| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.17863123813776935, | |
| "eval_steps": 25, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0017863123813776935, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0017863123813776935, | |
| "eval_loss": NaN, | |
| "eval_runtime": 222.0383, | |
| "eval_samples_per_second": 4.247, | |
| "eval_steps_per_second": 2.126, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003572624762755387, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00535893714413308, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.007145249525510774, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.008931561906888466, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01071787428826616, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.012504186669643855, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.014290499051021548, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01607681143239924, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.017863123813776933, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.019649436195154628, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.02143574857653232, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 2.3447, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.023222060957910014, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 109.6991, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02500837333928771, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0267946857206654, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 56.0876, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.028580998102043096, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 150.1053, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.030367310483420788, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.03215362286479848, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.033939935246176174, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.035726247627553866, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.037512560008931564, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.039298872390309256, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.04108518477168695, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 2936.1797, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.04287149715306464, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.04465780953444234, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.04465780953444234, | |
| "eval_loss": NaN, | |
| "eval_runtime": 58.9914, | |
| "eval_samples_per_second": 15.985, | |
| "eval_steps_per_second": 8.001, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.04644412191582003, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.04823043429719772, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.05001674667857542, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.05180305905995311, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 2650.417, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0535893714413308, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.055375683822708494, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 5.2058, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.05716199620408619, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 1481.9423, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.058948308585463884, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.060734620966841575, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.06252093334821927, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.06430724572959697, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.06609355811097466, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1108.9025, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.06787987049235235, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.06966618287373004, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.07145249525510773, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07323880763648544, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.0, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.07502512001786313, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.07681143239924082, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.0, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.07859774478061851, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0803840571619962, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0821703695433739, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.08395668192475159, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.0, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.08574299430612928, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.08752930668750698, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.08931561906888467, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08931561906888467, | |
| "eval_loss": NaN, | |
| "eval_runtime": 58.9667, | |
| "eval_samples_per_second": 15.992, | |
| "eval_steps_per_second": 8.005, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09110193145026237, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 10.501, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.09288824383164006, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.0, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.09467455621301775, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 0.0, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.09646086859439544, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.09824718097577313, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.10003349335715084, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.0, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.10181980573852853, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 10738.2832, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.10360611811990622, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 8.2575, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.10539243050128391, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 2.6762, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.1071787428826616, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1089650552640393, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.0, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.11075136764541699, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.0, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.11253768002679469, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.0, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.11432399240817238, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.11611030478955008, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.0, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.11789661717092777, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.11968292955230546, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 0.0, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.12146924193368315, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 6314.6611, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.12325555431506084, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.0, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.12504186669643855, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 3691.4143, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.12682817907781624, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 234.6905, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.12861449145919393, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 5145.4033, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.13040080384057162, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.1321871162219493, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.0, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.133973428603327, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.133973428603327, | |
| "eval_loss": NaN, | |
| "eval_runtime": 56.0405, | |
| "eval_samples_per_second": 16.827, | |
| "eval_steps_per_second": 8.422, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1357597409847047, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.6543469682057106e-05, | |
| "loss": 0.0, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.1375460533660824, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.526708147705013e-05, | |
| "loss": 0.0, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.13933236574746008, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.14111867812883777, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.2842758726130283e-05, | |
| "loss": 0.0, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.14290499051021546, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 0.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.14469130289159315, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 33.7649, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.14647761527297087, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.14826392765434857, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.548121372247918e-06, | |
| "loss": 5.1139, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.15005024003572626, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.0, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.15183655241710395, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 0.0, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.15362286479848164, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.852620357053651e-06, | |
| "loss": 0.0, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.15540917717985933, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.0, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.15719548956123702, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.322727117869951e-06, | |
| "loss": 0.0, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.15898180194261471, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.6408072716606346e-06, | |
| "loss": 0.0, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.1607681143239924, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1625544267053701, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.1643407390867478, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9369152030840556e-06, | |
| "loss": 0.0, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.16612705146812548, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.0, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.16791336384950317, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0926199633097157e-06, | |
| "loss": 0.0, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.16969967623088086, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.596123493895991e-07, | |
| "loss": 0.0, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.17148598861225856, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 15.2213, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.17327230099363627, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.7390523158633554e-07, | |
| "loss": 0.0, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.17505861337501397, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.2179748700879012e-07, | |
| "loss": 0.0, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.17684492575639166, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.0, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.17863123813776935, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17863123813776935, | |
| "eval_loss": NaN, | |
| "eval_runtime": 58.95, | |
| "eval_samples_per_second": 15.997, | |
| "eval_steps_per_second": 8.007, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.238916716560384e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |