| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 732, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 9.08570438897306, | |
| "learning_rate": 2.702702702702703e-07, | |
| "loss": 1.1396, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 9.679962246384667, | |
| "learning_rate": 1.3513513513513515e-06, | |
| "loss": 1.1779, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 10.102509592240429, | |
| "learning_rate": 2.702702702702703e-06, | |
| "loss": 1.1318, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 2.9319421528072973, | |
| "learning_rate": 4.0540540540540545e-06, | |
| "loss": 1.0028, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.3759757212785226, | |
| "learning_rate": 5.405405405405406e-06, | |
| "loss": 0.926, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.9504372792893874, | |
| "learning_rate": 6.7567567567567575e-06, | |
| "loss": 0.9044, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.2271132587175222, | |
| "learning_rate": 8.108108108108109e-06, | |
| "loss": 0.8775, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.8247319038566833, | |
| "learning_rate": 9.45945945945946e-06, | |
| "loss": 0.868, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.6839207898039522, | |
| "learning_rate": 1.0810810810810812e-05, | |
| "loss": 0.8626, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.6459463481231297, | |
| "learning_rate": 1.2162162162162164e-05, | |
| "loss": 0.8454, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.5903349665251353, | |
| "learning_rate": 1.3513513513513515e-05, | |
| "loss": 0.8437, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.5948338223404456, | |
| "learning_rate": 1.4864864864864865e-05, | |
| "loss": 0.8353, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.5895070475596026, | |
| "learning_rate": 1.6216216216216218e-05, | |
| "loss": 0.8222, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.6066852387492704, | |
| "learning_rate": 1.756756756756757e-05, | |
| "loss": 0.8144, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.602310053600098, | |
| "learning_rate": 1.891891891891892e-05, | |
| "loss": 0.821, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.5621279417664072, | |
| "learning_rate": 1.999988602302209e-05, | |
| "loss": 0.8133, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.6630630012402627, | |
| "learning_rate": 1.9995897101594454e-05, | |
| "loss": 0.8334, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.6077906778917731, | |
| "learning_rate": 1.99862119291555e-05, | |
| "loss": 0.804, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.5618472092387999, | |
| "learning_rate": 1.997083602488702e-05, | |
| "loss": 0.8193, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.6333601185724956, | |
| "learning_rate": 1.994977815088504e-05, | |
| "loss": 0.7945, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.6001887764474764, | |
| "learning_rate": 1.9923050307166655e-05, | |
| "loss": 0.8079, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.6114725172341102, | |
| "learning_rate": 1.989066772483171e-05, | |
| "loss": 0.8063, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.6410224419207078, | |
| "learning_rate": 1.9852648857383224e-05, | |
| "loss": 0.8079, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.6155008789371026, | |
| "learning_rate": 1.9809015370211505e-05, | |
| "loss": 0.8012, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.5701298765644415, | |
| "learning_rate": 1.9759792128247922e-05, | |
| "loss": 0.7858, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.5769994850814212, | |
| "learning_rate": 1.9705007181795416e-05, | |
| "loss": 0.7949, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.5989802045117304, | |
| "learning_rate": 1.964469175054377e-05, | |
| "loss": 0.7912, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.6158332171463875, | |
| "learning_rate": 1.9578880205778793e-05, | |
| "loss": 0.8, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.6045880045462032, | |
| "learning_rate": 1.950761005079556e-05, | |
| "loss": 0.7831, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.6022037882647541, | |
| "learning_rate": 1.9430921899526786e-05, | |
| "loss": 0.7967, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.5703057672919921, | |
| "learning_rate": 1.934885945339865e-05, | |
| "loss": 0.7851, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.5660007442989872, | |
| "learning_rate": 1.9261469476427122e-05, | |
| "loss": 0.7699, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.6052031318507304, | |
| "learning_rate": 1.916880176856909e-05, | |
| "loss": 0.787, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.5691776960805929, | |
| "learning_rate": 1.907090913734341e-05, | |
| "loss": 0.7851, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.6074889597948341, | |
| "learning_rate": 1.896784736773805e-05, | |
| "loss": 0.7753, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.5860864302531827, | |
| "learning_rate": 1.885967519042054e-05, | |
| "loss": 0.7871, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.569482390256513, | |
| "learning_rate": 1.8746454248269777e-05, | |
| "loss": 0.7601, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.5854282191729411, | |
| "learning_rate": 1.862824906124826e-05, | |
| "loss": 0.7819, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.6428086559905823, | |
| "learning_rate": 1.850512698963485e-05, | |
| "loss": 0.7831, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.5698221365710079, | |
| "learning_rate": 1.8377158195638877e-05, | |
| "loss": 0.7604, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.7373917549594062, | |
| "learning_rate": 1.8244415603417603e-05, | |
| "loss": 0.7587, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.585209211283403, | |
| "learning_rate": 1.8106974857519737e-05, | |
| "loss": 0.7804, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.6168183287266534, | |
| "learning_rate": 1.7964914279778716e-05, | |
| "loss": 0.7797, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.625316831394029, | |
| "learning_rate": 1.78183148246803e-05, | |
| "loss": 0.7583, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.5564471122778261, | |
| "learning_rate": 1.7667260033229953e-05, | |
| "loss": 0.7512, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.5890010216830359, | |
| "learning_rate": 1.751183598534625e-05, | |
| "loss": 0.7528, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.6009170802635413, | |
| "learning_rate": 1.7352131250807466e-05, | |
| "loss": 0.7626, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.5792443910657085, | |
| "learning_rate": 1.7188236838779297e-05, | |
| "loss": 0.7632, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.5716562571066842, | |
| "learning_rate": 1.702024614595248e-05, | |
| "loss": 0.7551, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.5703410316230326, | |
| "learning_rate": 1.6848254903319866e-05, | |
| "loss": 0.7427, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.6094207116793442, | |
| "learning_rate": 1.6672361121623238e-05, | |
| "loss": 0.7943, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.5915494337632395, | |
| "learning_rate": 1.6492665035501048e-05, | |
| "loss": 0.7373, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.5958263340568929, | |
| "learning_rate": 1.6309269046368777e-05, | |
| "loss": 0.7523, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.6569542499394689, | |
| "learning_rate": 1.612227766406461e-05, | |
| "loss": 0.755, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.5775231523651057, | |
| "learning_rate": 1.5931797447293553e-05, | |
| "loss": 0.7292, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.6322564996445303, | |
| "learning_rate": 1.5737936942904025e-05, | |
| "loss": 0.7476, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.5522985173213607, | |
| "learning_rate": 1.554080662403144e-05, | |
| "loss": 0.7427, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.6314561016931608, | |
| "learning_rate": 1.5340518827144145e-05, | |
| "loss": 0.7602, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.5865596324052452, | |
| "learning_rate": 1.5137187688027437e-05, | |
| "loss": 0.7496, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.5697597752265093, | |
| "learning_rate": 1.4930929076742317e-05, | |
| "loss": 0.7392, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.6129791987688828, | |
| "learning_rate": 1.4721860531595868e-05, | |
| "loss": 0.7537, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.5897396431474848, | |
| "learning_rate": 1.451010119216102e-05, | |
| "loss": 0.7583, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.5710567280058013, | |
| "learning_rate": 1.4295771731383799e-05, | |
| "loss": 0.7275, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.5953176828826594, | |
| "learning_rate": 1.4078994286816768e-05, | |
| "loss": 0.7453, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.6242032112691839, | |
| "learning_rate": 1.3859892391017867e-05, | |
| "loss": 0.7644, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.6014365104233058, | |
| "learning_rate": 1.3638590901154276e-05, | |
| "loss": 0.7532, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.5726135101483418, | |
| "learning_rate": 1.341521592785145e-05, | |
| "loss": 0.7663, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.6061053656685103, | |
| "learning_rate": 1.3189894763327851e-05, | |
| "loss": 0.7476, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.5910382927751472, | |
| "learning_rate": 1.2962755808856341e-05, | |
| "loss": 0.7418, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.6192359895872668, | |
| "learning_rate": 1.2733928501593587e-05, | |
| "loss": 0.7322, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.5754476677307057, | |
| "learning_rate": 1.2503543240819127e-05, | |
| "loss": 0.7505, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.5954914603467452, | |
| "learning_rate": 1.227173131362619e-05, | |
| "loss": 0.7316, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.636714901981685, | |
| "learning_rate": 1.2038624820106572e-05, | |
| "loss": 0.7483, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.5956828212931418, | |
| "learning_rate": 1.1804356598072223e-05, | |
| "loss": 0.7141, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.6438375663233106, | |
| "learning_rate": 1.1569060147356441e-05, | |
| "loss": 0.7589, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.5911783265386276, | |
| "learning_rate": 1.133286955373779e-05, | |
| "loss": 0.7485, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.5615839698644924, | |
| "learning_rate": 1.1095919412530136e-05, | |
| "loss": 0.75, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.5906964619382736, | |
| "learning_rate": 1.0858344751882304e-05, | |
| "loss": 0.7089, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.622095376177676, | |
| "learning_rate": 1.0620280955831088e-05, | |
| "loss": 0.7418, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.5629629729406987, | |
| "learning_rate": 1.038186368715145e-05, | |
| "loss": 0.7148, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.5596961155705783, | |
| "learning_rate": 1.0143228810047877e-05, | |
| "loss": 0.7226, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.5949479844701526, | |
| "learning_rate": 9.904512312730948e-06, | |
| "loss": 0.7446, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.5503557697571823, | |
| "learning_rate": 9.665850229923258e-06, | |
| "loss": 0.7294, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.5646735246212238, | |
| "learning_rate": 9.4273785653388e-06, | |
| "loss": 0.7399, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.6166612666135994, | |
| "learning_rate": 9.189233214180057e-06, | |
| "loss": 0.7464, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.6607877653922463, | |
| "learning_rate": 8.951549885696889e-06, | |
| "loss": 0.7104, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.5828440632986862, | |
| "learning_rate": 8.714464025851428e-06, | |
| "loss": 0.748, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.6692052000569471, | |
| "learning_rate": 8.478110740132971e-06, | |
| "loss": 0.7194, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.5925544656331848, | |
| "learning_rate": 8.242624716566928e-06, | |
| "loss": 0.7414, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.5681970472775137, | |
| "learning_rate": 8.008140148961642e-06, | |
| "loss": 0.7266, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.6022566260644191, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.7458, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.6280077977218962, | |
| "learning_rate": 7.542709227277396e-06, | |
| "loss": 0.7265, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.5662044508467975, | |
| "learning_rate": 7.312028103155426e-06, | |
| "loss": 0.7296, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.5875195935745249, | |
| "learning_rate": 7.0828787437645455e-06, | |
| "loss": 0.7666, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.5878339828851088, | |
| "learning_rate": 6.8553917319085676e-06, | |
| "loss": 0.7478, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.552328272875581, | |
| "learning_rate": 6.629696703087755e-06, | |
| "loss": 0.7237, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.6071223669952857, | |
| "learning_rate": 6.405922271624874e-06, | |
| "loss": 0.7357, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.6060821252643572, | |
| "learning_rate": 6.184195957373176e-06, | |
| "loss": 0.7178, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.543636900637841, | |
| "learning_rate": 5.964644113048079e-06, | |
| "loss": 0.7264, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.603596293760061, | |
| "learning_rate": 5.74739185222394e-06, | |
| "loss": 0.7415, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.5558735174394915, | |
| "learning_rate": 5.532562978036964e-06, | |
| "loss": 0.7431, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.5937971276012349, | |
| "learning_rate": 5.320279912634907e-06, | |
| "loss": 0.7316, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.5786141927192617, | |
| "learning_rate": 5.110663627413695e-06, | |
| "loss": 0.713, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.5809275145865697, | |
| "learning_rate": 4.903833574080825e-06, | |
| "loss": 0.7273, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.571032903937194, | |
| "learning_rate": 4.6999076165847214e-06, | |
| "loss": 0.7128, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.6308749156115376, | |
| "learning_rate": 4.499001963948929e-06, | |
| "loss": 0.727, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.5475000263845851, | |
| "learning_rate": 4.301231104049359e-06, | |
| "loss": 0.7236, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.5451654896042127, | |
| "learning_rate": 4.106707738372357e-06, | |
| "loss": 0.717, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.5895012216093011, | |
| "learning_rate": 3.915542717790759e-06, | |
| "loss": 0.7253, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.5991508019426129, | |
| "learning_rate": 3.727844979394526e-06, | |
| "loss": 0.7108, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.6024026652126097, | |
| "learning_rate": 3.543721484411976e-06, | |
| "loss": 0.7148, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.5479626437376189, | |
| "learning_rate": 3.3632771572569878e-06, | |
| "loss": 0.7205, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.5637951714589972, | |
| "learning_rate": 3.1866148257368666e-06, | |
| "loss": 0.7291, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.5971328138449835, | |
| "learning_rate": 3.0138351624550165e-06, | |
| "loss": 0.7242, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.5566178272277884, | |
| "learning_rate": 2.845036627441755e-06, | |
| "loss": 0.725, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.5747018707425353, | |
| "learning_rate": 2.6803154120460007e-06, | |
| "loss": 0.7232, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.583278019846911, | |
| "learning_rate": 2.5197653841197546e-06, | |
| "loss": 0.7593, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.5792687056211038, | |
| "learning_rate": 2.3634780345266805e-06, | |
| "loss": 0.7263, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.5936774049714754, | |
| "learning_rate": 2.211542425005223e-06, | |
| "loss": 0.7105, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.5618344372585139, | |
| "learning_rate": 2.064045137415982e-06, | |
| "loss": 0.7148, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.5753081059010117, | |
| "learning_rate": 1.9210702244022616e-06, | |
| "loss": 0.7313, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.5418687310157559, | |
| "learning_rate": 1.7826991614919264e-06, | |
| "loss": 0.7245, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.5828322063852037, | |
| "learning_rate": 1.6490108006678495e-06, | |
| "loss": 0.7527, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.5878067843611324, | |
| "learning_rate": 1.5200813254334013e-06, | |
| "loss": 0.7213, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.5816156229837385, | |
| "learning_rate": 1.3959842073986085e-06, | |
| "loss": 0.7134, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.5962667792304962, | |
| "learning_rate": 1.2767901644116943e-06, | |
| "loss": 0.7233, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.5399112934132381, | |
| "learning_rate": 1.1625671202598875e-06, | |
| "loss": 0.7179, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.5397855641716054, | |
| "learning_rate": 1.0533801659624531e-06, | |
| "loss": 0.7114, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.5936653550047086, | |
| "learning_rate": 9.492915226779809e-07, | |
| "loss": 0.7175, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.6099602078388837, | |
| "learning_rate": 8.503605062471187e-07, | |
| "loss": 0.7119, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.6177660757674084, | |
| "learning_rate": 7.566434933909006e-07, | |
| "loss": 0.7132, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.5217732163513131, | |
| "learning_rate": 6.681938895839746e-07, | |
| "loss": 0.7128, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.5808565758298766, | |
| "learning_rate": 5.850620986210198e-07, | |
| "loss": 0.6972, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.5829050029411651, | |
| "learning_rate": 5.072954938936925e-07, | |
| "loss": 0.7253, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.5481607356326386, | |
| "learning_rate": 4.3493839139447716e-07, | |
| "loss": 0.7222, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.5652882598658524, | |
| "learning_rate": 3.6803202446282217e-07, | |
| "loss": 0.6845, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.5796535522251339, | |
| "learning_rate": 3.0661452028795335e-07, | |
| "loss": 0.7148, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.5887255027538941, | |
| "learning_rate": 2.507208781817638e-07, | |
| "loss": 0.7224, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.5592326454997533, | |
| "learning_rate": 2.0038294963413251e-07, | |
| "loss": 0.7113, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.5776817470975002, | |
| "learning_rate": 1.556294201620734e-07, | |
| "loss": 0.7033, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.5696597587226961, | |
| "learning_rate": 1.1648579296304252e-07, | |
| "loss": 0.7277, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.5641386210392492, | |
| "learning_rate": 8.297437438170797e-08, | |
| "loss": 0.7256, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.5889080994491116, | |
| "learning_rate": 5.51142611984834e-08, | |
| "loss": 0.7279, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.5665974446341169, | |
| "learning_rate": 3.2921329747056527e-08, | |
| "loss": 0.7311, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.5822891522453822, | |
| "learning_rate": 1.6408226867118404e-08, | |
| "loss": 0.7131, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.5631956425423463, | |
| "learning_rate": 5.584362697453882e-09, | |
| "loss": 0.7071, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.563567530638776, | |
| "learning_rate": 4.5590531348227443e-10, | |
| "loss": 0.7119, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.7456573247909546, | |
| "eval_runtime": 9.4968, | |
| "eval_samples_per_second": 52.649, | |
| "eval_steps_per_second": 1.685, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 732, | |
| "total_flos": 106029000622080.0, | |
| "train_loss": 0.7605658245216953, | |
| "train_runtime": 6107.4706, | |
| "train_samples_per_second": 15.333, | |
| "train_steps_per_second": 0.12 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 732, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 106029000622080.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |