End of training
Browse files- all_results.json +10 -10
- eval_results.json +6 -6
- pytorch_model.bin +1 -1
- train_results.json +5 -5
- trainer_state.json +806 -14
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 1112,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 3.
|
| 8 |
-
"eval_wer":
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
"train_samples": 3336,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.27193310856819153,
|
| 4 |
+
"eval_runtime": 77.837,
|
| 5 |
"eval_samples": 1112,
|
| 6 |
+
"eval_samples_per_second": 14.286,
|
| 7 |
+
"eval_steps_per_second": 3.572,
|
| 8 |
+
"eval_wer": 0.3418834080717489,
|
| 9 |
+
"train_loss": 1.3617363874728863,
|
| 10 |
+
"train_runtime": 41188.5513,
|
| 11 |
"train_samples": 3336,
|
| 12 |
+
"train_samples_per_second": 8.099,
|
| 13 |
+
"train_steps_per_second": 0.252
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 1112,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 3.
|
| 8 |
-
"eval_wer":
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.27193310856819153,
|
| 4 |
+
"eval_runtime": 77.837,
|
| 5 |
"eval_samples": 1112,
|
| 6 |
+
"eval_samples_per_second": 14.286,
|
| 7 |
+
"eval_steps_per_second": 3.572,
|
| 8 |
+
"eval_wer": 0.3418834080717489
|
| 9 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262247537
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cbbbdffbb4174119b4bcfa2f71d7258342bc0f52541f29318f7eeda2b0a69be0
|
| 3 |
size 1262247537
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 3336,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"train_loss": 1.3617363874728863,
|
| 4 |
+
"train_runtime": 41188.5513,
|
| 5 |
"train_samples": 3336,
|
| 6 |
+
"train_samples_per_second": 8.099,
|
| 7 |
+
"train_steps_per_second": 0.252
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -10,28 +10,820 @@
|
|
| 10 |
{
|
| 11 |
"epoch": 0.96,
|
| 12 |
"learning_rate": 4.9500000000000004e-05,
|
| 13 |
-
"loss": 14.
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 1.92,
|
| 18 |
"learning_rate": 9.95e-05,
|
| 19 |
-
"loss": 4.
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch": 2.
|
| 24 |
-
"
|
| 25 |
-
"
|
| 26 |
-
"
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
}
|
| 31 |
],
|
| 32 |
-
"max_steps":
|
| 33 |
-
"num_train_epochs":
|
| 34 |
-
"total_flos":
|
| 35 |
"trial_name": null,
|
| 36 |
"trial_params": null
|
| 37 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 99.99760191846522,
|
| 5 |
+
"global_step": 10400,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 10 |
{
|
| 11 |
"epoch": 0.96,
|
| 12 |
"learning_rate": 4.9500000000000004e-05,
|
| 13 |
+
"loss": 14.732,
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 1.92,
|
| 18 |
"learning_rate": 9.95e-05,
|
| 19 |
+
"loss": 4.5706,
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 2.88,
|
| 24 |
+
"learning_rate": 0.0001495,
|
| 25 |
+
"loss": 3.4967,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 3.84,
|
| 30 |
+
"learning_rate": 0.00019950000000000002,
|
| 31 |
+
"loss": 3.3326,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 4.81,
|
| 36 |
+
"learning_rate": 0.0002495,
|
| 37 |
+
"loss": 3.0304,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.81,
|
| 42 |
+
"eval_loss": 1.5675914287567139,
|
| 43 |
+
"eval_runtime": 78.6195,
|
| 44 |
+
"eval_samples_per_second": 14.144,
|
| 45 |
+
"eval_steps_per_second": 3.536,
|
| 46 |
+
"eval_wer": 1.0554260089686098,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 5.77,
|
| 51 |
+
"learning_rate": 0.0002995,
|
| 52 |
+
"loss": 2.0254,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 6.73,
|
| 57 |
+
"learning_rate": 0.0003495,
|
| 58 |
+
"loss": 1.7226,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 7.69,
|
| 63 |
+
"learning_rate": 0.0003995,
|
| 64 |
+
"loss": 1.6212,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 8.65,
|
| 69 |
+
"learning_rate": 0.00044950000000000003,
|
| 70 |
+
"loss": 1.5773,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 9.61,
|
| 75 |
+
"learning_rate": 0.0004995,
|
| 76 |
+
"loss": 1.5263,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 9.61,
|
| 81 |
+
"eval_loss": 0.46925902366638184,
|
| 82 |
+
"eval_runtime": 52.7689,
|
| 83 |
+
"eval_samples_per_second": 21.073,
|
| 84 |
+
"eval_steps_per_second": 5.268,
|
| 85 |
+
"eval_wer": 0.8023318385650224,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 10.58,
|
| 90 |
+
"learning_rate": 0.0005495,
|
| 91 |
+
"loss": 1.5324,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 11.54,
|
| 96 |
+
"learning_rate": 0.0005995000000000001,
|
| 97 |
+
"loss": 1.5058,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 12.5,
|
| 102 |
+
"learning_rate": 0.0006495,
|
| 103 |
+
"loss": 1.512,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 13.46,
|
| 108 |
+
"learning_rate": 0.0006995,
|
| 109 |
+
"loss": 1.5222,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 14.42,
|
| 114 |
+
"learning_rate": 0.0007495000000000001,
|
| 115 |
+
"loss": 1.5299,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 14.42,
|
| 120 |
+
"eval_loss": 0.43676647543907166,
|
| 121 |
+
"eval_runtime": 52.3008,
|
| 122 |
+
"eval_samples_per_second": 21.262,
|
| 123 |
+
"eval_steps_per_second": 5.315,
|
| 124 |
+
"eval_wer": 0.7311210762331839,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 15.38,
|
| 129 |
+
"learning_rate": 0.0007995,
|
| 130 |
+
"loss": 1.5168,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 16.35,
|
| 135 |
+
"learning_rate": 0.0008495000000000001,
|
| 136 |
+
"loss": 1.5392,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 17.31,
|
| 141 |
+
"learning_rate": 0.0008995,
|
| 142 |
+
"loss": 1.5365,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 18.27,
|
| 147 |
+
"learning_rate": 0.0009495,
|
| 148 |
+
"loss": 1.4991,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 19.23,
|
| 153 |
+
"learning_rate": 0.0009995000000000002,
|
| 154 |
+
"loss": 1.5063,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 19.23,
|
| 159 |
+
"eval_loss": 0.43596330285072327,
|
| 160 |
+
"eval_runtime": 50.4313,
|
| 161 |
+
"eval_samples_per_second": 22.05,
|
| 162 |
+
"eval_steps_per_second": 5.512,
|
| 163 |
+
"eval_wer": 0.7302242152466367,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 20.19,
|
| 168 |
+
"learning_rate": 0.0009882142857142856,
|
| 169 |
+
"loss": 1.5283,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 21.15,
|
| 174 |
+
"learning_rate": 0.0009763095238095238,
|
| 175 |
+
"loss": 1.5146,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 22.12,
|
| 180 |
+
"learning_rate": 0.0009644047619047619,
|
| 181 |
+
"loss": 1.5236,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 23.08,
|
| 186 |
+
"learning_rate": 0.0009525,
|
| 187 |
+
"loss": 1.4972,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 24.04,
|
| 192 |
+
"learning_rate": 0.0009405952380952381,
|
| 193 |
+
"loss": 1.455,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 24.04,
|
| 198 |
+
"eval_loss": 0.4213278293609619,
|
| 199 |
+
"eval_runtime": 50.1354,
|
| 200 |
+
"eval_samples_per_second": 22.18,
|
| 201 |
+
"eval_steps_per_second": 5.545,
|
| 202 |
+
"eval_wer": 0.6692376681614349,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 25.0,
|
| 207 |
+
"learning_rate": 0.0009286904761904762,
|
| 208 |
+
"loss": 1.4424,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 25.96,
|
| 213 |
+
"learning_rate": 0.0009167857142857144,
|
| 214 |
+
"loss": 1.4434,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 26.92,
|
| 219 |
+
"learning_rate": 0.0009048809523809524,
|
| 220 |
+
"loss": 1.4513,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 27.88,
|
| 225 |
+
"learning_rate": 0.0008929761904761905,
|
| 226 |
+
"loss": 1.4328,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 28.84,
|
| 231 |
+
"learning_rate": 0.0008811904761904762,
|
| 232 |
+
"loss": 1.4755,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 28.84,
|
| 237 |
+
"eval_loss": 0.4329236149787903,
|
| 238 |
+
"eval_runtime": 50.0668,
|
| 239 |
+
"eval_samples_per_second": 22.21,
|
| 240 |
+
"eval_steps_per_second": 5.553,
|
| 241 |
+
"eval_wer": 0.5942600896860987,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 29.81,
|
| 246 |
+
"learning_rate": 0.0008692857142857144,
|
| 247 |
+
"loss": 1.4344,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 30.77,
|
| 252 |
+
"learning_rate": 0.0008573809523809523,
|
| 253 |
+
"loss": 1.4326,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 31.73,
|
| 258 |
+
"learning_rate": 0.0008454761904761905,
|
| 259 |
+
"loss": 1.4016,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 32.69,
|
| 264 |
+
"learning_rate": 0.0008335714285714285,
|
| 265 |
+
"loss": 1.3705,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 33.65,
|
| 270 |
+
"learning_rate": 0.0008216666666666667,
|
| 271 |
+
"loss": 1.352,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 33.65,
|
| 276 |
+
"eval_loss": 0.40741658210754395,
|
| 277 |
+
"eval_runtime": 76.8047,
|
| 278 |
+
"eval_samples_per_second": 14.478,
|
| 279 |
+
"eval_steps_per_second": 3.62,
|
| 280 |
+
"eval_wer": 0.5765022421524664,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 34.61,
|
| 285 |
+
"learning_rate": 0.0008097619047619048,
|
| 286 |
+
"loss": 1.3511,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 35.58,
|
| 291 |
+
"learning_rate": 0.0007978571428571428,
|
| 292 |
+
"loss": 1.3237,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 36.54,
|
| 297 |
+
"learning_rate": 0.000785952380952381,
|
| 298 |
+
"loss": 1.3165,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 37.5,
|
| 303 |
+
"learning_rate": 0.0007740476190476191,
|
| 304 |
+
"loss": 1.3342,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 38.46,
|
| 309 |
+
"learning_rate": 0.0007621428571428572,
|
| 310 |
+
"loss": 1.3122,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 38.46,
|
| 315 |
+
"eval_loss": 0.38659143447875977,
|
| 316 |
+
"eval_runtime": 80.318,
|
| 317 |
+
"eval_samples_per_second": 13.845,
|
| 318 |
+
"eval_steps_per_second": 3.461,
|
| 319 |
+
"eval_wer": 0.56304932735426,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 39.42,
|
| 324 |
+
"learning_rate": 0.0007502380952380953,
|
| 325 |
+
"loss": 1.3098,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 40.38,
|
| 330 |
+
"learning_rate": 0.0007383333333333334,
|
| 331 |
+
"loss": 1.3253,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 41.35,
|
| 336 |
+
"learning_rate": 0.0007264285714285714,
|
| 337 |
+
"loss": 1.2787,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 42.31,
|
| 342 |
+
"learning_rate": 0.0007146428571428572,
|
| 343 |
+
"loss": 1.274,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 43.27,
|
| 348 |
+
"learning_rate": 0.0007027380952380952,
|
| 349 |
+
"loss": 1.2799,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 43.27,
|
| 354 |
+
"eval_loss": 0.3859865069389343,
|
| 355 |
+
"eval_runtime": 77.9764,
|
| 356 |
+
"eval_samples_per_second": 14.261,
|
| 357 |
+
"eval_steps_per_second": 3.565,
|
| 358 |
+
"eval_wer": 0.5479820627802691,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 44.23,
|
| 363 |
+
"learning_rate": 0.0006908333333333333,
|
| 364 |
+
"loss": 1.2525,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 45.19,
|
| 369 |
+
"learning_rate": 0.0006789285714285714,
|
| 370 |
+
"loss": 1.2245,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 46.15,
|
| 375 |
+
"learning_rate": 0.0006670238095238096,
|
| 376 |
+
"loss": 1.2158,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 47.12,
|
| 381 |
+
"learning_rate": 0.0006551190476190476,
|
| 382 |
+
"loss": 1.2028,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 48.08,
|
| 387 |
+
"learning_rate": 0.0006432142857142857,
|
| 388 |
+
"loss": 1.212,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 48.08,
|
| 393 |
+
"eval_loss": 0.358958899974823,
|
| 394 |
+
"eval_runtime": 78.2642,
|
| 395 |
+
"eval_samples_per_second": 14.208,
|
| 396 |
+
"eval_steps_per_second": 3.552,
|
| 397 |
+
"eval_wer": 0.5316591928251121,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 49.04,
|
| 402 |
+
"learning_rate": 0.0006313095238095238,
|
| 403 |
+
"loss": 1.1808,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 50.0,
|
| 408 |
+
"learning_rate": 0.0006194047619047619,
|
| 409 |
+
"loss": 1.1783,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 50.96,
|
| 414 |
+
"learning_rate": 0.0006075000000000001,
|
| 415 |
+
"loss": 1.1675,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 51.92,
|
| 420 |
+
"learning_rate": 0.0005955952380952381,
|
| 421 |
+
"loss": 1.1867,
|
| 422 |
+
"step": 5400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 52.88,
|
| 426 |
+
"learning_rate": 0.0005836904761904763,
|
| 427 |
+
"loss": 1.1645,
|
| 428 |
+
"step": 5500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 52.88,
|
| 432 |
+
"eval_loss": 0.328298419713974,
|
| 433 |
+
"eval_runtime": 77.7284,
|
| 434 |
+
"eval_samples_per_second": 14.306,
|
| 435 |
+
"eval_steps_per_second": 3.577,
|
| 436 |
+
"eval_wer": 0.475695067264574,
|
| 437 |
+
"step": 5500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 53.84,
|
| 441 |
+
"learning_rate": 0.0005717857142857142,
|
| 442 |
+
"loss": 1.1407,
|
| 443 |
+
"step": 5600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 54.81,
|
| 447 |
+
"learning_rate": 0.0005598809523809523,
|
| 448 |
+
"loss": 1.1217,
|
| 449 |
+
"step": 5700
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 55.77,
|
| 453 |
+
"learning_rate": 0.0005479761904761905,
|
| 454 |
+
"loss": 1.118,
|
| 455 |
+
"step": 5800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 56.73,
|
| 459 |
+
"learning_rate": 0.0005360714285714285,
|
| 460 |
+
"loss": 1.0727,
|
| 461 |
+
"step": 5900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 57.69,
|
| 465 |
+
"learning_rate": 0.0005241666666666667,
|
| 466 |
+
"loss": 1.0854,
|
| 467 |
+
"step": 6000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 57.69,
|
| 471 |
+
"eval_loss": 0.3161650598049164,
|
| 472 |
+
"eval_runtime": 82.3243,
|
| 473 |
+
"eval_samples_per_second": 13.508,
|
| 474 |
+
"eval_steps_per_second": 3.377,
|
| 475 |
+
"eval_wer": 0.4686995515695067,
|
| 476 |
+
"step": 6000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 58.65,
|
| 480 |
+
"learning_rate": 0.0005122619047619048,
|
| 481 |
+
"loss": 1.0673,
|
| 482 |
+
"step": 6100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 59.61,
|
| 486 |
+
"learning_rate": 0.0005003571428571429,
|
| 487 |
+
"loss": 1.0595,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 60.58,
|
| 492 |
+
"learning_rate": 0.000488452380952381,
|
| 493 |
+
"loss": 1.0474,
|
| 494 |
+
"step": 6300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 61.54,
|
| 498 |
+
"learning_rate": 0.00047654761904761906,
|
| 499 |
+
"loss": 1.0424,
|
| 500 |
+
"step": 6400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 62.5,
|
| 504 |
+
"learning_rate": 0.00046464285714285715,
|
| 505 |
+
"loss": 1.0292,
|
| 506 |
+
"step": 6500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 62.5,
|
| 510 |
+
"eval_loss": 0.3125934600830078,
|
| 511 |
+
"eval_runtime": 78.0974,
|
| 512 |
+
"eval_samples_per_second": 14.239,
|
| 513 |
+
"eval_steps_per_second": 3.56,
|
| 514 |
+
"eval_wer": 0.44161434977578473,
|
| 515 |
+
"step": 6500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 63.46,
|
| 519 |
+
"learning_rate": 0.00045285714285714287,
|
| 520 |
+
"loss": 1.0259,
|
| 521 |
+
"step": 6600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 64.42,
|
| 525 |
+
"learning_rate": 0.00044095238095238096,
|
| 526 |
+
"loss": 1.0141,
|
| 527 |
+
"step": 6700
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 65.38,
|
| 531 |
+
"learning_rate": 0.00042916666666666667,
|
| 532 |
+
"loss": 0.9839,
|
| 533 |
+
"step": 6800
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 66.35,
|
| 537 |
+
"learning_rate": 0.00041726190476190476,
|
| 538 |
+
"loss": 0.9553,
|
| 539 |
+
"step": 6900
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 67.31,
|
| 543 |
+
"learning_rate": 0.00040535714285714285,
|
| 544 |
+
"loss": 0.9607,
|
| 545 |
+
"step": 7000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 67.31,
|
| 549 |
+
"eval_loss": 0.2990323007106781,
|
| 550 |
+
"eval_runtime": 78.5216,
|
| 551 |
+
"eval_samples_per_second": 14.162,
|
| 552 |
+
"eval_steps_per_second": 3.54,
|
| 553 |
+
"eval_wer": 0.40663677130044845,
|
| 554 |
+
"step": 7000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 68.27,
|
| 558 |
+
"learning_rate": 0.00039345238095238094,
|
| 559 |
+
"loss": 0.944,
|
| 560 |
+
"step": 7100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 69.23,
|
| 564 |
+
"learning_rate": 0.00038154761904761903,
|
| 565 |
+
"loss": 0.9203,
|
| 566 |
+
"step": 7200
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 70.19,
|
| 570 |
+
"learning_rate": 0.0003696428571428572,
|
| 571 |
+
"loss": 0.9164,
|
| 572 |
+
"step": 7300
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 71.15,
|
| 576 |
+
"learning_rate": 0.00035773809523809527,
|
| 577 |
+
"loss": 0.8996,
|
| 578 |
+
"step": 7400
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 72.12,
|
| 582 |
+
"learning_rate": 0.00034583333333333335,
|
| 583 |
+
"loss": 0.9156,
|
| 584 |
+
"step": 7500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 72.12,
|
| 588 |
+
"eval_loss": 0.28695425391197205,
|
| 589 |
+
"eval_runtime": 77.1644,
|
| 590 |
+
"eval_samples_per_second": 14.411,
|
| 591 |
+
"eval_steps_per_second": 3.603,
|
| 592 |
+
"eval_wer": 0.4008968609865471,
|
| 593 |
+
"step": 7500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 73.08,
|
| 597 |
+
"learning_rate": 0.0003339285714285714,
|
| 598 |
+
"loss": 0.8916,
|
| 599 |
+
"step": 7600
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 74.04,
|
| 603 |
+
"learning_rate": 0.00032202380952380953,
|
| 604 |
+
"loss": 0.8818,
|
| 605 |
+
"step": 7700
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 75.0,
|
| 609 |
+
"learning_rate": 0.0003101190476190476,
|
| 610 |
+
"loss": 0.8714,
|
| 611 |
+
"step": 7800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 75.96,
|
| 615 |
+
"learning_rate": 0.0002982142857142857,
|
| 616 |
+
"loss": 0.8529,
|
| 617 |
+
"step": 7900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 76.92,
|
| 621 |
+
"learning_rate": 0.0002863095238095238,
|
| 622 |
+
"loss": 0.8329,
|
| 623 |
+
"step": 8000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 76.92,
|
| 627 |
+
"eval_loss": 0.27907443046569824,
|
| 628 |
+
"eval_runtime": 76.6666,
|
| 629 |
+
"eval_samples_per_second": 14.504,
|
| 630 |
+
"eval_steps_per_second": 3.626,
|
| 631 |
+
"eval_wer": 0.39085201793721974,
|
| 632 |
+
"step": 8000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 77.88,
|
| 636 |
+
"learning_rate": 0.00027440476190476195,
|
| 637 |
+
"loss": 0.822,
|
| 638 |
+
"step": 8100
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 78.84,
|
| 642 |
+
"learning_rate": 0.00026250000000000004,
|
| 643 |
+
"loss": 0.8164,
|
| 644 |
+
"step": 8200
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 79.81,
|
| 648 |
+
"learning_rate": 0.0002505952380952381,
|
| 649 |
+
"loss": 0.8086,
|
| 650 |
+
"step": 8300
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 80.77,
|
| 654 |
+
"learning_rate": 0.00023869047619047622,
|
| 655 |
+
"loss": 0.7969,
|
| 656 |
+
"step": 8400
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 81.73,
|
| 660 |
+
"learning_rate": 0.00022678571428571428,
|
| 661 |
+
"loss": 0.7979,
|
| 662 |
+
"step": 8500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 81.73,
|
| 666 |
+
"eval_loss": 0.27704960107803345,
|
| 667 |
+
"eval_runtime": 76.769,
|
| 668 |
+
"eval_samples_per_second": 14.485,
|
| 669 |
+
"eval_steps_per_second": 3.621,
|
| 670 |
+
"eval_wer": 0.3669955156950673,
|
| 671 |
+
"step": 8500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 82.69,
|
| 675 |
+
"learning_rate": 0.0002148809523809524,
|
| 676 |
+
"loss": 0.7862,
|
| 677 |
+
"step": 8600
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 83.65,
|
| 681 |
+
"learning_rate": 0.0002029761904761905,
|
| 682 |
+
"loss": 0.7707,
|
| 683 |
+
"step": 8700
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 84.61,
|
| 687 |
+
"learning_rate": 0.00019107142857142855,
|
| 688 |
+
"loss": 0.7472,
|
| 689 |
+
"step": 8800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 85.58,
|
| 693 |
+
"learning_rate": 0.00017916666666666667,
|
| 694 |
+
"loss": 0.7583,
|
| 695 |
+
"step": 8900
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 86.54,
|
| 699 |
+
"learning_rate": 0.00016726190476190476,
|
| 700 |
+
"loss": 0.7144,
|
| 701 |
+
"step": 9000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 86.54,
|
| 705 |
+
"eval_loss": 0.2840667963027954,
|
| 706 |
+
"eval_runtime": 77.8679,
|
| 707 |
+
"eval_samples_per_second": 14.281,
|
| 708 |
+
"eval_steps_per_second": 3.57,
|
| 709 |
+
"eval_wer": 0.36609865470852015,
|
| 710 |
+
"step": 9000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 87.5,
|
| 714 |
+
"learning_rate": 0.00015535714285714287,
|
| 715 |
+
"loss": 0.7302,
|
| 716 |
+
"step": 9100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 88.46,
|
| 720 |
+
"learning_rate": 0.0001435714285714286,
|
| 721 |
+
"loss": 0.7214,
|
| 722 |
+
"step": 9200
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 89.42,
|
| 726 |
+
"learning_rate": 0.00013166666666666665,
|
| 727 |
+
"loss": 0.7179,
|
| 728 |
+
"step": 9300
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 90.38,
|
| 732 |
+
"learning_rate": 0.00011976190476190477,
|
| 733 |
+
"loss": 0.7037,
|
| 734 |
+
"step": 9400
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 91.35,
|
| 738 |
+
"learning_rate": 0.00010785714285714286,
|
| 739 |
+
"loss": 0.6997,
|
| 740 |
+
"step": 9500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 91.35,
|
| 744 |
+
"eval_loss": 0.2721162438392639,
|
| 745 |
+
"eval_runtime": 77.3441,
|
| 746 |
+
"eval_samples_per_second": 14.377,
|
| 747 |
+
"eval_steps_per_second": 3.594,
|
| 748 |
+
"eval_wer": 0.3485201793721973,
|
| 749 |
+
"step": 9500
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 92.31,
|
| 753 |
+
"learning_rate": 9.595238095238096e-05,
|
| 754 |
+
"loss": 0.676,
|
| 755 |
+
"step": 9600
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 93.27,
|
| 759 |
+
"learning_rate": 8.404761904761905e-05,
|
| 760 |
+
"loss": 0.6748,
|
| 761 |
+
"step": 9700
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 94.23,
|
| 765 |
+
"learning_rate": 7.214285714285715e-05,
|
| 766 |
+
"loss": 0.6668,
|
| 767 |
+
"step": 9800
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 95.19,
|
| 771 |
+
"learning_rate": 6.023809523809524e-05,
|
| 772 |
+
"loss": 0.662,
|
| 773 |
+
"step": 9900
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 96.15,
|
| 777 |
+
"learning_rate": 4.8333333333333334e-05,
|
| 778 |
+
"loss": 0.6568,
|
| 779 |
+
"step": 10000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 96.15,
|
| 783 |
+
"eval_loss": 0.2680588364601135,
|
| 784 |
+
"eval_runtime": 76.864,
|
| 785 |
+
"eval_samples_per_second": 14.467,
|
| 786 |
+
"eval_steps_per_second": 3.617,
|
| 787 |
+
"eval_wer": 0.34367713004484307,
|
| 788 |
+
"step": 10000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 97.12,
|
| 792 |
+
"learning_rate": 3.642857142857143e-05,
|
| 793 |
+
"loss": 0.6517,
|
| 794 |
+
"step": 10100
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 98.08,
|
| 798 |
+
"learning_rate": 2.4523809523809523e-05,
|
| 799 |
+
"loss": 0.6518,
|
| 800 |
+
"step": 10200
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 99.04,
|
| 804 |
+
"learning_rate": 1.2619047619047618e-05,
|
| 805 |
+
"loss": 0.6285,
|
| 806 |
+
"step": 10300
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 100.0,
|
| 810 |
+
"learning_rate": 7.142857142857143e-07,
|
| 811 |
+
"loss": 0.6282,
|
| 812 |
+
"step": 10400
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 100.0,
|
| 816 |
+
"step": 10400,
|
| 817 |
+
"total_flos": 4.70570622386933e+19,
|
| 818 |
+
"train_loss": 1.3617363874728863,
|
| 819 |
+
"train_runtime": 41188.5513,
|
| 820 |
+
"train_samples_per_second": 8.099,
|
| 821 |
+
"train_steps_per_second": 0.252
|
| 822 |
}
|
| 823 |
],
|
| 824 |
+
"max_steps": 10400,
|
| 825 |
+
"num_train_epochs": 100,
|
| 826 |
+
"total_flos": 4.70570622386933e+19,
|
| 827 |
"trial_name": null,
|
| 828 |
"trial_params": null
|
| 829 |
}
|