| { | |
| "best_metric": 0.9532240352938909, | |
| "best_model_checkpoint": "./output//20_roberta-large_nace_5__5e-6_0.01_0.06_07-21-22_10-40/checkpoint-40000", | |
| "epoch": 0.7299336672779861, | |
| "global_step": 40000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 6.082725060827251e-07, | |
| "loss": 0.3132, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_accuracy": 0.8434321142258856, | |
| "eval_f1": 0.8909110843109309, | |
| "eval_loss": 0.1265411078929901, | |
| "eval_roc_auc": 0.9213767705759509, | |
| "eval_runtime": 2893.0443, | |
| "eval_samples_per_second": 33.674, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.2165450121654502e-06, | |
| "loss": 0.1068, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_accuracy": 0.888648238059556, | |
| "eval_f1": 0.9273535044794836, | |
| "eval_loss": 0.07977746427059174, | |
| "eval_roc_auc": 0.9485323633587821, | |
| "eval_runtime": 2892.136, | |
| "eval_samples_per_second": 33.685, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.8248175182481753e-06, | |
| "loss": 0.0828, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_accuracy": 0.8895925929727677, | |
| "eval_f1": 0.9305848812431281, | |
| "eval_loss": 0.07206098735332489, | |
| "eval_roc_auc": 0.9527906922517034, | |
| "eval_runtime": 2892.9744, | |
| "eval_samples_per_second": 33.675, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.4330900243309005e-06, | |
| "loss": 0.0731, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_accuracy": 0.9002679093829873, | |
| "eval_f1": 0.937980881324318, | |
| "eval_loss": 0.06552038341760635, | |
| "eval_roc_auc": 0.9577804517072407, | |
| "eval_runtime": 2894.2574, | |
| "eval_samples_per_second": 33.66, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.0413625304136257e-06, | |
| "loss": 0.0696, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_accuracy": 0.9033781217601954, | |
| "eval_f1": 0.9405285597448666, | |
| "eval_loss": 0.06280936300754547, | |
| "eval_roc_auc": 0.9606146188592658, | |
| "eval_runtime": 2894.0668, | |
| "eval_samples_per_second": 33.662, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.6496350364963505e-06, | |
| "loss": 0.066, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_accuracy": 0.905287361041254, | |
| "eval_f1": 0.9417959549161978, | |
| "eval_loss": 0.05985909327864647, | |
| "eval_roc_auc": 0.9594101544601925, | |
| "eval_runtime": 2895.254, | |
| "eval_samples_per_second": 33.649, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.257907542579076e-06, | |
| "loss": 0.0637, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_accuracy": 0.9088081625111629, | |
| "eval_f1": 0.9438244788000874, | |
| "eval_loss": 0.05829664319753647, | |
| "eval_roc_auc": 0.9615365087786716, | |
| "eval_runtime": 2895.6689, | |
| "eval_samples_per_second": 33.644, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.866180048661801e-06, | |
| "loss": 0.0623, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 0.9128319356196303, | |
| "eval_f1": 0.9465757082857906, | |
| "eval_loss": 0.0566316694021225, | |
| "eval_roc_auc": 0.963087400509175, | |
| "eval_runtime": 2896.4086, | |
| "eval_samples_per_second": 33.635, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.969715206460757e-06, | |
| "loss": 0.0607, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.9125855821640099, | |
| "eval_f1": 0.9472601377088317, | |
| "eval_loss": 0.05489673465490341, | |
| "eval_roc_auc": 0.9637262945675201, | |
| "eval_runtime": 2897.0113, | |
| "eval_samples_per_second": 33.628, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.9308885480771105e-06, | |
| "loss": 0.0602, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_accuracy": 0.9147206454460537, | |
| "eval_f1": 0.9477220503970293, | |
| "eval_loss": 0.05647333711385727, | |
| "eval_roc_auc": 0.9643404592959164, | |
| "eval_runtime": 2898.0195, | |
| "eval_samples_per_second": 33.616, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.8920618896934645e-06, | |
| "loss": 0.0579, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.9125547879820572, | |
| "eval_f1": 0.9480550834363412, | |
| "eval_loss": 0.05308946967124939, | |
| "eval_roc_auc": 0.9659595651426466, | |
| "eval_runtime": 2898.0594, | |
| "eval_samples_per_second": 33.616, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.8532352313098175e-06, | |
| "loss": 0.0574, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_accuracy": 0.9161474425431888, | |
| "eval_f1": 0.949495831332664, | |
| "eval_loss": 0.051538560539484024, | |
| "eval_roc_auc": 0.9652335991771203, | |
| "eval_runtime": 2898.2265, | |
| "eval_samples_per_second": 33.614, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.8144085729261714e-06, | |
| "loss": 0.0566, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_accuracy": 0.9136941727143019, | |
| "eval_f1": 0.9496842358026173, | |
| "eval_loss": 0.05137912556529045, | |
| "eval_roc_auc": 0.968884227918182, | |
| "eval_runtime": 2899.198, | |
| "eval_samples_per_second": 33.603, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.775581914542525e-06, | |
| "loss": 0.0559, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_accuracy": 0.9152133523572946, | |
| "eval_f1": 0.9488828443753642, | |
| "eval_loss": 0.05188358202576637, | |
| "eval_roc_auc": 0.9643445491459475, | |
| "eval_runtime": 2898.1027, | |
| "eval_samples_per_second": 33.615, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.736755256158879e-06, | |
| "loss": 0.055, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_accuracy": 0.9187033596452511, | |
| "eval_f1": 0.9510342387062685, | |
| "eval_loss": 0.05020885169506073, | |
| "eval_roc_auc": 0.9663431709190323, | |
| "eval_runtime": 2897.1407, | |
| "eval_samples_per_second": 33.627, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.697928597775233e-06, | |
| "loss": 0.0547, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_accuracy": 0.9174921218217839, | |
| "eval_f1": 0.9501358873370442, | |
| "eval_loss": 0.05047479644417763, | |
| "eval_roc_auc": 0.9660343939821142, | |
| "eval_runtime": 2897.8899, | |
| "eval_samples_per_second": 33.618, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.659101939391586e-06, | |
| "loss": 0.0532, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_accuracy": 0.920027509469211, | |
| "eval_f1": 0.9515785386687373, | |
| "eval_loss": 0.04900650307536125, | |
| "eval_roc_auc": 0.9662507314169619, | |
| "eval_runtime": 2897.2054, | |
| "eval_samples_per_second": 33.626, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.62027528100794e-06, | |
| "loss": 0.0525, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_accuracy": 0.9201712156516562, | |
| "eval_f1": 0.9524362960165035, | |
| "eval_loss": 0.04841683432459831, | |
| "eval_roc_auc": 0.9675610696135818, | |
| "eval_runtime": 2898.5852, | |
| "eval_samples_per_second": 33.61, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.581448622624294e-06, | |
| "loss": 0.0535, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_accuracy": 0.9190934192833168, | |
| "eval_f1": 0.9519025007608072, | |
| "eval_loss": 0.04914608970284462, | |
| "eval_roc_auc": 0.9681339368085271, | |
| "eval_runtime": 2898.1713, | |
| "eval_samples_per_second": 33.615, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.542621964240648e-06, | |
| "loss": 0.0522, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_accuracy": 0.9198530091048132, | |
| "eval_f1": 0.9532240352938909, | |
| "eval_loss": 0.04747059941291809, | |
| "eval_roc_auc": 0.9699960586310394, | |
| "eval_runtime": 2897.6551, | |
| "eval_samples_per_second": 33.621, | |
| "step": 40000 | |
| } | |
| ], | |
| "max_steps": 273995, | |
| "num_train_epochs": 5, | |
| "total_flos": 6.9867777687552e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |