thangcd/thread-intelligence-classifier
016
1{2 "best_global_step": 338,3 "best_metric": 0.9925062447960034,4 "best_model_checkpoint": "./classifier_model_refined/checkpoint-338",5 "epoch": 3.0,6 "eval_steps": 500,7 "global_step": 1014,8 "is_hyper_param_search": false,9 "is_local_process_zero": true,10 "is_world_process_zero": true,11 "log_history": [12 {13 "epoch": 0.029585798816568046,14 "grad_norm": 12.594921112060547,15 "learning_rate": 9.911242603550295e-07,16 "loss": 0.1235574722290039,17 "step": 1018 },19 {20 "epoch": 0.05917159763313609,21 "grad_norm": 32.30978775024414,22 "learning_rate": 9.812623274161736e-07,23 "loss": 0.11867349147796631,24 "step": 2025 },26 {27 "epoch": 0.08875739644970414,28 "grad_norm": 0.43152111768722534,29 "learning_rate": 9.714003944773175e-07,30 "loss": 0.06737754344940186,31 "step": 3032 },33 {34 "epoch": 0.11834319526627218,35 "grad_norm": 6.851346969604492,36 "learning_rate": 9.615384615384615e-07,37 "loss": 0.037985888123512265,38 "step": 4039 },40 {41 "epoch": 0.14792899408284024,42 "grad_norm": 0.24064384400844574,43 "learning_rate": 9.516765285996054e-07,44 "loss": 0.08224716782569885,45 "step": 5046 },47 {48 "epoch": 0.17751479289940827,49 "grad_norm": 0.38599199056625366,50 "learning_rate": 9.418145956607495e-07,51 "loss": 0.0987737774848938,52 "step": 6053 },54 {55 "epoch": 0.20710059171597633,56 "grad_norm": 0.5154857039451599,57 "learning_rate": 9.319526627218934e-07,58 "loss": 0.03345215916633606,59 "step": 7060 },61 {62 "epoch": 0.23668639053254437,63 "grad_norm": 1.0237733125686646,64 "learning_rate": 9.220907297830375e-07,65 "loss": 0.08602827787399292,66 "step": 8067 },68 {69 "epoch": 0.26627218934911245,70 "grad_norm": 0.10272961854934692,71 "learning_rate": 9.122287968441814e-07,72 "loss": 0.05293607711791992,73 "step": 9074 },75 {76 "epoch": 0.2958579881656805,77 "grad_norm": 5.297066688537598,78 "learning_rate": 9.023668639053253e-07,79 "loss": 0.04965598285198212,80 "step": 10081 },82 {83 "epoch": 0.3254437869822485,84 "grad_norm": 0.5690721273422241,85 "learning_rate": 8.925049309664694e-07,86 "loss": 0.10500458478927613,87 "step": 11088 },89 {90 "epoch": 0.35502958579881655,91 "grad_norm": 0.13593363761901855,92 "learning_rate": 8.826429980276133e-07,93 "loss": 0.02019917368888855,94 "step": 12095 },96 {97 "epoch": 0.38461538461538464,98 "grad_norm": 10.624679565429688,99 "learning_rate": 8.727810650887574e-07,100 "loss": 0.07912254929542542,101 "step": 130102 },103 {104 "epoch": 0.41420118343195267,105 "grad_norm": 0.2176271677017212,106 "learning_rate": 8.629191321499013e-07,107 "loss": 0.06359128355979919,108 "step": 140109 },110 {111 "epoch": 0.4437869822485207,112 "grad_norm": 0.09295513480901718,113 "learning_rate": 8.530571992110453e-07,114 "loss": 0.07846295237541198,115 "step": 150116 },117 {118 "epoch": 0.47337278106508873,119 "grad_norm": 0.03902903571724892,120 "learning_rate": 8.431952662721893e-07,121 "loss": 0.0552996039390564,122 "step": 160123 },124 {125 "epoch": 0.5029585798816568,126 "grad_norm": 12.510440826416016,127 "learning_rate": 8.333333333333333e-07,128 "loss": 0.044957560300827024,129 "step": 170130 },131 {132 "epoch": 0.5325443786982249,133 "grad_norm": 1.6486430168151855,134 "learning_rate": 8.234714003944773e-07,135 "loss": 0.08355305790901184,136 "step": 180137 },138 {139 "epoch": 0.5621301775147929,140 "grad_norm": 0.09395692497491837,141 "learning_rate": 8.136094674556213e-07,142 "loss": 0.0977919340133667,143 "step": 190144 },145 {146 "epoch": 0.591715976331361,147 "grad_norm": 0.14703874289989471,148 "learning_rate": 8.037475345167653e-07,149 "loss": 0.06359317898750305,150 "step": 200151 },152 {153 "epoch": 0.621301775147929,154 "grad_norm": 0.04624412953853607,155 "learning_rate": 7.938856015779092e-07,156 "loss": 0.0349475234746933,157 "step": 210158 },159 {160 "epoch": 0.650887573964497,161 "grad_norm": 8.037376403808594,162 "learning_rate": 7.840236686390532e-07,163 "loss": 0.07437270879745483,164 "step": 220165 },166 {167 "epoch": 0.6804733727810651,168 "grad_norm": 25.54385757446289,169 "learning_rate": 7.741617357001972e-07,170 "loss": 0.03561083972454071,171 "step": 230172 },173 {174 "epoch": 0.7100591715976331,175 "grad_norm": 0.018464118242263794,176 "learning_rate": 7.642998027613412e-07,177 "loss": 0.07295619249343872,178 "step": 240179 },180 {181 "epoch": 0.7396449704142012,182 "grad_norm": 7.370178699493408,183 "learning_rate": 7.544378698224852e-07,184 "loss": 0.041477668285369876,185 "step": 250186 },187 {188 "epoch": 0.7692307692307693,189 "grad_norm": 0.0641414150595665,190 "learning_rate": 7.445759368836292e-07,191 "loss": 0.0911745309829712,192 "step": 260193 },194 {195 "epoch": 0.7988165680473372,196 "grad_norm": 5.256822109222412,197 "learning_rate": 7.347140039447731e-07,198 "loss": 0.02059374749660492,199 "step": 270200 },201 {202 "epoch": 0.8284023668639053,203 "grad_norm": 0.28701961040496826,204 "learning_rate": 7.248520710059172e-07,205 "loss": 0.07011921405792236,206 "step": 280207 },208 {209 "epoch": 0.8579881656804734,210 "grad_norm": 1.0779420137405396,211 "learning_rate": 7.149901380670611e-07,212 "loss": 0.00536484457552433,213 "step": 290214 },215 {216 "epoch": 0.8875739644970414,217 "grad_norm": 0.1912589967250824,218 "learning_rate": 7.051282051282052e-07,219 "loss": 0.0022948525846004488,220 "step": 300221 },222 {223 "epoch": 0.9171597633136095,224 "grad_norm": 0.008147371001541615,225 "learning_rate": 6.952662721893491e-07,226 "loss": 0.03232004940509796,227 "step": 310228 },229 {230 "epoch": 0.9467455621301775,231 "grad_norm": 0.08998975157737732,232 "learning_rate": 6.854043392504931e-07,233 "loss": 0.017145608365535737,234 "step": 320235 },236 {237 "epoch": 0.9763313609467456,238 "grad_norm": 0.19722725450992584,239 "learning_rate": 6.755424063116371e-07,240 "loss": 0.1359280228614807,241 "step": 330242 },243 {244 "epoch": 1.0,245 "eval_accuracy": 0.9925062447960034,246 "eval_f1_macro": 0.9946876395614233,247 "eval_loss": 0.06303828209638596,248 "eval_runtime": 2.8588,249 "eval_samples_per_second": 420.1,250 "eval_steps_per_second": 13.292,251 "step": 338252 },253 {254 "epoch": 1.0059171597633136,255 "grad_norm": 1.9466553926467896,256 "learning_rate": 6.65680473372781e-07,257 "loss": 0.048591965436935426,258 "step": 340259 },260 {261 "epoch": 1.0355029585798816,262 "grad_norm": 0.058112699538469315,263 "learning_rate": 6.558185404339251e-07,264 "loss": 0.04741050899028778,265 "step": 350266 },267 {268 "epoch": 1.0650887573964498,269 "grad_norm": 1.1703613996505737,270 "learning_rate": 6.45956607495069e-07,271 "loss": 0.08721850514411926,272 "step": 360273 },274 {275 "epoch": 1.0946745562130178,276 "grad_norm": 6.201245307922363,277 "learning_rate": 6.360946745562131e-07,278 "loss": 0.07950657606124878,279 "step": 370280 },281 {282 "epoch": 1.1242603550295858,283 "grad_norm": 0.28137311339378357,284 "learning_rate": 6.262327416173569e-07,285 "loss": 0.052055245637893675,286 "step": 380287 },288 {289 "epoch": 1.1538461538461537,290 "grad_norm": 0.40702566504478455,291 "learning_rate": 6.163708086785009e-07,292 "loss": 0.06641973257064819,293 "step": 390294 },295 {296 "epoch": 1.183431952662722,297 "grad_norm": 52.355533599853516,298 "learning_rate": 6.065088757396449e-07,299 "loss": 0.10939252376556396,300 "step": 400301 },302 {303 "epoch": 1.21301775147929,304 "grad_norm": 0.7765173316001892,305 "learning_rate": 5.966469428007889e-07,306 "loss": 0.10355955362319946,307 "step": 410308 },309 {310 "epoch": 1.242603550295858,311 "grad_norm": 0.11931727081537247,312 "learning_rate": 5.867850098619329e-07,313 "loss": 0.044220191240310666,314 "step": 420315 },316 {317 "epoch": 1.272189349112426,318 "grad_norm": 10.003737449645996,319 "learning_rate": 5.769230769230768e-07,320 "loss": 0.03412697315216064,321 "step": 430322 },323 {324 "epoch": 1.301775147928994,325 "grad_norm": 0.031009649857878685,326 "learning_rate": 5.670611439842209e-07,327 "loss": 0.06639747619628907,328 "step": 440329 },330 {331 "epoch": 1.331360946745562,332 "grad_norm": 0.012379486113786697,333 "learning_rate": 5.571992110453648e-07,334 "loss": 0.06767366528511047,335 "step": 450336 },337 {338 "epoch": 1.3609467455621302,339 "grad_norm": 5.0127339363098145,340 "learning_rate": 5.473372781065088e-07,341 "loss": 0.03919327259063721,342 "step": 460343 },344 {345 "epoch": 1.3905325443786982,346 "grad_norm": 6.062246322631836,347 "learning_rate": 5.374753451676528e-07,348 "loss": 0.09668261408805848,349 "step": 470350 },351 {352 "epoch": 1.4201183431952662,353 "grad_norm": 2.111865997314453,354 "learning_rate": 5.276134122287968e-07,355 "loss": 0.01744236797094345,356 "step": 480357 },358 {359 "epoch": 1.4497041420118344,360 "grad_norm": 15.700614929199219,361 "learning_rate": 5.177514792899408e-07,362 "loss": 0.06998544335365295,363 "step": 490364 },365 {366 "epoch": 1.4792899408284024,367 "grad_norm": 0.0927405133843422,368 "learning_rate": 5.078895463510848e-07,369 "loss": 0.05407346487045288,370 "step": 500371 },372 {373 "epoch": 1.5088757396449703,374 "grad_norm": 0.24365217983722687,375 "learning_rate": 4.980276134122287e-07,376 "loss": 0.04637296795845032,377 "step": 510378 },379 {380 "epoch": 1.5384615384615383,381 "grad_norm": 0.0884738638997078,382 "learning_rate": 4.881656804733728e-07,383 "loss": 0.021781933307647706,384 "step": 520385 },386 {387 "epoch": 1.5680473372781065,388 "grad_norm": 24.246849060058594,389 "learning_rate": 4.783037475345167e-07,390 "loss": 0.1513033390045166,391 "step": 530392 },393 {394 "epoch": 1.5976331360946747,395 "grad_norm": 0.018058154731988907,396 "learning_rate": 4.684418145956607e-07,397 "loss": 0.040793171525001524,398 "step": 540399 },400 {401 "epoch": 1.6272189349112427,402 "grad_norm": 0.010644642636179924,403 "learning_rate": 4.585798816568047e-07,404 "loss": 0.046628400683403015,405 "step": 550406 },407 {408 "epoch": 1.6568047337278107,409 "grad_norm": 0.008394652977585793,410 "learning_rate": 4.487179487179487e-07,411 "loss": 0.027513697743415833,412 "step": 560413 },414 {415 "epoch": 1.6863905325443787,416 "grad_norm": 0.10013745725154877,417 "learning_rate": 4.388560157790927e-07,418 "loss": 0.018340136110782623,419 "step": 570420 },421 {422 "epoch": 1.7159763313609466,423 "grad_norm": 1.7918286323547363,424 "learning_rate": 4.289940828402367e-07,425 "loss": 0.06239059567451477,426 "step": 580427 },428 {429 "epoch": 1.7455621301775148,430 "grad_norm": 1.6770535707473755,431 "learning_rate": 4.1913214990138063e-07,432 "loss": 0.0336042732000351,433 "step": 590434 },435 {436 "epoch": 1.7751479289940828,437 "grad_norm": 12.563508033752441,438 "learning_rate": 4.092702169625246e-07,439 "loss": 0.042831751704216006,440 "step": 600441 },442 {443 "epoch": 1.804733727810651,444 "grad_norm": 0.18050755560398102,445 "learning_rate": 3.994082840236686e-07,446 "loss": 0.014944742619991302,447 "step": 610448 },449 {450 "epoch": 1.834319526627219,451 "grad_norm": 0.026746636256575584,452 "learning_rate": 3.895463510848126e-07,453 "loss": 0.038342425227165224,454 "step": 620455 },456 {457 "epoch": 1.863905325443787,458 "grad_norm": 2.5095348358154297,459 "learning_rate": 3.796844181459566e-07,460 "loss": 0.0056992556899786,461 "step": 630462 },463 {464 "epoch": 1.893491124260355,465 "grad_norm": 0.16079524159431458,466 "learning_rate": 3.698224852071006e-07,467 "loss": 0.0316496878862381,468 "step": 640469 },470 {471 "epoch": 1.9230769230769231,472 "grad_norm": 0.190856471657753,473 "learning_rate": 3.599605522682446e-07,474 "loss": 0.0769759476184845,475 "step": 650476 },477 {478 "epoch": 1.952662721893491,479 "grad_norm": 0.9212966561317444,480 "learning_rate": 3.5009861932938857e-07,481 "loss": 0.0774189829826355,482 "step": 660483 },484 {485 "epoch": 1.9822485207100593,486 "grad_norm": 0.246255561709404,487 "learning_rate": 3.4023668639053256e-07,488 "loss": 0.1128537654876709,489 "step": 670490 },491 {492 "epoch": 2.0,493 "eval_accuracy": 0.9925062447960034,494 "eval_f1_macro": 0.9946876395614233,495 "eval_loss": 0.060506485402584076,496 "eval_runtime": 2.9071,497 "eval_samples_per_second": 413.134,498 "eval_steps_per_second": 13.072,499 "step": 676500 },501 {502 "epoch": 2.0118343195266273,503 "grad_norm": 6.575931072235107,504 "learning_rate": 3.3037475345167656e-07,505 "loss": 0.026142001152038574,506 "step": 680507 },508 {509 "epoch": 2.0414201183431953,510 "grad_norm": 0.8131992816925049,511 "learning_rate": 3.2051282051282055e-07,512 "loss": 0.12568552494049073,513 "step": 690514 },515 {516 "epoch": 2.0710059171597632,517 "grad_norm": 6.568769454956055,518 "learning_rate": 3.1065088757396443e-07,519 "loss": 0.03764350116252899,520 "step": 700521 },522 {523 "epoch": 2.100591715976331,524 "grad_norm": 3.0953428745269775,525 "learning_rate": 3.0078895463510843e-07,526 "loss": 0.018691258132457735,527 "step": 710528 },529 {530 "epoch": 2.1301775147928996,531 "grad_norm": 32.108238220214844,532 "learning_rate": 2.909270216962524e-07,533 "loss": 0.028234952688217164,534 "step": 720535 },536 {537 "epoch": 2.1597633136094676,538 "grad_norm": 0.10239103436470032,539 "learning_rate": 2.810650887573964e-07,540 "loss": 0.056781113147735596,541 "step": 730542 },543 {544 "epoch": 2.1893491124260356,545 "grad_norm": 10.961408615112305,546 "learning_rate": 2.712031558185404e-07,547 "loss": 0.025803667306900025,548 "step": 740549 },550 {551 "epoch": 2.2189349112426036,552 "grad_norm": 1.425737977027893,553 "learning_rate": 2.613412228796844e-07,554 "loss": 0.09297516942024231,555 "step": 750556 },557 {558 "epoch": 2.2485207100591715,559 "grad_norm": 0.018898094072937965,560 "learning_rate": 2.514792899408284e-07,561 "loss": 0.03459992706775665,562 "step": 760563 },564 {565 "epoch": 2.2781065088757395,566 "grad_norm": 0.41209477186203003,567 "learning_rate": 2.416173570019724e-07,568 "loss": 0.007052520662546158,569 "step": 770570 },571 {572 "epoch": 2.3076923076923075,573 "grad_norm": 0.6478250026702881,574 "learning_rate": 2.3175542406311634e-07,575 "loss": 0.09886857271194457,576 "step": 780577 },578 {579 "epoch": 2.337278106508876,580 "grad_norm": 10.342010498046875,581 "learning_rate": 2.2189349112426034e-07,582 "loss": 0.08295982480049133,583 "step": 790584 },585 {586 "epoch": 2.366863905325444,587 "grad_norm": 0.5843625068664551,588 "learning_rate": 2.1203155818540433e-07,589 "loss": 0.020692025125026704,590 "step": 800591 },592 {593 "epoch": 2.396449704142012,594 "grad_norm": 2.5893003940582275,595 "learning_rate": 2.0216962524654832e-07,596 "loss": 0.05822619199752808,597 "step": 810598 },599 {600 "epoch": 2.42603550295858,601 "grad_norm": 0.05523025244474411,602 "learning_rate": 1.9230769230769231e-07,603 "loss": 0.042874136567115785,604 "step": 820605 },606 {607 "epoch": 2.455621301775148,608 "grad_norm": 12.588403701782227,609 "learning_rate": 1.8244575936883628e-07,610 "loss": 0.04601215124130249,611 "step": 830612 },613 {614 "epoch": 2.485207100591716,615 "grad_norm": 0.22603930532932281,616 "learning_rate": 1.7258382642998027e-07,617 "loss": 0.05003575086593628,618 "step": 840619 },620 {621 "epoch": 2.5147928994082838,622 "grad_norm": 10.426587104797363,623 "learning_rate": 1.6272189349112426e-07,624 "loss": 0.021601344645023345,625 "step": 850626 },627 {628 "epoch": 2.544378698224852,629 "grad_norm": 0.01249838899821043,630 "learning_rate": 1.5285996055226823e-07,631 "loss": 0.04493778645992279,632 "step": 860633 },634 {635 "epoch": 2.57396449704142,636 "grad_norm": 0.6047437191009521,637 "learning_rate": 1.4299802761341222e-07,638 "loss": 0.017576351761817932,639 "step": 870640 },641 {642 "epoch": 2.603550295857988,643 "grad_norm": 0.22967278957366943,644 "learning_rate": 1.331360946745562e-07,645 "loss": 0.11544790267944335,646 "step": 880647 },648 {649 "epoch": 2.633136094674556,650 "grad_norm": 0.04233518987894058,651 "learning_rate": 1.232741617357002e-07,652 "loss": 0.008126095682382584,653 "step": 890654 },655 {656 "epoch": 2.662721893491124,657 "grad_norm": 13.421414375305176,658 "learning_rate": 1.1341222879684417e-07,659 "loss": 0.06348974704742431,660 "step": 900661 },662 {663 "epoch": 2.6923076923076925,664 "grad_norm": 17.87152099609375,665 "learning_rate": 1.0355029585798816e-07,666 "loss": 0.07953644394874573,667 "step": 910668 },669 {670 "epoch": 2.7218934911242605,671 "grad_norm": 0.36386340856552124,672 "learning_rate": 9.368836291913215e-08,673 "loss": 0.05966421365737915,674 "step": 920675 },676 {677 "epoch": 2.7514792899408285,678 "grad_norm": 6.722818374633789,679 "learning_rate": 8.382642998027613e-08,680 "loss": 0.06773669719696045,681 "step": 930682 },683 {684 "epoch": 2.7810650887573964,685 "grad_norm": 0.011403938755393028,686 "learning_rate": 7.396449704142011e-08,687 "loss": 0.01843421161174774,688 "step": 940689 },690 {691 "epoch": 2.8106508875739644,692 "grad_norm": 2.8068971633911133,693 "learning_rate": 6.410256410256409e-08,694 "loss": 0.037809357047080994,695 "step": 950696 },697 {698 "epoch": 2.8402366863905324,699 "grad_norm": 0.27246904373168945,700 "learning_rate": 5.424063116370808e-08,701 "loss": 0.06303403973579406,702 "step": 960703 },704 {705 "epoch": 2.8698224852071004,706 "grad_norm": 6.128849506378174,707 "learning_rate": 4.437869822485207e-08,708 "loss": 0.0703296720981598,709 "step": 970710 },711 {712 "epoch": 2.899408284023669,713 "grad_norm": 0.10920988023281097,714 "learning_rate": 3.4516765285996054e-08,715 "loss": 0.08795815706253052,716 "step": 980717 },718 {719 "epoch": 2.9289940828402368,720 "grad_norm": 0.7319285869598389,721 "learning_rate": 2.465483234714004e-08,722 "loss": 0.061006712913513186,723 "step": 990724 },725 {726 "epoch": 2.9585798816568047,727 "grad_norm": 0.07155957818031311,728 "learning_rate": 1.4792899408284022e-08,729 "loss": 0.03764758408069611,730 "step": 1000731 },732 {733 "epoch": 2.9881656804733727,734 "grad_norm": 0.8153097629547119,735 "learning_rate": 4.930966469428008e-09,736 "loss": 0.03423367142677307,737 "step": 1010738 },739 {740 "epoch": 3.0,741 "eval_accuracy": 0.9925062447960034,742 "eval_f1_macro": 0.9946876395614233,743 "eval_loss": 0.06174274533987045,744 "eval_runtime": 2.9029,745 "eval_samples_per_second": 413.719,746 "eval_steps_per_second": 13.09,747 "step": 1014748 }749 ],750 "logging_steps": 10,751 "max_steps": 1014,752 "num_input_tokens_seen": 0,753 "num_train_epochs": 3,754 "save_steps": 500,755 "stateful_callbacks": {756 "TrainerControl": {757 "args": {758 "should_epoch_stop": false,759 "should_evaluate": false,760 "should_log": false,761 "should_save": true,762 "should_training_stop": true763 },764 "attributes": {}765 }766 },767 "total_flos": 1073991376410624.0,768 "train_batch_size": 32,769 "trial_name": null,770 "trial_params": null771}772 