Commit
•
4eafb4f
1
Parent(s):
d5fed40
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- pytorch_model.bin +1 -1
- train_results.json +8 -0
- trainer_state.json +928 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.4525286555290222,
|
| 4 |
+
"eval_runtime": 319.7474,
|
| 5 |
+
"eval_samples": 10284,
|
| 6 |
+
"eval_samples_per_second": 32.163,
|
| 7 |
+
"eval_steps_per_second": 4.022,
|
| 8 |
+
"eval_wer": 0.4801227197959599,
|
| 9 |
+
"train_loss": 2.2287146272217635,
|
| 10 |
+
"train_runtime": 23537.8544,
|
| 11 |
+
"train_samples": 37318,
|
| 12 |
+
"train_samples_per_second": 15.854,
|
| 13 |
+
"train_steps_per_second": 0.495
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.4525286555290222,
|
| 4 |
+
"eval_runtime": 319.7474,
|
| 5 |
+
"eval_samples": 10284,
|
| 6 |
+
"eval_samples_per_second": 32.163,
|
| 7 |
+
"eval_steps_per_second": 4.022,
|
| 8 |
+
"eval_wer": 0.4801227197959599
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262313137
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32a6c4cdbb67e5027c4796ff7b91aac984f0b5ba33cdc497ef815a547b4926cc
|
| 3 |
size 1262313137
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 2.2287146272217635,
|
| 4 |
+
"train_runtime": 23537.8544,
|
| 5 |
+
"train_samples": 37318,
|
| 6 |
+
"train_samples_per_second": 15.854,
|
| 7 |
+
"train_steps_per_second": 0.495
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,928 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.99978563772776,
|
| 5 |
+
"global_step": 11660,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.09,
|
| 12 |
+
"learning_rate": 3.7125e-06,
|
| 13 |
+
"loss": 16.3142,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.17,
|
| 18 |
+
"learning_rate": 7.4625e-06,
|
| 19 |
+
"loss": 9.7017,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.26,
|
| 24 |
+
"learning_rate": 1.1212499999999998e-05,
|
| 25 |
+
"loss": 6.7433,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.34,
|
| 30 |
+
"learning_rate": 1.49625e-05,
|
| 31 |
+
"loss": 5.7032,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.43,
|
| 36 |
+
"learning_rate": 1.8712499999999997e-05,
|
| 37 |
+
"loss": 4.7972,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.43,
|
| 42 |
+
"eval_loss": 5.140112400054932,
|
| 43 |
+
"eval_runtime": 323.0627,
|
| 44 |
+
"eval_samples_per_second": 31.833,
|
| 45 |
+
"eval_steps_per_second": 3.981,
|
| 46 |
+
"eval_wer": 1.0,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.51,
|
| 51 |
+
"learning_rate": 2.2462499999999997e-05,
|
| 52 |
+
"loss": 4.1097,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.6,
|
| 57 |
+
"learning_rate": 2.6212499999999997e-05,
|
| 58 |
+
"loss": 3.6756,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.69,
|
| 63 |
+
"learning_rate": 2.99625e-05,
|
| 64 |
+
"loss": 3.5057,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.77,
|
| 69 |
+
"learning_rate": 3.37125e-05,
|
| 70 |
+
"loss": 3.4116,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.86,
|
| 75 |
+
"learning_rate": 3.7462499999999996e-05,
|
| 76 |
+
"loss": 3.3241,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.86,
|
| 81 |
+
"eval_loss": 3.3219833374023438,
|
| 82 |
+
"eval_runtime": 316.4143,
|
| 83 |
+
"eval_samples_per_second": 32.502,
|
| 84 |
+
"eval_steps_per_second": 4.064,
|
| 85 |
+
"eval_wer": 1.0,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.94,
|
| 90 |
+
"learning_rate": 4.12125e-05,
|
| 91 |
+
"loss": 3.2968,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.03,
|
| 96 |
+
"learning_rate": 4.4962499999999995e-05,
|
| 97 |
+
"loss": 3.2757,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.11,
|
| 102 |
+
"learning_rate": 4.871249999999999e-05,
|
| 103 |
+
"loss": 3.2343,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.2,
|
| 108 |
+
"learning_rate": 5.2462499999999994e-05,
|
| 109 |
+
"loss": 3.1872,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.29,
|
| 114 |
+
"learning_rate": 5.62125e-05,
|
| 115 |
+
"loss": 3.1432,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.29,
|
| 120 |
+
"eval_loss": 3.080551862716675,
|
| 121 |
+
"eval_runtime": 318.2154,
|
| 122 |
+
"eval_samples_per_second": 32.318,
|
| 123 |
+
"eval_steps_per_second": 4.041,
|
| 124 |
+
"eval_wer": 0.9998891086181085,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.37,
|
| 129 |
+
"learning_rate": 5.9962499999999994e-05,
|
| 130 |
+
"loss": 3.0996,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.46,
|
| 135 |
+
"learning_rate": 6.37125e-05,
|
| 136 |
+
"loss": 3.0582,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.54,
|
| 141 |
+
"learning_rate": 6.746249999999999e-05,
|
| 142 |
+
"loss": 3.0313,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.63,
|
| 147 |
+
"learning_rate": 7.121249999999999e-05,
|
| 148 |
+
"loss": 2.9824,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.72,
|
| 153 |
+
"learning_rate": 7.49625e-05,
|
| 154 |
+
"loss": 2.9297,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.72,
|
| 159 |
+
"eval_loss": 2.567805528640747,
|
| 160 |
+
"eval_runtime": 317.0688,
|
| 161 |
+
"eval_samples_per_second": 32.435,
|
| 162 |
+
"eval_steps_per_second": 4.056,
|
| 163 |
+
"eval_wer": 1.005729388064391,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.8,
|
| 168 |
+
"learning_rate": 7.423136645962732e-05,
|
| 169 |
+
"loss": 2.8167,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.89,
|
| 174 |
+
"learning_rate": 7.345496894409937e-05,
|
| 175 |
+
"loss": 2.6315,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.97,
|
| 180 |
+
"learning_rate": 7.26863354037267e-05,
|
| 181 |
+
"loss": 2.475,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 2.06,
|
| 186 |
+
"learning_rate": 7.190993788819875e-05,
|
| 187 |
+
"loss": 2.3571,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.14,
|
| 192 |
+
"learning_rate": 7.11335403726708e-05,
|
| 193 |
+
"loss": 2.2593,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.14,
|
| 198 |
+
"eval_loss": 1.1067590713500977,
|
| 199 |
+
"eval_runtime": 319.3729,
|
| 200 |
+
"eval_samples_per_second": 32.201,
|
| 201 |
+
"eval_steps_per_second": 4.027,
|
| 202 |
+
"eval_wer": 0.821834513094424,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 2.23,
|
| 207 |
+
"learning_rate": 7.035714285714285e-05,
|
| 208 |
+
"loss": 2.1855,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.32,
|
| 213 |
+
"learning_rate": 6.95807453416149e-05,
|
| 214 |
+
"loss": 2.1429,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 2.4,
|
| 219 |
+
"learning_rate": 6.880434782608696e-05,
|
| 220 |
+
"loss": 2.1143,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 2.49,
|
| 225 |
+
"learning_rate": 6.802795031055901e-05,
|
| 226 |
+
"loss": 2.081,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 2.57,
|
| 231 |
+
"learning_rate": 6.725155279503105e-05,
|
| 232 |
+
"loss": 2.0504,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.57,
|
| 237 |
+
"eval_loss": 0.7877965569496155,
|
| 238 |
+
"eval_runtime": 316.6537,
|
| 239 |
+
"eval_samples_per_second": 32.477,
|
| 240 |
+
"eval_steps_per_second": 4.061,
|
| 241 |
+
"eval_wer": 0.7113866967305524,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.66,
|
| 246 |
+
"learning_rate": 6.64751552795031e-05,
|
| 247 |
+
"loss": 2.014,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.74,
|
| 252 |
+
"learning_rate": 6.569875776397515e-05,
|
| 253 |
+
"loss": 1.9885,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.83,
|
| 258 |
+
"learning_rate": 6.49223602484472e-05,
