Training in progress, epoch 3, checkpoint
Browse files
last-checkpoint/.ipynb_checkpoints/trainer_state-checkpoint.json
ADDED
|
@@ -0,0 +1,333 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.9972065305108946,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 1004,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1986467192252778,
|
| 13 |
+
"grad_norm": 16.624483108520508,
|
| 14 |
+
"learning_rate": 9.800796812749005e-05,
|
| 15 |
+
"loss": 223.0731,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1986467192252778,
|
| 20 |
+
"eval_loss": 218.2069549560547,
|
| 21 |
+
"eval_runtime": 20.7815,
|
| 22 |
+
"eval_samples_per_second": 40.806,
|
| 23 |
+
"eval_steps_per_second": 20.403,
|
| 24 |
+
"step": 50
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.3972934384505556,
|
| 28 |
+
"grad_norm": 22.61164665222168,
|
| 29 |
+
"learning_rate": 9.601593625498009e-05,
|
| 30 |
+
"loss": 219.2387,
|
| 31 |
+
"step": 100
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.3972934384505556,
|
| 35 |
+
"eval_loss": 214.27130126953125,
|
| 36 |
+
"eval_runtime": 20.7861,
|
| 37 |
+
"eval_samples_per_second": 40.797,
|
| 38 |
+
"eval_steps_per_second": 20.398,
|
| 39 |
+
"step": 100
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.5959401576758334,
|
| 43 |
+
"grad_norm": 39.08198928833008,
|
| 44 |
+
"learning_rate": 9.402390438247013e-05,
|
| 45 |
+
"loss": 206.0933,
|
| 46 |
+
"step": 150
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.5959401576758334,
|
| 50 |
+
"eval_loss": 208.94505310058594,
|
| 51 |
+
"eval_runtime": 20.7825,
|
| 52 |
+
"eval_samples_per_second": 40.804,
|
| 53 |
+
"eval_steps_per_second": 20.402,
|
| 54 |
+
"step": 150
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.7945868769011112,
|
| 58 |
+
"grad_norm": 49.450714111328125,
|
| 59 |
+
"learning_rate": 9.203187250996016e-05,
|
| 60 |
+
"loss": 207.5961,
|
| 61 |
+
"step": 200
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.7945868769011112,
|
| 65 |
+
"eval_loss": 200.57440185546875,
|
| 66 |
+
"eval_runtime": 20.7763,
|
| 67 |
+
"eval_samples_per_second": 40.816,
|
| 68 |
+
"eval_steps_per_second": 20.408,
|
| 69 |
+
"step": 200
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.993233596126389,
|
| 73 |
+
"grad_norm": 88.86530303955078,
|
| 74 |
+
"learning_rate": 9.00398406374502e-05,
|
| 75 |
+
"loss": 197.4489,
|
| 76 |
+
"step": 250
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.993233596126389,
|
| 80 |
+
"eval_loss": 193.4808807373047,
|
| 81 |
+
"eval_runtime": 20.7884,
|
| 82 |
+
"eval_samples_per_second": 40.792,
|
| 83 |
+
"eval_steps_per_second": 20.396,
|
| 84 |
+
"step": 250
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.1946737848407722,
|
| 88 |
+
"grad_norm": 94.41268157958984,
|
| 89 |
+
"learning_rate": 8.804780876494024e-05,
|
| 90 |
+
"loss": 186.8238,
|
| 91 |
+
"step": 300
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 1.1946737848407722,
|
| 95 |
+
"eval_loss": 183.36685180664062,
|
| 96 |
+
"eval_runtime": 20.7922,
|
| 97 |
+
"eval_samples_per_second": 40.785,
|
| 98 |
+
"eval_steps_per_second": 20.392,
|
| 99 |
+
"step": 300
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.39332050406605,
|
| 103 |
+
"grad_norm": 115.2780532836914,
|
| 104 |
+
"learning_rate": 8.605577689243029e-05,
|
| 105 |
+
"loss": 176.8817,
|
| 106 |
+
"step": 350
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.39332050406605,
|
| 110 |
+
"eval_loss": 167.37599182128906,
|
| 111 |
+
"eval_runtime": 20.7875,
|
| 112 |
+
"eval_samples_per_second": 40.794,
|
| 113 |
+
"eval_steps_per_second": 20.397,
|
| 114 |
+
"step": 350
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.5919672232913278,
|
| 118 |
+
"grad_norm": 108.17979431152344,
|
| 119 |
+
"learning_rate": 8.406374501992032e-05,
|
| 120 |
+
"loss": 165.4761,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.5919672232913278,
|
| 125 |
+
"eval_loss": 148.36529541015625,
|
| 126 |
+
"eval_runtime": 20.7823,
|
| 127 |
+
"eval_samples_per_second": 40.804,
|
| 128 |
+
"eval_steps_per_second": 20.402,
|
| 129 |
+
"step": 400
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.7906139425166057,
|
| 133 |
+
"grad_norm": 181.7048797607422,
|
| 134 |
+
"learning_rate": 8.207171314741037e-05,
|
| 135 |
+
"loss": 149.9219,
|
| 136 |
+
"step": 450
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.7906139425166057,
|
| 140 |
+
"eval_loss": 144.7928924560547,
|
| 141 |
+
"eval_runtime": 20.7791,
|
| 142 |
+
"eval_samples_per_second": 40.81,
|
| 143 |
+
"eval_steps_per_second": 20.405,
|
| 144 |
+
"step": 450
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.9892606617418833,
|
| 148 |
+
"grad_norm": 637.717529296875,
|
| 149 |
+
"learning_rate": 8.00796812749004e-05,
|
| 150 |
+
"loss": 127.9147,
|
| 151 |
+
"step": 500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.9892606617418833,
|
| 155 |
+
"eval_loss": 113.41303253173828,
|
| 156 |
+
"eval_runtime": 20.7782,
|
| 157 |
+
"eval_samples_per_second": 40.812,
|
| 158 |
+
"eval_steps_per_second": 20.406,
|
| 159 |
+
"step": 500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.190700850456267,
|
| 163 |
+
"grad_norm": 1505.125244140625,
|
| 164 |
+
"learning_rate": 7.808764940239044e-05,
|
| 165 |
+
"loss": 106.9943,
|
| 166 |
+
"step": 550
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 2.190700850456267,
|
| 170 |
+
"eval_loss": 105.97330474853516,
|
| 171 |
+
"eval_runtime": 20.7864,
|
| 172 |
+
"eval_samples_per_second": 40.796,
|
| 173 |
+
"eval_steps_per_second": 20.398,
|
| 174 |
+
"step": 550
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.3893475696815445,
|
| 178 |
+
"grad_norm": 95.92086791992188,
|
| 179 |
+
"learning_rate": 7.609561752988048e-05,
|
| 180 |
+
"loss": 88.9087,
|
| 181 |
+
"step": 600
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.3893475696815445,
|
| 185 |
+
"eval_loss": 75.11954498291016,
|
| 186 |
+
"eval_runtime": 20.7814,
|
| 187 |
+
"eval_samples_per_second": 40.806,
|
| 188 |
+
"eval_steps_per_second": 20.403,
|
| 189 |
+
"step": 600
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 2.587994288906822,
|
| 193 |
+
"grad_norm": 484.91204833984375,
|
| 194 |
+
"learning_rate": 7.410358565737052e-05,
|
| 195 |
+
"loss": 95.5976,
|
| 196 |
+
"step": 650
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 2.587994288906822,
|
| 200 |
+
"eval_loss": 86.23988342285156,
|
| 201 |
+
"eval_runtime": 20.7754,
|
| 202 |
+
"eval_samples_per_second": 40.817,
|
| 203 |
+
"eval_steps_per_second": 20.409,
|
| 204 |
+
"step": 650
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.7866410081321,
|
| 208 |
+
"grad_norm": 46.8211669921875,
|
| 209 |
+
"learning_rate": 7.211155378486057e-05,
|
| 210 |
+
"loss": 83.614,
|
| 211 |
+
"step": 700
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 2.7866410081321,
|
| 215 |
+
"eval_loss": 63.02427673339844,
|
| 216 |
+
"eval_runtime": 20.7817,
|
| 217 |
+
"eval_samples_per_second": 40.805,
|
| 218 |
+
"eval_steps_per_second": 20.403,
|
| 219 |
+
"step": 700
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.985287727357378,
|
| 223 |
+
"grad_norm": 122.56575775146484,
|
| 224 |
+
"learning_rate": 7.01195219123506e-05,
|
| 225 |
+
"loss": 78.3681,
|
| 226 |
+
"step": 750
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 2.985287727357378,
|
| 230 |
+
"eval_loss": 36.502681732177734,
|
| 231 |
+
"eval_runtime": 20.778,
|
| 232 |
+
"eval_samples_per_second": 40.812,
|
| 233 |
+
"eval_steps_per_second": 20.406,
|
| 234 |
+
"step": 750
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 3.186727916071761,
|
| 238 |
+
"grad_norm": 352.10821533203125,
|
| 239 |
+
"learning_rate": 6.812749003984064e-05,
|
| 240 |
+
"loss": 38.2802,
|
| 241 |
+
"step": 800
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 3.186727916071761,
|
| 245 |
+
"eval_loss": 35.01713562011719,
|
| 246 |
+
"eval_runtime": 20.7907,
|
| 247 |
+
"eval_samples_per_second": 40.788,
|
| 248 |
+
"eval_steps_per_second": 20.394,
|
| 249 |
+
"step": 800
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 3.385374635297039,
|
| 253 |
+
"grad_norm": 97.40475463867188,
|
| 254 |
+
"learning_rate": 6.613545816733068e-05,
|
| 255 |
+
"loss": 27.1773,
|
| 256 |
+
"step": 850
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 3.385374635297039,
|
| 260 |
+
"eval_loss": 22.674545288085938,
|
| 261 |
+
"eval_runtime": 20.7869,
|
| 262 |
+
"eval_samples_per_second": 40.795,
|
| 263 |
+
"eval_steps_per_second": 20.397,
|
| 264 |
+
"step": 850
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.5840213545223167,
|
| 268 |
+
"grad_norm": 106.43690490722656,
|
| 269 |
+
"learning_rate": 6.414342629482072e-05,
|
| 270 |
+
"loss": 26.6469,
|
| 271 |
+
"step": 900
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 3.5840213545223167,
|
| 275 |
+
"eval_loss": 10.793374061584473,
|
| 276 |
+
"eval_runtime": 20.7814,
|
| 277 |
+
"eval_samples_per_second": 40.806,
|
| 278 |
+
"eval_steps_per_second": 20.403,
|
| 279 |
+
"step": 900
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 3.7826680737475944,
|
| 283 |
+
"grad_norm": 121.86284637451172,
|
| 284 |
+
"learning_rate": 6.215139442231077e-05,
|
| 285 |
+
"loss": 4.4939,
|
| 286 |
+
"step": 950
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 3.7826680737475944,
|
| 290 |
+
"eval_loss": -3.3120968341827393,
|
| 291 |
+
"eval_runtime": 20.7762,
|
| 292 |
+
"eval_samples_per_second": 40.816,
|
| 293 |
+
"eval_steps_per_second": 20.408,
|
| 294 |
+
"step": 950
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 3.9813147929728725,
|
| 298 |
+
"grad_norm": 148.8233642578125,
|
| 299 |
+
"learning_rate": 6.01593625498008e-05,
|
| 300 |
+
"loss": 19.1452,
|
| 301 |
+
"step": 1000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 3.9813147929728725,
|
| 305 |
+
"eval_loss": -23.70981788635254,
|
| 306 |
+
"eval_runtime": 20.7822,
|
| 307 |
+
"eval_samples_per_second": 40.804,
|
| 308 |
+
"eval_steps_per_second": 20.402,
|
| 309 |
+
"step": 1000
|
| 310 |
+
}
|
| 311 |
+
],
|
| 312 |
+
"logging_steps": 50,
|
| 313 |
+
"max_steps": 2510,
|
| 314 |
+
"num_input_tokens_seen": 0,
|
| 315 |
+
"num_train_epochs": 10,
|
| 316 |
+
"save_steps": 500,
|
| 317 |
+
"stateful_callbacks": {
|
| 318 |
+
"TrainerControl": {
|
| 319 |
+
"args": {
|
| 320 |
+
"should_epoch_stop": false,
|
| 321 |
+
"should_evaluate": false,
|
| 322 |
+
"should_log": false,
|
| 323 |
+
"should_save": true,
|
| 324 |
+
"should_training_stop": false
|
| 325 |
+
},
|
| 326 |
+
"attributes": {}
|
| 327 |
+
}
|
| 328 |
+
},
|
| 329 |
+
"total_flos": 0.0,
|
| 330 |
+
"train_batch_size": 2,
|
| 331 |
+
"trial_name": null,
|
| 332 |
+
"trial_params": null
|
| 333 |
+
}
|
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2610104820
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6549006c4278533d56c4a872630e58d454236c45e6b6f37240a67cc4a8d1149f
|
| 3 |
size 2610104820
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5210004271
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7ef81cbae86f3da4764bec2f9b3bc85de611d64cc8367d35920927dc54f9266
|
| 3 |
size 5210004271
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15006
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d3dc11eb7ffb894aa9af0579bc9e38efeb9cdc6220ba95d6dbad8bccb8be011a
|
| 3 |
size 15006
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fca60d3e211c90170c539f1ba6815cff92dba636daaf20fea2fac1d2459b3c9f
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 50,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -232,6 +232,81 @@
|
|
| 232 |
"eval_samples_per_second": 31.338,
|
| 233 |
"eval_steps_per_second": 15.669,
|
| 234 |
"step": 750
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 235 |
}
|
| 236 |
],
|
| 237 |
"logging_steps": 50,
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.9972065305108946,
|
| 5 |
"eval_steps": 50,
|
| 6 |
+
"global_step": 1004,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 232 |
"eval_samples_per_second": 31.338,
|
| 233 |
"eval_steps_per_second": 15.669,
|
| 234 |
"step": 750
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 3.186727916071761,
|
| 238 |
+
"grad_norm": 0.14813293516635895,
|
| 239 |
+
"learning_rate": 6.812749003984064e-05,
|
| 240 |
+
"loss": 0.0945,
|
| 241 |
+
"step": 800
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 3.186727916071761,
|
| 245 |
+
"eval_loss": 0.14082284271717072,
|
| 246 |
+
"eval_runtime": 27.0281,
|
| 247 |
+
"eval_samples_per_second": 31.375,
|
| 248 |
+
"eval_steps_per_second": 15.687,
|
| 249 |
+
"step": 800
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 3.385374635297039,
|
| 253 |
+
"grad_norm": 0.14571304619312286,
|
| 254 |
+
"learning_rate": 6.613545816733068e-05,
|
| 255 |
+
"loss": 0.085,
|
| 256 |
+
"step": 850
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 3.385374635297039,
|
| 260 |
+
"eval_loss": 0.13174280524253845,
|
| 261 |
+
"eval_runtime": 27.0267,
|
| 262 |
+
"eval_samples_per_second": 31.376,
|
| 263 |
+
"eval_steps_per_second": 15.688,
|
| 264 |
+
"step": 850
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.5840213545223167,
|
| 268 |
+
"grad_norm": 0.15489640831947327,
|
| 269 |
+
"learning_rate": 6.414342629482072e-05,
|
| 270 |
+
"loss": 0.0774,
|
| 271 |
+
"step": 900
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 3.5840213545223167,
|
| 275 |
+
"eval_loss": 0.12704746425151825,
|
| 276 |
+
"eval_runtime": 27.0289,
|
| 277 |
+
"eval_samples_per_second": 31.374,
|
| 278 |
+
"eval_steps_per_second": 15.687,
|
| 279 |
+
"step": 900
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 3.7826680737475944,
|
| 283 |
+
"grad_norm": 0.15014521777629852,
|
| 284 |
+
"learning_rate": 6.215139442231077e-05,
|
| 285 |
+
"loss": 0.0761,
|
| 286 |
+
"step": 950
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 3.7826680737475944,
|
| 290 |
+
"eval_loss": 0.12169376760721207,
|
| 291 |
+
"eval_runtime": 27.0479,
|
| 292 |
+
"eval_samples_per_second": 31.352,
|
| 293 |
+
"eval_steps_per_second": 15.676,
|
| 294 |
+
"step": 950
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 3.9813147929728725,
|
| 298 |
+
"grad_norm": 0.13176779448986053,
|
| 299 |
+
"learning_rate": 6.01593625498008e-05,
|
| 300 |
+
"loss": 0.071,
|
| 301 |
+
"step": 1000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 3.9813147929728725,
|
| 305 |
+
"eval_loss": 0.1125979870557785,
|
| 306 |
+
"eval_runtime": 27.0251,
|
| 307 |
+
"eval_samples_per_second": 31.378,
|
| 308 |
+
"eval_steps_per_second": 15.689,
|
| 309 |
+
"step": 1000
|
| 310 |
}
|
| 311 |
],
|
| 312 |
"logging_steps": 50,
|