| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.232791900634766, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5568, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7243107769423559, | |
| "eval_f1": 0.6144317942230656, | |
| "eval_loss": 0.48217353224754333, | |
| "eval_precision": 0.65566534914361, | |
| "eval_recall": 0.6074286233860702, | |
| "eval_runtime": 1.8124, | |
| "eval_samples_per_second": 220.155, | |
| "eval_steps_per_second": 27.588, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.6621267795562744, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4661, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7543859649122807, | |
| "eval_f1": 0.7304054054054054, | |
| "eval_loss": 0.44529902935028076, | |
| "eval_precision": 0.7240563585317666, | |
| "eval_recall": 0.7612293144208038, | |
| "eval_runtime": 1.7908, | |
| "eval_samples_per_second": 222.81, | |
| "eval_steps_per_second": 27.921, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.223342180252075, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3875, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8238834717707957, | |
| "eval_loss": 0.3446956276893616, | |
| "eval_precision": 0.8487520627062706, | |
| "eval_recall": 0.8074649936352064, | |
| "eval_runtime": 1.7891, | |
| "eval_samples_per_second": 223.02, | |
| "eval_steps_per_second": 27.947, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.7967684268951416, | |
| "learning_rate": 4e-05, | |
| "loss": 0.318, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8267427049559981, | |
| "eval_loss": 0.34423500299453735, | |
| "eval_precision": 0.8158466596088483, | |
| "eval_recall": 0.8436079287143117, | |
| "eval_runtime": 1.786, | |
| "eval_samples_per_second": 223.404, | |
| "eval_steps_per_second": 27.996, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.4756224453449249, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2855, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.826007326007326, | |
| "eval_loss": 0.3348763883113861, | |
| "eval_precision": 0.8157828282828283, | |
| "eval_recall": 0.8411074740861975, | |
| "eval_runtime": 1.7834, | |
| "eval_samples_per_second": 223.733, | |
| "eval_steps_per_second": 28.037, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.1933702230453491, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2638, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8177454831659652, | |
| "eval_loss": 0.35479673743247986, | |
| "eval_precision": 0.805161943319838, | |
| "eval_recall": 0.8472449536279323, | |
| "eval_runtime": 1.7837, | |
| "eval_samples_per_second": 223.688, | |
| "eval_steps_per_second": 28.031, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.3131154775619507, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2397, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8434065934065934, | |
| "eval_loss": 0.3253527581691742, | |
| "eval_precision": 0.8325757575757575, | |
| "eval_recall": 0.8592471358428806, | |
| "eval_runtime": 1.7938, | |
| "eval_samples_per_second": 222.438, | |
| "eval_steps_per_second": 27.874, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.750446319580078, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2428, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8654532336864889, | |
| "eval_loss": 0.2798740565776825, | |
| "eval_precision": 0.8804269882659713, | |
| "eval_recall": 0.8537461356610292, | |
| "eval_runtime": 1.7849, | |
| "eval_samples_per_second": 223.546, | |
| "eval_steps_per_second": 28.013, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 5.672217845916748, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2229, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8484099018899409, | |
| "eval_loss": 0.29030779004096985, | |
| "eval_precision": 0.8430645161290322, | |
| "eval_recall": 0.8545644662665939, | |
| "eval_runtime": 1.7837, | |
| "eval_samples_per_second": 223.689, | |
| "eval_steps_per_second": 28.031, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.1360011100769043, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2144, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8769602202215754, | |
| "eval_loss": 0.2583388686180115, | |
| "eval_precision": 0.8742831541218639, | |
| "eval_recall": 0.8797963266048372, | |
| "eval_runtime": 1.7944, | |
| "eval_samples_per_second": 222.362, | |
| "eval_steps_per_second": 27.865, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 4.677872657775879, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1967, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8622085718274466, | |
| "eval_loss": 0.27431806921958923, | |
| "eval_precision": 0.8530168716042322, | |
| "eval_recall": 0.8741589379887251, | |
| "eval_runtime": 1.7823, | |
| "eval_samples_per_second": 223.874, | |
| "eval_steps_per_second": 28.054, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 2.564518451690674, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1855, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8563451067988272, | |
| "eval_loss": 0.29132312536239624, | |
| "eval_precision": 0.8473119816985988, | |
| "eval_recall": 0.8681123840698308, | |
| "eval_runtime": 1.7831, | |
| "eval_samples_per_second": 223.773, | |
| "eval_steps_per_second": 28.042, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 2.314499855041504, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1761, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9022556390977443, | |
| "eval_f1": 0.8786430103333984, | |
| "eval_loss": 0.26596176624298096, | |
| "eval_precision": 0.8913001481099878, | |
| "eval_recall": 0.8683396981269322, | |
| "eval_runtime": 1.7906, | |
| "eval_samples_per_second": 222.829, | |
| "eval_steps_per_second": 27.923, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 7.584296226501465, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1733, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8622085718274466, | |
| "eval_loss": 0.28683483600616455, | |
| "eval_precision": 0.8530168716042322, | |
| "eval_recall": 0.8741589379887251, | |
| "eval_runtime": 1.7857, | |
| "eval_samples_per_second": 223.438, | |
| "eval_steps_per_second": 28.0, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.435178518295288, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1582, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8648373983739837, | |
| "eval_loss": 0.28010857105255127, | |
| "eval_precision": 0.8561154177433248, | |
| "eval_recall": 0.8759319876341153, | |
| "eval_runtime": 1.788, | |
| "eval_samples_per_second": 223.159, | |
| "eval_steps_per_second": 27.965, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.7755215167999268, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1537, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8550061050061051, | |
| "eval_loss": 0.30731528997421265, | |
| "eval_precision": 0.8437710437710437, | |
| "eval_recall": 0.8713402436806692, | |
| "eval_runtime": 1.7883, | |
| "eval_samples_per_second": 223.118, | |
| "eval_steps_per_second": 27.96, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 3.1951849460601807, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1537, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8781334505389722, | |
| "eval_loss": 0.2702249586582184, | |
| "eval_precision": 0.872316715542522, | |
| "eval_recall": 0.8847972358610656, | |
| "eval_runtime": 1.7845, | |
| "eval_samples_per_second": 223.597, | |
| "eval_steps_per_second": 28.02, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.639573097229004, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1461, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8760282890453928, | |
| "eval_loss": 0.29228049516677856, | |
| "eval_precision": 0.8682260305697083, | |
| "eval_recall": 0.8855246408437898, | |
| "eval_runtime": 1.8077, | |
| "eval_samples_per_second": 220.726, | |
| "eval_steps_per_second": 27.66, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.177137613296509, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1449, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8754533563232059, | |
| "eval_loss": 0.27906104922294617, | |
| "eval_precision": 0.8689781021897811, | |
| "eval_recall": 0.8830241862156756, | |
| "eval_runtime": 1.7931, | |
| "eval_samples_per_second": 222.52, | |
| "eval_steps_per_second": 27.885, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.6333515048027039, | |
| "learning_rate": 0.0, | |
| "loss": 0.1502, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8781334505389722, | |
| "eval_loss": 0.2797949016094208, | |
| "eval_precision": 0.872316715542522, | |
| "eval_recall": 0.8847972358610656, | |
| "eval_runtime": 1.7912, | |
| "eval_samples_per_second": 222.75, | |
| "eval_steps_per_second": 27.914, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.24180538537072355, | |
| "train_runtime": 620.8783, | |
| "train_samples_per_second": 117.189, | |
| "train_steps_per_second": 3.93 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 8444128359504000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |