| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 38601, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03885909691458771, | |
| "grad_norm": 0.45540618896484375, | |
| "learning_rate": 1.9740939353902752e-05, | |
| "loss": 0.4641, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07771819382917541, | |
| "grad_norm": 0.05885039269924164, | |
| "learning_rate": 1.94818787078055e-05, | |
| "loss": 0.0073, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11657729074376312, | |
| "grad_norm": 0.0643157884478569, | |
| "learning_rate": 1.922281806170825e-05, | |
| "loss": 0.0066, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.15543638765835083, | |
| "grad_norm": 0.025738418102264404, | |
| "learning_rate": 1.8963757415611e-05, | |
| "loss": 0.0055, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.19429548457293852, | |
| "grad_norm": 0.006490091327577829, | |
| "learning_rate": 1.8704696769513745e-05, | |
| "loss": 0.0018, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.23315458148752624, | |
| "grad_norm": 0.003259174292907119, | |
| "learning_rate": 1.8445636123416492e-05, | |
| "loss": 0.0017, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.27201367840211393, | |
| "grad_norm": 0.002379123354330659, | |
| "learning_rate": 1.8186575477319242e-05, | |
| "loss": 0.0004, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.31087277531670165, | |
| "grad_norm": 0.0013682579156011343, | |
| "learning_rate": 1.7927514831221992e-05, | |
| "loss": 0.0005, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.3497318722312893, | |
| "grad_norm": 0.003730255179107189, | |
| "learning_rate": 1.766845418512474e-05, | |
| "loss": 0.0061, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.38859096914587704, | |
| "grad_norm": 0.0010942475637421012, | |
| "learning_rate": 1.740939353902749e-05, | |
| "loss": 0.0015, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.42745006606046476, | |
| "grad_norm": 0.0026132045313715935, | |
| "learning_rate": 1.7150332892930236e-05, | |
| "loss": 0.002, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.4663091629750525, | |
| "grad_norm": 0.0005335228051990271, | |
| "learning_rate": 1.6891272246832986e-05, | |
| "loss": 0.0008, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.5051682598896402, | |
| "grad_norm": 0.0023255019914358854, | |
| "learning_rate": 1.6632211600735732e-05, | |
| "loss": 0.0009, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.5440273568042279, | |
| "grad_norm": 0.0013378489529713988, | |
| "learning_rate": 1.6373150954638482e-05, | |
| "loss": 0.0019, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.5828864537188155, | |
| "grad_norm": 0.0021079631987959146, | |
| "learning_rate": 1.611409030854123e-05, | |
| "loss": 0.0099, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.6217455506334033, | |
| "grad_norm": 0.00038598832907155156, | |
| "learning_rate": 1.585502966244398e-05, | |
| "loss": 0.0001, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.660604647547991, | |
| "grad_norm": 0.0005731793353334069, | |
| "learning_rate": 1.559596901634673e-05, | |
| "loss": 0.0013, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.6994637444625786, | |
| "grad_norm": 0.00023792145657353103, | |
| "learning_rate": 1.5336908370249476e-05, | |
| "loss": 0.0001, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.7383228413771664, | |
| "grad_norm": 0.0007500049541704357, | |
| "learning_rate": 1.5077847724152226e-05, | |
| "loss": 0.0025, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.7771819382917541, | |
| "grad_norm": 0.00044937318307347596, | |
| "learning_rate": 1.4818787078054974e-05, | |
| "loss": 0.0026, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.8160410352063419, | |
| "grad_norm": 0.0006425898754969239, | |
| "learning_rate": 1.4559726431957721e-05, | |
| "loss": 0.0006, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.8549001321209295, | |
| "grad_norm": 0.0006235554465092719, | |
| "learning_rate": 1.430066578586047e-05, | |
| "loss": 0.0012, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.8937592290355172, | |
| "grad_norm": 0.0004008706600870937, | |
| "learning_rate": 1.404160513976322e-05, | |
| "loss": 0.0017, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.932618325950105, | |
| "grad_norm": 0.0004875475133303553, | |
| "learning_rate": 1.3782544493665968e-05, | |
| "loss": 0.0001, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.9714774228646926, | |
| "grad_norm": 0.0015087984502315521, | |
| "learning_rate": 1.3523483847568716e-05, | |
| "loss": 0.0011, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9999115207373271, | |
| "eval_f1": 0.9998912136925701, | |
| "eval_loss": 0.0008539558621123433, | |
| "eval_precision": 0.9999032998912124, | |
| "eval_recall": 0.9998791277861045, | |
| "eval_runtime": 21.9409, | |
| "eval_samples_per_second": 930.682, | |
| "eval_steps_per_second": 116.358, | |
| "step": 12867 | |
| }, | |
| { | |
| "epoch": 1.0103365197792804, | |
| "grad_norm": 0.0007504357490688562, | |
| "learning_rate": 1.3264423201471466e-05, | |
| "loss": 0.0012, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.049195616693868, | |
| "grad_norm": 9.457436681259423e-05, | |
| "learning_rate": 1.3005362555374215e-05, | |
| "loss": 0.0001, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.0880547136084557, | |
| "grad_norm": 0.0005767066031694412, | |
| "learning_rate": 1.2746301909276963e-05, | |
| "loss": 0.0006, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.1269138105230434, | |
| "grad_norm": 0.0004582173132803291, | |
| "learning_rate": 1.2487241263179713e-05, | |
| "loss": 0.0045, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.165772907437631, | |
| "grad_norm": 0.0006144808721728623, | |
| "learning_rate": 1.222818061708246e-05, | |
| "loss": 0.0053, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.2046320043522187, | |
| "grad_norm": 0.0002801430528052151, | |
| "learning_rate": 1.1969119970985208e-05, | |
| "loss": 0.0005, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.2434911012668066, | |
| "grad_norm": 0.001027058344334364, | |
| "learning_rate": 1.1710059324887957e-05, | |
| "loss": 0.002, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.2823501981813943, | |
| "grad_norm": 0.00030287433764897287, | |
| "learning_rate": 1.1450998678790705e-05, | |
| "loss": 0.0001, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.321209295095982, | |
| "grad_norm": 9.348896855954081e-05, | |
| "learning_rate": 1.1191938032693455e-05, | |
| "loss": 0.0, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.3600683920105696, | |
| "grad_norm": 4.3804615415865555e-05, | |
| "learning_rate": 1.0932877386596204e-05, | |
| "loss": 0.0, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.3989274889251573, | |
| "grad_norm": 5.0469727284507826e-05, | |
| "learning_rate": 1.0673816740498952e-05, | |
| "loss": 0.0012, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.4377865858397452, | |
| "grad_norm": 8.691708353580907e-05, | |
| "learning_rate": 1.0414756094401702e-05, | |
| "loss": 0.0004, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.4766456827543328, | |
| "grad_norm": 6.273793405853212e-05, | |
| "learning_rate": 1.0155695448304449e-05, | |
| "loss": 0.0005, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.5155047796689205, | |
| "grad_norm": 9.822064748732373e-05, | |
| "learning_rate": 9.896634802207197e-06, | |
| "loss": 0.0, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.5543638765835082, | |
| "grad_norm": 0.00012862567382398993, | |
| "learning_rate": 9.637574156109945e-06, | |
| "loss": 0.0005, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.5932229734980958, | |
| "grad_norm": 0.013405167497694492, | |
| "learning_rate": 9.378513510012694e-06, | |
| "loss": 0.0005, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.6320820704126837, | |
| "grad_norm": 6.82063982822001e-05, | |
| "learning_rate": 9.119452863915444e-06, | |
| "loss": 0.0, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.6709411673272712, | |
| "grad_norm": 0.0001184305947390385, | |
| "learning_rate": 8.860392217818192e-06, | |
| "loss": 0.0006, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.709800264241859, | |
| "grad_norm": 0.0004617497615981847, | |
| "learning_rate": 8.60133157172094e-06, | |
| "loss": 0.0, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.7486593611564467, | |
| "grad_norm": 9.478950232733041e-05, | |
| "learning_rate": 8.342270925623689e-06, | |
| "loss": 0.0, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.7875184580710344, | |
| "grad_norm": 0.0005689120152965188, | |
| "learning_rate": 8.083210279526437e-06, | |
| "loss": 0.0013, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.8263775549856223, | |
| "grad_norm": 3.0777548090554774e-05, | |
| "learning_rate": 7.824149633429187e-06, | |
| "loss": 0.0002, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 1.8652366519002097, | |
| "grad_norm": 0.00029757359880022705, | |
| "learning_rate": 7.565088987331936e-06, | |
| "loss": 0.001, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.9040957488147976, | |
| "grad_norm": 0.0001594646309968084, | |
| "learning_rate": 7.306028341234683e-06, | |
| "loss": 0.0005, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 1.9429548457293853, | |
| "grad_norm": 3.8429996493505314e-05, | |
| "learning_rate": 7.046967695137432e-06, | |
| "loss": 0.0009, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.981813942643973, | |
| "grad_norm": 0.0016445047222077847, | |
| "learning_rate": 6.787907049040181e-06, | |
| "loss": 0.002, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9999188940092166, | |
| "eval_f1": 0.9998670389575854, | |
| "eval_loss": 0.0004210107435937971, | |
| "eval_precision": 0.9998791248640154, | |
| "eval_recall": 0.9998549533433254, | |
| "eval_runtime": 26.3407, | |
| "eval_samples_per_second": 775.225, | |
| "eval_steps_per_second": 96.922, | |
| "step": 25734 | |
| }, | |
| { | |
| "epoch": 2.020673039558561, | |
| "grad_norm": 0.010017353110015392, | |
| "learning_rate": 6.52884640294293e-06, | |
| "loss": 0.0015, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.0595321364731483, | |
| "grad_norm": 0.0025657187215983868, | |
| "learning_rate": 6.269785756845678e-06, | |
| "loss": 0.0028, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.098391233387736, | |
| "grad_norm": 0.00022906585945747793, | |
| "learning_rate": 6.010725110748426e-06, | |
| "loss": 0.0001, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.1372503303023236, | |
| "grad_norm": 6.71066518407315e-05, | |
| "learning_rate": 5.751664464651175e-06, | |
| "loss": 0.0006, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.1761094272169115, | |
| "grad_norm": 0.00021527826902456582, | |
| "learning_rate": 5.492603818553925e-06, | |
| "loss": 0.0, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.2149685241314994, | |
| "grad_norm": 9.48450033320114e-05, | |
| "learning_rate": 5.233543172456672e-06, | |
| "loss": 0.0005, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.253827621046087, | |
| "grad_norm": 0.00010121305240318179, | |
| "learning_rate": 4.9744825263594205e-06, | |
| "loss": 0.0, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.2926867179606747, | |
| "grad_norm": 8.369102579308674e-05, | |
| "learning_rate": 4.71542188026217e-06, | |
| "loss": 0.0, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.331545814875262, | |
| "grad_norm": 5.066412632004358e-05, | |
| "learning_rate": 4.456361234164918e-06, | |
| "loss": 0.0, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.37040491178985, | |
| "grad_norm": 0.00012291729217395186, | |
| "learning_rate": 4.197300588067667e-06, | |
| "loss": 0.0008, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.4092640087044375, | |
| "grad_norm": 0.00014957574603613466, | |
| "learning_rate": 3.938239941970416e-06, | |
| "loss": 0.0, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.4481231056190254, | |
| "grad_norm": 4.759750663652085e-05, | |
| "learning_rate": 3.679179295873164e-06, | |
| "loss": 0.0, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.4869822025336132, | |
| "grad_norm": 3.8031983422115445e-05, | |
| "learning_rate": 3.4201186497759125e-06, | |
| "loss": 0.0, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.5258412994482007, | |
| "grad_norm": 8.281264308607206e-05, | |
| "learning_rate": 3.1610580036786613e-06, | |
| "loss": 0.0, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.5647003963627886, | |
| "grad_norm": 8.475359209114686e-05, | |
| "learning_rate": 2.9019973575814097e-06, | |
| "loss": 0.0, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.603559493277376, | |
| "grad_norm": 8.979766425909474e-05, | |
| "learning_rate": 2.6429367114841585e-06, | |
| "loss": 0.0, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 2.642418590191964, | |
| "grad_norm": 0.0025826424825936556, | |
| "learning_rate": 2.3838760653869073e-06, | |
| "loss": 0.0007, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.681277687106552, | |
| "grad_norm": 8.839767542667687e-05, | |
| "learning_rate": 2.1248154192896557e-06, | |
| "loss": 0.0007, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 2.7201367840211392, | |
| "grad_norm": 3.709522934514098e-05, | |
| "learning_rate": 1.8657547731924045e-06, | |
| "loss": 0.0, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.758995880935727, | |
| "grad_norm": 0.00011620597069850191, | |
| "learning_rate": 1.6066941270951531e-06, | |
| "loss": 0.0005, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 2.7978549778503146, | |
| "grad_norm": 3.431947698118165e-05, | |
| "learning_rate": 1.3476334809979017e-06, | |
| "loss": 0.0, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.8367140747649024, | |
| "grad_norm": 8.798589260550216e-05, | |
| "learning_rate": 1.0885728349006503e-06, | |
| "loss": 0.0, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 2.8755731716794903, | |
| "grad_norm": 2.6514268029131927e-05, | |
| "learning_rate": 8.29512188803399e-07, | |
| "loss": 0.0, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.914432268594078, | |
| "grad_norm": 2.451765067235101e-05, | |
| "learning_rate": 5.704515427061476e-07, | |
| "loss": 0.0, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 2.9532913655086657, | |
| "grad_norm": 1.2271469131519552e-05, | |
| "learning_rate": 3.113908966088962e-07, | |
| "loss": 0.0, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.992150462423253, | |
| "grad_norm": 3.4994409361388534e-05, | |
| "learning_rate": 5.233025051164478e-08, | |
| "loss": 0.0005, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9999188940092166, | |
| "eval_f1": 0.9998549533433254, | |
| "eval_loss": 0.0004571246390696615, | |
| "eval_precision": 0.9998549533433254, | |
| "eval_recall": 0.9998549533433254, | |
| "eval_runtime": 40.6059, | |
| "eval_samples_per_second": 502.882, | |
| "eval_steps_per_second": 62.873, | |
| "step": 38601 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 38601, | |
| "total_flos": 1954036863526950.0, | |
| "train_loss": 0.007192428305773462, | |
| "train_runtime": 2269.357, | |
| "train_samples_per_second": 136.071, | |
| "train_steps_per_second": 17.01 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 38601, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1954036863526950.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |