Commit
•
828b542
1
Parent(s):
8de96ae
End of training
Browse files- all_results.json +10 -10
- eval_results.json +6 -6
- train_results.json +5 -5
- trainer_state.json +661 -46
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 4483,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 2.
|
| 8 |
-
"eval_wer": 2.
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
"train_samples": 10623,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.5350876450538635,
|
| 4 |
+
"eval_runtime": 237.6626,
|
| 5 |
"eval_samples": 4483,
|
| 6 |
+
"eval_samples_per_second": 18.863,
|
| 7 |
+
"eval_steps_per_second": 2.36,
|
| 8 |
+
"eval_wer": 2.6187722419928825,
|
| 9 |
+
"train_loss": 2.741275282851211,
|
| 10 |
+
"train_runtime": 31325.7849,
|
| 11 |
"train_samples": 10623,
|
| 12 |
+
"train_samples_per_second": 16.956,
|
| 13 |
+
"train_steps_per_second": 0.354
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 4483,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 2.
|
| 8 |
-
"eval_wer": 2.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.5350876450538635,
|
| 4 |
+
"eval_runtime": 237.6626,
|
| 5 |
"eval_samples": 4483,
|
| 6 |
+
"eval_samples_per_second": 18.863,
|
| 7 |
+
"eval_steps_per_second": 2.36,
|
| 8 |
+
"eval_wer": 2.6187722419928825
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 10623,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"train_loss": 2.741275282851211,
|
| 4 |
+
"train_runtime": 31325.7849,
|
| 5 |
"train_samples": 10623,
|
| 6 |
+
"train_samples_per_second": 16.956,
|
| 7 |
+
"train_steps_per_second": 0.354
|
| 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,
|
|
@@ -10,166 +10,781 @@
|
|
| 10 |
{
|
| 11 |
"epoch": 0.45,
|
| 12 |
"learning_rate": 3.6e-06,
|
| 13 |
-
"loss": 38.
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.9,
|
| 18 |
"learning_rate": 7.35e-06,
|
| 19 |
-
"loss": 24.
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 1.35,
|
| 24 |
"learning_rate": 1.1099999999999999e-05,
|
| 25 |
-
"loss": 15.
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 1.8,
|
| 30 |
"learning_rate": 1.485e-05,
|
| 31 |
-
"loss": 12.
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 2.25,
|
| 36 |
"learning_rate": 1.8599999999999998e-05,
|
| 37 |
-
"loss": 9.
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 2.7,
|
| 42 |
"learning_rate": 2.2349999999999998e-05,
|
| 43 |
-
"loss": 7.
|
| 44 |
"step": 600
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 3.15,
|
| 48 |
"learning_rate": 2.6099999999999997e-05,
|
| 49 |
-
"loss": 5.
|
| 50 |
"step": 700
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 3.6,
|
| 54 |
"learning_rate": 2.985e-05,
|
| 55 |
-
"loss": 4.
|
| 56 |
"step": 800
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 4.05,
|
| 60 |
"learning_rate": 3.36e-05,
|
| 61 |
-
"loss": 4.
|
| 62 |
"step": 900
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 4.5,
|
| 66 |
"learning_rate": 3.735e-05,
|
| 67 |
-
"loss": 4.
|
| 68 |
"step": 1000
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 4.5,
|
| 72 |
-
"eval_loss": 4.
|
| 73 |
-
"eval_runtime":
|
| 74 |
-
"eval_samples_per_second": 18.
|
| 75 |
-
"eval_steps_per_second": 2.
|
| 76 |
-
"eval_wer": 2.
|
| 77 |
"step": 1000
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 4.95,
|
| 81 |
"learning_rate": 4.11e-05,
|
| 82 |
-
"loss": 4.
|
| 83 |
"step": 1100
|
| 84 |
},
|
| 85 |
{
|
| 86 |
"epoch": 5.41,
|
| 87 |
"learning_rate": 4.484999999999999e-05,
|
| 88 |
-
"loss": 4.
|
| 89 |
"step": 1200
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 5.86,
|
| 93 |
"learning_rate": 4.8599999999999995e-05,
|
| 94 |
-
"loss": 4.
|
| 95 |
"step": 1300
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 6.31,
|
| 99 |
"learning_rate": 5.234999999999999e-05,
|
| 100 |
-
"loss": 4.
|
| 101 |
"step": 1400
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 6.76,
|
| 105 |
"learning_rate": 5.6099999999999995e-05,
|
| 106 |
-
"loss": 3.
|
| 107 |
"step": 1500
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 7.21,
|
| 111 |
"learning_rate": 5.985e-05,
|
| 112 |
-
"loss": 3.
|
| 113 |
"step": 1600
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 7.66,
|
| 117 |
"learning_rate": 6.359999999999999e-05,
|
| 118 |
-
"loss": 3.
|
| 119 |
"step": 1700
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 8.11,
|
| 123 |
"learning_rate": 6.735e-05,
|
| 124 |
-
"loss": 2.
|
| 125 |
"step": 1800
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 8.56,
|
| 129 |
"learning_rate": 7.11e-05,
|
| 130 |
-
"loss": 2.
|
| 131 |
"step": 1900
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 9.01,
|
| 135 |
"learning_rate": 7.484999999999999e-05,
|
| 136 |
-
"loss": 2.
|
| 137 |
"step": 2000
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 9.01,
|
| 141 |
-
"eval_loss": 1.
|
| 142 |
-
"eval_runtime": 237.
|
| 143 |
-
"eval_samples_per_second": 18.
|
| 144 |
-
"eval_steps_per_second": 2.
|
| 145 |
-
"eval_wer": 2.
|
| 146 |
"step": 2000
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 9.46,
|
| 150 |
-
"learning_rate":
|
| 151 |
-
"loss": 2.
|
| 152 |
"step": 2100
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 9.91,
|
| 156 |
-
"learning_rate":
|
| 157 |
-
"loss": 2.
|
| 158 |
"step": 2200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch": 10.
|
| 162 |
-
"
|
| 163 |
-
"
|
| 164 |
-
"
|
| 165 |
-
|
| 166 |
-
|
| 167 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 168 |
}
|
| 169 |
],
|
| 170 |
-
"max_steps":
|
| 171 |
-
"num_train_epochs":
|
| 172 |
-
"total_flos":
|
| 173 |
"trial_name": null,
|
| 174 |
"trial_params": null
|
| 175 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 50.0,
|
| 5 |
+
"global_step": 11100,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 10 |
{
|
| 11 |
"epoch": 0.45,
|
| 12 |
"learning_rate": 3.6e-06,
|
| 13 |
+
"loss": 38.551,
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.9,
|
| 18 |
"learning_rate": 7.35e-06,
|
| 19 |
+
"loss": 24.0076,
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 1.35,
|
| 24 |
"learning_rate": 1.1099999999999999e-05,
|
| 25 |
+
"loss": 15.8812,
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 1.8,
|
| 30 |
"learning_rate": 1.485e-05,
|
| 31 |
+
"loss": 12.8964,
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 2.25,
|
| 36 |
"learning_rate": 1.8599999999999998e-05,
|
| 37 |
+
"loss": 9.9344,
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 2.7,
|
| 42 |
"learning_rate": 2.2349999999999998e-05,
|
| 43 |
+
"loss": 7.1613,
|
| 44 |
"step": 600
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 3.15,
|
| 48 |
"learning_rate": 2.6099999999999997e-05,
|
| 49 |
+
"loss": 5.3772,
|
| 50 |
"step": 700
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 3.6,
|
| 54 |
"learning_rate": 2.985e-05,
|
| 55 |
+
"loss": 4.6404,
|
| 56 |
"step": 800
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 4.05,
|
| 60 |
"learning_rate": 3.36e-05,
|
| 61 |
+
"loss": 4.3831,
|
| 62 |
"step": 900
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 4.5,
|
| 66 |
"learning_rate": 3.735e-05,
|
| 67 |
+
"loss": 4.221,
|
| 68 |
"step": 1000
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 4.5,
|
| 72 |
+
"eval_loss": 4.1194539070129395,
|
| 73 |
+
"eval_runtime": 239.8438,
|
| 74 |
+
"eval_samples_per_second": 18.691,
|
| 75 |
+
"eval_steps_per_second": 2.339,
|
| 76 |
+
"eval_wer": 2.4023576512455516,
|
| 77 |
"step": 1000
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 4.95,
|
| 81 |
"learning_rate": 4.11e-05,
|
| 82 |
+
"loss": 4.1513,
|
| 83 |
"step": 1100
|
| 84 |
},
|
| 85 |
{
|
| 86 |
"epoch": 5.41,
|
| 87 |
"learning_rate": 4.484999999999999e-05,
|
| 88 |
+
"loss": 4.1055,
|
| 89 |
"step": 1200
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 5.86,
|
| 93 |
"learning_rate": 4.8599999999999995e-05,
|
| 94 |
+
"loss": 4.0677,
|
| 95 |
"step": 1300
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 6.31,
|
| 99 |
"learning_rate": 5.234999999999999e-05,
|
| 100 |
+
"loss": 4.0164,
|
| 101 |
"step": 1400
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 6.76,
|
| 105 |
"learning_rate": 5.6099999999999995e-05,
|
| 106 |
+
"loss": 3.9636,
|
| 107 |
"step": 1500
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 7.21,
|
| 111 |
"learning_rate": 5.985e-05,
|
| 112 |
+
"loss": 3.7377,
|
| 113 |
"step": 1600
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 7.66,
|
| 117 |
"learning_rate": 6.359999999999999e-05,
|
| 118 |
+
"loss": 3.1744,
|
| 119 |
"step": 1700
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 8.11,
|
| 123 |
"learning_rate": 6.735e-05,
|
| 124 |
+
"loss": 2.704,
|
| 125 |
"step": 1800
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 8.56,
|
| 129 |
"learning_rate": 7.11e-05,
|
| 130 |
+
"loss": 2.497,
|
| 131 |
"step": 1900
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 9.01,
|
| 135 |
"learning_rate": 7.484999999999999e-05,
|
| 136 |
+
"loss": 2.3597,
|
| 137 |
"step": 2000
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 9.01,
|
| 141 |
+
"eval_loss": 1.102413535118103,
|
| 142 |
+
"eval_runtime": 237.4335,
|
| 143 |
+
"eval_samples_per_second": 18.881,
|
| 144 |
+
"eval_steps_per_second": 2.363,
|
| 145 |
+
"eval_wer": 2.761788256227758,
|
| 146 |
"step": 2000
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 9.46,
|
| 150 |
+
"learning_rate": 7.42087912087912e-05,
|
| 151 |
+
"loss": 2.2495,
|
| 152 |
"step": 2100
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 9.91,
|
| 156 |
+
"learning_rate": 7.338461538461538e-05,
|
| 157 |
+
"loss": 2.1767,
|
| 158 |
"step": 2200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 10.36,
|
| 162 |
+
"learning_rate": 7.256043956043956e-05,
|
| 163 |
+
"loss": 2.1115,
|
| 164 |
+
"step": 2300
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 10.81,
|
| 168 |
+
"learning_rate": 7.173626373626374e-05,
|
| 169 |
+
"loss": 2.0653,
|
| 170 |
+
"step": 2400
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 11.26,
|
| 174 |
+
"learning_rate": 7.09120879120879e-05,
|
| 175 |
+
"loss": 2.0077,
|
| 176 |
+
"step": 2500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 11.71,
|
| 180 |
+
"learning_rate": 7.008791208791208e-05,
|
| 181 |
+
"loss": 1.9933,
|
| 182 |
+
"step": 2600
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 12.16,
|
| 186 |
+
"learning_rate": 6.926373626373626e-05,
|
| 187 |
+
"loss": 1.9756,
|
| 188 |
+
"step": 2700
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 12.61,
|
| 192 |
+
"learning_rate": 6.843956043956044e-05,
|
| 193 |
+
"loss": 1.9246,
|
| 194 |
+
"step": 2800
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 13.06,
|
| 198 |
+
"learning_rate": 6.761538461538461e-05,
|
| 199 |
+
"loss": 1.9044,
|
| 200 |
+
"step": 2900
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 13.51,
|
| 204 |
+
"learning_rate": 6.679120879120879e-05,
|
| 205 |
+
"loss": 1.8795,
|
| 206 |
+
"step": 3000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 13.51,
|
| 210 |
+
"eval_loss": 0.749809205532074,
|
| 211 |
+
"eval_runtime": 236.7828,
|
| 212 |
+
"eval_samples_per_second": 18.933,
|
| 213 |
+
"eval_steps_per_second": 2.369,
|
| 214 |
+
"eval_wer": 2.588523131672598,
|
| 215 |
+
"step": 3000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 13.96,
|
| 219 |
+
"learning_rate": 6.596703296703297e-05,
|
| 220 |
+
"loss": 1.8683,
|
| 221 |
+
"step": 3100
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 14.41,
|
| 225 |
+
"learning_rate": 6.514285714285713e-05,
|
| 226 |
+
"loss": 1.8367,
|
| 227 |
+
"step": 3200
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 14.86,
|
| 231 |
+
"learning_rate": 6.431868131868131e-05,
|
| 232 |
+
"loss": 1.8273,
|
| 233 |
+
"step": 3300
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 15.32,
|
| 237 |
+
"learning_rate": 6.349450549450548e-05,
|
| 238 |
+
"loss": 1.7743,
|
| 239 |
+
"step": 3400
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 15.77,
|
| 243 |
+
"learning_rate": 6.267032967032966e-05,
|
| 244 |
+
"loss": 1.7832,
|
| 245 |
+
"step": 3500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 16.22,
|
| 249 |
+
"learning_rate": 6.184615384615384e-05,
|
| 250 |
+
"loss": 1.7915,
|
| 251 |
+
"step": 3600
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 16.67,
|
| 255 |
+
"learning_rate": 6.102197802197801e-05,
|
| 256 |
+
"loss": 1.7488,
|
| 257 |
+
"step": 3700
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 17.12,
|
| 261 |
+
"learning_rate": 6.019780219780219e-05,
|
| 262 |
+
"loss": 1.7178,
|
| 263 |
+
"step": 3800
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 17.57,
|
| 267 |
+
"learning_rate": 5.937362637362637e-05,
|
| 268 |
+
"loss": 1.6903,
|
| 269 |
+
"step": 3900
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 18.02,
|
| 273 |
+
"learning_rate": 5.854945054945054e-05,
|
| 274 |
+
"loss": 1.7143,
|
| 275 |
+
"step": 4000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 18.02,
|
| 279 |
+
"eval_loss": 0.6538747549057007,
|
| 280 |
+
"eval_runtime": 236.0252,
|
| 281 |
+
"eval_samples_per_second": 18.994,
|
| 282 |
+
"eval_steps_per_second": 2.377,
|
| 283 |
+
"eval_wer": 2.5976423487544484,
|
| 284 |
+
"step": 4000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 18.47,
|
| 288 |
+
"learning_rate": 5.772527472527472e-05,
|
| 289 |
+
"loss": 1.7023,
|
| 290 |
+
"step": 4100
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 18.92,
|
| 294 |
+
"learning_rate": 5.69010989010989e-05,
|
| 295 |
+
"loss": 1.6887,
|
| 296 |
+
"step": 4200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 19.37,
|
| 300 |
+
"learning_rate": 5.607692307692307e-05,
|
| 301 |
+
"loss": 1.6596,
|
| 302 |
+
"step": 4300
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 19.82,
|
| 306 |
+
"learning_rate": 5.525274725274725e-05,
|
| 307 |
+
"loss": 1.6613,
|
| 308 |
+
"step": 4400
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 20.27,
|
| 312 |
+
"learning_rate": 5.442857142857143e-05,
|
| 313 |
+
"loss": 1.6575,
|
| 314 |
+
"step": 4500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 20.72,
|
| 318 |
+
"learning_rate": 5.36043956043956e-05,
|
| 319 |
+
"loss": 1.6486,
|
| 320 |
+
"step": 4600
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 21.17,
|
| 324 |
+
"learning_rate": 5.278021978021977e-05,
|
| 325 |
+
"loss": 1.6467,
|
| 326 |
+
"step": 4700
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 21.62,
|
| 330 |
+
"learning_rate": 5.195604395604395e-05,
|
| 331 |
+
"loss": 1.6206,
|
| 332 |
+
"step": 4800
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 22.07,
|
| 336 |
+
"learning_rate": 5.1140109890109885e-05,
|
| 337 |
+
"loss": 1.6048,
|
| 338 |
+
"step": 4900
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 22.52,
|
| 342 |
+
"learning_rate": 5.031593406593406e-05,
|
| 343 |
+
"loss": 1.6025,
|
| 344 |
+
"step": 5000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 22.52,
|
| 348 |
+
"eval_loss": 0.598939836025238,
|
| 349 |
+
"eval_runtime": 237.9623,
|
| 350 |
+
"eval_samples_per_second": 18.839,
|
| 351 |
+
"eval_steps_per_second": 2.358,
|
| 352 |
+
"eval_wer": 2.6034252669039146,
|
| 353 |
+
"step": 5000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 22.97,
|
| 357 |
+
"learning_rate": 4.949175824175824e-05,
|
| 358 |
+
"loss": 1.6151,
|
| 359 |
+
"step": 5100
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 23.42,
|
| 363 |
+
"learning_rate": 4.866758241758242e-05,
|
| 364 |
+
"loss": 1.5959,
|
| 365 |
+
"step": 5200
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 23.87,
|
| 369 |
+
"learning_rate": 4.784340659340658e-05,
|
| 370 |
+
"loss": 1.5845,
|
| 371 |
+
"step": 5300
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 24.32,
|
| 375 |
+
"learning_rate": 4.701923076923076e-05,
|
| 376 |
+
"loss": 1.5662,
|
| 377 |
+
"step": 5400
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 24.77,
|
| 381 |
+
"learning_rate": 4.619505494505494e-05,
|
| 382 |
+
"loss": 1.5759,
|
| 383 |
+
"step": 5500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 25.23,
|
| 387 |
+
"learning_rate": 4.5370879120879115e-05,
|
| 388 |
+
"loss": 1.5928,
|
| 389 |
+
"step": 5600
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 25.68,
|
| 393 |
+
"learning_rate": 4.4546703296703295e-05,
|
| 394 |
+
"loss": 1.5576,
|
| 395 |
+
"step": 5700
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 26.13,
|
| 399 |
+
"learning_rate": 4.3722527472527474e-05,
|
| 400 |
+
"loss": 1.5289,
|
| 401 |
+
"step": 5800
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 26.58,
|
| 405 |
+
"learning_rate": 4.289835164835164e-05,
|
| 406 |
+
"loss": 1.5381,
|
| 407 |
+
"step": 5900
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 27.03,
|
| 411 |
+
"learning_rate": 4.207417582417582e-05,
|
| 412 |
+
"loss": 1.5403,
|
| 413 |
+
"step": 6000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 27.03,
|
| 417 |
+
"eval_loss": 0.6035020351409912,
|
| 418 |
+
"eval_runtime": 237.8256,
|
| 419 |
+
"eval_samples_per_second": 18.85,
|
| 420 |
+
"eval_steps_per_second": 2.359,
|
| 421 |
+
"eval_wer": 2.69461743772242,
|
| 422 |
+
"step": 6000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 27.48,
|
| 426 |
+
"learning_rate": 4.125e-05,
|
| 427 |
+
"loss": 1.5314,
|
| 428 |
+
"step": 6100
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 27.93,
|
| 432 |
+
"learning_rate": 4.042582417582417e-05,
|
| 433 |
+
"loss": 1.5186,
|
| 434 |
+
"step": 6200
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 28.38,
|
| 438 |
+
"learning_rate": 3.960164835164835e-05,
|
| 439 |
+
"loss": 1.5097,
|
| 440 |
+
"step": 6300
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 28.83,
|
| 444 |
+
"learning_rate": 3.8777472527472525e-05,
|
| 445 |
+
"loss": 1.52,
|
| 446 |
+
"step": 6400
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 29.28,
|
| 450 |
+
"learning_rate": 3.79532967032967e-05,
|
| 451 |
+
"loss": 1.51,
|
| 452 |
+
"step": 6500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 29.73,
|
| 456 |
+
"learning_rate": 3.712912087912088e-05,
|
| 457 |
+
"loss": 1.5109,
|
| 458 |
+
"step": 6600
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 30.18,
|
| 462 |
+
"learning_rate": 3.630494505494505e-05,
|
| 463 |
+
"loss": 1.5219,
|
| 464 |
+
"step": 6700
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 30.63,
|
| 468 |
+
"learning_rate": 3.548076923076922e-05,
|
| 469 |
+
"loss": 1.5006,
|
| 470 |
+
"step": 6800
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 31.08,
|
| 474 |
+
"learning_rate": 3.46565934065934e-05,
|
| 475 |
+
"loss": 1.475,
|
| 476 |
+
"step": 6900
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 31.53,
|
| 480 |
+
"learning_rate": 3.383241758241758e-05,
|
| 481 |
+
"loss": 1.4773,
|
| 482 |
+
"step": 7000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 31.53,
|
| 486 |
+
"eval_loss": 0.5646877288818359,
|
| 487 |
+
"eval_runtime": 239.6564,
|
| 488 |
+
"eval_samples_per_second": 18.706,
|
| 489 |
+
"eval_steps_per_second": 2.341,
|
| 490 |
+
"eval_wer": 2.555827402135231,
|
| 491 |
+
"step": 7000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 31.98,
|
| 495 |
+
"learning_rate": 3.3008241758241755e-05,
|
| 496 |
+
"loss": 1.4921,
|
| 497 |
+
"step": 7100
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 32.43,
|
| 501 |
+
"learning_rate": 3.218406593406593e-05,
|
| 502 |
+
"loss": 1.4634,
|
| 503 |
+
"step": 7200
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 32.88,
|
| 507 |
+
"learning_rate": 3.135989010989011e-05,
|
| 508 |
+
"loss": 1.4716,
|
| 509 |
+
"step": 7300
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 33.33,
|
| 513 |
+
"learning_rate": 3.054395604395604e-05,
|
| 514 |
+
"loss": 1.4313,
|
| 515 |
+
"step": 7400
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 33.78,
|
| 519 |
+
"learning_rate": 2.971978021978022e-05,
|
| 520 |
+
"loss": 1.4702,
|
| 521 |
+
"step": 7500
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 34.23,
|
| 525 |
+
"learning_rate": 2.8895604395604393e-05,
|
| 526 |
+
"loss": 1.4679,
|
| 527 |
+
"step": 7600
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 34.68,
|
| 531 |
+
"learning_rate": 2.807142857142857e-05,
|
| 532 |
+
"loss": 1.4379,
|
| 533 |
+
"step": 7700
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 35.14,
|
| 537 |
+
"learning_rate": 2.7247252747252742e-05,
|
| 538 |
+
"loss": 1.4211,
|
| 539 |
+
"step": 7800
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 35.59,
|
| 543 |
+
"learning_rate": 2.642307692307692e-05,
|
| 544 |
+
"loss": 1.4295,
|
| 545 |
+
"step": 7900
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 36.04,
|
| 549 |
+
"learning_rate": 2.5598901098901098e-05,
|
| 550 |
+
"loss": 1.4228,
|
| 551 |
+
"step": 8000
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 36.04,
|
| 555 |
+
"eval_loss": 0.5477184057235718,
|
| 556 |
+
"eval_runtime": 235.9806,
|
| 557 |
+
"eval_samples_per_second": 18.997,
|
| 558 |
+
"eval_steps_per_second": 2.377,
|
| 559 |
+
"eval_wer": 2.5676156583629894,
|
| 560 |
+
"step": 8000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 36.49,
|
| 564 |
+
"learning_rate": 2.477472527472527e-05,
|
| 565 |
+
"loss": 1.4199,
|
| 566 |
+
"step": 8100
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 36.94,
|
| 570 |
+
"learning_rate": 2.395054945054945e-05,
|
| 571 |
+
"loss": 1.4317,
|
| 572 |
+
"step": 8200
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 37.39,
|
| 576 |
+
"learning_rate": 2.3126373626373626e-05,
|
| 577 |
+
"loss": 1.4048,
|
| 578 |
+
"step": 8300
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 37.84,
|
| 582 |
+
"learning_rate": 2.23021978021978e-05,
|
| 583 |
+
"loss": 1.4232,
|
| 584 |
+
"step": 8400
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 38.29,
|
| 588 |
+
"learning_rate": 2.147802197802198e-05,
|
| 589 |
+
"loss": 1.4076,
|
| 590 |
+
"step": 8500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 38.74,
|
| 594 |
+
"learning_rate": 2.065384615384615e-05,
|
| 595 |
+
"loss": 1.4009,
|
| 596 |
+
"step": 8600
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 39.19,
|
| 600 |
+
"learning_rate": 1.9829670329670328e-05,
|
| 601 |
+
"loss": 1.4173,
|
| 602 |
+
"step": 8700
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 39.64,
|
| 606 |
+
"learning_rate": 1.90054945054945e-05,
|
| 607 |
+
"loss": 1.4028,
|
| 608 |
+
"step": 8800
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 40.09,
|
| 612 |
+
"learning_rate": 1.818131868131868e-05,
|
| 613 |
+
"loss": 1.3694,
|
| 614 |
+
"step": 8900
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 40.54,
|
| 618 |
+
"learning_rate": 1.7357142857142856e-05,
|
| 619 |
+
"loss": 1.3801,
|
| 620 |
+
"step": 9000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 40.54,
|
| 624 |
+
"eval_loss": 0.5413390398025513,
|
| 625 |
+
"eval_runtime": 239.6653,
|
| 626 |
+
"eval_samples_per_second": 18.705,
|
| 627 |
+
"eval_steps_per_second": 2.341,
|
| 628 |
+
"eval_wer": 2.619217081850534,
|
| 629 |
+
"step": 9000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 40.99,
|
| 633 |
+
"learning_rate": 1.6532967032967033e-05,
|
| 634 |
+
"loss": 1.4054,
|
| 635 |
+
"step": 9100
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 41.44,
|
| 639 |
+
"learning_rate": 1.5708791208791206e-05,
|
| 640 |
+
"loss": 1.3823,
|
| 641 |
+
"step": 9200
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 41.89,
|
| 645 |
+
"learning_rate": 1.4884615384615383e-05,
|
| 646 |
+
"loss": 1.3902,
|
| 647 |
+
"step": 9300
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 42.34,
|
| 651 |
+
"learning_rate": 1.4060439560439558e-05,
|
| 652 |
+
"loss": 1.3568,
|
| 653 |
+
"step": 9400
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 42.79,
|
| 657 |
+
"learning_rate": 1.3236263736263736e-05,
|
| 658 |
+
"loss": 1.3766,
|
| 659 |
+
"step": 9500
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 43.24,
|
| 663 |
+
"learning_rate": 1.2412087912087912e-05,
|
| 664 |
+
"loss": 1.3883,
|
| 665 |
+
"step": 9600
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 43.69,
|
| 669 |
+
"learning_rate": 1.1587912087912087e-05,
|
| 670 |
+
"loss": 1.3666,
|
| 671 |
+
"step": 9700
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 44.14,
|
| 675 |
+
"learning_rate": 1.0763736263736263e-05,
|
| 676 |
+
"loss": 1.3718,
|
| 677 |
+
"step": 9800
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 44.59,
|
| 681 |
+
"learning_rate": 9.939560439560437e-06,
|
| 682 |
+
"loss": 1.367,
|
| 683 |
+
"step": 9900
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 45.05,
|
| 687 |
+
"learning_rate": 9.115384615384615e-06,
|
| 688 |
+
"loss": 1.3558,
|
| 689 |
+
"step": 10000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 45.05,
|
| 693 |
+
"eval_loss": 0.5343097448348999,
|
| 694 |
+
"eval_runtime": 238.241,
|
| 695 |
+
"eval_samples_per_second": 18.817,
|
| 696 |
+
"eval_steps_per_second": 2.355,
|
| 697 |
+
"eval_wer": 2.657473309608541,
|
| 698 |
+
"step": 10000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 45.5,
|
| 702 |
+
"learning_rate": 8.29120879120879e-06,
|
| 703 |
+
"loss": 1.3595,
|
| 704 |
+
"step": 10100
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 45.95,
|
| 708 |
+
"learning_rate": 7.467032967032966e-06,
|
| 709 |
+
"loss": 1.3697,
|
| 710 |
+
"step": 10200
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 46.4,
|
| 714 |
+
"learning_rate": 6.6428571428571416e-06,
|
| 715 |
+
"loss": 1.3423,
|
| 716 |
+
"step": 10300
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 46.85,
|
| 720 |
+
"learning_rate": 5.818681318681319e-06,
|
| 721 |
+
"loss": 1.341,
|
| 722 |
+
"step": 10400
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 47.3,
|
| 726 |
+
"learning_rate": 4.994505494505494e-06,
|
| 727 |
+
"loss": 1.3289,
|
| 728 |
+
"step": 10500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 47.75,
|
| 732 |
+
"learning_rate": 4.17032967032967e-06,
|
| 733 |
+
"loss": 1.3462,
|
| 734 |
+
"step": 10600
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 48.2,
|
| 738 |
+
"learning_rate": 3.3461538461538457e-06,
|
| 739 |
+
"loss": 1.3636,
|
| 740 |
+
"step": 10700
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 48.65,
|
| 744 |
+
"learning_rate": 2.521978021978022e-06,
|
| 745 |
+
"loss": 1.3366,
|
| 746 |
+
"step": 10800
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 49.1,
|
| 750 |
+
"learning_rate": 1.6978021978021975e-06,
|
| 751 |
+
"loss": 1.337,
|
| 752 |
+
"step": 10900
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 49.55,
|
| 756 |
+
"learning_rate": 8.818681318681318e-07,
|
| 757 |
+
"loss": 1.3298,
|
| 758 |
+
"step": 11000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 49.55,
|
| 762 |
+
"eval_loss": 0.5348710417747498,
|
| 763 |
+
"eval_runtime": 238.2007,
|
| 764 |
+
"eval_samples_per_second": 18.82,
|
| 765 |
+
"eval_steps_per_second": 2.355,
|
| 766 |
+
"eval_wer": 2.627446619217082,
|
| 767 |
+
"step": 11000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 50.0,
|
| 771 |
+
"learning_rate": 5.769230769230769e-08,
|
| 772 |
+
"loss": 1.3663,
|
| 773 |
+
"step": 11100
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 50.0,
|
| 777 |
+
"step": 11100,
|
| 778 |
+
"total_flos": 8.246726177897223e+19,
|
| 779 |
+
"train_loss": 2.741275282851211,
|
| 780 |
+
"train_runtime": 31325.7849,
|
| 781 |
+
"train_samples_per_second": 16.956,
|
| 782 |
+
"train_steps_per_second": 0.354
|
| 783 |
}
|
| 784 |
],
|
| 785 |
+
"max_steps": 11100,
|
| 786 |
+
"num_train_epochs": 50,
|
| 787 |
+
"total_flos": 8.246726177897223e+19,
|
| 788 |
"trial_name": null,
|
| 789 |
"trial_params": null
|
| 790 |
}
|