| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.16, |
| "eval_steps": 1000, |
| "global_step": 8000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.002, |
| "grad_norm": 0.7236698865890503, |
| "learning_rate": 0.000396, |
| "loss": 8.129180297851562, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.004, |
| "grad_norm": 0.6699316501617432, |
| "learning_rate": 0.0005999985696164921, |
| "loss": 6.683881225585938, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.006, |
| "grad_norm": 0.4809923470020294, |
| "learning_rate": 0.0005999867739541219, |
| "loss": 6.335140991210937, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.008, |
| "grad_norm": 0.6197829246520996, |
| "learning_rate": 0.0005999630639489827, |
| "loss": 6.119940185546875, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.01, |
| "grad_norm": 0.5310641527175903, |
| "learning_rate": 0.0005999274405427468, |
| "loss": 5.908866577148437, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.012, |
| "grad_norm": 0.4188089668750763, |
| "learning_rate": 0.0005998799051502414, |
| "loss": 5.712981567382813, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.014, |
| "grad_norm": 0.43575364351272583, |
| "learning_rate": 0.0005998204596593927, |
| "loss": 5.54151123046875, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.016, |
| "grad_norm": 0.3631345331668854, |
| "learning_rate": 0.0005997491064311517, |
| "loss": 5.377118530273438, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.018, |
| "grad_norm": 0.4753637909889221, |
| "learning_rate": 0.000599665848299399, |
| "loss": 5.2575732421875, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.02, |
| "grad_norm": 0.3885057270526886, |
| "learning_rate": 0.0005995706885708337, |
| "loss": 5.168424072265625, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.02, |
| "eval_accuracy": 0.21422370900467913, |
| "eval_loss": 5.0658464431762695, |
| "eval_runtime": 5.7327, |
| "eval_samples_per_second": 338.583, |
| "eval_steps_per_second": 21.281, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.022, |
| "grad_norm": 0.35898318886756897, |
| "learning_rate": 0.0005994636310248412, |
| "loss": 5.058907165527343, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.024, |
| "grad_norm": 0.36867955327033997, |
| "learning_rate": 0.000599344679913343, |
| "loss": 4.899845275878906, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.026, |
| "grad_norm": 0.3753994405269623, |
| "learning_rate": 0.0005992138399606286, |
| "loss": 4.828855285644531, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.028, |
| "grad_norm": 0.3426174521446228, |
| "learning_rate": 0.0005990711163631674, |
| "loss": 4.73110595703125, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.03, |
| "grad_norm": 0.34207016229629517, |
| "learning_rate": 0.000598916514789402, |
| "loss": 4.640643005371094, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.032, |
| "grad_norm": 0.40568745136260986, |
| "learning_rate": 0.0005987500413795236, |
| "loss": 4.591531677246094, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.034, |
| "grad_norm": 0.31946808099746704, |
| "learning_rate": 0.0005985717027452278, |
| "loss": 4.533291015625, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.036, |
| "grad_norm": 0.30501991510391235, |
| "learning_rate": 0.0005983815059694525, |
| "loss": 4.492686157226562, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.038, |
| "grad_norm": 0.33989349007606506, |
| "learning_rate": 0.0005981794586060958, |
| "loss": 4.41447509765625, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.04, |
| "grad_norm": 0.3303995430469513, |
| "learning_rate": 0.0005979655686797167, |
| "loss": 4.396521911621094, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.04, |
| "eval_accuracy": 0.2874403514237522, |
| "eval_loss": 4.30245304107666, |
| "eval_runtime": 5.7068, |
| "eval_samples_per_second": 340.121, |
| "eval_steps_per_second": 21.378, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.042, |
| "grad_norm": 0.46283307671546936, |
| "learning_rate": 0.0005977398446852161, |
| "loss": 4.379761352539062, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.044, |
| "grad_norm": 0.3387432396411896, |
| "learning_rate": 0.0005975022955874992, |
| "loss": 4.3154080200195315, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.046, |
| "grad_norm": 0.2719409465789795, |
| "learning_rate": 0.0005972529308211201, |
| "loss": 4.275622863769531, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.048, |
| "grad_norm": 0.2836616039276123, |
| "learning_rate": 0.0005969917602899061, |
| "loss": 4.258504333496094, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.05, |
| "grad_norm": 0.32210105657577515, |
| "learning_rate": 0.0005967187943665656, |
| "loss": 4.207368469238281, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.052, |
| "grad_norm": 0.30050167441368103, |
| "learning_rate": 0.0005964340438922749, |
| "loss": 4.2094793701171875, |
| "step": 2600 |
| }, |
| { |
| "epoch": 0.054, |
| "grad_norm": 0.3175508379936218, |
| "learning_rate": 0.0005961375201762486, |
| "loss": 4.171758117675782, |
| "step": 2700 |
| }, |
| { |
| "epoch": 0.056, |
| "grad_norm": 0.285259872674942, |
| "learning_rate": 0.0005958292349952897, |
| "loss": 4.1502682495117185, |
| "step": 2800 |
| }, |
| { |
| "epoch": 0.058, |
| "grad_norm": 0.2848680913448334, |
| "learning_rate": 0.0005955092005933225, |
| "loss": 4.128558959960937, |
| "step": 2900 |
| }, |
| { |
| "epoch": 0.06, |
| "grad_norm": 0.2977091073989868, |
| "learning_rate": 0.000595177429680906, |
| "loss": 4.134291076660157, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.06, |
| "eval_accuracy": 0.3091028793639367, |
| "eval_loss": 4.053634166717529, |
| "eval_runtime": 6.2016, |
| "eval_samples_per_second": 312.982, |
| "eval_steps_per_second": 19.672, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.062, |
| "grad_norm": 0.2866455018520355, |
| "learning_rate": 0.0005948339354347288, |
| "loss": 4.118316040039063, |
| "step": 3100 |
| }, |
| { |
| "epoch": 0.064, |
| "grad_norm": 0.2707529664039612, |
| "learning_rate": 0.0005944787314970865, |
| "loss": 4.104917602539063, |
| "step": 3200 |
| }, |
| { |
| "epoch": 0.066, |
| "grad_norm": 0.3138796091079712, |
| "learning_rate": 0.0005941118319753391, |
| "loss": 4.084369506835937, |
| "step": 3300 |
| }, |
| { |
| "epoch": 0.068, |
| "grad_norm": 0.2631393373012543, |
| "learning_rate": 0.0005937332514413514, |
| "loss": 4.044774475097657, |
| "step": 3400 |
| }, |
| { |
| "epoch": 0.07, |
| "grad_norm": 0.27509307861328125, |
| "learning_rate": 0.0005933430049309136, |
| "loss": 4.025779418945312, |
| "step": 3500 |
| }, |
| { |
| "epoch": 0.072, |
| "grad_norm": 0.33931243419647217, |
| "learning_rate": 0.0005929411079431449, |
| "loss": 4.004125366210937, |
| "step": 3600 |
| }, |
| { |
| "epoch": 0.074, |
| "grad_norm": 0.2669001519680023, |
| "learning_rate": 0.000592527576439877, |
| "loss": 4.012510070800781, |
| "step": 3700 |
| }, |
| { |
| "epoch": 0.076, |
| "grad_norm": 0.2727104425430298, |
| "learning_rate": 0.000592102426845021, |
| "loss": 3.9971853637695314, |
| "step": 3800 |
| }, |
| { |
| "epoch": 0.078, |
| "grad_norm": 0.25884559750556946, |
| "learning_rate": 0.0005916656760439145, |
| "loss": 3.9807003784179686, |
| "step": 3900 |
| }, |
| { |
| "epoch": 0.08, |
| "grad_norm": 0.3023005425930023, |
| "learning_rate": 0.0005912173413826517, |
| "loss": 3.9553176879882814, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.08, |
| "eval_accuracy": 0.32060359872601835, |
| "eval_loss": 3.923754930496216, |
| "eval_runtime": 6.2736, |
| "eval_samples_per_second": 309.393, |
| "eval_steps_per_second": 19.447, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.082, |
| "grad_norm": 0.2674967348575592, |
| "learning_rate": 0.0005907574406673935, |
| "loss": 3.9582391357421876, |
| "step": 4100 |
| }, |
| { |
| "epoch": 0.084, |
| "grad_norm": 0.25437116622924805, |
| "learning_rate": 0.0005902859921636611, |
| "loss": 3.942552185058594, |
| "step": 4200 |
| }, |
| { |
| "epoch": 0.086, |
| "grad_norm": 0.25480520725250244, |
| "learning_rate": 0.0005898030145956102, |
| "loss": 3.9550421142578127, |
| "step": 4300 |
| }, |
| { |
| "epoch": 0.088, |
| "grad_norm": 0.25695815682411194, |
| "learning_rate": 0.0005893085271452874, |
| "loss": 3.9156878662109373, |
| "step": 4400 |
| }, |
| { |
| "epoch": 0.09, |
| "grad_norm": 0.25535979866981506, |
| "learning_rate": 0.0005888025494518687, |
| "loss": 3.9294302368164065, |
| "step": 4500 |
| }, |
| { |
| "epoch": 0.092, |
| "grad_norm": 0.24557402729988098, |
| "learning_rate": 0.0005882851016108786, |
| "loss": 3.927276306152344, |
| "step": 4600 |
| }, |
| { |
| "epoch": 0.094, |
| "grad_norm": 0.26716557145118713, |
| "learning_rate": 0.0005877562041733932, |
| "loss": 3.904244384765625, |
| "step": 4700 |
| }, |
| { |
| "epoch": 0.096, |
| "grad_norm": 0.2525787651538849, |
| "learning_rate": 0.0005872158781452231, |
| "loss": 3.8892666625976564, |
| "step": 4800 |
| }, |
| { |
| "epoch": 0.098, |
| "grad_norm": 0.3125140964984894, |
| "learning_rate": 0.0005866641449860792, |
| "loss": 3.887486572265625, |
| "step": 4900 |
| }, |
| { |
| "epoch": 0.1, |
| "grad_norm": 0.2451365739107132, |
| "learning_rate": 0.0005861010266087211, |
| "loss": 3.875788269042969, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.1, |
| "eval_accuracy": 0.32875099183244255, |
| "eval_loss": 3.831427812576294, |
| "eval_runtime": 5.9329, |
| "eval_samples_per_second": 327.159, |
| "eval_steps_per_second": 20.563, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.102, |
| "grad_norm": 0.28080520033836365, |
| "learning_rate": 0.0005855265453780858, |
| "loss": 3.8711001586914064, |
| "step": 5100 |
| }, |
| { |
| "epoch": 0.104, |
| "grad_norm": 0.25133365392684937, |
| "learning_rate": 0.0005849407241104003, |
| "loss": 3.8725830078125, |
| "step": 5200 |
| }, |
| { |
| "epoch": 0.106, |
| "grad_norm": 0.27164554595947266, |
| "learning_rate": 0.000584343586072275, |
| "loss": 3.847073974609375, |
| "step": 5300 |
| }, |
| { |
| "epoch": 0.108, |
| "grad_norm": 0.2697848081588745, |
| "learning_rate": 0.0005837351549797795, |
| "loss": 3.79802978515625, |
| "step": 5400 |
| }, |
| { |
| "epoch": 0.11, |
| "grad_norm": 0.27253684401512146, |
| "learning_rate": 0.0005831154549975012, |
| "loss": 3.812421875, |
| "step": 5500 |
| }, |
| { |
| "epoch": 0.112, |
| "grad_norm": 0.27660876512527466, |
| "learning_rate": 0.0005824845107375853, |
| "loss": 3.7727542114257813, |
| "step": 5600 |
| }, |
| { |
| "epoch": 0.114, |
| "grad_norm": 0.2439262717962265, |
| "learning_rate": 0.0005818423472587571, |
| "loss": 3.7664410400390627, |
| "step": 5700 |
| }, |
| { |
| "epoch": 0.116, |
| "grad_norm": 0.24943126738071442, |
| "learning_rate": 0.0005811889900653269, |
| "loss": 3.7883651733398436, |
| "step": 5800 |
| }, |
| { |
| "epoch": 0.118, |
| "grad_norm": 0.2556191682815552, |
| "learning_rate": 0.0005805244651061771, |
| "loss": 3.7454019165039063, |
| "step": 5900 |
| }, |
| { |
| "epoch": 0.12, |
| "grad_norm": 0.27899155020713806, |
| "learning_rate": 0.0005798487987737322, |
| "loss": 3.767793884277344, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.12, |
| "eval_accuracy": 0.33610391076885543, |
| "eval_loss": 3.7605059146881104, |
| "eval_runtime": 5.697, |
| "eval_samples_per_second": 340.705, |
| "eval_steps_per_second": 21.415, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.122, |
| "grad_norm": 0.25070950388908386, |
| "learning_rate": 0.0005791620179029091, |
| "loss": 3.7485830688476565, |
| "step": 6100 |
| }, |
| { |
| "epoch": 0.124, |
| "grad_norm": 0.23322179913520813, |
| "learning_rate": 0.0005784641497700528, |
| "loss": 3.7345367431640626, |
| "step": 6200 |
| }, |
| { |
| "epoch": 0.126, |
| "grad_norm": 0.27075648307800293, |
| "learning_rate": 0.0005777552220918525, |
| "loss": 3.747908935546875, |
| "step": 6300 |
| }, |
| { |
| "epoch": 0.128, |
| "grad_norm": 0.22065384685993195, |
| "learning_rate": 0.0005770352630242405, |
| "loss": 3.7215350341796873, |
| "step": 6400 |
| }, |
| { |
| "epoch": 0.13, |
| "grad_norm": 0.2387268990278244, |
| "learning_rate": 0.0005763043011612747, |
| "loss": 3.7065789794921873, |
| "step": 6500 |
| }, |
| { |
| "epoch": 0.132, |
| "grad_norm": 0.2440839409828186, |
| "learning_rate": 0.0005755623655340019, |
| "loss": 3.7222140502929686, |
| "step": 6600 |
| }, |
| { |
| "epoch": 0.134, |
| "grad_norm": 0.7614469528198242, |
| "learning_rate": 0.000574809485609306, |
| "loss": 3.7069766235351564, |
| "step": 6700 |
| }, |
| { |
| "epoch": 0.136, |
| "grad_norm": 0.25670167803764343, |
| "learning_rate": 0.0005740456912887371, |
| "loss": 3.7104458618164062, |
| "step": 6800 |
| }, |
| { |
| "epoch": 0.138, |
| "grad_norm": 0.25214189291000366, |
| "learning_rate": 0.0005732710129073234, |
| "loss": 3.7040945434570314, |
| "step": 6900 |
| }, |
| { |
| "epoch": 0.14, |
| "grad_norm": 0.2262343317270279, |
| "learning_rate": 0.0005724854812323674, |
| "loss": 3.686309814453125, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.14, |
| "eval_accuracy": 0.3409998074307532, |
| "eval_loss": 3.705655097961426, |
| "eval_runtime": 6.3038, |
| "eval_samples_per_second": 307.912, |
| "eval_steps_per_second": 19.354, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.142, |
| "grad_norm": 0.2256837785243988, |
| "learning_rate": 0.0005716891274622232, |
| "loss": 3.6851046752929686, |
| "step": 7100 |
| }, |
| { |
| "epoch": 0.144, |
| "grad_norm": 0.22516745328903198, |
| "learning_rate": 0.0005708819832250578, |
| "loss": 3.6776641845703124, |
| "step": 7200 |
| }, |
| { |
| "epoch": 0.146, |
| "grad_norm": 0.28322234749794006, |
| "learning_rate": 0.000570064080577595, |
| "loss": 3.6858819580078124, |
| "step": 7300 |
| }, |
| { |
| "epoch": 0.148, |
| "grad_norm": 0.2589821219444275, |
| "learning_rate": 0.0005692354520038415, |
| "loss": 3.683369140625, |
| "step": 7400 |
| }, |
| { |
| "epoch": 0.15, |
| "grad_norm": 0.25471949577331543, |
| "learning_rate": 0.0005683961304137982, |
| "loss": 3.6476519775390623, |
| "step": 7500 |
| }, |
| { |
| "epoch": 0.152, |
| "grad_norm": 0.24686427414417267, |
| "learning_rate": 0.0005675461491421516, |
| "loss": 3.66244873046875, |
| "step": 7600 |
| }, |
| { |
| "epoch": 0.154, |
| "grad_norm": 0.23603662848472595, |
| "learning_rate": 0.0005666855419469506, |
| "loss": 3.6586270141601562, |
| "step": 7700 |
| }, |
| { |
| "epoch": 0.156, |
| "grad_norm": 0.23954837024211884, |
| "learning_rate": 0.0005658143430082659, |
| "loss": 3.6526480102539063, |
| "step": 7800 |
| }, |
| { |
| "epoch": 0.158, |
| "grad_norm": 0.24939873814582825, |
| "learning_rate": 0.0005649325869268321, |
| "loss": 3.6602520751953125, |
| "step": 7900 |
| }, |
| { |
| "epoch": 0.16, |
| "grad_norm": 0.22606751322746277, |
| "learning_rate": 0.0005640403087226739, |
| "loss": 3.6500091552734375, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_accuracy": 0.34401134847875336, |
| "eval_loss": 3.6704835891723633, |
| "eval_runtime": 5.9263, |
| "eval_samples_per_second": 327.521, |
| "eval_steps_per_second": 20.586, |
| "step": 8000 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 50000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9223372036854775807, |
| "save_steps": 2000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.5084035072e+17, |
| "train_batch_size": 120, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
