End of training
Browse files- all_results.json +10 -10
- eval_results.json +6 -6
- train_results.json +5 -5
- trainer_state.json +952 -34
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 360,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 3.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
"train_samples": 856,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 499.97,
|
| 3 |
+
"eval_loss": 1.0443373918533325,
|
| 4 |
+
"eval_runtime": 13.5396,
|
| 5 |
"eval_samples": 360,
|
| 6 |
+
"eval_samples_per_second": 26.589,
|
| 7 |
+
"eval_steps_per_second": 3.324,
|
| 8 |
+
"eval_wer": 0.5715405723245786,
|
| 9 |
+
"train_loss": 1.1417412889187153,
|
| 10 |
+
"train_runtime": 29000.6737,
|
| 11 |
"train_samples": 856,
|
| 12 |
+
"train_samples_per_second": 14.758,
|
| 13 |
+
"train_steps_per_second": 0.448
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 360,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 3.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 499.97,
|
| 3 |
+
"eval_loss": 1.0443373918533325,
|
| 4 |
+
"eval_runtime": 13.5396,
|
| 5 |
"eval_samples": 360,
|
| 6 |
+
"eval_samples_per_second": 26.589,
|
| 7 |
+
"eval_steps_per_second": 3.324,
|
| 8 |
+
"eval_wer": 0.5715405723245786
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 856,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 499.97,
|
| 3 |
+
"train_loss": 1.1417412889187153,
|
| 4 |
+
"train_runtime": 29000.6737,
|
| 5 |
"train_samples": 856,
|
| 6 |
+
"train_samples_per_second": 14.758,
|
| 7 |
+
"train_steps_per_second": 0.448
|
| 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,
|
|
@@ -10,112 +10,1030 @@
|
|
| 10 |
{
|
| 11 |
"epoch": 3.82,
|
| 12 |
"learning_rate": 3.7125e-06,
|
| 13 |
-
"loss": 25.
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 7.67,
|
| 18 |
"learning_rate": 7.4625e-06,
|
| 19 |
-
"loss": 12.
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 11.52,
|
| 24 |
"learning_rate": 1.1212499999999998e-05,
|
| 25 |
-
"loss": 7.
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 15.37,
|
| 30 |
"learning_rate": 1.49625e-05,
|
| 31 |
-
"loss": 5.
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 19.22,
|
| 36 |
"learning_rate": 1.8712499999999997e-05,
|
| 37 |
-
"loss": 4.
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 19.22,
|
| 42 |
-
"eval_loss": 4.
|
| 43 |
-
"eval_runtime": 13.
|
| 44 |
-
"eval_samples_per_second": 26.
|
| 45 |
-
"eval_steps_per_second": 3.
|
| 46 |
"eval_wer": 1.0,
|
| 47 |
"step": 500
|
| 48 |
},
|
| 49 |
{
|
| 50 |
"epoch": 23.07,
|
| 51 |
"learning_rate": 2.2462499999999997e-05,
|
| 52 |
-
"loss": 3.
|
| 53 |
"step": 600
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 26.9,
|
| 57 |
"learning_rate": 2.6212499999999997e-05,
|
| 58 |
-
"loss": 3.
|
| 59 |
"step": 700
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 30.75,
|
| 63 |
"learning_rate": 2.99625e-05,
|
| 64 |
-
"loss": 3.
|
| 65 |
"step": 800
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 34.6,
|
| 69 |
"learning_rate": 3.37125e-05,
|
| 70 |
-
"loss": 3.
|
| 71 |
"step": 900
|
| 72 |
},
|
| 73 |
{
|
| 74 |
"epoch": 38.45,
|
| 75 |
"learning_rate": 3.7462499999999996e-05,
|
| 76 |
-
"loss": 3.
|
| 77 |
"step": 1000
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 38.45,
|
| 81 |
-
"eval_loss": 3.
|
| 82 |
-
"eval_runtime": 13.
|
| 83 |
-
"eval_samples_per_second": 26.
|
| 84 |
-
"eval_steps_per_second": 3.
|
| 85 |
-
"eval_wer": 0
|
| 86 |
"step": 1000
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 42.3,
|
| 90 |
"learning_rate": 4.12125e-05,
|
| 91 |
-
"loss": 3.
|
| 92 |
"step": 1100
|
| 93 |
},
|
| 94 |
{
|
| 95 |
"epoch": 46.15,
|
| 96 |
"learning_rate": 4.4962499999999995e-05,
|
| 97 |
-
"loss": 3.
|
| 98 |
"step": 1200
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 49.97,
|
| 102 |
"learning_rate": 4.871249999999999e-05,
|
| 103 |
-
"loss":
|
| 104 |
"step": 1300
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch":
|
| 108 |
-
"
|
| 109 |
-
"
|
| 110 |
-
"
|
| 111 |
-
|
| 112 |
-
|
| 113 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 114 |
}
|
| 115 |
],
|
| 116 |
-
"max_steps":
|
| 117 |
-
"num_train_epochs":
|
| 118 |
-
"total_flos": 6.
|
| 119 |
"trial_name": null,
|
| 120 |
"trial_params": null
|
| 121 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 499.97196261682245,
|
| 5 |
+
"global_step": 13000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 10 |
{
|
| 11 |
"epoch": 3.82,
|
| 12 |
"learning_rate": 3.7125e-06,
|
| 13 |
+
"loss": 25.9068,
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 7.67,
|
| 18 |
"learning_rate": 7.4625e-06,
|
| 19 |
+
"loss": 12.5481,
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 11.52,
|
| 24 |
"learning_rate": 1.1212499999999998e-05,
|
| 25 |
+
"loss": 7.2704,
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 15.37,
|
| 30 |
"learning_rate": 1.49625e-05,
|
| 31 |
+
"loss": 5.7673,
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 19.22,
|
| 36 |
"learning_rate": 1.8712499999999997e-05,
|
| 37 |
+
"loss": 4.6694,
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 19.22,
|
| 42 |
+
"eval_loss": 4.045496463775635,
|
| 43 |
+
"eval_runtime": 13.5539,
|
| 44 |
+
"eval_samples_per_second": 26.561,
|
| 45 |
+
"eval_steps_per_second": 3.32,
|
| 46 |
"eval_wer": 1.0,
|
| 47 |
"step": 500
|
| 48 |
},
|
| 49 |
{
|
| 50 |
"epoch": 23.07,
|
| 51 |
"learning_rate": 2.2462499999999997e-05,
|
| 52 |
+
"loss": 3.9455,
|
| 53 |
"step": 600
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 26.9,
|
| 57 |
"learning_rate": 2.6212499999999997e-05,
|
| 58 |
+
"loss": 3.583,
|
| 59 |
"step": 700
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 30.75,
|
| 63 |
"learning_rate": 2.99625e-05,
|
| 64 |
+
"loss": 3.5138,
|
| 65 |
"step": 800
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 34.6,
|
| 69 |
"learning_rate": 3.37125e-05,
|
| 70 |
+
"loss": 3.4479,
|
| 71 |
"step": 900
|
| 72 |
},
|
| 73 |
{
|
| 74 |
"epoch": 38.45,
|
| 75 |
"learning_rate": 3.7462499999999996e-05,
|
| 76 |
+
"loss": 3.3907,
|
| 77 |
"step": 1000
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 38.45,
|
| 81 |
+
"eval_loss": 3.283632516860962,
|
| 82 |
+
"eval_runtime": 13.4556,
|
| 83 |
+
"eval_samples_per_second": 26.755,
|
| 84 |
+
"eval_steps_per_second": 3.344,
|
| 85 |
+
"eval_wer": 1.0,
|
| 86 |
"step": 1000
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 42.3,
|
| 90 |
"learning_rate": 4.12125e-05,
|
| 91 |
+
"loss": 3.3556,
|
| 92 |
"step": 1100
|
| 93 |
},
|
| 94 |
{
|
| 95 |
"epoch": 46.15,
|
| 96 |
"learning_rate": 4.4962499999999995e-05,
|
| 97 |
+
"loss": 3.3117,
|
| 98 |
"step": 1200
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 49.97,
|
| 102 |
"learning_rate": 4.871249999999999e-05,
|
| 103 |
+
"loss": 3.0309,
|
| 104 |
"step": 1300
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 53.82,
|
| 108 |
+
"learning_rate": 5.2462499999999994e-05,
|
| 109 |
+
"loss": 2.5273,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 57.67,
|
| 114 |
+
"learning_rate": 5.62125e-05,
|
| 115 |
+
"loss": 2.0866,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 57.67,
|
| 120 |
+
"eval_loss": 1.2787948846817017,
|
| 121 |
+
"eval_runtime": 13.4904,
|
| 122 |
+
"eval_samples_per_second": 26.686,
|
| 123 |
+
"eval_steps_per_second": 3.336,
|
| 124 |
+
"eval_wer": 0.7714621716973736,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 61.52,
|
| 129 |
+
"learning_rate": 5.9962499999999994e-05,
|
| 130 |
+
"loss": 1.8478,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 65.37,
|
| 135 |
+
"learning_rate": 6.37125e-05,
|
| 136 |
+
"loss": 1.7048,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 69.22,
|
| 141 |
+
"learning_rate": 6.7425e-05,
|
| 142 |
+
"loss": 1.6017,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 73.07,
|
| 147 |
+
"learning_rate": 7.11375e-05,
|
| 148 |
+
"loss": 1.4931,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 76.9,
|
| 153 |
+
"learning_rate": 7.48875e-05,
|
| 154 |
+
"loss": 1.4106,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 76.9,
|
| 159 |
+
"eval_loss": 0.7866009473800659,
|
| 160 |
+
"eval_runtime": 13.5793,
|
| 161 |
+
"eval_samples_per_second": 26.511,
|
| 162 |
+
"eval_steps_per_second": 3.314,
|
| 163 |
+
"eval_wer": 0.689141513132105,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 80.75,
|
| 168 |
+
"learning_rate": 7.433863636363635e-05,
|
| 169 |
+
"loss": 1.3808,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 84.6,
|
| 174 |
+
"learning_rate": 7.365681818181818e-05,
|
| 175 |
+
"loss": 1.3114,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 88.45,
|
| 180 |
+
"learning_rate": 7.2975e-05,
|
| 181 |
+
"loss": 1.2661,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 92.3,
|
| 186 |
+
"learning_rate": 7.229318181818181e-05,
|
| 187 |
+
"loss": 1.2103,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 96.15,
|
| 192 |
+
"learning_rate": 7.161136363636363e-05,
|
| 193 |
+
"loss": 1.1711,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 96.15,
|
| 198 |
+
"eval_loss": 0.655582070350647,
|
| 199 |
+
"eval_runtime": 13.3729,
|
| 200 |
+
"eval_samples_per_second": 26.92,
|
| 201 |
+
"eval_steps_per_second": 3.365,
|
| 202 |
+
"eval_wer": 0.6272050176401411,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 99.97,
|
| 207 |
+
"learning_rate": 7.092954545454545e-05,
|
| 208 |
+
"loss": 1.1192,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 103.82,
|
| 213 |
+
"learning_rate": 7.024772727272726e-05,
|
| 214 |
+
"loss": 1.0931,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 107.67,
|
| 219 |
+
"learning_rate": 6.956590909090908e-05,
|
| 220 |
+
"loss": 1.0688,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 111.52,
|
| 225 |
+
"learning_rate": 6.88840909090909e-05,
|
| 226 |
+
"loss": 1.0559,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 115.37,
|
| 231 |
+
"learning_rate": 6.82090909090909e-05,
|
| 232 |
+
"loss": 1.038,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 115.37,
|
| 237 |
+
"eval_loss": 0.6194556951522827,
|
| 238 |
+
"eval_runtime": 15.6261,
|
| 239 |
+
"eval_samples_per_second": 23.038,
|
| 240 |
+
"eval_steps_per_second": 2.88,
|
| 241 |
+
"eval_wer": 0.5680125441003528,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 119.22,
|
| 246 |
+
"learning_rate": 6.752727272727272e-05,
|
| 247 |
+
"loss": 1.0038,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 123.07,
|
| 252 |
+
"learning_rate": 6.684545454545454e-05,
|
| 253 |
+
"loss": 0.9754,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 126.9,
|
| 258 |
+
"learning_rate": 6.616363636363636e-05,
|
| 259 |
+
"loss": 0.9372,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 130.75,
|
| 264 |
+
"learning_rate": 6.548181818181817e-05,
|
| 265 |
+
"loss": 0.9232,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 134.6,
|
| 270 |
+
"learning_rate": 6.479999999999999e-05,
|
| 271 |
+
"loss": 0.8989,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 134.6,
|
| 276 |
+
"eval_loss": 0.6562529802322388,
|
| 277 |
+
"eval_runtime": 13.2747,
|
| 278 |
+
"eval_samples_per_second": 27.119,
|
| 279 |
+
"eval_steps_per_second": 3.39,
|
| 280 |
+
"eval_wer": 0.560172481379851,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 138.45,
|
| 285 |
+
"learning_rate": 6.411818181818182e-05,
|
| 286 |
+
"loss": 0.882,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 142.3,
|
| 291 |
+
"learning_rate": 6.343636363636364e-05,
|
| 292 |
+
"loss": 0.868,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 146.15,
|
| 297 |
+
"learning_rate": 6.275454545454545e-05,
|
| 298 |
+
"loss": 0.8438,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 149.97,
|
| 303 |
+
"learning_rate": 6.207272727272727e-05,
|
| 304 |
+
"loss": 0.8209,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 153.82,
|
| 309 |
+
"learning_rate": 6.139090909090908e-05,
|
| 310 |
+
"loss": 0.8021,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 153.82,
|
| 315 |
+
"eval_loss": 0.6643799543380737,
|
| 316 |
+
"eval_runtime": 13.2721,
|
| 317 |
+
"eval_samples_per_second": 27.125,
|
| 318 |
+
"eval_steps_per_second": 3.391,
|
| 319 |
+
"eval_wer": 0.5327322618580949,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 157.67,
|
| 324 |
+
"learning_rate": 6.07090909090909e-05,
|
| 325 |
+
"loss": 0.7682,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 161.52,
|
| 330 |
+
"learning_rate": 6.0027272727272725e-05,
|
| 331 |
+
"loss": 0.7496,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 165.37,
|
| 336 |
+
"learning_rate": 5.934545454545454e-05,
|
| 337 |
+
"loss": 0.7427,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 169.22,
|
| 342 |
+
"learning_rate": 5.866363636363636e-05,
|
| 343 |
+
"loss": 0.7336,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 173.07,
|
| 348 |
+
"learning_rate": 5.7981818181818174e-05,
|
| 349 |
+
"loss": 0.7161,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 173.07,
|
| 354 |
+
"eval_loss": 0.6844300031661987,
|
| 355 |
+
"eval_runtime": 13.407,
|
| 356 |
+
"eval_samples_per_second": 26.852,
|
| 357 |
+
"eval_steps_per_second": 3.356,
|
| 358 |
+
"eval_wer": 0.5252842022736182,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 176.9,
|
| 363 |
+
"learning_rate": 5.73e-05,
|
| 364 |
+
"loss": 0.6994,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 180.75,
|
| 369 |
+
"learning_rate": 5.6618181818181806e-05,
|
| 370 |
+
"loss": 0.6778,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 184.6,
|
| 375 |
+
"learning_rate": 5.593636363636363e-05,
|
| 376 |
+
"loss": 0.6709,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 188.45,
|
| 381 |
+
"learning_rate": 5.5254545454545446e-05,
|
| 382 |
+
"loss": 0.6551,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 192.3,
|
| 387 |
+
"learning_rate": 5.457272727272727e-05,
|
| 388 |
+
"loss": 0.6449,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 192.3,
|
| 393 |
+
"eval_loss": 0.7017893195152283,
|
| 394 |
+
"eval_runtime": 13.3622,
|
| 395 |
+
"eval_samples_per_second": 26.942,
|
| 396 |
+
"eval_steps_per_second": 3.368,
|
| 397 |
+
"eval_wer": 0.5331242649941199,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 196.15,
|
| 402 |
+
"learning_rate": 5.389090909090909e-05,
|
| 403 |
+
"loss": 0.6229,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 199.97,
|
| 408 |
+
"learning_rate": 5.32090909090909e-05,
|
| 409 |
+
"loss": 0.5977,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 203.82,
|
| 414 |
+
"learning_rate": 5.2527272727272725e-05,
|
| 415 |
+
"loss": 0.5985,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 207.67,
|
| 420 |
+
"learning_rate": 5.185227272727272e-05,
|
| 421 |
+
"loss": 0.5819,
|
| 422 |
+
"step": 5400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 211.52,
|
| 426 |
+
"learning_rate": 5.117045454545454e-05,
|
| 427 |
+
"loss": 0.5659,
|
| 428 |
+
"step": 5500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 211.52,
|
| 432 |
+
"eval_loss": 0.7450936436653137,
|
| 433 |
+
"eval_runtime": 13.5305,
|
| 434 |
+
"eval_samples_per_second": 26.606,
|
| 435 |
+
"eval_steps_per_second": 3.326,
|
| 436 |
+
"eval_wer": 0.546452371618973,
|
| 437 |
+
"step": 5500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 215.37,
|
| 441 |
+
"learning_rate": 5.048863636363636e-05,
|
| 442 |
+
"loss": 0.5615,
|
| 443 |
+
"step": 5600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 219.22,
|
| 447 |
+
"learning_rate": 4.980681818181818e-05,
|
| 448 |
+
"loss": 0.5475,
|
| 449 |
+
"step": 5700
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 223.07,
|
| 453 |
+
"learning_rate": 4.9124999999999995e-05,
|
| 454 |
+
"loss": 0.529,
|
| 455 |
+
"step": 5800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 226.9,
|
| 459 |
+
"learning_rate": 4.844318181818181e-05,
|
| 460 |
+
"loss": 0.5217,
|
| 461 |
+
"step": 5900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 230.75,
|
| 465 |
+
"learning_rate": 4.7761363636363634e-05,
|
| 466 |
+
"loss": 0.5118,
|
| 467 |
+
"step": 6000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 230.75,
|
| 471 |
+
"eval_loss": 0.7857084274291992,
|
| 472 |
+
"eval_runtime": 13.2806,
|
| 473 |
+
"eval_samples_per_second": 27.107,
|
| 474 |
+
"eval_steps_per_second": 3.388,
|
| 475 |
+
"eval_wer": 0.5386123088984712,
|
| 476 |
+
"step": 6000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 234.6,
|
| 480 |
+
"learning_rate": 4.707954545454545e-05,
|
| 481 |
+
"loss": 0.4996,
|
| 482 |
+
"step": 6100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 238.45,
|
| 486 |
+
"learning_rate": 4.639772727272727e-05,
|
| 487 |
+
"loss": 0.4808,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 242.3,
|
| 492 |
+
"learning_rate": 4.5715909090909083e-05,
|
| 493 |
+
"loss": 0.4632,
|
| 494 |
+
"step": 6300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 246.15,
|
| 498 |
+
"learning_rate": 4.5034090909090907e-05,
|
| 499 |
+
"loss": 0.4643,
|
| 500 |
+
"step": 6400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 249.97,
|
| 504 |
+
"learning_rate": 4.435227272727272e-05,
|
| 505 |
+
"loss": 0.4385,
|
| 506 |
+
"step": 6500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 249.97,
|
| 510 |
+
"eval_loss": 0.8062427043914795,
|
| 511 |
+
"eval_runtime": 13.503,
|
| 512 |
+
"eval_samples_per_second": 26.661,
|
| 513 |
+
"eval_steps_per_second": 3.333,
|
| 514 |
+
"eval_wer": 0.5382203057624461,
|
| 515 |
+
"step": 6500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 253.82,
|
| 519 |
+
"learning_rate": 4.367045454545454e-05,
|
| 520 |
+
"loss": 0.439,
|
| 521 |
+
"step": 6600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 257.67,
|
| 525 |
+
"learning_rate": 4.298863636363636e-05,
|
| 526 |
+
"loss": 0.4274,
|
| 527 |
+
"step": 6700
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 261.52,
|
| 531 |
+
"learning_rate": 4.230681818181818e-05,
|
| 532 |
+
"loss": 0.4177,
|
| 533 |
+
"step": 6800
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 265.37,
|
| 537 |
+
"learning_rate": 4.1625e-05,
|
| 538 |
+
"loss": 0.411,
|
| 539 |
+
"step": 6900
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 269.22,
|
| 543 |
+
"learning_rate": 4.094318181818181e-05,
|
| 544 |
+
"loss": 0.3984,
|
| 545 |
+
"step": 7000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 269.22,
|
| 549 |
+
"eval_loss": 0.8315810561180115,
|
| 550 |
+
"eval_runtime": 13.3889,
|
| 551 |
+
"eval_samples_per_second": 26.888,
|
| 552 |
+
"eval_steps_per_second": 3.361,
|
| 553 |
+
"eval_wer": 0.5621324970599765,
|
| 554 |
+
"step": 7000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 273.07,
|
| 558 |
+
"learning_rate": 4.0261363636363635e-05,
|
| 559 |
+
"loss": 0.3935,
|
| 560 |
+
"step": 7100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 276.9,
|
| 564 |
+
"learning_rate": 3.957954545454545e-05,
|
| 565 |
+
"loss": 0.3842,
|
| 566 |
+
"step": 7200
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 280.75,
|
| 570 |
+
"learning_rate": 3.8897727272727274e-05,
|
| 571 |
+
"loss": 0.3756,
|
| 572 |
+
"step": 7300
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 284.6,
|
| 576 |
+
"learning_rate": 3.8215909090909084e-05,
|
| 577 |
+
"loss": 0.3671,
|
| 578 |
+
"step": 7400
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 288.45,
|
| 582 |
+
"learning_rate": 3.754090909090909e-05,
|
| 583 |
+
"loss": 0.3666,
|
| 584 |
+
"step": 7500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 288.45,
|
| 588 |
+
"eval_loss": 0.8735576868057251,
|
| 589 |
+
"eval_runtime": 13.3799,
|
| 590 |
+
"eval_samples_per_second": 26.906,
|
| 591 |
+
"eval_steps_per_second": 3.363,
|
| 592 |
+
"eval_wer": 0.5503724029792239,
|
| 593 |
+
"step": 7500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 292.3,
|
| 597 |
+
"learning_rate": 3.6859090909090904e-05,
|
| 598 |
+
"loss": 0.3454,
|
| 599 |
+
"step": 7600
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 296.15,
|
| 603 |
+
"learning_rate": 3.617727272727273e-05,
|
| 604 |
+
"loss": 0.3486,
|
| 605 |
+
"step": 7700
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 299.97,
|
| 609 |
+
"learning_rate": 3.5495454545454544e-05,
|
| 610 |
+
"loss": 0.3347,
|
| 611 |
+
"step": 7800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 303.82,
|
| 615 |
+
"learning_rate": 3.481363636363636e-05,
|
| 616 |
+
"loss": 0.3262,
|
| 617 |
+
"step": 7900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 307.67,
|
| 621 |
+
"learning_rate": 3.413181818181818e-05,
|
| 622 |
+
"loss": 0.3256,
|
| 623 |
+
"step": 8000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 307.67,
|
| 627 |
+
"eval_loss": 0.9133402705192566,
|
| 628 |
+
"eval_runtime": 13.8839,
|
| 629 |
+
"eval_samples_per_second": 25.929,
|
| 630 |
+
"eval_steps_per_second": 3.241,
|
| 631 |
+
"eval_wer": 0.568796550372403,
|
| 632 |
+
"step": 8000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 311.52,
|
| 636 |
+
"learning_rate": 3.345e-05,
|
| 637 |
+
"loss": 0.3062,
|
| 638 |
+
"step": 8100
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 315.37,
|
| 642 |
+
"learning_rate": 3.2768181818181816e-05,
|
| 643 |
+
"loss": 0.3053,
|
| 644 |
+
"step": 8200
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 319.22,
|
| 648 |
+
"learning_rate": 3.208636363636363e-05,
|
| 649 |
+
"loss": 0.3003,
|
| 650 |
+
"step": 8300
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 323.07,
|
| 654 |
+
"learning_rate": 3.140454545454545e-05,
|
| 655 |
+
"loss": 0.2975,
|
| 656 |
+
"step": 8400
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 326.9,
|
| 660 |
+
"learning_rate": 3.072272727272727e-05,
|
| 661 |
+
"loss": 0.289,
|
| 662 |
+
"step": 8500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 326.9,
|
| 666 |
+
"eval_loss": 0.9556481242179871,
|
| 667 |
+
"eval_runtime": 13.3982,
|
| 668 |
+
"eval_samples_per_second": 26.869,
|
| 669 |
+
"eval_steps_per_second": 3.359,
|
| 670 |
+
"eval_wer": 0.5684045472363779,
|
| 671 |
+
"step": 8500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 330.75,
|
| 675 |
+
"learning_rate": 3.004090909090909e-05,
|
| 676 |
+
"loss": 0.2802,
|
| 677 |
+
"step": 8600
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 334.6,
|
| 681 |
+
"learning_rate": 2.9359090909090905e-05,
|
| 682 |
+
"loss": 0.2736,
|
| 683 |
+
"step": 8700
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 338.45,
|
| 687 |
+
"learning_rate": 2.8677272727272725e-05,
|
| 688 |
+
"loss": 0.2683,
|
| 689 |
+
"step": 8800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 342.3,
|
| 693 |
+
"learning_rate": 2.799545454545454e-05,
|
| 694 |
+
"loss": 0.2681,
|
| 695 |
+
"step": 8900
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 346.15,
|
| 699 |
+
"learning_rate": 2.731363636363636e-05,
|
| 700 |
+
"loss": 0.2663,
|
| 701 |
+
"step": 9000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 346.15,
|
| 705 |
+
"eval_loss": 0.9343604445457458,
|
| 706 |
+
"eval_runtime": 13.4662,
|
| 707 |
+
"eval_samples_per_second": 26.734,
|
| 708 |
+
"eval_steps_per_second": 3.342,
|
| 709 |
+
"eval_wer": 0.5707565660525284,
|
| 710 |
+
"step": 9000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 349.97,
|
| 714 |
+
"learning_rate": 2.6631818181818177e-05,
|
| 715 |
+
"loss": 0.2597,
|
| 716 |
+
"step": 9100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 353.82,
|
| 720 |
+
"learning_rate": 2.5949999999999997e-05,
|
| 721 |
+
"loss": 0.2636,
|
| 722 |
+
"step": 9200
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 357.67,
|
| 726 |
+
"learning_rate": 2.5268181818181817e-05,
|
| 727 |
+
"loss": 0.2538,
|
| 728 |
+
"step": 9300
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 361.52,
|
| 732 |
+
"learning_rate": 2.4586363636363637e-05,
|
| 733 |
+
"loss": 0.2505,
|
| 734 |
+
"step": 9400
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 365.37,
|
| 738 |
+
"learning_rate": 2.3904545454545453e-05,
|
| 739 |
+
"loss": 0.2445,
|
| 740 |
+
"step": 9500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 365.37,
|
| 744 |
+
"eval_loss": 0.9471534490585327,
|
| 745 |
+
"eval_runtime": 13.371,
|
| 746 |
+
"eval_samples_per_second": 26.924,
|
| 747 |
+
"eval_steps_per_second": 3.365,
|
| 748 |
+
"eval_wer": 0.5589964719717758,
|
| 749 |
+
"step": 9500
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 369.22,
|
| 753 |
+
"learning_rate": 2.3222727272727273e-05,
|
| 754 |
+
"loss": 0.2397,
|
| 755 |
+
"step": 9600
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 373.07,
|
| 759 |
+
"learning_rate": 2.254090909090909e-05,
|
| 760 |
+
"loss": 0.2395,
|
| 761 |
+
"step": 9700
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 376.9,
|
| 765 |
+
"learning_rate": 2.185909090909091e-05,
|
| 766 |
+
"loss": 0.2277,
|
| 767 |
+
"step": 9800
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 380.75,
|
| 771 |
+
"learning_rate": 2.1184090909090906e-05,
|
| 772 |
+
"loss": 0.2305,
|
| 773 |
+
"step": 9900
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 384.6,
|
| 777 |
+
"learning_rate": 2.0509090909090907e-05,
|
| 778 |
+
"loss": 0.2289,
|
| 779 |
+
"step": 10000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 384.6,
|
| 783 |
+
"eval_loss": 0.9713197350502014,
|
| 784 |
+
"eval_runtime": 13.5048,
|
| 785 |
+
"eval_samples_per_second": 26.657,
|
| 786 |
+
"eval_steps_per_second": 3.332,
|
| 787 |
+
"eval_wer": 0.5672285378283026,
|
| 788 |
+
"step": 10000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 388.45,
|
| 792 |
+
"learning_rate": 1.9827272727272727e-05,
|
| 793 |
+
"loss": 0.2139,
|
| 794 |
+
"step": 10100
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 392.3,
|
| 798 |
+
"learning_rate": 1.9145454545454543e-05,
|
| 799 |
+
"loss": 0.2214,
|
| 800 |
+
"step": 10200
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 396.15,
|
| 804 |
+
"learning_rate": 1.846363636363636e-05,
|
| 805 |
+
"loss": 0.2053,
|
| 806 |
+
"step": 10300
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 399.97,
|
| 810 |
+
"learning_rate": 1.7781818181818183e-05,
|
| 811 |
+
"loss": 0.2007,
|
| 812 |
+
"step": 10400
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 403.82,
|
| 816 |
+
"learning_rate": 1.71e-05,
|
| 817 |
+
"loss": 0.2048,
|
| 818 |
+
"step": 10500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 403.82,
|
| 822 |
+
"eval_loss": 0.9977540969848633,
|
| 823 |
+
"eval_runtime": 15.7748,
|
| 824 |
+
"eval_samples_per_second": 22.821,
|
| 825 |
+
"eval_steps_per_second": 2.853,
|
| 826 |
+
"eval_wer": 0.5762446099568796,
|
| 827 |
+
"step": 10500
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 407.67,
|
| 831 |
+
"learning_rate": 1.641818181818182e-05,
|
| 832 |
+
"loss": 0.1995,
|
| 833 |
+
"step": 10600
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 411.52,
|
| 837 |
+
"learning_rate": 1.5736363636363635e-05,
|
| 838 |
+
"loss": 0.1994,
|
| 839 |
+
"step": 10700
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 415.37,
|
| 843 |
+
"learning_rate": 1.5054545454545453e-05,
|
| 844 |
+
"loss": 0.2005,
|
| 845 |
+
"step": 10800
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 419.22,
|
| 849 |
+
"learning_rate": 1.4372727272727271e-05,
|
| 850 |
+
"loss": 0.1988,
|
| 851 |
+
"step": 10900
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 423.07,
|
| 855 |
+
"learning_rate": 1.369090909090909e-05,
|
| 856 |
+
"loss": 0.1857,
|
| 857 |
+
"step": 11000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 423.07,
|
| 861 |
+
"eval_loss": 1.0229874849319458,
|
| 862 |
+
"eval_runtime": 13.4188,
|
| 863 |
+
"eval_samples_per_second": 26.828,
|
| 864 |
+
"eval_steps_per_second": 3.354,
|
| 865 |
+
"eval_wer": 0.5797726381811055,
|
| 866 |
+
"step": 11000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 426.9,
|
| 870 |
+
"learning_rate": 1.3009090909090907e-05,
|
| 871 |
+
"loss": 0.1793,
|
| 872 |
+
"step": 11100
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 430.75,
|
| 876 |
+
"learning_rate": 1.2327272727272726e-05,
|
| 877 |
+
"loss": 0.1835,
|
| 878 |
+
"step": 11200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 434.6,
|
| 882 |
+
"learning_rate": 1.1645454545454545e-05,
|
| 883 |
+
"loss": 0.1729,
|
| 884 |
+
"step": 11300
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 438.45,
|
| 888 |
+
"learning_rate": 1.0963636363636363e-05,
|
| 889 |
+
"loss": 0.1751,
|
| 890 |
+
"step": 11400
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 442.3,
|
| 894 |
+
"learning_rate": 1.0281818181818181e-05,
|
| 895 |
+
"loss": 0.1751,
|
| 896 |
+
"step": 11500
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 442.3,
|
| 900 |
+
"eval_loss": 1.0409098863601685,
|
| 901 |
+
"eval_runtime": 13.4411,
|
| 902 |
+
"eval_samples_per_second": 26.784,
|
| 903 |
+
"eval_steps_per_second": 3.348,
|
| 904 |
+
"eval_wer": 0.5754606036848294,
|
| 905 |
+
"step": 11500
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 446.15,
|
| 909 |
+
"learning_rate": 9.6e-06,
|
| 910 |
+
"loss": 0.1743,
|
| 911 |
+
"step": 11600
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 449.97,
|
| 915 |
+
"learning_rate": 8.918181818181818e-06,
|
| 916 |
+
"loss": 0.1704,
|
| 917 |
+
"step": 11700
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 453.82,
|
| 921 |
+
"learning_rate": 8.236363636363636e-06,
|
| 922 |
+
"loss": 0.1691,
|
| 923 |
+
"step": 11800
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 457.67,
|
| 927 |
+
"learning_rate": 7.554545454545453e-06,
|
| 928 |
+
"loss": 0.1635,
|
| 929 |
+
"step": 11900
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 461.52,
|
| 933 |
+
"learning_rate": 6.872727272727273e-06,
|
| 934 |
+
"loss": 0.1688,
|
| 935 |
+
"step": 12000
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 461.52,
|
| 939 |
+
"eval_loss": 1.0444588661193848,
|
| 940 |
+
"eval_runtime": 13.5066,
|
| 941 |
+
"eval_samples_per_second": 26.654,
|
| 942 |
+
"eval_steps_per_second": 3.332,
|
| 943 |
+
"eval_wer": 0.5727165817326538,
|
| 944 |
+
"step": 12000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 465.37,
|
| 948 |
+
"learning_rate": 6.190909090909091e-06,
|
| 949 |
+
"loss": 0.1666,
|
| 950 |
+
"step": 12100
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 469.22,
|
| 954 |
+
"learning_rate": 5.509090909090908e-06,
|
| 955 |
+
"loss": 0.1619,
|
| 956 |
+
"step": 12200
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 473.07,
|
| 960 |
+
"learning_rate": 4.827272727272726e-06,
|
| 961 |
+
"loss": 0.161,
|
| 962 |
+
"step": 12300
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 476.9,
|
| 966 |
+
"learning_rate": 4.145454545454545e-06,
|
| 967 |
+
"loss": 0.1527,
|
| 968 |
+
"step": 12400
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 480.75,
|
| 972 |
+
"learning_rate": 3.4636363636363636e-06,
|
| 973 |
+
"loss": 0.1633,
|
| 974 |
+
"step": 12500
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 480.75,
|
| 978 |
+
"eval_loss": 1.048424482345581,
|
| 979 |
+
"eval_runtime": 13.4794,
|
| 980 |
+
"eval_samples_per_second": 26.707,
|
| 981 |
+
"eval_steps_per_second": 3.338,
|
| 982 |
+
"eval_wer": 0.5738925911407291,
|
| 983 |
+
"step": 12500
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 484.6,
|
| 987 |
+
"learning_rate": 2.7818181818181817e-06,
|
| 988 |
+
"loss": 0.1604,
|
| 989 |
+
"step": 12600
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 488.45,
|
| 993 |
+
"learning_rate": 2.1e-06,
|
| 994 |
+
"loss": 0.156,
|
| 995 |
+
"step": 12700
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 492.3,
|
| 999 |
+
"learning_rate": 1.418181818181818e-06,
|
| 1000 |
+
"loss": 0.1579,
|
| 1001 |
+
"step": 12800
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 496.15,
|
| 1005 |
+
"learning_rate": 7.363636363636363e-07,
|
| 1006 |
+
"loss": 0.1557,
|
| 1007 |
+
"step": 12900
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 499.97,
|
| 1011 |
+
"learning_rate": 5.454545454545454e-08,
|
| 1012 |
+
"loss": 0.1488,
|
| 1013 |
+
"step": 13000
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 499.97,
|
| 1017 |
+
"eval_loss": 1.0443373918533325,
|
| 1018 |
+
"eval_runtime": 13.3989,
|
| 1019 |
+
"eval_samples_per_second": 26.868,
|
| 1020 |
+
"eval_steps_per_second": 3.358,
|
| 1021 |
+
"eval_wer": 0.5715405723245786,
|
| 1022 |
+
"step": 13000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 499.97,
|
| 1026 |
+
"step": 13000,
|
| 1027 |
+
"total_flos": 6.567667376811244e+19,
|
| 1028 |
+
"train_loss": 1.1417412889187153,
|
| 1029 |
+
"train_runtime": 29000.6737,
|
| 1030 |
+
"train_samples_per_second": 14.758,
|
| 1031 |
+
"train_steps_per_second": 0.448
|
| 1032 |
}
|
| 1033 |
],
|
| 1034 |
+
"max_steps": 13000,
|
| 1035 |
+
"num_train_epochs": 500,
|
| 1036 |
+
"total_flos": 6.567667376811244e+19,
|
| 1037 |
"trial_name": null,
|
| 1038 |
"trial_params": null
|
| 1039 |
}
|