Commit
•
c0d4c3c
1
Parent(s):
f2e517b
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +415 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.2756262719631195,
|
| 4 |
+
"eval_runtime": 36.5594,
|
| 5 |
+
"eval_samples": 1193,
|
| 6 |
+
"eval_samples_per_second": 32.632,
|
| 7 |
+
"eval_steps_per_second": 1.039,
|
| 8 |
+
"eval_wer": 0.22793148880105402,
|
| 9 |
+
"train_loss": 1.2146642443028892,
|
| 10 |
+
"train_runtime": 11060.029,
|
| 11 |
+
"train_samples": 2606,
|
| 12 |
+
"train_samples_per_second": 23.562,
|
| 13 |
+
"train_steps_per_second": 0.741
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.2756262719631195,
|
| 4 |
+
"eval_runtime": 36.5594,
|
| 5 |
+
"eval_samples": 1193,
|
| 6 |
+
"eval_samples_per_second": 32.632,
|
| 7 |
+
"eval_steps_per_second": 1.039,
|
| 8 |
+
"eval_wer": 0.22793148880105402
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"train_loss": 1.2146642443028892,
|
| 4 |
+
"train_runtime": 11060.029,
|
| 5 |
+
"train_samples": 2606,
|
| 6 |
+
"train_samples_per_second": 23.562,
|
| 7 |
+
"train_steps_per_second": 0.741
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,415 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"global_step": 8200,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 2.44,
|
| 12 |
+
"learning_rate": 1.4058000000000002e-05,
|
| 13 |
+
"loss": 8.5696,
|
| 14 |
+
"step": 200
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 4.88,
|
| 18 |
+
"learning_rate": 2.8258000000000003e-05,
|
| 19 |
+
"loss": 3.3881,
|
| 20 |
+
"step": 400
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 6.1,
|
| 24 |
+
"eval_loss": 2.9710121154785156,
|
| 25 |
+
"eval_runtime": 37.4848,
|
| 26 |
+
"eval_samples_per_second": 31.826,
|
| 27 |
+
"eval_steps_per_second": 1.014,
|
| 28 |
+
"eval_wer": 1.0,
|
| 29 |
+
"step": 500
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 7.32,
|
| 33 |
+
"learning_rate": 4.2458e-05,
|
| 34 |
+
"loss": 2.9695,
|
| 35 |
+
"step": 600
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 9.76,
|
| 39 |
+
"learning_rate": 5.6658000000000005e-05,
|
| 40 |
+
"loss": 2.8538,
|
| 41 |
+
"step": 800
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 12.2,
|
| 45 |
+
"learning_rate": 7.0858e-05,
|
| 46 |
+
"loss": 2.6401,
|
| 47 |
+
"step": 1000
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 12.2,
|
| 51 |
+
"eval_loss": 1.7677034139633179,
|
| 52 |
+
"eval_runtime": 36.6194,
|
| 53 |
+
"eval_samples_per_second": 32.578,
|
| 54 |
+
"eval_steps_per_second": 1.038,
|
| 55 |
+
"eval_wer": 0.9733567559654516,
|
| 56 |
+
"step": 1000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 14.63,
|
| 60 |
+
"learning_rate": 6.904750000000001e-05,
|
| 61 |
+
"loss": 1.81,
|
| 62 |
+
"step": 1200
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 17.07,
|
| 66 |
+
"learning_rate": 6.707527777777778e-05,
|
| 67 |
+
"loss": 1.5152,
|
| 68 |
+
"step": 1400
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 18.29,
|
| 72 |
+
"eval_loss": 0.5564321875572205,
|
| 73 |
+
"eval_runtime": 36.9716,
|
| 74 |
+
"eval_samples_per_second": 32.268,
|
| 75 |
+
"eval_steps_per_second": 1.028,
|
| 76 |
+
"eval_wer": 0.6010832967354707,
|
| 77 |
+
"step": 1500
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 19.51,
|
| 81 |
+
"learning_rate": 6.510305555555556e-05,
|
| 82 |
+
"loss": 1.386,
|
| 83 |
+
"step": 1600
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 21.95,
|
| 87 |
+
"learning_rate": 6.313083333333334e-05,
|
| 88 |
+
"loss": 1.3129,
|
| 89 |
+
"step": 1800
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 24.39,
|
| 93 |
+
"learning_rate": 6.115861111111111e-05,
|
| 94 |
+
"loss": 1.2191,
|
| 95 |
+
"step": 2000
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 24.39,
|
| 99 |
+
"eval_loss": 0.43192508816719055,
|
| 100 |
+
"eval_runtime": 36.8038,
|
| 101 |
+
"eval_samples_per_second": 32.415,
|
| 102 |
+
"eval_steps_per_second": 1.033,
|
| 103 |
+
"eval_wer": 0.4390279607670912,
|
| 104 |
+
"step": 2000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 26.83,
|
| 108 |
+
"learning_rate": 5.918638888888889e-05,
|
| 109 |
+
"loss": 1.1031,
|
| 110 |
+
"step": 2200
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 29.27,
|
| 114 |
+
"learning_rate": 5.7214166666666665e-05,
|
| 115 |
+
"loss": 1.0237,
|
| 116 |
+
"step": 2400
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 30.49,
|
| 120 |
+
"eval_loss": 0.3140852749347687,
|
| 121 |
+
"eval_runtime": 36.7964,
|
| 122 |
+
"eval_samples_per_second": 32.422,
|
| 123 |
+
"eval_steps_per_second": 1.033,
|
| 124 |
+
"eval_wer": 0.31752305665349145,
|
| 125 |
+
"step": 2500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 31.71,
|
| 129 |
+
"learning_rate": 5.5241944444444446e-05,
|
| 130 |
+
"loss": 0.9707,
|
| 131 |
+
"step": 2600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 34.15,
|
| 135 |
+
"learning_rate": 5.326972222222223e-05,
|
| 136 |
+
"loss": 0.9189,
|
| 137 |
+
"step": 2800
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 36.59,
|
| 141 |
+
"learning_rate": 5.129750000000001e-05,
|
| 142 |
+
"loss": 0.8892,
|
| 143 |
+
"step": 3000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 36.59,
|
| 147 |
+
"eval_loss": 0.27478542923927307,
|
| 148 |
+
"eval_runtime": 37.4013,
|
| 149 |
+
"eval_samples_per_second": 31.897,
|
| 150 |
+
"eval_steps_per_second": 1.016,
|
| 151 |
+
"eval_wer": 0.2689210950080515,
|
| 152 |
+
"step": 3000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 39.02,
|
| 156 |
+
"learning_rate": 4.932527777777778e-05,
|
| 157 |
+
"loss": 0.8573,
|
| 158 |
+
"step": 3200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 41.46,
|
| 162 |
+
"learning_rate": 4.735305555555556e-05,
|
| 163 |
+
"loss": 0.8296,
|
| 164 |
+
"step": 3400
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 42.68,
|
| 168 |
+
"eval_loss": 0.26798102259635925,
|
| 169 |
+
"eval_runtime": 37.6187,
|
| 170 |
+
"eval_samples_per_second": 31.713,
|
| 171 |
+
"eval_steps_per_second": 1.01,
|
| 172 |
+
"eval_wer": 0.2534036012296882,
|
| 173 |
+
"step": 3500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 43.9,
|
| 177 |
+
"learning_rate": 4.538083333333334e-05,
|
| 178 |
+
"loss": 0.8047,
|
| 179 |
+
"step": 3600
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 46.34,
|
| 183 |
+
"learning_rate": 4.340861111111111e-05,
|
| 184 |
+
"loss": 0.7829,
|
| 185 |
+
"step": 3800
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 48.78,
|
| 189 |
+
"learning_rate": 4.1436388888888886e-05,
|
| 190 |
+
"loss": 0.7602,
|
| 191 |
+
"step": 4000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 48.78,
|
| 195 |
+
"eval_loss": 0.28198665380477905,
|
| 196 |
+
"eval_runtime": 36.6815,
|
| 197 |
+
"eval_samples_per_second": 32.523,
|
| 198 |
+
"eval_steps_per_second": 1.036,
|
| 199 |
+
"eval_wer": 0.25062216366564194,
|
| 200 |
+
"step": 4000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 51.22,
|
| 204 |
+
"learning_rate": 3.946416666666667e-05,
|
| 205 |
+
"loss": 0.7536,
|
| 206 |
+
"step": 4200
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 53.66,
|
| 210 |
+
"learning_rate": 3.7501805555555555e-05,
|
| 211 |
+
"loss": 0.7186,
|
| 212 |
+
"step": 4400
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 54.88,
|
| 216 |
+
"eval_loss": 0.2672194242477417,
|
| 217 |
+
"eval_runtime": 36.55,
|
| 218 |
+
"eval_samples_per_second": 32.64,
|
| 219 |
+
"eval_steps_per_second": 1.04,
|
| 220 |
+
"eval_wer": 0.23978919631093545,
|
| 221 |
+
"step": 4500
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 56.1,
|
| 225 |
+
"learning_rate": 3.552958333333333e-05,
|
| 226 |
+
"loss": 0.7166,
|
| 227 |
+
"step": 4600
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 58.54,
|
| 231 |
+
"learning_rate": 3.355736111111111e-05,
|
| 232 |
+
"loss": 0.706,
|
| 233 |
+
"step": 4800
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 60.98,
|
| 237 |
+
"learning_rate": 3.1595000000000005e-05,
|
| 238 |
+
"loss": 0.6887,
|
| 239 |
+
"step": 5000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 60.98,
|
| 243 |
+
"eval_loss": 0.27292799949645996,
|
| 244 |
+
"eval_runtime": 36.0168,
|
| 245 |
+
"eval_samples_per_second": 33.123,
|
| 246 |
+
"eval_steps_per_second": 1.055,
|
| 247 |
+
"eval_wer": 0.24022837066315328,
|
| 248 |
+
"step": 5000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 63.41,
|
| 252 |
+
"learning_rate": 2.962277777777778e-05,
|
| 253 |
+
"loss": 0.6664,
|
| 254 |
+
"step": 5200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 65.85,
|
| 258 |
+
"learning_rate": 2.7650555555555557e-05,
|
| 259 |
+
"loss": 0.6507,
|
| 260 |
+
"step": 5400
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 67.07,
|
| 264 |
+
"eval_loss": 0.27668464183807373,
|
| 265 |
+
"eval_runtime": 37.7209,
|
| 266 |
+
"eval_samples_per_second": 31.627,
|
| 267 |
+
"eval_steps_per_second": 1.007,
|
| 268 |
+
"eval_wer": 0.23612941004245352,
|
| 269 |
+
"step": 5500
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 68.29,
|
| 273 |
+
"learning_rate": 2.5678333333333338e-05,
|
| 274 |
+
"loss": 0.6418,
|
| 275 |
+
"step": 5600
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 70.73,
|
| 279 |
+
"learning_rate": 2.3706111111111112e-05,
|
| 280 |
+
"loss": 0.6432,
|
| 281 |
+
"step": 5800
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 73.17,
|
| 285 |
+
"learning_rate": 2.173388888888889e-05,
|
| 286 |
+
"loss": 0.6226,
|
| 287 |
+
"step": 6000
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 73.17,
|
| 291 |
+
"eval_loss": 0.2817266285419464,
|
| 292 |
+
"eval_runtime": 36.1511,
|
| 293 |
+
"eval_samples_per_second": 33.0,
|
| 294 |
+
"eval_steps_per_second": 1.051,
|
| 295 |
+
"eval_wer": 0.233201581027668,
|
| 296 |
+
"step": 6000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 75.61,
|
| 300 |
+
"learning_rate": 1.9761666666666667e-05,
|
| 301 |
+
"loss": 0.6183,
|
| 302 |
+
"step": 6200
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 78.05,
|
| 306 |
+
"learning_rate": 1.7789444444444445e-05,
|
| 307 |
+
"loss": 0.6024,
|
| 308 |
+
"step": 6400
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 79.27,
|
| 312 |
+
"eval_loss": 0.26792478561401367,
|
| 313 |
+
"eval_runtime": 36.4629,
|
| 314 |
+
"eval_samples_per_second": 32.718,
|
| 315 |
+
"eval_steps_per_second": 1.042,
|
| 316 |
+
"eval_wer": 0.22793148880105402,
|
| 317 |
+
"step": 6500
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 80.49,
|
| 321 |
+
"learning_rate": 1.5817222222222223e-05,
|
| 322 |
+
"loss": 0.6028,
|
| 323 |
+
"step": 6600
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 82.93,
|
| 327 |
+
"learning_rate": 1.3845000000000002e-05,
|
| 328 |
+
"loss": 0.5855,
|
| 329 |
+
"step": 6800
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 85.37,
|
| 333 |
+
"learning_rate": 1.1872777777777778e-05,
|
| 334 |
+
"loss": 0.5787,
|
| 335 |
+
"step": 7000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 85.37,
|
| 339 |
+
"eval_loss": 0.28374168276786804,
|
| 340 |
+
"eval_runtime": 36.0785,
|
| 341 |
+
"eval_samples_per_second": 33.067,
|
| 342 |
+
"eval_steps_per_second": 1.053,
|
| 343 |
+
"eval_wer": 0.23159127506953595,
|
| 344 |
+
"step": 7000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 87.8,
|
| 348 |
+
"learning_rate": 9.910416666666668e-06,
|
| 349 |
+
"loss": 0.5824,
|
| 350 |
+
"step": 7200
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 90.24,
|
| 354 |
+
"learning_rate": 7.938194444444445e-06,
|
| 355 |
+
"loss": 0.5744,
|
| 356 |
+
"step": 7400
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 91.46,
|
| 360 |
+
"eval_loss": 0.28380560874938965,
|
| 361 |
+
"eval_runtime": 35.876,
|
| 362 |
+
"eval_samples_per_second": 33.253,
|
| 363 |
+
"eval_steps_per_second": 1.059,
|
| 364 |
+
"eval_wer": 0.22837066315327184,
|
| 365 |
+
"step": 7500
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 92.68,
|
| 369 |
+
"learning_rate": 5.965972222222223e-06,
|
| 370 |
+
"loss": 0.569,
|
| 371 |
+
"step": 7600
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 95.12,
|
| 375 |
+
"learning_rate": 3.9937500000000005e-06,
|
| 376 |
+
"loss": 0.561,
|
| 377 |
+
"step": 7800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 97.56,
|
| 381 |
+
"learning_rate": 2.0215277777777777e-06,
|
| 382 |
+
"loss": 0.5556,
|
| 383 |
+
"step": 8000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 97.56,
|
| 387 |
+
"eval_loss": 0.2762868106365204,
|
| 388 |
+
"eval_runtime": 36.6971,
|
| 389 |
+
"eval_samples_per_second": 32.509,
|
| 390 |
+
"eval_steps_per_second": 1.036,
|
| 391 |
+
"eval_wer": 0.2280778802517933,
|
| 392 |
+
"step": 8000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 100.0,
|
| 396 |
+
"learning_rate": 4.930555555555556e-08,
|
| 397 |
+
"loss": 0.558,
|
| 398 |
+
"step": 8200
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 100.0,
|
| 402 |
+
"step": 8200,
|
| 403 |
+
"total_flos": 2.609429292408082e+19,
|
| 404 |
+
"train_loss": 1.2146642443028892,
|
| 405 |
+
"train_runtime": 11060.029,
|
| 406 |
+
"train_samples_per_second": 23.562,
|
| 407 |
+
"train_steps_per_second": 0.741
|
| 408 |
+
}
|
| 409 |
+
],
|
| 410 |
+
"max_steps": 8200,
|
| 411 |
+
"num_train_epochs": 100,
|
| 412 |
+
"total_flos": 2.609429292408082e+19,
|
| 413 |
+
"trial_name": null,
|
| 414 |
+
"trial_params": null
|
| 415 |
+
}
|