Upload checkpoint-1000/trainer_state.json with huggingface_hub
Browse files
checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,367 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.0308293530178028,
|
| 3 |
+
"best_model_checkpoint": "./whisper-tiny-qlora/checkpoint-600",
|
| 4 |
+
"epoch": 0.6146281499692686,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 1000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.015365703749231715,
|
| 13 |
+
"grad_norm": 1.955202579498291,
|
| 14 |
+
"learning_rate": 4.800000000000001e-07,
|
| 15 |
+
"loss": 2.4438,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03073140749846343,
|
| 20 |
+
"grad_norm": 1.465623140335083,
|
| 21 |
+
"learning_rate": 9.800000000000001e-07,
|
| 22 |
+
"loss": 2.0995,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.046097111247695145,
|
| 27 |
+
"grad_norm": 1.489674687385559,
|
| 28 |
+
"learning_rate": 1.48e-06,
|
| 29 |
+
"loss": 2.4302,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06146281499692686,
|
| 34 |
+
"grad_norm": 1.7222563028335571,
|
| 35 |
+
"learning_rate": 1.98e-06,
|
| 36 |
+
"loss": 2.2989,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07682851874615858,
|
| 41 |
+
"grad_norm": 2.750751256942749,
|
| 42 |
+
"learning_rate": 2.4800000000000004e-06,
|
| 43 |
+
"loss": 2.2729,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.09219422249539029,
|
| 48 |
+
"grad_norm": 2.440593957901001,
|
| 49 |
+
"learning_rate": 2.9800000000000003e-06,
|
| 50 |
+
"loss": 2.0241,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.107559926244622,
|
| 55 |
+
"grad_norm": 2.425014019012451,
|
| 56 |
+
"learning_rate": 3.48e-06,
|
| 57 |
+
"loss": 2.6992,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.12292562999385372,
|
| 62 |
+
"grad_norm": 1.8003255128860474,
|
| 63 |
+
"learning_rate": 3.980000000000001e-06,
|
| 64 |
+
"loss": 2.0456,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.12292562999385372,
|
| 69 |
+
"eval_loss": 2.245432138442993,
|
| 70 |
+
"eval_runtime": 51.8391,
|
| 71 |
+
"eval_samples_per_second": 7.851,
|
| 72 |
+
"eval_steps_per_second": 0.251,
|
| 73 |
+
"eval_wer": 1.084672166739036,
|
| 74 |
+
"step": 200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.13829133374308544,
|
| 78 |
+
"grad_norm": 1.9000691175460815,
|
| 79 |
+
"learning_rate": 4.48e-06,
|
| 80 |
+
"loss": 2.611,
|
| 81 |
+
"step": 225
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.15365703749231716,
|
| 85 |
+
"grad_norm": 1.4311325550079346,
|
| 86 |
+
"learning_rate": 4.960000000000001e-06,
|
| 87 |
+
"loss": 2.4044,
|
| 88 |
+
"step": 250
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.16902274124154887,
|
| 92 |
+
"grad_norm": 2.4365923404693604,
|
| 93 |
+
"learning_rate": 5.460000000000001e-06,
|
| 94 |
+
"loss": 2.1992,
|
| 95 |
+
"step": 275
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.18438844499078058,
|
| 99 |
+
"grad_norm": 2.4958994388580322,
|
| 100 |
+
"learning_rate": 5.9600000000000005e-06,
|
| 101 |
+
"loss": 2.5628,
|
| 102 |
+
"step": 300
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.1997541487400123,
|
| 106 |
+
"grad_norm": 1.8009036779403687,
|
| 107 |
+
"learning_rate": 6.460000000000001e-06,
|
| 108 |
+
"loss": 1.8897,
|
| 109 |
+
"step": 325
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.215119852489244,
|
| 113 |
+
"grad_norm": 1.8635932207107544,
|
| 114 |
+
"learning_rate": 6.96e-06,
|
| 115 |
+
"loss": 2.2831,
|
| 116 |
+
"step": 350
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.23048555623847572,
|
| 120 |
+
"grad_norm": 1.8662347793579102,
|
| 121 |
+
"learning_rate": 7.4600000000000006e-06,
|
| 122 |
+
"loss": 2.0718,
|
| 123 |
+
"step": 375
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.24585125998770743,
|
| 127 |
+
"grad_norm": 2.4394640922546387,
|
| 128 |
+
"learning_rate": 7.960000000000002e-06,
|
| 129 |
+
"loss": 2.1359,
|
| 130 |
+
"step": 400
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.24585125998770743,
|
| 134 |
+
"eval_loss": 2.1913249492645264,
|
| 135 |
+
"eval_runtime": 49.901,
|
| 136 |
+
"eval_samples_per_second": 8.156,
|
| 137 |
+
"eval_steps_per_second": 0.261,
|
| 138 |
+
"eval_wer": 1.1096396005210596,
|
| 139 |
+
"step": 400
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.26121696373693915,
|
| 143 |
+
"grad_norm": 1.6949491500854492,
|
| 144 |
+
"learning_rate": 8.46e-06,
|
| 145 |
+
"loss": 2.0452,
|
| 146 |
+
"step": 425
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.2765826674861709,
|
| 150 |
+
"grad_norm": 1.273695468902588,
|
| 151 |
+
"learning_rate": 8.96e-06,
|
| 152 |
+
"loss": 2.0868,
|
| 153 |
+
"step": 450
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.29194837123540257,
|
| 157 |
+
"grad_norm": 2.592191457748413,
|
| 158 |
+
"learning_rate": 9.460000000000001e-06,
|
| 159 |
+
"loss": 2.028,
|
| 160 |
+
"step": 475
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.3073140749846343,
|
| 164 |
+
"grad_norm": 2.204875946044922,
|
| 165 |
+
"learning_rate": 9.960000000000001e-06,
|
| 166 |
+
"loss": 2.3593,
|
| 167 |
+
"step": 500
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.322679778733866,
|
| 171 |
+
"grad_norm": 2.3962159156799316,
|
| 172 |
+
"learning_rate": 9.54e-06,
|
| 173 |
+
"loss": 2.2455,
|
| 174 |
+
"step": 525
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.33804548248309774,
|
| 178 |
+
"grad_norm": 1.6109391450881958,
|
| 179 |
+
"learning_rate": 9.040000000000002e-06,
|
| 180 |
+
"loss": 2.1865,
|
| 181 |
+
"step": 550
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.3534111862323294,
|
| 185 |
+
"grad_norm": 1.053389310836792,
|
| 186 |
+
"learning_rate": 8.540000000000001e-06,
|
| 187 |
+
"loss": 2.0772,
|
| 188 |
+
"step": 575
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.36877688998156116,
|
| 192 |
+
"grad_norm": 4.344364166259766,
|
| 193 |
+
"learning_rate": 8.040000000000001e-06,
|
| 194 |
+
"loss": 1.7806,
|
| 195 |
+
"step": 600
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.36877688998156116,
|
| 199 |
+
"eval_loss": 2.0955584049224854,
|
| 200 |
+
"eval_runtime": 47.7981,
|
| 201 |
+
"eval_samples_per_second": 8.515,
|
| 202 |
+
"eval_steps_per_second": 0.272,
|
| 203 |
+
"eval_wer": 1.0308293530178028,
|
| 204 |
+
"step": 600
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.38414259373079285,
|
| 208 |
+
"grad_norm": 2.7162230014801025,
|
| 209 |
+
"learning_rate": 7.540000000000001e-06,
|
| 210 |
+
"loss": 1.9152,
|
| 211 |
+
"step": 625
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.3995082974800246,
|
| 215 |
+
"grad_norm": 3.225233793258667,
|
| 216 |
+
"learning_rate": 7.04e-06,
|
| 217 |
+
"loss": 2.2311,
|
| 218 |
+
"step": 650
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.4148740012292563,
|
| 222 |
+
"grad_norm": 3.202869176864624,
|
| 223 |
+
"learning_rate": 6.540000000000001e-06,
|
| 224 |
+
"loss": 2.0565,
|
| 225 |
+
"step": 675
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.430239704978488,
|
| 229 |
+
"grad_norm": 2.291895627975464,
|
| 230 |
+
"learning_rate": 6.040000000000001e-06,
|
| 231 |
+
"loss": 2.0733,
|
| 232 |
+
"step": 700
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.44560540872771975,
|
| 236 |
+
"grad_norm": 2.5822622776031494,
|
| 237 |
+
"learning_rate": 5.540000000000001e-06,
|
| 238 |
+
"loss": 1.8574,
|
| 239 |
+
"step": 725
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.46097111247695144,
|
| 243 |
+
"grad_norm": 4.14776611328125,
|
| 244 |
+
"learning_rate": 5.04e-06,
|
| 245 |
+
"loss": 2.0769,
|
| 246 |
+
"step": 750
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.4763368162261832,
|
| 250 |
+
"grad_norm": 4.232518196105957,
|
| 251 |
+
"learning_rate": 4.540000000000001e-06,
|
| 252 |
+
"loss": 2.1455,
|
| 253 |
+
"step": 775
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.49170251997541486,
|
| 257 |
+
"grad_norm": 4.926309585571289,
|
| 258 |
+
"learning_rate": 4.04e-06,
|
| 259 |
+
"loss": 2.1011,
|
| 260 |
+
"step": 800
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.49170251997541486,
|
| 264 |
+
"eval_loss": 2.02160906791687,
|
| 265 |
+
"eval_runtime": 49.8742,
|
| 266 |
+
"eval_samples_per_second": 8.161,
|
| 267 |
+
"eval_steps_per_second": 0.261,
|
| 268 |
+
"eval_wer": 1.138080764220582,
|
| 269 |
+
"step": 800
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.5070682237246465,
|
| 273 |
+
"grad_norm": 1.896835446357727,
|
| 274 |
+
"learning_rate": 3.54e-06,
|
| 275 |
+
"loss": 1.8221,
|
| 276 |
+
"step": 825
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.5224339274738783,
|
| 280 |
+
"grad_norm": 4.965891361236572,
|
| 281 |
+
"learning_rate": 3.04e-06,
|
| 282 |
+
"loss": 2.0295,
|
| 283 |
+
"step": 850
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.53779963122311,
|
| 287 |
+
"grad_norm": 2.1091415882110596,
|
| 288 |
+
"learning_rate": 2.5400000000000002e-06,
|
| 289 |
+
"loss": 1.8927,
|
| 290 |
+
"step": 875
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.5531653349723418,
|
| 294 |
+
"grad_norm": 4.7148590087890625,
|
| 295 |
+
"learning_rate": 2.04e-06,
|
| 296 |
+
"loss": 1.7901,
|
| 297 |
+
"step": 900
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.5685310387215734,
|
| 301 |
+
"grad_norm": 4.210736274719238,
|
| 302 |
+
"learning_rate": 1.54e-06,
|
| 303 |
+
"loss": 1.7876,
|
| 304 |
+
"step": 925
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.5838967424708051,
|
| 308 |
+
"grad_norm": 2.350623607635498,
|
| 309 |
+
"learning_rate": 1.04e-06,
|
| 310 |
+
"loss": 1.9305,
|
| 311 |
+
"step": 950
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.5992624462200369,
|
| 315 |
+
"grad_norm": 1.7559255361557007,
|
| 316 |
+
"learning_rate": 5.4e-07,
|
| 317 |
+
"loss": 1.8409,
|
| 318 |
+
"step": 975
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.6146281499692686,
|
| 322 |
+
"grad_norm": 2.374906301498413,
|
| 323 |
+
"learning_rate": 4e-08,
|
| 324 |
+
"loss": 2.1429,
|
| 325 |
+
"step": 1000
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.6146281499692686,
|
| 329 |
+
"eval_loss": 1.9964325428009033,
|
| 330 |
+
"eval_runtime": 50.5419,
|
| 331 |
+
"eval_samples_per_second": 8.053,
|
| 332 |
+
"eval_steps_per_second": 0.257,
|
| 333 |
+
"eval_wer": 1.150238818931828,
|
| 334 |
+
"step": 1000
|
| 335 |
+
}
|
| 336 |
+
],
|
| 337 |
+
"logging_steps": 25,
|
| 338 |
+
"max_steps": 1000,
|
| 339 |
+
"num_input_tokens_seen": 0,
|
| 340 |
+
"num_train_epochs": 1,
|
| 341 |
+
"save_steps": 200,
|
| 342 |
+
"stateful_callbacks": {
|
| 343 |
+
"EarlyStoppingCallback": {
|
| 344 |
+
"args": {
|
| 345 |
+
"early_stopping_patience": 3,
|
| 346 |
+
"early_stopping_threshold": 0.0
|
| 347 |
+
},
|
| 348 |
+
"attributes": {
|
| 349 |
+
"early_stopping_patience_counter": 0
|
| 350 |
+
}
|
| 351 |
+
},
|
| 352 |
+
"TrainerControl": {
|
| 353 |
+
"args": {
|
| 354 |
+
"should_epoch_stop": false,
|
| 355 |
+
"should_evaluate": false,
|
| 356 |
+
"should_log": false,
|
| 357 |
+
"should_save": true,
|
| 358 |
+
"should_training_stop": true
|
| 359 |
+
},
|
| 360 |
+
"attributes": {}
|
| 361 |
+
}
|
| 362 |
+
},
|
| 363 |
+
"total_flos": 3.084742656e+16,
|
| 364 |
+
"train_batch_size": 1,
|
| 365 |
+
"trial_name": null,
|
| 366 |
+
"trial_params": null
|
| 367 |
+
}
|