Commit
•
57ff16f
1
Parent(s):
451f2a6
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +577 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.20929841697216034,
|
| 4 |
+
"eval_runtime": 43.7778,
|
| 5 |
+
"eval_samples": 1184,
|
| 6 |
+
"eval_samples_per_second": 27.046,
|
| 7 |
+
"eval_steps_per_second": 1.69,
|
| 8 |
+
"eval_wer": 0.190736096769435,
|
| 9 |
+
"train_loss": 1.1495286922454835,
|
| 10 |
+
"train_runtime": 11669.3128,
|
| 11 |
+
"train_samples": 2558,
|
| 12 |
+
"train_samples_per_second": 21.921,
|
| 13 |
+
"train_steps_per_second": 0.686
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.20929841697216034,
|
| 4 |
+
"eval_runtime": 43.7778,
|
| 5 |
+
"eval_samples": 1184,
|
| 6 |
+
"eval_samples_per_second": 27.046,
|
| 7 |
+
"eval_steps_per_second": 1.69,
|
| 8 |
+
"eval_wer": 0.190736096769435
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"train_loss": 1.1495286922454835,
|
| 4 |
+
"train_runtime": 11669.3128,
|
| 5 |
+
"train_samples": 2558,
|
| 6 |
+
"train_samples_per_second": 21.921,
|
| 7 |
+
"train_steps_per_second": 0.686
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,577 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"global_step": 8000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.25,
|
| 12 |
+
"learning_rate": 6.789999999999999e-06,
|
| 13 |
+
"loss": 15.499,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 2.5,
|
| 18 |
+
"learning_rate": 1.379e-05,
|
| 19 |
+
"loss": 5.6988,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 3.75,
|
| 24 |
+
"learning_rate": 2.0789999999999996e-05,
|
| 25 |
+
"loss": 3.9135,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 5.0,
|
| 30 |
+
"learning_rate": 2.779e-05,
|
| 31 |
+
"loss": 3.315,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 6.25,
|
| 36 |
+
"learning_rate": 3.479e-05,
|
| 37 |
+
"loss": 3.0713,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 7.5,
|
| 42 |
+
"learning_rate": 4.178999999999999e-05,
|
| 43 |
+
"loss": 2.9727,
|
| 44 |
+
"step": 600
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 8.75,
|
| 48 |
+
"learning_rate": 4.878999999999999e-05,
|
| 49 |
+
"loss": 2.8912,
|
| 50 |
+
"step": 700
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 10.0,
|
| 54 |
+
"learning_rate": 5.579e-05,
|
| 55 |
+
"loss": 2.8514,
|
| 56 |
+
"step": 800
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 11.25,
|
| 60 |
+
"learning_rate": 6.279e-05,
|
| 61 |
+
"loss": 2.5765,
|
| 62 |
+
"step": 900
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 12.5,
|
| 66 |
+
"learning_rate": 6.979e-05,
|
| 67 |
+
"loss": 1.785,
|
| 68 |
+
"step": 1000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 12.5,
|
| 72 |
+
"eval_loss": 0.7464718222618103,
|
| 73 |
+
"eval_runtime": 42.2657,
|
| 74 |
+
"eval_samples_per_second": 28.013,
|
| 75 |
+
"eval_steps_per_second": 1.751,
|
| 76 |
+
"eval_wer": 0.6812214190883611,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 13.75,
|
| 81 |
+
"learning_rate": 6.903e-05,
|
| 82 |
+
"loss": 1.4773,
|
| 83 |
+
"step": 1100
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 15.0,
|
| 87 |
+
"learning_rate": 6.803e-05,
|
| 88 |
+
"loss": 1.3507,
|
| 89 |
+
"step": 1200
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 16.25,
|
| 93 |
+
"learning_rate": 6.702999999999999e-05,
|
| 94 |
+
"loss": 1.2818,
|
| 95 |
+
"step": 1300
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 17.5,
|
| 99 |
+
"learning_rate": 6.602999999999999e-05,
|
| 100 |
+
"loss": 1.233,
|
| 101 |
+
"step": 1400
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 18.75,
|
| 105 |
+
"learning_rate": 6.502999999999999e-05,
|
| 106 |
+
"loss": 1.161,
|
| 107 |
+
"step": 1500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 20.0,
|
| 111 |
+
"learning_rate": 6.403e-05,
|
| 112 |
+
"loss": 1.1004,
|
| 113 |
+
"step": 1600
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 21.25,
|
| 117 |
+
"learning_rate": 6.303e-05,
|
| 118 |
+
"loss": 1.0246,
|
| 119 |
+
"step": 1700
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 22.5,
|
| 123 |
+
"learning_rate": 6.202999999999999e-05,
|
| 124 |
+
"loss": 0.9693,
|
| 125 |
+
"step": 1800
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 23.75,
|
| 129 |
+
"learning_rate": 6.103e-05,
|
| 130 |
+
"loss": 0.9507,
|
| 131 |
+
"step": 1900
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 25.0,
|
| 135 |
+
"learning_rate": 6.002999999999999e-05,
|
| 136 |
+
"loss": 0.8989,
|
| 137 |
+
"step": 2000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 25.0,
|
| 141 |
+
"eval_loss": 0.24948318302631378,
|
| 142 |
+
"eval_runtime": 40.7976,
|
| 143 |
+
"eval_samples_per_second": 29.021,
|
| 144 |
+
"eval_steps_per_second": 1.814,
|
| 145 |
+
"eval_wer": 0.27319663667207555,
|
| 146 |
+
"step": 2000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 26.25,
|
| 150 |
+
"learning_rate": 5.904e-05,
|
| 151 |
+
"loss": 0.8743,
|
| 152 |
+
"step": 2100
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 27.5,
|
| 156 |
+
"learning_rate": 5.804e-05,
|
| 157 |
+
"loss": 0.8558,
|
| 158 |
+
"step": 2200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 28.75,
|
| 162 |
+
"learning_rate": 5.7039999999999996e-05,
|
| 163 |
+
"loss": 0.8228,
|
| 164 |
+
"step": 2300
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 30.0,
|
| 168 |
+
"learning_rate": 5.604e-05,
|
| 169 |
+
"loss": 0.8154,
|
| 170 |
+
"step": 2400
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 31.25,
|
| 174 |
+
"learning_rate": 5.5039999999999995e-05,
|
| 175 |
+
"loss": 0.7932,
|
| 176 |
+
"step": 2500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 32.5,
|
| 180 |
+
"learning_rate": 5.404e-05,
|
| 181 |
+
"loss": 0.7755,
|
| 182 |
+
"step": 2600
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 33.75,
|
| 186 |
+
"learning_rate": 5.3039999999999994e-05,
|
| 187 |
+
"loss": 0.7585,
|
| 188 |
+
"step": 2700
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 35.0,
|
| 192 |
+
"learning_rate": 5.2039999999999996e-05,
|
| 193 |
+
"loss": 0.7472,
|
| 194 |
+
"step": 2800
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 36.25,
|
| 198 |
+
"learning_rate": 5.103999999999999e-05,
|
| 199 |
+
"loss": 0.734,
|
| 200 |
+
"step": 2900
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 37.5,
|
| 204 |
+
"learning_rate": 5.0039999999999995e-05,
|
| 205 |
+
"loss": 0.7118,
|
| 206 |
+
"step": 3000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 37.5,
|
| 210 |
+
"eval_loss": 0.21255508065223694,
|
| 211 |
+
"eval_runtime": 40.7166,
|
| 212 |
+
"eval_samples_per_second": 29.079,
|
| 213 |
+
"eval_steps_per_second": 1.817,
|
| 214 |
+
"eval_wer": 0.22835226434577371,
|
| 215 |
+
"step": 3000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 38.75,
|
| 219 |
+
"learning_rate": 4.904e-05,
|
| 220 |
+
"loss": 0.7163,
|
| 221 |
+
"step": 3100
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 40.0,
|
| 225 |
+
"learning_rate": 4.8039999999999994e-05,
|
| 226 |
+
"loss": 0.7066,
|
| 227 |
+
"step": 3200
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 41.25,
|
| 231 |
+
"learning_rate": 4.704e-05,
|
| 232 |
+
"loss": 0.7067,
|
| 233 |
+
"step": 3300
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 42.5,
|
| 237 |
+
"learning_rate": 4.603999999999999e-05,
|
| 238 |
+
"loss": 0.688,
|
| 239 |
+
"step": 3400
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 43.75,
|
| 243 |
+
"learning_rate": 4.5039999999999996e-05,
|
| 244 |
+
"loss": 0.6777,
|
| 245 |
+
"step": 3500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 45.0,
|
| 249 |
+
"learning_rate": 4.403999999999999e-05,
|
| 250 |
+
"loss": 0.6621,
|
| 251 |
+
"step": 3600
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 46.25,
|
| 255 |
+
"learning_rate": 4.3039999999999994e-05,
|
| 256 |
+
"loss": 0.6494,
|
| 257 |
+
"step": 3700
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 47.5,
|
| 261 |
+
"learning_rate": 4.203999999999999e-05,
|
| 262 |
+
"loss": 0.646,
|
| 263 |
+
"step": 3800
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 48.75,
|
| 267 |
+
"learning_rate": 4.104e-05,
|
| 268 |
+
"loss": 0.6356,
|
| 269 |
+
"step": 3900
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 50.0,
|
| 273 |
+
"learning_rate": 4.0039999999999996e-05,
|
| 274 |
+
"loss": 0.6367,
|
| 275 |
+
"step": 4000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 50.0,
|
| 279 |
+
"eval_loss": 0.20487311482429504,
|
| 280 |
+
"eval_runtime": 43.4458,
|
| 281 |
+
"eval_samples_per_second": 27.252,
|
| 282 |
+
"eval_steps_per_second": 1.703,
|
| 283 |
+
"eval_wer": 0.20489747750405665,
|
| 284 |
+
"step": 4000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 51.25,
|
| 288 |
+
"learning_rate": 3.904e-05,
|
| 289 |
+
"loss": 0.6272,
|
| 290 |
+
"step": 4100
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 52.5,
|
| 294 |
+
"learning_rate": 3.804e-05,
|
| 295 |
+
"loss": 0.6187,
|
| 296 |
+
"step": 4200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 53.75,
|
| 300 |
+
"learning_rate": 3.704e-05,
|
| 301 |
+
"loss": 0.6116,
|
| 302 |
+
"step": 4300
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 55.0,
|
| 306 |
+
"learning_rate": 3.604e-05,
|
| 307 |
+
"loss": 0.6074,
|
| 308 |
+
"step": 4400
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 56.25,
|
| 312 |
+
"learning_rate": 3.504999999999999e-05,
|
| 313 |
+
"loss": 0.5945,
|
| 314 |
+
"step": 4500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 57.5,
|
| 318 |
+
"learning_rate": 3.4049999999999994e-05,
|
| 319 |
+
"loss": 0.6074,
|
| 320 |
+
"step": 4600
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 58.75,
|
| 324 |
+
"learning_rate": 3.305e-05,
|
| 325 |
+
"loss": 0.6011,
|
| 326 |
+
"step": 4700
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 60.0,
|
| 330 |
+
"learning_rate": 3.205e-05,
|
| 331 |
+
"loss": 0.5815,
|
| 332 |
+
"step": 4800
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 61.25,
|
| 336 |
+
"learning_rate": 3.1049999999999996e-05,
|
| 337 |
+
"loss": 0.5704,
|
| 338 |
+
"step": 4900
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 62.5,
|
| 342 |
+
"learning_rate": 3.0049999999999995e-05,
|
| 343 |
+
"loss": 0.5763,
|
| 344 |
+
"step": 5000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 62.5,
|
| 348 |
+
"eval_loss": 0.21160605549812317,
|
| 349 |
+
"eval_runtime": 41.2219,
|
| 350 |
+
"eval_samples_per_second": 28.723,
|
| 351 |
+
"eval_steps_per_second": 1.795,
|
| 352 |
+
"eval_wer": 0.20548753503466588,
|
| 353 |
+
"step": 5000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 63.75,
|
| 357 |
+
"learning_rate": 2.9049999999999995e-05,
|
| 358 |
+
"loss": 0.5742,
|
| 359 |
+
"step": 5100
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 65.0,
|
| 363 |
+
"learning_rate": 2.8049999999999997e-05,
|
| 364 |
+
"loss": 0.58,
|
| 365 |
+
"step": 5200
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 66.25,
|
| 369 |
+
"learning_rate": 2.705e-05,
|
| 370 |
+
"loss": 0.5542,
|
| 371 |
+
"step": 5300
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 67.5,
|
| 375 |
+
"learning_rate": 2.605e-05,
|
| 376 |
+
"loss": 0.5472,
|
| 377 |
+
"step": 5400
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 68.75,
|
| 381 |
+
"learning_rate": 2.505e-05,
|
| 382 |
+
"loss": 0.5433,
|
| 383 |
+
"step": 5500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 70.0,
|
| 387 |
+
"learning_rate": 2.405e-05,
|
| 388 |
+
"loss": 0.5448,
|
| 389 |
+
"step": 5600
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 71.25,
|
| 393 |
+
"learning_rate": 2.3049999999999998e-05,
|
| 394 |
+
"loss": 0.5469,
|
| 395 |
+
"step": 5700
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 72.5,
|
| 399 |
+
"learning_rate": 2.2049999999999997e-05,
|
| 400 |
+
"loss": 0.5424,
|
| 401 |
+
"step": 5800
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 73.75,
|
| 405 |
+
"learning_rate": 2.1049999999999997e-05,
|
| 406 |
+
"loss": 0.5221,
|
| 407 |
+
"step": 5900
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 75.0,
|
| 411 |
+
"learning_rate": 2.0049999999999996e-05,
|
| 412 |
+
"loss": 0.5196,
|
| 413 |
+
"step": 6000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 75.0,
|
| 417 |
+
"eval_loss": 0.21106497943401337,
|
| 418 |
+
"eval_runtime": 41.193,
|
| 419 |
+
"eval_samples_per_second": 28.743,
|
| 420 |
+
"eval_steps_per_second": 1.796,
|
| 421 |
+
"eval_wer": 0.19103112553473964,
|
| 422 |
+
"step": 6000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 76.25,
|
| 426 |
+
"learning_rate": 1.905e-05,
|
| 427 |
+
"loss": 0.5215,
|
| 428 |
+
"step": 6100
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 77.5,
|
| 432 |
+
"learning_rate": 1.8049999999999998e-05,
|
| 433 |
+
"loss": 0.5141,
|
| 434 |
+
"step": 6200
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 78.75,
|
| 438 |
+
"learning_rate": 1.7049999999999998e-05,
|
| 439 |
+
"loss": 0.5087,
|
| 440 |
+
"step": 6300
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 80.0,
|
| 444 |
+
"learning_rate": 1.6049999999999997e-05,
|
| 445 |
+
"loss": 0.5136,
|
| 446 |
+
"step": 6400
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 81.25,
|
| 450 |
+
"learning_rate": 1.5049999999999998e-05,
|
| 451 |
+
"loss": 0.4973,
|
| 452 |
+
"step": 6500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 82.5,
|
| 456 |
+
"learning_rate": 1.4049999999999998e-05,
|
| 457 |
+
"loss": 0.5064,
|
| 458 |
+
"step": 6600
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 83.75,
|
| 462 |
+
"learning_rate": 1.3049999999999999e-05,
|
| 463 |
+
"loss": 0.4985,
|
| 464 |
+
"step": 6700
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 85.0,
|
| 468 |
+
"learning_rate": 1.205e-05,
|
| 469 |
+
"loss": 0.4926,
|
| 470 |
+
"step": 6800
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 86.25,
|
| 474 |
+
"learning_rate": 1.105e-05,
|
| 475 |
+
"loss": 0.4973,
|
| 476 |
+
"step": 6900
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 87.5,
|
| 480 |
+
"learning_rate": 1.0049999999999999e-05,
|
| 481 |
+
"loss": 0.4949,
|
| 482 |
+
"step": 7000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 87.5,
|
| 486 |
+
"eval_loss": 0.21311460435390472,
|
| 487 |
+
"eval_runtime": 43.2323,
|
| 488 |
+
"eval_samples_per_second": 27.387,
|
| 489 |
+
"eval_steps_per_second": 1.712,
|
| 490 |
+
"eval_wer": 0.19309632689187195,
|
| 491 |
+
"step": 7000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 88.75,
|
| 495 |
+
"learning_rate": 9.05e-06,
|
| 496 |
+
"loss": 0.5008,
|
| 497 |
+
"step": 7100
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 90.0,
|
| 501 |
+
"learning_rate": 8.05e-06,
|
| 502 |
+
"loss": 0.4834,
|
| 503 |
+
"step": 7200
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 91.25,
|
| 507 |
+
"learning_rate": 7.049999999999999e-06,
|
| 508 |
+
"loss": 0.4717,
|
| 509 |
+
"step": 7300
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 92.5,
|
| 513 |
+
"learning_rate": 6.049999999999999e-06,
|
| 514 |
+
"loss": 0.485,
|
| 515 |
+
"step": 7400
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 93.75,
|
| 519 |
+
"learning_rate": 5.05e-06,
|
| 520 |
+
"loss": 0.4835,
|
| 521 |
+
"step": 7500
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 95.0,
|
| 525 |
+
"learning_rate": 4.049999999999999e-06,
|
| 526 |
+
"loss": 0.4787,
|
| 527 |
+
"step": 7600
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 96.25,
|
| 531 |
+
"learning_rate": 3.05e-06,
|
| 532 |
+
"loss": 0.4747,
|
| 533 |
+
"step": 7700
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 97.5,
|
| 537 |
+
"learning_rate": 2.05e-06,
|
| 538 |
+
"loss": 0.4804,
|
| 539 |
+
"step": 7800
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 98.75,
|
| 543 |
+
"learning_rate": 1.05e-06,
|
| 544 |
+
"loss": 0.4731,
|
| 545 |
+
"step": 7900
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 100.0,
|
| 549 |
+
"learning_rate": 6e-08,
|
| 550 |
+
"loss": 0.4797,
|
| 551 |
+
"step": 8000
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 100.0,
|
| 555 |
+
"eval_loss": 0.20929841697216034,
|
| 556 |
+
"eval_runtime": 42.6795,
|
| 557 |
+
"eval_samples_per_second": 27.742,
|
| 558 |
+
"eval_steps_per_second": 1.734,
|
| 559 |
+
"eval_wer": 0.190736096769435,
|
| 560 |
+
"step": 8000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 100.0,
|
| 564 |
+
"step": 8000,
|
| 565 |
+
"total_flos": 2.997146399308124e+19,
|
| 566 |
+
"train_loss": 1.1495286922454835,
|
| 567 |
+
"train_runtime": 11669.3128,
|
| 568 |
+
"train_samples_per_second": 21.921,
|
| 569 |
+
"train_steps_per_second": 0.686
|
| 570 |
+
}
|
| 571 |
+
],
|
| 572 |
+
"max_steps": 8000,
|
| 573 |
+
"num_train_epochs": 100,
|
| 574 |
+
"total_flos": 2.997146399308124e+19,
|
| 575 |
+
"trial_name": null,
|
| 576 |
+
"trial_params": null
|
| 577 |
+
}
|