| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5807622504537205, | |
| "eval_steps": 25, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005807622504537205, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005807622504537205, | |
| "eval_loss": NaN, | |
| "eval_runtime": 95.3201, | |
| "eval_samples_per_second": 3.042, | |
| "eval_steps_per_second": 1.521, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01161524500907441, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.017422867513611617, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.02323049001814882, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 3896.2488, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.029038112522686024, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 657.6468, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03484573502722323, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 2.4658, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.040653357531760435, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 3.41, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.04646098003629764, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 35.8081, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.052268602540834846, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 843.6981, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05807622504537205, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06388384754990925, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.06969147005444647, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.07549909255898367, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.08130671506352087, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 21.9721, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.08711433756805807, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09292196007259527, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 15.3499, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.09872958257713249, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.10453720508166969, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1103448275862069, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.1161524500907441, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12196007259528131, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.1277676950998185, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 61.7461, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.1335753176043557, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 539.3531, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.13938294010889293, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.14519056261343014, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 33.2995, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.14519056261343014, | |
| "eval_loss": NaN, | |
| "eval_runtime": 9.6805, | |
| "eval_samples_per_second": 29.957, | |
| "eval_steps_per_second": 14.979, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.15099818511796734, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.15680580762250454, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 5.59, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.16261343012704174, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1611.2329, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.16842105263157894, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 135.0654, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.17422867513611615, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 165.8565, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18003629764065335, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.18584392014519055, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 81.9507, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.19165154264972778, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.19745916515426498, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 6.6547, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.20326678765880218, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.20907441016333939, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 7.5238, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2148820326678766, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.2206896551724138, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 68.1653, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.226497277676951, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.2323049001814882, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2381125226860254, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.0, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.24392014519056263, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.24972776769509983, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 707.1777, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.255535390199637, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 195.0476, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.2613430127041742, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.9188, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2671506352087114, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.27295825771324866, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.7395, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.27876588021778587, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 8.7959, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.28457350272232307, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.29038112522686027, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.29038112522686027, | |
| "eval_loss": NaN, | |
| "eval_runtime": 9.683, | |
| "eval_samples_per_second": 29.949, | |
| "eval_steps_per_second": 14.975, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2961887477313975, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 5.5278, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.3019963702359347, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 2660.9358, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.3078039927404719, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 25.8831, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.3136116152450091, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 108.3971, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.3194192377495463, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 215.6494, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.3252268602540835, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 3366.7695, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.3310344827586207, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 26.9873, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.3368421052631579, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.0, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.3426497277676951, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 0.0, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.3484573502722323, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 7.1294, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3542649727767695, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 666.5757, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.3600725952813067, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.0, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.3658802177858439, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.3673, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.3716878402903811, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.37749546279491836, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 293.4128, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.38330308529945556, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 1924.7764, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.38911070780399276, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 0.0, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.39491833030852996, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.0, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.40072595281306717, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.0, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.40653357531760437, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 447.9875, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.41234119782214157, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 1.1965, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.41814882032667877, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 636.1243, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.423956442831216, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 58.9024, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.4297640653357532, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 1028.7271, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.4355716878402904, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 109.8508, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4355716878402904, | |
| "eval_loss": NaN, | |
| "eval_runtime": 9.6787, | |
| "eval_samples_per_second": 29.963, | |
| "eval_steps_per_second": 14.981, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4413793103448276, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.6543469682057106e-05, | |
| "loss": 65.3235, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.4471869328493648, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.526708147705013e-05, | |
| "loss": 0.1186, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.452994555353902, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 58.8162, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.4588021778584392, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.2842758726130283e-05, | |
| "loss": 21.2564, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.4646098003629764, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 0.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.4704174228675136, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.0, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.4762250453720508, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.482032667876588, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.548121372247918e-06, | |
| "loss": 0.0, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.48784029038112525, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 2906.1953, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.49364791288566245, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 0.4129, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.49945553539019966, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.852620357053651e-06, | |
| "loss": 658.3425, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.5052631578947369, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.0, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.511070780399274, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.322727117869951e-06, | |
| "loss": 0.0, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.5168784029038113, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.6408072716606346e-06, | |
| "loss": 2626.5364, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.5226860254083484, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 101.4575, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5284936479128857, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.9226, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.5343012704174228, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9369152030840556e-06, | |
| "loss": 0.0, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.5401088929219601, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.0, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.5459165154264973, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0926199633097157e-06, | |
| "loss": 0.0, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.5517241379310345, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.596123493895991e-07, | |
| "loss": 13.896, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.5575317604355717, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.5633393829401089, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.7390523158633554e-07, | |
| "loss": 0.0, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.5691470054446461, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.2179748700879012e-07, | |
| "loss": 0.4251, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.5749546279491833, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 2705.8181, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.5807622504537205, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5807622504537205, | |
| "eval_loss": NaN, | |
| "eval_runtime": 9.6878, | |
| "eval_samples_per_second": 29.934, | |
| "eval_steps_per_second": 14.967, | |
| "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": 1.183490272961495e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |