End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +1975 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9086582463847702,
|
| 4 |
+
"eval_loss": 0.2568952143192291,
|
| 5 |
+
"eval_runtime": 15.8199,
|
| 6 |
+
"eval_samples": 5463,
|
| 7 |
+
"eval_samples_per_second": 345.324,
|
| 8 |
+
"eval_steps_per_second": 43.173,
|
| 9 |
+
"train_loss": 0.10566671685770447,
|
| 10 |
+
"train_runtime": 14393.9986,
|
| 11 |
+
"train_samples": 104743,
|
| 12 |
+
"train_samples_per_second": 72.769,
|
| 13 |
+
"train_steps_per_second": 4.548
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9086582463847702,
|
| 4 |
+
"eval_loss": 0.2568952143192291,
|
| 5 |
+
"eval_runtime": 15.8199,
|
| 6 |
+
"eval_samples": 5463,
|
| 7 |
+
"eval_samples_per_second": 345.324,
|
| 8 |
+
"eval_steps_per_second": 43.173
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 0.10566671685770447,
|
| 4 |
+
"train_runtime": 14393.9986,
|
| 5 |
+
"train_samples": 104743,
|
| 6 |
+
"train_samples_per_second": 72.769,
|
| 7 |
+
"train_steps_per_second": 4.548
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1975 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.2568952143192291,
|
| 3 |
+
"best_model_checkpoint": "./results_train/bert-base-cased/qnli/checkpoint-13000",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 65470,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.08,
|
| 12 |
+
"learning_rate": 2.545176889793841e-06,
|
| 13 |
+
"loss": 0.6529,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.08,
|
| 18 |
+
"eval_accuracy": 0.7733845872231375,
|
| 19 |
+
"eval_loss": 0.4945374131202698,
|
| 20 |
+
"eval_runtime": 15.9632,
|
| 21 |
+
"eval_samples_per_second": 342.225,
|
| 22 |
+
"eval_steps_per_second": 42.786,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.15,
|
| 27 |
+
"learning_rate": 5.090353779587682e-06,
|
| 28 |
+
"loss": 0.4731,
|
| 29 |
+
"step": 1000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.15,
|
| 33 |
+
"eval_accuracy": 0.8405637927878455,
|
| 34 |
+
"eval_loss": 0.38875827193260193,
|
| 35 |
+
"eval_runtime": 16.0177,
|
| 36 |
+
"eval_samples_per_second": 341.06,
|
| 37 |
+
"eval_steps_per_second": 42.64,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.23,
|
| 42 |
+
"learning_rate": 7.635530669381522e-06,
|
| 43 |
+
"loss": 0.4113,
|
| 44 |
+
"step": 1500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.23,
|
| 48 |
+
"eval_accuracy": 0.8522789676002197,
|
| 49 |
+
"eval_loss": 0.3604711890220642,
|
| 50 |
+
"eval_runtime": 15.9651,
|
| 51 |
+
"eval_samples_per_second": 342.185,
|
| 52 |
+
"eval_steps_per_second": 42.781,
|
| 53 |
+
"step": 1500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.31,
|
| 57 |
+
"learning_rate": 1.0180707559175364e-05,
|
| 58 |
+
"loss": 0.4006,
|
| 59 |
+
"step": 2000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.31,
|
| 63 |
+
"eval_accuracy": 0.8649093904448105,
|
| 64 |
+
"eval_loss": 0.3175490200519562,
|
| 65 |
+
"eval_runtime": 16.0149,
|
| 66 |
+
"eval_samples_per_second": 341.119,
|
| 67 |
+
"eval_steps_per_second": 42.648,
|
| 68 |
+
"step": 2000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.38,
|
| 72 |
+
"learning_rate": 1.2725884448969203e-05,
|
| 73 |
+
"loss": 0.3615,
|
| 74 |
+
"step": 2500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.38,
|
| 78 |
+
"eval_accuracy": 0.867838184147904,
|
| 79 |
+
"eval_loss": 0.32201188802719116,
|
| 80 |
+
"eval_runtime": 15.9586,
|
| 81 |
+
"eval_samples_per_second": 342.323,
|
| 82 |
+
"eval_steps_per_second": 42.798,
|
| 83 |
+
"step": 2500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.46,
|
| 87 |
+
"learning_rate": 1.5271061338763045e-05,
|
| 88 |
+
"loss": 0.3662,
|
| 89 |
+
"step": 3000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.46,
|
| 93 |
+
"eval_accuracy": 0.866739886509244,
|
| 94 |
+
"eval_loss": 0.3197564482688904,
|
| 95 |
+
"eval_runtime": 16.037,
|
| 96 |
+
"eval_samples_per_second": 340.649,
|
| 97 |
+
"eval_steps_per_second": 42.589,
|
| 98 |
+
"step": 3000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.53,
|
| 102 |
+
"learning_rate": 1.7816238228556887e-05,
|
| 103 |
+
"loss": 0.3578,
|
| 104 |
+
"step": 3500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.53,
|
| 108 |
+
"eval_accuracy": 0.8779059125022881,
|
| 109 |
+
"eval_loss": 0.29696720838546753,
|
| 110 |
+
"eval_runtime": 15.9685,
|
| 111 |
+
"eval_samples_per_second": 342.112,
|
| 112 |
+
"eval_steps_per_second": 42.772,
|
| 113 |
+
"step": 3500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.61,
|
| 117 |
+
"learning_rate": 1.997692595180449e-05,
|
| 118 |
+
"loss": 0.3349,
|
| 119 |
+
"step": 4000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.61,
|
| 123 |
+
"eval_accuracy": 0.8833974006955885,
|
| 124 |
+
"eval_loss": 0.28639864921569824,
|
| 125 |
+
"eval_runtime": 16.0264,
|
| 126 |
+
"eval_samples_per_second": 340.875,
|
| 127 |
+
"eval_steps_per_second": 42.617,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.69,
|
| 132 |
+
"learning_rate": 1.9814432654652997e-05,
|
| 133 |
+
"loss": 0.3548,
|
| 134 |
+
"step": 4500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.69,
|
| 138 |
+
"eval_accuracy": 0.8945634266886326,
|
| 139 |
+
"eval_loss": 0.2664399743080139,
|
| 140 |
+
"eval_runtime": 15.9996,
|
| 141 |
+
"eval_samples_per_second": 341.447,
|
| 142 |
+
"eval_steps_per_second": 42.689,
|
| 143 |
+
"step": 4500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.76,
|
| 147 |
+
"learning_rate": 1.9651939357501505e-05,
|
| 148 |
+
"loss": 0.3254,
|
| 149 |
+
"step": 5000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.76,
|
| 153 |
+
"eval_accuracy": 0.8929159802306426,
|
| 154 |
+
"eval_loss": 0.2650640904903412,
|
| 155 |
+
"eval_runtime": 16.02,
|
| 156 |
+
"eval_samples_per_second": 341.011,
|
| 157 |
+
"eval_steps_per_second": 42.634,
|
| 158 |
+
"step": 5000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.84,
|
| 162 |
+
"learning_rate": 1.9489446060350013e-05,
|
| 163 |
+
"loss": 0.3212,
|
| 164 |
+
"step": 5500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.84,
|
| 168 |
+
"eval_accuracy": 0.8817499542375984,
|
| 169 |
+
"eval_loss": 0.2942565679550171,
|
| 170 |
+
"eval_runtime": 15.947,
|
| 171 |
+
"eval_samples_per_second": 342.573,
|
| 172 |
+
"eval_steps_per_second": 42.829,
|
| 173 |
+
"step": 5500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.92,
|
| 177 |
+
"learning_rate": 1.932695276319852e-05,
|
| 178 |
+
"loss": 0.3195,
|
| 179 |
+
"step": 6000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.92,
|
| 183 |
+
"eval_accuracy": 0.8960278235401794,
|
| 184 |
+
"eval_loss": 0.26299455761909485,
|
| 185 |
+
"eval_runtime": 15.9626,
|
| 186 |
+
"eval_samples_per_second": 342.237,
|
| 187 |
+
"eval_steps_per_second": 42.787,
|
| 188 |
+
"step": 6000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.99,
|
| 192 |
+
"learning_rate": 1.916445946604703e-05,
|
| 193 |
+
"loss": 0.3044,
|
| 194 |
+
"step": 6500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.99,
|
| 198 |
+
"eval_accuracy": 0.8881566904631155,
|
| 199 |
+
"eval_loss": 0.27664485573768616,
|
| 200 |
+
"eval_runtime": 15.9599,
|
| 201 |
+
"eval_samples_per_second": 342.295,
|
| 202 |
+
"eval_steps_per_second": 42.795,
|
| 203 |
+
"step": 6500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 1.07,
|
| 207 |
+
"learning_rate": 1.9001966168895533e-05,
|
| 208 |
+
"loss": 0.245,
|
| 209 |
+
"step": 7000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.07,
|
| 213 |
+
"eval_accuracy": 0.9013362621270364,
|
| 214 |
+
"eval_loss": 0.3268437087535858,
|
| 215 |
+
"eval_runtime": 16.0333,
|
| 216 |
+
"eval_samples_per_second": 340.728,
|
| 217 |
+
"eval_steps_per_second": 42.599,
|
| 218 |
+
"step": 7000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.15,
|
| 222 |
+
"learning_rate": 1.883947287174404e-05,
|
| 223 |
+
"loss": 0.2415,
|
| 224 |
+
"step": 7500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.15,
|
| 228 |
+
"eval_accuracy": 0.8991396668497162,
|
| 229 |
+
"eval_loss": 0.29871582984924316,
|
| 230 |
+
"eval_runtime": 15.9621,
|
| 231 |
+
"eval_samples_per_second": 342.249,
|
| 232 |
+
"eval_steps_per_second": 42.789,
|
| 233 |
+
"step": 7500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.22,
|
| 237 |
+
"learning_rate": 1.867697957459255e-05,
|
| 238 |
+
"loss": 0.248,
|
| 239 |
+
"step": 8000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 1.22,
|
| 243 |
+
"eval_accuracy": 0.8812008054182684,
|
| 244 |
+
"eval_loss": 0.3372972905635834,
|
| 245 |
+
"eval_runtime": 16.0184,
|
| 246 |
+
"eval_samples_per_second": 341.046,
|
| 247 |
+
"eval_steps_per_second": 42.639,
|
| 248 |
+
"step": 8000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.3,
|
| 252 |
+
"learning_rate": 1.8514486277441056e-05,
|
| 253 |
+
"loss": 0.2328,
|
| 254 |
+
"step": 8500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.3,
|
| 258 |
+
"eval_accuracy": 0.8808347062053816,
|
| 259 |
+
"eval_loss": 0.34742510318756104,
|
| 260 |
+
"eval_runtime": 15.9539,
|
| 261 |
+
"eval_samples_per_second": 342.424,
|
| 262 |
+
"eval_steps_per_second": 42.811,
|
| 263 |
+
"step": 8500
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.37,
|
| 267 |
+
"learning_rate": 1.8351992980289564e-05,
|
| 268 |
+
"loss": 0.2427,
|
| 269 |
+
"step": 9000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.37,
|
| 273 |
+
"eval_accuracy": 0.8912685337726524,
|
| 274 |
+
"eval_loss": 0.2986329197883606,
|
| 275 |
+
"eval_runtime": 15.9879,
|
| 276 |
+
"eval_samples_per_second": 341.695,
|
| 277 |
+
"eval_steps_per_second": 42.72,
|
| 278 |
+
"step": 9000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.45,
|
| 282 |
+
"learning_rate": 1.8189499683138072e-05,
|
| 283 |
+
"loss": 0.243,
|
| 284 |
+
"step": 9500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.45,
|
| 288 |
+
"eval_accuracy": 0.8982244188174996,
|
| 289 |
+
"eval_loss": 0.2834506034851074,
|
| 290 |
+
"eval_runtime": 15.9775,
|
| 291 |
+
"eval_samples_per_second": 341.919,
|
| 292 |
+
"eval_steps_per_second": 42.748,
|
| 293 |
+
"step": 9500
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 1.53,
|
| 297 |
+
"learning_rate": 1.802700638598658e-05,
|
| 298 |
+
"loss": 0.2375,
|
| 299 |
+
"step": 10000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 1.53,
|
| 303 |
+
"eval_accuracy": 0.8909024345597657,
|
| 304 |
+
"eval_loss": 0.2893359661102295,
|
| 305 |
+
"eval_runtime": 15.9492,
|
| 306 |
+
"eval_samples_per_second": 342.525,
|
| 307 |
+
"eval_steps_per_second": 42.823,
|
| 308 |
+
"step": 10000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 1.6,
|
| 312 |
+
"learning_rate": 1.7864513088835088e-05,
|
| 313 |
+
"loss": 0.2253,
|
| 314 |
+
"step": 10500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 1.6,
|
| 318 |
+
"eval_accuracy": 0.8960278235401794,
|
| 319 |
+
"eval_loss": 0.2941761910915375,
|
| 320 |
+
"eval_runtime": 15.9667,
|
| 321 |
+
"eval_samples_per_second": 342.15,
|
| 322 |
+
"eval_steps_per_second": 42.777,
|
| 323 |
+
"step": 10500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 1.68,
|
| 327 |
+
"learning_rate": 1.7702019791683592e-05,
|
| 328 |
+
"loss": 0.2429,
|
| 329 |
+
"step": 11000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.68,
|
| 333 |
+
"eval_accuracy": 0.9028006589785832,
|
| 334 |
+
"eval_loss": 0.28518712520599365,
|
| 335 |
+
"eval_runtime": 16.0239,
|
| 336 |
+
"eval_samples_per_second": 340.928,
|
| 337 |
+
"eval_steps_per_second": 42.624,
|
| 338 |
+
"step": 11000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.76,
|
| 342 |
+
"learning_rate": 1.75395264945321e-05,
|
| 343 |
+
"loss": 0.238,
|
| 344 |
+
"step": 11500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 1.76,
|
| 348 |
+
"eval_accuracy": 0.9037159070108,
|
| 349 |
+
"eval_loss": 0.2671617865562439,
|
| 350 |
+
"eval_runtime": 15.9753,
|
| 351 |
+
"eval_samples_per_second": 341.966,
|
| 352 |
+
"eval_steps_per_second": 42.754,
|
| 353 |
+
"step": 11500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.83,
|
| 357 |
+
"learning_rate": 1.737703319738061e-05,
|
| 358 |
+
"loss": 0.2344,
|
| 359 |
+
"step": 12000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.83,
|
| 363 |
+
"eval_accuracy": 0.9020684605528098,
|
| 364 |
+
"eval_loss": 0.267158567905426,
|
| 365 |
+
"eval_runtime": 16.0345,
|
| 366 |
+
"eval_samples_per_second": 340.703,
|
| 367 |
+
"eval_steps_per_second": 42.596,
|
| 368 |
+
"step": 12000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.91,
|
| 372 |
+
"learning_rate": 1.7214539900229116e-05,
|
| 373 |
+
"loss": 0.2368,
|
| 374 |
+
"step": 12500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.91,
|
| 378 |
+
"eval_accuracy": 0.9018854109463664,
|
| 379 |
+
"eval_loss": 0.2753552496433258,
|
| 380 |
+
"eval_runtime": 15.9627,
|
| 381 |
+
"eval_samples_per_second": 342.234,
|
| 382 |
+
"eval_steps_per_second": 42.787,
|
| 383 |
+
"step": 12500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.99,
|
| 387 |
+
"learning_rate": 1.7052046603077624e-05,
|
| 388 |
+
"loss": 0.2424,
|
| 389 |
+
"step": 13000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.99,
|
| 393 |
+
"eval_accuracy": 0.9086582463847702,
|
| 394 |
+
"eval_loss": 0.2568952143192291,
|
| 395 |
+
"eval_runtime": 15.9721,
|
| 396 |
+
"eval_samples_per_second": 342.034,
|
| 397 |
+
"eval_steps_per_second": 42.762,
|
| 398 |
+
"step": 13000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 2.06,
|
| 402 |
+
"learning_rate": 1.6889553305926132e-05,
|
| 403 |
+
"loss": 0.1542,
|
| 404 |
+
"step": 13500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 2.06,
|
| 408 |
+
"eval_accuracy": 0.9026176093721399,
|
| 409 |
+
"eval_loss": 0.38070985674858093,
|
| 410 |
+
"eval_runtime": 15.9896,
|
| 411 |
+
"eval_samples_per_second": 341.659,
|
| 412 |
+
"eval_steps_per_second": 42.715,
|
| 413 |
+
"step": 13500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 2.14,
|
| 417 |
+
"learning_rate": 1.672706000877464e-05,
|
| 418 |
+
"loss": 0.1416,
|
| 419 |
+
"step": 14000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 2.14,
|
| 423 |
+
"eval_accuracy": 0.8980413692110562,
|
| 424 |
+
"eval_loss": 0.3915307819843292,
|
| 425 |
+
"eval_runtime": 15.9993,
|
| 426 |
+
"eval_samples_per_second": 341.452,
|
| 427 |
+
"eval_steps_per_second": 42.689,
|
| 428 |
+
"step": 14000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 2.21,
|
| 432 |
+
"learning_rate": 1.6564566711623148e-05,
|
| 433 |
+
"loss": 0.1445,
|
| 434 |
+
"step": 14500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 2.21,
|
| 438 |
+
"eval_accuracy": 0.8984074684239429,
|
| 439 |
+
"eval_loss": 0.42921990156173706,
|
| 440 |
+
"eval_runtime": 15.964,
|
| 441 |
+
"eval_samples_per_second": 342.208,
|
| 442 |
+
"eval_steps_per_second": 42.784,
|
| 443 |
+
"step": 14500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 2.29,
|
| 447 |
+
"learning_rate": 1.6402073414471655e-05,
|
| 448 |
+
"loss": 0.1631,
|
| 449 |
+
"step": 15000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 2.29,
|
| 453 |
+
"eval_accuracy": 0.8971261211788395,
|
| 454 |
+
"eval_loss": 0.4096887409687042,
|
| 455 |
+
"eval_runtime": 16.0404,
|
| 456 |
+
"eval_samples_per_second": 340.578,
|
| 457 |
+
"eval_steps_per_second": 42.58,
|
| 458 |
+
"step": 15000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 2.37,
|
| 462 |
+
"learning_rate": 1.6239580117320163e-05,
|
| 463 |
+
"loss": 0.1512,
|
| 464 |
+
"step": 15500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 2.37,
|
| 468 |
+
"eval_accuracy": 0.9011532125205931,
|
| 469 |
+
"eval_loss": 0.3880288898944855,
|
| 470 |
+
"eval_runtime": 15.9791,
|
| 471 |
+
"eval_samples_per_second": 341.885,
|
| 472 |
+
"eval_steps_per_second": 42.743,
|
| 473 |
+
"step": 15500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.44,
|
| 477 |
+
"learning_rate": 1.607708682016867e-05,
|
| 478 |
+
"loss": 0.1624,
|
| 479 |
+
"step": 16000
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 2.44,
|
| 483 |
+
"eval_accuracy": 0.8954786747208493,
|
| 484 |
+
"eval_loss": 0.40831083059310913,
|
| 485 |
+
"eval_runtime": 16.0179,
|
| 486 |
+
"eval_samples_per_second": 341.056,
|
| 487 |
+
"eval_steps_per_second": 42.64,
|
| 488 |
+
"step": 16000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 2.52,
|
| 492 |
+
"learning_rate": 1.5914593523017176e-05,
|
| 493 |
+
"loss": 0.1616,
|
| 494 |
+
"step": 16500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.52,
|
| 498 |
+
"eval_accuracy": 0.9038989566172433,
|
| 499 |
+
"eval_loss": 0.3950469195842743,
|
| 500 |
+
"eval_runtime": 15.9181,
|
| 501 |
+
"eval_samples_per_second": 343.195,
|
| 502 |
+
"eval_steps_per_second": 42.907,
|
| 503 |
+
"step": 16500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 2.6,
|
| 507 |
+
"learning_rate": 1.5752100225865684e-05,
|
| 508 |
+
"loss": 0.1587,
|
| 509 |
+
"step": 17000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 2.6,
|
| 513 |
+
"eval_accuracy": 0.9103056928427604,
|
| 514 |
+
"eval_loss": 0.35793736577033997,
|
| 515 |
+
"eval_runtime": 15.9998,
|
| 516 |
+
"eval_samples_per_second": 341.441,
|
| 517 |
+
"eval_steps_per_second": 42.688,
|
| 518 |
+
"step": 17000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.67,
|
| 522 |
+
"learning_rate": 1.558960692871419e-05,
|
| 523 |
+
"loss": 0.1615,
|
| 524 |
+
"step": 17500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 2.67,
|
| 528 |
+
"eval_accuracy": 0.9011532125205931,
|
| 529 |
+
"eval_loss": 0.39305415749549866,
|
| 530 |
+
"eval_runtime": 15.9412,
|
| 531 |
+
"eval_samples_per_second": 342.697,
|
| 532 |
+
"eval_steps_per_second": 42.845,
|
| 533 |
+
"step": 17500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 2.75,
|
| 537 |
+
"learning_rate": 1.54271136315627e-05,
|
| 538 |
+
"loss": 0.1623,
|
| 539 |
+
"step": 18000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 2.75,
|
| 543 |
+
"eval_accuracy": 0.90591250228812,
|
| 544 |
+
"eval_loss": 0.3697090446949005,
|
| 545 |
+
"eval_runtime": 15.9757,
|
| 546 |
+
"eval_samples_per_second": 341.957,
|
| 547 |
+
"eval_steps_per_second": 42.752,
|
| 548 |
+
"step": 18000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 2.83,
|
| 552 |
+
"learning_rate": 1.5264620334411207e-05,
|
| 553 |
+
"loss": 0.1687,
|
| 554 |
+
"step": 18500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 2.83,
|
| 558 |
+
"eval_accuracy": 0.9037159070108,
|
| 559 |
+
"eval_loss": 0.34730929136276245,
|
| 560 |
+
"eval_runtime": 15.964,
|
| 561 |
+
"eval_samples_per_second": 342.208,
|
| 562 |
+
"eval_steps_per_second": 42.784,
|
| 563 |
+
"step": 18500
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 2.9,
|
| 567 |
+
"learning_rate": 1.5102127037259715e-05,
|
| 568 |
+
"loss": 0.1627,
|
| 569 |
+
"step": 19000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 2.9,
|
| 573 |
+
"eval_accuracy": 0.8982244188174996,
|
| 574 |
+
"eval_loss": 0.3851012885570526,
|
| 575 |
+
"eval_runtime": 15.9956,
|
| 576 |
+
"eval_samples_per_second": 341.531,
|
| 577 |
+
"eval_steps_per_second": 42.699,
|
| 578 |
+
"step": 19000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.98,
|
| 582 |
+
"learning_rate": 1.4939633740108221e-05,
|
| 583 |
+
"loss": 0.1593,
|
| 584 |
+
"step": 19500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 2.98,
|
| 588 |
+
"eval_accuracy": 0.9018854109463664,
|
| 589 |
+
"eval_loss": 0.4038921296596527,
|
| 590 |
+
"eval_runtime": 15.9569,
|
| 591 |
+
"eval_samples_per_second": 342.36,
|
| 592 |
+
"eval_steps_per_second": 42.803,
|
| 593 |
+
"step": 19500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 3.05,
|
| 597 |
+
"learning_rate": 1.477714044295673e-05,
|
| 598 |
+
"loss": 0.1136,
|
| 599 |
+
"step": 20000
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 3.05,
|
| 603 |
+
"eval_accuracy": 0.9024345597656965,
|
| 604 |
+
"eval_loss": 0.48348602652549744,
|
| 605 |
+
"eval_runtime": 15.9762,
|
| 606 |
+
"eval_samples_per_second": 341.946,
|
| 607 |
+
"eval_steps_per_second": 42.751,
|
| 608 |
+
"step": 20000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 3.13,
|
| 612 |
+
"learning_rate": 1.4614647145805237e-05,
|
| 613 |
+
"loss": 0.0965,
|
| 614 |
+
"step": 20500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 3.13,
|
| 618 |
+
"eval_accuracy": 0.9011532125205931,
|
| 619 |
+
"eval_loss": 0.5060518383979797,
|
| 620 |
+
"eval_runtime": 15.9664,
|
| 621 |
+
"eval_samples_per_second": 342.155,
|
| 622 |
+
"eval_steps_per_second": 42.777,
|
| 623 |
+
"step": 20500
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 3.21,
|
| 627 |
+
"learning_rate": 1.4452153848653745e-05,
|
| 628 |
+
"loss": 0.0931,
|
| 629 |
+
"step": 21000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 3.21,
|
| 633 |
+
"eval_accuracy": 0.8991396668497162,
|
| 634 |
+
"eval_loss": 0.5278568267822266,
|
| 635 |
+
"eval_runtime": 16.0118,
|
| 636 |
+
"eval_samples_per_second": 341.186,
|
| 637 |
+
"eval_steps_per_second": 42.656,
|
| 638 |
+
"step": 21000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 3.28,
|
| 642 |
+
"learning_rate": 1.4289660551502251e-05,
|
| 643 |
+
"loss": 0.0993,
|
| 644 |
+
"step": 21500
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 3.28,
|
| 648 |
+
"eval_accuracy": 0.9018854109463664,
|
| 649 |
+
"eval_loss": 0.48559123277664185,
|
| 650 |
+
"eval_runtime": 15.9958,
|
| 651 |
+
"eval_samples_per_second": 341.526,
|
| 652 |
+
"eval_steps_per_second": 42.699,
|
| 653 |
+
"step": 21500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 3.36,
|
| 657 |
+
"learning_rate": 1.412716725435076e-05,
|
| 658 |
+
"loss": 0.1187,
|
| 659 |
+
"step": 22000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 3.36,
|
| 663 |
+
"eval_accuracy": 0.9031667581914699,
|
| 664 |
+
"eval_loss": 0.48830872774124146,
|
| 665 |
+
"eval_runtime": 16.0824,
|
| 666 |
+
"eval_samples_per_second": 339.689,
|
| 667 |
+
"eval_steps_per_second": 42.469,
|
| 668 |
+
"step": 22000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 3.44,
|
| 672 |
+
"learning_rate": 1.3964673957199267e-05,
|
| 673 |
+
"loss": 0.1008,
|
| 674 |
+
"step": 22500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 3.44,
|
| 678 |
+
"eval_accuracy": 0.9015193117334798,
|
| 679 |
+
"eval_loss": 0.5054051280021667,
|
| 680 |
+
"eval_runtime": 15.9998,
|
| 681 |
+
"eval_samples_per_second": 341.442,
|
| 682 |
+
"eval_steps_per_second": 42.688,
|
| 683 |
+
"step": 22500
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 3.51,
|
| 687 |
+
"learning_rate": 1.3802180660047775e-05,
|
| 688 |
+
"loss": 0.1013,
|
| 689 |
+
"step": 23000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 3.51,
|
| 693 |
+
"eval_accuracy": 0.9022515101592532,
|
| 694 |
+
"eval_loss": 0.5024679899215698,
|
| 695 |
+
"eval_runtime": 16.0433,
|
| 696 |
+
"eval_samples_per_second": 340.516,
|
| 697 |
+
"eval_steps_per_second": 42.572,
|
| 698 |
+
"step": 23000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 3.59,
|
| 702 |
+
"learning_rate": 1.3639687362896281e-05,
|
| 703 |
+
"loss": 0.1092,
|
| 704 |
+
"step": 23500
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 3.59,
|
| 708 |
+
"eval_accuracy": 0.8985905180303863,
|
| 709 |
+
"eval_loss": 0.44847574830055237,
|
| 710 |
+
"eval_runtime": 15.9971,
|
| 711 |
+
"eval_samples_per_second": 341.499,
|
| 712 |
+
"eval_steps_per_second": 42.695,
|
| 713 |
+
"step": 23500
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 3.67,
|
| 717 |
+
"learning_rate": 1.347719406574479e-05,
|
| 718 |
+
"loss": 0.1135,
|
| 719 |
+
"step": 24000
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 3.67,
|
| 723 |
+
"eval_accuracy": 0.8976752699981695,
|
| 724 |
+
"eval_loss": 0.512342095375061,
|
| 725 |
+
"eval_runtime": 16.034,
|
| 726 |
+
"eval_samples_per_second": 340.714,
|
| 727 |
+
"eval_steps_per_second": 42.597,
|
| 728 |
+
"step": 24000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 3.74,
|
| 732 |
+
"learning_rate": 1.3314700768593297e-05,
|
| 733 |
+
"loss": 0.1042,
|
| 734 |
+
"step": 24500
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 3.74,
|
| 738 |
+
"eval_accuracy": 0.9009701629141498,
|
| 739 |
+
"eval_loss": 0.4883783161640167,
|
| 740 |
+
"eval_runtime": 16.5969,
|
| 741 |
+
"eval_samples_per_second": 329.158,
|
| 742 |
+
"eval_steps_per_second": 41.152,
|
| 743 |
+
"step": 24500
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 3.82,
|
| 747 |
+
"learning_rate": 1.3152207471441804e-05,
|
| 748 |
+
"loss": 0.1178,
|
| 749 |
+
"step": 25000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 3.82,
|
| 753 |
+
"eval_accuracy": 0.900604063701263,
|
| 754 |
+
"eval_loss": 0.4129727780818939,
|
| 755 |
+
"eval_runtime": 16.0114,
|
| 756 |
+
"eval_samples_per_second": 341.195,
|
| 757 |
+
"eval_steps_per_second": 42.657,
|
| 758 |
+
"step": 25000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 3.89,
|
| 762 |
+
"learning_rate": 1.298971417429031e-05,
|
| 763 |
+
"loss": 0.1041,
|
| 764 |
+
"step": 25500
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 3.89,
|
| 768 |
+
"eval_accuracy": 0.8951125755079626,
|
| 769 |
+
"eval_loss": 0.4847397208213806,
|
| 770 |
+
"eval_runtime": 15.9759,
|
| 771 |
+
"eval_samples_per_second": 341.952,
|
| 772 |
+
"eval_steps_per_second": 42.752,
|
| 773 |
+
"step": 25500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 3.97,
|
| 777 |
+
"learning_rate": 1.282722087713882e-05,
|
| 778 |
+
"loss": 0.1025,
|
| 779 |
+
"step": 26000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 3.97,
|
| 783 |
+
"eval_accuracy": 0.8985905180303863,
|
| 784 |
+
"eval_loss": 0.4706151783466339,
|
| 785 |
+
"eval_runtime": 15.9981,
|
| 786 |
+
"eval_samples_per_second": 341.478,
|
| 787 |
+
"eval_steps_per_second": 42.693,
|
| 788 |
+
"step": 26000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 4.05,
|
| 792 |
+
"learning_rate": 1.2664727579987326e-05,
|
| 793 |
+
"loss": 0.0833,
|
| 794 |
+
"step": 26500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 4.05,
|
| 798 |
+
"eval_accuracy": 0.90536335346879,
|
| 799 |
+
"eval_loss": 0.5018308162689209,
|
| 800 |
+
"eval_runtime": 15.9984,
|
| 801 |
+
"eval_samples_per_second": 341.472,
|
| 802 |
+
"eval_steps_per_second": 42.692,
|
| 803 |
+
"step": 26500
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 4.12,
|
| 807 |
+
"learning_rate": 1.2502234282835834e-05,
|
| 808 |
+
"loss": 0.0632,
|
| 809 |
+
"step": 27000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 4.12,
|
| 813 |
+
"eval_accuracy": 0.9055464030752334,
|
| 814 |
+
"eval_loss": 0.5425559282302856,
|
| 815 |
+
"eval_runtime": 15.9902,
|
| 816 |
+
"eval_samples_per_second": 341.646,
|
| 817 |
+
"eval_steps_per_second": 42.714,
|
| 818 |
+
"step": 27000
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 4.2,
|
| 822 |
+
"learning_rate": 1.233974098568434e-05,
|
| 823 |
+
"loss": 0.0613,
|
| 824 |
+
"step": 27500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 4.2,
|
| 828 |
+
"eval_accuracy": 0.9020684605528098,
|
| 829 |
+
"eval_loss": 0.5629337430000305,
|
| 830 |
+
"eval_runtime": 16.0205,
|
| 831 |
+
"eval_samples_per_second": 341.001,
|
| 832 |
+
"eval_steps_per_second": 42.633,
|
| 833 |
+
"step": 27500
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 4.28,
|
| 837 |
+
"learning_rate": 1.217724768853285e-05,
|
| 838 |
+
"loss": 0.0703,
|
| 839 |
+
"step": 28000
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 4.28,
|
| 843 |
+
"eval_accuracy": 0.9033498077979132,
|
| 844 |
+
"eval_loss": 0.5763201713562012,
|
| 845 |
+
"eval_runtime": 16.043,
|
| 846 |
+
"eval_samples_per_second": 340.522,
|
| 847 |
+
"eval_steps_per_second": 42.573,
|
| 848 |
+
"step": 28000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 4.35,
|
| 852 |
+
"learning_rate": 1.2014754391381356e-05,
|
| 853 |
+
"loss": 0.0687,
|
| 854 |
+
"step": 28500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 4.35,
|
| 858 |
+
"eval_accuracy": 0.9038989566172433,
|
| 859 |
+
"eval_loss": 0.5274228453636169,
|
| 860 |
+
"eval_runtime": 16.0055,
|
| 861 |
+
"eval_samples_per_second": 341.319,
|
| 862 |
+
"eval_steps_per_second": 42.673,
|
| 863 |
+
"step": 28500
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 4.43,
|
| 867 |
+
"learning_rate": 1.1852261094229864e-05,
|
| 868 |
+
"loss": 0.0707,
|
| 869 |
+
"step": 29000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 4.43,
|
| 873 |
+
"eval_accuracy": 0.9018854109463664,
|
| 874 |
+
"eval_loss": 0.5681867599487305,
|
| 875 |
+
"eval_runtime": 16.0832,
|
| 876 |
+
"eval_samples_per_second": 339.671,
|
| 877 |
+
"eval_steps_per_second": 42.467,
|
| 878 |
+
"step": 29000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 4.51,
|
| 882 |
+
"learning_rate": 1.168976779707837e-05,
|
| 883 |
+
"loss": 0.0717,
|
| 884 |
+
"step": 29500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 4.51,
|
| 888 |
+
"eval_accuracy": 0.9004210140948197,
|
| 889 |
+
"eval_loss": 0.5382347702980042,
|
| 890 |
+
"eval_runtime": 16.0114,
|
| 891 |
+
"eval_samples_per_second": 341.195,
|
| 892 |
+
"eval_steps_per_second": 42.657,
|
| 893 |
+
"step": 29500
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 4.58,
|
| 897 |
+
"learning_rate": 1.152727449992688e-05,
|
| 898 |
+
"loss": 0.0692,
|
| 899 |
+
"step": 30000
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 4.58,
|
| 903 |
+
"eval_accuracy": 0.9009701629141498,
|
| 904 |
+
"eval_loss": 0.5901069045066833,
|
| 905 |
+
"eval_runtime": 15.9825,
|
| 906 |
+
"eval_samples_per_second": 341.812,
|
| 907 |
+
"eval_steps_per_second": 42.734,
|
| 908 |
+
"step": 30000
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 4.66,
|
| 912 |
+
"learning_rate": 1.1364781202775386e-05,
|
| 913 |
+
"loss": 0.0701,
|
| 914 |
+
"step": 30500
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 4.66,
|
| 918 |
+
"eval_accuracy": 0.8989566172432729,
|
| 919 |
+
"eval_loss": 0.5817105770111084,
|
| 920 |
+
"eval_runtime": 16.0192,
|
| 921 |
+
"eval_samples_per_second": 341.027,
|
| 922 |
+
"eval_steps_per_second": 42.636,
|
| 923 |
+
"step": 30500
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 4.73,
|
| 927 |
+
"learning_rate": 1.1202287905623894e-05,
|
| 928 |
+
"loss": 0.0708,
|
| 929 |
+
"step": 31000
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 4.73,
|
| 933 |
+
"eval_accuracy": 0.9018854109463664,
|
| 934 |
+
"eval_loss": 0.5580092668533325,
|
| 935 |
+
"eval_runtime": 16.0391,
|
| 936 |
+
"eval_samples_per_second": 340.605,
|
| 937 |
+
"eval_steps_per_second": 42.583,
|
| 938 |
+
"step": 31000
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 4.81,
|
| 942 |
+
"learning_rate": 1.10397946084724e-05,
|
| 943 |
+
"loss": 0.07,
|
| 944 |
+
"step": 31500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 4.81,
|
| 948 |
+
"eval_accuracy": 0.8989566172432729,
|
| 949 |
+
"eval_loss": 0.5640453100204468,
|
| 950 |
+
"eval_runtime": 15.9857,
|
| 951 |
+
"eval_samples_per_second": 341.744,
|
| 952 |
+
"eval_steps_per_second": 42.726,
|
| 953 |
+
"step": 31500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 4.89,
|
| 957 |
+
"learning_rate": 1.087730131132091e-05,
|
| 958 |
+
"loss": 0.0725,
|
| 959 |
+
"step": 32000
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 4.89,
|
| 963 |
+
"eval_accuracy": 0.8993227164561596,
|
| 964 |
+
"eval_loss": 0.5767560005187988,
|
| 965 |
+
"eval_runtime": 16.012,
|
| 966 |
+
"eval_samples_per_second": 341.182,
|
| 967 |
+
"eval_steps_per_second": 42.656,
|
| 968 |
+
"step": 32000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 4.96,
|
| 972 |
+
"learning_rate": 1.0714808014169416e-05,
|
| 973 |
+
"loss": 0.0701,
|
| 974 |
+
"step": 32500
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 4.96,
|
| 978 |
+
"eval_accuracy": 0.9035328574043566,
|
| 979 |
+
"eval_loss": 0.5289304852485657,
|
| 980 |
+
"eval_runtime": 16.0251,
|
| 981 |
+
"eval_samples_per_second": 340.903,
|
| 982 |
+
"eval_steps_per_second": 42.621,
|
| 983 |
+
"step": 32500
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 5.04,
|
| 987 |
+
"learning_rate": 1.0552314717017924e-05,
|
| 988 |
+
"loss": 0.0441,
|
| 989 |
+
"step": 33000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 5.04,
|
| 993 |
+
"eval_accuracy": 0.9009701629141498,
|
| 994 |
+
"eval_loss": 0.6401296854019165,
|
| 995 |
+
"eval_runtime": 16.0047,
|
| 996 |
+
"eval_samples_per_second": 341.338,
|
| 997 |
+
"eval_steps_per_second": 42.675,
|
| 998 |
+
"step": 33000
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 5.12,
|
| 1002 |
+
"learning_rate": 1.038982141986643e-05,
|
| 1003 |
+
"loss": 0.0388,
|
| 1004 |
+
"step": 33500
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 5.12,
|
| 1008 |
+
"eval_accuracy": 0.8991396668497162,
|
| 1009 |
+
"eval_loss": 0.644571840763092,
|
| 1010 |
+
"eval_runtime": 15.9856,
|
| 1011 |
+
"eval_samples_per_second": 341.745,
|
| 1012 |
+
"eval_steps_per_second": 42.726,
|
| 1013 |
+
"step": 33500
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 5.19,
|
| 1017 |
+
"learning_rate": 1.022732812271494e-05,
|
| 1018 |
+
"loss": 0.0417,
|
| 1019 |
+
"step": 34000
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 5.19,
|
| 1023 |
+
"eval_accuracy": 0.9038989566172433,
|
| 1024 |
+
"eval_loss": 0.6326994299888611,
|
| 1025 |
+
"eval_runtime": 15.9965,
|
| 1026 |
+
"eval_samples_per_second": 341.512,
|
| 1027 |
+
"eval_steps_per_second": 42.697,
|
| 1028 |
+
"step": 34000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 5.27,
|
| 1032 |
+
"learning_rate": 1.0064834825563446e-05,
|
| 1033 |
+
"loss": 0.039,
|
| 1034 |
+
"step": 34500
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 5.27,
|
| 1038 |
+
"eval_accuracy": 0.90481420464946,
|
| 1039 |
+
"eval_loss": 0.6385270357131958,
|
| 1040 |
+
"eval_runtime": 15.9731,
|
| 1041 |
+
"eval_samples_per_second": 342.013,
|
| 1042 |
+
"eval_steps_per_second": 42.759,
|
| 1043 |
+
"step": 34500
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 5.35,
|
| 1047 |
+
"learning_rate": 9.902341528411954e-06,
|
| 1048 |
+
"loss": 0.0407,
|
| 1049 |
+
"step": 35000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 5.35,
|
| 1053 |
+
"eval_accuracy": 0.9029837085850265,
|
| 1054 |
+
"eval_loss": 0.6509894728660583,
|
| 1055 |
+
"eval_runtime": 16.0054,
|
| 1056 |
+
"eval_samples_per_second": 341.322,
|
| 1057 |
+
"eval_steps_per_second": 42.673,
|
| 1058 |
+
"step": 35000
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 5.42,
|
| 1062 |
+
"learning_rate": 9.739848231260461e-06,
|
| 1063 |
+
"loss": 0.0446,
|
| 1064 |
+
"step": 35500
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 5.42,
|
| 1068 |
+
"eval_accuracy": 0.9029837085850265,
|
| 1069 |
+
"eval_loss": 0.5787823796272278,
|
| 1070 |
+
"eval_runtime": 16.0265,
|
| 1071 |
+
"eval_samples_per_second": 340.872,
|
| 1072 |
+
"eval_steps_per_second": 42.617,
|
| 1073 |
+
"step": 35500
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 5.5,
|
| 1077 |
+
"learning_rate": 9.57735493410897e-06,
|
| 1078 |
+
"loss": 0.0422,
|
| 1079 |
+
"step": 36000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 5.5,
|
| 1083 |
+
"eval_accuracy": 0.9024345597656965,
|
| 1084 |
+
"eval_loss": 0.6722646951675415,
|
| 1085 |
+
"eval_runtime": 16.1928,
|
| 1086 |
+
"eval_samples_per_second": 337.372,
|
| 1087 |
+
"eval_steps_per_second": 42.179,
|
| 1088 |
+
"step": 36000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 5.58,
|
| 1092 |
+
"learning_rate": 9.414861636957477e-06,
|
| 1093 |
+
"loss": 0.04,
|
| 1094 |
+
"step": 36500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 5.58,
|
| 1098 |
+
"eval_accuracy": 0.9037159070108,
|
| 1099 |
+
"eval_loss": 0.6601631045341492,
|
| 1100 |
+
"eval_runtime": 15.9359,
|
| 1101 |
+
"eval_samples_per_second": 342.812,
|
| 1102 |
+
"eval_steps_per_second": 42.859,
|
| 1103 |
+
"step": 36500
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 5.65,
|
| 1107 |
+
"learning_rate": 9.252368339805983e-06,
|
| 1108 |
+
"loss": 0.0514,
|
| 1109 |
+
"step": 37000
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 5.65,
|
| 1113 |
+
"eval_accuracy": 0.9024345597656965,
|
| 1114 |
+
"eval_loss": 0.6407366394996643,
|
| 1115 |
+
"eval_runtime": 16.0327,
|
| 1116 |
+
"eval_samples_per_second": 340.742,
|
| 1117 |
+
"eval_steps_per_second": 42.601,
|
| 1118 |
+
"step": 37000
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 5.73,
|
| 1122 |
+
"learning_rate": 9.089875042654491e-06,
|
| 1123 |
+
"loss": 0.0462,
|
| 1124 |
+
"step": 37500
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 5.73,
|
| 1128 |
+
"eval_accuracy": 0.90481420464946,
|
| 1129 |
+
"eval_loss": 0.6144635677337646,
|
| 1130 |
+
"eval_runtime": 16.0227,
|
| 1131 |
+
"eval_samples_per_second": 340.955,
|
| 1132 |
+
"eval_steps_per_second": 42.627,
|
| 1133 |
+
"step": 37500
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 5.8,
|
| 1137 |
+
"learning_rate": 8.927381745502999e-06,
|
| 1138 |
+
"loss": 0.0479,
|
| 1139 |
+
"step": 38000
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 5.8,
|
| 1143 |
+
"eval_accuracy": 0.9007871133077063,
|
| 1144 |
+
"eval_loss": 0.5881094336509705,
|
| 1145 |
+
"eval_runtime": 16.0441,
|
| 1146 |
+
"eval_samples_per_second": 340.5,
|
| 1147 |
+
"eval_steps_per_second": 42.57,
|
| 1148 |
+
"step": 38000
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 5.88,
|
| 1152 |
+
"learning_rate": 8.764888448351507e-06,
|
| 1153 |
+
"loss": 0.0503,
|
| 1154 |
+
"step": 38500
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 5.88,
|
| 1158 |
+
"eval_accuracy": 0.9007871133077063,
|
| 1159 |
+
"eval_loss": 0.6001153588294983,
|
| 1160 |
+
"eval_runtime": 15.9844,
|
| 1161 |
+
"eval_samples_per_second": 341.77,
|
| 1162 |
+
"eval_steps_per_second": 42.729,
|
| 1163 |
+
"step": 38500
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 5.96,
|
| 1167 |
+
"learning_rate": 8.602395151200013e-06,
|
| 1168 |
+
"loss": 0.0385,
|
| 1169 |
+
"step": 39000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 5.96,
|
| 1173 |
+
"eval_accuracy": 0.9051803038623467,
|
| 1174 |
+
"eval_loss": 0.6463525295257568,
|
| 1175 |
+
"eval_runtime": 16.0778,
|
| 1176 |
+
"eval_samples_per_second": 339.785,
|
| 1177 |
+
"eval_steps_per_second": 42.481,
|
| 1178 |
+
"step": 39000
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 6.03,
|
| 1182 |
+
"learning_rate": 8.439901854048521e-06,
|
| 1183 |
+
"loss": 0.0436,
|
| 1184 |
+
"step": 39500
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 6.03,
|
| 1188 |
+
"eval_accuracy": 0.9038989566172433,
|
| 1189 |
+
"eval_loss": 0.6683228611946106,
|
| 1190 |
+
"eval_runtime": 15.9874,
|
| 1191 |
+
"eval_samples_per_second": 341.707,
|
| 1192 |
+
"eval_steps_per_second": 42.721,
|
| 1193 |
+
"step": 39500
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 6.11,
|
| 1197 |
+
"learning_rate": 8.277408556897029e-06,
|
| 1198 |
+
"loss": 0.0296,
|
| 1199 |
+
"step": 40000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 6.11,
|
| 1203 |
+
"eval_accuracy": 0.9037159070108,
|
| 1204 |
+
"eval_loss": 0.696530282497406,
|
| 1205 |
+
"eval_runtime": 15.9785,
|
| 1206 |
+
"eval_samples_per_second": 341.896,
|
| 1207 |
+
"eval_steps_per_second": 42.745,
|
| 1208 |
+
"step": 40000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 6.19,
|
| 1212 |
+
"learning_rate": 8.114915259745537e-06,
|
| 1213 |
+
"loss": 0.0275,
|
| 1214 |
+
"step": 40500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 6.19,
|
| 1218 |
+
"eval_accuracy": 0.9038989566172433,
|
| 1219 |
+
"eval_loss": 0.7192733883857727,
|
| 1220 |
+
"eval_runtime": 15.9602,
|
| 1221 |
+
"eval_samples_per_second": 342.288,
|
| 1222 |
+
"eval_steps_per_second": 42.794,
|
| 1223 |
+
"step": 40500
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 6.26,
|
| 1227 |
+
"learning_rate": 7.952421962594043e-06,
|
| 1228 |
+
"loss": 0.0258,
|
| 1229 |
+
"step": 41000
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 6.26,
|
| 1233 |
+
"eval_accuracy": 0.9038989566172433,
|
| 1234 |
+
"eval_loss": 0.7229453921318054,
|
| 1235 |
+
"eval_runtime": 15.9822,
|
| 1236 |
+
"eval_samples_per_second": 341.817,
|
| 1237 |
+
"eval_steps_per_second": 42.735,
|
| 1238 |
+
"step": 41000
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 6.34,
|
| 1242 |
+
"learning_rate": 7.789928665442551e-06,
|
| 1243 |
+
"loss": 0.0223,
|
| 1244 |
+
"step": 41500
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 6.34,
|
| 1248 |
+
"eval_accuracy": 0.9073768991396669,
|
| 1249 |
+
"eval_loss": 0.697367250919342,
|
| 1250 |
+
"eval_runtime": 15.9944,
|
| 1251 |
+
"eval_samples_per_second": 341.557,
|
| 1252 |
+
"eval_steps_per_second": 42.702,
|
| 1253 |
+
"step": 41500
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 6.42,
|
| 1257 |
+
"learning_rate": 7.627435368291059e-06,
|
| 1258 |
+
"loss": 0.0268,
|
| 1259 |
+
"step": 42000
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 6.42,
|
| 1263 |
+
"eval_accuracy": 0.9028006589785832,
|
| 1264 |
+
"eval_loss": 0.704451858997345,
|
| 1265 |
+
"eval_runtime": 16.0075,
|
| 1266 |
+
"eval_samples_per_second": 341.277,
|
| 1267 |
+
"eval_steps_per_second": 42.667,
|
| 1268 |
+
"step": 42000
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 6.49,
|
| 1272 |
+
"learning_rate": 7.464942071139566e-06,
|
| 1273 |
+
"loss": 0.0297,
|
| 1274 |
+
"step": 42500
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 6.49,
|
| 1278 |
+
"eval_accuracy": 0.9055464030752334,
|
| 1279 |
+
"eval_loss": 0.72894686460495,
|
| 1280 |
+
"eval_runtime": 15.9896,
|
| 1281 |
+
"eval_samples_per_second": 341.659,
|
| 1282 |
+
"eval_steps_per_second": 42.715,
|
| 1283 |
+
"step": 42500
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 6.57,
|
| 1287 |
+
"learning_rate": 7.302448773988074e-06,
|
| 1288 |
+
"loss": 0.0295,
|
| 1289 |
+
"step": 43000
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 6.57,
|
| 1293 |
+
"eval_accuracy": 0.9049972542559034,
|
| 1294 |
+
"eval_loss": 0.6809965372085571,
|
| 1295 |
+
"eval_runtime": 16.0,
|
| 1296 |
+
"eval_samples_per_second": 341.438,
|
| 1297 |
+
"eval_steps_per_second": 42.688,
|
| 1298 |
+
"step": 43000
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 6.64,
|
| 1302 |
+
"learning_rate": 7.139955476836581e-06,
|
| 1303 |
+
"loss": 0.0265,
|
| 1304 |
+
"step": 43500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 6.64,
|
| 1308 |
+
"eval_accuracy": 0.9064616511074501,
|
| 1309 |
+
"eval_loss": 0.6833410263061523,
|
| 1310 |
+
"eval_runtime": 15.9737,
|
| 1311 |
+
"eval_samples_per_second": 341.999,
|
| 1312 |
+
"eval_steps_per_second": 42.758,
|
| 1313 |
+
"step": 43500
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 6.72,
|
| 1317 |
+
"learning_rate": 6.9774621796850885e-06,
|
| 1318 |
+
"loss": 0.0268,
|
| 1319 |
+
"step": 44000
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 6.72,
|
| 1323 |
+
"eval_accuracy": 0.9035328574043566,
|
| 1324 |
+
"eval_loss": 0.7155495882034302,
|
| 1325 |
+
"eval_runtime": 16.0634,
|
| 1326 |
+
"eval_samples_per_second": 340.089,
|
| 1327 |
+
"eval_steps_per_second": 42.519,
|
| 1328 |
+
"step": 44000
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 6.8,
|
| 1332 |
+
"learning_rate": 6.8149688825335956e-06,
|
| 1333 |
+
"loss": 0.0293,
|
| 1334 |
+
"step": 44500
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 6.8,
|
| 1338 |
+
"eval_accuracy": 0.9015193117334798,
|
| 1339 |
+
"eval_loss": 0.7632263898849487,
|
| 1340 |
+
"eval_runtime": 15.9819,
|
| 1341 |
+
"eval_samples_per_second": 341.824,
|
| 1342 |
+
"eval_steps_per_second": 42.736,
|
| 1343 |
+
"step": 44500
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 6.87,
|
| 1347 |
+
"learning_rate": 6.6524755853821034e-06,
|
| 1348 |
+
"loss": 0.0289,
|
| 1349 |
+
"step": 45000
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 6.87,
|
| 1353 |
+
"eval_accuracy": 0.9033498077979132,
|
| 1354 |
+
"eval_loss": 0.7229273319244385,
|
| 1355 |
+
"eval_runtime": 16.0622,
|
| 1356 |
+
"eval_samples_per_second": 340.116,
|
| 1357 |
+
"eval_steps_per_second": 42.522,
|
| 1358 |
+
"step": 45000
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 6.95,
|
| 1362 |
+
"learning_rate": 6.4899822882306105e-06,
|
| 1363 |
+
"loss": 0.0286,
|
| 1364 |
+
"step": 45500
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 6.95,
|
| 1368 |
+
"eval_accuracy": 0.90536335346879,
|
| 1369 |
+
"eval_loss": 0.6670674681663513,
|
| 1370 |
+
"eval_runtime": 15.9524,
|
| 1371 |
+
"eval_samples_per_second": 342.456,
|
| 1372 |
+
"eval_steps_per_second": 42.815,
|
| 1373 |
+
"step": 45500
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 7.03,
|
| 1377 |
+
"learning_rate": 6.327488991079118e-06,
|
| 1378 |
+
"loss": 0.024,
|
| 1379 |
+
"step": 46000
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 7.03,
|
| 1383 |
+
"eval_accuracy": 0.9033498077979132,
|
| 1384 |
+
"eval_loss": 0.7658922672271729,
|
| 1385 |
+
"eval_runtime": 16.0192,
|
| 1386 |
+
"eval_samples_per_second": 341.028,
|
| 1387 |
+
"eval_steps_per_second": 42.636,
|
| 1388 |
+
"step": 46000
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 7.1,
|
| 1392 |
+
"learning_rate": 6.164995693927625e-06,
|
| 1393 |
+
"loss": 0.0141,
|
| 1394 |
+
"step": 46500
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 7.1,
|
| 1398 |
+
"eval_accuracy": 0.9022515101592532,
|
| 1399 |
+
"eval_loss": 0.7980563044548035,
|
| 1400 |
+
"eval_runtime": 16.0395,
|
| 1401 |
+
"eval_samples_per_second": 340.597,
|
| 1402 |
+
"eval_steps_per_second": 42.582,
|
| 1403 |
+
"step": 46500
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 7.18,
|
| 1407 |
+
"learning_rate": 6.002502396776133e-06,
|
| 1408 |
+
"loss": 0.0212,
|
| 1409 |
+
"step": 47000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 7.18,
|
| 1413 |
+
"eval_accuracy": 0.9088412959912137,
|
| 1414 |
+
"eval_loss": 0.7021394968032837,
|
| 1415 |
+
"eval_runtime": 15.9988,
|
| 1416 |
+
"eval_samples_per_second": 341.463,
|
| 1417 |
+
"eval_steps_per_second": 42.691,
|
| 1418 |
+
"step": 47000
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 7.26,
|
| 1422 |
+
"learning_rate": 5.84000909962464e-06,
|
| 1423 |
+
"loss": 0.0183,
|
| 1424 |
+
"step": 47500
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 7.26,
|
| 1428 |
+
"eval_accuracy": 0.909573494416987,
|
| 1429 |
+
"eval_loss": 0.7122032046318054,
|
| 1430 |
+
"eval_runtime": 16.0037,
|
| 1431 |
+
"eval_samples_per_second": 341.358,
|
| 1432 |
+
"eval_steps_per_second": 42.678,
|
| 1433 |
+
"step": 47500
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 7.33,
|
| 1437 |
+
"learning_rate": 5.677515802473148e-06,
|
| 1438 |
+
"loss": 0.0221,
|
| 1439 |
+
"step": 48000
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 7.33,
|
| 1443 |
+
"eval_accuracy": 0.9064616511074501,
|
| 1444 |
+
"eval_loss": 0.7080280780792236,
|
| 1445 |
+
"eval_runtime": 16.0516,
|
| 1446 |
+
"eval_samples_per_second": 340.339,
|
| 1447 |
+
"eval_steps_per_second": 42.55,
|
| 1448 |
+
"step": 48000
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 7.41,
|
| 1452 |
+
"learning_rate": 5.515022505321655e-06,
|
| 1453 |
+
"loss": 0.0146,
|
| 1454 |
+
"step": 48500
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 7.41,
|
| 1458 |
+
"eval_accuracy": 0.9073768991396669,
|
| 1459 |
+
"eval_loss": 0.7343564033508301,
|
| 1460 |
+
"eval_runtime": 16.0182,
|
| 1461 |
+
"eval_samples_per_second": 341.049,
|
| 1462 |
+
"eval_steps_per_second": 42.639,
|
| 1463 |
+
"step": 48500
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 7.48,
|
| 1467 |
+
"learning_rate": 5.352529208170163e-06,
|
| 1468 |
+
"loss": 0.0181,
|
| 1469 |
+
"step": 49000
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 7.48,
|
| 1473 |
+
"eval_accuracy": 0.9104887424492037,
|
| 1474 |
+
"eval_loss": 0.7273329496383667,
|
| 1475 |
+
"eval_runtime": 15.9794,
|
| 1476 |
+
"eval_samples_per_second": 341.877,
|
| 1477 |
+
"eval_steps_per_second": 42.742,
|
| 1478 |
+
"step": 49000
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 7.56,
|
| 1482 |
+
"learning_rate": 5.19003591101867e-06,
|
| 1483 |
+
"loss": 0.0161,
|
| 1484 |
+
"step": 49500
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 7.56,
|
| 1488 |
+
"eval_accuracy": 0.9082921471718836,
|
| 1489 |
+
"eval_loss": 0.73320472240448,
|
| 1490 |
+
"eval_runtime": 15.9747,
|
| 1491 |
+
"eval_samples_per_second": 341.978,
|
| 1492 |
+
"eval_steps_per_second": 42.755,
|
| 1493 |
+
"step": 49500
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 7.64,
|
| 1497 |
+
"learning_rate": 5.027542613867178e-06,
|
| 1498 |
+
"loss": 0.0193,
|
| 1499 |
+
"step": 50000
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 7.64,
|
| 1503 |
+
"eval_accuracy": 0.9093904448105437,
|
| 1504 |
+
"eval_loss": 0.7117404937744141,
|
| 1505 |
+
"eval_runtime": 16.028,
|
| 1506 |
+
"eval_samples_per_second": 340.841,
|
| 1507 |
+
"eval_steps_per_second": 42.613,
|
| 1508 |
+
"step": 50000
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 7.71,
|
| 1512 |
+
"learning_rate": 4.865049316715686e-06,
|
| 1513 |
+
"loss": 0.0165,
|
| 1514 |
+
"step": 50500
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 7.71,
|
| 1518 |
+
"eval_accuracy": 0.9070107999267801,
|
| 1519 |
+
"eval_loss": 0.7796688675880432,
|
| 1520 |
+
"eval_runtime": 16.0196,
|
| 1521 |
+
"eval_samples_per_second": 341.02,
|
| 1522 |
+
"eval_steps_per_second": 42.635,
|
| 1523 |
+
"step": 50500
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 7.79,
|
| 1527 |
+
"learning_rate": 4.702556019564194e-06,
|
| 1528 |
+
"loss": 0.0173,
|
| 1529 |
+
"step": 51000
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 7.79,
|
| 1533 |
+
"eval_accuracy": 0.9106717920556471,
|
| 1534 |
+
"eval_loss": 0.7127623558044434,
|
| 1535 |
+
"eval_runtime": 16.0489,
|
| 1536 |
+
"eval_samples_per_second": 340.398,
|
| 1537 |
+
"eval_steps_per_second": 42.558,
|
| 1538 |
+
"step": 51000
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 7.87,
|
| 1542 |
+
"learning_rate": 4.540062722412701e-06,
|
| 1543 |
+
"loss": 0.0158,
|
| 1544 |
+
"step": 51500
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 7.87,
|
| 1548 |
+
"eval_accuracy": 0.9110378912685337,
|
| 1549 |
+
"eval_loss": 0.7419800758361816,
|
| 1550 |
+
"eval_runtime": 15.9669,
|
| 1551 |
+
"eval_samples_per_second": 342.144,
|
| 1552 |
+
"eval_steps_per_second": 42.776,
|
| 1553 |
+
"step": 51500
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 7.94,
|
| 1557 |
+
"learning_rate": 4.377569425261209e-06,
|
| 1558 |
+
"loss": 0.0157,
|
| 1559 |
+
"step": 52000
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 7.94,
|
| 1563 |
+
"eval_accuracy": 0.9128683873329673,
|
| 1564 |
+
"eval_loss": 0.7197983860969543,
|
| 1565 |
+
"eval_runtime": 16.0117,
|
| 1566 |
+
"eval_samples_per_second": 341.189,
|
| 1567 |
+
"eval_steps_per_second": 42.656,
|
| 1568 |
+
"step": 52000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 8.02,
|
| 1572 |
+
"learning_rate": 4.215076128109716e-06,
|
| 1573 |
+
"loss": 0.0142,
|
| 1574 |
+
"step": 52500
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 8.02,
|
| 1578 |
+
"eval_accuracy": 0.9114039904814205,
|
| 1579 |
+
"eval_loss": 0.7032836675643921,
|
| 1580 |
+
"eval_runtime": 15.9817,
|
| 1581 |
+
"eval_samples_per_second": 341.829,
|
| 1582 |
+
"eval_steps_per_second": 42.736,
|
| 1583 |
+
"step": 52500
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 8.1,
|
| 1587 |
+
"learning_rate": 4.0525828309582235e-06,
|
| 1588 |
+
"loss": 0.0107,
|
| 1589 |
+
"step": 53000
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 8.1,
|
| 1593 |
+
"eval_accuracy": 0.9108548416620904,
|
| 1594 |
+
"eval_loss": 0.776072084903717,
|
| 1595 |
+
"eval_runtime": 16.015,
|
| 1596 |
+
"eval_samples_per_second": 341.118,
|
| 1597 |
+
"eval_steps_per_second": 42.648,
|
| 1598 |
+
"step": 53000
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 8.17,
|
| 1602 |
+
"learning_rate": 3.8900895338067306e-06,
|
| 1603 |
+
"loss": 0.0122,
|
| 1604 |
+
"step": 53500
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 8.17,
|
| 1608 |
+
"eval_accuracy": 0.9110378912685337,
|
| 1609 |
+
"eval_loss": 0.7777400612831116,
|
| 1610 |
+
"eval_runtime": 15.947,
|
| 1611 |
+
"eval_samples_per_second": 342.572,
|
| 1612 |
+
"eval_steps_per_second": 42.829,
|
| 1613 |
+
"step": 53500
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 8.25,
|
| 1617 |
+
"learning_rate": 3.727596236655238e-06,
|
| 1618 |
+
"loss": 0.013,
|
| 1619 |
+
"step": 54000
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 8.25,
|
| 1623 |
+
"eval_accuracy": 0.9075599487461102,
|
| 1624 |
+
"eval_loss": 0.7880035042762756,
|
| 1625 |
+
"eval_runtime": 15.9944,
|
| 1626 |
+
"eval_samples_per_second": 341.558,
|
| 1627 |
+
"eval_steps_per_second": 42.703,
|
| 1628 |
+
"step": 54000
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 8.32,
|
| 1632 |
+
"learning_rate": 3.5651029395037455e-06,
|
| 1633 |
+
"loss": 0.0092,
|
| 1634 |
+
"step": 54500
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 8.32,
|
| 1638 |
+
"eval_accuracy": 0.9073768991396669,
|
| 1639 |
+
"eval_loss": 0.8070389032363892,
|
| 1640 |
+
"eval_runtime": 15.9764,
|
| 1641 |
+
"eval_samples_per_second": 341.942,
|
| 1642 |
+
"eval_steps_per_second": 42.751,
|
| 1643 |
+
"step": 54500
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 8.4,
|
| 1647 |
+
"learning_rate": 3.402609642352253e-06,
|
| 1648 |
+
"loss": 0.0102,
|
| 1649 |
+
"step": 55000
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 8.4,
|
| 1653 |
+
"eval_accuracy": 0.9055464030752334,
|
| 1654 |
+
"eval_loss": 0.8112868666648865,
|
| 1655 |
+
"eval_runtime": 16.2072,
|
| 1656 |
+
"eval_samples_per_second": 337.073,
|
| 1657 |
+
"eval_steps_per_second": 42.142,
|
| 1658 |
+
"step": 55000
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 8.48,
|
| 1662 |
+
"learning_rate": 3.2401163452007604e-06,
|
| 1663 |
+
"loss": 0.0099,
|
| 1664 |
+
"step": 55500
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 8.48,
|
| 1668 |
+
"eval_accuracy": 0.9079260479589969,
|
| 1669 |
+
"eval_loss": 0.8152795433998108,
|
| 1670 |
+
"eval_runtime": 16.1355,
|
| 1671 |
+
"eval_samples_per_second": 338.57,
|
| 1672 |
+
"eval_steps_per_second": 42.329,
|
| 1673 |
+
"step": 55500
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 8.55,
|
| 1677 |
+
"learning_rate": 3.0776230480492682e-06,
|
| 1678 |
+
"loss": 0.0087,
|
| 1679 |
+
"step": 56000
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 8.55,
|
| 1683 |
+
"eval_accuracy": 0.9112209408749771,
|
| 1684 |
+
"eval_loss": 0.8045311570167542,
|
| 1685 |
+
"eval_runtime": 16.0534,
|
| 1686 |
+
"eval_samples_per_second": 340.302,
|
| 1687 |
+
"eval_steps_per_second": 42.546,
|
| 1688 |
+
"step": 56000
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 8.63,
|
| 1692 |
+
"learning_rate": 2.9151297508977757e-06,
|
| 1693 |
+
"loss": 0.0133,
|
| 1694 |
+
"step": 56500
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 8.63,
|
| 1698 |
+
"eval_accuracy": 0.9081090975654402,
|
| 1699 |
+
"eval_loss": 0.8173390030860901,
|
| 1700 |
+
"eval_runtime": 15.9908,
|
| 1701 |
+
"eval_samples_per_second": 341.634,
|
| 1702 |
+
"eval_steps_per_second": 42.712,
|
| 1703 |
+
"step": 56500
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 8.71,
|
| 1707 |
+
"learning_rate": 2.752636453746283e-06,
|
| 1708 |
+
"loss": 0.017,
|
| 1709 |
+
"step": 57000
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"epoch": 8.71,
|
| 1713 |
+
"eval_accuracy": 0.9101226432363171,
|
| 1714 |
+
"eval_loss": 0.7646049857139587,
|
| 1715 |
+
"eval_runtime": 16.0101,
|
| 1716 |
+
"eval_samples_per_second": 341.222,
|
| 1717 |
+
"eval_steps_per_second": 42.661,
|
| 1718 |
+
"step": 57000
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 8.78,
|
| 1722 |
+
"learning_rate": 2.5901431565947906e-06,
|
| 1723 |
+
"loss": 0.0093,
|
| 1724 |
+
"step": 57500
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 8.78,
|
| 1728 |
+
"eval_accuracy": 0.9119531393007505,
|
| 1729 |
+
"eval_loss": 0.7681165337562561,
|
