| { | |
| "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": 3.6051695346832275, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.556, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7167919799498746, | |
| "eval_f1": 0.6641239002659476, | |
| "eval_loss": 0.5324748158454895, | |
| "eval_precision": 0.6617008797653958, | |
| "eval_recall": 0.6671212947808693, | |
| "eval_runtime": 5.0818, | |
| "eval_samples_per_second": 78.516, | |
| "eval_steps_per_second": 9.839, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.362328290939331, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5103, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7719298245614035, | |
| "eval_f1": 0.6523624874335775, | |
| "eval_loss": 0.4822019934654236, | |
| "eval_precision": 0.7714565527065527, | |
| "eval_recall": 0.6386161120203673, | |
| "eval_runtime": 5.0545, | |
| "eval_samples_per_second": 78.94, | |
| "eval_steps_per_second": 9.892, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.656192779541016, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4637, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8045112781954887, | |
| "eval_f1": 0.7479591836734694, | |
| "eval_loss": 0.42453086376190186, | |
| "eval_precision": 0.771505376344086, | |
| "eval_recall": 0.7341789416257501, | |
| "eval_runtime": 5.075, | |
| "eval_samples_per_second": 78.62, | |
| "eval_steps_per_second": 9.852, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.4060696363449097, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4173, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8245614035087719, | |
| "eval_f1": 0.7873004752040941, | |
| "eval_loss": 0.38980478048324585, | |
| "eval_precision": 0.7887936313533375, | |
| "eval_recall": 0.7858701582105838, | |
| "eval_runtime": 5.064, | |
| "eval_samples_per_second": 78.791, | |
| "eval_steps_per_second": 9.874, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.650151491165161, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3674, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.7999105055663995, | |
| "eval_loss": 0.3571353554725647, | |
| "eval_precision": 0.8058980811575966, | |
| "eval_recall": 0.794735406437534, | |
| "eval_runtime": 5.0507, | |
| "eval_samples_per_second": 78.999, | |
| "eval_steps_per_second": 9.9, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.5915608406066895, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3484, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8029928975654221, | |
| "eval_loss": 0.3431943356990814, | |
| "eval_precision": 0.8037650785914463, | |
| "eval_recall": 0.8022367703218767, | |
| "eval_runtime": 5.135, | |
| "eval_samples_per_second": 77.703, | |
| "eval_steps_per_second": 9.737, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.9446077346801758, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3247, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.816408656658218, | |
| "eval_loss": 0.3298611342906952, | |
| "eval_precision": 0.8270654903728508, | |
| "eval_recall": 0.8078741589379888, | |
| "eval_runtime": 5.0753, | |
| "eval_samples_per_second": 78.617, | |
| "eval_steps_per_second": 9.852, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 9.143532752990723, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3102, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8228198648441347, | |
| "eval_loss": 0.3259894549846649, | |
| "eval_precision": 0.8509591907917684, | |
| "eval_recall": 0.8049645390070922, | |
| "eval_runtime": 5.087, | |
| "eval_samples_per_second": 78.435, | |
| "eval_steps_per_second": 9.829, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.818455696105957, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2991, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.821647309770462, | |
| "eval_loss": 0.31378358602523804, | |
| "eval_precision": 0.8349087353324641, | |
| "eval_recall": 0.8114202582287688, | |
| "eval_runtime": 5.0565, | |
| "eval_samples_per_second": 78.909, | |
| "eval_steps_per_second": 9.888, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.8728206157684326, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.29, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8180088078011953, | |
| "eval_loss": 0.31225934624671936, | |
| "eval_precision": 0.8323930726843348, | |
| "eval_recall": 0.8071467539552646, | |
| "eval_runtime": 5.0562, | |
| "eval_samples_per_second": 78.913, | |
| "eval_steps_per_second": 9.889, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.6045541763305664, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2778, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8377065410088949, | |
| "eval_loss": 0.30650559067726135, | |
| "eval_precision": 0.8423344947735192, | |
| "eval_recall": 0.8335151845790143, | |
| "eval_runtime": 5.0839, | |
| "eval_samples_per_second": 78.483, | |
| "eval_steps_per_second": 9.835, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.5327140092849731, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2702, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.821647309770462, | |
| "eval_loss": 0.3005804717540741, | |
| "eval_precision": 0.8349087353324641, | |
| "eval_recall": 0.8114202582287688, | |
| "eval_runtime": 5.0646, | |
| "eval_samples_per_second": 78.782, | |
| "eval_steps_per_second": 9.872, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 5.446022987365723, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2664, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8298403801632752, | |
| "eval_loss": 0.29961732029914856, | |
| "eval_precision": 0.8315523576240049, | |
| "eval_recall": 0.8281960356428442, | |
| "eval_runtime": 5.0899, | |
| "eval_samples_per_second": 78.39, | |
| "eval_steps_per_second": 9.823, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 2.8527348041534424, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.264, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8476882658063937, | |
| "eval_loss": 0.2987360656261444, | |
| "eval_precision": 0.8437296561519796, | |
| "eval_recall": 0.8520640116384797, | |
| "eval_runtime": 5.0668, | |
| "eval_samples_per_second": 78.748, | |
| "eval_steps_per_second": 9.868, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 2.7352912425994873, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.254, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8522278069611882, | |
| "eval_loss": 0.2951277792453766, | |
| "eval_precision": 0.8513631702756499, | |
| "eval_recall": 0.8531096563011457, | |
| "eval_runtime": 5.222, | |
| "eval_samples_per_second": 76.408, | |
| "eval_steps_per_second": 9.575, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 2.442108154296875, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2571, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8350789627607721, | |
| "eval_loss": 0.2944652736186981, | |
| "eval_precision": 0.8463358876939919, | |
| "eval_recall": 0.8260138206946717, | |
| "eval_runtime": 5.0611, | |
| "eval_samples_per_second": 78.837, | |
| "eval_steps_per_second": 9.879, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 3.852628231048584, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2511, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8454251965513313, | |
| "eval_loss": 0.2917979061603546, | |
| "eval_precision": 0.8463049835506276, | |
| "eval_recall": 0.8445626477541371, | |
| "eval_runtime": 5.0682, | |
| "eval_samples_per_second": 78.727, | |
| "eval_steps_per_second": 9.866, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 5.464624881744385, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2574, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8472902633190447, | |
| "eval_loss": 0.29094478487968445, | |
| "eval_precision": 0.8510272912927781, | |
| "eval_recall": 0.8438352427714131, | |
| "eval_runtime": 5.1003, | |
| "eval_samples_per_second": 78.231, | |
| "eval_steps_per_second": 9.803, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 2.1257989406585693, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2508, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.850729517396184, | |
| "eval_loss": 0.29074448347091675, | |
| "eval_precision": 0.8535087719298247, | |
| "eval_recall": 0.8481087470449173, | |
| "eval_runtime": 5.1136, | |
| "eval_samples_per_second": 78.027, | |
| "eval_steps_per_second": 9.778, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.8716917037963867, | |
| "learning_rate": 0.0, | |
| "loss": 0.2536, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.850729517396184, | |
| "eval_loss": 0.29076284170150757, | |
| "eval_precision": 0.8535087719298247, | |
| "eval_recall": 0.8481087470449173, | |
| "eval_runtime": 5.0627, | |
| "eval_samples_per_second": 78.811, | |
| "eval_steps_per_second": 9.876, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7624554283800000.0, | |
| "train_loss": 0.32445813476062213, | |
| "train_runtime": 1939.7236, | |
| "train_samples_per_second": 37.583, | |
| "train_steps_per_second": 1.258 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7624554283800000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |