{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.09528346831824679, "eval_steps": 9, "global_step": 100, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0009528346831824678, "grad_norm": 1.3286488056182861, "learning_rate": 1e-05, "loss": 2.8561, "step": 1 }, { "epoch": 0.0009528346831824678, "eval_loss": 3.010896921157837, "eval_runtime": 196.4019, "eval_samples_per_second": 4.501, "eval_steps_per_second": 0.565, "step": 1 }, { "epoch": 0.0019056693663649356, "grad_norm": 1.1131958961486816, "learning_rate": 2e-05, "loss": 2.8611, "step": 2 }, { "epoch": 0.0028585040495474035, "grad_norm": 1.1074174642562866, "learning_rate": 3e-05, "loss": 2.9567, "step": 3 }, { "epoch": 0.003811338732729871, "grad_norm": 0.99653559923172, "learning_rate": 4e-05, "loss": 2.8704, "step": 4 }, { "epoch": 0.004764173415912339, "grad_norm": 1.1130872964859009, "learning_rate": 5e-05, "loss": 2.8391, "step": 5 }, { "epoch": 0.005717008099094807, "grad_norm": 1.083756446838379, "learning_rate": 6e-05, "loss": 2.7191, "step": 6 }, { "epoch": 0.006669842782277275, "grad_norm": 1.2542109489440918, "learning_rate": 7e-05, "loss": 2.8155, "step": 7 }, { "epoch": 0.007622677465459742, "grad_norm": 1.3546979427337646, "learning_rate": 8e-05, "loss": 2.6735, "step": 8 }, { "epoch": 0.00857551214864221, "grad_norm": 1.3032156229019165, "learning_rate": 9e-05, "loss": 2.7692, "step": 9 }, { "epoch": 0.00857551214864221, "eval_loss": 2.807802438735962, "eval_runtime": 196.863, "eval_samples_per_second": 4.49, "eval_steps_per_second": 0.564, "step": 9 }, { "epoch": 0.009528346831824679, "grad_norm": 1.346268892288208, "learning_rate": 0.0001, "loss": 2.8929, "step": 10 }, { "epoch": 0.010481181515007145, "grad_norm": 1.1175191402435303, "learning_rate": 9.99695413509548e-05, "loss": 2.7615, "step": 11 }, { "epoch": 0.011434016198189614, "grad_norm": 1.2937767505645752, "learning_rate": 9.987820251299122e-05, "loss": 2.4941, "step": 12 }, { "epoch": 0.012386850881372083, "grad_norm": 1.0705084800720215, "learning_rate": 9.972609476841367e-05, "loss": 2.3596, "step": 13 }, { "epoch": 0.01333968556455455, "grad_norm": 1.4327889680862427, "learning_rate": 9.951340343707852e-05, "loss": 2.6066, "step": 14 }, { "epoch": 0.014292520247737018, "grad_norm": 1.4437620639801025, "learning_rate": 9.924038765061042e-05, "loss": 2.4794, "step": 15 }, { "epoch": 0.015245354930919485, "grad_norm": 1.3408985137939453, "learning_rate": 9.890738003669029e-05, "loss": 2.4767, "step": 16 }, { "epoch": 0.016198189614101955, "grad_norm": 1.2504230737686157, "learning_rate": 9.851478631379982e-05, "loss": 2.4222, "step": 17 }, { "epoch": 0.01715102429728442, "grad_norm": 1.209808588027954, "learning_rate": 9.806308479691595e-05, "loss": 2.4661, "step": 18 }, { "epoch": 0.01715102429728442, "eval_loss": 2.5225956439971924, "eval_runtime": 196.935, "eval_samples_per_second": 4.489, "eval_steps_per_second": 0.564, "step": 18 }, { "epoch": 0.01810385898046689, "grad_norm": 1.1820844411849976, "learning_rate": 9.755282581475769e-05, "loss": 2.3123, "step": 19 }, { "epoch": 0.019056693663649357, "grad_norm": 1.272624135017395, "learning_rate": 9.698463103929542e-05, "loss": 2.4863, "step": 20 }, { "epoch": 0.020009528346831826, "grad_norm": 1.213018774986267, "learning_rate": 9.635919272833938e-05, "loss": 2.4414, "step": 21 }, { "epoch": 0.02096236303001429, "grad_norm": 1.7545298337936401, "learning_rate": 9.567727288213005e-05, "loss": 2.6525, "step": 22 }, { "epoch": 0.02191519771319676, "grad_norm": 1.255905270576477, "learning_rate": 9.493970231495835e-05, "loss": 2.0139, "step": 23 }, { "epoch": 0.022868032396379228, "grad_norm": 1.409415602684021, "learning_rate": 9.414737964294636e-05, "loss": 2.8552, "step": 24 }, { "epoch": 0.023820867079561697, "grad_norm": 1.1865147352218628, "learning_rate": 9.330127018922194e-05, "loss": 2.2278, "step": 25 }, { "epoch": 0.024773701762744165, "grad_norm": 1.18612802028656, "learning_rate": 9.24024048078213e-05, "loss": 2.4295, "step": 26 }, { "epoch": 0.02572653644592663, "grad_norm": 1.541218638420105, "learning_rate": 9.145187862775209e-05, "loss": 2.7706, "step": 27 }, { "epoch": 0.02572653644592663, "eval_loss": 2.387563943862915, "eval_runtime": 197.0083, "eval_samples_per_second": 4.487, "eval_steps_per_second": 0.563, "step": 27 }, { "epoch": 0.0266793711291091, "grad_norm": 1.0855358839035034, "learning_rate": 9.045084971874738e-05, "loss": 2.0015, "step": 28 }, { "epoch": 0.027632205812291567, "grad_norm": 1.234771728515625, "learning_rate": 8.940053768033609e-05, "loss": 2.3806, "step": 29 }, { "epoch": 0.028585040495474036, "grad_norm": 1.0970972776412964, "learning_rate": 8.83022221559489e-05, "loss": 2.1708, "step": 30 }, { "epoch": 0.029537875178656504, "grad_norm": 1.7669090032577515, "learning_rate": 8.715724127386972e-05, "loss": 2.7497, "step": 31 }, { "epoch": 0.03049070986183897, "grad_norm": 1.653584361076355, "learning_rate": 8.596699001693255e-05, "loss": 2.4642, "step": 32 }, { "epoch": 0.03144354454502144, "grad_norm": 1.5698741674423218, "learning_rate": 8.473291852294987e-05, "loss": 2.5928, "step": 33 }, { "epoch": 0.03239637922820391, "grad_norm": 1.7401469945907593, "learning_rate": 8.345653031794292e-05, "loss": 2.3844, "step": 34 }, { "epoch": 0.03334921391138637, "grad_norm": 1.6176362037658691, "learning_rate": 8.213938048432697e-05, "loss": 2.3984, "step": 35 }, { "epoch": 0.03430204859456884, "grad_norm": 1.7408335208892822, "learning_rate": 8.07830737662829e-05, "loss": 2.5587, "step": 36 }, { "epoch": 0.03430204859456884, "eval_loss": 2.3208231925964355, "eval_runtime": 196.6199, "eval_samples_per_second": 4.496, "eval_steps_per_second": 0.565, "step": 36 }, { "epoch": 0.03525488327775131, "grad_norm": 3.2627251148223877, "learning_rate": 7.938926261462366e-05, "loss": 2.228, "step": 37 }, { "epoch": 0.03620771796093378, "grad_norm": 1.0993157625198364, "learning_rate": 7.795964517353735e-05, "loss": 2.3224, "step": 38 }, { "epoch": 0.037160552644116246, "grad_norm": 1.1583936214447021, "learning_rate": 7.649596321166024e-05, "loss": 2.1978, "step": 39 }, { "epoch": 0.038113387327298714, "grad_norm": 1.2586085796356201, "learning_rate": 7.500000000000001e-05, "loss": 2.2668, "step": 40 }, { "epoch": 0.03906622201048118, "grad_norm": 1.080640196800232, "learning_rate": 7.347357813929454e-05, "loss": 2.2083, "step": 41 }, { "epoch": 0.04001905669366365, "grad_norm": 1.2006219625473022, "learning_rate": 7.191855733945387e-05, "loss": 1.9749, "step": 42 }, { "epoch": 0.04097189137684612, "grad_norm": 1.1608589887619019, "learning_rate": 7.033683215379002e-05, "loss": 2.2486, "step": 43 }, { "epoch": 0.04192472606002858, "grad_norm": 1.5159680843353271, "learning_rate": 6.873032967079561e-05, "loss": 2.3073, "step": 44 }, { "epoch": 0.04287756074321105, "grad_norm": 1.2097727060317993, "learning_rate": 6.710100716628344e-05, "loss": 2.1224, "step": 45 }, { "epoch": 0.04287756074321105, "eval_loss": 2.284031867980957, "eval_runtime": 196.629, "eval_samples_per_second": 4.496, "eval_steps_per_second": 0.565, "step": 45 }, { "epoch": 0.04383039542639352, "grad_norm": 2.138612985610962, "learning_rate": 6.545084971874738e-05, "loss": 2.2678, "step": 46 }, { "epoch": 0.04478323010957599, "grad_norm": 1.2521336078643799, "learning_rate": 6.378186779084995e-05, "loss": 2.1841, "step": 47 }, { "epoch": 0.045736064792758456, "grad_norm": 1.2867101430892944, "learning_rate": 6.209609477998338e-05, "loss": 2.3342, "step": 48 }, { "epoch": 0.046688899475940925, "grad_norm": 1.0387842655181885, "learning_rate": 6.0395584540887963e-05, "loss": 2.1413, "step": 49 }, { "epoch": 0.04764173415912339, "grad_norm": 0.9112858176231384, "learning_rate": 5.868240888334653e-05, "loss": 1.9995, "step": 50 }, { "epoch": 0.04859456884230586, "grad_norm": 1.1231671571731567, "learning_rate": 5.695865504800327e-05, "loss": 2.2521, "step": 51 }, { "epoch": 0.04954740352548833, "grad_norm": 1.1129883527755737, "learning_rate": 5.522642316338268e-05, "loss": 2.0542, "step": 52 }, { "epoch": 0.0505002382086708, "grad_norm": 1.109358310699463, "learning_rate": 5.348782368720626e-05, "loss": 2.04, "step": 53 }, { "epoch": 0.05145307289185326, "grad_norm": 1.1299806833267212, "learning_rate": 5.174497483512506e-05, "loss": 2.4423, "step": 54 }, { "epoch": 0.05145307289185326, "eval_loss": 2.2711129188537598, "eval_runtime": 196.6182, "eval_samples_per_second": 4.496, "eval_steps_per_second": 0.565, "step": 54 }, { "epoch": 0.05240590757503573, "grad_norm": 1.216711401939392, "learning_rate": 5e-05, "loss": 2.1914, "step": 55 }, { "epoch": 0.0533587422582182, "grad_norm": 1.4715956449508667, "learning_rate": 4.825502516487497e-05, "loss": 2.6387, "step": 56 }, { "epoch": 0.054311576941400666, "grad_norm": 1.202911615371704, "learning_rate": 4.6512176312793736e-05, "loss": 2.2061, "step": 57 }, { "epoch": 0.055264411624583135, "grad_norm": 11.649187088012695, "learning_rate": 4.477357683661734e-05, "loss": 2.0835, "step": 58 }, { "epoch": 0.0562172463077656, "grad_norm": 1.2017055749893188, "learning_rate": 4.3041344951996746e-05, "loss": 2.2761, "step": 59 }, { "epoch": 0.05717008099094807, "grad_norm": 1.1830456256866455, "learning_rate": 4.131759111665349e-05, "loss": 2.4208, "step": 60 }, { "epoch": 0.05812291567413054, "grad_norm": 1.0666215419769287, "learning_rate": 3.960441545911204e-05, "loss": 2.2889, "step": 61 }, { "epoch": 0.05907575035731301, "grad_norm": 0.9857567548751831, "learning_rate": 3.790390522001662e-05, "loss": 2.2491, "step": 62 }, { "epoch": 0.06002858504049548, "grad_norm": 0.9194086194038391, "learning_rate": 3.6218132209150045e-05, "loss": 2.028, "step": 63 }, { "epoch": 0.06002858504049548, "eval_loss": 2.2625362873077393, "eval_runtime": 196.687, "eval_samples_per_second": 4.494, "eval_steps_per_second": 0.564, "step": 63 }, { "epoch": 0.06098141972367794, "grad_norm": 5.810688495635986, "learning_rate": 3.4549150281252636e-05, "loss": 2.3557, "step": 64 }, { "epoch": 0.06193425440686041, "grad_norm": 1.6563860177993774, "learning_rate": 3.289899283371657e-05, "loss": 2.1718, "step": 65 }, { "epoch": 0.06288708909004288, "grad_norm": 4.58107852935791, "learning_rate": 3.12696703292044e-05, "loss": 2.2625, "step": 66 }, { "epoch": 0.06383992377322535, "grad_norm": 1.1236543655395508, "learning_rate": 2.9663167846209998e-05, "loss": 2.1118, "step": 67 }, { "epoch": 0.06479275845640782, "grad_norm": 1.2911332845687866, "learning_rate": 2.8081442660546125e-05, "loss": 2.2574, "step": 68 }, { "epoch": 0.06574559313959027, "grad_norm": 1.0672364234924316, "learning_rate": 2.6526421860705473e-05, "loss": 2.1681, "step": 69 }, { "epoch": 0.06669842782277274, "grad_norm": 1.161694049835205, "learning_rate": 2.500000000000001e-05, "loss": 2.3036, "step": 70 }, { "epoch": 0.06765126250595521, "grad_norm": 1.2077661752700806, "learning_rate": 2.350403678833976e-05, "loss": 2.204, "step": 71 }, { "epoch": 0.06860409718913768, "grad_norm": 1.1666839122772217, "learning_rate": 2.2040354826462668e-05, "loss": 2.2889, "step": 72 }, { "epoch": 0.06860409718913768, "eval_loss": 2.254430055618286, "eval_runtime": 196.7106, "eval_samples_per_second": 4.494, "eval_steps_per_second": 0.564, "step": 72 }, { "epoch": 0.06955693187232015, "grad_norm": 0.7700322866439819, "learning_rate": 2.061073738537635e-05, "loss": 2.0121, "step": 73 }, { "epoch": 0.07050976655550262, "grad_norm": 0.9282349944114685, "learning_rate": 1.9216926233717085e-05, "loss": 2.1442, "step": 74 }, { "epoch": 0.07146260123868509, "grad_norm": 0.8676728010177612, "learning_rate": 1.7860619515673033e-05, "loss": 2.102, "step": 75 }, { "epoch": 0.07241543592186755, "grad_norm": 1.164281964302063, "learning_rate": 1.6543469682057106e-05, "loss": 2.2321, "step": 76 }, { "epoch": 0.07336827060505002, "grad_norm": 1.1541085243225098, "learning_rate": 1.526708147705013e-05, "loss": 2.215, "step": 77 }, { "epoch": 0.07432110528823249, "grad_norm": 1.1044031381607056, "learning_rate": 1.4033009983067452e-05, "loss": 1.9906, "step": 78 }, { "epoch": 0.07527393997141496, "grad_norm": 1.034180998802185, "learning_rate": 1.2842758726130283e-05, "loss": 1.9801, "step": 79 }, { "epoch": 0.07622677465459743, "grad_norm": 1.1965882778167725, "learning_rate": 1.1697777844051105e-05, "loss": 2.0424, "step": 80 }, { "epoch": 0.0771796093377799, "grad_norm": 1.2659209966659546, "learning_rate": 1.0599462319663905e-05, "loss": 2.1702, "step": 81 }, { "epoch": 0.0771796093377799, "eval_loss": 2.248025894165039, "eval_runtime": 196.9804, "eval_samples_per_second": 4.488, "eval_steps_per_second": 0.564, "step": 81 }, { "epoch": 0.07813244402096237, "grad_norm": 1.0619964599609375, "learning_rate": 9.549150281252633e-06, "loss": 2.2545, "step": 82 }, { "epoch": 0.07908527870414483, "grad_norm": 1.0333913564682007, "learning_rate": 8.548121372247918e-06, "loss": 2.0858, "step": 83 }, { "epoch": 0.0800381133873273, "grad_norm": 0.9673925042152405, "learning_rate": 7.597595192178702e-06, "loss": 2.3828, "step": 84 }, { "epoch": 0.08099094807050977, "grad_norm": 0.9544613361358643, "learning_rate": 6.698729810778065e-06, "loss": 2.3138, "step": 85 }, { "epoch": 0.08194378275369224, "grad_norm": 1.184772253036499, "learning_rate": 5.852620357053651e-06, "loss": 2.4817, "step": 86 }, { "epoch": 0.08289661743687471, "grad_norm": 1.039474606513977, "learning_rate": 5.060297685041659e-06, "loss": 2.1223, "step": 87 }, { "epoch": 0.08384945212005716, "grad_norm": 0.9785050749778748, "learning_rate": 4.322727117869951e-06, "loss": 2.0992, "step": 88 }, { "epoch": 0.08480228680323963, "grad_norm": 1.0777572393417358, "learning_rate": 3.6408072716606346e-06, "loss": 2.2065, "step": 89 }, { "epoch": 0.0857551214864221, "grad_norm": 0.8153870105743408, "learning_rate": 3.0153689607045845e-06, "loss": 1.8947, "step": 90 }, { "epoch": 0.0857551214864221, "eval_loss": 2.246095657348633, "eval_runtime": 196.6499, "eval_samples_per_second": 4.495, "eval_steps_per_second": 0.564, "step": 90 }, { "epoch": 0.08670795616960457, "grad_norm": 1.0481244325637817, "learning_rate": 2.4471741852423237e-06, "loss": 2.349, "step": 91 }, { "epoch": 0.08766079085278704, "grad_norm": 4.479108810424805, "learning_rate": 1.9369152030840556e-06, "loss": 2.1547, "step": 92 }, { "epoch": 0.0886136255359695, "grad_norm": 1.07948637008667, "learning_rate": 1.4852136862001764e-06, "loss": 2.2365, "step": 93 }, { "epoch": 0.08956646021915197, "grad_norm": 0.9758604168891907, "learning_rate": 1.0926199633097157e-06, "loss": 2.2499, "step": 94 }, { "epoch": 0.09051929490233444, "grad_norm": 0.9797371029853821, "learning_rate": 7.596123493895991e-07, "loss": 1.9102, "step": 95 }, { "epoch": 0.09147212958551691, "grad_norm": 1.0941691398620605, "learning_rate": 4.865965629214819e-07, "loss": 2.0948, "step": 96 }, { "epoch": 0.09242496426869938, "grad_norm": 1.081665277481079, "learning_rate": 2.7390523158633554e-07, "loss": 2.4104, "step": 97 }, { "epoch": 0.09337779895188185, "grad_norm": 1.2182639837265015, "learning_rate": 1.2179748700879012e-07, "loss": 2.2791, "step": 98 }, { "epoch": 0.09433063363506432, "grad_norm": 1.0702403783798218, "learning_rate": 3.04586490452119e-08, "loss": 2.3174, "step": 99 }, { "epoch": 0.09433063363506432, "eval_loss": 2.2456421852111816, "eval_runtime": 196.6996, "eval_samples_per_second": 4.494, "eval_steps_per_second": 0.564, "step": 99 }, { "epoch": 0.09528346831824679, "grad_norm": 1.2310587167739868, "learning_rate": 0.0, "loss": 2.3505, "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": 1.483774567120896e+17, "train_batch_size": 8, "trial_name": null, "trial_params": null }