End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- runs/Feb08_00-21-15_dante/events.out.tfevents.1644352578.dante.3611933.2 +3 -0
- train_results.json +8 -0
- trainer_state.json +1831 -0
- wandb/run-20220208_002217-138c86e3/files/output.log +2 -2
- wandb/run-20220208_002217-138c86e3/files/wandb-summary.json +0 -0
- wandb/run-20220208_002217-138c86e3/logs/debug-internal.log +2 -2
- wandb/run-20220208_002217-138c86e3/run-138c86e3.wandb +2 -2
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_loss": 0.16563400626182556,
|
| 4 |
+
"eval_runtime": 381.3008,
|
| 5 |
+
"eval_samples": 5437,
|
| 6 |
+
"eval_samples_per_second": 14.259,
|
| 7 |
+
"eval_steps_per_second": 0.892,
|
| 8 |
+
"eval_wer": 0.09330600958945748,
|
| 9 |
+
"train_loss": 0.08625626374438786,
|
| 10 |
+
"train_runtime": 75520.7008,
|
| 11 |
+
"train_samples": 49645,
|
| 12 |
+
"train_samples_per_second": 9.861,
|
| 13 |
+
"train_steps_per_second": 0.308
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_loss": 0.16563400626182556,
|
| 4 |
+
"eval_runtime": 381.3008,
|
| 5 |
+
"eval_samples": 5437,
|
| 6 |
+
"eval_samples_per_second": 14.259,
|
| 7 |
+
"eval_steps_per_second": 0.892,
|
| 8 |
+
"eval_wer": 0.09330600958945748
|
| 9 |
+
}
|
runs/Feb08_00-21-15_dante/events.out.tfevents.1644352578.dante.3611933.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d06bec12e5db21a7b32296666cd0ec65b0369bca3f31b5a31700b2610732c1b8
|
| 3 |
+
size 364
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"train_loss": 0.08625626374438786,
|
| 4 |
+
"train_runtime": 75520.7008,
|
| 5 |
+
"train_samples": 49645,
|
| 6 |
+
"train_samples_per_second": 9.861,
|
| 7 |
+
"train_steps_per_second": 0.308
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1831 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 14.999677731227845,
|
| 5 |
+
"global_step": 23265,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.06,
|
| 12 |
+
"learning_rate": 1e-05,
|
| 13 |
+
"loss": 0.0958,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.13,
|
| 18 |
+
"learning_rate": 1.9900000000000003e-05,
|
| 19 |
+
"loss": 0.0914,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.19,
|
| 24 |
+
"learning_rate": 2.9900000000000002e-05,
|
| 25 |
+
"loss": 0.0893,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.26,
|
| 30 |
+
"learning_rate": 3.99e-05,
|
| 31 |
+
"loss": 0.0925,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.32,
|
| 36 |
+
"learning_rate": 4.99e-05,
|
| 37 |
+
"loss": 0.0969,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.32,
|
| 42 |
+
"eval_loss": 0.17728671431541443,
|
| 43 |
+
"eval_runtime": 380.6151,
|
| 44 |
+
"eval_samples_per_second": 14.285,
|
| 45 |
+
"eval_steps_per_second": 0.893,
|
| 46 |
+
"eval_wer": 0.10537617539490327,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.39,
|
| 51 |
+
"learning_rate": 4.978256094882495e-05,
|
| 52 |
+
"loss": 0.0982,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.45,
|
| 57 |
+
"learning_rate": 4.956292554359763e-05,
|
| 58 |
+
"loss": 0.0982,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.52,
|
| 63 |
+
"learning_rate": 4.9343290138370304e-05,
|
| 64 |
+
"loss": 0.0939,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.58,
|
| 69 |
+
"learning_rate": 4.9125851087195255e-05,
|
| 70 |
+
"loss": 0.0957,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.64,
|
| 75 |
+
"learning_rate": 4.8906215681967934e-05,
|
| 76 |
+
"loss": 0.0929,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.64,
|
| 81 |
+
"eval_loss": 0.16721801459789276,
|
| 82 |
+
"eval_runtime": 381.6558,
|
| 83 |
+
"eval_samples_per_second": 14.246,
|
| 84 |
+
"eval_steps_per_second": 0.891,
|
| 85 |
+
"eval_wer": 0.1061409787521939,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.71,
|
| 90 |
+
"learning_rate": 4.8688776630792884e-05,
|
| 91 |
+
"loss": 0.0938,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.77,
|
| 96 |
+
"learning_rate": 4.8469141225565564e-05,
|
| 97 |
+
"loss": 0.088,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.84,
|
| 102 |
+
"learning_rate": 4.8251702174390514e-05,
|
| 103 |
+
"loss": 0.1015,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.9,
|
| 108 |
+
"learning_rate": 4.803206676916319e-05,
|
| 109 |
+
"loss": 0.0963,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.97,
|
| 114 |
+
"learning_rate": 4.7812431363935866e-05,
|
| 115 |
+
"loss": 0.1018,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.97,
|
| 120 |
+
"eval_loss": 0.17695783078670502,
|
| 121 |
+
"eval_runtime": 379.2394,
|
| 122 |
+
"eval_samples_per_second": 14.337,
|
| 123 |
+
"eval_steps_per_second": 0.897,
|
| 124 |
+
"eval_wer": 0.10669987351329091,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.03,
|
| 129 |
+
"learning_rate": 4.7592795958708545e-05,
|
| 130 |
+
"loss": 0.0899,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.1,
|
| 135 |
+
"learning_rate": 4.7373160553481224e-05,
|
| 136 |
+
"loss": 0.0898,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.16,
|
| 141 |
+
"learning_rate": 4.7153525148253903e-05,
|
| 142 |
+
"loss": 0.0884,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.22,
|
| 147 |
+
"learning_rate": 4.6933889743026576e-05,
|
| 148 |
+
"loss": 0.0914,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.29,
|
| 153 |
+
"learning_rate": 4.6714254337799255e-05,
|
| 154 |
+
"loss": 0.0871,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.29,
|
| 159 |
+
"eval_loss": 0.18323425948619843,
|
| 160 |
+
"eval_runtime": 384.6744,
|
| 161 |
+
"eval_samples_per_second": 14.134,
|
| 162 |
+
"eval_steps_per_second": 0.884,
|
| 163 |
+
"eval_wer": 0.1086805181052487,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.35,
|
| 168 |
+
"learning_rate": 4.6494618932571934e-05,
|
| 169 |
+
"loss": 0.0918,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.42,
|
| 174 |
+
"learning_rate": 4.6274983527344614e-05,
|
| 175 |
+
"loss": 0.0909,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.48,
|
| 180 |
+
"learning_rate": 4.6057544476169564e-05,
|
| 181 |
+
"loss": 0.0988,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.55,
|
| 186 |
+
"learning_rate": 4.5837909070942236e-05,
|
| 187 |
+
"loss": 0.0967,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.61,
|
| 192 |
+
"learning_rate": 4.5618273665714916e-05,
|
| 193 |
+
"loss": 0.0908,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.61,
|
| 198 |
+
"eval_loss": 0.1829613745212555,
|
| 199 |
+
"eval_runtime": 380.5109,
|
| 200 |
+
"eval_samples_per_second": 14.289,
|
| 201 |
+
"eval_steps_per_second": 0.894,
|
| 202 |
+
"eval_wer": 0.11013168344985146,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 1.68,
|
| 207 |
+
"learning_rate": 4.539863826048759e-05,
|
| 208 |
+
"loss": 0.0901,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.74,
|
| 213 |
+
"learning_rate": 4.5179002855260274e-05,
|
| 214 |
+
"loss": 0.0934,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 1.81,
|
| 219 |
+
"learning_rate": 4.495936745003295e-05,
|
| 220 |
+
"loss": 0.0958,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 1.87,
|
| 225 |
+
"learning_rate": 4.4739732044805626e-05,
|
| 226 |
+
"loss": 0.0909,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.93,
|
| 231 |
+
"learning_rate": 4.45200966395783e-05,
|
| 232 |
+
"loss": 0.0975,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.93,
|
| 237 |
+
"eval_loss": 0.1847657710313797,
|
| 238 |
+
"eval_runtime": 376.8362,
|
| 239 |
+
"eval_samples_per_second": 14.428,
|
| 240 |
+
"eval_steps_per_second": 0.902,
|
| 241 |
+
"eval_wer": 0.11004343690862561,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.0,
|
| 246 |
+
"learning_rate": 4.4300461234350984e-05,
|
| 247 |
+
"loss": 0.0977,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.06,
|
| 252 |
+
"learning_rate": 4.408082582912366e-05,
|
| 253 |
+
"loss": 0.0949,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.13,
|
| 258 |
+
"learning_rate": 4.3861190423896336e-05,
|
| 259 |
+
"loss": 0.0916,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.19,
|
| 264 |
+
"learning_rate": 4.364155501866901e-05,
|
| 265 |
+
"loss": 0.0963,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.26,
|
| 270 |
+
"learning_rate": 4.342191961344169e-05,
|
| 271 |
+
"loss": 0.0936,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.26,
|
| 276 |
+
"eval_loss": 0.18531930446624756,
|
| 277 |
+
"eval_runtime": 380.7297,
|
| 278 |
+
"eval_samples_per_second": 14.28,
|
| 279 |
+
"eval_steps_per_second": 0.893,
|
| 280 |
+
"eval_wer": 0.1112690833145401,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.32,
|
| 285 |
+
"learning_rate": 4.320228420821437e-05,
|
| 286 |
+
"loss": 0.0955,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 2.39,
|
| 291 |
+
"learning_rate": 4.298264880298704e-05,
|
| 292 |
+
"loss": 0.093,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 2.45,
|
| 297 |
+
"learning_rate": 4.276301339775972e-05,
|
| 298 |
+
"loss": 0.0924,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.51,
|
| 303 |
+
"learning_rate": 4.254557434658467e-05,
|
| 304 |
+
"loss": 0.1065,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.58,
|
| 309 |
+
"learning_rate": 4.232593894135735e-05,
|
| 310 |
+
"loss": 0.1025,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 2.58,
|
| 315 |
+
"eval_loss": 0.19584240019321442,
|
| 316 |
+
"eval_runtime": 374.8969,
|
| 317 |
+
"eval_samples_per_second": 14.503,
|
| 318 |
+
"eval_steps_per_second": 0.907,
|
| 319 |
+
"eval_wer": 0.11489699667604694,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 2.64,
|
| 324 |
+
"learning_rate": 4.210630353613003e-05,
|
| 325 |
+
"loss": 0.101,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.71,
|
| 330 |
+
"learning_rate": 4.18866681309027e-05,
|
| 331 |
+
"loss": 0.1038,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 2.77,
|
| 336 |
+
"learning_rate": 4.166703272567538e-05,
|
| 337 |
+
"loss": 0.1043,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.84,
|
| 342 |
+
"learning_rate": 4.144739732044806e-05,
|
| 343 |
+
"loss": 0.1057,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 2.9,
|
| 348 |
+
"learning_rate": 4.122776191522074e-05,
|
| 349 |
+
"loss": 0.0989,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 2.9,
|
| 354 |
+
"eval_loss": 0.17756135761737823,
|
| 355 |
+
"eval_runtime": 376.0235,
|
| 356 |
+
"eval_samples_per_second": 14.459,
|
| 357 |
+
"eval_steps_per_second": 0.904,
|
| 358 |
+
"eval_wer": 0.11230843146675556,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 2.97,
|
| 363 |
+
"learning_rate": 4.100812650999341e-05,
|
| 364 |
+
"loss": 0.0944,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 3.03,
|
| 369 |
+
"learning_rate": 4.078849110476609e-05,
|
| 370 |
+
"loss": 0.0911,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 3.09,
|
| 375 |
+
"learning_rate": 4.056885569953877e-05,
|
| 376 |
+
"loss": 0.0903,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 3.16,
|
| 381 |
+
"learning_rate": 4.034922029431145e-05,
|
| 382 |
+
"loss": 0.093,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 3.22,
|
| 387 |
+
"learning_rate": 4.012958488908412e-05,
|
| 388 |
+
"loss": 0.0946,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 3.22,
|
| 393 |
+
"eval_loss": 0.18248289823532104,
|
| 394 |
+
"eval_runtime": 377.2331,
|
| 395 |
+
"eval_samples_per_second": 14.413,
|
| 396 |
+
"eval_steps_per_second": 0.901,
|
| 397 |
+
"eval_wer": 0.10971986625746419,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 3.29,
|
| 402 |
+
"learning_rate": 3.99099494838568e-05,
|
| 403 |
+
"loss": 0.0885,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 3.35,
|
| 408 |
+
"learning_rate": 3.969031407862948e-05,
|
| 409 |
+
"loss": 0.0951,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 3.42,
|
| 414 |
+
"learning_rate": 3.947067867340216e-05,
|
| 415 |
+
"loss": 0.0829,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 3.48,
|
| 420 |
+
"learning_rate": 3.925104326817483e-05,
|
| 421 |
+
"loss": 0.0864,
|
| 422 |
+
"step": 5400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 3.55,
|
| 426 |
+
"learning_rate": 3.903140786294751e-05,
|
| 427 |
+
"loss": 0.0859,
|
| 428 |
+
"step": 5500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 3.55,
|
| 432 |
+
"eval_loss": 0.1863715648651123,
|
| 433 |
+
"eval_runtime": 375.9814,
|
| 434 |
+
"eval_samples_per_second": 14.461,
|
| 435 |
+
"eval_steps_per_second": 0.904,
|
| 436 |
+
"eval_wer": 0.10721954758939865,
|
| 437 |
+
"step": 5500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 3.61,
|
| 441 |
+
"learning_rate": 3.881177245772018e-05,
|
| 442 |
+
"loss": 0.0835,
|
| 443 |
+
"step": 5600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 3.67,
|
| 447 |
+
"learning_rate": 3.859213705249287e-05,
|
| 448 |
+
"loss": 0.0818,
|
| 449 |
+
"step": 5700
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 3.74,
|
| 453 |
+
"learning_rate": 3.837250164726554e-05,
|
| 454 |
+
"loss": 0.0829,
|
| 455 |
+
"step": 5800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 3.8,
|
| 459 |
+
"learning_rate": 3.815286624203822e-05,
|
| 460 |
+
"loss": 0.0811,
|
| 461 |
+
"step": 5900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 3.87,
|
| 465 |
+
"learning_rate": 3.793323083681089e-05,
|
| 466 |
+
"loss": 0.0867,
|
| 467 |
+
"step": 6000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 3.87,
|
| 471 |
+
"eval_loss": 0.18856672942638397,
|
| 472 |
+
"eval_runtime": 378.3989,
|
| 473 |
+
"eval_samples_per_second": 14.368,
|
| 474 |
+
"eval_steps_per_second": 0.899,
|
| 475 |
+
"eval_wer": 0.10813142851539902,
|
| 476 |
+
"step": 6000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 3.93,
|
| 480 |
+
"learning_rate": 3.771359543158357e-05,
|
| 481 |
+
"loss": 0.084,
|
| 482 |
+
"step": 6100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 4.0,
|
| 486 |
+
"learning_rate": 3.749396002635625e-05,
|
| 487 |
+
"loss": 0.0865,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 4.06,
|
| 492 |
+
"learning_rate": 3.7274324621128924e-05,
|
| 493 |
+
"loss": 0.0816,
|
| 494 |
+
"step": 6300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 4.13,
|
| 498 |
+
"learning_rate": 3.70546892159016e-05,
|
| 499 |
+
"loss": 0.0808,
|
| 500 |
+
"step": 6400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 4.19,
|
| 504 |
+
"learning_rate": 3.683505381067428e-05,
|
| 505 |
+
"loss": 0.0783,
|
| 506 |
+
"step": 6500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 4.19,
|
| 510 |
+
"eval_loss": 0.18829604983329773,
|
| 511 |
+
"eval_runtime": 376.7422,
|
| 512 |
+
"eval_samples_per_second": 14.432,
|
| 513 |
+
"eval_steps_per_second": 0.902,
|
| 514 |
+
"eval_wer": 0.10633708217714022,
|
| 515 |
+
"step": 6500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 4.26,
|
| 519 |
+
"learning_rate": 3.661541840544696e-05,
|
| 520 |
+
"loss": 0.0808,
|
| 521 |
+
"step": 6600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 4.32,
|
| 525 |
+
"learning_rate": 3.6395783000219634e-05,
|
| 526 |
+
"loss": 0.0824,
|
| 527 |
+
"step": 6700
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 4.38,
|
| 531 |
+
"learning_rate": 3.617614759499231e-05,
|
| 532 |
+
"loss": 0.084,
|
| 533 |
+
"step": 6800
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 4.45,
|
| 537 |
+
"learning_rate": 3.595651218976499e-05,
|
| 538 |
+
"loss": 0.0831,
|
| 539 |
+
"step": 6900
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 4.51,
|
| 543 |
+
"learning_rate": 3.573687678453767e-05,
|
| 544 |
+
"loss": 0.0804,
|
| 545 |
+
"step": 7000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 4.51,
|
| 549 |
+
"eval_loss": 0.18314477801322937,
|
| 550 |
+
"eval_runtime": 381.141,
|
| 551 |
+
"eval_samples_per_second": 14.265,
|
| 552 |
+
"eval_steps_per_second": 0.892,
|
| 553 |
+
"eval_wer": 0.10627825114965633,
|
| 554 |
+
"step": 7000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 4.58,
|
| 558 |
+
"learning_rate": 3.551943773336262e-05,
|
| 559 |
+
"loss": 0.0816,
|
| 560 |
+
"step": 7100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 4.64,
|
| 564 |
+
"learning_rate": 3.5299802328135294e-05,
|
| 565 |
+
"loss": 0.0814,
|
| 566 |
+
"step": 7200
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 4.71,
|
| 570 |
+
"learning_rate": 3.5080166922907974e-05,
|
| 571 |
+
"loss": 0.0773,
|
| 572 |
+
"step": 7300
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 4.77,
|
| 576 |
+
"learning_rate": 3.486053151768065e-05,
|
| 577 |
+
"loss": 0.086,
|
| 578 |
+
"step": 7400
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 4.84,
|
| 582 |
+
"learning_rate": 3.464089611245333e-05,
|
| 583 |
+
"loss": 0.0797,
|
| 584 |
+
"step": 7500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 4.84,
|
| 588 |
+
"eval_loss": 0.18838059902191162,
|
| 589 |
+
"eval_runtime": 376.2869,
|
| 590 |
+
"eval_samples_per_second": 14.449,
|
| 591 |
+
"eval_steps_per_second": 0.904,
|
| 592 |
+
"eval_wer": 0.1058272132722798,
|
| 593 |
+
"step": 7500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 4.9,
|
| 597 |
+
"learning_rate": 3.4421260707226005e-05,
|
| 598 |
+
"loss": 0.0841,
|
| 599 |
+
"step": 7600
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 4.96,
|
| 603 |
+
"learning_rate": 3.4201625301998684e-05,
|
| 604 |
+
"loss": 0.0817,
|
| 605 |
+
"step": 7700
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 5.03,
|
| 609 |
+
"learning_rate": 3.398198989677136e-05,
|
| 610 |
+
"loss": 0.0808,
|
| 611 |
+
"step": 7800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 5.09,
|
| 615 |
+
"learning_rate": 3.376235449154404e-05,
|
| 616 |
+
"loss": 0.0758,
|
| 617 |
+
"step": 7900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 5.16,
|
| 621 |
+
"learning_rate": 3.3542719086316715e-05,
|
| 622 |
+
"loss": 0.0705,
|
| 623 |
+
"step": 8000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 5.16,
|
| 627 |
+
"eval_loss": 0.1801859736442566,
|
| 628 |
+
"eval_runtime": 378.8319,
|
| 629 |
+
"eval_samples_per_second": 14.352,
|
| 630 |
+
"eval_steps_per_second": 0.897,
|
| 631 |
+
"eval_wer": 0.10566052536107542,
|
| 632 |
+
"step": 8000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 5.22,
|
| 636 |
+
"learning_rate": 3.3323083681089394e-05,
|
| 637 |
+
"loss": 0.0804,
|
| 638 |
+
"step": 8100
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 5.29,
|
| 642 |
+
"learning_rate": 3.310344827586207e-05,
|
| 643 |
+
"loss": 0.0753,
|
| 644 |
+
"step": 8200
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 5.35,
|
| 648 |
+
"learning_rate": 3.288381287063475e-05,
|
| 649 |
+
"loss": 0.072,
|
| 650 |
+
"step": 8300
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 5.42,
|
| 654 |
+
"learning_rate": 3.2664177465407425e-05,
|
| 655 |
+
"loss": 0.078,
|
| 656 |
+
"step": 8400
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 5.48,
|
| 660 |
+
"learning_rate": 3.2444542060180104e-05,
|
| 661 |
+
"loss": 0.0795,
|
| 662 |
+
"step": 8500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 5.48,
|
| 666 |
+
"eval_loss": 0.18536211550235748,
|
| 667 |
+
"eval_runtime": 379.793,
|
| 668 |
+
"eval_samples_per_second": 14.316,
|
| 669 |
+
"eval_steps_per_second": 0.895,
|
| 670 |
+
"eval_wer": 0.10382695833782737,
|
| 671 |
+
"step": 8500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 5.54,
|
| 675 |
+
"learning_rate": 3.222490665495278e-05,
|
| 676 |
+
"loss": 0.0748,
|
| 677 |
+
"step": 8600
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 5.61,
|
| 681 |
+
"learning_rate": 3.2005271249725456e-05,
|
| 682 |
+
"loss": 0.0731,
|
| 683 |
+
"step": 8700
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 5.67,
|
| 687 |
+
"learning_rate": 3.1785635844498135e-05,
|
| 688 |
+
"loss": 0.0766,
|
| 689 |
+
"step": 8800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 5.74,
|
| 693 |
+
"learning_rate": 3.156600043927081e-05,
|
| 694 |
+
"loss": 0.072,
|
| 695 |
+
"step": 8900
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 5.8,
|
| 699 |
+
"learning_rate": 3.134636503404349e-05,
|
| 700 |
+
"loss": 0.0711,
|
| 701 |
+
"step": 9000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 5.8,
|
| 705 |
+
"eval_loss": 0.17664773762226105,
|
| 706 |
+
"eval_runtime": 376.7989,
|
| 707 |
+
"eval_samples_per_second": 14.429,
|
| 708 |
+
"eval_steps_per_second": 0.902,
|
| 709 |
+
"eval_wer": 0.10324845323423573,
|
| 710 |
+
"step": 9000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 5.87,
|
| 714 |
+
"learning_rate": 3.1126729628816166e-05,
|
| 715 |
+
"loss": 0.0951,
|
| 716 |
+
"step": 9100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 5.93,
|
| 720 |
+
"learning_rate": 3.0907094223588845e-05,
|
| 721 |
+
"loss": 0.095,
|
| 722 |
+
"step": 9200
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 6.0,
|
| 726 |
+
"learning_rate": 3.068745881836152e-05,
|
| 727 |
+
"loss": 0.0945,
|
| 728 |
+
"step": 9300
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 6.06,
|
| 732 |
+
"learning_rate": 3.04678234131342e-05,
|
| 733 |
+
"loss": 0.0892,
|
| 734 |
+
"step": 9400
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 6.13,
|
| 738 |
+
"learning_rate": 3.0248188007906873e-05,
|
| 739 |
+
"loss": 0.0973,
|
| 740 |
+
"step": 9500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 6.13,
|
| 744 |
+
"eval_loss": 0.16625218093395233,
|
| 745 |
+
"eval_runtime": 378.912,
|
| 746 |
+
"eval_samples_per_second": 14.349,
|
| 747 |
+
"eval_steps_per_second": 0.897,
|
| 748 |
+
"eval_wer": 0.10142469138223499,
|
| 749 |
+
"step": 9500
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 6.19,
|
| 753 |
+
"learning_rate": 3.0028552602679556e-05,
|
| 754 |
+
"loss": 0.0911,
|
| 755 |
+
"step": 9600
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 6.25,
|
| 759 |
+
"learning_rate": 2.9808917197452228e-05,
|
| 760 |
+
"loss": 0.0897,
|
| 761 |
+
"step": 9700
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 6.32,
|
| 765 |
+
"learning_rate": 2.958928179222491e-05,
|
| 766 |
+
"loss": 0.085,
|
| 767 |
+
"step": 9800
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 6.38,
|
| 771 |
+
"learning_rate": 2.9369646386997583e-05,
|
| 772 |
+
"loss": 0.0897,
|
| 773 |
+
"step": 9900
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 6.45,
|
| 777 |
+
"learning_rate": 2.9150010981770266e-05,
|
| 778 |
+
"loss": 0.087,
|
| 779 |
+
"step": 10000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 6.45,
|
| 783 |
+
"eval_loss": 0.1664039045572281,
|
| 784 |
+
"eval_runtime": 378.5078,
|
| 785 |
+
"eval_samples_per_second": 14.364,
|
| 786 |
+
"eval_steps_per_second": 0.898,
|
| 787 |
+
"eval_wer": 0.10139527586849305,
|
| 788 |
+
"step": 10000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 6.51,
|
| 792 |
+
"learning_rate": 2.893037557654294e-05,
|
| 793 |
+
"loss": 0.091,
|
| 794 |
+
"step": 10100
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 6.58,
|
| 798 |
+
"learning_rate": 2.871074017131562e-05,
|
| 799 |
+
"loss": 0.0877,
|
| 800 |
+
"step": 10200
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 6.64,
|
| 804 |
+
"learning_rate": 2.8491104766088293e-05,
|
| 805 |
+
"loss": 0.0865,
|
| 806 |
+
"step": 10300
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 6.71,
|
| 810 |
+
"learning_rate": 2.8271469360860976e-05,
|
| 811 |
+
"loss": 0.0904,
|
| 812 |
+
"step": 10400
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 6.77,
|
| 816 |
+
"learning_rate": 2.805183395563365e-05,
|
| 817 |
+
"loss": 0.0962,
|
| 818 |
+
"step": 10500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 6.77,
|
| 822 |
+
"eval_loss": 0.16310061514377594,
|
| 823 |
+
"eval_runtime": 377.2788,
|
| 824 |
+
"eval_samples_per_second": 14.411,
|
| 825 |
+
"eval_steps_per_second": 0.901,
|
| 826 |
+
"eval_wer": 0.10092462764862188,
|
| 827 |
+
"step": 10500
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 6.83,
|
| 831 |
+
"learning_rate": 2.7832198550406324e-05,
|
| 832 |
+
"loss": 0.0946,
|
| 833 |
+
"step": 10600
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 6.9,
|
| 837 |
+
"learning_rate": 2.7612563145179004e-05,
|
| 838 |
+
"loss": 0.0894,
|
| 839 |
+
"step": 10700
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 6.96,
|
| 843 |
+
"learning_rate": 2.739292773995168e-05,
|
| 844 |
+
"loss": 0.0908,
|
| 845 |
+
"step": 10800
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 7.03,
|
| 849 |
+
"learning_rate": 2.717329233472436e-05,
|
| 850 |
+
"loss": 0.0864,
|
| 851 |
+
"step": 10900
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 7.09,
|
| 855 |
+
"learning_rate": 2.6953656929497035e-05,
|
| 856 |
+
"loss": 0.0857,
|
| 857 |
+
"step": 11000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 7.09,
|
| 861 |
+
"eval_loss": 0.16591086983680725,
|
| 862 |
+
"eval_runtime": 375.5053,
|
| 863 |
+
"eval_samples_per_second": 14.479,
|
| 864 |
+
"eval_steps_per_second": 0.905,
|
| 865 |
+
"eval_wer": 0.10021865531881514,
|
| 866 |
+
"step": 11000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 7.16,
|
| 870 |
+
"learning_rate": 2.6734021524269714e-05,
|
| 871 |
+
"loss": 0.0941,
|
| 872 |
+
"step": 11100
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 7.22,
|
| 876 |
+
"learning_rate": 2.651438611904239e-05,
|
| 877 |
+
"loss": 0.0853,
|
| 878 |
+
"step": 11200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 7.29,
|
| 882 |
+
"learning_rate": 2.629475071381507e-05,
|
| 883 |
+
"loss": 0.0835,
|
| 884 |
+
"step": 11300
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 7.35,
|
| 888 |
+
"learning_rate": 2.6075115308587745e-05,
|
| 889 |
+
"loss": 0.0834,
|
| 890 |
+
"step": 11400
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 7.41,
|
| 894 |
+
"learning_rate": 2.5855479903360424e-05,
|
| 895 |
+
"loss": 0.0882,
|
| 896 |
+
"step": 11500
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 7.41,
|
| 900 |
+
"eval_loss": 0.16680683195590973,
|
| 901 |
+
"eval_runtime": 375.61,
|
| 902 |
+
"eval_samples_per_second": 14.475,
|
| 903 |
+
"eval_steps_per_second": 0.905,
|
| 904 |
+
"eval_wer": 0.10069910870993362,
|
| 905 |
+
"step": 11500
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 7.48,
|
| 909 |
+
"learning_rate": 2.56358444981331e-05,
|
| 910 |
+
"loss": 0.0817,
|
| 911 |
+
"step": 11600
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 7.54,
|
| 915 |
+
"learning_rate": 2.541620909290578e-05,
|
| 916 |
+
"loss": 0.0797,
|
| 917 |
+
"step": 11700
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 7.61,
|
| 921 |
+
"learning_rate": 2.5196573687678455e-05,
|
| 922 |
+
"loss": 0.0844,
|
| 923 |
+
"step": 11800
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 7.67,
|
| 927 |
+
"learning_rate": 2.497693828245113e-05,
|
| 928 |
+
"loss": 0.0812,
|
| 929 |
+
"step": 11900
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 7.74,
|
| 933 |
+
"learning_rate": 2.475730287722381e-05,
|
| 934 |
+
"loss": 0.0784,
|
| 935 |
+
"step": 12000
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 7.74,
|
| 939 |
+
"eval_loss": 0.16882041096687317,
|
| 940 |
+
"eval_runtime": 382.3729,
|
| 941 |
+
"eval_samples_per_second": 14.219,
|
| 942 |
+
"eval_steps_per_second": 0.889,
|
| 943 |
+
"eval_wer": 0.09955190367399766,
|
| 944 |
+
"step": 12000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 7.8,
|
| 948 |
+
"learning_rate": 2.4537667471996486e-05,
|
| 949 |
+
"loss": 0.0828,
|
| 950 |
+
"step": 12100
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 7.87,
|
| 954 |
+
"learning_rate": 2.4318032066769165e-05,
|
| 955 |
+
"loss": 0.0841,
|
| 956 |
+
"step": 12200
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 7.93,
|
| 960 |
+
"learning_rate": 2.409839666154184e-05,
|
| 961 |
+
"loss": 0.0836,
|
| 962 |
+
"step": 12300
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 7.99,
|
| 966 |
+
"learning_rate": 2.387876125631452e-05,
|
| 967 |
+
"loss": 0.0763,
|
| 968 |
+
"step": 12400
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 8.06,
|
| 972 |
+
"learning_rate": 2.3659125851087196e-05,
|
| 973 |
+
"loss": 0.0838,
|
| 974 |
+
"step": 12500
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 8.06,
|
| 978 |
+
"eval_loss": 0.16750921308994293,
|
| 979 |
+
"eval_runtime": 379.814,
|
| 980 |
+
"eval_samples_per_second": 14.315,
|
| 981 |
+
"eval_steps_per_second": 0.895,
|
| 982 |
+
"eval_wer": 0.09836547795307245,
|
| 983 |
+
"step": 12500
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 8.12,
|
| 987 |
+
"learning_rate": 2.3439490445859875e-05,
|
| 988 |
+
"loss": 0.0792,
|
| 989 |
+
"step": 12600
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 8.19,
|
| 993 |
+
"learning_rate": 2.321985504063255e-05,
|
| 994 |
+
"loss": 0.079,
|
| 995 |
+
"step": 12700
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 8.25,
|
| 999 |
+
"learning_rate": 2.300021963540523e-05,
|
| 1000 |
+
"loss": 0.0833,
|
| 1001 |
+
"step": 12800
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 8.32,
|
| 1005 |
+
"learning_rate": 2.2780584230177906e-05,
|
| 1006 |
+
"loss": 0.0757,
|
| 1007 |
+
"step": 12900
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 8.38,
|
| 1011 |
+
"learning_rate": 2.2560948824950586e-05,
|
| 1012 |
+
"loss": 0.0863,
|
| 1013 |
+
"step": 13000
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 8.38,
|
| 1017 |
+
"eval_loss": 0.16389821469783783,
|
| 1018 |
+
"eval_runtime": 379.0093,
|
| 1019 |
+
"eval_samples_per_second": 14.345,
|
| 1020 |
+
"eval_steps_per_second": 0.897,
|
| 1021 |
+
"eval_wer": 0.09786541421945934,
|
| 1022 |
+
"step": 13000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 8.45,
|
| 1026 |
+
"learning_rate": 2.234131341972326e-05,
|
| 1027 |
+
"loss": 0.0811,
|
| 1028 |
+
"step": 13100
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 8.51,
|
| 1032 |
+
"learning_rate": 2.2121678014495937e-05,
|
| 1033 |
+
"loss": 0.0775,
|
| 1034 |
+
"step": 13200
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 8.57,
|
| 1038 |
+
"learning_rate": 2.1902042609268617e-05,
|
| 1039 |
+
"loss": 0.0774,
|
| 1040 |
+
"step": 13300
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 8.64,
|
| 1044 |
+
"learning_rate": 2.1682407204041292e-05,
|
| 1045 |
+
"loss": 0.0791,
|
| 1046 |
+
"step": 13400
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 8.7,
|
| 1050 |
+
"learning_rate": 2.1462771798813968e-05,
|
| 1051 |
+
"loss": 0.0763,
|
| 1052 |
+
"step": 13500
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 8.7,
|
| 1056 |
+
"eval_loss": 0.16381791234016418,
|
| 1057 |
+
"eval_runtime": 379.2251,
|
| 1058 |
+
"eval_samples_per_second": 14.337,
|
| 1059 |
+
"eval_steps_per_second": 0.897,
|
| 1060 |
+
"eval_wer": 0.0980222969594164,
|
| 1061 |
+
"step": 13500
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 8.77,
|
| 1065 |
+
"learning_rate": 2.1243136393586648e-05,
|
| 1066 |
+
"loss": 0.0814,
|
| 1067 |
+
"step": 13600
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 8.83,
|
| 1071 |
+
"learning_rate": 2.1023500988359323e-05,
|
| 1072 |
+
"loss": 0.0852,
|
| 1073 |
+
"step": 13700
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 8.9,
|
| 1077 |
+
"learning_rate": 2.0803865583132003e-05,
|
| 1078 |
+
"loss": 0.0839,
|
| 1079 |
+
"step": 13800
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 8.96,
|
| 1083 |
+
"learning_rate": 2.058423017790468e-05,
|
| 1084 |
+
"loss": 0.0803,
|
| 1085 |
+
"step": 13900
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 9.03,
|
| 1089 |
+
"learning_rate": 2.0364594772677358e-05,
|
| 1090 |
+
"loss": 0.0822,
|
| 1091 |
+
"step": 14000
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 9.03,
|
| 1095 |
+
"eval_loss": 0.17092539370059967,
|
| 1096 |
+
"eval_runtime": 375.1283,
|
| 1097 |
+
"eval_samples_per_second": 14.494,
|
| 1098 |
+
"eval_steps_per_second": 0.906,
|
| 1099 |
+
"eval_wer": 0.09723788325963113,
|
| 1100 |
+
"step": 14000
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 9.09,
|
| 1104 |
+
"learning_rate": 2.0144959367450034e-05,
|
| 1105 |
+
"loss": 0.0793,
|
| 1106 |
+
"step": 14100
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 9.16,
|
| 1110 |
+
"learning_rate": 1.992532396222271e-05,
|
| 1111 |
+
"loss": 0.0821,
|
| 1112 |
+
"step": 14200
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 9.22,
|
| 1116 |
+
"learning_rate": 1.970568855699539e-05,
|
| 1117 |
+
"loss": 0.081,
|
| 1118 |
+
"step": 14300
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 9.28,
|
| 1122 |
+
"learning_rate": 1.9486053151768065e-05,
|
| 1123 |
+
"loss": 0.0763,
|
| 1124 |
+
"step": 14400
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 9.35,
|
| 1128 |
+
"learning_rate": 1.9266417746540744e-05,
|
| 1129 |
+
"loss": 0.0769,
|
| 1130 |
+
"step": 14500
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 9.35,
|
| 1134 |
+
"eval_loss": 0.16998609900474548,
|
| 1135 |
+
"eval_runtime": 376.06,
|
| 1136 |
+
"eval_samples_per_second": 14.458,
|
| 1137 |
+
"eval_steps_per_second": 0.904,
|
| 1138 |
+
"eval_wer": 0.09645346955984586,
|
| 1139 |
+
"step": 14500
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 9.41,
|
| 1143 |
+
"learning_rate": 1.904678234131342e-05,
|
| 1144 |
+
"loss": 0.0781,
|
| 1145 |
+
"step": 14600
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 9.48,
|
| 1149 |
+
"learning_rate": 1.88271469360861e-05,
|
| 1150 |
+
"loss": 0.0771,
|
| 1151 |
+
"step": 14700
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 9.54,
|
| 1155 |
+
"learning_rate": 1.8607511530858775e-05,
|
| 1156 |
+
"loss": 0.0806,
|
| 1157 |
+
"step": 14800
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 9.61,
|
| 1161 |
+
"learning_rate": 1.8387876125631454e-05,
|
| 1162 |
+
"loss": 0.074,
|
| 1163 |
+
"step": 14900
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 9.67,
|
| 1167 |
+
"learning_rate": 1.816824072040413e-05,
|
| 1168 |
+
"loss": 0.0838,
|
| 1169 |
+
"step": 15000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 9.67,
|
| 1173 |
+
"eval_loss": 0.17028485238552094,
|
| 1174 |
+
"eval_runtime": 378.355,
|
| 1175 |
+
"eval_samples_per_second": 14.37,
|
| 1176 |
+
"eval_steps_per_second": 0.899,
|
| 1177 |
+
"eval_wer": 0.09736535048584624,
|
| 1178 |
+
"step": 15000
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 9.74,
|
| 1182 |
+
"learning_rate": 1.794860531517681e-05,
|
| 1183 |
+
"loss": 0.0737,
|
| 1184 |
+
"step": 15100
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 9.8,
|
| 1188 |
+
"learning_rate": 1.7728969909949485e-05,
|
| 1189 |
+
"loss": 0.0697,
|
| 1190 |
+
"step": 15200
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 9.86,
|
| 1194 |
+
"learning_rate": 1.750933450472216e-05,
|
| 1195 |
+
"loss": 0.0789,
|
| 1196 |
+
"step": 15300
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 9.93,
|
| 1200 |
+
"learning_rate": 1.7289699099494837e-05,
|
| 1201 |
+
"loss": 0.0827,
|
| 1202 |
+
"step": 15400
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 9.99,
|
| 1206 |
+
"learning_rate": 1.7070063694267516e-05,
|
| 1207 |
+
"loss": 0.0799,
|
| 1208 |
+
"step": 15500
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 9.99,
|
| 1212 |
+
"eval_loss": 0.16668923199176788,
|
| 1213 |
+
"eval_runtime": 383.8462,
|
| 1214 |
+
"eval_samples_per_second": 14.165,
|
| 1215 |
+
"eval_steps_per_second": 0.886,
|
| 1216 |
+
"eval_wer": 0.09569847137380254,
|
| 1217 |
+
"step": 15500
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 10.06,
|
| 1221 |
+
"learning_rate": 1.6850428289040192e-05,
|
| 1222 |
+
"loss": 0.0752,
|
| 1223 |
+
"step": 15600
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 10.12,
|
| 1227 |
+
"learning_rate": 1.663079288381287e-05,
|
| 1228 |
+
"loss": 0.0689,
|
| 1229 |
+
"step": 15700
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 10.19,
|
| 1233 |
+
"learning_rate": 1.6411157478585547e-05,
|
| 1234 |
+
"loss": 0.0701,
|
| 1235 |
+
"step": 15800
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 10.25,
|
| 1239 |
+
"learning_rate": 1.6191522073358226e-05,
|
| 1240 |
+
"loss": 0.0718,
|
| 1241 |
+
"step": 15900
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 10.32,
|
| 1245 |
+
"learning_rate": 1.5971886668130902e-05,
|
| 1246 |
+
"loss": 0.0712,
|
| 1247 |
+
"step": 16000
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 10.32,
|
| 1251 |
+
"eval_loss": 0.17544205486774445,
|
| 1252 |
+
"eval_runtime": 378.572,
|
| 1253 |
+
"eval_samples_per_second": 14.362,
|
| 1254 |
+
"eval_steps_per_second": 0.898,
|
| 1255 |
+
"eval_wer": 0.09600243168246933,
|
| 1256 |
+
"step": 16000
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 10.38,
|
| 1260 |
+
"learning_rate": 1.575225126290358e-05,
|
| 1261 |
+
"loss": 0.076,
|
| 1262 |
+
"step": 16100
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 10.44,
|
| 1266 |
+
"learning_rate": 1.5532615857676257e-05,
|
| 1267 |
+
"loss": 0.0792,
|
| 1268 |
+
"step": 16200
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 10.51,
|
| 1272 |
+
"learning_rate": 1.5312980452448936e-05,
|
| 1273 |
+
"loss": 0.0764,
|
| 1274 |
+
"step": 16300
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 10.57,
|
| 1278 |
+
"learning_rate": 1.5093345047221614e-05,
|
| 1279 |
+
"loss": 0.0747,
|
| 1280 |
+
"step": 16400
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 10.64,
|
| 1284 |
+
"learning_rate": 1.4873709641994291e-05,
|
| 1285 |
+
"loss": 0.0737,
|
| 1286 |
+
"step": 16500
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 10.64,
|
| 1290 |
+
"eval_loss": 0.17247016727924347,
|
| 1291 |
+
"eval_runtime": 379.1086,
|
| 1292 |
+
"eval_samples_per_second": 14.342,
|
| 1293 |
+
"eval_steps_per_second": 0.897,
|
| 1294 |
+
"eval_wer": 0.09679665055350192,
|
| 1295 |
+
"step": 16500
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 10.7,
|
| 1299 |
+
"learning_rate": 1.4654074236766967e-05,
|
| 1300 |
+
"loss": 0.0704,
|
| 1301 |
+
"step": 16600
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 10.77,
|
| 1305 |
+
"learning_rate": 1.4434438831539645e-05,
|
| 1306 |
+
"loss": 0.0746,
|
| 1307 |
+
"step": 16700
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 10.83,
|
| 1311 |
+
"learning_rate": 1.4214803426312322e-05,
|
| 1312 |
+
"loss": 0.0747,
|
| 1313 |
+
"step": 16800
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 10.9,
|
| 1317 |
+
"learning_rate": 1.3995168021085e-05,
|
| 1318 |
+
"loss": 0.0746,
|
| 1319 |
+
"step": 16900
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 10.96,
|
| 1323 |
+
"learning_rate": 1.3775532615857678e-05,
|
| 1324 |
+
"loss": 0.0851,
|
| 1325 |
+
"step": 17000
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 10.96,
|
| 1329 |
+
"eval_loss": 0.17331218719482422,
|
| 1330 |
+
"eval_runtime": 384.1505,
|
| 1331 |
+
"eval_samples_per_second": 14.153,
|
| 1332 |
+
"eval_steps_per_second": 0.885,
|
| 1333 |
+
"eval_wer": 0.09575730240128644,
|
| 1334 |
+
"step": 17000
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 11.03,
|
| 1338 |
+
"learning_rate": 1.3555897210630355e-05,
|
| 1339 |
+
"loss": 0.0776,
|
| 1340 |
+
"step": 17100
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 11.09,
|
| 1344 |
+
"learning_rate": 1.3336261805403031e-05,
|
| 1345 |
+
"loss": 0.0712,
|
| 1346 |
+
"step": 17200
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 11.15,
|
| 1350 |
+
"learning_rate": 1.3116626400175708e-05,
|
| 1351 |
+
"loss": 0.0798,
|
| 1352 |
+
"step": 17300
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 11.22,
|
| 1356 |
+
"learning_rate": 1.2896990994948386e-05,
|
| 1357 |
+
"loss": 0.0752,
|
| 1358 |
+
"step": 17400
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 11.28,
|
| 1362 |
+
"learning_rate": 1.2677355589721064e-05,
|
| 1363 |
+
"loss": 0.076,
|
| 1364 |
+
"step": 17500
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 11.28,
|
| 1368 |
+
"eval_loss": 0.16816101968288422,
|
| 1369 |
+
"eval_runtime": 375.8451,
|
| 1370 |
+
"eval_samples_per_second": 14.466,
|
| 1371 |
+
"eval_steps_per_second": 0.905,
|
| 1372 |
+
"eval_wer": 0.09541412140763038,
|
| 1373 |
+
"step": 17500
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 11.35,
|
| 1377 |
+
"learning_rate": 1.2457720184493741e-05,
|
| 1378 |
+
"loss": 0.0785,
|
| 1379 |
+
"step": 17600
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 11.41,
|
| 1383 |
+
"learning_rate": 1.2238084779266417e-05,
|
| 1384 |
+
"loss": 0.0787,
|
| 1385 |
+
"step": 17700
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 11.48,
|
| 1389 |
+
"learning_rate": 1.2018449374039095e-05,
|
| 1390 |
+
"loss": 0.073,
|
| 1391 |
+
"step": 17800
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 11.54,
|
| 1395 |
+
"learning_rate": 1.1798813968811772e-05,
|
| 1396 |
+
"loss": 0.0686,
|
| 1397 |
+
"step": 17900
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 11.61,
|
| 1401 |
+
"learning_rate": 1.157917856358445e-05,
|
| 1402 |
+
"loss": 0.0712,
|
| 1403 |
+
"step": 18000
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 11.61,
|
| 1407 |
+
"eval_loss": 0.17131488025188446,
|
| 1408 |
+
"eval_runtime": 375.69,
|
| 1409 |
+
"eval_samples_per_second": 14.472,
|
| 1410 |
+
"eval_steps_per_second": 0.905,
|
| 1411 |
+
"eval_wer": 0.09434535774167296,
|
| 1412 |
+
"step": 18000
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 11.67,
|
| 1416 |
+
"learning_rate": 1.1359543158357127e-05,
|
| 1417 |
+
"loss": 0.0737,
|
| 1418 |
+
"step": 18100
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 11.73,
|
| 1422 |
+
"learning_rate": 1.1139907753129805e-05,
|
| 1423 |
+
"loss": 0.0781,
|
| 1424 |
+
"step": 18200
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 11.8,
|
| 1428 |
+
"learning_rate": 1.0920272347902482e-05,
|
| 1429 |
+
"loss": 0.084,
|
| 1430 |
+
"step": 18300
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 11.86,
|
| 1434 |
+
"learning_rate": 1.0702833296727433e-05,
|
| 1435 |
+
"loss": 0.0841,
|
| 1436 |
+
"step": 18400
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 11.93,
|
| 1440 |
+
"learning_rate": 1.048319789150011e-05,
|
| 1441 |
+
"loss": 0.0745,
|
| 1442 |
+
"step": 18500
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 11.93,
|
| 1446 |
+
"eval_loss": 0.16615267097949982,
|
| 1447 |
+
"eval_runtime": 379.2398,
|
| 1448 |
+
"eval_samples_per_second": 14.337,
|
| 1449 |
+
"eval_steps_per_second": 0.897,
|
| 1450 |
+
"eval_wer": 0.09507094041397433,
|
| 1451 |
+
"step": 18500
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 11.99,
|
| 1455 |
+
"learning_rate": 1.0263562486272788e-05,
|
| 1456 |
+
"loss": 0.0881,
|
| 1457 |
+
"step": 18600
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 12.06,
|
| 1461 |
+
"learning_rate": 1.0043927081045465e-05,
|
| 1462 |
+
"loss": 0.0801,
|
| 1463 |
+
"step": 18700
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 12.12,
|
| 1467 |
+
"learning_rate": 9.824291675818143e-06,
|
| 1468 |
+
"loss": 0.0863,
|
| 1469 |
+
"step": 18800
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 12.19,
|
| 1473 |
+
"learning_rate": 9.60465627059082e-06,
|
| 1474 |
+
"loss": 0.0786,
|
| 1475 |
+
"step": 18900
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 12.25,
|
| 1479 |
+
"learning_rate": 9.385020865363498e-06,
|
| 1480 |
+
"loss": 0.0864,
|
| 1481 |
+
"step": 19000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 12.25,
|
| 1485 |
+
"eval_loss": 0.16919353604316711,
|
| 1486 |
+
"eval_runtime": 377.3159,
|
| 1487 |
+
"eval_samples_per_second": 14.41,
|
| 1488 |
+
"eval_steps_per_second": 0.901,
|
| 1489 |
+
"eval_wer": 0.09465912322158707,
|
| 1490 |
+
"step": 19000
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 12.31,
|
| 1494 |
+
"learning_rate": 9.165385460136176e-06,
|
| 1495 |
+
"loss": 0.0973,
|
| 1496 |
+
"step": 19100
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 12.38,
|
| 1500 |
+
"learning_rate": 8.945750054908853e-06,
|
| 1501 |
+
"loss": 0.0999,
|
| 1502 |
+
"step": 19200
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 12.44,
|
| 1506 |
+
"learning_rate": 8.726114649681529e-06,
|
| 1507 |
+
"loss": 0.0922,
|
| 1508 |
+
"step": 19300
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 12.51,
|
| 1512 |
+
"learning_rate": 8.506479244454206e-06,
|
| 1513 |
+
"loss": 0.0918,
|
| 1514 |
+
"step": 19400
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 12.57,
|
| 1518 |
+
"learning_rate": 8.286843839226884e-06,
|
| 1519 |
+
"loss": 0.0937,
|
| 1520 |
+
"step": 19500
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 12.57,
|
| 1524 |
+
"eval_loss": 0.16241294145584106,
|
| 1525 |
+
"eval_runtime": 377.2732,
|
| 1526 |
+
"eval_samples_per_second": 14.411,
|
| 1527 |
+
"eval_steps_per_second": 0.901,
|
| 1528 |
+
"eval_wer": 0.09427672154294174,
|
| 1529 |
+
"step": 19500
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 12.64,
|
| 1533 |
+
"learning_rate": 8.067208433999562e-06,
|
| 1534 |
+
"loss": 0.0916,
|
| 1535 |
+
"step": 19600
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 12.7,
|
| 1539 |
+
"learning_rate": 7.847573028772239e-06,
|
| 1540 |
+
"loss": 0.0941,
|
| 1541 |
+
"step": 19700
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 12.77,
|
| 1545 |
+
"learning_rate": 7.627937623544916e-06,
|
| 1546 |
+
"loss": 0.0919,
|
| 1547 |
+
"step": 19800
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 12.83,
|
| 1551 |
+
"learning_rate": 7.408302218317593e-06,
|
| 1552 |
+
"loss": 0.0943,
|
| 1553 |
+
"step": 19900
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 12.89,
|
| 1557 |
+
"learning_rate": 7.188666813090271e-06,
|
| 1558 |
+
"loss": 0.0915,
|
| 1559 |
+
"step": 20000
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 12.89,
|
| 1563 |
+
"eval_loss": 0.1677888035774231,
|
| 1564 |
+
"eval_runtime": 382.7478,
|
| 1565 |
+
"eval_samples_per_second": 14.205,
|
| 1566 |
+
"eval_steps_per_second": 0.888,
|
| 1567 |
+
"eval_wer": 0.09419828017296322,
|
| 1568 |
+
"step": 20000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 12.96,
|
| 1572 |
+
"learning_rate": 6.9690314078629485e-06,
|
| 1573 |
+
"loss": 0.0882,
|
| 1574 |
+
"step": 20100
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 13.02,
|
| 1578 |
+
"learning_rate": 6.749396002635624e-06,
|
| 1579 |
+
"loss": 0.0974,
|
| 1580 |
+
"step": 20200
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 13.09,
|
| 1584 |
+
"learning_rate": 6.529760597408302e-06,
|
| 1585 |
+
"loss": 0.0955,
|
| 1586 |
+
"step": 20300
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 13.15,
|
| 1590 |
+
"learning_rate": 6.3101251921809794e-06,
|
| 1591 |
+
"loss": 0.0992,
|
| 1592 |
+
"step": 20400
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 13.22,
|
| 1596 |
+
"learning_rate": 6.092686141005931e-06,
|
| 1597 |
+
"loss": 0.0926,
|
| 1598 |
+
"step": 20500
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 13.22,
|
| 1602 |
+
"eval_loss": 0.16411229968070984,
|
| 1603 |
+
"eval_runtime": 381.1998,
|
| 1604 |
+
"eval_samples_per_second": 14.263,
|
| 1605 |
+
"eval_steps_per_second": 0.892,
|
| 1606 |
+
"eval_wer": 0.09449243531038269,
|
| 1607 |
+
"step": 20500
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 13.28,
|
| 1611 |
+
"learning_rate": 5.873050735778607e-06,
|
| 1612 |
+
"loss": 0.0939,
|
| 1613 |
+
"step": 20600
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 13.35,
|
| 1617 |
+
"learning_rate": 5.655611684603558e-06,
|
| 1618 |
+
"loss": 0.0974,
|
| 1619 |
+
"step": 20700
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 13.41,
|
| 1623 |
+
"learning_rate": 5.435976279376235e-06,
|
| 1624 |
+
"loss": 0.0957,
|
| 1625 |
+
"step": 20800
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 13.48,
|
| 1629 |
+
"learning_rate": 5.216340874148913e-06,
|
| 1630 |
+
"loss": 0.0909,
|
| 1631 |
+
"step": 20900
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 13.54,
|
| 1635 |
+
"learning_rate": 4.99670546892159e-06,
|
| 1636 |
+
"loss": 0.0912,
|
| 1637 |
+
"step": 21000
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 13.54,
|
| 1641 |
+
"eval_loss": 0.1665380299091339,
|
| 1642 |
+
"eval_runtime": 376.5271,
|
| 1643 |
+
"eval_samples_per_second": 14.44,
|
| 1644 |
+
"eval_steps_per_second": 0.903,
|
| 1645 |
+
"eval_wer": 0.09374724229558669,
|
| 1646 |
+
"step": 21000
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 13.6,
|
| 1650 |
+
"learning_rate": 4.777070063694268e-06,
|
| 1651 |
+
"loss": 0.09,
|
| 1652 |
+
"step": 21100
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 13.67,
|
| 1656 |
+
"learning_rate": 4.557434658466945e-06,
|
| 1657 |
+
"loss": 0.0938,
|
| 1658 |
+
"step": 21200
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 13.73,
|
| 1662 |
+
"learning_rate": 4.337799253239622e-06,
|
| 1663 |
+
"loss": 0.0867,
|
| 1664 |
+
"step": 21300
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 13.8,
|
| 1668 |
+
"learning_rate": 4.1181638480123e-06,
|
| 1669 |
+
"loss": 0.0844,
|
| 1670 |
+
"step": 21400
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 13.86,
|
| 1674 |
+
"learning_rate": 3.898528442784977e-06,
|
| 1675 |
+
"loss": 0.0917,
|
| 1676 |
+
"step": 21500
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 13.86,
|
| 1680 |
+
"eval_loss": 0.16477684676647186,
|
| 1681 |
+
"eval_runtime": 375.5992,
|
| 1682 |
+
"eval_samples_per_second": 14.476,
|
| 1683 |
+
"eval_steps_per_second": 0.905,
|
| 1684 |
+
"eval_wer": 0.09364919058311354,
|
| 1685 |
+
"step": 21500
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 13.93,
|
| 1689 |
+
"learning_rate": 3.678893037557654e-06,
|
| 1690 |
+
"loss": 0.0936,
|
| 1691 |
+
"step": 21600
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 13.99,
|
| 1695 |
+
"learning_rate": 3.4592576323303315e-06,
|
| 1696 |
+
"loss": 0.0902,
|
| 1697 |
+
"step": 21700
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 14.06,
|
| 1701 |
+
"learning_rate": 3.239622227103009e-06,
|
| 1702 |
+
"loss": 0.0994,
|
| 1703 |
+
"step": 21800
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 14.12,
|
| 1707 |
+
"learning_rate": 3.0199868218756866e-06,
|
| 1708 |
+
"loss": 0.0948,
|
| 1709 |
+
"step": 21900
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"epoch": 14.18,
|
| 1713 |
+
"learning_rate": 2.8003514166483638e-06,
|
| 1714 |
+
"loss": 0.094,
|
| 1715 |
+
"step": 22000
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 14.18,
|
| 1719 |
+
"eval_loss": 0.16351106762886047,
|
| 1720 |
+
"eval_runtime": 381.719,
|
| 1721 |
+
"eval_samples_per_second": 14.243,
|
| 1722 |
+
"eval_steps_per_second": 0.891,
|
| 1723 |
+
"eval_wer": 0.09346289232941453,
|
| 1724 |
+
"step": 22000
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 14.25,
|
| 1728 |
+
"learning_rate": 2.5807160114210413e-06,
|
| 1729 |
+
"loss": 0.0943,
|
| 1730 |
+
"step": 22100
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 14.31,
|
| 1734 |
+
"learning_rate": 2.3610806061937184e-06,
|
| 1735 |
+
"loss": 0.0941,
|
| 1736 |
+
"step": 22200
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 14.38,
|
| 1740 |
+
"learning_rate": 2.141445200966396e-06,
|
| 1741 |
+
"loss": 0.0868,
|
| 1742 |
+
"step": 22300
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 14.44,
|
| 1746 |
+
"learning_rate": 1.9218097957390735e-06,
|
| 1747 |
+
"loss": 0.0896,
|
| 1748 |
+
"step": 22400
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 14.51,
|
| 1752 |
+
"learning_rate": 1.7021743905117505e-06,
|
| 1753 |
+
"loss": 0.0864,
|
| 1754 |
+
"step": 22500
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 14.51,
|
| 1758 |
+
"eval_loss": 0.16783463954925537,
|
| 1759 |
+
"eval_runtime": 377.482,
|
| 1760 |
+
"eval_samples_per_second": 14.403,
|
| 1761 |
+
"eval_steps_per_second": 0.901,
|
| 1762 |
+
"eval_wer": 0.09338445095943601,
|
| 1763 |
+
"step": 22500
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 14.57,
|
| 1767 |
+
"learning_rate": 1.482538985284428e-06,
|
| 1768 |
+
"loss": 0.0935,
|
| 1769 |
+
"step": 22600
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 14.64,
|
| 1773 |
+
"learning_rate": 1.2629035800571053e-06,
|
| 1774 |
+
"loss": 0.0913,
|
| 1775 |
+
"step": 22700
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 14.7,
|
| 1779 |
+
"learning_rate": 1.0432681748297827e-06,
|
| 1780 |
+
"loss": 0.093,
|
| 1781 |
+
"step": 22800
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 14.76,
|
| 1785 |
+
"learning_rate": 8.236327696024599e-07,
|
| 1786 |
+
"loss": 0.0871,
|
| 1787 |
+
"step": 22900
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 14.83,
|
| 1791 |
+
"learning_rate": 6.039973643751374e-07,
|
| 1792 |
+
"loss": 0.0899,
|
| 1793 |
+
"step": 23000
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 14.83,
|
| 1797 |
+
"eval_loss": 0.16628585755825043,
|
| 1798 |
+
"eval_runtime": 378.1045,
|
| 1799 |
+
"eval_samples_per_second": 14.38,
|
| 1800 |
+
"eval_steps_per_second": 0.899,
|
| 1801 |
+
"eval_wer": 0.09316873719199506,
|
| 1802 |
+
"step": 23000
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 14.89,
|
| 1806 |
+
"learning_rate": 3.8436195914781466e-07,
|
| 1807 |
+
"loss": 0.099,
|
| 1808 |
+
"step": 23100
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 14.96,
|
| 1812 |
+
"learning_rate": 1.64726553920492e-07,
|
| 1813 |
+
"loss": 0.0964,
|
| 1814 |
+
"step": 23200
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 15.0,
|
| 1818 |
+
"step": 23265,
|
| 1819 |
+
"total_flos": 1.618046661664003e+20,
|
| 1820 |
+
"train_loss": 0.08625626374438786,
|
| 1821 |
+
"train_runtime": 75520.7008,
|
| 1822 |
+
"train_samples_per_second": 9.861,
|
| 1823 |
+
"train_steps_per_second": 0.308
|
| 1824 |
+
}
|
| 1825 |
+
],
|
| 1826 |
+
"max_steps": 23265,
|
| 1827 |
+
"num_train_epochs": 15,
|
| 1828 |
+
"total_flos": 1.618046661664003e+20,
|
| 1829 |
+
"trial_name": null,
|
| 1830 |
+
"trial_params": null
|
| 1831 |
+
}
|
wandb/run-20220208_002217-138c86e3/files/output.log
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b0be90b9f80bee7ec20639fb6e8497e532e496cd58fe07a74434804f393c16f
|
| 3 |
+
size 16924848
|
wandb/run-20220208_002217-138c86e3/files/wandb-summary.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
wandb/run-20220208_002217-138c86e3/logs/debug-internal.log
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:47d636d95cc3e2ac41ad1c139ebee5a91c0e2f25797f00dd56996c5a641f5082
|
| 3 |
+
size 7626602
|
wandb/run-20220208_002217-138c86e3/run-138c86e3.wandb
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1cc2938dd990bfb4884453df5c1fb3c85ad70487e5b3dac844bd803e3c698bc4
|
| 3 |
+
size 197621755
|