| { | |
| "best_metric": 0.6367893815040588, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-500", | |
| "epoch": 0.018298261665141813, | |
| "eval_steps": 50, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.659652333028362e-05, | |
| "eval_loss": 3.3232226371765137, | |
| "eval_runtime": 654.1201, | |
| "eval_samples_per_second": 17.59, | |
| "eval_steps_per_second": 4.398, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00036596523330283625, | |
| "grad_norm": 3.221409559249878, | |
| "learning_rate": 4.34e-05, | |
| "loss": 2.2773, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0007319304666056725, | |
| "grad_norm": 2.2033233642578125, | |
| "learning_rate": 8.68e-05, | |
| "loss": 1.6412, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0010978956999085087, | |
| "grad_norm": 2.0530292987823486, | |
| "learning_rate": 0.0001302, | |
| "loss": 1.2863, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.001463860933211345, | |
| "grad_norm": 2.043330430984497, | |
| "learning_rate": 0.0001736, | |
| "loss": 1.1969, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0018298261665141812, | |
| "grad_norm": 5.248676300048828, | |
| "learning_rate": 0.000217, | |
| "loss": 1.1612, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0018298261665141812, | |
| "eval_loss": 1.7948706150054932, | |
| "eval_runtime": 654.0329, | |
| "eval_samples_per_second": 17.592, | |
| "eval_steps_per_second": 4.399, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0021957913998170175, | |
| "grad_norm": 1.6893078088760376, | |
| "learning_rate": 0.00021673569945319091, | |
| "loss": 1.2685, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0025617566331198535, | |
| "grad_norm": 1.2931175231933594, | |
| "learning_rate": 0.00021594408545846038, | |
| "loss": 1.0429, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.00292772186642269, | |
| "grad_norm": 1.124497652053833, | |
| "learning_rate": 0.0002146290146796179, | |
| "loss": 1.0055, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.003293687099725526, | |
| "grad_norm": 2.295246124267578, | |
| "learning_rate": 0.0002127968940093076, | |
| "loss": 1.0593, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0036596523330283625, | |
| "grad_norm": 2.5044991970062256, | |
| "learning_rate": 0.00021045664935527106, | |
| "loss": 1.0974, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0036596523330283625, | |
| "eval_loss": 1.6511973142623901, | |
| "eval_runtime": 653.846, | |
| "eval_samples_per_second": 17.597, | |
| "eval_steps_per_second": 4.4, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0040256175663311985, | |
| "grad_norm": 1.1868317127227783, | |
| "learning_rate": 0.00020761968215422217, | |
| "loss": 1.1934, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.004391582799634035, | |
| "grad_norm": 1.1751034259796143, | |
| "learning_rate": 0.00020429981382519356, | |
| "loss": 0.8967, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.004757548032936871, | |
| "grad_norm": 1.8134667873382568, | |
| "learning_rate": 0.00020051321843297219, | |
| "loss": 0.959, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.005123513266239707, | |
| "grad_norm": 1.202273964881897, | |
| "learning_rate": 0.0001962783438896818, | |
| "loss": 1.0112, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.0054894784995425435, | |
| "grad_norm": 2.1259782314300537, | |
| "learning_rate": 0.0001916158220784091, | |
| "loss": 0.8748, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0054894784995425435, | |
| "eval_loss": 1.6277738809585571, | |
| "eval_runtime": 654.0264, | |
| "eval_samples_per_second": 17.593, | |
| "eval_steps_per_second": 4.399, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.00585544373284538, | |
| "grad_norm": 0.8180252313613892, | |
| "learning_rate": 0.00018654836833674362, | |
| "loss": 1.1652, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.006221408966148216, | |
| "grad_norm": 1.229128360748291, | |
| "learning_rate": 0.0001811006707899361, | |
| "loss": 0.9004, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.006587374199451052, | |
| "grad_norm": 2.126133680343628, | |
| "learning_rate": 0.0001752992700728339, | |
| "loss": 0.8799, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.0069533394327538885, | |
| "grad_norm": 1.0177737474441528, | |
| "learning_rate": 0.00016917243002657602, | |
| "loss": 0.8698, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.007319304666056725, | |
| "grad_norm": 2.6272943019866943, | |
| "learning_rate": 0.00016275, | |
| "loss": 0.8727, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.007319304666056725, | |
| "eval_loss": 1.3153941631317139, | |
| "eval_runtime": 653.5309, | |
| "eval_samples_per_second": 17.606, | |
| "eval_steps_per_second": 4.402, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.0076852698993595606, | |
| "grad_norm": 0.8264740109443665, | |
| "learning_rate": 0.0001560632694266149, | |
| "loss": 0.9661, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.008051235132662397, | |
| "grad_norm": 0.7382209897041321, | |
| "learning_rate": 0.00014914481538562646, | |
| "loss": 0.8098, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.008417200365965233, | |
| "grad_norm": 0.9931471943855286, | |
| "learning_rate": 0.0001420283438896818, | |
| "loss": 0.7825, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.00878316559926807, | |
| "grad_norm": 0.7864680886268616, | |
| "learning_rate": 0.00013474852567256393, | |
| "loss": 0.8832, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.009149130832570906, | |
| "grad_norm": 2.8572616577148438, | |
| "learning_rate": 0.00012734082727686196, | |
| "loss": 0.8983, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.009149130832570906, | |
| "eval_loss": 1.1016149520874023, | |
| "eval_runtime": 654.6016, | |
| "eval_samples_per_second": 17.577, | |
| "eval_steps_per_second": 4.395, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.009515096065873741, | |
| "grad_norm": 0.7957597970962524, | |
| "learning_rate": 0.0001198413382645404, | |
| "loss": 0.9094, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.009881061299176578, | |
| "grad_norm": 0.9204810261726379, | |
| "learning_rate": 0.00011228659539222137, | |
| "loss": 0.8334, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.010247026532479414, | |
| "grad_norm": 1.0104973316192627, | |
| "learning_rate": 0.00010471340460777866, | |
| "loss": 0.7847, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.01061299176578225, | |
| "grad_norm": 0.8759304285049438, | |
| "learning_rate": 9.715866173545961e-05, | |
| "loss": 0.7903, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.010978956999085087, | |
| "grad_norm": 3.211150646209717, | |
| "learning_rate": 8.965917272313806e-05, | |
| "loss": 0.7951, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.010978956999085087, | |
| "eval_loss": 0.9889414310455322, | |
| "eval_runtime": 653.2449, | |
| "eval_samples_per_second": 17.614, | |
| "eval_steps_per_second": 4.404, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.011344922232387923, | |
| "grad_norm": 0.8451036810874939, | |
| "learning_rate": 8.225147432743606e-05, | |
| "loss": 0.7332, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.01171088746569076, | |
| "grad_norm": 0.6947381496429443, | |
| "learning_rate": 7.497165611031821e-05, | |
| "loss": 0.7797, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.012076852698993596, | |
| "grad_norm": 0.7813195586204529, | |
| "learning_rate": 6.785518461437353e-05, | |
| "loss": 0.7427, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.012442817932296431, | |
| "grad_norm": 1.1668095588684082, | |
| "learning_rate": 6.093673057338509e-05, | |
| "loss": 0.6847, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.012808783165599268, | |
| "grad_norm": 1.675475001335144, | |
| "learning_rate": 5.4250000000000024e-05, | |
| "loss": 0.7452, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.012808783165599268, | |
| "eval_loss": 0.799678385257721, | |
| "eval_runtime": 653.6373, | |
| "eval_samples_per_second": 17.603, | |
| "eval_steps_per_second": 4.402, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.013174748398902104, | |
| "grad_norm": 0.8130257725715637, | |
| "learning_rate": 4.782756997342398e-05, | |
| "loss": 0.6847, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.01354071363220494, | |
| "grad_norm": 0.8550499081611633, | |
| "learning_rate": 4.170072992716607e-05, | |
| "loss": 0.6897, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.013906678865507777, | |
| "grad_norm": 0.7728462815284729, | |
| "learning_rate": 3.5899329210063916e-05, | |
| "loss": 0.7088, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.014272644098810613, | |
| "grad_norm": 1.0523146390914917, | |
| "learning_rate": 3.045163166325637e-05, | |
| "loss": 0.6873, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.01463860933211345, | |
| "grad_norm": 2.0533359050750732, | |
| "learning_rate": 2.5384177921590895e-05, | |
| "loss": 0.6468, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.01463860933211345, | |
| "eval_loss": 0.7176544666290283, | |
| "eval_runtime": 654.5529, | |
| "eval_samples_per_second": 17.578, | |
| "eval_steps_per_second": 4.395, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.015004574565416285, | |
| "grad_norm": 0.7370195984840393, | |
| "learning_rate": 2.0721656110318213e-05, | |
| "loss": 0.6352, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.015370539798719121, | |
| "grad_norm": 0.7632651329040527, | |
| "learning_rate": 1.6486781567027783e-05, | |
| "loss": 0.6466, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.01573650503202196, | |
| "grad_norm": 0.9370796084403992, | |
| "learning_rate": 1.2700186174806422e-05, | |
| "loss": 0.6818, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.016102470265324794, | |
| "grad_norm": 0.8790647387504578, | |
| "learning_rate": 9.380317845777794e-06, | |
| "loss": 0.7192, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.01646843549862763, | |
| "grad_norm": 2.4070212841033936, | |
| "learning_rate": 6.543350644728947e-06, | |
| "loss": 0.6286, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.01646843549862763, | |
| "eval_loss": 0.6477181315422058, | |
| "eval_runtime": 654.5484, | |
| "eval_samples_per_second": 17.579, | |
| "eval_steps_per_second": 4.395, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.016834400731930467, | |
| "grad_norm": 0.7986826300621033, | |
| "learning_rate": 4.2031059906924e-06, | |
| "loss": 0.5613, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.0172003659652333, | |
| "grad_norm": 0.9289596080780029, | |
| "learning_rate": 2.3709853203820825e-06, | |
| "loss": 0.6961, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.01756633119853614, | |
| "grad_norm": 0.9395463466644287, | |
| "learning_rate": 1.0559145415396157e-06, | |
| "loss": 0.6394, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.017932296431838975, | |
| "grad_norm": 0.949378252029419, | |
| "learning_rate": 2.643005468090745e-07, | |
| "loss": 0.6338, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.018298261665141813, | |
| "grad_norm": 1.6931450366973877, | |
| "learning_rate": 0.0, | |
| "loss": 0.6262, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.018298261665141813, | |
| "eval_loss": 0.6367893815040588, | |
| "eval_runtime": 654.042, | |
| "eval_samples_per_second": 17.592, | |
| "eval_steps_per_second": 4.399, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.3877540397056e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |