| { | |
| "best_metric": 0.8286852589641435, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-ul9reh0d/checkpoint-1600", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 1600, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 3.052316665649414, | |
| "learning_rate": 6.753956614871026e-05, | |
| "loss": 0.5937, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.5294617414474487, | |
| "learning_rate": 6.720017134394288e-05, | |
| "loss": 0.5648, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 2.700227737426758, | |
| "learning_rate": 6.686077653917548e-05, | |
| "loss": 0.6581, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 12.69797134399414, | |
| "learning_rate": 6.65213817344081e-05, | |
| "loss": 0.5975, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 10.163844108581543, | |
| "learning_rate": 6.618198692964071e-05, | |
| "loss": 0.3418, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 23.86635971069336, | |
| "learning_rate": 6.584259212487333e-05, | |
| "loss": 0.7951, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "grad_norm": 5.886571884155273, | |
| "learning_rate": 6.550319732010593e-05, | |
| "loss": 0.7399, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 18.702909469604492, | |
| "learning_rate": 6.516380251533855e-05, | |
| "loss": 0.4083, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "grad_norm": 6.397231101989746, | |
| "learning_rate": 6.482440771057116e-05, | |
| "loss": 0.6293, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 39.70441436767578, | |
| "learning_rate": 6.448501290580378e-05, | |
| "loss": 0.6267, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.275, | |
| "grad_norm": 1.6805046796798706, | |
| "learning_rate": 6.414561810103638e-05, | |
| "loss": 0.5261, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 2.5493907928466797, | |
| "learning_rate": 6.380622329626899e-05, | |
| "loss": 0.2222, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.325, | |
| "grad_norm": 5.368661403656006, | |
| "learning_rate": 6.346682849150162e-05, | |
| "loss": 0.5268, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 4.198462963104248, | |
| "learning_rate": 6.312743368673422e-05, | |
| "loss": 0.5109, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 6.481055736541748, | |
| "learning_rate": 6.278803888196683e-05, | |
| "loss": 0.3406, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.7471745014190674, | |
| "learning_rate": 6.244864407719945e-05, | |
| "loss": 0.6461, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.425, | |
| "grad_norm": 16.66990089416504, | |
| "learning_rate": 6.210924927243205e-05, | |
| "loss": 0.4364, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 7.193353176116943, | |
| "learning_rate": 6.176985446766467e-05, | |
| "loss": 0.3659, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.475, | |
| "grad_norm": 8.753779411315918, | |
| "learning_rate": 6.143045966289728e-05, | |
| "loss": 0.5441, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.7563661336898804, | |
| "learning_rate": 6.10910648581299e-05, | |
| "loss": 0.4186, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.525, | |
| "grad_norm": 27.28961753845215, | |
| "learning_rate": 6.07516700533625e-05, | |
| "loss": 0.394, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.9820290803909302, | |
| "learning_rate": 6.041227524859511e-05, | |
| "loss": 0.9692, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.575, | |
| "grad_norm": 0.795172393321991, | |
| "learning_rate": 6.0072880443827725e-05, | |
| "loss": 0.4171, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.29914233088493347, | |
| "learning_rate": 5.973348563906034e-05, | |
| "loss": 0.467, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 8.065807342529297, | |
| "learning_rate": 5.939409083429295e-05, | |
| "loss": 1.0716, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 3.9354641437530518, | |
| "learning_rate": 5.905469602952556e-05, | |
| "loss": 0.3503, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.675, | |
| "grad_norm": 0.45260441303253174, | |
| "learning_rate": 5.871530122475817e-05, | |
| "loss": 0.7493, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.5366302728652954, | |
| "learning_rate": 5.8375906419990785e-05, | |
| "loss": 0.346, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.725, | |
| "grad_norm": 8.400691032409668, | |
| "learning_rate": 5.80365116152234e-05, | |
| "loss": 0.3399, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 7.528047561645508, | |
| "learning_rate": 5.769711681045601e-05, | |
| "loss": 0.5184, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.775, | |
| "grad_norm": 0.5284883975982666, | |
| "learning_rate": 5.7357722005688614e-05, | |
| "loss": 0.3518, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.5432015657424927, | |
| "learning_rate": 5.7018327200921226e-05, | |
| "loss": 0.4533, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.825, | |
| "grad_norm": 8.873420715332031, | |
| "learning_rate": 5.667893239615384e-05, | |
| "loss": 0.6322, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 15.452073097229004, | |
| "learning_rate": 5.633953759138645e-05, | |
| "loss": 0.5022, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 5.867374897003174, | |
| "learning_rate": 5.600014278661906e-05, | |
| "loss": 0.2418, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 24.693492889404297, | |
| "learning_rate": 5.5660747981851674e-05, | |
| "loss": 0.5568, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.925, | |
| "grad_norm": 2.5772554874420166, | |
| "learning_rate": 5.5321353177084286e-05, | |
| "loss": 0.5199, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 1.09758460521698, | |
| "learning_rate": 5.4981958372316905e-05, | |
| "loss": 0.4895, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.975, | |
| "grad_norm": 0.9194037914276123, | |
| "learning_rate": 5.464256356754952e-05, | |
| "loss": 0.5017, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.7973395586013794, | |
| "learning_rate": 5.430316876278213e-05, | |
| "loss": 0.316, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8425, | |
| "eval_f1": 0.7407407407407407, | |
| "eval_loss": 0.4195494055747986, | |
| "eval_precision": 0.75, | |
| "eval_recall": 0.7317073170731707, | |
| "eval_runtime": 1.5136, | |
| "eval_samples_per_second": 264.267, | |
| "eval_steps_per_second": 16.517, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.025, | |
| "grad_norm": 0.25087594985961914, | |
| "learning_rate": 5.396377395801474e-05, | |
| "loss": 0.2648, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 5.209116458892822, | |
| "learning_rate": 5.362437915324735e-05, | |
| "loss": 0.4875, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.075, | |
| "grad_norm": 0.28788337111473083, | |
| "learning_rate": 5.3284984348479965e-05, | |
| "loss": 0.5706, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.34227511286735535, | |
| "learning_rate": 5.2945589543712577e-05, | |
| "loss": 0.2126, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 9.716309547424316, | |
| "learning_rate": 5.260619473894518e-05, | |
| "loss": 0.489, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 10.545299530029297, | |
| "learning_rate": 5.2266799934177794e-05, | |
| "loss": 0.4637, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.175, | |
| "grad_norm": 0.4732460677623749, | |
| "learning_rate": 5.1927405129410406e-05, | |
| "loss": 0.4207, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 7.716385841369629, | |
| "learning_rate": 5.158801032464302e-05, | |
| "loss": 0.4326, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.225, | |
| "grad_norm": 4.098447799682617, | |
| "learning_rate": 5.124861551987563e-05, | |
| "loss": 0.3346, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 15.903754234313965, | |
| "learning_rate": 5.090922071510824e-05, | |
| "loss": 0.5305, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.275, | |
| "grad_norm": 0.19262875616550446, | |
| "learning_rate": 5.0569825910340854e-05, | |
| "loss": 0.2946, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.18385837972164154, | |
| "learning_rate": 5.0230431105573466e-05, | |
| "loss": 0.2572, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.325, | |
| "grad_norm": 0.1092706024646759, | |
| "learning_rate": 4.989103630080608e-05, | |
| "loss": 0.1818, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 0.2516274154186249, | |
| "learning_rate": 4.955164149603869e-05, | |
| "loss": 0.4868, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 0.19831004738807678, | |
| "learning_rate": 4.92122466912713e-05, | |
| "loss": 0.268, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.16449308395385742, | |
| "learning_rate": 4.887285188650391e-05, | |
| "loss": 0.1882, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.425, | |
| "grad_norm": 8.629518508911133, | |
| "learning_rate": 4.853345708173652e-05, | |
| "loss": 0.6056, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 2.007662057876587, | |
| "learning_rate": 4.819406227696913e-05, | |
| "loss": 0.232, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.475, | |
| "grad_norm": 0.3099694550037384, | |
| "learning_rate": 4.785466747220174e-05, | |
| "loss": 0.2788, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 40.103694915771484, | |
| "learning_rate": 4.7515272667434355e-05, | |
| "loss": 0.1362, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.525, | |
| "grad_norm": 31.98729133605957, | |
| "learning_rate": 4.717587786266697e-05, | |
| "loss": 0.5487, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 2.4679489135742188, | |
| "learning_rate": 4.683648305789958e-05, | |
| "loss": 0.5429, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.575, | |
| "grad_norm": 24.133512496948242, | |
| "learning_rate": 4.64970882531322e-05, | |
| "loss": 0.4373, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.29410362243652344, | |
| "learning_rate": 4.615769344836481e-05, | |
| "loss": 0.4658, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 3.7344865798950195, | |
| "learning_rate": 4.581829864359742e-05, | |
| "loss": 0.0568, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 0.154106006026268, | |
| "learning_rate": 4.5478903838830034e-05, | |
| "loss": 0.1266, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.675, | |
| "grad_norm": 62.301597595214844, | |
| "learning_rate": 4.5139509034062646e-05, | |
| "loss": 0.3428, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 0.08602792769670486, | |
| "learning_rate": 4.480011422929526e-05, | |
| "loss": 0.641, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.725, | |
| "grad_norm": 0.13891366124153137, | |
| "learning_rate": 4.446071942452787e-05, | |
| "loss": 0.0499, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.22815102338790894, | |
| "learning_rate": 4.4121324619760475e-05, | |
| "loss": 0.8891, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.775, | |
| "grad_norm": 0.284967839717865, | |
| "learning_rate": 4.378192981499309e-05, | |
| "loss": 0.5238, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.2135128527879715, | |
| "learning_rate": 4.34425350102257e-05, | |
| "loss": 0.4598, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.825, | |
| "grad_norm": 0.2596323788166046, | |
| "learning_rate": 4.310314020545831e-05, | |
| "loss": 0.2411, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 0.12039212137460709, | |
| "learning_rate": 4.276374540069092e-05, | |
| "loss": 0.5047, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 7.840117454528809, | |
| "learning_rate": 4.2424350595923535e-05, | |
| "loss": 0.2763, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 29.95262908935547, | |
| "learning_rate": 4.208495579115615e-05, | |
| "loss": 0.3228, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.925, | |
| "grad_norm": 0.21045206487178802, | |
| "learning_rate": 4.174556098638876e-05, | |
| "loss": 0.1931, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 0.31024569272994995, | |
| "learning_rate": 4.140616618162137e-05, | |
| "loss": 0.463, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.975, | |
| "grad_norm": 0.5587632656097412, | |
| "learning_rate": 4.106677137685398e-05, | |
| "loss": 0.1264, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 6.912115573883057, | |
| "learning_rate": 4.0727376572086595e-05, | |
| "loss": 0.4146, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.875, | |
| "eval_f1": 0.8106060606060606, | |
| "eval_loss": 0.48966696858406067, | |
| "eval_precision": 0.7588652482269503, | |
| "eval_recall": 0.8699186991869918, | |
| "eval_runtime": 1.5322, | |
| "eval_samples_per_second": 261.066, | |
| "eval_steps_per_second": 16.317, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.025, | |
| "grad_norm": 9.648056030273438, | |
| "learning_rate": 4.03879817673192e-05, | |
| "loss": 0.1043, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 0.14807315170764923, | |
| "learning_rate": 4.004858696255181e-05, | |
| "loss": 0.1021, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.075, | |
| "grad_norm": 1.4937807321548462, | |
| "learning_rate": 3.9709192157784424e-05, | |
| "loss": 0.4885, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 0.08521637320518494, | |
| "learning_rate": 3.9369797353017036e-05, | |
| "loss": 0.0906, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.125, | |
| "grad_norm": 0.1001555472612381, | |
| "learning_rate": 3.903040254824965e-05, | |
| "loss": 0.2574, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 0.04532875493168831, | |
| "learning_rate": 3.869100774348226e-05, | |
| "loss": 0.1966, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.175, | |
| "grad_norm": 0.048231206834316254, | |
| "learning_rate": 3.835161293871487e-05, | |
| "loss": 0.2154, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.1648254096508026, | |
| "learning_rate": 3.801221813394749e-05, | |
| "loss": 0.673, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.225, | |
| "grad_norm": 0.1614377647638321, | |
| "learning_rate": 3.76728233291801e-05, | |
| "loss": 0.1303, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 0.12167342752218246, | |
| "learning_rate": 3.7333428524412715e-05, | |
| "loss": 0.0674, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.275, | |
| "grad_norm": 142.9597930908203, | |
| "learning_rate": 3.699403371964533e-05, | |
| "loss": 0.5005, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 0.40126386284828186, | |
| "learning_rate": 3.665463891487794e-05, | |
| "loss": 0.0134, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.325, | |
| "grad_norm": 0.027333322912454605, | |
| "learning_rate": 3.631524411011055e-05, | |
| "loss": 0.2041, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 98.71678924560547, | |
| "learning_rate": 3.597584930534316e-05, | |
| "loss": 0.0721, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.375, | |
| "grad_norm": 0.028235169127583504, | |
| "learning_rate": 3.563645450057577e-05, | |
| "loss": 0.7024, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.030322661623358727, | |
| "learning_rate": 3.529705969580838e-05, | |
| "loss": 0.0277, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.425, | |
| "grad_norm": 15.570178985595703, | |
| "learning_rate": 3.495766489104099e-05, | |
| "loss": 0.3353, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 0.08365435898303986, | |
| "learning_rate": 3.4618270086273604e-05, | |
| "loss": 0.1971, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.475, | |
| "grad_norm": 0.02198941260576248, | |
| "learning_rate": 3.4278875281506216e-05, | |
| "loss": 0.3403, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.22028592228889465, | |
| "learning_rate": 3.393948047673883e-05, | |
| "loss": 0.1051, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.525, | |
| "grad_norm": 0.18908417224884033, | |
| "learning_rate": 3.360008567197144e-05, | |
| "loss": 0.2455, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 0.04513182491064072, | |
| "learning_rate": 3.326069086720405e-05, | |
| "loss": 0.1396, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.575, | |
| "grad_norm": 0.10827507078647614, | |
| "learning_rate": 3.2921296062436664e-05, | |
| "loss": 0.2477, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 0.41028207540512085, | |
| "learning_rate": 3.2581901257669276e-05, | |
| "loss": 0.1829, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.625, | |
| "grad_norm": 0.08836387097835541, | |
| "learning_rate": 3.224250645290189e-05, | |
| "loss": 0.2214, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 0.21761733293533325, | |
| "learning_rate": 3.190311164813449e-05, | |
| "loss": 0.2421, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.675, | |
| "grad_norm": 0.10760723054409027, | |
| "learning_rate": 3.156371684336711e-05, | |
| "loss": 0.0054, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 0.05878245085477829, | |
| "learning_rate": 3.1224322038599724e-05, | |
| "loss": 0.2122, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.725, | |
| "grad_norm": 0.18703211843967438, | |
| "learning_rate": 3.0884927233832336e-05, | |
| "loss": 0.1254, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 0.011866235174238682, | |
| "learning_rate": 3.054553242906495e-05, | |
| "loss": 0.1307, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.775, | |
| "grad_norm": 0.04749550670385361, | |
| "learning_rate": 3.0206137624297556e-05, | |
| "loss": 0.1111, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.06841249763965607, | |
| "learning_rate": 2.986674281953017e-05, | |
| "loss": 0.1828, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.825, | |
| "grad_norm": 0.020818322896957397, | |
| "learning_rate": 2.952734801476278e-05, | |
| "loss": 0.1772, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 0.1497318297624588, | |
| "learning_rate": 2.9187953209995392e-05, | |
| "loss": 0.5667, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.875, | |
| "grad_norm": 0.028121085837483406, | |
| "learning_rate": 2.8848558405228004e-05, | |
| "loss": 0.1637, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 0.15947194397449493, | |
| "learning_rate": 2.8509163600460613e-05, | |
| "loss": 0.1511, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.925, | |
| "grad_norm": 0.13598938286304474, | |
| "learning_rate": 2.8169768795693225e-05, | |
| "loss": 0.4279, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 0.09932050853967667, | |
| "learning_rate": 2.7830373990925837e-05, | |
| "loss": 0.2489, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.975, | |
| "grad_norm": 0.08448607474565506, | |
| "learning_rate": 2.7490979186158452e-05, | |
| "loss": 0.4415, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.4404872953891754, | |
| "learning_rate": 2.7151584381391064e-05, | |
| "loss": 0.0094, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.88, | |
| "eval_f1": 0.7966101694915254, | |
| "eval_loss": 0.543876051902771, | |
| "eval_precision": 0.831858407079646, | |
| "eval_recall": 0.7642276422764228, | |
| "eval_runtime": 1.5751, | |
| "eval_samples_per_second": 253.947, | |
| "eval_steps_per_second": 15.872, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.025, | |
| "grad_norm": 0.08914393186569214, | |
| "learning_rate": 2.6812189576623676e-05, | |
| "loss": 0.1131, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 0.07447825372219086, | |
| "learning_rate": 2.6472794771856288e-05, | |
| "loss": 0.0121, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 3.075, | |
| "grad_norm": 0.03912358358502388, | |
| "learning_rate": 2.6133399967088897e-05, | |
| "loss": 0.1766, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 13.849862098693848, | |
| "learning_rate": 2.579400516232151e-05, | |
| "loss": 0.2275, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 0.2011016458272934, | |
| "learning_rate": 2.545461035755412e-05, | |
| "loss": 0.2427, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 0.08804444223642349, | |
| "learning_rate": 2.5115215552786733e-05, | |
| "loss": 0.155, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 3.175, | |
| "grad_norm": 0.045624420046806335, | |
| "learning_rate": 2.4775820748019345e-05, | |
| "loss": 0.2603, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.10040140151977539, | |
| "learning_rate": 2.4436425943251953e-05, | |
| "loss": 0.0033, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 3.225, | |
| "grad_norm": 0.05396701395511627, | |
| "learning_rate": 2.4097031138484565e-05, | |
| "loss": 0.0031, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 0.06577543169260025, | |
| "learning_rate": 2.3757636333717177e-05, | |
| "loss": 0.1364, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.275, | |
| "grad_norm": 0.05943725258111954, | |
| "learning_rate": 2.341824152894979e-05, | |
| "loss": 0.2291, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 37.863914489746094, | |
| "learning_rate": 2.3078846724182405e-05, | |
| "loss": 0.2501, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 3.325, | |
| "grad_norm": 0.20205795764923096, | |
| "learning_rate": 2.2739451919415017e-05, | |
| "loss": 0.0971, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "grad_norm": 0.11639241874217987, | |
| "learning_rate": 2.240005711464763e-05, | |
| "loss": 0.2033, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 3.375, | |
| "grad_norm": 0.06607875972986221, | |
| "learning_rate": 2.2060662309880237e-05, | |
| "loss": 0.0093, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 0.21797184646129608, | |
| "learning_rate": 2.172126750511285e-05, | |
| "loss": 0.1164, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 3.425, | |
| "grad_norm": 0.0903061032295227, | |
| "learning_rate": 2.138187270034546e-05, | |
| "loss": 0.0025, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "grad_norm": 0.02832476980984211, | |
| "learning_rate": 2.1042477895578073e-05, | |
| "loss": 0.2609, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 3.475, | |
| "grad_norm": 0.0343223437666893, | |
| "learning_rate": 2.0703083090810685e-05, | |
| "loss": 0.0021, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 0.020103761926293373, | |
| "learning_rate": 2.0363688286043297e-05, | |
| "loss": 0.0046, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.525, | |
| "grad_norm": 0.13800527155399323, | |
| "learning_rate": 2.0024293481275906e-05, | |
| "loss": 0.3184, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "grad_norm": 0.024840140715241432, | |
| "learning_rate": 1.9684898676508518e-05, | |
| "loss": 0.2926, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 3.575, | |
| "grad_norm": 0.030249882489442825, | |
| "learning_rate": 1.934550387174113e-05, | |
| "loss": 0.0027, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.019687151536345482, | |
| "learning_rate": 1.9006109066973745e-05, | |
| "loss": 0.0026, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 3.625, | |
| "grad_norm": 0.017106233164668083, | |
| "learning_rate": 1.8666714262206357e-05, | |
| "loss": 0.0023, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "grad_norm": 0.01799299567937851, | |
| "learning_rate": 1.832731945743897e-05, | |
| "loss": 0.0019, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 3.675, | |
| "grad_norm": 0.024008911103010178, | |
| "learning_rate": 1.798792465267158e-05, | |
| "loss": 0.0532, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "grad_norm": 0.1859898418188095, | |
| "learning_rate": 1.764852984790419e-05, | |
| "loss": 0.1062, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 3.725, | |
| "grad_norm": 13.769889831542969, | |
| "learning_rate": 1.7309135043136802e-05, | |
| "loss": 0.1398, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.015940789133310318, | |
| "learning_rate": 1.6969740238369414e-05, | |
| "loss": 0.0017, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.775, | |
| "grad_norm": 0.37872692942619324, | |
| "learning_rate": 1.6630345433602026e-05, | |
| "loss": 0.105, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 0.020107530057430267, | |
| "learning_rate": 1.6290950628834638e-05, | |
| "loss": 0.1373, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 3.825, | |
| "grad_norm": 0.23860657215118408, | |
| "learning_rate": 1.5951555824067247e-05, | |
| "loss": 0.0012, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "grad_norm": 0.017826450988650322, | |
| "learning_rate": 1.5612161019299862e-05, | |
| "loss": 0.0029, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 3.875, | |
| "grad_norm": 0.18682965636253357, | |
| "learning_rate": 1.5272766214532474e-05, | |
| "loss": 0.3438, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "grad_norm": 0.03294467553496361, | |
| "learning_rate": 1.4933371409765084e-05, | |
| "loss": 0.1961, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 3.925, | |
| "grad_norm": 0.01622551679611206, | |
| "learning_rate": 1.4593976604997696e-05, | |
| "loss": 0.0016, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "grad_norm": 0.24873243272304535, | |
| "learning_rate": 1.4254581800230306e-05, | |
| "loss": 0.0017, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 3.975, | |
| "grad_norm": 0.5810887813568115, | |
| "learning_rate": 1.3915186995462918e-05, | |
| "loss": 0.1409, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.049918390810489655, | |
| "learning_rate": 1.3575792190695532e-05, | |
| "loss": 0.0492, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8925, | |
| "eval_f1": 0.8286852589641435, | |
| "eval_loss": 0.6152504682540894, | |
| "eval_precision": 0.8125, | |
| "eval_recall": 0.8455284552845529, | |
| "eval_runtime": 1.5277, | |
| "eval_samples_per_second": 261.825, | |
| "eval_steps_per_second": 16.364, | |
| "step": 1600 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 847261481803776.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 6.787896095347766e-05, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 4, | |
| "seed": 39 | |
| } | |
| } | |