{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.1029336078229542, "eval_steps": 9, "global_step": 100, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.001029336078229542, "grad_norm": 0.23270708322525024, "learning_rate": 1e-05, "loss": 4.8778, "step": 1 }, { "epoch": 0.001029336078229542, "eval_loss": 2.449375629425049, "eval_runtime": 226.7141, "eval_samples_per_second": 3.608, "eval_steps_per_second": 0.454, "step": 1 }, { "epoch": 0.002058672156459084, "grad_norm": 0.2403886765241623, "learning_rate": 2e-05, "loss": 4.6643, "step": 2 }, { "epoch": 0.003088008234688626, "grad_norm": 0.21336641907691956, "learning_rate": 3e-05, "loss": 4.8765, "step": 3 }, { "epoch": 0.004117344312918168, "grad_norm": 0.24259218573570251, "learning_rate": 4e-05, "loss": 5.0207, "step": 4 }, { "epoch": 0.0051466803911477095, "grad_norm": 0.2876662611961365, "learning_rate": 5e-05, "loss": 5.0594, "step": 5 }, { "epoch": 0.006176016469377252, "grad_norm": 0.3250068724155426, "learning_rate": 6e-05, "loss": 4.8713, "step": 6 }, { "epoch": 0.007205352547606794, "grad_norm": 0.2040358930826187, "learning_rate": 7e-05, "loss": 4.711, "step": 7 }, { "epoch": 0.008234688625836336, "grad_norm": 0.34615465998649597, "learning_rate": 8e-05, "loss": 4.8104, "step": 8 }, { "epoch": 0.009264024704065878, "grad_norm": 0.3927066922187805, "learning_rate": 9e-05, "loss": 5.0474, "step": 9 }, { "epoch": 0.009264024704065878, "eval_loss": 2.4394149780273438, "eval_runtime": 226.8989, "eval_samples_per_second": 3.605, "eval_steps_per_second": 0.454, "step": 9 }, { "epoch": 0.010293360782295419, "grad_norm": 0.2759287655353546, "learning_rate": 0.0001, "loss": 4.8251, "step": 10 }, { "epoch": 0.011322696860524962, "grad_norm": 0.35022222995758057, "learning_rate": 9.99695413509548e-05, "loss": 5.2115, "step": 11 }, { "epoch": 0.012352032938754504, "grad_norm": 0.2925807535648346, "learning_rate": 9.987820251299122e-05, "loss": 4.8341, "step": 12 }, { "epoch": 0.013381369016984045, "grad_norm": 0.27655553817749023, "learning_rate": 9.972609476841367e-05, "loss": 4.5449, "step": 13 }, { "epoch": 0.014410705095213588, "grad_norm": 0.3062187135219574, "learning_rate": 9.951340343707852e-05, "loss": 4.8675, "step": 14 }, { "epoch": 0.01544004117344313, "grad_norm": 0.3353738486766815, "learning_rate": 9.924038765061042e-05, "loss": 4.7423, "step": 15 }, { "epoch": 0.016469377251672673, "grad_norm": 0.3210652470588684, "learning_rate": 9.890738003669029e-05, "loss": 5.0688, "step": 16 }, { "epoch": 0.017498713329902212, "grad_norm": 0.36584898829460144, "learning_rate": 9.851478631379982e-05, "loss": 4.8991, "step": 17 }, { "epoch": 0.018528049408131755, "grad_norm": 0.31348302960395813, "learning_rate": 9.806308479691595e-05, "loss": 4.7511, "step": 18 }, { "epoch": 0.018528049408131755, "eval_loss": 2.4040560722351074, "eval_runtime": 226.9454, "eval_samples_per_second": 3.604, "eval_steps_per_second": 0.454, "step": 18 }, { "epoch": 0.0195573854863613, "grad_norm": 0.31781822443008423, "learning_rate": 9.755282581475769e-05, "loss": 4.818, "step": 19 }, { "epoch": 0.020586721564590838, "grad_norm": 0.27385666966438293, "learning_rate": 9.698463103929542e-05, "loss": 4.8567, "step": 20 }, { "epoch": 0.02161605764282038, "grad_norm": 0.3134090304374695, "learning_rate": 9.635919272833938e-05, "loss": 4.8306, "step": 21 }, { "epoch": 0.022645393721049924, "grad_norm": 0.32903534173965454, "learning_rate": 9.567727288213005e-05, "loss": 4.6411, "step": 22 }, { "epoch": 0.023674729799279464, "grad_norm": 0.5250512361526489, "learning_rate": 9.493970231495835e-05, "loss": 4.9593, "step": 23 }, { "epoch": 0.024704065877509007, "grad_norm": 0.31589871644973755, "learning_rate": 9.414737964294636e-05, "loss": 4.7378, "step": 24 }, { "epoch": 0.02573340195573855, "grad_norm": 0.2427978217601776, "learning_rate": 9.330127018922194e-05, "loss": 4.471, "step": 25 }, { "epoch": 0.02676273803396809, "grad_norm": 0.3135339617729187, "learning_rate": 9.24024048078213e-05, "loss": 4.7955, "step": 26 }, { "epoch": 0.027792074112197633, "grad_norm": 0.3900616466999054, "learning_rate": 9.145187862775209e-05, "loss": 5.0081, "step": 27 }, { "epoch": 0.027792074112197633, "eval_loss": 2.386911153793335, "eval_runtime": 227.018, "eval_samples_per_second": 3.603, "eval_steps_per_second": 0.454, "step": 27 }, { "epoch": 0.028821410190427176, "grad_norm": 0.26265034079551697, "learning_rate": 9.045084971874738e-05, "loss": 4.799, "step": 28 }, { "epoch": 0.029850746268656716, "grad_norm": 0.30579325556755066, "learning_rate": 8.940053768033609e-05, "loss": 4.5961, "step": 29 }, { "epoch": 0.03088008234688626, "grad_norm": 0.42424359917640686, "learning_rate": 8.83022221559489e-05, "loss": 4.8045, "step": 30 }, { "epoch": 0.0319094184251158, "grad_norm": 0.3576556444168091, "learning_rate": 8.715724127386972e-05, "loss": 4.8802, "step": 31 }, { "epoch": 0.032938754503345345, "grad_norm": 0.3096839189529419, "learning_rate": 8.596699001693255e-05, "loss": 4.7263, "step": 32 }, { "epoch": 0.033968090581574885, "grad_norm": 0.425644189119339, "learning_rate": 8.473291852294987e-05, "loss": 4.6257, "step": 33 }, { "epoch": 0.034997426659804425, "grad_norm": 0.3519304394721985, "learning_rate": 8.345653031794292e-05, "loss": 4.7576, "step": 34 }, { "epoch": 0.03602676273803397, "grad_norm": 0.38826555013656616, "learning_rate": 8.213938048432697e-05, "loss": 4.7818, "step": 35 }, { "epoch": 0.03705609881626351, "grad_norm": 0.33783453702926636, "learning_rate": 8.07830737662829e-05, "loss": 4.7301, "step": 36 }, { "epoch": 0.03705609881626351, "eval_loss": 2.3819499015808105, "eval_runtime": 226.8579, "eval_samples_per_second": 3.606, "eval_steps_per_second": 0.454, "step": 36 }, { "epoch": 0.03808543489449305, "grad_norm": 0.30728089809417725, "learning_rate": 7.938926261462366e-05, "loss": 4.6269, "step": 37 }, { "epoch": 0.0391147709727226, "grad_norm": 0.26561239361763, "learning_rate": 7.795964517353735e-05, "loss": 4.8023, "step": 38 }, { "epoch": 0.04014410705095214, "grad_norm": 0.333359956741333, "learning_rate": 7.649596321166024e-05, "loss": 4.6354, "step": 39 }, { "epoch": 0.041173443129181676, "grad_norm": 0.280263215303421, "learning_rate": 7.500000000000001e-05, "loss": 4.9582, "step": 40 }, { "epoch": 0.04220277920741122, "grad_norm": 0.3637993633747101, "learning_rate": 7.347357813929454e-05, "loss": 5.0495, "step": 41 }, { "epoch": 0.04323211528564076, "grad_norm": 0.3035355806350708, "learning_rate": 7.191855733945387e-05, "loss": 4.8531, "step": 42 }, { "epoch": 0.0442614513638703, "grad_norm": 0.2905413806438446, "learning_rate": 7.033683215379002e-05, "loss": 4.8934, "step": 43 }, { "epoch": 0.04529078744209985, "grad_norm": 0.2680681049823761, "learning_rate": 6.873032967079561e-05, "loss": 4.8269, "step": 44 }, { "epoch": 0.04632012352032939, "grad_norm": 0.2697928547859192, "learning_rate": 6.710100716628344e-05, "loss": 4.5012, "step": 45 }, { "epoch": 0.04632012352032939, "eval_loss": 2.379274845123291, "eval_runtime": 226.8561, "eval_samples_per_second": 3.606, "eval_steps_per_second": 0.454, "step": 45 }, { "epoch": 0.04734945959855893, "grad_norm": 0.34642162919044495, "learning_rate": 6.545084971874738e-05, "loss": 4.8511, "step": 46 }, { "epoch": 0.048378795676788475, "grad_norm": 0.3296074867248535, "learning_rate": 6.378186779084995e-05, "loss": 4.8204, "step": 47 }, { "epoch": 0.049408131755018014, "grad_norm": 0.27953025698661804, "learning_rate": 6.209609477998338e-05, "loss": 5.0973, "step": 48 }, { "epoch": 0.050437467833247554, "grad_norm": 0.3788706660270691, "learning_rate": 6.0395584540887963e-05, "loss": 5.0161, "step": 49 }, { "epoch": 0.0514668039114771, "grad_norm": 0.32087740302085876, "learning_rate": 5.868240888334653e-05, "loss": 4.8925, "step": 50 }, { "epoch": 0.05249613998970664, "grad_norm": 0.2698979675769806, "learning_rate": 5.695865504800327e-05, "loss": 4.6946, "step": 51 }, { "epoch": 0.05352547606793618, "grad_norm": 0.27260327339172363, "learning_rate": 5.522642316338268e-05, "loss": 4.601, "step": 52 }, { "epoch": 0.054554812146165726, "grad_norm": 0.3249811828136444, "learning_rate": 5.348782368720626e-05, "loss": 4.6817, "step": 53 }, { "epoch": 0.055584148224395266, "grad_norm": 0.4115317463874817, "learning_rate": 5.174497483512506e-05, "loss": 4.8402, "step": 54 }, { "epoch": 0.055584148224395266, "eval_loss": 2.3773033618927, "eval_runtime": 227.1671, "eval_samples_per_second": 3.601, "eval_steps_per_second": 0.453, "step": 54 }, { "epoch": 0.056613484302624806, "grad_norm": 0.31797289848327637, "learning_rate": 5e-05, "loss": 4.866, "step": 55 }, { "epoch": 0.05764282038085435, "grad_norm": 0.32362496852874756, "learning_rate": 4.825502516487497e-05, "loss": 4.5958, "step": 56 }, { "epoch": 0.05867215645908389, "grad_norm": 0.4479215741157532, "learning_rate": 4.6512176312793736e-05, "loss": 4.8453, "step": 57 }, { "epoch": 0.05970149253731343, "grad_norm": 0.25469648838043213, "learning_rate": 4.477357683661734e-05, "loss": 4.6875, "step": 58 }, { "epoch": 0.06073082861554297, "grad_norm": 0.2602284550666809, "learning_rate": 4.3041344951996746e-05, "loss": 4.6748, "step": 59 }, { "epoch": 0.06176016469377252, "grad_norm": 0.309254914522171, "learning_rate": 4.131759111665349e-05, "loss": 4.8994, "step": 60 }, { "epoch": 0.06278950077200206, "grad_norm": 0.352872759103775, "learning_rate": 3.960441545911204e-05, "loss": 4.8684, "step": 61 }, { "epoch": 0.0638188368502316, "grad_norm": 0.33318430185317993, "learning_rate": 3.790390522001662e-05, "loss": 4.6973, "step": 62 }, { "epoch": 0.06484817292846114, "grad_norm": 0.4939478039741516, "learning_rate": 3.6218132209150045e-05, "loss": 4.7247, "step": 63 }, { "epoch": 0.06484817292846114, "eval_loss": 2.3757121562957764, "eval_runtime": 226.8372, "eval_samples_per_second": 3.606, "eval_steps_per_second": 0.454, "step": 63 }, { "epoch": 0.06587750900669069, "grad_norm": 0.27650734782218933, "learning_rate": 3.4549150281252636e-05, "loss": 4.8668, "step": 64 }, { "epoch": 0.06690684508492023, "grad_norm": 0.2447676956653595, "learning_rate": 3.289899283371657e-05, "loss": 4.6292, "step": 65 }, { "epoch": 0.06793618116314977, "grad_norm": 0.40559127926826477, "learning_rate": 3.12696703292044e-05, "loss": 4.6037, "step": 66 }, { "epoch": 0.06896551724137931, "grad_norm": 0.32476454973220825, "learning_rate": 2.9663167846209998e-05, "loss": 4.8826, "step": 67 }, { "epoch": 0.06999485331960885, "grad_norm": 0.25857582688331604, "learning_rate": 2.8081442660546125e-05, "loss": 4.8219, "step": 68 }, { "epoch": 0.07102418939783839, "grad_norm": 0.3146672546863556, "learning_rate": 2.6526421860705473e-05, "loss": 4.9364, "step": 69 }, { "epoch": 0.07205352547606794, "grad_norm": 0.3049579858779907, "learning_rate": 2.500000000000001e-05, "loss": 4.7155, "step": 70 }, { "epoch": 0.07308286155429748, "grad_norm": 0.27620747685432434, "learning_rate": 2.350403678833976e-05, "loss": 4.8771, "step": 71 }, { "epoch": 0.07411219763252702, "grad_norm": 0.33244985342025757, "learning_rate": 2.2040354826462668e-05, "loss": 4.8509, "step": 72 }, { "epoch": 0.07411219763252702, "eval_loss": 2.3749494552612305, "eval_runtime": 226.9106, "eval_samples_per_second": 3.605, "eval_steps_per_second": 0.454, "step": 72 }, { "epoch": 0.07514153371075656, "grad_norm": 0.2920321524143219, "learning_rate": 2.061073738537635e-05, "loss": 4.8178, "step": 73 }, { "epoch": 0.0761708697889861, "grad_norm": 0.2564603388309479, "learning_rate": 1.9216926233717085e-05, "loss": 4.6999, "step": 74 }, { "epoch": 0.07720020586721564, "grad_norm": 0.3616389036178589, "learning_rate": 1.7860619515673033e-05, "loss": 4.5835, "step": 75 }, { "epoch": 0.0782295419454452, "grad_norm": 0.3594980537891388, "learning_rate": 1.6543469682057106e-05, "loss": 4.6926, "step": 76 }, { "epoch": 0.07925887802367473, "grad_norm": 0.30547279119491577, "learning_rate": 1.526708147705013e-05, "loss": 4.7796, "step": 77 }, { "epoch": 0.08028821410190427, "grad_norm": 0.35963600873947144, "learning_rate": 1.4033009983067452e-05, "loss": 4.5412, "step": 78 }, { "epoch": 0.08131755018013381, "grad_norm": 0.26262137293815613, "learning_rate": 1.2842758726130283e-05, "loss": 4.5385, "step": 79 }, { "epoch": 0.08234688625836335, "grad_norm": 0.3326081931591034, "learning_rate": 1.1697777844051105e-05, "loss": 4.7268, "step": 80 }, { "epoch": 0.08337622233659289, "grad_norm": 0.29198020696640015, "learning_rate": 1.0599462319663905e-05, "loss": 4.7795, "step": 81 }, { "epoch": 0.08337622233659289, "eval_loss": 2.3745617866516113, "eval_runtime": 226.9222, "eval_samples_per_second": 3.605, "eval_steps_per_second": 0.454, "step": 81 }, { "epoch": 0.08440555841482245, "grad_norm": 0.36414095759391785, "learning_rate": 9.549150281252633e-06, "loss": 4.7554, "step": 82 }, { "epoch": 0.08543489449305199, "grad_norm": 0.3461471199989319, "learning_rate": 8.548121372247918e-06, "loss": 4.788, "step": 83 }, { "epoch": 0.08646423057128153, "grad_norm": 0.36194470524787903, "learning_rate": 7.597595192178702e-06, "loss": 5.0131, "step": 84 }, { "epoch": 0.08749356664951106, "grad_norm": 0.3344295918941498, "learning_rate": 6.698729810778065e-06, "loss": 4.9249, "step": 85 }, { "epoch": 0.0885229027277406, "grad_norm": 0.29325899481773376, "learning_rate": 5.852620357053651e-06, "loss": 4.4008, "step": 86 }, { "epoch": 0.08955223880597014, "grad_norm": 0.31407222151756287, "learning_rate": 5.060297685041659e-06, "loss": 4.6769, "step": 87 }, { "epoch": 0.0905815748841997, "grad_norm": 0.44068512320518494, "learning_rate": 4.322727117869951e-06, "loss": 4.9277, "step": 88 }, { "epoch": 0.09161091096242924, "grad_norm": 0.3176312744617462, "learning_rate": 3.6408072716606346e-06, "loss": 4.8676, "step": 89 }, { "epoch": 0.09264024704065878, "grad_norm": 0.3227611780166626, "learning_rate": 3.0153689607045845e-06, "loss": 4.7992, "step": 90 }, { "epoch": 0.09264024704065878, "eval_loss": 2.3742470741271973, "eval_runtime": 226.8864, "eval_samples_per_second": 3.605, "eval_steps_per_second": 0.454, "step": 90 }, { "epoch": 0.09366958311888832, "grad_norm": 0.3470034897327423, "learning_rate": 2.4471741852423237e-06, "loss": 4.8775, "step": 91 }, { "epoch": 0.09469891919711786, "grad_norm": 0.3345381021499634, "learning_rate": 1.9369152030840556e-06, "loss": 4.6053, "step": 92 }, { "epoch": 0.0957282552753474, "grad_norm": 0.2730475962162018, "learning_rate": 1.4852136862001764e-06, "loss": 4.6352, "step": 93 }, { "epoch": 0.09675759135357695, "grad_norm": 0.3089372515678406, "learning_rate": 1.0926199633097157e-06, "loss": 4.7786, "step": 94 }, { "epoch": 0.09778692743180649, "grad_norm": 0.3256579339504242, "learning_rate": 7.596123493895991e-07, "loss": 4.6533, "step": 95 }, { "epoch": 0.09881626351003603, "grad_norm": 0.4511566460132599, "learning_rate": 4.865965629214819e-07, "loss": 4.9058, "step": 96 }, { "epoch": 0.09984559958826557, "grad_norm": 0.2858843207359314, "learning_rate": 2.7390523158633554e-07, "loss": 4.4558, "step": 97 }, { "epoch": 0.10087493566649511, "grad_norm": 0.38588446378707886, "learning_rate": 1.2179748700879012e-07, "loss": 4.9628, "step": 98 }, { "epoch": 0.10190427174472465, "grad_norm": 0.29587432742118835, "learning_rate": 3.04586490452119e-08, "loss": 4.683, "step": 99 }, { "epoch": 0.10190427174472465, "eval_loss": 2.3742330074310303, "eval_runtime": 226.8851, "eval_samples_per_second": 3.605, "eval_steps_per_second": 0.454, "step": 99 }, { "epoch": 0.1029336078229542, "grad_norm": 0.36477580666542053, "learning_rate": 0.0, "loss": 4.9429, "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": 1.1925729759461376e+17, "train_batch_size": 8, "trial_name": null, "trial_params": null }