| { | |
| "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": 3.6468505859375, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5544, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7218045112781954, | |
| "eval_f1": 0.6745776909116292, | |
| "eval_loss": 0.5251643657684326, | |
| "eval_precision": 0.6704603946239633, | |
| "eval_recall": 0.6806692125841063, | |
| "eval_runtime": 5.1543, | |
| "eval_samples_per_second": 77.411, | |
| "eval_steps_per_second": 9.701, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.0383548736572266, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4974, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7844611528822055, | |
| "eval_f1": 0.6839633068081344, | |
| "eval_loss": 0.4520864188671112, | |
| "eval_precision": 0.7792397660818713, | |
| "eval_recall": 0.6649845426441171, | |
| "eval_runtime": 5.0577, | |
| "eval_samples_per_second": 78.89, | |
| "eval_steps_per_second": 9.886, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.7733473777771, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4206, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.782902317244911, | |
| "eval_loss": 0.36416465044021606, | |
| "eval_precision": 0.8034409776746092, | |
| "eval_recall": 0.7694126204764502, | |
| "eval_runtime": 5.0665, | |
| "eval_samples_per_second": 78.752, | |
| "eval_steps_per_second": 9.869, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.6582493782043457, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3657, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8122237052238519, | |
| "eval_loss": 0.33759891986846924, | |
| "eval_precision": 0.8184491978609625, | |
| "eval_recall": 0.8068285142753229, | |
| "eval_runtime": 5.0866, | |
| "eval_samples_per_second": 78.442, | |
| "eval_steps_per_second": 9.83, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.0865817070007324, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3188, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8167483159828537, | |
| "eval_loss": 0.3197535574436188, | |
| "eval_precision": 0.8201621387462095, | |
| "eval_recall": 0.8136024731769412, | |
| "eval_runtime": 5.1213, | |
| "eval_samples_per_second": 77.91, | |
| "eval_steps_per_second": 9.763, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.176858425140381, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3069, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8255172205802521, | |
| "eval_loss": 0.3127439022064209, | |
| "eval_precision": 0.8239495798319327, | |
| "eval_recall": 0.8271503909801782, | |
| "eval_runtime": 5.0741, | |
| "eval_samples_per_second": 78.635, | |
| "eval_steps_per_second": 9.854, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.9114988446235657, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2838, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8359744037230948, | |
| "eval_loss": 0.30525630712509155, | |
| "eval_precision": 0.8448835433371515, | |
| "eval_recall": 0.828514275322786, | |
| "eval_runtime": 5.0555, | |
| "eval_samples_per_second": 78.924, | |
| "eval_steps_per_second": 9.89, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.085402965545654, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2699, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8403693509153758, | |
| "eval_loss": 0.2976122498512268, | |
| "eval_precision": 0.8647333925035843, | |
| "eval_recall": 0.8238316057464994, | |
| "eval_runtime": 5.0901, | |
| "eval_samples_per_second": 78.388, | |
| "eval_steps_per_second": 9.823, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.7740559577941895, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2614, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8342105263157895, | |
| "eval_loss": 0.2887146770954132, | |
| "eval_precision": 0.8398085585585586, | |
| "eval_recall": 0.82924168030551, | |
| "eval_runtime": 5.0975, | |
| "eval_samples_per_second": 78.274, | |
| "eval_steps_per_second": 9.809, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.8042409420013428, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2515, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8298403801632752, | |
| "eval_loss": 0.2852274477481842, | |
| "eval_precision": 0.8315523576240049, | |
| "eval_recall": 0.8281960356428442, | |
| "eval_runtime": 5.0847, | |
| "eval_samples_per_second": 78.471, | |
| "eval_steps_per_second": 9.833, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.159557819366455, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2453, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.844327731092437, | |
| "eval_loss": 0.2800367474555969, | |
| "eval_precision": 0.8411320530352577, | |
| "eval_recall": 0.8477905073649754, | |
| "eval_runtime": 5.0685, | |
| "eval_samples_per_second": 78.721, | |
| "eval_steps_per_second": 9.865, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.643934965133667, | |
| "learning_rate": 2e-05, | |
| "loss": 0.236, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8502252252252251, | |
| "eval_loss": 0.27178239822387695, | |
| "eval_precision": 0.863265306122449, | |
| "eval_recall": 0.8398799781778505, | |
| "eval_runtime": 5.0641, | |
| "eval_samples_per_second": 78.79, | |
| "eval_steps_per_second": 9.873, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 2.2930312156677246, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.227, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8541488972828073, | |
| "eval_loss": 0.2711658477783203, | |
| "eval_precision": 0.8559859154929578, | |
| "eval_recall": 0.8523822513184216, | |
| "eval_runtime": 5.1106, | |
| "eval_samples_per_second": 78.073, | |
| "eval_steps_per_second": 9.784, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 3.1738922595977783, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.227, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8478885630498534, | |
| "eval_loss": 0.27570030093193054, | |
| "eval_precision": 0.8385980403326877, | |
| "eval_recall": 0.8602927805055465, | |
| "eval_runtime": 5.1226, | |
| "eval_samples_per_second": 77.891, | |
| "eval_steps_per_second": 9.761, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 2.479130506515503, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2171, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8622085718274466, | |
| "eval_loss": 0.27079105377197266, | |
| "eval_precision": 0.8530168716042322, | |
| "eval_recall": 0.8741589379887251, | |
| "eval_runtime": 5.0709, | |
| "eval_samples_per_second": 78.684, | |
| "eval_steps_per_second": 9.86, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 2.9859211444854736, | |
| "learning_rate": 1e-05, | |
| "loss": 0.214, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8629148629148629, | |
| "eval_loss": 0.2631725072860718, | |
| "eval_precision": 0.8657894736842104, | |
| "eval_recall": 0.860201854882706, | |
| "eval_runtime": 5.0662, | |
| "eval_samples_per_second": 78.757, | |
| "eval_steps_per_second": 9.869, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 4.846933364868164, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2124, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.858259325044405, | |
| "eval_loss": 0.26394030451774597, | |
| "eval_precision": 0.8573798178418481, | |
| "eval_recall": 0.8591562102200401, | |
| "eval_runtime": 5.0705, | |
| "eval_samples_per_second": 78.691, | |
| "eval_steps_per_second": 9.861, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 6.029839515686035, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2166, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8636104675452922, | |
| "eval_loss": 0.2631914019584656, | |
| "eval_precision": 0.8645363713902765, | |
| "eval_recall": 0.8627023095108202, | |
| "eval_runtime": 5.0927, | |
| "eval_samples_per_second": 78.348, | |
| "eval_steps_per_second": 9.818, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 2.059579372406006, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2086, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8575487105473052, | |
| "eval_loss": 0.2630336880683899, | |
| "eval_precision": 0.8584592421103936, | |
| "eval_recall": 0.8566557555919259, | |
| "eval_runtime": 5.1111, | |
| "eval_samples_per_second": 78.066, | |
| "eval_steps_per_second": 9.783, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 2.347975254058838, | |
| "learning_rate": 0.0, | |
| "loss": 0.2113, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8629148629148629, | |
| "eval_loss": 0.26311013102531433, | |
| "eval_precision": 0.8657894736842104, | |
| "eval_recall": 0.860201854882706, | |
| "eval_runtime": 5.0704, | |
| "eval_samples_per_second": 78.692, | |
| "eval_steps_per_second": 9.861, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7650353185560000.0, | |
| "train_loss": 0.28728070728114397, | |
| "train_runtime": 1942.6788, | |
| "train_samples_per_second": 37.526, | |
| "train_steps_per_second": 1.256 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7650353185560000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |