| { | |
| "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": 5.112319469451904, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5509, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7393483709273183, | |
| "eval_f1": 0.6507070707070707, | |
| "eval_loss": 0.4983255863189697, | |
| "eval_precision": 0.6800605637083625, | |
| "eval_recall": 0.6405710129114385, | |
| "eval_runtime": 1.7657, | |
| "eval_samples_per_second": 225.971, | |
| "eval_steps_per_second": 28.317, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.6866044998168945, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4511, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7769423558897243, | |
| "eval_f1": 0.7593078346448687, | |
| "eval_loss": 0.4377373456954956, | |
| "eval_precision": 0.7546743295019157, | |
| "eval_recall": 0.8021913075104565, | |
| "eval_runtime": 1.769, | |
| "eval_samples_per_second": 225.555, | |
| "eval_steps_per_second": 28.265, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.584764242172241, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.368, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8196102381877741, | |
| "eval_loss": 0.32603567838668823, | |
| "eval_precision": 0.8381270903010034, | |
| "eval_recall": 0.8064193489725404, | |
| "eval_runtime": 1.7715, | |
| "eval_samples_per_second": 225.23, | |
| "eval_steps_per_second": 28.224, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.8483095169067383, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3019, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8333281762485303, | |
| "eval_loss": 0.30364951491355896, | |
| "eval_precision": 0.8410471369819678, | |
| "eval_recall": 0.8267412256773959, | |
| "eval_runtime": 1.7702, | |
| "eval_samples_per_second": 225.393, | |
| "eval_steps_per_second": 28.245, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.774143934249878, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2668, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8424651921601347, | |
| "eval_loss": 0.31921207904815674, | |
| "eval_precision": 0.8372140762463343, | |
| "eval_recall": 0.8485179123476996, | |
| "eval_runtime": 1.7714, | |
| "eval_samples_per_second": 225.248, | |
| "eval_steps_per_second": 28.227, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.2327117919921875, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2471, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8380263497804185, | |
| "eval_loss": 0.30589351058006287, | |
| "eval_precision": 0.830503344095941, | |
| "eval_recall": 0.8474722676850337, | |
| "eval_runtime": 1.7732, | |
| "eval_samples_per_second": 225.015, | |
| "eval_steps_per_second": 28.197, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.5115749835968018, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2422, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8524146298159436, | |
| "eval_loss": 0.2949831783771515, | |
| "eval_precision": 0.8451250578971746, | |
| "eval_recall": 0.8613384251682124, | |
| "eval_runtime": 1.7731, | |
| "eval_samples_per_second": 225.024, | |
| "eval_steps_per_second": 28.198, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.2918312549591064, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2258, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8454251965513313, | |
| "eval_loss": 0.29280924797058105, | |
| "eval_precision": 0.8463049835506276, | |
| "eval_recall": 0.8445626477541371, | |
| "eval_runtime": 1.7799, | |
| "eval_samples_per_second": 224.171, | |
| "eval_steps_per_second": 28.092, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 5.160737037658691, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2054, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8533986527862829, | |
| "eval_loss": 0.30492648482322693, | |
| "eval_precision": 0.8572003218020917, | |
| "eval_recall": 0.8498817966903074, | |
| "eval_runtime": 1.779, | |
| "eval_samples_per_second": 224.288, | |
| "eval_steps_per_second": 28.106, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.917464017868042, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2009, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8488361520276414, | |
| "eval_loss": 0.30127042531967163, | |
| "eval_precision": 0.8488361520276414, | |
| "eval_recall": 0.8488361520276414, | |
| "eval_runtime": 1.7757, | |
| "eval_samples_per_second": 224.7, | |
| "eval_steps_per_second": 28.158, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 6.667805194854736, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1755, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.858259325044405, | |
| "eval_loss": 0.30701279640197754, | |
| "eval_precision": 0.8573798178418481, | |
| "eval_recall": 0.8591562102200401, | |
| "eval_runtime": 1.7942, | |
| "eval_samples_per_second": 222.38, | |
| "eval_steps_per_second": 27.867, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 8.611730575561523, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1821, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8568221901555235, | |
| "eval_loss": 0.2995355427265167, | |
| "eval_precision": 0.8596491228070176, | |
| "eval_recall": 0.8541553009638116, | |
| "eval_runtime": 1.7796, | |
| "eval_samples_per_second": 224.202, | |
| "eval_steps_per_second": 28.095, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 2.71295428276062, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1652, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.866029197080292, | |
| "eval_loss": 0.3272043764591217, | |
| "eval_precision": 0.8552631578947368, | |
| "eval_recall": 0.8809328968903437, | |
| "eval_runtime": 1.7775, | |
| "eval_samples_per_second": 224.467, | |
| "eval_steps_per_second": 28.129, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 5.373868942260742, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1566, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8897243107769424, | |
| "eval_f1": 0.8718540145985401, | |
| "eval_loss": 0.33357149362564087, | |
| "eval_precision": 0.8609022556390977, | |
| "eval_recall": 0.886979450809238, | |
| "eval_runtime": 1.7836, | |
| "eval_samples_per_second": 223.703, | |
| "eval_steps_per_second": 28.033, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 5.369639873504639, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1634, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8622899159663866, | |
| "eval_loss": 0.314995676279068, | |
| "eval_precision": 0.8589244307033712, | |
| "eval_recall": 0.8659301691216585, | |
| "eval_runtime": 1.777, | |
| "eval_samples_per_second": 224.539, | |
| "eval_steps_per_second": 28.138, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 6.779192924499512, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1496, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8696722245432793, | |
| "eval_loss": 0.3320792317390442, | |
| "eval_precision": 0.8706135006701596, | |
| "eval_recall": 0.8687488634297145, | |
| "eval_runtime": 1.7833, | |
| "eval_samples_per_second": 223.741, | |
| "eval_steps_per_second": 28.038, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 1.1515932083129883, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1355, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8616171059774413, | |
| "eval_loss": 0.32759982347488403, | |
| "eval_precision": 0.859873949579832, | |
| "eval_recall": 0.8634297144935443, | |
| "eval_runtime": 1.7782, | |
| "eval_samples_per_second": 224.387, | |
| "eval_steps_per_second": 28.119, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 1.6571087837219238, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1477, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8563025210084034, | |
| "eval_loss": 0.33653610944747925, | |
| "eval_precision": 0.8529936381473334, | |
| "eval_recall": 0.8598836152027641, | |
| "eval_runtime": 1.7851, | |
| "eval_samples_per_second": 223.518, | |
| "eval_steps_per_second": 28.01, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 2.6701011657714844, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1317, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.858259325044405, | |
| "eval_loss": 0.3385031819343567, | |
| "eval_precision": 0.8573798178418481, | |
| "eval_recall": 0.8591562102200401, | |
| "eval_runtime": 1.7765, | |
| "eval_samples_per_second": 224.597, | |
| "eval_steps_per_second": 28.145, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.197312593460083, | |
| "learning_rate": 0.0, | |
| "loss": 0.1267, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.858259325044405, | |
| "eval_loss": 0.3389217257499695, | |
| "eval_precision": 0.8573798178418481, | |
| "eval_recall": 0.8591562102200401, | |
| "eval_runtime": 1.7779, | |
| "eval_samples_per_second": 224.423, | |
| "eval_steps_per_second": 28.123, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.22970127551282038, | |
| "train_runtime": 621.3103, | |
| "train_samples_per_second": 117.107, | |
| "train_steps_per_second": 3.927 | |
| } | |
| ], | |
| "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 | |
| } | |