"eval_runtime": 284.3732,
|
| 279 |
+
"eval_samples_per_second": 70.33,
|
| 280 |
+
"eval_steps_per_second": 8.791,
|
| 281 |
+
"step": 512
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.8448,
|
| 285 |
+
"grad_norm": 10.704535484313965,
|
| 286 |
+
"learning_rate": 4.16e-05,
|
| 287 |
+
"loss": 0.1693,
|
| 288 |
+
"step": 528
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.8704,
|
| 292 |
+
"grad_norm": 8.581585884094238,
|
| 293 |
+
"learning_rate": 4.1344e-05,
|
| 294 |
+
"loss": 0.1326,
|
| 295 |
+
"step": 544
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.896,
|
| 299 |
+
"grad_norm": 17.267370223999023,
|
| 300 |
+
"learning_rate": 4.1088000000000004e-05,
|
| 301 |
+
"loss": 0.1503,
|
| 302 |
+
"step": 560
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.9216,
|
| 306 |
+
"grad_norm": 8.601036071777344,
|
| 307 |
+
"learning_rate": 4.0832000000000005e-05,
|
| 308 |
+
"loss": 0.1338,
|
| 309 |
+
"step": 576
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.9472,
|
| 313 |
+
"grad_norm": 18.970775604248047,
|
| 314 |
+
"learning_rate": 4.0576000000000006e-05,
|
| 315 |
+
"loss": 0.1366,
|
| 316 |
+
"step": 592
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.9728,
|
| 320 |
+
"grad_norm": 6.951398849487305,
|
| 321 |
+
"learning_rate": 4.032e-05,
|
| 322 |
+
"loss": 0.1658,
|
| 323 |
+
"step": 608
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.9984,
|
| 327 |
+
"grad_norm": 14.18687915802002,
|
| 328 |
+
"learning_rate": 4.0064e-05,
|
| 329 |
+
"loss": 0.1244,
|
| 330 |
+
"step": 624
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.024,
|
| 334 |
+
"grad_norm": 13.171905517578125,
|
| 335 |
+
"learning_rate": 3.9808e-05,
|
| 336 |
+
"loss": 0.1662,
|
| 337 |
+
"step": 640
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 1.024,
|
| 341 |
+
"eval_accuracy": 0.94265,
|
| 342 |
+
"eval_f1": 0.9433015607417117,
|
| 343 |
+
"eval_loss": 0.14737515151500702,
|
| 344 |
+
"eval_precision": 0.9436242855642897,
|
| 345 |
+
"eval_recall": 0.94265,
|
| 346 |
+
"eval_runtime": 284.6998,
|
| 347 |
+
"eval_samples_per_second": 70.249,
|
| 348 |
+
"eval_steps_per_second": 8.781,
|
| 349 |
+
"step": 640
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.0496,
|
| 353 |
+
"grad_norm": 26.666791915893555,
|
| 354 |
+
"learning_rate": 3.9552000000000003e-05,
|
| 355 |
+
"loss": 0.1282,
|
| 356 |
+
"step": 656
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 1.0752,
|
| 360 |
+
"grad_norm": 11.280467987060547,
|
| 361 |
+
"learning_rate": 3.9296e-05,
|
| 362 |
+
"loss": 0.1114,
|
| 363 |
+
"step": 672
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 1.1008,
|
| 367 |
+
"grad_norm": 7.130141258239746,
|
| 368 |
+
"learning_rate": 3.9040000000000006e-05,
|
| 369 |
+
"loss": 0.1101,
|
| 370 |
+
"step": 688
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.1264,
|
| 374 |
+
"grad_norm": 20.67454719543457,
|
| 375 |
+
"learning_rate": 3.878400000000001e-05,
|
| 376 |
+
"loss": 0.0994,
|
| 377 |
+
"step": 704
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.152,
|
| 381 |
+
"grad_norm": 12.668320655822754,
|
| 382 |
+
"learning_rate": 3.8528e-05,
|
| 383 |
+
"loss": 0.1557,
|
| 384 |
+
"step": 720
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.1776,
|
| 388 |
+
"grad_norm": 14.848088264465332,
|
| 389 |
+
"learning_rate": 3.8272e-05,
|
| 390 |
+
"loss": 0.1284,
|
| 391 |
+
"step": 736
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.2032,
|
| 395 |
+
"grad_norm": 11.244462966918945,
|
| 396 |
+
"learning_rate": 3.8016e-05,
|
| 397 |
+
"loss": 0.1082,
|
| 398 |
+
"step": 752
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.2288000000000001,
|
| 402 |
+
"grad_norm": 6.845088005065918,
|
| 403 |
+
"learning_rate": 3.776e-05,
|
| 404 |
+
"loss": 0.1257,
|
| 405 |
+
"step": 768
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 1.2288000000000001,
|
| 409 |
+
"eval_accuracy": 0.9578,
|
| 410 |
+
"eval_f1": 0.9573020715487597,
|
| 411 |
+
"eval_loss": 0.11855915188789368,
|
| 412 |
+
"eval_precision": 0.9603942863304861,
|
| 413 |
+
"eval_recall": 0.9578,
|
| 414 |
+
"eval_runtime": 282.4873,
|
| 415 |
+
"eval_samples_per_second": 70.8,
|
| 416 |
+
"eval_steps_per_second": 8.85,
|
| 417 |
+
"step": 768
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.2544,
|
| 421 |
+
"grad_norm": 20.235675811767578,
|
| 422 |
+
"learning_rate": 3.7504e-05,
|
| 423 |
+
"loss": 0.1093,
|
| 424 |
+
"step": 784
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 1.28,
|
| 428 |
+
"grad_norm": 11.13853931427002,
|
| 429 |
+
"learning_rate": 3.7248e-05,
|
| 430 |
+
"loss": 0.1085,
|
| 431 |
+
"step": 800
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 1.3056,
|
| 435 |
+
"grad_norm": 8.145424842834473,
|
| 436 |
+
"learning_rate": 3.699200000000001e-05,
|
| 437 |
+
"loss": 0.1143,
|
| 438 |
+
"step": 816
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 1.3312,
|
| 442 |
+
"grad_norm": 7.372732162475586,
|
| 443 |
+
"learning_rate": 3.6736e-05,
|
| 444 |
+
"loss": 0.1001,
|
| 445 |
+
"step": 832
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 1.3568,
|
| 449 |
+
"grad_norm": 15.379700660705566,
|
| 450 |
+
"learning_rate": 3.648e-05,
|
| 451 |
+
"loss": 0.1251,
|
| 452 |
+
"step": 848
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.3824,
|
| 456 |
+
"grad_norm": 10.976445198059082,
|
| 457 |
+
"learning_rate": 3.6224000000000004e-05,
|
| 458 |
+
"loss": 0.0952,
|
| 459 |
+
"step": 864
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.408,
|
| 463 |
+
"grad_norm": 16.909706115722656,
|
| 464 |
+
"learning_rate": 3.5968e-05,
|
| 465 |
+
"loss": 0.1199,
|
| 466 |
+
"step": 880
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 1.4336,
|
| 470 |
+
"grad_norm": 17.032054901123047,
|
| 471 |
+
"learning_rate": 3.5712e-05,
|
| 472 |
+
"loss": 0.1215,
|
| 473 |
+
"step": 896
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.4336,
|
| 477 |
+
"eval_accuracy": 0.9556,
|
| 478 |
+
"eval_f1": 0.9559899884084567,
|
| 479 |
+
"eval_loss": 0.12017901241779327,
|
| 480 |
+
"eval_precision": 0.956118373847158,
|
| 481 |
+
"eval_recall": 0.9556,
|
| 482 |
+
"eval_runtime": 283.5072,
|
| 483 |
+
"eval_samples_per_second": 70.545,
|
| 484 |
+
"eval_steps_per_second": 8.818,
|
| 485 |
+
"step": 896
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.4592,
|
| 489 |
+
"grad_norm": 29.543867111206055,
|
| 490 |
+
"learning_rate": 3.5456e-05,
|
| 491 |
+
"loss": 0.116,
|
| 492 |
+
"step": 912
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.4848,
|
| 496 |
+
"grad_norm": 9.891976356506348,
|
| 497 |
+
"learning_rate": 3.52e-05,
|
| 498 |
+
"loss": 0.106,
|
| 499 |
+
"step": 928
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.5104,
|
| 503 |
+
"grad_norm": 10.544168472290039,
|
| 504 |
+
"learning_rate": 3.4943999999999996e-05,
|
| 505 |
+
"loss": 0.0881,
|
| 506 |
+
"step": 944
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.536,
|
| 510 |
+
"grad_norm": 7.543793678283691,
|
| 511 |
+
"learning_rate": 3.4688000000000004e-05,
|
| 512 |
+
"loss": 0.0995,
|
| 513 |
+
"step": 960
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.5615999999999999,
|
| 517 |
+
"grad_norm": 13.808887481689453,
|
| 518 |
+
"learning_rate": 3.4432000000000005e-05,
|
| 519 |
+
"loss": 0.0864,
|
| 520 |
+
"step": 976
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.5872000000000002,
|
| 524 |
+
"grad_norm": 17.851945877075195,
|
| 525 |
+
"learning_rate": 3.4176e-05,
|
| 526 |
+
"loss": 0.1073,
|
| 527 |
+
"step": 992
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.6128,
|
| 531 |
+
"grad_norm": 12.95018196105957,
|
| 532 |
+
"learning_rate": 3.392e-05,
|
| 533 |
+
"loss": 0.0776,
|
| 534 |
+
"step": 1008
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.6383999999999999,
|
| 538 |
+
"grad_norm": 4.886596202850342,
|
| 539 |
+
"learning_rate": 3.3664e-05,
|
| 540 |
+
"loss": 0.0917,
|
| 541 |
+
"step": 1024
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.6383999999999999,
|
| 545 |
+
"eval_accuracy": 0.9611,
|
| 546 |
+
"eval_f1": 0.9611166892749899,
|
| 547 |
+
"eval_loss": 0.10448341071605682,
|
| 548 |
+
"eval_precision": 0.9610692328386163,
|
| 549 |
+
"eval_recall": 0.9611,
|
| 550 |
+
"eval_runtime": 281.5004,
|
| 551 |
+
"eval_samples_per_second": 71.048,
|
| 552 |
+
"eval_steps_per_second": 8.881,
|
| 553 |
+
"step": 1024
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.6640000000000001,
|
| 557 |
+
"grad_norm": 10.328375816345215,
|
| 558 |
+
"learning_rate": 3.3408e-05,
|
| 559 |
+
"loss": 0.1157,
|
| 560 |
+
"step": 1040
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.6896,
|
| 564 |
+
"grad_norm": 12.873481750488281,
|
| 565 |
+
"learning_rate": 3.3152e-05,
|
| 566 |
+
"loss": 0.0981,
|
| 567 |
+
"step": 1056
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 1.7151999999999998,
|
| 571 |
+
"grad_norm": 8.897006034851074,
|
| 572 |
+
"learning_rate": 3.2896e-05,
|
| 573 |
+
"loss": 0.0922,
|
| 574 |
+
"step": 1072
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 1.7408000000000001,
|
| 578 |
+
"grad_norm": 10.708754539489746,
|
| 579 |
+
"learning_rate": 3.2640000000000006e-05,
|
| 580 |
+
"loss": 0.0981,
|
| 581 |
+
"step": 1088
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 1.7664,
|
| 585 |
+
"grad_norm": 5.463771343231201,
|
| 586 |
+
"learning_rate": 3.2384e-05,
|
| 587 |
+
"loss": 0.099,
|
| 588 |
+
"step": 1104
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 1.792,
|
| 592 |
+
"grad_norm": 9.746946334838867,
|
| 593 |
+
"learning_rate": 3.2128e-05,
|
| 594 |
+
"loss": 0.0931,
|
| 595 |
+
"step": 1120
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 1.8176,
|
| 599 |
+
"grad_norm": 10.415765762329102,
|
| 600 |
+
"learning_rate": 3.1872e-05,
|
| 601 |
+
"loss": 0.1072,
|
| 602 |
+
"step": 1136
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.8432,
|
| 606 |
+
"grad_norm": 15.399734497070312,
|
| 607 |
+
"learning_rate": 3.1616000000000004e-05,
|
| 608 |
+
"loss": 0.1256,
|
| 609 |
+
"step": 1152
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.8432,
|
| 613 |
+
"eval_accuracy": 0.96325,
|
| 614 |
+
"eval_f1": 0.9629677002648841,
|
| 615 |
+
"eval_loss": 0.09713909775018692,
|
| 616 |
+
"eval_precision": 0.9645426077641248,
|
| 617 |
+
"eval_recall": 0.96325,
|
| 618 |
+
"eval_runtime": 278.3663,
|
| 619 |
+
"eval_samples_per_second": 71.848,
|
| 620 |
+
"eval_steps_per_second": 8.981,
|
| 621 |
+
"step": 1152
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 1.8688,
|
| 625 |
+
"grad_norm": 18.964815139770508,
|
| 626 |
+
"learning_rate": 3.136e-05,
|
| 627 |
+
"loss": 0.0833,
|
| 628 |
+
"step": 1168
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 1.8944,
|
| 632 |
+
"grad_norm": 19.32863426208496,
|
| 633 |
+
"learning_rate": 3.1104e-05,
|
| 634 |
+
"loss": 0.1033,
|
| 635 |
+
"step": 1184
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 1.92,
|
| 639 |
+
"grad_norm": 13.44693660736084,
|
| 640 |
+
"learning_rate": 3.0848e-05,
|
| 641 |
+
"loss": 0.082,
|
| 642 |
+
"step": 1200
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 1.9456,
|
| 646 |
+
"grad_norm": 6.130032539367676,
|
| 647 |
+
"learning_rate": 3.0592e-05,
|
| 648 |
+
"loss": 0.1034,
|
| 649 |
+
"step": 1216
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 1.9712,
|
| 653 |
+
"grad_norm": 25.041440963745117,
|
| 654 |
+
"learning_rate": 3.0336000000000002e-05,
|
| 655 |
+
"loss": 0.0925,
|
| 656 |
+
"step": 1232
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 1.9968,
|
| 660 |
+
"grad_norm": 21.437124252319336,
|
| 661 |
+
"learning_rate": 3.0080000000000003e-05,
|
| 662 |
+
"loss": 0.073,
|
| 663 |
+
"step": 1248
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 2.0224,
|
| 667 |
+
"grad_norm": 5.39111852645874,
|
| 668 |
+
"learning_rate": 2.9824e-05,
|
| 669 |
+
"loss": 0.0664,
|
| 670 |
+
"step": 1264
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 2.048,
|
| 674 |
+
"grad_norm": 23.514789581298828,
|
| 675 |
+
"learning_rate": 2.9568000000000002e-05,
|
| 676 |
+
"loss": 0.0676,
|
| 677 |
+
"step": 1280
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 2.048,
|
| 681 |
+
"eval_accuracy": 0.9487,
|
| 682 |
+
"eval_f1": 0.9477085652077598,
|
| 683 |
+
"eval_loss": 0.1523836851119995,
|
| 684 |
+
"eval_precision": 0.954484900593093,
|
| 685 |
+
"eval_recall": 0.9487,
|
| 686 |
+
"eval_runtime": 281.1485,
|
| 687 |
+
"eval_samples_per_second": 71.137,
|
| 688 |
+
"eval_steps_per_second": 8.892,
|
| 689 |
+
"step": 1280
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 2.0736,
|
| 693 |
+
"grad_norm": 13.302830696105957,
|
| 694 |
+
"learning_rate": 2.9312e-05,
|
| 695 |
+
"loss": 0.0653,
|
| 696 |
+
"step": 1296
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 2.0992,
|
| 700 |
+
"grad_norm": 3.786924123764038,
|
| 701 |
+
"learning_rate": 2.9056e-05,
|
| 702 |
+
"loss": 0.1038,
|
| 703 |
+
"step": 1312
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 2.1248,
|
| 707 |
+
"grad_norm": 13.609587669372559,
|
| 708 |
+
"learning_rate": 2.88e-05,
|
| 709 |
+
"loss": 0.0521,
|
| 710 |
+
"step": 1328
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.1504,
|
| 714 |
+
"grad_norm": 12.58696174621582,
|
| 715 |
+
"learning_rate": 2.8544000000000003e-05,
|
| 716 |
+
"loss": 0.0736,
|
| 717 |
+
"step": 1344
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 2.176,
|
| 721 |
+
"grad_norm": 8.014899253845215,
|
| 722 |
+
"learning_rate": 2.8288000000000004e-05,
|
| 723 |
+
"loss": 0.0594,
|
| 724 |
+
"step": 1360
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 2.2016,
|
| 728 |
+
"grad_norm": 43.778987884521484,
|
| 729 |
+
"learning_rate": 2.8032000000000002e-05,
|
| 730 |
+
"loss": 0.0668,
|
| 731 |
+
"step": 1376
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 2.2272,
|
| 735 |
+
"grad_norm": 16.46095085144043,
|
| 736 |
+
"learning_rate": 2.7776000000000003e-05,
|
| 737 |
+
"loss": 0.0521,
|
| 738 |
+
"step": 1392
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 2.2528,
|
| 742 |
+
"grad_norm": 14.682294845581055,
|
| 743 |
+
"learning_rate": 2.752e-05,
|
| 744 |
+
"loss": 0.0458,
|
| 745 |
+
"step": 1408
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 2.2528,
|
| 749 |
+
"eval_accuracy": 0.9641,
|
| 750 |
+
"eval_f1": 0.9643031804596063,
|
| 751 |
+
"eval_loss": 0.1148511990904808,
|
| 752 |
+
"eval_precision": 0.9642012329521269,
|
| 753 |
+
"eval_recall": 0.9641,
|
| 754 |
+
"eval_runtime": 278.1179,
|
| 755 |
+
"eval_samples_per_second": 71.912,
|
| 756 |
+
"eval_steps_per_second": 8.989,
|
| 757 |
+
"step": 1408
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 2.2784,
|
| 761 |
+
"grad_norm": 16.87139892578125,
|
| 762 |
+
"learning_rate": 2.7264000000000002e-05,
|
| 763 |
+
"loss": 0.0573,
|
| 764 |
+
"step": 1424
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 2.304,
|
| 768 |
+
"grad_norm": 15.872309684753418,
|
| 769 |
+
"learning_rate": 2.7008e-05,
|
| 770 |
+
"loss": 0.0572,
|
| 771 |
+
"step": 1440
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 2.3296,
|
| 775 |
+
"grad_norm": 7.687776565551758,
|
| 776 |
+
"learning_rate": 2.6752e-05,
|
| 777 |
+
"loss": 0.0425,
|
| 778 |
+
"step": 1456
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 2.3552,
|
| 782 |
+
"grad_norm": 4.777771949768066,
|
| 783 |
+
"learning_rate": 2.6496e-05,
|
| 784 |
+
"loss": 0.0428,
|
| 785 |
+
"step": 1472
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 2.3808,
|
| 789 |
+
"grad_norm": 13.205519676208496,
|
| 790 |
+
"learning_rate": 2.6240000000000003e-05,
|
| 791 |
+
"loss": 0.0637,
|
| 792 |
+
"step": 1488
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 2.4064,
|
| 796 |
+
"grad_norm": 11.858043670654297,
|
| 797 |
+
"learning_rate": 2.5984000000000004e-05,
|
| 798 |
+
"loss": 0.0967,
|
| 799 |
+
"step": 1504
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 2.432,
|
| 803 |
+
"grad_norm": 9.521114349365234,
|
| 804 |
+
"learning_rate": 2.5728e-05,
|
| 805 |
+
"loss": 0.0656,
|
| 806 |
+
"step": 1520
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.4576000000000002,
|
| 810 |
+
"grad_norm": 12.698617935180664,
|
| 811 |
+
"learning_rate": 2.5472000000000003e-05,
|
| 812 |
+
"loss": 0.0462,
|
| 813 |
+
"step": 1536
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 2.4576000000000002,
|
| 817 |
+
"eval_accuracy": 0.96305,
|
| 818 |
+
"eval_f1": 0.9632364288008687,
|
| 819 |
+
"eval_loss": 0.12331197410821915,
|
| 820 |
+
"eval_precision": 0.9630746783109622,
|
| 821 |
+
"eval_recall": 0.96305,
|
| 822 |
+
"eval_runtime": 281.1929,
|
| 823 |
+
"eval_samples_per_second": 71.126,
|
| 824 |
+
"eval_steps_per_second": 8.891,
|
| 825 |
+
"step": 1536
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 2.4832,
|
| 829 |
+
"grad_norm": 9.886771202087402,
|
| 830 |
+
"learning_rate": 2.5216e-05,
|
| 831 |
+
"loss": 0.0601,
|
| 832 |
+
"step": 1552
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 2.5088,
|
| 836 |
+
"grad_norm": 10.044967651367188,
|
| 837 |
+
"learning_rate": 2.496e-05,
|
| 838 |
+
"loss": 0.0513,
|
| 839 |
+
"step": 1568
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 2.5343999999999998,
|
| 843 |
+
"grad_norm": 6.7596282958984375,
|
| 844 |
+
"learning_rate": 2.4704000000000003e-05,
|
| 845 |
+
"loss": 0.0553,
|
| 846 |
+
"step": 1584
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 2.56,
|
| 850 |
+
"grad_norm": 11.670364379882812,
|
| 851 |
+
"learning_rate": 2.4448e-05,
|
| 852 |
+
"loss": 0.055,
|
| 853 |
+
"step": 1600
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 2.5856,
|
| 857 |
+
"grad_norm": 11.451431274414062,
|
| 858 |
+
"learning_rate": 2.4192e-05,
|
| 859 |
+
"loss": 0.0596,
|
| 860 |
+
"step": 1616
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 2.6112,
|
| 864 |
+
"grad_norm": 10.542276382446289,
|
| 865 |
+
"learning_rate": 2.3936e-05,
|
| 866 |
+
"loss": 0.0736,
|
| 867 |
+
"step": 1632
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 2.6368,
|
| 871 |
+
"grad_norm": 5.758074760437012,
|
| 872 |
+
"learning_rate": 2.3680000000000004e-05,
|
| 873 |
+
"loss": 0.051,
|
| 874 |
+
"step": 1648
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 2.6624,
|
| 878 |
+
"grad_norm": 10.516912460327148,
|
| 879 |
+
"learning_rate": 2.3424e-05,
|
| 880 |
+
"loss": 0.0453,
|
| 881 |
+
"step": 1664
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 2.6624,
|
| 885 |
+
"eval_accuracy": 0.96715,
|
| 886 |
+
"eval_f1": 0.9669626011452112,
|
| 887 |
+
"eval_loss": 0.10297044366598129,
|
| 888 |
+
"eval_precision": 0.967941822724822,
|
| 889 |
+
"eval_recall": 0.96715,
|
| 890 |
+
"eval_runtime": 279.502,
|
| 891 |
+
"eval_samples_per_second": 71.556,
|
| 892 |
+
"eval_steps_per_second": 8.944,
|
| 893 |
+
"step": 1664
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 2.6879999999999997,
|
| 897 |
+
"grad_norm": 12.392754554748535,
|
| 898 |
+
"learning_rate": 2.3168000000000002e-05,
|
| 899 |
+
"loss": 0.0553,
|
| 900 |
+
"step": 1680
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 2.7136,
|
| 904 |
+
"grad_norm": 17.765338897705078,
|
| 905 |
+
"learning_rate": 2.2912e-05,
|
| 906 |
+
"loss": 0.0532,
|
| 907 |
+
"step": 1696
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 2.7392,
|
| 911 |
+
"grad_norm": 9.730329513549805,
|
| 912 |
+
"learning_rate": 2.2656e-05,
|
| 913 |
+
"loss": 0.0371,
|
| 914 |
+
"step": 1712
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 2.7648,
|
| 918 |
+
"grad_norm": 13.465132713317871,
|
| 919 |
+
"learning_rate": 2.2400000000000002e-05,
|
| 920 |
+
"loss": 0.0596,
|
| 921 |
+
"step": 1728
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 2.7904,
|
| 925 |
+
"grad_norm": 27.750364303588867,
|
| 926 |
+
"learning_rate": 2.2144e-05,
|
| 927 |
+
"loss": 0.0484,
|
| 928 |
+
"step": 1744
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 2.816,
|
| 932 |
+
"grad_norm": 3.0103964805603027,
|
| 933 |
+
"learning_rate": 2.1888e-05,
|
| 934 |
+
"loss": 0.0701,
|
| 935 |
+
"step": 1760
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 2.8416,
|
| 939 |
+
"grad_norm": 10.96268081665039,
|
| 940 |
+
"learning_rate": 2.1632000000000002e-05,
|
| 941 |
+
"loss": 0.0533,
|
| 942 |
+
"step": 1776
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 2.8672,
|
| 946 |
+
"grad_norm": 12.069666862487793,
|
| 947 |
+
"learning_rate": 2.1376e-05,
|
| 948 |
+
"loss": 0.0631,
|
| 949 |
+
"step": 1792
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 2.8672,
|
| 953 |
+
"eval_accuracy": 0.967,
|
| 954 |
+
"eval_f1": 0.9671810050273864,
|
| 955 |
+
"eval_loss": 0.08957825601100922,
|
| 956 |
+
"eval_precision": 0.9671147410539459,
|
| 957 |
+
"eval_recall": 0.967,
|
| 958 |
+
"eval_runtime": 277.5967,
|
| 959 |
+
"eval_samples_per_second": 72.047,
|
| 960 |
+
"eval_steps_per_second": 9.006,
|
| 961 |
+
"step": 1792
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.8928000000000003,
|
| 965 |
+
"grad_norm": 13.720245361328125,
|
| 966 |
+
"learning_rate": 2.112e-05,
|
| 967 |
+
"loss": 0.0535,
|
| 968 |
+
"step": 1808
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.9184,
|
| 972 |
+
"grad_norm": 6.320708751678467,
|
| 973 |
+
"learning_rate": 2.0864e-05,
|
| 974 |
+
"loss": 0.0649,
|
| 975 |
+
"step": 1824
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.944,
|
| 979 |
+
"grad_norm": 5.463020324707031,
|
| 980 |
+
"learning_rate": 2.0608000000000003e-05,
|
| 981 |
+
"loss": 0.048,
|
| 982 |
+
"step": 1840
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.9696,
|
| 986 |
+
"grad_norm": 4.433831691741943,
|
| 987 |
+
"learning_rate": 2.0352e-05,
|
| 988 |
+
"loss": 0.0586,
|
| 989 |
+
"step": 1856
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.9952,
|
| 993 |
+
"grad_norm": 17.39836883544922,
|
| 994 |
+
"learning_rate": 2.0096000000000002e-05,
|
| 995 |
+
"loss": 0.049,
|
| 996 |
+
"step": 1872
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 3.0208,
|
| 1000 |
+
"grad_norm": 14.647570610046387,
|
| 1001 |
+
"learning_rate": 1.984e-05,
|
| 1002 |
+
"loss": 0.0342,
|
| 1003 |
+
"step": 1888
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 3.0464,
|
| 1007 |
+
"grad_norm": 14.295393943786621,
|
| 1008 |
+
"learning_rate": 1.9584e-05,
|
| 1009 |
+
"loss": 0.0388,
|
| 1010 |
+
"step": 1904
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 3.072,
|
| 1014 |
+
"grad_norm": 11.012173652648926,
|
| 1015 |
+
"learning_rate": 1.9328000000000002e-05,
|
| 1016 |
+
"loss": 0.0358,
|
| 1017 |
+
"step": 1920
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 3.072,
|
| 1021 |
+
"eval_accuracy": 0.9735,
|
| 1022 |
+
"eval_f1": 0.9734189039509762,
|
| 1023 |
+
"eval_loss": 0.09655638784170151,
|
| 1024 |
+
"eval_precision": 0.973788868583104,
|
| 1025 |
+
"eval_recall": 0.9735,
|
| 1026 |
+
"eval_runtime": 279.0333,
|
| 1027 |
+
"eval_samples_per_second": 71.676,
|
| 1028 |
+
"eval_steps_per_second": 8.96,
|
| 1029 |
+
"step": 1920
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 3.0976,
|
| 1033 |
+
"grad_norm": 9.099358558654785,
|
| 1034 |
+
"learning_rate": 1.9072e-05,
|
| 1035 |
+
"loss": 0.0188,
|
| 1036 |
+
"step": 1936
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 3.1232,
|
| 1040 |
+
"grad_norm": 13.887655258178711,
|
| 1041 |
+
"learning_rate": 1.8816e-05,
|
| 1042 |
+
"loss": 0.0317,
|
| 1043 |
+
"step": 1952
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 3.1488,
|
| 1047 |
+
"grad_norm": 19.007081985473633,
|
| 1048 |
+
"learning_rate": 1.856e-05,
|
| 1049 |
+
"loss": 0.0259,
|
| 1050 |
+
"step": 1968
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 3.1744,
|
| 1054 |
+
"grad_norm": 11.445087432861328,
|
| 1055 |
+
"learning_rate": 1.8304000000000003e-05,
|
| 1056 |
+
"loss": 0.0372,
|
| 1057 |
+
"step": 1984
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 3.2,
|
| 1061 |
+
"grad_norm": 20.101608276367188,
|
| 1062 |
+
"learning_rate": 1.8048e-05,
|
| 1063 |
+
"loss": 0.0223,
|
| 1064 |
+
"step": 2000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 3.2256,
|
| 1068 |
+
"grad_norm": 20.436317443847656,
|
| 1069 |
+
"learning_rate": 1.7792000000000002e-05,
|
| 1070 |
+
"loss": 0.0274,
|
| 1071 |
+
"step": 2016
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 3.2512,
|
| 1075 |
+
"grad_norm": 11.57426929473877,
|
| 1076 |
+
"learning_rate": 1.7536e-05,
|
| 1077 |
+
"loss": 0.03,
|
| 1078 |
+
"step": 2032
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 3.2768,
|
| 1082 |
+
"grad_norm": 10.924423217773438,
|
| 1083 |
+
"learning_rate": 1.728e-05,
|
| 1084 |
+
"loss": 0.0229,
|
| 1085 |
+
"step": 2048
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 3.2768,
|
| 1089 |
+
"eval_accuracy": 0.96745,
|
| 1090 |
+
"eval_f1": 0.9676427047404297,
|
| 1091 |
+
"eval_loss": 0.12504249811172485,
|
| 1092 |
+
"eval_precision": 0.967625634978291,
|
| 1093 |
+
"eval_recall": 0.96745,
|
| 1094 |
+
"eval_runtime": 280.1996,
|
| 1095 |
+
"eval_samples_per_second": 71.378,
|
| 1096 |
+
"eval_steps_per_second": 8.922,
|
| 1097 |
+
"step": 2048
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 3.3024,
|
| 1101 |
+
"grad_norm": 26.28428077697754,
|
| 1102 |
+
"learning_rate": 1.7024e-05,
|
| 1103 |
+
"loss": 0.0282,
|
| 1104 |
+
"step": 2064
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 3.328,
|
| 1108 |
+
"grad_norm": 1.8928290605545044,
|
| 1109 |
+
"learning_rate": 1.6768e-05,
|
| 1110 |
+
"loss": 0.0245,
|
| 1111 |
+
"step": 2080
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 3.3536,
|
| 1115 |
+
"grad_norm": 2.3847615718841553,
|
| 1116 |
+
"learning_rate": 1.6512e-05,
|
| 1117 |
+
"loss": 0.0273,
|
| 1118 |
+
"step": 2096
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 3.3792,
|
| 1122 |
+
"grad_norm": 43.50973892211914,
|
| 1123 |
+
"learning_rate": 1.6256e-05,
|
| 1124 |
+
"loss": 0.0485,
|
| 1125 |
+
"step": 2112
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 3.4048,
|
| 1129 |
+
"grad_norm": 3.1951754093170166,
|
| 1130 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 1131 |
+
"loss": 0.0312,
|
| 1132 |
+
"step": 2128
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 3.4304,
|
| 1136 |
+
"grad_norm": 8.364577293395996,
|
| 1137 |
+
"learning_rate": 1.5744e-05,
|
| 1138 |
+
"loss": 0.0308,
|
| 1139 |
+
"step": 2144
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 3.456,
|
| 1143 |
+
"grad_norm": 15.640523910522461,
|
| 1144 |
+
"learning_rate": 1.5487999999999998e-05,
|
| 1145 |
+
"loss": 0.0174,
|
| 1146 |
+
"step": 2160
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 3.4816,
|
| 1150 |
+
"grad_norm": 17.16481590270996,
|
| 1151 |
+
"learning_rate": 1.5232000000000003e-05,
|
| 1152 |
+
"loss": 0.0272,
|
| 1153 |
+
"step": 2176
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 3.4816,
|
| 1157 |
+
"eval_accuracy": 0.96915,
|
| 1158 |
+
"eval_f1": 0.9692869954564767,
|
| 1159 |
+
"eval_loss": 0.11478219926357269,
|
| 1160 |
+
"eval_precision": 0.9691824640029512,
|
| 1161 |
+
"eval_recall": 0.96915,
|
| 1162 |
+
"eval_runtime": 281.6325,
|
| 1163 |
+
"eval_samples_per_second": 71.015,
|
| 1164 |
+
"eval_steps_per_second": 8.877,
|
| 1165 |
+
"step": 2176
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 3.5072,
|
| 1169 |
+
"grad_norm": 1.0738009214401245,
|
| 1170 |
+
"learning_rate": 1.4976000000000002e-05,
|
| 1171 |
+
"loss": 0.0255,
|
| 1172 |
+
"step": 2192
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 3.5328,
|
| 1176 |
+
"grad_norm": 41.19829177856445,
|
| 1177 |
+
"learning_rate": 1.472e-05,
|
| 1178 |
+
"loss": 0.031,
|
| 1179 |
+
"step": 2208
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 3.5584,
|
| 1183 |
+
"grad_norm": 15.848438262939453,
|
| 1184 |
+
"learning_rate": 1.4463999999999999e-05,
|
| 1185 |
+
"loss": 0.0377,
|
| 1186 |
+
"step": 2224
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 3.584,
|
| 1190 |
+
"grad_norm": 2.093379020690918,
|
| 1191 |
+
"learning_rate": 1.4208000000000002e-05,
|
| 1192 |
+
"loss": 0.0244,
|
| 1193 |
+
"step": 2240
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 3.6096,
|
| 1197 |
+
"grad_norm": 8.310689926147461,
|
| 1198 |
+
"learning_rate": 1.3952000000000001e-05,
|
| 1199 |
+
"loss": 0.0203,
|
| 1200 |
+
"step": 2256
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 3.6352,
|
| 1204 |
+
"grad_norm": 24.2000675201416,
|
| 1205 |
+
"learning_rate": 1.3696e-05,
|
| 1206 |
+
"loss": 0.0408,
|
| 1207 |
+
"step": 2272
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 3.6608,
|
| 1211 |
+
"grad_norm": 13.623507499694824,
|
| 1212 |
+
"learning_rate": 1.344e-05,
|
| 1213 |
+
"loss": 0.0214,
|
| 1214 |
+
"step": 2288
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 3.6864,
|
| 1218 |
+
"grad_norm": 18.911640167236328,
|
| 1219 |
+
"learning_rate": 1.3184000000000001e-05,
|
| 1220 |
+
"loss": 0.0253,
|
| 1221 |
+
"step": 2304
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 3.6864,
|
| 1225 |
+
"eval_accuracy": 0.97565,
|
| 1226 |
+
"eval_f1": 0.9755495544379561,
|
| 1227 |
+
"eval_loss": 0.11298450827598572,
|
| 1228 |
+
"eval_precision": 0.9760774584693096,
|
| 1229 |
+
"eval_recall": 0.97565,
|
| 1230 |
+
"eval_runtime": 280.3093,
|
| 1231 |
+
"eval_samples_per_second": 71.35,
|
| 1232 |
+
"eval_steps_per_second": 8.919,
|
| 1233 |
+
"step": 2304
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 3.7119999999999997,
|
| 1237 |
+
"grad_norm": 1.1341925859451294,
|
| 1238 |
+
"learning_rate": 1.2928e-05,
|
| 1239 |
+
"loss": 0.0197,
|
| 1240 |
+
"step": 2320
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 3.7376,
|
| 1244 |
+
"grad_norm": 13.4550142288208,
|
| 1245 |
+
"learning_rate": 1.2672e-05,
|
| 1246 |
+
"loss": 0.0361,
|
| 1247 |
+
"step": 2336
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 3.7632,
|
| 1251 |
+
"grad_norm": 15.61780834197998,
|
| 1252 |
+
"learning_rate": 1.2416000000000001e-05,
|
| 1253 |
+
"loss": 0.0393,
|
| 1254 |
+
"step": 2352
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 3.7888,
|
| 1258 |
+
"grad_norm": 13.786069869995117,
|
| 1259 |
+
"learning_rate": 1.216e-05,
|
| 1260 |
+
"loss": 0.0442,
|
| 1261 |
+
"step": 2368
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 3.8144,
|
| 1265 |
+
"grad_norm": 17.309276580810547,
|
| 1266 |
+
"learning_rate": 1.1904000000000002e-05,
|
| 1267 |
+
"loss": 0.0208,
|
| 1268 |
+
"step": 2384
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 3.84,
|
| 1272 |
+
"grad_norm": 2.970080614089966,
|
| 1273 |
+
"learning_rate": 1.1648000000000001e-05,
|
| 1274 |
+
"loss": 0.0335,
|
| 1275 |
+
"step": 2400
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 3.8656,
|
| 1279 |
+
"grad_norm": 14.395270347595215,
|
| 1280 |
+
"learning_rate": 1.1392e-05,
|
| 1281 |
+
"loss": 0.0296,
|
| 1282 |
+
"step": 2416
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 3.8912,
|
| 1286 |
+
"grad_norm": 33.41287612915039,
|
| 1287 |
+
"learning_rate": 1.1136e-05,
|
| 1288 |
+
"loss": 0.0249,
|
| 1289 |
+
"step": 2432
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 3.8912,
|
| 1293 |
+
"eval_accuracy": 0.97155,
|
| 1294 |
+
"eval_f1": 0.971651659709645,
|
| 1295 |
+
"eval_loss": 0.1090923324227333,
|
| 1296 |
+
"eval_precision": 0.9715421641079485,
|
| 1297 |
+
"eval_recall": 0.97155,
|
| 1298 |
+
"eval_runtime": 282.2386,
|
| 1299 |
+
"eval_samples_per_second": 70.862,
|
| 1300 |
+
"eval_steps_per_second": 8.858,
|
| 1301 |
+
"step": 2432
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 3.9168,
|
| 1305 |
+
"grad_norm": 5.735462188720703,
|
| 1306 |
+
"learning_rate": 1.088e-05,
|
| 1307 |
+
"loss": 0.0246,
|
| 1308 |
+
"step": 2448
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 3.9424,
|
| 1312 |
+
"grad_norm": 9.76904582977295,
|
| 1313 |
+
"learning_rate": 1.0624e-05,
|
| 1314 |
+
"loss": 0.0189,
|
| 1315 |
+
"step": 2464
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 3.968,
|
| 1319 |
+
"grad_norm": 21.996170043945312,
|
| 1320 |
+
"learning_rate": 1.0368e-05,
|
| 1321 |
+
"loss": 0.0232,
|
| 1322 |
+
"step": 2480
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 3.9936,
|
| 1326 |
+
"grad_norm": 12.781206130981445,
|
| 1327 |
+
"learning_rate": 1.0112e-05,
|
| 1328 |
+
"loss": 0.0315,
|
| 1329 |
+
"step": 2496
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 4.0192,
|
| 1333 |
+
"grad_norm": 1.531118631362915,
|
| 1334 |
+
"learning_rate": 9.856e-06,
|
| 1335 |
+
"loss": 0.0092,
|
| 1336 |
+
"step": 2512
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 4.0448,
|
| 1340 |
+
"grad_norm": 4.409332752227783,
|
| 1341 |
+
"learning_rate": 9.600000000000001e-06,
|
| 1342 |
+
"loss": 0.0066,
|
| 1343 |
+
"step": 2528
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 4.0704,
|
| 1347 |
+
"grad_norm": 19.169538497924805,
|
| 1348 |
+
"learning_rate": 9.344e-06,
|
| 1349 |
+
"loss": 0.0131,
|
| 1350 |
+
"step": 2544
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 4.096,
|
| 1354 |
+
"grad_norm": 2.931497573852539,
|
| 1355 |
+
"learning_rate": 9.088e-06,
|
| 1356 |
+
"loss": 0.0049,
|
| 1357 |
+
"step": 2560
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 4.096,
|
| 1361 |
+
"eval_accuracy": 0.97555,
|
| 1362 |
+
"eval_f1": 0.9755704234666928,
|
| 1363 |
+
"eval_loss": 0.1419830322265625,
|
| 1364 |
+
"eval_precision": 0.9755193219728142,
|
| 1365 |
+
"eval_recall": 0.97555,
|
| 1366 |
+
"eval_runtime": 279.3028,
|
| 1367 |
+
"eval_samples_per_second": 71.607,
|
| 1368 |
+
"eval_steps_per_second": 8.951,
|
| 1369 |
+
"step": 2560
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 4.1216,
|
| 1373 |
+
"grad_norm": 56.46807861328125,
|
| 1374 |
+
"learning_rate": 8.832e-06,
|
| 1375 |
+
"loss": 0.0083,
|
| 1376 |
+
"step": 2576
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 4.1472,
|
| 1380 |
+
"grad_norm": 7.309184551239014,
|
| 1381 |
+
"learning_rate": 8.576e-06,
|
| 1382 |
+
"loss": 0.009,
|
| 1383 |
+
"step": 2592
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 4.1728,
|
| 1387 |
+
"grad_norm": 0.3025994896888733,
|
| 1388 |
+
"learning_rate": 8.32e-06,
|
| 1389 |
+
"loss": 0.0117,
|
| 1390 |
+
"step": 2608
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 4.1984,
|
| 1394 |
+
"grad_norm": 2.3376545906066895,
|
| 1395 |
+
"learning_rate": 8.064000000000001e-06,
|
| 1396 |
+
"loss": 0.0158,
|
| 1397 |
+
"step": 2624
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 4.224,
|
| 1401 |
+
"grad_norm": 0.007340936455875635,
|
| 1402 |
+
"learning_rate": 7.808e-06,
|
| 1403 |
+
"loss": 0.0063,
|
| 1404 |
+
"step": 2640
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 4.2496,
|
| 1408 |
+
"grad_norm": 49.095787048339844,
|
| 1409 |
+
"learning_rate": 7.552000000000001e-06,
|
| 1410 |
+
"loss": 0.0086,
|
| 1411 |
+
"step": 2656
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 4.2752,
|
| 1415 |
+
"grad_norm": 0.7440342307090759,
|
| 1416 |
+
"learning_rate": 7.296e-06,
|
| 1417 |
+
"loss": 0.0067,
|
| 1418 |
+
"step": 2672
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 4.3008,
|
| 1422 |
+
"grad_norm": 0.016900308430194855,
|
| 1423 |
+
"learning_rate": 7.04e-06,
|
| 1424 |
+
"loss": 0.0159,
|
| 1425 |
+
"step": 2688
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 4.3008,
|
| 1429 |
+
"eval_accuracy": 0.9775,
|
| 1430 |
+
"eval_f1": 0.9774367192636422,
|
| 1431 |
+
"eval_loss": 0.14225812256336212,
|
| 1432 |
+
"eval_precision": 0.977734598030591,
|
| 1433 |
+
"eval_recall": 0.9775,
|
| 1434 |
+
"eval_runtime": 279.9457,
|
| 1435 |
+
"eval_samples_per_second": 71.442,
|
| 1436 |
+
"eval_steps_per_second": 8.93,
|
| 1437 |
+
"step": 2688
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 4.3264,
|
| 1441 |
+
"grad_norm": 0.04318913444876671,
|
| 1442 |
+
"learning_rate": 6.784e-06,
|
| 1443 |
+
"loss": 0.0113,
|
| 1444 |
+
"step": 2704
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 4.352,
|
| 1448 |
+
"grad_norm": 24.88129234313965,
|
| 1449 |
+
"learning_rate": 6.528000000000001e-06,
|
| 1450 |
+
"loss": 0.0064,
|
| 1451 |
+
"step": 2720
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 4.3776,
|
| 1455 |
+
"grad_norm": 0.3640088140964508,
|
| 1456 |
+
"learning_rate": 6.272e-06,
|
| 1457 |
+
"loss": 0.0105,
|
| 1458 |
+
"step": 2736
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 4.4032,
|
| 1462 |
+
"grad_norm": 0.05440501868724823,
|
| 1463 |
+
"learning_rate": 6.0160000000000005e-06,
|
| 1464 |
+
"loss": 0.0179,
|
| 1465 |
+
"step": 2752
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 4.4288,
|
| 1469 |
+
"grad_norm": 0.35142025351524353,
|
| 1470 |
+
"learning_rate": 5.76e-06,
|
| 1471 |
+
"loss": 0.0052,
|
| 1472 |
+
"step": 2768
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 4.4544,
|
| 1476 |
+
"grad_norm": 17.978708267211914,
|
| 1477 |
+
"learning_rate": 5.504e-06,
|
| 1478 |
+
"loss": 0.0052,
|
| 1479 |
+
"step": 2784
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 4.48,
|
| 1483 |
+
"grad_norm": 0.02156807854771614,
|
| 1484 |
+
"learning_rate": 5.248e-06,
|
| 1485 |
+
"loss": 0.007,
|
| 1486 |
+
"step": 2800
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 4.5056,
|
| 1490 |
+
"grad_norm": 0.010784478858113289,
|
| 1491 |
+
"learning_rate": 4.992e-06,
|
| 1492 |
+
"loss": 0.0026,
|
| 1493 |
+
"step": 2816
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 4.5056,
|
| 1497 |
+
"eval_accuracy": 0.9774,
|
| 1498 |
+
"eval_f1": 0.9773457166351697,
|
| 1499 |
+
"eval_loss": 0.1453547328710556,
|
| 1500 |
+
"eval_precision": 0.9775879326882146,
|
| 1501 |
+
"eval_recall": 0.9774,
|
| 1502 |
+
"eval_runtime": 282.3251,
|
| 1503 |
+
"eval_samples_per_second": 70.84,
|
| 1504 |
+
"eval_steps_per_second": 8.855,
|
| 1505 |
+
"step": 2816
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 4.5312,
|
| 1509 |
+
"grad_norm": 27.997182846069336,
|
| 1510 |
+
"learning_rate": 4.736e-06,
|
| 1511 |
+
"loss": 0.0098,
|
| 1512 |
+
"step": 2832
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 4.5568,
|
| 1516 |
+
"grad_norm": 0.01749483309686184,
|
| 1517 |
+
"learning_rate": 4.48e-06,
|
| 1518 |
+
"loss": 0.0065,
|
| 1519 |
+
"step": 2848
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 4.5824,
|
| 1523 |
+
"grad_norm": 0.020929008722305298,
|
| 1524 |
+
"learning_rate": 4.2240000000000006e-06,
|
| 1525 |
+
"loss": 0.0023,
|
| 1526 |
+
"step": 2864
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 4.608,
|
| 1530 |
+
"grad_norm": 25.257774353027344,
|
| 1531 |
+
"learning_rate": 3.968e-06,
|
| 1532 |
+
"loss": 0.0093,
|
| 1533 |
+
"step": 2880
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 4.6336,
|
| 1537 |
+
"grad_norm": 0.07572409510612488,
|
| 1538 |
+
"learning_rate": 3.712e-06,
|
| 1539 |
+
"loss": 0.0043,
|
| 1540 |
+
"step": 2896
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 4.6592,
|
| 1544 |
+
"grad_norm": 0.010246412828564644,
|
| 1545 |
+
"learning_rate": 3.456e-06,
|
| 1546 |
+
"loss": 0.004,
|
| 1547 |
+
"step": 2912
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 4.6848,
|
| 1551 |
+
"grad_norm": 6.125358581542969,
|
| 1552 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 1553 |
+
"loss": 0.0053,
|
| 1554 |
+
"step": 2928
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 4.7104,
|
| 1558 |
+
"grad_norm": 10.814197540283203,
|
| 1559 |
+
"learning_rate": 2.944e-06,
|
| 1560 |
+
"loss": 0.0059,
|
| 1561 |
+
"step": 2944
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 4.7104,
|
| 1565 |
+
"eval_accuracy": 0.9785,
|
| 1566 |
+
"eval_f1": 0.9784932444241875,
|
| 1567 |
+
"eval_loss": 0.14451073110103607,
|
| 1568 |
+
"eval_precision": 0.9785149243735634,
|
| 1569 |
+
"eval_recall": 0.9785,
|
| 1570 |
+
"eval_runtime": 276.5263,
|
| 1571 |
+
"eval_samples_per_second": 72.326,
|
| 1572 |
+
"eval_steps_per_second": 9.041,
|
| 1573 |
+
"step": 2944
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 4.736,
|
| 1577 |
+
"grad_norm": 31.85561180114746,
|
| 1578 |
+
"learning_rate": 2.6880000000000004e-06,
|
| 1579 |
+
"loss": 0.0073,
|
| 1580 |
+
"step": 2960
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 4.7616,
|
| 1584 |
+
"grad_norm": 6.0043182373046875,
|
| 1585 |
+
"learning_rate": 2.432e-06,
|
| 1586 |
+
"loss": 0.0197,
|
| 1587 |
+
"step": 2976
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 4.7872,
|
| 1591 |
+
"grad_norm": 2.259871244430542,
|
| 1592 |
+
"learning_rate": 2.1760000000000004e-06,
|
| 1593 |
+
"loss": 0.0044,
|
| 1594 |
+
"step": 2992
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 4.8128,
|
| 1598 |
+
"grad_norm": 24.58582305908203,
|
| 1599 |
+
"learning_rate": 1.92e-06,
|
| 1600 |
+
"loss": 0.0173,
|
| 1601 |
+
"step": 3008
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 4.8384,
|
| 1605 |
+
"grad_norm": 18.684513092041016,
|
| 1606 |
+
"learning_rate": 1.6639999999999999e-06,
|
| 1607 |
+
"loss": 0.0017,
|
| 1608 |
+
"step": 3024
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 4.864,
|
| 1612 |
+
"grad_norm": 0.005354017019271851,
|
| 1613 |
+
"learning_rate": 1.4080000000000001e-06,
|
| 1614 |
+
"loss": 0.0033,
|
| 1615 |
+
"step": 3040
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 4.8896,
|
| 1619 |
+
"grad_norm": 5.149356365203857,
|
| 1620 |
+
"learning_rate": 1.1520000000000002e-06,
|
| 1621 |
+
"loss": 0.0138,
|
| 1622 |
+
"step": 3056
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 4.9152000000000005,
|
| 1626 |
+
"grad_norm": 0.019161969423294067,
|
| 1627 |
+
"learning_rate": 8.96e-07,
|
| 1628 |
+
"loss": 0.0011,
|
| 1629 |
+
"step": 3072
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 4.9152000000000005,
|
| 1633 |
+
"eval_accuracy": 0.9781,
|
| 1634 |
+
"eval_f1": 0.9780509764198836,
|
| 1635 |
+
"eval_loss": 0.13070251047611237,
|
| 1636 |
+
"eval_precision": 0.9782672483328131,
|
| 1637 |
+
"eval_recall": 0.9781,
|
| 1638 |
+
"eval_runtime": 276.4215,
|
| 1639 |
+
"eval_samples_per_second": 72.353,
|
| 1640 |
+
"eval_steps_per_second": 9.044,
|
| 1641 |
+
"step": 3072
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 4.9408,
|
| 1645 |
+
"grad_norm": 6.959413528442383,
|
| 1646 |
+
"learning_rate": 6.4e-07,
|
| 1647 |
+
"loss": 0.01,
|
| 1648 |
+
"step": 3088
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 4.9664,
|
| 1652 |
+
"grad_norm": 26.619112014770508,
|
| 1653 |
+
"learning_rate": 3.8400000000000005e-07,
|
| 1654 |
+
"loss": 0.008,
|
| 1655 |
+
"step": 3104
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 4.992,
|
| 1659 |
+
"grad_norm": 2.544370412826538,
|
| 1660 |
+
"learning_rate": 1.28e-07,
|
| 1661 |
+
"loss": 0.0028,
|
| 1662 |
+
"step": 3120
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 5.0,
|
| 1666 |
+
"step": 3125,
|
| 1667 |
+
"total_flos": 1.9922182705152e+18,
|
| 1668 |
+
"train_loss": 0.09219677979975939,
|
| 1669 |
+
"train_runtime": 12464.8013,
|
| 1670 |
+
"train_samples_per_second": 32.09,
|
| 1671 |
+
"train_steps_per_second": 0.251
|
| 1672 |
+
}
|
| 1673 |
+
],
|
| 1674 |
+
"logging_steps": 16,
|
| 1675 |
+
"max_steps": 3125,
|
| 1676 |
+
"num_input_tokens_seen": 0,
|
| 1677 |
+
"num_train_epochs": 5,
|
| 1678 |
+
"save_steps": 128,
|
| 1679 |
+
"stateful_callbacks": {
|
| 1680 |
+
"TrainerControl": {
|
| 1681 |
+
"args": {
|
| 1682 |
+
"should_epoch_stop": false,
|
| 1683 |
+
"should_evaluate": false,
|
| 1684 |
+
"should_log": false,
|
| 1685 |
+
"should_save": true,
|
| 1686 |
+
"should_training_stop": true
|
| 1687 |
+
},
|
| 1688 |
+
"attributes": {}
|
| 1689 |
+
}
|
| 1690 |
+
},
|
| 1691 |
+
"total_flos": 1.9922182705152e+18,
|
| 1692 |
+
"train_batch_size": 128,
|
| 1693 |
+
"trial_name": null,
|
| 1694 |
+
"trial_params": null
|
| 1695 |
+
}
|