| 1730 |
+
"eval_runtime": 15.9413,
|
| 1731 |
+
"eval_samples_per_second": 342.694,
|
| 1732 |
+
"eval_steps_per_second": 42.845,
|
| 1733 |
+
"step": 57500
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 8.86,
|
| 1737 |
+
"learning_rate": 2.427649859443298e-06,
|
| 1738 |
+
"loss": 0.0085,
|
| 1739 |
+
"step": 58000
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 8.86,
|
| 1743 |
+
"eval_accuracy": 0.9082921471718836,
|
| 1744 |
+
"eval_loss": 0.8067189455032349,
|
| 1745 |
+
"eval_runtime": 16.0314,
|
| 1746 |
+
"eval_samples_per_second": 340.768,
|
| 1747 |
+
"eval_steps_per_second": 42.604,
|
| 1748 |
+
"step": 58000
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 8.94,
|
| 1752 |
+
"learning_rate": 2.2651565622918055e-06,
|
| 1753 |
+
"loss": 0.0137,
|
| 1754 |
+
"step": 58500
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 8.94,
|
| 1758 |
+
"eval_accuracy": 0.9106717920556471,
|
| 1759 |
+
"eval_loss": 0.7645466327667236,
|
| 1760 |
+
"eval_runtime": 15.9969,
|
| 1761 |
+
"eval_samples_per_second": 341.504,
|
| 1762 |
+
"eval_steps_per_second": 42.696,
|
| 1763 |
+
"step": 58500
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 9.01,
|
| 1767 |
+
"learning_rate": 2.102663265140313e-06,
|
| 1768 |
+
"loss": 0.0099,
|
| 1769 |
+
"step": 59000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 9.01,
|
| 1773 |
+
"eval_accuracy": 0.9123192385136372,
|
| 1774 |
+
"eval_loss": 0.7742260098457336,
|
| 1775 |
+
"eval_runtime": 16.0529,
|
| 1776 |
+
"eval_samples_per_second": 340.313,
|
| 1777 |
+
"eval_steps_per_second": 42.547,
|
| 1778 |
+
"step": 59000
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 9.09,
|
| 1782 |
+
"learning_rate": 1.9401699679888204e-06,
|
| 1783 |
+
"loss": 0.0074,
|
| 1784 |
+
"step": 59500
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 9.09,
|
| 1788 |
+
"eval_accuracy": 0.90591250228812,
|
| 1789 |
+
"eval_loss": 0.8052135705947876,
|
| 1790 |
+
"eval_runtime": 15.9958,
|
| 1791 |
+
"eval_samples_per_second": 341.527,
|
| 1792 |
+
"eval_steps_per_second": 42.699,
|
| 1793 |
+
"step": 59500
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 9.16,
|
| 1797 |
+
"learning_rate": 1.777676670837328e-06,
|
| 1798 |
+
"loss": 0.0057,
|
| 1799 |
+
"step": 60000
|
| 1800 |
+
},
|
| 1801 |
+
{
|
| 1802 |
+
"epoch": 9.16,
|
| 1803 |
+
"eval_accuracy": 0.9132344865458539,
|
| 1804 |
+
"eval_loss": 0.790639340877533,
|
| 1805 |
+
"eval_runtime": 16.0441,
|
| 1806 |
+
"eval_samples_per_second": 340.499,
|
| 1807 |
+
"eval_steps_per_second": 42.57,
|
| 1808 |
+
"step": 60000
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 9.24,
|
| 1812 |
+
"learning_rate": 1.6151833736858355e-06,
|
| 1813 |
+
"loss": 0.0046,
|
| 1814 |
+
"step": 60500
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 9.24,
|
| 1818 |
+
"eval_accuracy": 0.9136005857587406,
|
| 1819 |
+
"eval_loss": 0.7914955019950867,
|
| 1820 |
+
"eval_runtime": 15.9828,
|
| 1821 |
+
"eval_samples_per_second": 341.806,
|
| 1822 |
+
"eval_steps_per_second": 42.734,
|
| 1823 |
+
"step": 60500
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 9.32,
|
| 1827 |
+
"learning_rate": 1.452690076534343e-06,
|
| 1828 |
+
"loss": 0.0036,
|
| 1829 |
+
"step": 61000
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 9.32,
|
| 1833 |
+
"eval_accuracy": 0.9101226432363171,
|
| 1834 |
+
"eval_loss": 0.8114162087440491,
|
| 1835 |
+
"eval_runtime": 16.0918,
|
| 1836 |
+
"eval_samples_per_second": 339.489,
|
| 1837 |
+
"eval_steps_per_second": 42.444,
|
| 1838 |
+
"step": 61000
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 9.39,
|
| 1842 |
+
"learning_rate": 1.2901967793828504e-06,
|
| 1843 |
+
"loss": 0.0036,
|
| 1844 |
+
"step": 61500
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 9.39,
|
| 1848 |
+
"eval_accuracy": 0.9086582463847702,
|
| 1849 |
+
"eval_loss": 0.8381993174552917,
|
| 1850 |
+
"eval_runtime": 16.0882,
|
| 1851 |
+
"eval_samples_per_second": 339.565,
|
| 1852 |
+
"eval_steps_per_second": 42.453,
|
| 1853 |
+
"step": 61500
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"epoch": 9.47,
|
| 1857 |
+
"learning_rate": 1.127703482231358e-06,
|
| 1858 |
+
"loss": 0.0079,
|
| 1859 |
+
"step": 62000
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 9.47,
|
| 1863 |
+
"eval_accuracy": 0.9110378912685337,
|
| 1864 |
+
"eval_loss": 0.8012856841087341,
|
| 1865 |
+
"eval_runtime": 16.0031,
|
| 1866 |
+
"eval_samples_per_second": 341.371,
|
| 1867 |
+
"eval_steps_per_second": 42.679,
|
| 1868 |
+
"step": 62000
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 9.55,
|
| 1872 |
+
"learning_rate": 9.652101850798656e-07,
|
| 1873 |
+
"loss": 0.0079,
|
| 1874 |
+
"step": 62500
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 9.55,
|
| 1878 |
+
"eval_accuracy": 0.9099395936298736,
|
| 1879 |
+
"eval_loss": 0.8116302490234375,
|
| 1880 |
+
"eval_runtime": 16.0181,
|
| 1881 |
+
"eval_samples_per_second": 341.052,
|
| 1882 |
+
"eval_steps_per_second": 42.639,
|
| 1883 |
+
"step": 62500
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 9.62,
|
| 1887 |
+
"learning_rate": 8.02716887928373e-07,
|
| 1888 |
+
"loss": 0.0054,
|
| 1889 |
+
"step": 63000
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 9.62,
|
| 1893 |
+
"eval_accuracy": 0.9110378912685337,
|
| 1894 |
+
"eval_loss": 0.8142715096473694,
|
| 1895 |
+
"eval_runtime": 16.0228,
|
| 1896 |
+
"eval_samples_per_second": 340.951,
|
| 1897 |
+
"eval_steps_per_second": 42.627,
|
| 1898 |
+
"step": 63000
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 9.7,
|
| 1902 |
+
"learning_rate": 6.402235907768806e-07,
|
| 1903 |
+
"loss": 0.0036,
|
| 1904 |
+
"step": 63500
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 9.7,
|
| 1908 |
+
"eval_accuracy": 0.9114039904814205,
|
| 1909 |
+
"eval_loss": 0.8137949109077454,
|
| 1910 |
+
"eval_runtime": 16.0052,
|
| 1911 |
+
"eval_samples_per_second": 341.327,
|
| 1912 |
+
"eval_steps_per_second": 42.674,
|
| 1913 |
+
"step": 63500
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 9.78,
|
| 1917 |
+
"learning_rate": 4.777302936253879e-07,
|
| 1918 |
+
"loss": 0.0053,
|
| 1919 |
+
"step": 64000
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 9.78,
|
| 1923 |
+
"eval_accuracy": 0.9106717920556471,
|
| 1924 |
+
"eval_loss": 0.8157973289489746,
|
| 1925 |
+
"eval_runtime": 16.048,
|
| 1926 |
+
"eval_samples_per_second": 340.415,
|
| 1927 |
+
"eval_steps_per_second": 42.56,
|
| 1928 |
+
"step": 64000
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 9.85,
|
| 1932 |
+
"learning_rate": 3.1523699647389547e-07,
|
| 1933 |
+
"loss": 0.0035,
|
| 1934 |
+
"step": 64500
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 9.85,
|
| 1938 |
+
"eval_accuracy": 0.9112209408749771,
|
| 1939 |
+
"eval_loss": 0.814604640007019,
|
| 1940 |
+
"eval_runtime": 16.0042,
|
| 1941 |
+
"eval_samples_per_second": 341.348,
|
| 1942 |
+
"eval_steps_per_second": 42.676,
|
| 1943 |
+
"step": 64500
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 9.93,
|
| 1947 |
+
"learning_rate": 1.5274369932240295e-07,
|
| 1948 |
+
"loss": 0.0051,
|
| 1949 |
+
"step": 65000
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 9.93,
|
| 1953 |
+
"eval_accuracy": 0.9106717920556471,
|
| 1954 |
+
"eval_loss": 0.8167830109596252,
|
| 1955 |
+
"eval_runtime": 16.0336,
|
| 1956 |
+
"eval_samples_per_second": 340.722,
|
| 1957 |
+
"eval_steps_per_second": 42.598,
|
| 1958 |
+
"step": 65000
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 10.0,
|
| 1962 |
+
"step": 65470,
|
| 1963 |
+
"total_flos": 6.88976031789312e+16,
|
| 1964 |
+
"train_loss": 0.10566671685770447,
|
| 1965 |
+
"train_runtime": 14393.9986,
|
| 1966 |
+
"train_samples_per_second": 72.769,
|
| 1967 |
+
"train_steps_per_second": 4.548
|
| 1968 |
+
}
|
| 1969 |
+
],
|
| 1970 |
+
"max_steps": 65470,
|
| 1971 |
+
"num_train_epochs": 10,
|
| 1972 |
+
"total_flos": 6.88976031789312e+16,
|
| 1973 |
+
"trial_name": null,
|
| 1974 |
+
"trial_params": null
|
| 1975 |
+
}
|