|
| 259 |
+
"loss": 1.9532,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.92,
|
| 264 |
+
"learning_rate": 6.414596273291925e-05,
|
| 265 |
+
"loss": 1.936,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 3.0,
|
| 270 |
+
"learning_rate": 6.33695652173913e-05,
|
| 271 |
+
"loss": 1.937,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.0,
|
| 276 |
+
"eval_loss": 0.6955012679100037,
|
| 277 |
+
"eval_runtime": 319.8543,
|
| 278 |
+
"eval_samples_per_second": 32.152,
|
| 279 |
+
"eval_steps_per_second": 4.021,
|
| 280 |
+
"eval_wer": 0.6449812408745633,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.09,
|
| 285 |
+
"learning_rate": 6.259316770186334e-05,
|
| 286 |
+
"loss": 1.9074,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 3.17,
|
| 291 |
+
"learning_rate": 6.18167701863354e-05,
|
| 292 |
+
"loss": 1.8975,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.26,
|
| 297 |
+
"learning_rate": 6.104037267080745e-05,
|
| 298 |
+
"loss": 1.8628,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 3.34,
|
| 303 |
+
"learning_rate": 6.02639751552795e-05,
|
| 304 |
+
"loss": 1.8605,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 3.43,
|
| 309 |
+
"learning_rate": 5.948757763975155e-05,
|
| 310 |
+
"loss": 1.8491,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 3.43,
|
| 315 |
+
"eval_loss": 0.645221471786499,
|
| 316 |
+
"eval_runtime": 318.4935,
|
| 317 |
+
"eval_samples_per_second": 32.29,
|
| 318 |
+
"eval_steps_per_second": 4.038,
|
| 319 |
+
"eval_wer": 0.6303620603618756,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 3.52,
|
| 324 |
+
"learning_rate": 5.87111801242236e-05,
|
| 325 |
+
"loss": 1.85,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 3.6,
|
| 330 |
+
"learning_rate": 5.7934782608695654e-05,
|
| 331 |
+
"loss": 1.8298,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 3.69,
|
| 336 |
+
"learning_rate": 5.715838509316769e-05,
|
| 337 |
+
"loss": 1.8108,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.77,
|
| 342 |
+
"learning_rate": 5.6381987577639744e-05,
|
| 343 |
+
"loss": 1.8136,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.86,
|
| 348 |
+
"learning_rate": 5.5605590062111795e-05,
|
| 349 |
+
"loss": 1.803,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.86,
|
| 354 |
+
"eval_loss": 0.5961059927940369,
|
| 355 |
+
"eval_runtime": 322.5472,
|
| 356 |
+
"eval_samples_per_second": 31.884,
|
| 357 |
+
"eval_steps_per_second": 3.987,
|
| 358 |
+
"eval_wer": 0.6041547304415326,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 3.94,
|
| 363 |
+
"learning_rate": 5.482919254658385e-05,
|
| 364 |
+
"loss": 1.7829,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 4.03,
|
| 369 |
+
"learning_rate": 5.40527950310559e-05,
|
| 370 |
+
"loss": 1.7755,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 4.12,
|
| 375 |
+
"learning_rate": 5.327639751552795e-05,
|
| 376 |
+
"loss": 1.7805,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 4.2,
|
| 381 |
+
"learning_rate": 5.2499999999999995e-05,
|
| 382 |
+
"loss": 1.7647,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 4.29,
|
| 387 |
+
"learning_rate": 5.173136645962733e-05,
|
| 388 |
+
"loss": 1.7545,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 4.29,
|
| 393 |
+
"eval_loss": 0.5550380945205688,
|
| 394 |
+
"eval_runtime": 317.6329,
|
| 395 |
+
"eval_samples_per_second": 32.377,
|
| 396 |
+
"eval_steps_per_second": 4.049,
|
| 397 |
+
"eval_wer": 0.5747500323433197,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.37,
|
| 402 |
+
"learning_rate": 5.095496894409938e-05,
|
| 403 |
+
"loss": 1.735,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.46,
|
| 408 |
+
"learning_rate": 5.017857142857142e-05,
|
| 409 |
+
"loss": 1.7365,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 4.55,
|
| 414 |
+
"learning_rate": 4.940217391304347e-05,
|
| 415 |
+
"loss": 1.734,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.63,
|
| 420 |
+
"learning_rate": 4.862577639751552e-05,
|
| 421 |
+
"loss": 1.7254,
|
| 422 |
+
"step": 5400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 4.72,
|
| 426 |
+
"learning_rate": 4.7849378881987574e-05,
|
| 427 |
+
"loss": 1.7045,
|
| 428 |
+
"step": 5500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.72,
|
| 432 |
+
"eval_loss": 0.5373523831367493,
|
| 433 |
+
"eval_runtime": 318.8226,
|
| 434 |
+
"eval_samples_per_second": 32.256,
|
| 435 |
+
"eval_steps_per_second": 4.034,
|
| 436 |
+
"eval_wer": 0.5743064668157539,
|
| 437 |
+
"step": 5500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 4.8,
|
| 441 |
+
"learning_rate": 4.7072981366459626e-05,
|
| 442 |
+
"loss": 1.7031,
|
| 443 |
+
"step": 5600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 4.89,
|
| 447 |
+
"learning_rate": 4.629658385093168e-05,
|
| 448 |
+
"loss": 1.7135,
|
| 449 |
+
"step": 5700
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 4.97,
|
| 453 |
+
"learning_rate": 4.5520186335403715e-05,
|
| 454 |
+
"loss": 1.7053,
|
| 455 |
+
"step": 5800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 5.06,
|
| 459 |
+
"learning_rate": 4.474378881987577e-05,
|
| 460 |
+
"loss": 1.7078,
|
| 461 |
+
"step": 5900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 5.15,
|
| 465 |
+
"learning_rate": 4.396739130434782e-05,
|
| 466 |
+
"loss": 1.6733,
|
| 467 |
+
"step": 6000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 5.15,
|
| 471 |
+
"eval_loss": 0.5336768627166748,
|
| 472 |
+
"eval_runtime": 321.7447,
|
| 473 |
+
"eval_samples_per_second": 31.963,
|
| 474 |
+
"eval_steps_per_second": 3.997,
|
| 475 |
+
"eval_wer": 0.5403737039569741,
|
| 476 |
+
"step": 6000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 5.23,
|
| 480 |
+
"learning_rate": 4.319099378881987e-05,
|
| 481 |
+
"loss": 1.6737,
|
| 482 |
+
"step": 6100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 5.32,
|
| 486 |
+
"learning_rate": 4.241459627329192e-05,
|
| 487 |
+
"loss": 1.6939,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 5.4,
|
| 492 |
+
"learning_rate": 4.1638198757763974e-05,
|
| 493 |
+
"loss": 1.674,
|
| 494 |
+
"step": 6300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 5.49,
|
| 498 |
+
"learning_rate": 4.0861801242236026e-05,
|
| 499 |
+
"loss": 1.6685,
|
| 500 |
+
"step": 6400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 5.57,
|
| 504 |
+
"learning_rate": 4.008540372670807e-05,
|
| 505 |
+
"loss": 1.6761,
|
| 506 |
+
"step": 6500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 5.57,
|
| 510 |
+
"eval_loss": 0.5054484009742737,
|
| 511 |
+
"eval_runtime": 316.5106,
|
| 512 |
+
"eval_samples_per_second": 32.492,
|
| 513 |
+
"eval_steps_per_second": 4.063,
|
| 514 |
+
"eval_wer": 0.5265862088084721,
|
| 515 |
+
"step": 6500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 5.66,
|
| 519 |
+
"learning_rate": 3.930900621118012e-05,
|
| 520 |
+
"loss": 1.6729,
|
| 521 |
+
"step": 6600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 5.75,
|
| 525 |
+
"learning_rate": 3.853260869565217e-05,
|
| 526 |
+
"loss": 1.6521,
|
| 527 |
+
"step": 6700
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 5.83,
|
| 531 |
+
"learning_rate": 3.775621118012422e-05,
|
| 532 |
+
"loss": 1.6876,
|
| 533 |
+
"step": 6800
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 5.92,
|
| 537 |
+
"learning_rate": 3.697981366459627e-05,
|
| 538 |
+
"loss": 1.6499,
|
| 539 |
+
"step": 6900
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 6.0,
|
| 543 |
+
"learning_rate": 3.620341614906832e-05,
|
| 544 |
+
"loss": 1.655,
|
| 545 |
+
"step": 7000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 6.0,
|
| 549 |
+
"eval_loss": 0.492550253868103,
|
| 550 |
+
"eval_runtime": 321.8948,
|
| 551 |
+
"eval_samples_per_second": 31.948,
|
| 552 |
+
"eval_steps_per_second": 3.995,
|
| 553 |
+
"eval_wer": 0.5243314173766795,
|
| 554 |
+
"step": 7000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 6.09,
|
| 558 |
+
"learning_rate": 3.543478260869565e-05,
|
| 559 |
+
"loss": 1.6501,
|
| 560 |
+
"step": 7100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 6.17,
|
| 564 |
+
"learning_rate": 3.4658385093167694e-05,
|
| 565 |
+
"loss": 1.653,
|
| 566 |
+
"step": 7200
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 6.26,
|
| 570 |
+
"learning_rate": 3.3881987577639746e-05,
|
| 571 |
+
"loss": 1.6381,
|
| 572 |
+
"step": 7300
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 6.35,
|
| 576 |
+
"learning_rate": 3.31055900621118e-05,
|
| 577 |
+
"loss": 1.6231,
|
| 578 |
+
"step": 7400
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 6.43,
|
| 582 |
+
"learning_rate": 3.232919254658385e-05,
|
| 583 |
+
"loss": 1.6252,
|
| 584 |
+
"step": 7500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 6.43,
|
| 588 |
+
"eval_loss": 0.49458181858062744,
|
| 589 |
+
"eval_runtime": 317.6079,
|
| 590 |
+
"eval_samples_per_second": 32.38,
|
| 591 |
+
"eval_steps_per_second": 4.049,
|
| 592 |
+
"eval_wer": 0.5182508732696324,
|
| 593 |
+
"step": 7500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 6.52,
|
| 597 |
+
"learning_rate": 3.1552795031055894e-05,
|
| 598 |
+
"loss": 1.6409,
|
| 599 |
+
"step": 7600
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 6.6,
|
| 603 |
+
"learning_rate": 3.0776397515527946e-05,
|
| 604 |
+
"loss": 1.6256,
|
| 605 |
+
"step": 7700
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 6.69,
|
| 609 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 610 |
+
"loss": 1.6113,
|
| 611 |
+
"step": 7800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 6.78,
|
| 615 |
+
"learning_rate": 2.9223602484472046e-05,
|
| 616 |
+
"loss": 1.6096,
|
| 617 |
+
"step": 7900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 6.86,
|
| 621 |
+
"learning_rate": 2.8447204968944097e-05,
|
| 622 |
+
"loss": 1.6209,
|
| 623 |
+
"step": 8000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 6.86,
|
| 627 |
+
"eval_loss": 0.491542249917984,
|
| 628 |
+
"eval_runtime": 319.2369,
|
| 629 |
+
"eval_samples_per_second": 32.214,
|
| 630 |
+
"eval_steps_per_second": 4.028,
|
| 631 |
+
"eval_wer": 0.5193597870885468,
|
| 632 |
+
"step": 8000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 6.95,
|
| 636 |
+
"learning_rate": 2.767080745341615e-05,
|
| 637 |
+
"loss": 1.5917,
|
| 638 |
+
"step": 8100
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 7.03,
|
| 642 |
+
"learning_rate": 2.6894409937888194e-05,
|
| 643 |
+
"loss": 1.6049,
|
| 644 |
+
"step": 8200
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 7.12,
|
| 648 |
+
"learning_rate": 2.6118012422360246e-05,
|
| 649 |
+
"loss": 1.5891,
|
| 650 |
+
"step": 8300
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 7.2,
|
| 654 |
+
"learning_rate": 2.5341614906832297e-05,
|
| 655 |
+
"loss": 1.5915,
|
| 656 |
+
"step": 8400
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 7.29,
|
| 660 |
+
"learning_rate": 2.4565217391304346e-05,
|
| 661 |
+
"loss": 1.5772,
|
| 662 |
+
"step": 8500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 7.29,
|
| 666 |
+
"eval_loss": 0.47245877981185913,
|
| 667 |
+
"eval_runtime": 314.2017,
|
| 668 |
+
"eval_samples_per_second": 32.731,
|
| 669 |
+
"eval_steps_per_second": 4.093,
|
| 670 |
+
"eval_wer": 0.5103775851553404,
|
| 671 |
+
"step": 8500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 7.38,
|
| 675 |
+
"learning_rate": 2.3788819875776394e-05,
|
| 676 |
+
"loss": 1.5932,
|
| 677 |
+
"step": 8600
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 7.46,
|
| 681 |
+
"learning_rate": 2.3012422360248446e-05,
|
| 682 |
+
"loss": 1.574,
|
| 683 |
+
"step": 8700
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 7.55,
|
| 687 |
+
"learning_rate": 2.2236024844720494e-05,
|
| 688 |
+
"loss": 1.5818,
|
| 689 |
+
"step": 8800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 7.63,
|
| 693 |
+
"learning_rate": 2.1459627329192546e-05,
|
| 694 |
+
"loss": 1.5852,
|
| 695 |
+
"step": 8900
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 7.72,
|
| 699 |
+
"learning_rate": 2.0683229813664594e-05,
|
| 700 |
+
"loss": 1.5602,
|
| 701 |
+
"step": 9000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 7.72,
|
| 705 |
+
"eval_loss": 0.47263726592063904,
|
| 706 |
+
"eval_runtime": 323.393,
|
| 707 |
+
"eval_samples_per_second": 31.8,
|
| 708 |
+
"eval_steps_per_second": 3.977,
|
| 709 |
+
"eval_wer": 0.509656791173046,
|
| 710 |
+
"step": 9000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 7.8,
|
| 714 |
+
"learning_rate": 1.9906832298136646e-05,
|
| 715 |
+
"loss": 1.5737,
|
| 716 |
+
"step": 9100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 7.89,
|
| 720 |
+
"learning_rate": 1.9130434782608694e-05,
|
| 721 |
+
"loss": 1.5894,
|
| 722 |
+
"step": 9200
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 7.98,
|
| 726 |
+
"learning_rate": 1.8361801242236024e-05,
|
| 727 |
+
"loss": 1.5694,
|
| 728 |
+
"step": 9300
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 8.06,
|
| 732 |
+
"learning_rate": 1.7585403726708073e-05,
|
| 733 |
+
"loss": 1.537,
|
| 734 |
+
"step": 9400
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 8.15,
|
| 738 |
+
"learning_rate": 1.680900621118012e-05,
|
| 739 |
+
"loss": 1.5783,
|
| 740 |
+
"step": 9500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 8.15,
|
| 744 |
+
"eval_loss": 0.4666772186756134,
|
| 745 |
+
"eval_runtime": 318.3833,
|
| 746 |
+
"eval_samples_per_second": 32.301,
|
| 747 |
+
"eval_steps_per_second": 4.039,
|
| 748 |
+
"eval_wer": 0.49557358567283344,
|
| 749 |
+
"step": 9500
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 8.23,
|
| 753 |
+
"learning_rate": 1.604037267080745e-05,
|
| 754 |
+
"loss": 1.5432,
|
| 755 |
+
"step": 9600
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 8.32,
|
| 759 |
+
"learning_rate": 1.5263975155279503e-05,
|
| 760 |
+
"loss": 1.5439,
|
| 761 |
+
"step": 9700
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 8.4,
|
| 765 |
+
"learning_rate": 1.4487577639751551e-05,
|
| 766 |
+
"loss": 1.5589,
|
| 767 |
+
"step": 9800
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 8.49,
|
| 771 |
+
"learning_rate": 1.3711180124223601e-05,
|
| 772 |
+
"loss": 1.5573,
|
| 773 |
+
"step": 9900
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 8.58,
|
| 777 |
+
"learning_rate": 1.2934782608695651e-05,
|
| 778 |
+
"loss": 1.5442,
|
| 779 |
+
"step": 10000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 8.58,
|
| 783 |
+
"eval_loss": 0.46852901577949524,
|
| 784 |
+
"eval_runtime": 317.9433,
|
| 785 |
+
"eval_samples_per_second": 32.345,
|
| 786 |
+
"eval_steps_per_second": 4.045,
|
| 787 |
+
"eval_wer": 0.49374387787162477,
|
| 788 |
+
"step": 10000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 8.66,
|
| 792 |
+
"learning_rate": 1.2158385093167701e-05,
|
| 793 |
+
"loss": 1.5679,
|
| 794 |
+
"step": 10100
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 8.75,
|
| 798 |
+
"learning_rate": 1.1381987577639751e-05,
|
| 799 |
+
"loss": 1.5604,
|
| 800 |
+
"step": 10200
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 8.83,
|
| 804 |
+
"learning_rate": 1.06055900621118e-05,
|
| 805 |
+
"loss": 1.5458,
|
| 806 |
+
"step": 10300
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 8.92,
|
| 810 |
+
"learning_rate": 9.829192546583851e-06,
|
| 811 |
+
"loss": 1.5501,
|
| 812 |
+
"step": 10400
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 9.01,
|
| 816 |
+
"learning_rate": 9.0527950310559e-06,
|
| 817 |
+
"loss": 1.5597,
|
| 818 |
+
"step": 10500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 9.01,
|
| 822 |
+
"eval_loss": 0.4707895815372467,
|
| 823 |
+
"eval_runtime": 323.6683,
|
| 824 |
+
"eval_samples_per_second": 31.773,
|
| 825 |
+
"eval_steps_per_second": 3.973,
|
| 826 |
+
"eval_wer": 0.495665995157743,
|
| 827 |
+
"step": 10500
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 9.09,
|
| 831 |
+
"learning_rate": 8.27639751552795e-06,
|
| 832 |
+
"loss": 1.5189,
|
| 833 |
+
"step": 10600
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 9.18,
|
| 837 |
+
"learning_rate": 7.499999999999999e-06,
|
| 838 |
+
"loss": 1.5272,
|
| 839 |
+
"step": 10700
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 9.26,
|
| 843 |
+
"learning_rate": 6.7236024844720485e-06,
|
| 844 |
+
"loss": 1.5319,
|
| 845 |
+
"step": 10800
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 9.35,
|
| 849 |
+
"learning_rate": 5.9472049689440985e-06,
|
| 850 |
+
"loss": 1.5175,
|
| 851 |
+
"step": 10900
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 9.43,
|
| 855 |
+
"learning_rate": 5.1708074534161485e-06,
|
| 856 |
+
"loss": 1.5406,
|
| 857 |
+
"step": 11000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 9.43,
|
| 861 |
+
"eval_loss": 0.45390617847442627,
|
| 862 |
+
"eval_runtime": 320.1014,
|
| 863 |
+
"eval_samples_per_second": 32.127,
|
| 864 |
+
"eval_steps_per_second": 4.017,
|
| 865 |
+
"eval_wer": 0.48104681464505517,
|
| 866 |
+
"step": 11000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 9.52,
|
| 870 |
+
"learning_rate": 4.3944099378881985e-06,
|
| 871 |
+
"loss": 1.5326,
|
| 872 |
+
"step": 11100
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 9.61,
|
| 876 |
+
"learning_rate": 3.6180124223602485e-06,
|
| 877 |
+
"loss": 1.5142,
|
| 878 |
+
"step": 11200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 9.69,
|
| 882 |
+
"learning_rate": 2.8416149068322976e-06,
|
| 883 |
+
"loss": 1.5195,
|
| 884 |
+
"step": 11300
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 9.78,
|
| 888 |
+
"learning_rate": 2.0652173913043476e-06,
|
| 889 |
+
"loss": 1.5147,
|
| 890 |
+
"step": 11400
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 9.86,
|
| 894 |
+
"learning_rate": 1.2888198757763974e-06,
|
| 895 |
+
"loss": 1.5274,
|
| 896 |
+
"step": 11500
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 9.86,
|
| 900 |
+
"eval_loss": 0.45016008615493774,
|
| 901 |
+
"eval_runtime": 317.8491,
|
| 902 |
+
"eval_samples_per_second": 32.355,
|
| 903 |
+
"eval_steps_per_second": 4.046,
|
| 904 |
+
"eval_wer": 0.4782560482007873,
|
| 905 |
+
"step": 11500
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 9.95,
|
| 909 |
+
"learning_rate": 5.124223602484471e-07,
|
| 910 |
+
"loss": 1.5295,
|
| 911 |
+
"step": 11600
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 10.0,
|
| 915 |
+
"step": 11660,
|
| 916 |
+
"total_flos": 3.1970153172049187e+19,
|
| 917 |
+
"train_loss": 2.2287146272217635,
|
| 918 |
+
"train_runtime": 23537.8544,
|
| 919 |
+
"train_samples_per_second": 15.854,
|
| 920 |
+
"train_steps_per_second": 0.495
|
| 921 |
+
}
|
| 922 |
+
],
|
| 923 |
+
"max_steps": 11660,
|
| 924 |
+
"num_train_epochs": 10,
|
| 925 |
+
"total_flos": 3.1970153172049187e+19,
|
| 926 |
+
"trial_name": null,
|
| 927 |
+
"trial_params": null
|
| 928 |
+
}
|