| { | |
| "best_global_step": 675, | |
| "best_metric": 0.6912277936935425, | |
| "best_model_checkpoint": "checkpoints/rft-finetune-llama-3.1-8b-gsm8k/gsm8k/finetune-llama-3.1-8b-gsm8k-step-1/checkpoint-675", | |
| "epoch": 1.0, | |
| "eval_steps": 75, | |
| "global_step": 744, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.006720430107526882, | |
| "grad_norm": 14.59550096035499, | |
| "learning_rate": 5.333333333333333e-09, | |
| "loss": 1.1893, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.013440860215053764, | |
| "grad_norm": 15.552214878649293, | |
| "learning_rate": 1.1999999999999998e-08, | |
| "loss": 1.1733, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.020161290322580645, | |
| "grad_norm": 16.885206147000304, | |
| "learning_rate": 1.8666666666666665e-08, | |
| "loss": 1.1934, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.026881720430107527, | |
| "grad_norm": 15.221916137414837, | |
| "learning_rate": 2.5333333333333335e-08, | |
| "loss": 1.2152, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.033602150537634407, | |
| "grad_norm": 15.887269098975871, | |
| "learning_rate": 3.2e-08, | |
| "loss": 1.2051, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.04032258064516129, | |
| "grad_norm": 14.254790289709463, | |
| "learning_rate": 3.866666666666666e-08, | |
| "loss": 1.1997, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04704301075268817, | |
| "grad_norm": 14.583670649351852, | |
| "learning_rate": 4.533333333333333e-08, | |
| "loss": 1.1812, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.053763440860215055, | |
| "grad_norm": 14.695116879291863, | |
| "learning_rate": 5.2e-08, | |
| "loss": 1.195, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.06048387096774194, | |
| "grad_norm": 14.808886364711231, | |
| "learning_rate": 5.866666666666666e-08, | |
| "loss": 1.19, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06720430107526881, | |
| "grad_norm": 14.279020954727395, | |
| "learning_rate": 6.533333333333332e-08, | |
| "loss": 1.1758, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0739247311827957, | |
| "grad_norm": 15.710193271146736, | |
| "learning_rate": 7.2e-08, | |
| "loss": 1.1654, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.08064516129032258, | |
| "grad_norm": 12.823166713617123, | |
| "learning_rate": 7.866666666666666e-08, | |
| "loss": 1.0956, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08736559139784947, | |
| "grad_norm": 13.123484893071117, | |
| "learning_rate": 8.533333333333333e-08, | |
| "loss": 1.0939, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.09408602150537634, | |
| "grad_norm": 13.804028711760814, | |
| "learning_rate": 9.2e-08, | |
| "loss": 1.0806, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.10080645161290322, | |
| "grad_norm": 14.362479314024831, | |
| "learning_rate": 9.866666666666666e-08, | |
| "loss": 1.0428, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.10080645161290322, | |
| "eval_loss": 1.0135635137557983, | |
| "eval_runtime": 64.0958, | |
| "eval_samples_per_second": 128.277, | |
| "eval_steps_per_second": 2.013, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.10752688172043011, | |
| "grad_norm": 12.804236982581402, | |
| "learning_rate": 9.940209267563527e-08, | |
| "loss": 0.9585, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.11424731182795698, | |
| "grad_norm": 11.669278429338862, | |
| "learning_rate": 9.865470852017936e-08, | |
| "loss": 0.9112, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.12096774193548387, | |
| "grad_norm": 11.081808260451195, | |
| "learning_rate": 9.790732436472347e-08, | |
| "loss": 0.877, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.12768817204301075, | |
| "grad_norm": 16.49147962749734, | |
| "learning_rate": 9.715994020926755e-08, | |
| "loss": 0.8504, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.13440860215053763, | |
| "grad_norm": 13.95628735293844, | |
| "learning_rate": 9.641255605381165e-08, | |
| "loss": 0.7931, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.14112903225806453, | |
| "grad_norm": 8.601123066406517, | |
| "learning_rate": 9.566517189835575e-08, | |
| "loss": 0.711, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.1478494623655914, | |
| "grad_norm": 3.7294133148506337, | |
| "learning_rate": 9.491778774289984e-08, | |
| "loss": 0.6981, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.15456989247311828, | |
| "grad_norm": 2.723138779265988, | |
| "learning_rate": 9.417040358744395e-08, | |
| "loss": 0.6953, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.16129032258064516, | |
| "grad_norm": 2.2560223539756423, | |
| "learning_rate": 9.342301943198804e-08, | |
| "loss": 0.6846, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.16801075268817203, | |
| "grad_norm": 2.2521211316559433, | |
| "learning_rate": 9.267563527653213e-08, | |
| "loss": 0.6696, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.17473118279569894, | |
| "grad_norm": 2.2818251781816588, | |
| "learning_rate": 9.192825112107622e-08, | |
| "loss": 0.6809, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.1814516129032258, | |
| "grad_norm": 2.44710559044971, | |
| "learning_rate": 9.118086696562033e-08, | |
| "loss": 0.6841, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.1881720430107527, | |
| "grad_norm": 2.350544704349094, | |
| "learning_rate": 9.043348281016442e-08, | |
| "loss": 0.6766, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.19489247311827956, | |
| "grad_norm": 1.9640158046302565, | |
| "learning_rate": 8.968609865470852e-08, | |
| "loss": 0.6878, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.20161290322580644, | |
| "grad_norm": 2.0782198781659353, | |
| "learning_rate": 8.893871449925261e-08, | |
| "loss": 0.6636, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.20161290322580644, | |
| "eval_loss": 0.6971947550773621, | |
| "eval_runtime": 56.1638, | |
| "eval_samples_per_second": 146.393, | |
| "eval_steps_per_second": 2.297, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.20833333333333334, | |
| "grad_norm": 2.1288692138731458, | |
| "learning_rate": 8.819133034379672e-08, | |
| "loss": 0.6866, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.21505376344086022, | |
| "grad_norm": 2.2152810121981568, | |
| "learning_rate": 8.74439461883408e-08, | |
| "loss": 0.6884, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.2217741935483871, | |
| "grad_norm": 1.9700569516052628, | |
| "learning_rate": 8.66965620328849e-08, | |
| "loss": 0.6793, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.22849462365591397, | |
| "grad_norm": 2.0535897043325724, | |
| "learning_rate": 8.5949177877429e-08, | |
| "loss": 0.6777, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.23521505376344087, | |
| "grad_norm": 2.011966313411605, | |
| "learning_rate": 8.520179372197309e-08, | |
| "loss": 0.6596, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.24193548387096775, | |
| "grad_norm": 2.0907303106285164, | |
| "learning_rate": 8.445440956651718e-08, | |
| "loss": 0.6713, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.24865591397849462, | |
| "grad_norm": 1.9870643151130227, | |
| "learning_rate": 8.370702541106129e-08, | |
| "loss": 0.6835, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.2553763440860215, | |
| "grad_norm": 2.1110324099341797, | |
| "learning_rate": 8.295964125560538e-08, | |
| "loss": 0.6691, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.2620967741935484, | |
| "grad_norm": 2.337028306260816, | |
| "learning_rate": 8.221225710014947e-08, | |
| "loss": 0.6706, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.26881720430107525, | |
| "grad_norm": 1.9987914444654455, | |
| "learning_rate": 8.146487294469356e-08, | |
| "loss": 0.6769, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.27553763440860213, | |
| "grad_norm": 2.1558155606241884, | |
| "learning_rate": 8.071748878923767e-08, | |
| "loss": 0.6709, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.28225806451612906, | |
| "grad_norm": 2.09004359899187, | |
| "learning_rate": 7.997010463378176e-08, | |
| "loss": 0.6784, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.28897849462365593, | |
| "grad_norm": 2.285145499206384, | |
| "learning_rate": 7.922272047832586e-08, | |
| "loss": 0.6715, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.2956989247311828, | |
| "grad_norm": 2.027191135310369, | |
| "learning_rate": 7.847533632286996e-08, | |
| "loss": 0.6612, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.3024193548387097, | |
| "grad_norm": 2.0151475678114705, | |
| "learning_rate": 7.772795216741404e-08, | |
| "loss": 0.6676, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.3024193548387097, | |
| "eval_loss": 0.6939067244529724, | |
| "eval_runtime": 56.1322, | |
| "eval_samples_per_second": 146.476, | |
| "eval_steps_per_second": 2.298, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.30913978494623656, | |
| "grad_norm": 1.9995401054673094, | |
| "learning_rate": 7.698056801195815e-08, | |
| "loss": 0.6593, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.31586021505376344, | |
| "grad_norm": 2.0110294954158454, | |
| "learning_rate": 7.623318385650224e-08, | |
| "loss": 0.6563, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.3225806451612903, | |
| "grad_norm": 2.059981924661176, | |
| "learning_rate": 7.548579970104633e-08, | |
| "loss": 0.6764, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.3293010752688172, | |
| "grad_norm": 2.149345633891912, | |
| "learning_rate": 7.473841554559043e-08, | |
| "loss": 0.6645, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.33602150537634407, | |
| "grad_norm": 2.101850865821983, | |
| "learning_rate": 7.399103139013453e-08, | |
| "loss": 0.6612, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.34274193548387094, | |
| "grad_norm": 2.1254515490062245, | |
| "learning_rate": 7.324364723467862e-08, | |
| "loss": 0.6773, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.34946236559139787, | |
| "grad_norm": 2.08043259609194, | |
| "learning_rate": 7.249626307922272e-08, | |
| "loss": 0.6826, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.35618279569892475, | |
| "grad_norm": 1.972159359225326, | |
| "learning_rate": 7.174887892376681e-08, | |
| "loss": 0.6736, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.3629032258064516, | |
| "grad_norm": 2.020214538372761, | |
| "learning_rate": 7.100149476831092e-08, | |
| "loss": 0.6628, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.3696236559139785, | |
| "grad_norm": 2.119471507884024, | |
| "learning_rate": 7.0254110612855e-08, | |
| "loss": 0.6628, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.3763440860215054, | |
| "grad_norm": 2.2339027083129803, | |
| "learning_rate": 6.95067264573991e-08, | |
| "loss": 0.6461, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.38306451612903225, | |
| "grad_norm": 2.179556694440606, | |
| "learning_rate": 6.87593423019432e-08, | |
| "loss": 0.6685, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.3897849462365591, | |
| "grad_norm": 2.0979269829823477, | |
| "learning_rate": 6.801195814648729e-08, | |
| "loss": 0.6688, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.396505376344086, | |
| "grad_norm": 2.145759818638723, | |
| "learning_rate": 6.72645739910314e-08, | |
| "loss": 0.6582, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.4032258064516129, | |
| "grad_norm": 2.0232579108607998, | |
| "learning_rate": 6.651718983557549e-08, | |
| "loss": 0.6849, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4032258064516129, | |
| "eval_loss": 0.6928321123123169, | |
| "eval_runtime": 56.1826, | |
| "eval_samples_per_second": 146.344, | |
| "eval_steps_per_second": 2.296, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4099462365591398, | |
| "grad_norm": 2.215763186449215, | |
| "learning_rate": 6.576980568011958e-08, | |
| "loss": 0.6772, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.4166666666666667, | |
| "grad_norm": 1.9274901065614445, | |
| "learning_rate": 6.502242152466367e-08, | |
| "loss": 0.691, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.42338709677419356, | |
| "grad_norm": 2.108143183803596, | |
| "learning_rate": 6.427503736920778e-08, | |
| "loss": 0.6628, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.43010752688172044, | |
| "grad_norm": 2.207966702792586, | |
| "learning_rate": 6.352765321375186e-08, | |
| "loss": 0.6768, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.4368279569892473, | |
| "grad_norm": 2.096647992540451, | |
| "learning_rate": 6.278026905829596e-08, | |
| "loss": 0.6667, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.4435483870967742, | |
| "grad_norm": 2.1336095512755113, | |
| "learning_rate": 6.203288490284006e-08, | |
| "loss": 0.6581, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.45026881720430106, | |
| "grad_norm": 2.0159699059441976, | |
| "learning_rate": 6.128550074738415e-08, | |
| "loss": 0.6743, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.45698924731182794, | |
| "grad_norm": 1.9979710525534193, | |
| "learning_rate": 6.053811659192824e-08, | |
| "loss": 0.6638, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.4637096774193548, | |
| "grad_norm": 2.049384366870693, | |
| "learning_rate": 5.979073243647235e-08, | |
| "loss": 0.679, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.47043010752688175, | |
| "grad_norm": 2.147547124197116, | |
| "learning_rate": 5.9043348281016435e-08, | |
| "loss": 0.6719, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.4771505376344086, | |
| "grad_norm": 1.994604705245644, | |
| "learning_rate": 5.8295964125560534e-08, | |
| "loss": 0.6622, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.4838709677419355, | |
| "grad_norm": 2.035462459844569, | |
| "learning_rate": 5.754857997010463e-08, | |
| "loss": 0.6464, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.4905913978494624, | |
| "grad_norm": 2.042123968621037, | |
| "learning_rate": 5.6801195814648727e-08, | |
| "loss": 0.6627, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.49731182795698925, | |
| "grad_norm": 2.14484478856972, | |
| "learning_rate": 5.605381165919282e-08, | |
| "loss": 0.6522, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.5040322580645161, | |
| "grad_norm": 2.0812193752637462, | |
| "learning_rate": 5.530642750373692e-08, | |
| "loss": 0.6858, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.5040322580645161, | |
| "eval_loss": 0.6924068927764893, | |
| "eval_runtime": 56.1089, | |
| "eval_samples_per_second": 146.536, | |
| "eval_steps_per_second": 2.299, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.510752688172043, | |
| "grad_norm": 2.1209544150897757, | |
| "learning_rate": 5.455904334828101e-08, | |
| "loss": 0.6451, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.5174731182795699, | |
| "grad_norm": 2.045170736604748, | |
| "learning_rate": 5.381165919282511e-08, | |
| "loss": 0.6624, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.5241935483870968, | |
| "grad_norm": 2.10236284260544, | |
| "learning_rate": 5.306427503736921e-08, | |
| "loss": 0.6741, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.5309139784946236, | |
| "grad_norm": 2.1314029429521244, | |
| "learning_rate": 5.2316890881913303e-08, | |
| "loss": 0.6642, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.5376344086021505, | |
| "grad_norm": 2.186951067665247, | |
| "learning_rate": 5.15695067264574e-08, | |
| "loss": 0.6836, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5443548387096774, | |
| "grad_norm": 2.052806350954309, | |
| "learning_rate": 5.082212257100149e-08, | |
| "loss": 0.672, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 0.5510752688172043, | |
| "grad_norm": 2.071189416663902, | |
| "learning_rate": 5.0074738415545595e-08, | |
| "loss": 0.6677, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.5577956989247311, | |
| "grad_norm": 2.0118486297022833, | |
| "learning_rate": 4.932735426008968e-08, | |
| "loss": 0.6639, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 0.5645161290322581, | |
| "grad_norm": 2.0632350448343333, | |
| "learning_rate": 4.8579970104633774e-08, | |
| "loss": 0.6754, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.571236559139785, | |
| "grad_norm": 2.251163063469807, | |
| "learning_rate": 4.7832585949177874e-08, | |
| "loss": 0.7003, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.5779569892473119, | |
| "grad_norm": 2.0661342085354697, | |
| "learning_rate": 4.708520179372197e-08, | |
| "loss": 0.6904, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.5846774193548387, | |
| "grad_norm": 2.1557028238109517, | |
| "learning_rate": 4.6337817638266066e-08, | |
| "loss": 0.6318, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.5913978494623656, | |
| "grad_norm": 2.1516536049922452, | |
| "learning_rate": 4.5590433482810165e-08, | |
| "loss": 0.6552, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.5981182795698925, | |
| "grad_norm": 2.086935790084074, | |
| "learning_rate": 4.484304932735426e-08, | |
| "loss": 0.6778, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 0.6048387096774194, | |
| "grad_norm": 2.146673007914857, | |
| "learning_rate": 4.409566517189836e-08, | |
| "loss": 0.6536, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.6048387096774194, | |
| "eval_loss": 0.6919218897819519, | |
| "eval_runtime": 56.1387, | |
| "eval_samples_per_second": 146.459, | |
| "eval_steps_per_second": 2.298, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.6115591397849462, | |
| "grad_norm": 2.1253216816067066, | |
| "learning_rate": 4.334828101644245e-08, | |
| "loss": 0.6511, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.6182795698924731, | |
| "grad_norm": 2.071737757767298, | |
| "learning_rate": 4.260089686098654e-08, | |
| "loss": 0.6592, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 2.0932212158820396, | |
| "learning_rate": 4.185351270553064e-08, | |
| "loss": 0.6692, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 0.6317204301075269, | |
| "grad_norm": 2.0910146959652307, | |
| "learning_rate": 4.1106128550074736e-08, | |
| "loss": 0.6683, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.6384408602150538, | |
| "grad_norm": 2.062557270254667, | |
| "learning_rate": 4.0358744394618835e-08, | |
| "loss": 0.6656, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.6451612903225806, | |
| "grad_norm": 2.017329078129843, | |
| "learning_rate": 3.961136023916293e-08, | |
| "loss": 0.6719, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.6518817204301075, | |
| "grad_norm": 2.2036353665493755, | |
| "learning_rate": 3.886397608370702e-08, | |
| "loss": 0.682, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 0.6586021505376344, | |
| "grad_norm": 2.007622155160972, | |
| "learning_rate": 3.811659192825112e-08, | |
| "loss": 0.6486, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.6653225806451613, | |
| "grad_norm": 2.2339142859626553, | |
| "learning_rate": 3.736920777279521e-08, | |
| "loss": 0.6542, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 0.6720430107526881, | |
| "grad_norm": 2.1968231695887823, | |
| "learning_rate": 3.662182361733931e-08, | |
| "loss": 0.6701, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.678763440860215, | |
| "grad_norm": 2.311102123656589, | |
| "learning_rate": 3.5874439461883405e-08, | |
| "loss": 0.6541, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 0.6854838709677419, | |
| "grad_norm": 2.162411566728306, | |
| "learning_rate": 3.51270553064275e-08, | |
| "loss": 0.6609, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.6922043010752689, | |
| "grad_norm": 2.0777352845523156, | |
| "learning_rate": 3.43796711509716e-08, | |
| "loss": 0.6654, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 0.6989247311827957, | |
| "grad_norm": 2.0466624810725613, | |
| "learning_rate": 3.36322869955157e-08, | |
| "loss": 0.6637, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.7056451612903226, | |
| "grad_norm": 2.046164435882464, | |
| "learning_rate": 3.288490284005979e-08, | |
| "loss": 0.6672, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.7056451612903226, | |
| "eval_loss": 0.691673994064331, | |
| "eval_runtime": 56.2018, | |
| "eval_samples_per_second": 146.294, | |
| "eval_steps_per_second": 2.295, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.7123655913978495, | |
| "grad_norm": 2.098676297617592, | |
| "learning_rate": 3.213751868460389e-08, | |
| "loss": 0.6581, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.7190860215053764, | |
| "grad_norm": 1.9986286959461719, | |
| "learning_rate": 3.139013452914798e-08, | |
| "loss": 0.6658, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 0.7258064516129032, | |
| "grad_norm": 2.1195333643509935, | |
| "learning_rate": 3.0642750373692075e-08, | |
| "loss": 0.669, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.7325268817204301, | |
| "grad_norm": 2.055327535910759, | |
| "learning_rate": 2.9895366218236174e-08, | |
| "loss": 0.6663, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 0.739247311827957, | |
| "grad_norm": 2.074614293697511, | |
| "learning_rate": 2.9147982062780267e-08, | |
| "loss": 0.6627, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.7459677419354839, | |
| "grad_norm": 2.0360906767754, | |
| "learning_rate": 2.8400597907324363e-08, | |
| "loss": 0.6619, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 0.7526881720430108, | |
| "grad_norm": 2.098550310165075, | |
| "learning_rate": 2.765321375186846e-08, | |
| "loss": 0.6791, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.7594086021505376, | |
| "grad_norm": 2.056390058629133, | |
| "learning_rate": 2.6905829596412556e-08, | |
| "loss": 0.6569, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 0.7661290322580645, | |
| "grad_norm": 2.2413055374611632, | |
| "learning_rate": 2.6158445440956652e-08, | |
| "loss": 0.6704, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.7728494623655914, | |
| "grad_norm": 2.054426280985883, | |
| "learning_rate": 2.5411061285500745e-08, | |
| "loss": 0.6563, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.7795698924731183, | |
| "grad_norm": 2.0006598598366345, | |
| "learning_rate": 2.466367713004484e-08, | |
| "loss": 0.6567, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.7862903225806451, | |
| "grad_norm": 2.2751507511393925, | |
| "learning_rate": 2.3916292974588937e-08, | |
| "loss": 0.6693, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 0.793010752688172, | |
| "grad_norm": 2.1645437231829456, | |
| "learning_rate": 2.3168908819133033e-08, | |
| "loss": 0.6643, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.7997311827956989, | |
| "grad_norm": 2.0220952097349176, | |
| "learning_rate": 2.242152466367713e-08, | |
| "loss": 0.66, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 0.8064516129032258, | |
| "grad_norm": 2.078077165465023, | |
| "learning_rate": 2.1674140508221225e-08, | |
| "loss": 0.6559, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8064516129032258, | |
| "eval_loss": 0.6913867592811584, | |
| "eval_runtime": 56.1429, | |
| "eval_samples_per_second": 146.448, | |
| "eval_steps_per_second": 2.298, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8131720430107527, | |
| "grad_norm": 2.3106172558361884, | |
| "learning_rate": 2.092675635276532e-08, | |
| "loss": 0.6828, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 0.8198924731182796, | |
| "grad_norm": 2.094033510896202, | |
| "learning_rate": 2.0179372197309417e-08, | |
| "loss": 0.6635, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.8266129032258065, | |
| "grad_norm": 2.0658322655908665, | |
| "learning_rate": 1.943198804185351e-08, | |
| "loss": 0.6753, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 2.17250427254849, | |
| "learning_rate": 1.8684603886397606e-08, | |
| "loss": 0.6727, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.8400537634408602, | |
| "grad_norm": 2.259179736932614, | |
| "learning_rate": 1.7937219730941703e-08, | |
| "loss": 0.6959, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.8467741935483871, | |
| "grad_norm": 2.0612328204646815, | |
| "learning_rate": 1.71898355754858e-08, | |
| "loss": 0.6584, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.853494623655914, | |
| "grad_norm": 2.1679992601621154, | |
| "learning_rate": 1.6442451420029895e-08, | |
| "loss": 0.6508, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 0.8602150537634409, | |
| "grad_norm": 1.9886841572370688, | |
| "learning_rate": 1.569506726457399e-08, | |
| "loss": 0.6729, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.8669354838709677, | |
| "grad_norm": 2.1506038261656184, | |
| "learning_rate": 1.4947683109118087e-08, | |
| "loss": 0.6608, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 0.8736559139784946, | |
| "grad_norm": 1.9864785884472476, | |
| "learning_rate": 1.4200298953662182e-08, | |
| "loss": 0.6628, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.8803763440860215, | |
| "grad_norm": 2.1120434621533284, | |
| "learning_rate": 1.3452914798206278e-08, | |
| "loss": 0.6838, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 0.8870967741935484, | |
| "grad_norm": 1.9981627858324142, | |
| "learning_rate": 1.2705530642750372e-08, | |
| "loss": 0.6561, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.8938172043010753, | |
| "grad_norm": 2.050666914131394, | |
| "learning_rate": 1.1958146487294468e-08, | |
| "loss": 0.6491, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 0.9005376344086021, | |
| "grad_norm": 2.159702446356377, | |
| "learning_rate": 1.1210762331838565e-08, | |
| "loss": 0.6783, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.907258064516129, | |
| "grad_norm": 2.0164611126323697, | |
| "learning_rate": 1.046337817638266e-08, | |
| "loss": 0.6472, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.907258064516129, | |
| "eval_loss": 0.6912277936935425, | |
| "eval_runtime": 56.1813, | |
| "eval_samples_per_second": 146.348, | |
| "eval_steps_per_second": 2.296, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.9139784946236559, | |
| "grad_norm": 2.1472859721487683, | |
| "learning_rate": 9.715994020926755e-09, | |
| "loss": 0.6621, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.9206989247311828, | |
| "grad_norm": 2.138066510167175, | |
| "learning_rate": 8.968609865470851e-09, | |
| "loss": 0.6396, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 0.9274193548387096, | |
| "grad_norm": 2.2998888062886538, | |
| "learning_rate": 8.221225710014947e-09, | |
| "loss": 0.6603, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.9341397849462365, | |
| "grad_norm": 2.0384421650226527, | |
| "learning_rate": 7.473841554559044e-09, | |
| "loss": 0.6384, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 0.9408602150537635, | |
| "grad_norm": 2.0927886382588827, | |
| "learning_rate": 6.726457399103139e-09, | |
| "loss": 0.6557, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.9475806451612904, | |
| "grad_norm": 2.02850385366728, | |
| "learning_rate": 5.979073243647234e-09, | |
| "loss": 0.6663, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 0.9543010752688172, | |
| "grad_norm": 2.2433567010088797, | |
| "learning_rate": 5.23168908819133e-09, | |
| "loss": 0.6796, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.9610215053763441, | |
| "grad_norm": 2.057838483040146, | |
| "learning_rate": 4.484304932735426e-09, | |
| "loss": 0.6626, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 0.967741935483871, | |
| "grad_norm": 2.1280821159120786, | |
| "learning_rate": 3.736920777279522e-09, | |
| "loss": 0.633, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.9744623655913979, | |
| "grad_norm": 2.019967451723735, | |
| "learning_rate": 2.989536621823617e-09, | |
| "loss": 0.6718, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.9811827956989247, | |
| "grad_norm": 2.1683941089274477, | |
| "learning_rate": 2.242152466367713e-09, | |
| "loss": 0.6578, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.9879032258064516, | |
| "grad_norm": 2.015824272032049, | |
| "learning_rate": 1.4947683109118085e-09, | |
| "loss": 0.6781, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 0.9946236559139785, | |
| "grad_norm": 2.1171098020664325, | |
| "learning_rate": 7.473841554559043e-10, | |
| "loss": 0.6829, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 744, | |
| "total_flos": 1582562992455680.0, | |
| "train_loss": 0.7245069735793657, | |
| "train_runtime": 3975.4106, | |
| "train_samples_per_second": 11.971, | |
| "train_steps_per_second": 0.187 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 744, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 75, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1582562992455680.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |