Commit
•
6fdd07c
1
Parent(s):
3791502
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +409 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 200.0,
|
| 3 |
+
"eval_loss": 1.0355710983276367,
|
| 4 |
+
"eval_runtime": 14.0252,
|
| 5 |
+
"eval_samples": 294,
|
| 6 |
+
"eval_samples_per_second": 20.962,
|
| 7 |
+
"eval_steps_per_second": 0.713,
|
| 8 |
+
"eval_wer": 0.6523702031602708,
|
| 9 |
+
"train_loss": 0.8014375554598295,
|
| 10 |
+
"train_runtime": 12807.5215,
|
| 11 |
+
"train_samples": 828,
|
| 12 |
+
"train_samples_per_second": 12.93,
|
| 13 |
+
"train_steps_per_second": 0.812
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 200.0,
|
| 3 |
+
"eval_loss": 1.0355710983276367,
|
| 4 |
+
"eval_runtime": 14.0252,
|
| 5 |
+
"eval_samples": 294,
|
| 6 |
+
"eval_samples_per_second": 20.962,
|
| 7 |
+
"eval_steps_per_second": 0.713,
|
| 8 |
+
"eval_wer": 0.6523702031602708
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 200.0,
|
| 3 |
+
"train_loss": 0.8014375554598295,
|
| 4 |
+
"train_runtime": 12807.5215,
|
| 5 |
+
"train_samples": 828,
|
| 6 |
+
"train_samples_per_second": 12.93,
|
| 7 |
+
"train_steps_per_second": 0.812
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,409 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 200.0,
|
| 5 |
+
"global_step": 10400,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 5.77,
|
| 12 |
+
"learning_rate": 0.000148,
|
| 13 |
+
"loss": 5.649,
|
| 14 |
+
"step": 300
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 9.62,
|
| 18 |
+
"eval_loss": 3.0038444995880127,
|
| 19 |
+
"eval_runtime": 14.1536,
|
| 20 |
+
"eval_samples_per_second": 20.772,
|
| 21 |
+
"eval_steps_per_second": 0.707,
|
| 22 |
+
"eval_wer": 1.0,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 11.54,
|
| 27 |
+
"learning_rate": 0.00029800000000000003,
|
| 28 |
+
"loss": 2.9978,
|
| 29 |
+
"step": 600
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 17.31,
|
| 33 |
+
"learning_rate": 0.00039600000000000003,
|
| 34 |
+
"loss": 1.6272,
|
| 35 |
+
"step": 900
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 19.23,
|
| 39 |
+
"eval_loss": 0.7362223267555237,
|
| 40 |
+
"eval_runtime": 13.8179,
|
| 41 |
+
"eval_samples_per_second": 21.277,
|
| 42 |
+
"eval_steps_per_second": 0.724,
|
| 43 |
+
"eval_wer": 0.781941309255079,
|
| 44 |
+
"step": 1000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 23.08,
|
| 48 |
+
"learning_rate": 0.0003835,
|
| 49 |
+
"loss": 1.2822,
|
| 50 |
+
"step": 1200
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 28.85,
|
| 54 |
+
"learning_rate": 0.000371,
|
| 55 |
+
"loss": 1.1354,
|
| 56 |
+
"step": 1500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 28.85,
|
| 60 |
+
"eval_loss": 0.6409761309623718,
|
| 61 |
+
"eval_runtime": 13.8178,
|
| 62 |
+
"eval_samples_per_second": 21.277,
|
| 63 |
+
"eval_steps_per_second": 0.724,
|
| 64 |
+
"eval_wer": 0.7110609480812641,
|
| 65 |
+
"step": 1500
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 34.62,
|
| 69 |
+
"learning_rate": 0.00035850000000000004,
|
| 70 |
+
"loss": 1.0424,
|
| 71 |
+
"step": 1800
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 38.46,
|
| 75 |
+
"eval_loss": 0.6907294392585754,
|
| 76 |
+
"eval_runtime": 13.9619,
|
| 77 |
+
"eval_samples_per_second": 21.057,
|
| 78 |
+
"eval_steps_per_second": 0.716,
|
| 79 |
+
"eval_wer": 0.7431151241534989,
|
| 80 |
+
"step": 2000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 40.38,
|
| 84 |
+
"learning_rate": 0.000346,
|
| 85 |
+
"loss": 0.9872,
|
| 86 |
+
"step": 2100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 46.15,
|
| 90 |
+
"learning_rate": 0.00033350000000000003,
|
| 91 |
+
"loss": 0.9293,
|
| 92 |
+
"step": 2400
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 48.08,
|
| 96 |
+
"eval_loss": 0.7248561978340149,
|
| 97 |
+
"eval_runtime": 14.1849,
|
| 98 |
+
"eval_samples_per_second": 20.726,
|
| 99 |
+
"eval_steps_per_second": 0.705,
|
| 100 |
+
"eval_wer": 0.7101580135440181,
|
| 101 |
+
"step": 2500
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 51.92,
|
| 105 |
+
"learning_rate": 0.000321,
|
| 106 |
+
"loss": 0.8747,
|
| 107 |
+
"step": 2700
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 57.69,
|
| 111 |
+
"learning_rate": 0.0003085,
|
| 112 |
+
"loss": 0.8246,
|
| 113 |
+
"step": 3000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 57.69,
|
| 117 |
+
"eval_loss": 0.7421836853027344,
|
| 118 |
+
"eval_runtime": 14.4192,
|
| 119 |
+
"eval_samples_per_second": 20.39,
|
| 120 |
+
"eval_steps_per_second": 0.694,
|
| 121 |
+
"eval_wer": 0.6966139954853273,
|
| 122 |
+
"step": 3000
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 63.46,
|
| 126 |
+
"learning_rate": 0.000296,
|
| 127 |
+
"loss": 0.7837,
|
| 128 |
+
"step": 3300
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 67.31,
|
| 132 |
+
"eval_loss": 0.7412946820259094,
|
| 133 |
+
"eval_runtime": 14.1823,
|
| 134 |
+
"eval_samples_per_second": 20.73,
|
| 135 |
+
"eval_steps_per_second": 0.705,
|
| 136 |
+
"eval_wer": 0.6812641083521445,
|
| 137 |
+
"step": 3500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 69.23,
|
| 141 |
+
"learning_rate": 0.0002835,
|
| 142 |
+
"loss": 0.7527,
|
| 143 |
+
"step": 3600
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 75.0,
|
| 147 |
+
"learning_rate": 0.00027100000000000003,
|
| 148 |
+
"loss": 0.7147,
|
| 149 |
+
"step": 3900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 76.92,
|
| 153 |
+
"eval_loss": 0.7873469591140747,
|
| 154 |
+
"eval_runtime": 13.9067,
|
| 155 |
+
"eval_samples_per_second": 21.141,
|
| 156 |
+
"eval_steps_per_second": 0.719,
|
| 157 |
+
"eval_wer": 0.6930022573363431,
|
| 158 |
+
"step": 4000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 80.77,
|
| 162 |
+
"learning_rate": 0.0002585,
|
| 163 |
+
"loss": 0.6779,
|
| 164 |
+
"step": 4200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 86.54,
|
| 168 |
+
"learning_rate": 0.000246,
|
| 169 |
+
"loss": 0.6276,
|
| 170 |
+
"step": 4500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 86.54,
|
| 174 |
+
"eval_loss": 0.8037810921669006,
|
| 175 |
+
"eval_runtime": 14.0837,
|
| 176 |
+
"eval_samples_per_second": 20.875,
|
| 177 |
+
"eval_steps_per_second": 0.71,
|
| 178 |
+
"eval_wer": 0.6677200902934537,
|
| 179 |
+
"step": 4500
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 92.31,
|
| 183 |
+
"learning_rate": 0.0002335,
|
| 184 |
+
"loss": 0.6041,
|
| 185 |
+
"step": 4800
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 96.15,
|
| 189 |
+
"eval_loss": 0.8240488767623901,
|
| 190 |
+
"eval_runtime": 13.9338,
|
| 191 |
+
"eval_samples_per_second": 21.1,
|
| 192 |
+
"eval_steps_per_second": 0.718,
|
| 193 |
+
"eval_wer": 0.6830699774266366,
|
| 194 |
+
"step": 5000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 98.08,
|
| 198 |
+
"learning_rate": 0.000221,
|
| 199 |
+
"loss": 0.5588,
|
| 200 |
+
"step": 5100
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 103.85,
|
| 204 |
+
"learning_rate": 0.0002085,
|
| 205 |
+
"loss": 0.5336,
|
| 206 |
+
"step": 5400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 105.77,
|
| 210 |
+
"eval_loss": 0.8747946619987488,
|
| 211 |
+
"eval_runtime": 13.9807,
|
| 212 |
+
"eval_samples_per_second": 21.029,
|
| 213 |
+
"eval_steps_per_second": 0.715,
|
| 214 |
+
"eval_wer": 0.6749435665914221,
|
| 215 |
+
"step": 5500
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 109.62,
|
| 219 |
+
"learning_rate": 0.000196,
|
| 220 |
+
"loss": 0.5,
|
| 221 |
+
"step": 5700
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 115.38,
|
| 225 |
+
"learning_rate": 0.00018350000000000002,
|
| 226 |
+
"loss": 0.4705,
|
| 227 |
+
"step": 6000
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 115.38,
|
| 231 |
+
"eval_loss": 0.9005643129348755,
|
| 232 |
+
"eval_runtime": 13.867,
|
| 233 |
+
"eval_samples_per_second": 21.201,
|
| 234 |
+
"eval_steps_per_second": 0.721,
|
| 235 |
+
"eval_wer": 0.6496613995485327,
|
| 236 |
+
"step": 6000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 121.15,
|
| 240 |
+
"learning_rate": 0.00017104166666666667,
|
| 241 |
+
"loss": 0.43,
|
| 242 |
+
"step": 6300
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 125.0,
|
| 246 |
+
"eval_loss": 0.8953593969345093,
|
| 247 |
+
"eval_runtime": 13.9621,
|
| 248 |
+
"eval_samples_per_second": 21.057,
|
| 249 |
+
"eval_steps_per_second": 0.716,
|
| 250 |
+
"eval_wer": 0.655079006772009,
|
| 251 |
+
"step": 6500
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 126.92,
|
| 255 |
+
"learning_rate": 0.00015854166666666667,
|
| 256 |
+
"loss": 0.4068,
|
| 257 |
+
"step": 6600
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 132.69,
|
| 261 |
+
"learning_rate": 0.0001460416666666667,
|
| 262 |
+
"loss": 0.3859,
|
| 263 |
+
"step": 6900
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 134.62,
|
| 267 |
+
"eval_loss": 0.9073536396026611,
|
| 268 |
+
"eval_runtime": 13.9518,
|
| 269 |
+
"eval_samples_per_second": 21.072,
|
| 270 |
+
"eval_steps_per_second": 0.717,
|
| 271 |
+
"eval_wer": 0.6613995485327314,
|
| 272 |
+
"step": 7000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 138.46,
|
| 276 |
+
"learning_rate": 0.00013354166666666668,
|
| 277 |
+
"loss": 0.3622,
|
| 278 |
+
"step": 7200
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 144.23,
|
| 282 |
+
"learning_rate": 0.00012104166666666668,
|
| 283 |
+
"loss": 0.3342,
|
| 284 |
+
"step": 7500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 144.23,
|
| 288 |
+
"eval_loss": 0.9693499803543091,
|
| 289 |
+
"eval_runtime": 13.8467,
|
| 290 |
+
"eval_samples_per_second": 21.233,
|
| 291 |
+
"eval_steps_per_second": 0.722,
|
| 292 |
+
"eval_wer": 0.6559819413092551,
|
| 293 |
+
"step": 7500
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 150.0,
|
| 297 |
+
"learning_rate": 0.00010854166666666667,
|
| 298 |
+
"loss": 0.3155,
|
| 299 |
+
"step": 7800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 153.85,
|
| 303 |
+
"eval_loss": 1.0072590112686157,
|
| 304 |
+
"eval_runtime": 13.7449,
|
| 305 |
+
"eval_samples_per_second": 21.39,
|
| 306 |
+
"eval_steps_per_second": 0.728,
|
| 307 |
+
"eval_wer": 0.6690744920993228,
|
| 308 |
+
"step": 8000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 155.77,
|
| 312 |
+
"learning_rate": 9.604166666666668e-05,
|
| 313 |
+
"loss": 0.2894,
|
| 314 |
+
"step": 8100
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 161.54,
|
| 318 |
+
"learning_rate": 8.358333333333334e-05,
|
| 319 |
+
"loss": 0.2673,
|
| 320 |
+
"step": 8400
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 163.46,
|
| 324 |
+
"eval_loss": 1.0170269012451172,
|
| 325 |
+
"eval_runtime": 14.0595,
|
| 326 |
+
"eval_samples_per_second": 20.911,
|
| 327 |
+
"eval_steps_per_second": 0.711,
|
| 328 |
+
"eval_wer": 0.6632054176072235,
|
| 329 |
+
"step": 8500
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 167.31,
|
| 333 |
+
"learning_rate": 7.108333333333333e-05,
|
| 334 |
+
"loss": 0.2517,
|
| 335 |
+
"step": 8700
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 173.08,
|
| 339 |
+
"learning_rate": 5.858333333333333e-05,
|
| 340 |
+
"loss": 0.2409,
|
| 341 |
+
"step": 9000
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 173.08,
|
| 345 |
+
"eval_loss": 1.0304286479949951,
|
| 346 |
+
"eval_runtime": 13.8942,
|
| 347 |
+
"eval_samples_per_second": 21.16,
|
| 348 |
+
"eval_steps_per_second": 0.72,
|
| 349 |
+
"eval_wer": 0.6708803611738149,
|
| 350 |
+
"step": 9000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 178.85,
|
| 354 |
+
"learning_rate": 4.608333333333333e-05,
|
| 355 |
+
"loss": 0.2189,
|
| 356 |
+
"step": 9300
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 182.69,
|
| 360 |
+
"eval_loss": 0.9965260624885559,
|
| 361 |
+
"eval_runtime": 14.2442,
|
| 362 |
+
"eval_samples_per_second": 20.64,
|
| 363 |
+
"eval_steps_per_second": 0.702,
|
| 364 |
+
"eval_wer": 0.654627539503386,
|
| 365 |
+
"step": 9500
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 184.62,
|
| 369 |
+
"learning_rate": 3.3625000000000004e-05,
|
| 370 |
+
"loss": 0.203,
|
| 371 |
+
"step": 9600
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 190.38,
|
| 375 |
+
"learning_rate": 2.1125000000000002e-05,
|
| 376 |
+
"loss": 0.1973,
|
| 377 |
+
"step": 9900
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 192.31,
|
| 381 |
+
"eval_loss": 1.0360474586486816,
|
| 382 |
+
"eval_runtime": 14.1087,
|
| 383 |
+
"eval_samples_per_second": 20.838,
|
| 384 |
+
"eval_steps_per_second": 0.709,
|
| 385 |
+
"eval_wer": 0.655079006772009,
|
| 386 |
+
"step": 10000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 196.15,
|
| 390 |
+
"learning_rate": 8.625e-06,
|
| 391 |
+
"loss": 0.1881,
|
| 392 |
+
"step": 10200
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 200.0,
|
| 396 |
+
"step": 10400,
|
| 397 |
+
"total_flos": 3.0179570579437056e+19,
|
| 398 |
+
"train_loss": 0.8014375554598295,
|
| 399 |
+
"train_runtime": 12807.5215,
|
| 400 |
+
"train_samples_per_second": 12.93,
|
| 401 |
+
"train_steps_per_second": 0.812
|
| 402 |
+
}
|
| 403 |
+
],
|
| 404 |
+
"max_steps": 10400,
|
| 405 |
+
"num_train_epochs": 200,
|
| 406 |
+
"total_flos": 3.0179570579437056e+19,
|
| 407 |
+
"trial_name": null,
|
| 408 |
+
"trial_params": null
|
| 409 |
+
}
|