| { | |
| "best_metric": 0.7983870967741935, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-qvrnjpot/checkpoint-800", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 800, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.7631841897964478, | |
| "learning_rate": 6.160315650515145e-05, | |
| "loss": 0.617, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 2.8386940956115723, | |
| "learning_rate": 6.082336971394699e-05, | |
| "loss": 0.5112, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 2.7043747901916504, | |
| "learning_rate": 6.004358292274255e-05, | |
| "loss": 0.4761, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 2.6273417472839355, | |
| "learning_rate": 5.9263796131538093e-05, | |
| "loss": 0.5074, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 5.05076265335083, | |
| "learning_rate": 5.8484009340333646e-05, | |
| "loss": 0.3454, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 6.585201263427734, | |
| "learning_rate": 5.77042225491292e-05, | |
| "loss": 0.4747, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 11.389758110046387, | |
| "learning_rate": 5.6924435757924745e-05, | |
| "loss": 0.5088, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 17.25305938720703, | |
| "learning_rate": 5.6144648966720304e-05, | |
| "loss": 0.3702, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 4.934868335723877, | |
| "learning_rate": 5.536486217551585e-05, | |
| "loss": 0.2679, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 8.351325035095215, | |
| "learning_rate": 5.45850753843114e-05, | |
| "loss": 0.4069, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 8.08082389831543, | |
| "learning_rate": 5.3805288593106956e-05, | |
| "loss": 0.5033, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 1.6448326110839844, | |
| "learning_rate": 5.30255018019025e-05, | |
| "loss": 0.3679, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 2.436563014984131, | |
| "learning_rate": 5.224571501069806e-05, | |
| "loss": 0.506, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.7677550315856934, | |
| "learning_rate": 5.146592821949361e-05, | |
| "loss": 0.351, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 5.091623783111572, | |
| "learning_rate": 5.068614142828916e-05, | |
| "loss": 0.4919, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 12.155316352844238, | |
| "learning_rate": 4.990635463708471e-05, | |
| "loss": 0.4279, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 2.3314337730407715, | |
| "learning_rate": 4.912656784588026e-05, | |
| "loss": 0.4913, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 6.5636210441589355, | |
| "learning_rate": 4.834678105467582e-05, | |
| "loss": 0.3929, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 2.8381412029266357, | |
| "learning_rate": 4.7566994263471364e-05, | |
| "loss": 0.2571, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.629071056842804, | |
| "learning_rate": 4.678720747226692e-05, | |
| "loss": 0.2667, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8525, | |
| "eval_f1": 0.7704280155642024, | |
| "eval_loss": 0.35932326316833496, | |
| "eval_precision": 0.7388059701492538, | |
| "eval_recall": 0.8048780487804879, | |
| "eval_runtime": 1.5245, | |
| "eval_samples_per_second": 262.373, | |
| "eval_steps_per_second": 16.398, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 0.6173528432846069, | |
| "learning_rate": 4.600742068106247e-05, | |
| "loss": 0.1887, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.25395506620407104, | |
| "learning_rate": 4.5227633889858016e-05, | |
| "loss": 0.2535, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 25.887020111083984, | |
| "learning_rate": 4.4447847098653575e-05, | |
| "loss": 0.1868, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.13451418280601501, | |
| "learning_rate": 4.366806030744912e-05, | |
| "loss": 0.3065, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.36101168394088745, | |
| "learning_rate": 4.2888273516244674e-05, | |
| "loss": 0.249, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 7.472021579742432, | |
| "learning_rate": 4.2108486725040227e-05, | |
| "loss": 0.2548, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 8.727078437805176, | |
| "learning_rate": 4.132869993383577e-05, | |
| "loss": 0.2963, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 13.003727912902832, | |
| "learning_rate": 4.054891314263133e-05, | |
| "loss": 0.3822, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.511164665222168, | |
| "learning_rate": 3.976912635142688e-05, | |
| "loss": 0.2591, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.3373236656188965, | |
| "learning_rate": 3.898933956022243e-05, | |
| "loss": 0.2428, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 1.4264609813690186, | |
| "learning_rate": 3.8209552769017983e-05, | |
| "loss": 0.3269, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.13309431076049805, | |
| "learning_rate": 3.742976597781353e-05, | |
| "loss": 0.2393, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 10.90920639038086, | |
| "learning_rate": 3.664997918660909e-05, | |
| "loss": 0.306, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 13.362682342529297, | |
| "learning_rate": 3.5870192395404635e-05, | |
| "loss": 0.3569, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 4.543549537658691, | |
| "learning_rate": 3.509040560420019e-05, | |
| "loss": 0.3563, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 4.552388668060303, | |
| "learning_rate": 3.431061881299574e-05, | |
| "loss": 0.3519, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 6.557765960693359, | |
| "learning_rate": 3.3530832021791286e-05, | |
| "loss": 0.2307, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 8.15699577331543, | |
| "learning_rate": 3.2751045230586846e-05, | |
| "loss": 0.2675, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 0.4172225892543793, | |
| "learning_rate": 3.197125843938239e-05, | |
| "loss": 0.1495, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.5596890449523926, | |
| "learning_rate": 3.1191471648177945e-05, | |
| "loss": 0.3743, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.88, | |
| "eval_f1": 0.7948717948717948, | |
| "eval_loss": 0.3793269395828247, | |
| "eval_precision": 0.8378378378378378, | |
| "eval_recall": 0.7560975609756098, | |
| "eval_runtime": 1.5273, | |
| "eval_samples_per_second": 261.897, | |
| "eval_steps_per_second": 16.369, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 0.26775601506233215, | |
| "learning_rate": 3.0411684856973494e-05, | |
| "loss": 0.1442, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 0.1751158982515335, | |
| "learning_rate": 2.9631898065769047e-05, | |
| "loss": 0.1926, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 0.27150827646255493, | |
| "learning_rate": 2.88521112745646e-05, | |
| "loss": 0.1173, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 11.543705940246582, | |
| "learning_rate": 2.8072324483360152e-05, | |
| "loss": 0.2252, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 1.7563936710357666, | |
| "learning_rate": 2.72925376921557e-05, | |
| "loss": 0.2043, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 0.22426572442054749, | |
| "learning_rate": 2.651275090095125e-05, | |
| "loss": 0.089, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 0.09884244948625565, | |
| "learning_rate": 2.5732964109746804e-05, | |
| "loss": 0.0607, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.5133824348449707, | |
| "learning_rate": 2.4953177318542356e-05, | |
| "loss": 0.2, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 1.0894460678100586, | |
| "learning_rate": 2.417339052733791e-05, | |
| "loss": 0.2041, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 20.391416549682617, | |
| "learning_rate": 2.339360373613346e-05, | |
| "loss": 0.3932, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 1.9286866188049316, | |
| "learning_rate": 2.2613816944929008e-05, | |
| "loss": 0.0993, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 0.09835375845432281, | |
| "learning_rate": 2.183403015372456e-05, | |
| "loss": 0.0856, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 0.2403247058391571, | |
| "learning_rate": 2.1054243362520113e-05, | |
| "loss": 0.1441, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 0.10543715208768845, | |
| "learning_rate": 2.0274456571315666e-05, | |
| "loss": 0.1147, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 10.961527824401855, | |
| "learning_rate": 1.9494669780111215e-05, | |
| "loss": 0.2312, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 29.004623413085938, | |
| "learning_rate": 1.8714882988906765e-05, | |
| "loss": 0.2722, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 0.12045635282993317, | |
| "learning_rate": 1.7935096197702317e-05, | |
| "loss": 0.336, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 20.202497482299805, | |
| "learning_rate": 1.715530940649787e-05, | |
| "loss": 0.0886, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 5.752193450927734, | |
| "learning_rate": 1.6375522615293423e-05, | |
| "loss": 0.1171, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 25.30577278137207, | |
| "learning_rate": 1.5595735824088972e-05, | |
| "loss": 0.1935, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.88, | |
| "eval_f1": 0.7931034482758621, | |
| "eval_loss": 0.4657374620437622, | |
| "eval_precision": 0.8440366972477065, | |
| "eval_recall": 0.7479674796747967, | |
| "eval_runtime": 1.5234, | |
| "eval_samples_per_second": 262.573, | |
| "eval_steps_per_second": 16.411, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 0.058765921741724014, | |
| "learning_rate": 1.4815949032884523e-05, | |
| "loss": 0.0065, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 0.05543431267142296, | |
| "learning_rate": 1.4036162241680076e-05, | |
| "loss": 0.1243, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 4.4582014083862305, | |
| "learning_rate": 1.3256375450475625e-05, | |
| "loss": 0.0124, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.7271829843521118, | |
| "learning_rate": 1.2476588659271178e-05, | |
| "loss": 0.0167, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 0.07129786163568497, | |
| "learning_rate": 1.169680186806673e-05, | |
| "loss": 0.0047, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 0.040439434349536896, | |
| "learning_rate": 1.091701507686228e-05, | |
| "loss": 0.0031, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "grad_norm": 0.032109107822179794, | |
| "learning_rate": 1.0137228285657833e-05, | |
| "loss": 0.0047, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 0.068085677921772, | |
| "learning_rate": 9.357441494453382e-06, | |
| "loss": 0.1646, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "grad_norm": 0.031069407239556313, | |
| "learning_rate": 8.577654703248935e-06, | |
| "loss": 0.267, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 0.11077981442213058, | |
| "learning_rate": 7.797867912044486e-06, | |
| "loss": 0.0763, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "grad_norm": 0.0862860158085823, | |
| "learning_rate": 7.018081120840038e-06, | |
| "loss": 0.0946, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.03846760839223862, | |
| "learning_rate": 6.238294329635589e-06, | |
| "loss": 0.1012, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "grad_norm": 0.045732282102108, | |
| "learning_rate": 5.45850753843114e-06, | |
| "loss": 0.0695, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "grad_norm": 0.044342800974845886, | |
| "learning_rate": 4.678720747226691e-06, | |
| "loss": 0.0064, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.06440162658691406, | |
| "learning_rate": 3.898933956022243e-06, | |
| "loss": 0.0971, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 0.04895680025219917, | |
| "learning_rate": 3.1191471648177945e-06, | |
| "loss": 0.1129, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "grad_norm": 20.34134864807129, | |
| "learning_rate": 2.3393603736133456e-06, | |
| "loss": 0.1191, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "grad_norm": 0.12530703842639923, | |
| "learning_rate": 1.5595735824088973e-06, | |
| "loss": 0.0551, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "grad_norm": 0.05208105966448784, | |
| "learning_rate": 7.797867912044486e-07, | |
| "loss": 0.0498, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.04303634166717529, | |
| "learning_rate": 0.0, | |
| "loss": 0.0832, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.875, | |
| "eval_f1": 0.7983870967741935, | |
| "eval_loss": 0.5619988441467285, | |
| "eval_precision": 0.792, | |
| "eval_recall": 0.8048780487804879, | |
| "eval_runtime": 1.5731, | |
| "eval_samples_per_second": 254.276, | |
| "eval_steps_per_second": 15.892, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 847261481803776.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 6.238294329635589e-05, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 4, | |
| "per_device_train_batch_size": 8, | |
| "seed": 31 | |
| } | |
| } | |