Ramnathan
commited on
Commit
·
364e224
1
Parent(s):
6e5e690
Update from
Browse files- trainer_state.json +436 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,436 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 8.658008658008658,
|
| 5 |
+
"global_step": 2000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.22,
|
| 12 |
+
"learning_rate": 8.333333333333334e-06,
|
| 13 |
+
"loss": 7866.0069,
|
| 14 |
+
"step": 50
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.43,
|
| 18 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 19 |
+
"loss": 4137.7106,
|
| 20 |
+
"step": 100
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.43,
|
| 24 |
+
"eval_loss": 3112.050537109375,
|
| 25 |
+
"eval_runtime": 179.5895,
|
| 26 |
+
"eval_samples_per_second": 9.355,
|
| 27 |
+
"eval_steps_per_second": 0.468,
|
| 28 |
+
"eval_wer": 1.0,
|
| 29 |
+
"step": 100
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.65,
|
| 33 |
+
"learning_rate": 2.5e-05,
|
| 34 |
+
"loss": 3114.5012,
|
| 35 |
+
"step": 150
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.87,
|
| 39 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 40 |
+
"loss": 2980.0003,
|
| 41 |
+
"step": 200
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.87,
|
| 45 |
+
"eval_loss": 2958.919921875,
|
| 46 |
+
"eval_runtime": 167.1595,
|
| 47 |
+
"eval_samples_per_second": 10.05,
|
| 48 |
+
"eval_steps_per_second": 0.503,
|
| 49 |
+
"eval_wer": 1.0,
|
| 50 |
+
"step": 200
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.08,
|
| 54 |
+
"learning_rate": 4.1666666666666665e-05,
|
| 55 |
+
"loss": 3155.9772,
|
| 56 |
+
"step": 250
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.3,
|
| 60 |
+
"learning_rate": 5e-05,
|
| 61 |
+
"loss": 2779.3762,
|
| 62 |
+
"step": 300
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.3,
|
| 66 |
+
"eval_loss": 2743.244140625,
|
| 67 |
+
"eval_runtime": 177.6493,
|
| 68 |
+
"eval_samples_per_second": 9.457,
|
| 69 |
+
"eval_steps_per_second": 0.473,
|
| 70 |
+
"eval_wer": 1.0,
|
| 71 |
+
"step": 300
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 1.52,
|
| 75 |
+
"learning_rate": 5.833333333333333e-05,
|
| 76 |
+
"loss": 2540.3578,
|
| 77 |
+
"step": 350
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.73,
|
| 81 |
+
"learning_rate": 6.666666666666667e-05,
|
| 82 |
+
"loss": 1841.6328,
|
| 83 |
+
"step": 400
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.73,
|
| 87 |
+
"eval_loss": 1341.6024169921875,
|
| 88 |
+
"eval_runtime": 181.145,
|
| 89 |
+
"eval_samples_per_second": 9.274,
|
| 90 |
+
"eval_steps_per_second": 0.464,
|
| 91 |
+
"eval_wer": 0.9295629466739967,
|
| 92 |
+
"step": 400
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.95,
|
| 96 |
+
"learning_rate": 7.5e-05,
|
| 97 |
+
"loss": 1248.4787,
|
| 98 |
+
"step": 450
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 2.16,
|
| 102 |
+
"learning_rate": 8.333333333333333e-05,
|
| 103 |
+
"loss": 1043.6277,
|
| 104 |
+
"step": 500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.16,
|
| 108 |
+
"eval_loss": 793.3690185546875,
|
| 109 |
+
"eval_runtime": 175.8809,
|
| 110 |
+
"eval_samples_per_second": 9.552,
|
| 111 |
+
"eval_steps_per_second": 0.478,
|
| 112 |
+
"eval_wer": 0.5867234744365035,
|
| 113 |
+
"step": 500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.38,
|
| 117 |
+
"learning_rate": 9.166666666666667e-05,
|
| 118 |
+
"loss": 827.3966,
|
| 119 |
+
"step": 550
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 2.6,
|
| 123 |
+
"learning_rate": 0.0001,
|
| 124 |
+
"loss": 834.6795,
|
| 125 |
+
"step": 600
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.6,
|
| 129 |
+
"eval_loss": 582.5656127929688,
|
| 130 |
+
"eval_runtime": 180.7161,
|
| 131 |
+
"eval_samples_per_second": 9.296,
|
| 132 |
+
"eval_steps_per_second": 0.465,
|
| 133 |
+
"eval_wer": 0.44612424409015944,
|
| 134 |
+
"step": 600
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.81,
|
| 138 |
+
"learning_rate": 0.00010833333333333334,
|
| 139 |
+
"loss": 757.1235,
|
| 140 |
+
"step": 650
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 3.03,
|
| 144 |
+
"learning_rate": 0.00011666666666666667,
|
| 145 |
+
"loss": 681.7599,
|
| 146 |
+
"step": 700
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 3.03,
|
| 150 |
+
"eval_loss": 502.809814453125,
|
| 151 |
+
"eval_runtime": 167.2696,
|
| 152 |
+
"eval_samples_per_second": 10.044,
|
| 153 |
+
"eval_steps_per_second": 0.502,
|
| 154 |
+
"eval_wer": 0.40702308960967565,
|
| 155 |
+
"step": 700
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.25,
|
| 159 |
+
"learning_rate": 0.000125,
|
| 160 |
+
"loss": 566.5348,
|
| 161 |
+
"step": 750
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 3.46,
|
| 165 |
+
"learning_rate": 0.00013333333333333334,
|
| 166 |
+
"loss": 499.7937,
|
| 167 |
+
"step": 800
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 3.46,
|
| 171 |
+
"eval_loss": 472.62139892578125,
|
| 172 |
+
"eval_runtime": 167.1792,
|
| 173 |
+
"eval_samples_per_second": 10.049,
|
| 174 |
+
"eval_steps_per_second": 0.502,
|
| 175 |
+
"eval_wer": 0.3989142385926333,
|
| 176 |
+
"step": 800
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 3.68,
|
| 180 |
+
"learning_rate": 0.00014166666666666668,
|
| 181 |
+
"loss": 436.0044,
|
| 182 |
+
"step": 850
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 3.9,
|
| 186 |
+
"learning_rate": 0.00015,
|
| 187 |
+
"loss": 392.5165,
|
| 188 |
+
"step": 900
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 3.9,
|
| 192 |
+
"eval_loss": 408.5908203125,
|
| 193 |
+
"eval_runtime": 174.8526,
|
| 194 |
+
"eval_samples_per_second": 9.608,
|
| 195 |
+
"eval_steps_per_second": 0.48,
|
| 196 |
+
"eval_wer": 0.34579439252336447,
|
| 197 |
+
"step": 900
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 4.11,
|
| 201 |
+
"learning_rate": 0.00015833333333333332,
|
| 202 |
+
"loss": 388.1201,
|
| 203 |
+
"step": 950
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 4.33,
|
| 207 |
+
"learning_rate": 0.00016666666666666666,
|
| 208 |
+
"loss": 351.2061,
|
| 209 |
+
"step": 1000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 4.33,
|
| 213 |
+
"eval_loss": 433.3687438964844,
|
| 214 |
+
"eval_runtime": 170.1742,
|
| 215 |
+
"eval_samples_per_second": 9.872,
|
| 216 |
+
"eval_steps_per_second": 0.494,
|
| 217 |
+
"eval_wer": 0.32449147883452445,
|
| 218 |
+
"step": 1000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 4.55,
|
| 222 |
+
"learning_rate": 0.000175,
|
| 223 |
+
"loss": 371.0807,
|
| 224 |
+
"step": 1050
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 4.76,
|
| 228 |
+
"learning_rate": 0.00018333333333333334,
|
| 229 |
+
"loss": 332.344,
|
| 230 |
+
"step": 1100
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 4.76,
|
| 234 |
+
"eval_loss": 412.85247802734375,
|
| 235 |
+
"eval_runtime": 189.4552,
|
| 236 |
+
"eval_samples_per_second": 8.868,
|
| 237 |
+
"eval_steps_per_second": 0.443,
|
| 238 |
+
"eval_wer": 0.32909565695437054,
|
| 239 |
+
"step": 1100
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 4.98,
|
| 243 |
+
"learning_rate": 0.00019166666666666667,
|
| 244 |
+
"loss": 332.4827,
|
| 245 |
+
"step": 1150
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 5.19,
|
| 249 |
+
"learning_rate": 0.0002,
|
| 250 |
+
"loss": 308.6846,
|
| 251 |
+
"step": 1200
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 5.19,
|
| 255 |
+
"eval_loss": 440.1231994628906,
|
| 256 |
+
"eval_runtime": 168.6145,
|
| 257 |
+
"eval_samples_per_second": 9.964,
|
| 258 |
+
"eval_steps_per_second": 0.498,
|
| 259 |
+
"eval_wer": 0.3194062671797691,
|
| 260 |
+
"step": 1200
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 5.41,
|
| 264 |
+
"learning_rate": 0.00020833333333333335,
|
| 265 |
+
"loss": 298.0805,
|
| 266 |
+
"step": 1250
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 5.63,
|
| 270 |
+
"learning_rate": 0.00021666666666666668,
|
| 271 |
+
"loss": 318.8863,
|
| 272 |
+
"step": 1300
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 5.63,
|
| 276 |
+
"eval_loss": 436.9075012207031,
|
| 277 |
+
"eval_runtime": 166.5693,
|
| 278 |
+
"eval_samples_per_second": 10.086,
|
| 279 |
+
"eval_steps_per_second": 0.504,
|
| 280 |
+
"eval_wer": 0.325109950522265,
|
| 281 |
+
"step": 1300
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 5.84,
|
| 285 |
+
"learning_rate": 0.00022500000000000002,
|
| 286 |
+
"loss": 317.8166,
|
| 287 |
+
"step": 1350
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 6.06,
|
| 291 |
+
"learning_rate": 0.00023333333333333333,
|
| 292 |
+
"loss": 303.7214,
|
| 293 |
+
"step": 1400
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 6.06,
|
| 297 |
+
"eval_loss": 427.9447937011719,
|
| 298 |
+
"eval_runtime": 201.1021,
|
| 299 |
+
"eval_samples_per_second": 8.354,
|
| 300 |
+
"eval_steps_per_second": 0.418,
|
| 301 |
+
"eval_wer": 0.32387300714678396,
|
| 302 |
+
"step": 1400
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 6.28,
|
| 306 |
+
"learning_rate": 0.00024166666666666667,
|
| 307 |
+
"loss": 291.1951,
|
| 308 |
+
"step": 1450
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 6.49,
|
| 312 |
+
"learning_rate": 0.00025,
|
| 313 |
+
"loss": 282.5848,
|
| 314 |
+
"step": 1500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 6.49,
|
| 318 |
+
"eval_loss": 426.2596130371094,
|
| 319 |
+
"eval_runtime": 183.3654,
|
| 320 |
+
"eval_samples_per_second": 9.162,
|
| 321 |
+
"eval_steps_per_second": 0.458,
|
| 322 |
+
"eval_wer": 0.30648708081363385,
|
| 323 |
+
"step": 1500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 6.71,
|
| 327 |
+
"learning_rate": 0.00025833333333333334,
|
| 328 |
+
"loss": 307.7221,
|
| 329 |
+
"step": 1550
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 6.93,
|
| 333 |
+
"learning_rate": 0.0002666666666666667,
|
| 334 |
+
"loss": 276.1861,
|
| 335 |
+
"step": 1600
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 6.93,
|
| 339 |
+
"eval_loss": 495.3029479980469,
|
| 340 |
+
"eval_runtime": 202.5189,
|
| 341 |
+
"eval_samples_per_second": 8.296,
|
| 342 |
+
"eval_steps_per_second": 0.415,
|
| 343 |
+
"eval_wer": 0.3147333699835074,
|
| 344 |
+
"step": 1600
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 7.14,
|
| 348 |
+
"learning_rate": 0.000275,
|
| 349 |
+
"loss": 267.0663,
|
| 350 |
+
"step": 1650
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 7.36,
|
| 354 |
+
"learning_rate": 0.00028333333333333335,
|
| 355 |
+
"loss": 268.4355,
|
| 356 |
+
"step": 1700
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 7.36,
|
| 360 |
+
"eval_loss": 458.63690185546875,
|
| 361 |
+
"eval_runtime": 177.1854,
|
| 362 |
+
"eval_samples_per_second": 9.482,
|
| 363 |
+
"eval_steps_per_second": 0.474,
|
| 364 |
+
"eval_wer": 0.32497251236943375,
|
| 365 |
+
"step": 1700
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 7.58,
|
| 369 |
+
"learning_rate": 0.0002916666666666667,
|
| 370 |
+
"loss": 280.6286,
|
| 371 |
+
"step": 1750
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 7.79,
|
| 375 |
+
"learning_rate": 0.0003,
|
| 376 |
+
"loss": 277.7815,
|
| 377 |
+
"step": 1800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 7.79,
|
| 381 |
+
"eval_loss": 540.147705078125,
|
| 382 |
+
"eval_runtime": 176.8625,
|
| 383 |
+
"eval_samples_per_second": 9.499,
|
| 384 |
+
"eval_steps_per_second": 0.475,
|
| 385 |
+
"eval_wer": 0.34819956019791093,
|
| 386 |
+
"step": 1800
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 8.01,
|
| 390 |
+
"learning_rate": 0.00030833333333333337,
|
| 391 |
+
"loss": 271.2937,
|
| 392 |
+
"step": 1850
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 8.23,
|
| 396 |
+
"learning_rate": 0.00031666666666666665,
|
| 397 |
+
"loss": 247.8558,
|
| 398 |
+
"step": 1900
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 8.23,
|
| 402 |
+
"eval_loss": 457.4529113769531,
|
| 403 |
+
"eval_runtime": 181.8308,
|
| 404 |
+
"eval_samples_per_second": 9.239,
|
| 405 |
+
"eval_steps_per_second": 0.462,
|
| 406 |
+
"eval_wer": 0.33981583287520617,
|
| 407 |
+
"step": 1900
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 8.44,
|
| 411 |
+
"learning_rate": 0.00032500000000000004,
|
| 412 |
+
"loss": 271.9148,
|
| 413 |
+
"step": 1950
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 8.66,
|
| 417 |
+
"learning_rate": 0.0003333333333333333,
|
| 418 |
+
"loss": 325.8626,
|
| 419 |
+
"step": 2000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 8.66,
|
| 423 |
+
"eval_loss": 450.06524658203125,
|
| 424 |
+
"eval_runtime": 162.9106,
|
| 425 |
+
"eval_samples_per_second": 10.312,
|
| 426 |
+
"eval_steps_per_second": 0.516,
|
| 427 |
+
"eval_wer": 0.3224986256184717,
|
| 428 |
+
"step": 2000
|
| 429 |
+
}
|
| 430 |
+
],
|
| 431 |
+
"max_steps": 2310,
|
| 432 |
+
"num_train_epochs": 10,
|
| 433 |
+
"total_flos": 1.1409490176804864e+18,
|
| 434 |
+
"trial_name": null,
|
| 435 |
+
"trial_params": null
|
| 436 |
+
}
|