| { | |
| "best_metric": 2.5461041927337646, | |
| "best_model_checkpoint": "car_models_image_detection/checkpoint-32650", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 32650, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15313935681470137, | |
| "grad_norm": 1.442165732383728, | |
| "learning_rate": 8.875766871165644e-06, | |
| "loss": 5.7501, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.30627871362940273, | |
| "grad_norm": 1.5294591188430786, | |
| "learning_rate": 8.737730061349693e-06, | |
| "loss": 5.66, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.45941807044410415, | |
| "grad_norm": 1.7971255779266357, | |
| "learning_rate": 8.599693251533743e-06, | |
| "loss": 5.5686, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6125574272588055, | |
| "grad_norm": 1.7858527898788452, | |
| "learning_rate": 8.461656441717792e-06, | |
| "loss": 5.4714, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7656967840735069, | |
| "grad_norm": 1.6970139741897583, | |
| "learning_rate": 8.32361963190184e-06, | |
| "loss": 5.3715, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.9188361408882083, | |
| "grad_norm": 1.9722312688827515, | |
| "learning_rate": 8.18558282208589e-06, | |
| "loss": 5.2844, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.20365025344993468, | |
| "eval_loss": 5.212795734405518, | |
| "eval_model_preparation_time": 0.0035, | |
| "eval_runtime": 948.361, | |
| "eval_samples_per_second": 73.431, | |
| "eval_steps_per_second": 9.179, | |
| "step": 3265 | |
| }, | |
| { | |
| "epoch": 1.0719754977029097, | |
| "grad_norm": 2.317599296569824, | |
| "learning_rate": 8.047546012269938e-06, | |
| "loss": 5.1866, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.225114854517611, | |
| "grad_norm": 1.872455358505249, | |
| "learning_rate": 7.909509202453989e-06, | |
| "loss": 5.1003, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.3782542113323124, | |
| "grad_norm": 2.192427396774292, | |
| "learning_rate": 7.771472392638037e-06, | |
| "loss": 5.0184, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.5313935681470139, | |
| "grad_norm": 2.1170101165771484, | |
| "learning_rate": 7.633435582822086e-06, | |
| "loss": 4.9414, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.6845329249617151, | |
| "grad_norm": 2.908296823501587, | |
| "learning_rate": 7.495398773006135e-06, | |
| "loss": 4.863, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.8376722817764164, | |
| "grad_norm": 3.4065804481506348, | |
| "learning_rate": 7.3573619631901846e-06, | |
| "loss": 4.779, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.9908116385911179, | |
| "grad_norm": 2.6088645458221436, | |
| "learning_rate": 7.219325153374233e-06, | |
| "loss": 4.7011, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.31753758669710935, | |
| "eval_loss": 4.691849231719971, | |
| "eval_model_preparation_time": 0.0035, | |
| "eval_runtime": 713.5751, | |
| "eval_samples_per_second": 97.592, | |
| "eval_steps_per_second": 12.199, | |
| "step": 6530 | |
| }, | |
| { | |
| "epoch": 2.1439509954058193, | |
| "grad_norm": 2.4476678371429443, | |
| "learning_rate": 7.081288343558283e-06, | |
| "loss": 4.5978, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.2970903522205206, | |
| "grad_norm": 3.1015167236328125, | |
| "learning_rate": 6.9432515337423315e-06, | |
| "loss": 4.53, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.450229709035222, | |
| "grad_norm": 3.4957149028778076, | |
| "learning_rate": 6.80521472392638e-06, | |
| "loss": 4.437, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.6033690658499236, | |
| "grad_norm": 3.372995138168335, | |
| "learning_rate": 6.66717791411043e-06, | |
| "loss": 4.3886, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.756508422664625, | |
| "grad_norm": 2.965057849884033, | |
| "learning_rate": 6.529141104294479e-06, | |
| "loss": 4.3013, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.909647779479326, | |
| "grad_norm": 3.4717111587524414, | |
| "learning_rate": 6.391104294478528e-06, | |
| "loss": 4.2275, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.3973635462887175, | |
| "eval_loss": 4.212672710418701, | |
| "eval_model_preparation_time": 0.0035, | |
| "eval_runtime": 702.2225, | |
| "eval_samples_per_second": 99.169, | |
| "eval_steps_per_second": 12.396, | |
| "step": 9795 | |
| }, | |
| { | |
| "epoch": 3.0627871362940278, | |
| "grad_norm": 3.6480553150177, | |
| "learning_rate": 6.253067484662577e-06, | |
| "loss": 4.1497, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.215926493108729, | |
| "grad_norm": 2.944525718688965, | |
| "learning_rate": 6.1150306748466255e-06, | |
| "loss": 4.0626, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.3690658499234303, | |
| "grad_norm": 3.6789698600769043, | |
| "learning_rate": 5.976993865030675e-06, | |
| "loss": 3.9861, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.522205206738132, | |
| "grad_norm": 3.7942678928375244, | |
| "learning_rate": 5.838957055214725e-06, | |
| "loss": 3.9302, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.6753445635528332, | |
| "grad_norm": 2.889557123184204, | |
| "learning_rate": 5.700920245398773e-06, | |
| "loss": 3.8667, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.8284839203675345, | |
| "grad_norm": 4.875265121459961, | |
| "learning_rate": 5.562883435582822e-06, | |
| "loss": 3.7844, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.9816232771822357, | |
| "grad_norm": 4.279688835144043, | |
| "learning_rate": 5.424846625766871e-06, | |
| "loss": 3.7406, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.4593833914904005, | |
| "eval_loss": 3.774059772491455, | |
| "eval_model_preparation_time": 0.0035, | |
| "eval_runtime": 705.2879, | |
| "eval_samples_per_second": 98.738, | |
| "eval_steps_per_second": 12.342, | |
| "step": 13060 | |
| }, | |
| { | |
| "epoch": 4.134762633996937, | |
| "grad_norm": 4.041619777679443, | |
| "learning_rate": 5.286809815950921e-06, | |
| "loss": 3.6478, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.287901990811639, | |
| "grad_norm": 5.805275917053223, | |
| "learning_rate": 5.14877300613497e-06, | |
| "loss": 3.58, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.44104134762634, | |
| "grad_norm": 3.922706365585327, | |
| "learning_rate": 5.010736196319019e-06, | |
| "loss": 3.543, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.594180704441041, | |
| "grad_norm": 3.550119638442993, | |
| "learning_rate": 4.872699386503067e-06, | |
| "loss": 3.4574, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.747320061255743, | |
| "grad_norm": 5.960729122161865, | |
| "learning_rate": 4.734662576687116e-06, | |
| "loss": 3.4116, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.900459418070444, | |
| "grad_norm": 4.563659191131592, | |
| "learning_rate": 4.5966257668711664e-06, | |
| "loss": 3.3544, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5196800643317682, | |
| "eval_loss": 3.398108959197998, | |
| "eval_model_preparation_time": 0.0035, | |
| "eval_runtime": 700.4362, | |
| "eval_samples_per_second": 99.422, | |
| "eval_steps_per_second": 12.428, | |
| "step": 16325 | |
| }, | |
| { | |
| "epoch": 5.053598774885145, | |
| "grad_norm": 5.071287631988525, | |
| "learning_rate": 4.458588957055215e-06, | |
| "loss": 3.2835, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.206738131699847, | |
| "grad_norm": 5.6899003982543945, | |
| "learning_rate": 4.320552147239264e-06, | |
| "loss": 3.2203, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.359877488514548, | |
| "grad_norm": 5.455111980438232, | |
| "learning_rate": 4.1825153374233126e-06, | |
| "loss": 3.1822, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.51301684532925, | |
| "grad_norm": 7.294569969177246, | |
| "learning_rate": 4.044478527607362e-06, | |
| "loss": 3.1201, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.666156202143951, | |
| "grad_norm": 4.797060012817383, | |
| "learning_rate": 3.906441717791411e-06, | |
| "loss": 3.0764, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 5.819295558958652, | |
| "grad_norm": 9.175103187561035, | |
| "learning_rate": 3.7684049079754604e-06, | |
| "loss": 3.0317, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 5.972434915773354, | |
| "grad_norm": 5.574986934661865, | |
| "learning_rate": 3.630368098159509e-06, | |
| "loss": 3.0015, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5613951952210686, | |
| "eval_loss": 3.0899131298065186, | |
| "eval_model_preparation_time": 0.0035, | |
| "eval_runtime": 705.6276, | |
| "eval_samples_per_second": 98.691, | |
| "eval_steps_per_second": 12.337, | |
| "step": 19590 | |
| }, | |
| { | |
| "epoch": 6.1255742725880555, | |
| "grad_norm": 6.0056352615356445, | |
| "learning_rate": 3.4923312883435583e-06, | |
| "loss": 2.921, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.278713629402756, | |
| "grad_norm": 12.546520233154297, | |
| "learning_rate": 3.3542944785276074e-06, | |
| "loss": 2.8971, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 6.431852986217458, | |
| "grad_norm": 4.947099208831787, | |
| "learning_rate": 3.2162576687116565e-06, | |
| "loss": 2.8553, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 6.584992343032159, | |
| "grad_norm": 8.054845809936523, | |
| "learning_rate": 3.0782208588957057e-06, | |
| "loss": 2.8151, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 6.738131699846861, | |
| "grad_norm": 6.752479076385498, | |
| "learning_rate": 2.940184049079755e-06, | |
| "loss": 2.7747, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 6.891271056661562, | |
| "grad_norm": 6.941985607147217, | |
| "learning_rate": 2.8021472392638035e-06, | |
| "loss": 2.7401, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5933313229655797, | |
| "eval_loss": 2.853682279586792, | |
| "eval_model_preparation_time": 0.0035, | |
| "eval_runtime": 697.9759, | |
| "eval_samples_per_second": 99.773, | |
| "eval_steps_per_second": 12.472, | |
| "step": 22855 | |
| }, | |
| { | |
| "epoch": 7.044410413476263, | |
| "grad_norm": 4.710619926452637, | |
| "learning_rate": 2.664110429447853e-06, | |
| "loss": 2.7003, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 7.197549770290965, | |
| "grad_norm": 5.312741756439209, | |
| "learning_rate": 2.5260736196319018e-06, | |
| "loss": 2.6404, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 7.3506891271056665, | |
| "grad_norm": 6.6971845626831055, | |
| "learning_rate": 2.3880368098159513e-06, | |
| "loss": 2.6377, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 7.503828483920367, | |
| "grad_norm": 6.931722164154053, | |
| "learning_rate": 2.25e-06, | |
| "loss": 2.5981, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 7.656967840735069, | |
| "grad_norm": 7.1908040046691895, | |
| "learning_rate": 2.111963190184049e-06, | |
| "loss": 2.5702, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 7.810107197549771, | |
| "grad_norm": 5.63743257522583, | |
| "learning_rate": 1.9739263803680983e-06, | |
| "loss": 2.5676, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 7.9632465543644715, | |
| "grad_norm": 6.103708267211914, | |
| "learning_rate": 1.8358895705521473e-06, | |
| "loss": 2.5566, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6183460417294907, | |
| "eval_loss": 2.6820664405822754, | |
| "eval_model_preparation_time": 0.0035, | |
| "eval_runtime": 747.6734, | |
| "eval_samples_per_second": 93.141, | |
| "eval_steps_per_second": 11.643, | |
| "step": 26120 | |
| }, | |
| { | |
| "epoch": 8.116385911179172, | |
| "grad_norm": 7.664525508880615, | |
| "learning_rate": 1.6978527607361964e-06, | |
| "loss": 2.4933, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 8.269525267993874, | |
| "grad_norm": 8.328133583068848, | |
| "learning_rate": 1.5598159509202455e-06, | |
| "loss": 2.4762, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 8.422664624808576, | |
| "grad_norm": 8.003011703491211, | |
| "learning_rate": 1.4217791411042944e-06, | |
| "loss": 2.4585, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 8.575803981623277, | |
| "grad_norm": 6.738711833953857, | |
| "learning_rate": 1.2837423312883436e-06, | |
| "loss": 2.4542, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 8.728943338437979, | |
| "grad_norm": 6.110630512237549, | |
| "learning_rate": 1.1457055214723925e-06, | |
| "loss": 2.4222, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 8.88208269525268, | |
| "grad_norm": 5.14086389541626, | |
| "learning_rate": 1.0076687116564419e-06, | |
| "loss": 2.4216, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6327058113987851, | |
| "eval_loss": 2.581773042678833, | |
| "eval_model_preparation_time": 0.0035, | |
| "eval_runtime": 754.0093, | |
| "eval_samples_per_second": 92.358, | |
| "eval_steps_per_second": 11.545, | |
| "step": 29385 | |
| }, | |
| { | |
| "epoch": 9.03522205206738, | |
| "grad_norm": 7.1302971839904785, | |
| "learning_rate": 8.696319018404908e-07, | |
| "loss": 2.4115, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 9.188361408882082, | |
| "grad_norm": 5.568242073059082, | |
| "learning_rate": 7.315950920245399e-07, | |
| "loss": 2.3616, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 9.341500765696784, | |
| "grad_norm": 4.705562591552734, | |
| "learning_rate": 5.93558282208589e-07, | |
| "loss": 2.3858, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 9.494640122511486, | |
| "grad_norm": 6.591212749481201, | |
| "learning_rate": 4.5552147239263803e-07, | |
| "loss": 2.3935, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 9.647779479326188, | |
| "grad_norm": 7.563518524169922, | |
| "learning_rate": 3.174846625766871e-07, | |
| "loss": 2.3611, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 9.800918836140887, | |
| "grad_norm": 7.720861434936523, | |
| "learning_rate": 1.794478527607362e-07, | |
| "loss": 2.3667, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 9.95405819295559, | |
| "grad_norm": 5.801350116729736, | |
| "learning_rate": 4.141104294478528e-08, | |
| "loss": 2.3529, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6372434986142822, | |
| "eval_loss": 2.5461041927337646, | |
| "eval_model_preparation_time": 0.0035, | |
| "eval_runtime": 700.2631, | |
| "eval_samples_per_second": 99.447, | |
| "eval_steps_per_second": 12.431, | |
| "step": 32650 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 32650, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 8.117946756716657e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |