End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +1285 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9323394495412844,
|
| 4 |
+
"eval_loss": 0.19522710144519806,
|
| 5 |
+
"eval_runtime": 2.4841,
|
| 6 |
+
"eval_samples": 872,
|
| 7 |
+
"eval_samples_per_second": 351.035,
|
| 8 |
+
"eval_steps_per_second": 43.879,
|
| 9 |
+
"train_loss": 0.10745611605338416,
|
| 10 |
+
"train_runtime": 8358.8854,
|
| 11 |
+
"train_samples": 67349,
|
| 12 |
+
"train_samples_per_second": 80.572,
|
| 13 |
+
"train_steps_per_second": 5.037
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9323394495412844,
|
| 4 |
+
"eval_loss": 0.19522710144519806,
|
| 5 |
+
"eval_runtime": 2.4841,
|
| 6 |
+
"eval_samples": 872,
|
| 7 |
+
"eval_samples_per_second": 351.035,
|
| 8 |
+
"eval_steps_per_second": 43.879
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 0.10745611605338416,
|
| 4 |
+
"train_runtime": 8358.8854,
|
| 5 |
+
"train_samples": 67349,
|
| 6 |
+
"train_samples_per_second": 80.572,
|
| 7 |
+
"train_steps_per_second": 5.037
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1285 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.19522710144519806,
|
| 3 |
+
"best_model_checkpoint": "./results_train/roberta-base/sst2/checkpoint-3500",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 42100,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.12,
|
| 12 |
+
"learning_rate": 3.9588281868566905e-06,
|
| 13 |
+
"loss": 0.575,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.12,
|
| 18 |
+
"eval_accuracy": 0.9071100917431193,
|
| 19 |
+
"eval_loss": 0.2664913535118103,
|
| 20 |
+
"eval_runtime": 2.4433,
|
| 21 |
+
"eval_samples_per_second": 356.901,
|
| 22 |
+
"eval_steps_per_second": 44.613,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.24,
|
| 27 |
+
"learning_rate": 7.917656373713381e-06,
|
| 28 |
+
"loss": 0.2989,
|
| 29 |
+
"step": 1000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.24,
|
| 33 |
+
"eval_accuracy": 0.9220183486238532,
|
| 34 |
+
"eval_loss": 0.20883557200431824,
|
| 35 |
+
"eval_runtime": 2.4454,
|
| 36 |
+
"eval_samples_per_second": 356.584,
|
| 37 |
+
"eval_steps_per_second": 44.573,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.36,
|
| 42 |
+
"learning_rate": 1.1876484560570072e-05,
|
| 43 |
+
"loss": 0.2725,
|
| 44 |
+
"step": 1500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.36,
|
| 48 |
+
"eval_accuracy": 0.9243119266055045,
|
| 49 |
+
"eval_loss": 0.25596883893013,
|
| 50 |
+
"eval_runtime": 2.451,
|
| 51 |
+
"eval_samples_per_second": 355.775,
|
| 52 |
+
"eval_steps_per_second": 44.472,
|
| 53 |
+
"step": 1500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.48,
|
| 57 |
+
"learning_rate": 1.5835312747426762e-05,
|
| 58 |
+
"loss": 0.2814,
|
| 59 |
+
"step": 2000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.48,
|
| 63 |
+
"eval_accuracy": 0.926605504587156,
|
| 64 |
+
"eval_loss": 0.20158442854881287,
|
| 65 |
+
"eval_runtime": 2.462,
|
| 66 |
+
"eval_samples_per_second": 354.188,
|
| 67 |
+
"eval_steps_per_second": 44.274,
|
| 68 |
+
"step": 2000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.59,
|
| 72 |
+
"learning_rate": 1.9794140934283453e-05,
|
| 73 |
+
"loss": 0.2586,
|
| 74 |
+
"step": 2500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.59,
|
| 78 |
+
"eval_accuracy": 0.9174311926605505,
|
| 79 |
+
"eval_loss": 0.22930225729942322,
|
| 80 |
+
"eval_runtime": 2.4517,
|
| 81 |
+
"eval_samples_per_second": 355.671,
|
| 82 |
+
"eval_steps_per_second": 44.459,
|
| 83 |
+
"step": 2500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.71,
|
| 87 |
+
"learning_rate": 1.9760448779501697e-05,
|
| 88 |
+
"loss": 0.2536,
|
| 89 |
+
"step": 3000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.71,
|
| 93 |
+
"eval_accuracy": 0.9323394495412844,
|
| 94 |
+
"eval_loss": 0.23396578431129456,
|
| 95 |
+
"eval_runtime": 2.4584,
|
| 96 |
+
"eval_samples_per_second": 354.697,
|
| 97 |
+
"eval_steps_per_second": 44.337,
|
| 98 |
+
"step": 3000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.83,
|
| 102 |
+
"learning_rate": 1.95077576186385e-05,
|
| 103 |
+
"loss": 0.2494,
|
| 104 |
+
"step": 3500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.83,
|
| 108 |
+
"eval_accuracy": 0.9323394495412844,
|
| 109 |
+
"eval_loss": 0.19522710144519806,
|
| 110 |
+
"eval_runtime": 2.4521,
|
| 111 |
+
"eval_samples_per_second": 355.616,
|
| 112 |
+
"eval_steps_per_second": 44.452,
|
| 113 |
+
"step": 3500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.95,
|
| 117 |
+
"learning_rate": 1.925506645777531e-05,
|
| 118 |
+
"loss": 0.2396,
|
| 119 |
+
"step": 4000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.95,
|
| 123 |
+
"eval_accuracy": 0.9323394495412844,
|
| 124 |
+
"eval_loss": 0.24936608970165253,
|
| 125 |
+
"eval_runtime": 2.4569,
|
| 126 |
+
"eval_samples_per_second": 354.916,
|
| 127 |
+
"eval_steps_per_second": 44.365,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.07,
|
| 132 |
+
"learning_rate": 1.9002375296912114e-05,
|
| 133 |
+
"loss": 0.2123,
|
| 134 |
+
"step": 4500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.07,
|
| 138 |
+
"eval_accuracy": 0.9380733944954128,
|
| 139 |
+
"eval_loss": 0.21870844066143036,
|
| 140 |
+
"eval_runtime": 2.449,
|
| 141 |
+
"eval_samples_per_second": 356.068,
|
| 142 |
+
"eval_steps_per_second": 44.509,
|
| 143 |
+
"step": 4500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.19,
|
| 147 |
+
"learning_rate": 1.874968413604892e-05,
|
| 148 |
+
"loss": 0.2042,
|
| 149 |
+
"step": 5000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.19,
|
| 153 |
+
"eval_accuracy": 0.9151376146788991,
|
| 154 |
+
"eval_loss": 0.2811821401119232,
|
| 155 |
+
"eval_runtime": 2.4602,
|
| 156 |
+
"eval_samples_per_second": 354.439,
|
| 157 |
+
"eval_steps_per_second": 44.305,
|
| 158 |
+
"step": 5000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.31,
|
| 162 |
+
"learning_rate": 1.849699297518573e-05,
|
| 163 |
+
"loss": 0.2083,
|
| 164 |
+
"step": 5500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.31,
|
| 168 |
+
"eval_accuracy": 0.9346330275229358,
|
| 169 |
+
"eval_loss": 0.27386215329170227,
|
| 170 |
+
"eval_runtime": 2.5255,
|
| 171 |
+
"eval_samples_per_second": 345.272,
|
| 172 |
+
"eval_steps_per_second": 43.159,
|
| 173 |
+
"step": 5500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.43,
|
| 177 |
+
"learning_rate": 1.8244301814322537e-05,
|
| 178 |
+
"loss": 0.2041,
|
| 179 |
+
"step": 6000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.43,
|
| 183 |
+
"eval_accuracy": 0.9380733944954128,
|
| 184 |
+
"eval_loss": 0.20871196687221527,
|
| 185 |
+
"eval_runtime": 2.4547,
|
| 186 |
+
"eval_samples_per_second": 355.241,
|
| 187 |
+
"eval_steps_per_second": 44.405,
|
| 188 |
+
"step": 6000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.54,
|
| 192 |
+
"learning_rate": 1.7991610653459345e-05,
|
| 193 |
+
"loss": 0.1969,
|
| 194 |
+
"step": 6500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.54,
|
| 198 |
+
"eval_accuracy": 0.9254587155963303,
|
| 199 |
+
"eval_loss": 0.25904807448387146,
|
| 200 |
+
"eval_runtime": 2.4532,
|
| 201 |
+
"eval_samples_per_second": 355.448,
|
| 202 |
+
"eval_steps_per_second": 44.431,
|
| 203 |
+
"step": 6500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 1.66,
|
| 207 |
+
"learning_rate": 1.773891949259615e-05,
|
| 208 |
+
"loss": 0.1982,
|
| 209 |
+
"step": 7000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.66,
|
| 213 |
+
"eval_accuracy": 0.930045871559633,
|
| 214 |
+
"eval_loss": 0.2444588840007782,
|
| 215 |
+
"eval_runtime": 2.4545,
|
| 216 |
+
"eval_samples_per_second": 355.268,
|
| 217 |
+
"eval_steps_per_second": 44.409,
|
| 218 |
+
"step": 7000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.78,
|
| 222 |
+
"learning_rate": 1.7486228331732958e-05,
|
| 223 |
+
"loss": 0.1943,
|
| 224 |
+
"step": 7500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.78,
|
| 228 |
+
"eval_accuracy": 0.926605504587156,
|
| 229 |
+
"eval_loss": 0.2798321545124054,
|
| 230 |
+
"eval_runtime": 2.4455,
|
| 231 |
+
"eval_samples_per_second": 356.567,
|
| 232 |
+
"eval_steps_per_second": 44.571,
|
| 233 |
+
"step": 7500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.9,
|
| 237 |
+
"learning_rate": 1.7233537170869766e-05,
|
| 238 |
+
"loss": 0.1848,
|
| 239 |
+
"step": 8000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 1.9,
|
| 243 |
+
"eval_accuracy": 0.9311926605504587,
|
| 244 |
+
"eval_loss": 0.2844010591506958,
|
| 245 |
+
"eval_runtime": 2.4586,
|
| 246 |
+
"eval_samples_per_second": 354.679,
|
| 247 |
+
"eval_steps_per_second": 44.335,
|
| 248 |
+
"step": 8000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.02,
|
| 252 |
+
"learning_rate": 1.698084601000657e-05,
|
| 253 |
+
"loss": 0.1788,
|
| 254 |
+
"step": 8500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.02,
|
| 258 |
+
"eval_accuracy": 0.9254587155963303,
|
| 259 |
+
"eval_loss": 0.2998378872871399,
|
| 260 |
+
"eval_runtime": 2.446,
|
| 261 |
+
"eval_samples_per_second": 356.496,
|
| 262 |
+
"eval_steps_per_second": 44.562,
|
| 263 |
+
"step": 8500
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.14,
|
| 267 |
+
"learning_rate": 1.672815484914338e-05,
|
| 268 |
+
"loss": 0.1623,
|
| 269 |
+
"step": 9000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 2.14,
|
| 273 |
+
"eval_accuracy": 0.9392201834862385,
|
| 274 |
+
"eval_loss": 0.2695905268192291,
|
| 275 |
+
"eval_runtime": 2.4607,
|
| 276 |
+
"eval_samples_per_second": 354.365,
|
| 277 |
+
"eval_steps_per_second": 44.296,
|
| 278 |
+
"step": 9000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.26,
|
| 282 |
+
"learning_rate": 1.6475463688280183e-05,
|
| 283 |
+
"loss": 0.1499,
|
| 284 |
+
"step": 9500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.26,
|
| 288 |
+
"eval_accuracy": 0.9277522935779816,
|
| 289 |
+
"eval_loss": 0.25331878662109375,
|
| 290 |
+
"eval_runtime": 2.4449,
|
| 291 |
+
"eval_samples_per_second": 356.659,
|
| 292 |
+
"eval_steps_per_second": 44.582,
|
| 293 |
+
"step": 9500
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 2.38,
|
| 297 |
+
"learning_rate": 1.622277252741699e-05,
|
| 298 |
+
"loss": 0.1426,
|
| 299 |
+
"step": 10000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.38,
|
| 303 |
+
"eval_accuracy": 0.930045871559633,
|
| 304 |
+
"eval_loss": 0.29705262184143066,
|
| 305 |
+
"eval_runtime": 2.4651,
|
| 306 |
+
"eval_samples_per_second": 353.733,
|
| 307 |
+
"eval_steps_per_second": 44.217,
|
| 308 |
+
"step": 10000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.49,
|
| 312 |
+
"learning_rate": 1.59700813665538e-05,
|
| 313 |
+
"loss": 0.1479,
|
| 314 |
+
"step": 10500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.49,
|
| 318 |
+
"eval_accuracy": 0.9357798165137615,
|
| 319 |
+
"eval_loss": 0.25958266854286194,
|
| 320 |
+
"eval_runtime": 2.4502,
|
| 321 |
+
"eval_samples_per_second": 355.883,
|
| 322 |
+
"eval_steps_per_second": 44.485,
|
| 323 |
+
"step": 10500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 2.61,
|
| 327 |
+
"learning_rate": 1.5717390205690607e-05,
|
| 328 |
+
"loss": 0.1405,
|
| 329 |
+
"step": 11000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.61,
|
| 333 |
+
"eval_accuracy": 0.9254587155963303,
|
| 334 |
+
"eval_loss": 0.2944609522819519,
|
| 335 |
+
"eval_runtime": 2.4554,
|
| 336 |
+
"eval_samples_per_second": 355.141,
|
| 337 |
+
"eval_steps_per_second": 44.393,
|
| 338 |
+
"step": 11000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.73,
|
| 342 |
+
"learning_rate": 1.5464699044827415e-05,
|
| 343 |
+
"loss": 0.1577,
|
| 344 |
+
"step": 11500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 2.73,
|
| 348 |
+
"eval_accuracy": 0.9002293577981652,
|
| 349 |
+
"eval_loss": 0.40612396597862244,
|
| 350 |
+
"eval_runtime": 2.4539,
|
| 351 |
+
"eval_samples_per_second": 355.36,
|
| 352 |
+
"eval_steps_per_second": 44.42,
|
| 353 |
+
"step": 11500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 2.85,
|
| 357 |
+
"learning_rate": 1.521200788396422e-05,
|
| 358 |
+
"loss": 0.1521,
|
| 359 |
+
"step": 12000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 2.85,
|
| 363 |
+
"eval_accuracy": 0.9334862385321101,
|
| 364 |
+
"eval_loss": 0.2724354565143585,
|
| 365 |
+
"eval_runtime": 2.4461,
|
| 366 |
+
"eval_samples_per_second": 356.483,
|
| 367 |
+
"eval_steps_per_second": 44.56,
|
| 368 |
+
"step": 12000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.97,
|
| 372 |
+
"learning_rate": 1.4959316723101027e-05,
|
| 373 |
+
"loss": 0.1426,
|
| 374 |
+
"step": 12500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 2.97,
|
| 378 |
+
"eval_accuracy": 0.9426605504587156,
|
| 379 |
+
"eval_loss": 0.27123740315437317,
|
| 380 |
+
"eval_runtime": 2.4449,
|
| 381 |
+
"eval_samples_per_second": 356.655,
|
| 382 |
+
"eval_steps_per_second": 44.582,
|
| 383 |
+
"step": 12500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 3.09,
|
| 387 |
+
"learning_rate": 1.4706625562237835e-05,
|
| 388 |
+
"loss": 0.1206,
|
| 389 |
+
"step": 13000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 3.09,
|
| 393 |
+
"eval_accuracy": 0.9357798165137615,
|
| 394 |
+
"eval_loss": 0.2954227328300476,
|
| 395 |
+
"eval_runtime": 2.467,
|
| 396 |
+
"eval_samples_per_second": 353.464,
|
| 397 |
+
"eval_steps_per_second": 44.183,
|
| 398 |
+
"step": 13000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 3.21,
|
| 402 |
+
"learning_rate": 1.4453934401374641e-05,
|
| 403 |
+
"loss": 0.1074,
|
| 404 |
+
"step": 13500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 3.21,
|
| 408 |
+
"eval_accuracy": 0.9392201834862385,
|
| 409 |
+
"eval_loss": 0.2653304934501648,
|
| 410 |
+
"eval_runtime": 2.4486,
|
| 411 |
+
"eval_samples_per_second": 356.118,
|
| 412 |
+
"eval_steps_per_second": 44.515,
|
| 413 |
+
"step": 13500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 3.33,
|
| 417 |
+
"learning_rate": 1.420124324051145e-05,
|
| 418 |
+
"loss": 0.112,
|
| 419 |
+
"step": 14000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 3.33,
|
| 423 |
+
"eval_accuracy": 0.9346330275229358,
|
| 424 |
+
"eval_loss": 0.2777578830718994,
|
| 425 |
+
"eval_runtime": 2.4566,
|
| 426 |
+
"eval_samples_per_second": 354.969,
|
| 427 |
+
"eval_steps_per_second": 44.371,
|
| 428 |
+
"step": 14000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 3.44,
|
| 432 |
+
"learning_rate": 1.3948552079648254e-05,
|
| 433 |
+
"loss": 0.1147,
|
| 434 |
+
"step": 14500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 3.44,
|
| 438 |
+
"eval_accuracy": 0.9311926605504587,
|
| 439 |
+
"eval_loss": 0.3704558312892914,
|
| 440 |
+
"eval_runtime": 2.4454,
|
| 441 |
+
"eval_samples_per_second": 356.589,
|
| 442 |
+
"eval_steps_per_second": 44.574,
|
| 443 |
+
"step": 14500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 3.56,
|
| 447 |
+
"learning_rate": 1.3695860918785062e-05,
|
| 448 |
+
"loss": 0.1196,
|
| 449 |
+
"step": 15000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 3.56,
|
| 453 |
+
"eval_accuracy": 0.9346330275229358,
|
| 454 |
+
"eval_loss": 0.2889645993709564,
|
| 455 |
+
"eval_runtime": 2.4563,
|
| 456 |
+
"eval_samples_per_second": 354.999,
|
| 457 |
+
"eval_steps_per_second": 44.375,
|
| 458 |
+
"step": 15000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 3.68,
|
| 462 |
+
"learning_rate": 1.344316975792187e-05,
|
| 463 |
+
"loss": 0.1159,
|
| 464 |
+
"step": 15500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 3.68,
|
| 468 |
+
"eval_accuracy": 0.926605504587156,
|
| 469 |
+
"eval_loss": 0.3448694944381714,
|
| 470 |
+
"eval_runtime": 2.4429,
|
| 471 |
+
"eval_samples_per_second": 356.949,
|
| 472 |
+
"eval_steps_per_second": 44.619,
|
| 473 |
+
"step": 15500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 3.8,
|
| 477 |
+
"learning_rate": 1.3190478597058676e-05,
|
| 478 |
+
"loss": 0.119,
|
| 479 |
+
"step": 16000
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 3.8,
|
| 483 |
+
"eval_accuracy": 0.9334862385321101,
|
| 484 |
+
"eval_loss": 0.3207152187824249,
|
| 485 |
+
"eval_runtime": 2.461,
|
| 486 |
+
"eval_samples_per_second": 354.323,
|
| 487 |
+
"eval_steps_per_second": 44.29,
|
| 488 |
+
"step": 16000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 3.92,
|
| 492 |
+
"learning_rate": 1.2937787436195484e-05,
|
| 493 |
+
"loss": 0.1268,
|
| 494 |
+
"step": 16500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 3.92,
|
| 498 |
+
"eval_accuracy": 0.9311926605504587,
|
| 499 |
+
"eval_loss": 0.3234628736972809,
|
| 500 |
+
"eval_runtime": 2.4504,
|
| 501 |
+
"eval_samples_per_second": 355.858,
|
| 502 |
+
"eval_steps_per_second": 44.482,
|
| 503 |
+
"step": 16500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 4.04,
|
| 507 |
+
"learning_rate": 1.2685096275332289e-05,
|
| 508 |
+
"loss": 0.1074,
|
| 509 |
+
"step": 17000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 4.04,
|
| 513 |
+
"eval_accuracy": 0.9334862385321101,
|
| 514 |
+
"eval_loss": 0.3650290369987488,
|
| 515 |
+
"eval_runtime": 2.456,
|
| 516 |
+
"eval_samples_per_second": 355.052,
|
| 517 |
+
"eval_steps_per_second": 44.382,
|
| 518 |
+
"step": 17000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 4.16,
|
| 522 |
+
"learning_rate": 1.2432405114469096e-05,
|
| 523 |
+
"loss": 0.0805,
|
| 524 |
+
"step": 17500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 4.16,
|
| 528 |
+
"eval_accuracy": 0.9380733944954128,
|
| 529 |
+
"eval_loss": 0.33378419280052185,
|
| 530 |
+
"eval_runtime": 2.4457,
|
| 531 |
+
"eval_samples_per_second": 356.538,
|
| 532 |
+
"eval_steps_per_second": 44.567,
|
| 533 |
+
"step": 17500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 4.28,
|
| 537 |
+
"learning_rate": 1.2179713953605903e-05,
|
| 538 |
+
"loss": 0.0838,
|
| 539 |
+
"step": 18000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 4.28,
|
| 543 |
+
"eval_accuracy": 0.9208715596330275,
|
| 544 |
+
"eval_loss": 0.4302394688129425,
|
| 545 |
+
"eval_runtime": 2.4587,
|
| 546 |
+
"eval_samples_per_second": 354.661,
|
| 547 |
+
"eval_steps_per_second": 44.333,
|
| 548 |
+
"step": 18000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 4.39,
|
| 552 |
+
"learning_rate": 1.192702279274271e-05,
|
| 553 |
+
"loss": 0.0848,
|
| 554 |
+
"step": 18500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 4.39,
|
| 558 |
+
"eval_accuracy": 0.9323394495412844,
|
| 559 |
+
"eval_loss": 0.40956971049308777,
|
| 560 |
+
"eval_runtime": 2.4483,
|
| 561 |
+
"eval_samples_per_second": 356.162,
|
| 562 |
+
"eval_steps_per_second": 44.52,
|
| 563 |
+
"step": 18500
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 4.51,
|
| 567 |
+
"learning_rate": 1.1674331631879519e-05,
|
| 568 |
+
"loss": 0.0922,
|
| 569 |
+
"step": 19000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 4.51,
|
| 573 |
+
"eval_accuracy": 0.9369266055045872,
|
| 574 |
+
"eval_loss": 0.3332035541534424,
|
| 575 |
+
"eval_runtime": 2.4597,
|
| 576 |
+
"eval_samples_per_second": 354.511,
|
| 577 |
+
"eval_steps_per_second": 44.314,
|
| 578 |
+
"step": 19000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 4.63,
|
| 582 |
+
"learning_rate": 1.1421640471016325e-05,
|
| 583 |
+
"loss": 0.091,
|
| 584 |
+
"step": 19500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 4.63,
|
| 588 |
+
"eval_accuracy": 0.9438073394495413,
|
| 589 |
+
"eval_loss": 0.3024330735206604,
|
| 590 |
+
"eval_runtime": 2.4457,
|
| 591 |
+
"eval_samples_per_second": 356.542,
|
| 592 |
+
"eval_steps_per_second": 44.568,
|
| 593 |
+
"step": 19500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 4.75,
|
| 597 |
+
"learning_rate": 1.1168949310153133e-05,
|
| 598 |
+
"loss": 0.0977,
|
| 599 |
+
"step": 20000
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 4.75,
|
| 603 |
+
"eval_accuracy": 0.9495412844036697,
|
| 604 |
+
"eval_loss": 0.2673788070678711,
|
| 605 |
+
"eval_runtime": 2.4587,
|
| 606 |
+
"eval_samples_per_second": 354.654,
|
| 607 |
+
"eval_steps_per_second": 44.332,
|
| 608 |
+
"step": 20000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 4.87,
|
| 612 |
+
"learning_rate": 1.0916258149289937e-05,
|
| 613 |
+
"loss": 0.0897,
|
| 614 |
+
"step": 20500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 4.87,
|
| 618 |
+
"eval_accuracy": 0.930045871559633,
|
| 619 |
+
"eval_loss": 0.39930590987205505,
|
| 620 |
+
"eval_runtime": 2.4473,
|
| 621 |
+
"eval_samples_per_second": 356.313,
|
| 622 |
+
"eval_steps_per_second": 44.539,
|
| 623 |
+
"step": 20500
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 4.99,
|
| 627 |
+
"learning_rate": 1.0663566988426745e-05,
|
| 628 |
+
"loss": 0.1013,
|
| 629 |
+
"step": 21000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 4.99,
|
| 633 |
+
"eval_accuracy": 0.9288990825688074,
|
| 634 |
+
"eval_loss": 0.322666198015213,
|
| 635 |
+
"eval_runtime": 2.4496,
|
| 636 |
+
"eval_samples_per_second": 355.981,
|
| 637 |
+
"eval_steps_per_second": 44.498,
|
| 638 |
+
"step": 21000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 5.11,
|
| 642 |
+
"learning_rate": 1.0410875827563553e-05,
|
| 643 |
+
"loss": 0.0671,
|
| 644 |
+
"step": 21500
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 5.11,
|
| 648 |
+
"eval_accuracy": 0.9426605504587156,
|
| 649 |
+
"eval_loss": 0.3374435603618622,
|
| 650 |
+
"eval_runtime": 2.4457,
|
| 651 |
+
"eval_samples_per_second": 356.54,
|
| 652 |
+
"eval_steps_per_second": 44.567,
|
| 653 |
+
"step": 21500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 5.23,
|
| 657 |
+
"learning_rate": 1.015818466670036e-05,
|
| 658 |
+
"loss": 0.0671,
|
| 659 |
+
"step": 22000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 5.23,
|
| 663 |
+
"eval_accuracy": 0.9277522935779816,
|
| 664 |
+
"eval_loss": 0.4108366072177887,
|
| 665 |
+
"eval_runtime": 2.4551,
|
| 666 |
+
"eval_samples_per_second": 355.179,
|
| 667 |
+
"eval_steps_per_second": 44.397,
|
| 668 |
+
"step": 22000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 5.34,
|
| 672 |
+
"learning_rate": 9.905493505837167e-06,
|
| 673 |
+
"loss": 0.0652,
|
| 674 |
+
"step": 22500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 5.34,
|
| 678 |
+
"eval_accuracy": 0.9380733944954128,
|
| 679 |
+
"eval_loss": 0.3549734652042389,
|
| 680 |
+
"eval_runtime": 2.4475,
|
| 681 |
+
"eval_samples_per_second": 356.289,
|
| 682 |
+
"eval_steps_per_second": 44.536,
|
| 683 |
+
"step": 22500
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 5.46,
|
| 687 |
+
"learning_rate": 9.652802344973974e-06,
|
| 688 |
+
"loss": 0.0664,
|
| 689 |
+
"step": 23000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 5.46,
|
| 693 |
+
"eval_accuracy": 0.9357798165137615,
|
| 694 |
+
"eval_loss": 0.339821994304657,
|
| 695 |
+
"eval_runtime": 2.4559,
|
| 696 |
+
"eval_samples_per_second": 355.062,
|
| 697 |
+
"eval_steps_per_second": 44.383,
|
| 698 |
+
"step": 23000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 5.58,
|
| 702 |
+
"learning_rate": 9.40011118411078e-06,
|
| 703 |
+
"loss": 0.0742,
|
| 704 |
+
"step": 23500
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 5.58,
|
| 708 |
+
"eval_accuracy": 0.9380733944954128,
|
| 709 |
+
"eval_loss": 0.3286002278327942,
|
| 710 |
+
"eval_runtime": 2.4471,
|
| 711 |
+
"eval_samples_per_second": 356.342,
|
| 712 |
+
"eval_steps_per_second": 44.543,
|
| 713 |
+
"step": 23500
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 5.7,
|
| 717 |
+
"learning_rate": 9.147420023247588e-06,
|
| 718 |
+
"loss": 0.0758,
|
| 719 |
+
"step": 24000
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 5.7,
|
| 723 |
+
"eval_accuracy": 0.9311926605504587,
|
| 724 |
+
"eval_loss": 0.32764118909835815,
|
| 725 |
+
"eval_runtime": 2.4639,
|
| 726 |
+
"eval_samples_per_second": 353.904,
|
| 727 |
+
"eval_steps_per_second": 44.238,
|
| 728 |
+
"step": 24000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 5.82,
|
| 732 |
+
"learning_rate": 8.894728862384394e-06,
|
| 733 |
+
"loss": 0.075,
|
| 734 |
+
"step": 24500
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 5.82,
|
| 738 |
+
"eval_accuracy": 0.9369266055045872,
|
| 739 |
+
"eval_loss": 0.32022935152053833,
|
| 740 |
+
"eval_runtime": 2.4503,
|
| 741 |
+
"eval_samples_per_second": 355.874,
|
| 742 |
+
"eval_steps_per_second": 44.484,
|
| 743 |
+
"step": 24500
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 5.94,
|
| 747 |
+
"learning_rate": 8.642037701521202e-06,
|
| 748 |
+
"loss": 0.0686,
|
| 749 |
+
"step": 25000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 5.94,
|
| 753 |
+
"eval_accuracy": 0.9415137614678899,
|
| 754 |
+
"eval_loss": 0.3481292426586151,
|
| 755 |
+
"eval_runtime": 2.4555,
|
| 756 |
+
"eval_samples_per_second": 355.12,
|
| 757 |
+
"eval_steps_per_second": 44.39,
|
| 758 |
+
"step": 25000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 6.06,
|
| 762 |
+
"learning_rate": 8.389346540658008e-06,
|
| 763 |
+
"loss": 0.0729,
|
| 764 |
+
"step": 25500
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 6.06,
|
| 768 |
+
"eval_accuracy": 0.9334862385321101,
|
| 769 |
+
"eval_loss": 0.38161903619766235,
|
| 770 |
+
"eval_runtime": 2.4476,
|
| 771 |
+
"eval_samples_per_second": 356.27,
|
| 772 |
+
"eval_steps_per_second": 44.534,
|
| 773 |
+
"step": 25500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 6.18,
|
| 777 |
+
"learning_rate": 8.136655379794816e-06,
|
| 778 |
+
"loss": 0.0568,
|
| 779 |
+
"step": 26000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 6.18,
|
| 783 |
+
"eval_accuracy": 0.9380733944954128,
|
| 784 |
+
"eval_loss": 0.31324318051338196,
|
| 785 |
+
"eval_runtime": 2.4707,
|
| 786 |
+
"eval_samples_per_second": 352.935,
|
| 787 |
+
"eval_steps_per_second": 44.117,
|
| 788 |
+
"step": 26000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 6.29,
|
| 792 |
+
"learning_rate": 7.883964218931623e-06,
|
| 793 |
+
"loss": 0.0529,
|
| 794 |
+
"step": 26500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 6.29,
|
| 798 |
+
"eval_accuracy": 0.930045871559633,
|
| 799 |
+
"eval_loss": 0.3756808340549469,
|
| 800 |
+
"eval_runtime": 2.4544,
|
| 801 |
+
"eval_samples_per_second": 355.287,
|
| 802 |
+
"eval_steps_per_second": 44.411,
|
| 803 |
+
"step": 26500
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 6.41,
|
| 807 |
+
"learning_rate": 7.631273058068429e-06,
|
| 808 |
+
"loss": 0.0506,
|
| 809 |
+
"step": 27000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 6.41,
|
| 813 |
+
"eval_accuracy": 0.9380733944954128,
|
| 814 |
+
"eval_loss": 0.33958113193511963,
|
| 815 |
+
"eval_runtime": 2.4531,
|
| 816 |
+
"eval_samples_per_second": 355.471,
|
| 817 |
+
"eval_steps_per_second": 44.434,
|
| 818 |
+
"step": 27000
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 6.53,
|
| 822 |
+
"learning_rate": 7.378581897205236e-06,
|
| 823 |
+
"loss": 0.0476,
|
| 824 |
+
"step": 27500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 6.53,
|
| 828 |
+
"eval_accuracy": 0.9403669724770642,
|
| 829 |
+
"eval_loss": 0.3641544580459595,
|
| 830 |
+
"eval_runtime": 2.4417,
|
| 831 |
+
"eval_samples_per_second": 357.132,
|
| 832 |
+
"eval_steps_per_second": 44.641,
|
| 833 |
+
"step": 27500
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 6.65,
|
| 837 |
+
"learning_rate": 7.125890736342044e-06,
|
| 838 |
+
"loss": 0.0555,
|
| 839 |
+
"step": 28000
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 6.65,
|
| 843 |
+
"eval_accuracy": 0.9403669724770642,
|
| 844 |
+
"eval_loss": 0.34298017621040344,
|
| 845 |
+
"eval_runtime": 2.4463,
|
| 846 |
+
"eval_samples_per_second": 356.452,
|
| 847 |
+
"eval_steps_per_second": 44.556,
|
| 848 |
+
"step": 28000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 6.77,
|
| 852 |
+
"learning_rate": 6.87319957547885e-06,
|
| 853 |
+
"loss": 0.0574,
|
| 854 |
+
"step": 28500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 6.77,
|
| 858 |
+
"eval_accuracy": 0.9392201834862385,
|
| 859 |
+
"eval_loss": 0.3401435613632202,
|
| 860 |
+
"eval_runtime": 2.4439,
|
| 861 |
+
"eval_samples_per_second": 356.811,
|
| 862 |
+
"eval_steps_per_second": 44.601,
|
| 863 |
+
"step": 28500
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 6.89,
|
| 867 |
+
"learning_rate": 6.620508414615657e-06,
|
| 868 |
+
"loss": 0.0524,
|
| 869 |
+
"step": 29000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 6.89,
|
| 873 |
+
"eval_accuracy": 0.9346330275229358,
|
| 874 |
+
"eval_loss": 0.33783158659935,
|
| 875 |
+
"eval_runtime": 2.4521,
|
| 876 |
+
"eval_samples_per_second": 355.616,
|
| 877 |
+
"eval_steps_per_second": 44.452,
|
| 878 |
+
"step": 29000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 7.01,
|
| 882 |
+
"learning_rate": 6.367817253752464e-06,
|
| 883 |
+
"loss": 0.0492,
|
| 884 |
+
"step": 29500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 7.01,
|
| 888 |
+
"eval_accuracy": 0.9380733944954128,
|
| 889 |
+
"eval_loss": 0.3833492398262024,
|
| 890 |
+
"eval_runtime": 2.4457,
|
| 891 |
+
"eval_samples_per_second": 356.538,
|
| 892 |
+
"eval_steps_per_second": 44.567,
|
| 893 |
+
"step": 29500
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 7.13,
|
| 897 |
+
"learning_rate": 6.1151260928892706e-06,
|
| 898 |
+
"loss": 0.039,
|
| 899 |
+
"step": 30000
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 7.13,
|
| 903 |
+
"eval_accuracy": 0.9346330275229358,
|
| 904 |
+
"eval_loss": 0.3346712589263916,
|
| 905 |
+
"eval_runtime": 2.4434,
|
| 906 |
+
"eval_samples_per_second": 356.873,
|
| 907 |
+
"eval_steps_per_second": 44.609,
|
| 908 |
+
"step": 30000
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 7.24,
|
| 912 |
+
"learning_rate": 5.8624349320260785e-06,
|
| 913 |
+
"loss": 0.0411,
|
| 914 |
+
"step": 30500
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 7.24,
|
| 918 |
+
"eval_accuracy": 0.9334862385321101,
|
| 919 |
+
"eval_loss": 0.4404141902923584,
|
| 920 |
+
"eval_runtime": 2.4419,
|
| 921 |
+
"eval_samples_per_second": 357.102,
|
| 922 |
+
"eval_steps_per_second": 44.638,
|
| 923 |
+
"step": 30500
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 7.36,
|
| 927 |
+
"learning_rate": 5.609743771162886e-06,
|
| 928 |
+
"loss": 0.0412,
|
| 929 |
+
"step": 31000
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 7.36,
|
| 933 |
+
"eval_accuracy": 0.9380733944954128,
|
| 934 |
+
"eval_loss": 0.36179476976394653,
|
| 935 |
+
"eval_runtime": 2.4414,
|
| 936 |
+
"eval_samples_per_second": 357.173,
|
| 937 |
+
"eval_steps_per_second": 44.647,
|
| 938 |
+
"step": 31000
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 7.48,
|
| 942 |
+
"learning_rate": 5.357052610299692e-06,
|
| 943 |
+
"loss": 0.0477,
|
| 944 |
+
"step": 31500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 7.48,
|
| 948 |
+
"eval_accuracy": 0.9380733944954128,
|
| 949 |
+
"eval_loss": 0.3806387484073639,
|
| 950 |
+
"eval_runtime": 2.4471,
|
| 951 |
+
"eval_samples_per_second": 356.337,
|
| 952 |
+
"eval_steps_per_second": 44.542,
|
| 953 |
+
"step": 31500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 7.6,
|
| 957 |
+
"learning_rate": 5.104361449436499e-06,
|
| 958 |
+
"loss": 0.0435,
|
| 959 |
+
"step": 32000
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 7.6,
|
| 963 |
+
"eval_accuracy": 0.9334862385321101,
|
| 964 |
+
"eval_loss": 0.39115917682647705,
|
| 965 |
+
"eval_runtime": 2.4665,
|
| 966 |
+
"eval_samples_per_second": 353.536,
|
| 967 |
+
"eval_steps_per_second": 44.192,
|
| 968 |
+
"step": 32000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 7.72,
|
| 972 |
+
"learning_rate": 4.851670288573306e-06,
|
| 973 |
+
"loss": 0.0443,
|
| 974 |
+
"step": 32500
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 7.72,
|
| 978 |
+
"eval_accuracy": 0.9392201834862385,
|
| 979 |
+
"eval_loss": 0.39003145694732666,
|
| 980 |
+
"eval_runtime": 2.4534,
|
| 981 |
+
"eval_samples_per_second": 355.426,
|
| 982 |
+
"eval_steps_per_second": 44.428,
|
| 983 |
+
"step": 32500
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 7.84,
|
| 987 |
+
"learning_rate": 4.598979127710113e-06,
|
| 988 |
+
"loss": 0.0421,
|
| 989 |
+
"step": 33000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 7.84,
|
| 993 |
+
"eval_accuracy": 0.9369266055045872,
|
| 994 |
+
"eval_loss": 0.4152164161205292,
|
| 995 |
+
"eval_runtime": 2.4525,
|
| 996 |
+
"eval_samples_per_second": 355.556,
|
| 997 |
+
"eval_steps_per_second": 44.445,
|
| 998 |
+
"step": 33000
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 7.96,
|
| 1002 |
+
"learning_rate": 4.34628796684692e-06,
|
| 1003 |
+
"loss": 0.0495,
|
| 1004 |
+
"step": 33500
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 7.96,
|
| 1008 |
+
"eval_accuracy": 0.9288990825688074,
|
| 1009 |
+
"eval_loss": 0.3831779360771179,
|
| 1010 |
+
"eval_runtime": 2.447,
|
| 1011 |
+
"eval_samples_per_second": 356.361,
|
| 1012 |
+
"eval_steps_per_second": 44.545,
|
| 1013 |
+
"step": 33500
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 8.08,
|
| 1017 |
+
"learning_rate": 4.093596805983727e-06,
|
| 1018 |
+
"loss": 0.0293,
|
| 1019 |
+
"step": 34000
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 8.08,
|
| 1023 |
+
"eval_accuracy": 0.9346330275229358,
|
| 1024 |
+
"eval_loss": 0.44268128275871277,
|
| 1025 |
+
"eval_runtime": 2.4587,
|
| 1026 |
+
"eval_samples_per_second": 354.661,
|
| 1027 |
+
"eval_steps_per_second": 44.333,
|
| 1028 |
+
"step": 34000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 8.19,
|
| 1032 |
+
"learning_rate": 3.840905645120534e-06,
|
| 1033 |
+
"loss": 0.0253,
|
| 1034 |
+
"step": 34500
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 8.19,
|
| 1038 |
+
"eval_accuracy": 0.9380733944954128,
|
| 1039 |
+
"eval_loss": 0.44246163964271545,
|
| 1040 |
+
"eval_runtime": 2.4427,
|
| 1041 |
+
"eval_samples_per_second": 356.983,
|
| 1042 |
+
"eval_steps_per_second": 44.623,
|
| 1043 |
+
"step": 34500
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 8.31,
|
| 1047 |
+
"learning_rate": 3.5882144842573407e-06,
|
| 1048 |
+
"loss": 0.0407,
|
| 1049 |
+
"step": 35000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 8.31,
|
| 1053 |
+
"eval_accuracy": 0.9357798165137615,
|
| 1054 |
+
"eval_loss": 0.41019341349601746,
|
| 1055 |
+
"eval_runtime": 2.453,
|
| 1056 |
+
"eval_samples_per_second": 355.477,
|
| 1057 |
+
"eval_steps_per_second": 44.435,
|
| 1058 |
+
"step": 35000
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 8.43,
|
| 1062 |
+
"learning_rate": 3.3355233233941482e-06,
|
| 1063 |
+
"loss": 0.0311,
|
| 1064 |
+
"step": 35500
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 8.43,
|
| 1068 |
+
"eval_accuracy": 0.9369266055045872,
|
| 1069 |
+
"eval_loss": 0.44467687606811523,
|
| 1070 |
+
"eval_runtime": 2.4425,
|
| 1071 |
+
"eval_samples_per_second": 357.013,
|
| 1072 |
+
"eval_steps_per_second": 44.627,
|
| 1073 |
+
"step": 35500
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 8.55,
|
| 1077 |
+
"learning_rate": 3.082832162530955e-06,
|
| 1078 |
+
"loss": 0.0291,
|
| 1079 |
+
"step": 36000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 8.55,
|
| 1083 |
+
"eval_accuracy": 0.9346330275229358,
|
| 1084 |
+
"eval_loss": 0.46120545268058777,
|
| 1085 |
+
"eval_runtime": 2.4514,
|
| 1086 |
+
"eval_samples_per_second": 355.714,
|
| 1087 |
+
"eval_steps_per_second": 44.464,
|
| 1088 |
+
"step": 36000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 8.67,
|
| 1092 |
+
"learning_rate": 2.8301410016677616e-06,
|
| 1093 |
+
"loss": 0.035,
|
| 1094 |
+
"step": 36500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 8.67,
|
| 1098 |
+
"eval_accuracy": 0.9346330275229358,
|
| 1099 |
+
"eval_loss": 0.4240852892398834,
|
| 1100 |
+
"eval_runtime": 2.4477,
|
| 1101 |
+
"eval_samples_per_second": 356.249,
|
| 1102 |
+
"eval_steps_per_second": 44.531,
|
| 1103 |
+
"step": 36500
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 8.79,
|
| 1107 |
+
"learning_rate": 2.577449840804569e-06,
|
| 1108 |
+
"loss": 0.0381,
|
| 1109 |
+
"step": 37000
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 8.79,
|
| 1113 |
+
"eval_accuracy": 0.9311926605504587,
|
| 1114 |
+
"eval_loss": 0.41976186633110046,
|
| 1115 |
+
"eval_runtime": 2.4523,
|
| 1116 |
+
"eval_samples_per_second": 355.586,
|
| 1117 |
+
"eval_steps_per_second": 44.448,
|
| 1118 |
+
"step": 37000
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 8.91,
|
| 1122 |
+
"learning_rate": 2.3247586799413758e-06,
|
| 1123 |
+
"loss": 0.0234,
|
| 1124 |
+
"step": 37500
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 8.91,
|
| 1128 |
+
"eval_accuracy": 0.9369266055045872,
|
| 1129 |
+
"eval_loss": 0.4344768822193146,
|
| 1130 |
+
"eval_runtime": 2.4469,
|
| 1131 |
+
"eval_samples_per_second": 356.366,
|
| 1132 |
+
"eval_steps_per_second": 44.546,
|
| 1133 |
+
"step": 37500
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 9.03,
|
| 1137 |
+
"learning_rate": 2.072067519078183e-06,
|
| 1138 |
+
"loss": 0.0311,
|
| 1139 |
+
"step": 38000
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 9.03,
|
| 1143 |
+
"eval_accuracy": 0.9311926605504587,
|
| 1144 |
+
"eval_loss": 0.45580777525901794,
|
| 1145 |
+
"eval_runtime": 2.4545,
|
| 1146 |
+
"eval_samples_per_second": 355.27,
|
| 1147 |
+
"eval_steps_per_second": 44.409,
|
| 1148 |
+
"step": 38000
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 9.14,
|
| 1152 |
+
"learning_rate": 1.8193763582149898e-06,
|
| 1153 |
+
"loss": 0.028,
|
| 1154 |
+
"step": 38500
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 9.14,
|
| 1158 |
+
"eval_accuracy": 0.9380733944954128,
|
| 1159 |
+
"eval_loss": 0.42450448870658875,
|
| 1160 |
+
"eval_runtime": 2.4449,
|
| 1161 |
+
"eval_samples_per_second": 356.658,
|
| 1162 |
+
"eval_steps_per_second": 44.582,
|
| 1163 |
+
"step": 38500
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 9.26,
|
| 1167 |
+
"learning_rate": 1.5666851973517969e-06,
|
| 1168 |
+
"loss": 0.0213,
|
| 1169 |
+
"step": 39000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 9.26,
|
| 1173 |
+
"eval_accuracy": 0.9380733944954128,
|
| 1174 |
+
"eval_loss": 0.446162611246109,
|
| 1175 |
+
"eval_runtime": 2.4606,
|
| 1176 |
+
"eval_samples_per_second": 354.384,
|
| 1177 |
+
"eval_steps_per_second": 44.298,
|
| 1178 |
+
"step": 39000
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 9.38,
|
| 1182 |
+
"learning_rate": 1.3139940364886035e-06,
|
| 1183 |
+
"loss": 0.0276,
|
| 1184 |
+
"step": 39500
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 9.38,
|
| 1188 |
+
"eval_accuracy": 0.9380733944954128,
|
| 1189 |
+
"eval_loss": 0.42100322246551514,
|
| 1190 |
+
"eval_runtime": 2.4512,
|
| 1191 |
+
"eval_samples_per_second": 355.743,
|
| 1192 |
+
"eval_steps_per_second": 44.468,
|
| 1193 |
+
"step": 39500
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 9.5,
|
| 1197 |
+
"learning_rate": 1.0613028756254106e-06,
|
| 1198 |
+
"loss": 0.0183,
|
| 1199 |
+
"step": 40000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 9.5,
|
| 1203 |
+
"eval_accuracy": 0.9403669724770642,
|
| 1204 |
+
"eval_loss": 0.43098002672195435,
|
| 1205 |
+
"eval_runtime": 2.45,
|
| 1206 |
+
"eval_samples_per_second": 355.922,
|
| 1207 |
+
"eval_steps_per_second": 44.49,
|
| 1208 |
+
"step": 40000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 9.62,
|
| 1212 |
+
"learning_rate": 8.086117147622177e-07,
|
| 1213 |
+
"loss": 0.0184,
|
| 1214 |
+
"step": 40500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 9.62,
|
| 1218 |
+
"eval_accuracy": 0.9403669724770642,
|
| 1219 |
+
"eval_loss": 0.4437469244003296,
|
| 1220 |
+
"eval_runtime": 2.4461,
|
| 1221 |
+
"eval_samples_per_second": 356.492,
|
| 1222 |
+
"eval_steps_per_second": 44.561,
|
| 1223 |
+
"step": 40500
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 9.74,
|
| 1227 |
+
"learning_rate": 5.559205538990246e-07,
|
| 1228 |
+
"loss": 0.0296,
|
| 1229 |
+
"step": 41000
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 9.74,
|
| 1233 |
+
"eval_accuracy": 0.9392201834862385,
|
| 1234 |
+
"eval_loss": 0.43114030361175537,
|
| 1235 |
+
"eval_runtime": 2.4504,
|
| 1236 |
+
"eval_samples_per_second": 355.859,
|
| 1237 |
+
"eval_steps_per_second": 44.482,
|
| 1238 |
+
"step": 41000
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 9.86,
|
| 1242 |
+
"learning_rate": 3.0322939303583163e-07,
|
| 1243 |
+
"loss": 0.019,
|
| 1244 |
+
"step": 41500
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 9.86,
|
| 1248 |
+
"eval_accuracy": 0.9415137614678899,
|
| 1249 |
+
"eval_loss": 0.42435380816459656,
|
| 1250 |
+
"eval_runtime": 2.4473,
|
| 1251 |
+
"eval_samples_per_second": 356.311,
|
| 1252 |
+
"eval_steps_per_second": 44.539,
|
| 1253 |
+
"step": 41500
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 9.98,
|
| 1257 |
+
"learning_rate": 5.053823217263861e-08,
|
| 1258 |
+
"loss": 0.0245,
|
| 1259 |
+
"step": 42000
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 9.98,
|
| 1263 |
+
"eval_accuracy": 0.9415137614678899,
|
| 1264 |
+
"eval_loss": 0.42697247862815857,
|
| 1265 |
+
"eval_runtime": 2.46,
|
| 1266 |
+
"eval_samples_per_second": 354.474,
|
| 1267 |
+
"eval_steps_per_second": 44.309,
|
| 1268 |
+
"step": 42000
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 10.0,
|
| 1272 |
+
"step": 42100,
|
| 1273 |
+
"total_flos": 4.43006661686016e+16,
|
| 1274 |
+
"train_loss": 0.10745611605338416,
|
| 1275 |
+
"train_runtime": 8358.8854,
|
| 1276 |
+
"train_samples_per_second": 80.572,
|
| 1277 |
+
"train_steps_per_second": 5.037
|
| 1278 |
+
}
|
| 1279 |
+
],
|
| 1280 |
+
"max_steps": 42100,
|
| 1281 |
+
"num_train_epochs": 10,
|
| 1282 |
+
"total_flos": 4.43006661686016e+16,
|
| 1283 |
+
"trial_name": null,
|
| 1284 |
+
"trial_params": null
|
| 1285 |
+
}
|