agentlans/multilingual-e5-small-aligned-quality
024
1{2 "best_metric": 0.19583497941493988,3 "best_model_checkpoint": "multilingual-e5-small-aligned-quality-20241214-new/checkpoint-23439",4 "epoch": 3.0,5 "eval_steps": 500,6 "global_step": 23439,7 "is_hyper_param_search": false,8 "is_local_process_zero": true,9 "is_world_process_zero": true,10 "log_history": [11 {12 "epoch": 0.06399590426212723,13 "grad_norm": 1.3940926790237427,14 "learning_rate": 4.8933401595631215e-05,15 "loss": 0.3708,16 "step": 50017 },18 {19 "epoch": 0.12799180852425446,20 "grad_norm": 1.3164088726043701,21 "learning_rate": 4.786680319126243e-05,22 "loss": 0.3132,23 "step": 100024 },25 {26 "epoch": 0.19198771278638166,27 "grad_norm": 1.3515187501907349,28 "learning_rate": 4.680020478689364e-05,29 "loss": 0.2973,30 "step": 150031 },32 {33 "epoch": 0.2559836170485089,34 "grad_norm": 1.7389737367630005,35 "learning_rate": 4.573360638252485e-05,36 "loss": 0.2877,37 "step": 200038 },39 {40 "epoch": 0.3199795213106361,41 "grad_norm": 1.2677313089370728,42 "learning_rate": 4.4667007978156063e-05,43 "loss": 0.2803,44 "step": 250045 },46 {47 "epoch": 0.3839754255727633,48 "grad_norm": 1.3927785158157349,49 "learning_rate": 4.360040957378728e-05,50 "loss": 0.2743,51 "step": 300052 },53 {54 "epoch": 0.4479713298348906,55 "grad_norm": 1.628334879875183,56 "learning_rate": 4.2533811169418495e-05,57 "loss": 0.2747,58 "step": 350059 },60 {61 "epoch": 0.5119672340970178,62 "grad_norm": 1.4009227752685547,63 "learning_rate": 4.146721276504971e-05,64 "loss": 0.2672,65 "step": 400066 },67 {68 "epoch": 0.575963138359145,69 "grad_norm": 1.4522840976715088,70 "learning_rate": 4.040061436068092e-05,71 "loss": 0.264,72 "step": 450073 },74 {75 "epoch": 0.6399590426212722,76 "grad_norm": 1.5413362979888916,77 "learning_rate": 3.933401595631213e-05,78 "loss": 0.2588,79 "step": 500080 },81 {82 "epoch": 0.7039549468833994,83 "grad_norm": 1.2421332597732544,84 "learning_rate": 3.8267417551943344e-05,85 "loss": 0.253,86 "step": 550087 },88 {89 "epoch": 0.7679508511455266,90 "grad_norm": 1.3509016036987305,91 "learning_rate": 3.7200819147574556e-05,92 "loss": 0.2539,93 "step": 600094 },95 {96 "epoch": 0.831946755407654,97 "grad_norm": 1.9007371664047241,98 "learning_rate": 3.613422074320577e-05,99 "loss": 0.2516,100 "step": 6500101 },102 {103 "epoch": 0.8959426596697811,104 "grad_norm": 1.3137174844741821,105 "learning_rate": 3.506762233883698e-05,106 "loss": 0.2464,107 "step": 7000108 },109 {110 "epoch": 0.9599385639319084,111 "grad_norm": 1.5280767679214478,112 "learning_rate": 3.400102393446819e-05,113 "loss": 0.2436,114 "step": 7500115 },116 {117 "epoch": 1.0,118 "eval_loss": 0.2296382337808609,119 "eval_mse": 0.22963822030546574,120 "eval_runtime": 103.0849,121 "eval_samples_per_second": 1766.611,122 "eval_steps_per_second": 220.828,123 "step": 7813124 },125 {126 "epoch": 1.0239344681940357,127 "grad_norm": 1.2579801082611084,128 "learning_rate": 3.293442553009941e-05,129 "loss": 0.2307,130 "step": 8000131 },132 {133 "epoch": 1.0879303724561629,134 "grad_norm": 1.494954228401184,135 "learning_rate": 3.1867827125730624e-05,136 "loss": 0.2053,137 "step": 8500138 },139 {140 "epoch": 1.15192627671829,141 "grad_norm": 1.6762491464614868,142 "learning_rate": 3.0801228721361836e-05,143 "loss": 0.2021,144 "step": 9000145 },146 {147 "epoch": 1.2159221809804173,148 "grad_norm": 1.3708257675170898,149 "learning_rate": 2.9734630316993045e-05,150 "loss": 0.2034,151 "step": 9500152 },153 {154 "epoch": 1.2799180852425445,155 "grad_norm": 1.57413649559021,156 "learning_rate": 2.866803191262426e-05,157 "loss": 0.2016,158 "step": 10000159 },160 {161 "epoch": 1.3439139895046717,162 "grad_norm": 1.1981782913208008,163 "learning_rate": 2.7601433508255476e-05,164 "loss": 0.2014,165 "step": 10500166 },167 {168 "epoch": 1.4079098937667989,169 "grad_norm": 2.0707337856292725,170 "learning_rate": 2.6534835103886685e-05,171 "loss": 0.1996,172 "step": 11000173 },174 {175 "epoch": 1.471905798028926,176 "grad_norm": 1.3037790060043335,177 "learning_rate": 2.54682366995179e-05,178 "loss": 0.1998,179 "step": 11500180 },181 {182 "epoch": 1.5359017022910533,183 "grad_norm": 1.331900715827942,184 "learning_rate": 2.4401638295149112e-05,185 "loss": 0.1995,186 "step": 12000187 },188 {189 "epoch": 1.5998976065531805,190 "grad_norm": 1.6664810180664062,191 "learning_rate": 2.3335039890780325e-05,192 "loss": 0.1995,193 "step": 12500194 },195 {196 "epoch": 1.6638935108153077,197 "grad_norm": 1.518689751625061,198 "learning_rate": 2.2268441486411537e-05,199 "loss": 0.1981,200 "step": 13000201 },202 {203 "epoch": 1.727889415077435,204 "grad_norm": 1.4883790016174316,205 "learning_rate": 2.120184308204275e-05,206 "loss": 0.1954,207 "step": 13500208 },209 {210 "epoch": 1.7918853193395623,211 "grad_norm": 1.3662298917770386,212 "learning_rate": 2.0135244677673965e-05,213 "loss": 0.1957,214 "step": 14000215 },216 {217 "epoch": 1.8558812236016895,218 "grad_norm": 1.4358400106430054,219 "learning_rate": 1.9068646273305177e-05,220 "loss": 0.1943,221 "step": 14500222 },223 {224 "epoch": 1.9198771278638167,225 "grad_norm": 1.3663334846496582,226 "learning_rate": 1.800204786893639e-05,227 "loss": 0.1959,228 "step": 15000229 },230 {231 "epoch": 1.983873032125944,232 "grad_norm": 1.3956917524337769,233 "learning_rate": 1.69354494645676e-05,234 "loss": 0.1927,235 "step": 15500236 },237 {238 "epoch": 2.0,239 "eval_loss": 0.2078620195388794,240 "eval_mse": 0.20786202507040266,241 "eval_runtime": 99.6454,242 "eval_samples_per_second": 1827.591,243 "eval_steps_per_second": 228.45,244 "step": 15626245 },246 {247 "epoch": 2.0478689363880713,248 "grad_norm": 1.112040638923645,249 "learning_rate": 1.5868851060198814e-05,250 "loss": 0.1722,251 "step": 16000252 },253 {254 "epoch": 2.1118648406501985,255 "grad_norm": 1.627584457397461,256 "learning_rate": 1.480225265583003e-05,257 "loss": 0.1648,258 "step": 16500259 },260 {261 "epoch": 2.1758607449123257,262 "grad_norm": 1.6189327239990234,263 "learning_rate": 1.3735654251461241e-05,264 "loss": 0.1637,265 "step": 17000266 },267 {268 "epoch": 2.239856649174453,269 "grad_norm": 1.2161808013916016,270 "learning_rate": 1.2669055847092454e-05,271 "loss": 0.1644,272 "step": 17500273 },274 {275 "epoch": 2.30385255343658,276 "grad_norm": 1.4423365592956543,277 "learning_rate": 1.1602457442723666e-05,278 "loss": 0.163,279 "step": 18000280 },281 {282 "epoch": 2.3678484576987073,283 "grad_norm": 1.298816204071045,284 "learning_rate": 1.053585903835488e-05,285 "loss": 0.1626,286 "step": 18500287 },288 {289 "epoch": 2.4318443619608345,290 "grad_norm": 1.2976384162902832,291 "learning_rate": 9.469260633986092e-06,292 "loss": 0.1627,293 "step": 19000294 },295 {296 "epoch": 2.4958402662229617,297 "grad_norm": 1.39438796043396,298 "learning_rate": 8.402662229617304e-06,299 "loss": 0.1633,300 "step": 19500301 },302 {303 "epoch": 2.559836170485089,304 "grad_norm": 1.342234492301941,305 "learning_rate": 7.336063825248518e-06,306 "loss": 0.162,307 "step": 20000308 },309 {310 "epoch": 2.623832074747216,311 "grad_norm": 1.3450177907943726,312 "learning_rate": 6.26946542087973e-06,313 "loss": 0.1619,314 "step": 20500315 },316 {317 "epoch": 2.6878279790093433,318 "grad_norm": 1.3149842023849487,319 "learning_rate": 5.202867016510943e-06,320 "loss": 0.1608,321 "step": 21000322 },323 {324 "epoch": 2.7518238832714705,325 "grad_norm": 1.3171745538711548,326 "learning_rate": 4.1362686121421564e-06,327 "loss": 0.1622,328 "step": 21500329 },330 {331 "epoch": 2.8158197875335977,332 "grad_norm": 1.0992498397827148,333 "learning_rate": 3.069670207773369e-06,334 "loss": 0.1608,335 "step": 22000336 },337 {338 "epoch": 2.879815691795725,339 "grad_norm": 1.3956327438354492,340 "learning_rate": 2.003071803404582e-06,341 "loss": 0.161,342 "step": 22500343 },344 {345 "epoch": 2.943811596057852,346 "grad_norm": 1.531733512878418,347 "learning_rate": 9.364733990357951e-07,348 "loss": 0.1615,349 "step": 23000350 },351 {352 "epoch": 3.0,353 "eval_loss": 0.19583497941493988,354 "eval_mse": 0.19583499065839644,355 "eval_runtime": 99.9929,356 "eval_samples_per_second": 1821.239,357 "eval_steps_per_second": 227.656,358 "step": 23439359 },360 {361 "epoch": 3.0,362 "step": 23439,363 "total_flos": 4.9403660544e+16,364 "train_loss": 0.21201796470442558,365 "train_runtime": 3304.0326,366 "train_samples_per_second": 907.981,367 "train_steps_per_second": 7.094368 }369 ],370 "logging_steps": 500,371 "max_steps": 23439,372 "num_input_tokens_seen": 0,373 "num_train_epochs": 3,374 "save_steps": 500,375 "stateful_callbacks": {376 "TrainerControl": {377 "args": {378 "should_epoch_stop": false,379 "should_evaluate": false,380 "should_log": false,381 "should_save": true,382 "should_training_stop": true383 },384 "attributes": {}385 }386 },387 "total_flos": 4.9403660544e+16,388 "train_batch_size": 128,389 "trial_name": null,390 "trial_params": null391}392 