| { | |
| "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": 2.953822374343872, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5448, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7243107769423559, | |
| "eval_f1": 0.6567661411411412, | |
| "eval_loss": 0.5046952366828918, | |
| "eval_precision": 0.6629251700680272, | |
| "eval_recall": 0.6524368066921259, | |
| "eval_runtime": 5.0686, | |
| "eval_samples_per_second": 78.72, | |
| "eval_steps_per_second": 9.865, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.6171164512634277, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4527, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7944862155388471, | |
| "eval_f1": 0.775152564736929, | |
| "eval_loss": 0.4319760501384735, | |
| "eval_precision": 0.7667055167055168, | |
| "eval_recall": 0.8121022004000726, | |
| "eval_runtime": 5.0454, | |
| "eval_samples_per_second": 79.082, | |
| "eval_steps_per_second": 9.91, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.439111709594727, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3603, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.7984931903796002, | |
| "eval_loss": 0.33696243166923523, | |
| "eval_precision": 0.8393082695594026, | |
| "eval_recall": 0.7768230587379523, | |
| "eval_runtime": 5.0496, | |
| "eval_samples_per_second": 79.016, | |
| "eval_steps_per_second": 9.902, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.9325271248817444, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3081, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8461962888779714, | |
| "eval_loss": 0.29947736859321594, | |
| "eval_precision": 0.8453465227094517, | |
| "eval_recall": 0.8470631023822512, | |
| "eval_runtime": 5.0527, | |
| "eval_samples_per_second": 78.967, | |
| "eval_steps_per_second": 9.896, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.1945856809616089, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2793, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8456742372671576, | |
| "eval_loss": 0.3008015751838684, | |
| "eval_precision": 0.8536697247706422, | |
| "eval_recall": 0.8388343335151845, | |
| "eval_runtime": 5.0466, | |
| "eval_samples_per_second": 79.063, | |
| "eval_steps_per_second": 9.908, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.8788158893585205, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2526, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8411818738518064, | |
| "eval_loss": 0.2986568510532379, | |
| "eval_precision": 0.8448542607834644, | |
| "eval_recall": 0.8377886888525186, | |
| "eval_runtime": 5.0571, | |
| "eval_samples_per_second": 78.899, | |
| "eval_steps_per_second": 9.887, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.2110929489135742, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2478, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.846679266293906, | |
| "eval_loss": 0.3030013144016266, | |
| "eval_precision": 0.8609191655801824, | |
| "eval_recall": 0.8356064739043463, | |
| "eval_runtime": 5.0491, | |
| "eval_samples_per_second": 79.023, | |
| "eval_steps_per_second": 9.903, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.5065371990203857, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2337, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8350789627607721, | |
| "eval_loss": 0.2974022924900055, | |
| "eval_precision": 0.8463358876939919, | |
| "eval_recall": 0.8260138206946717, | |
| "eval_runtime": 5.0637, | |
| "eval_samples_per_second": 78.796, | |
| "eval_steps_per_second": 9.874, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.127802848815918, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.217, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8395201930584144, | |
| "eval_loss": 0.2773597240447998, | |
| "eval_precision": 0.8562091503267973, | |
| "eval_recall": 0.8270594653573378, | |
| "eval_runtime": 5.0754, | |
| "eval_samples_per_second": 78.615, | |
| "eval_steps_per_second": 9.851, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.408611297607422, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1966, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.844327731092437, | |
| "eval_loss": 0.2846048176288605, | |
| "eval_precision": 0.8411320530352577, | |
| "eval_recall": 0.8477905073649754, | |
| "eval_runtime": 5.0388, | |
| "eval_samples_per_second": 79.185, | |
| "eval_steps_per_second": 9.923, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.3639743328094482, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.199, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8545433391506689, | |
| "eval_loss": 0.2909916341304779, | |
| "eval_precision": 0.8639270714012982, | |
| "eval_recall": 0.846653937079469, | |
| "eval_runtime": 5.0638, | |
| "eval_samples_per_second": 78.794, | |
| "eval_steps_per_second": 9.874, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 3.4640684127807617, | |
| "learning_rate": 2e-05, | |
| "loss": 0.187, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.846679266293906, | |
| "eval_loss": 0.2870851755142212, | |
| "eval_precision": 0.8609191655801824, | |
| "eval_recall": 0.8356064739043463, | |
| "eval_runtime": 5.0487, | |
| "eval_samples_per_second": 79.031, | |
| "eval_steps_per_second": 9.904, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 2.2104856967926025, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1812, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8526315789473684, | |
| "eval_loss": 0.2812826931476593, | |
| "eval_precision": 0.8585304054054055, | |
| "eval_recall": 0.8473813420621932, | |
| "eval_runtime": 5.0484, | |
| "eval_samples_per_second": 79.034, | |
| "eval_steps_per_second": 9.904, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 3.044590950012207, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1633, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8596342841745197, | |
| "eval_loss": 0.2956988215446472, | |
| "eval_precision": 0.8555364857667042, | |
| "eval_recall": 0.8641571194762684, | |
| "eval_runtime": 5.0481, | |
| "eval_samples_per_second": 79.04, | |
| "eval_steps_per_second": 9.905, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.9169403314590454, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1607, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8696722245432793, | |
| "eval_loss": 0.28752827644348145, | |
| "eval_precision": 0.8706135006701596, | |
| "eval_recall": 0.8687488634297145, | |
| "eval_runtime": 5.0457, | |
| "eval_samples_per_second": 79.078, | |
| "eval_steps_per_second": 9.91, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.9245001077651978, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1584, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8560793854229822, | |
| "eval_loss": 0.2859440743923187, | |
| "eval_precision": 0.8609538327526132, | |
| "eval_recall": 0.8516548463356974, | |
| "eval_runtime": 5.086, | |
| "eval_samples_per_second": 78.451, | |
| "eval_steps_per_second": 9.831, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 3.737988233566284, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1535, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8609292598654301, | |
| "eval_loss": 0.29243767261505127, | |
| "eval_precision": 0.8609292598654301, | |
| "eval_recall": 0.8609292598654301, | |
| "eval_runtime": 5.0479, | |
| "eval_samples_per_second": 79.042, | |
| "eval_steps_per_second": 9.905, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.7625505924224854, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1432, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8616171059774413, | |
| "eval_loss": 0.29657039046287537, | |
| "eval_precision": 0.859873949579832, | |
| "eval_recall": 0.8634297144935443, | |
| "eval_runtime": 5.0616, | |
| "eval_samples_per_second": 78.829, | |
| "eval_steps_per_second": 9.878, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.5611210465431213, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1466, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8568221901555235, | |
| "eval_loss": 0.29467570781707764, | |
| "eval_precision": 0.8596491228070176, | |
| "eval_recall": 0.8541553009638116, | |
| "eval_runtime": 5.0458, | |
| "eval_samples_per_second": 79.075, | |
| "eval_steps_per_second": 9.909, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 7.14449405670166, | |
| "learning_rate": 0.0, | |
| "loss": 0.1411, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8609292598654301, | |
| "eval_loss": 0.2950780689716339, | |
| "eval_precision": 0.8609292598654301, | |
| "eval_recall": 0.8609292598654301, | |
| "eval_runtime": 5.0549, | |
| "eval_samples_per_second": 78.933, | |
| "eval_steps_per_second": 9.891, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8460375995160000.0, | |
| "train_loss": 0.2363556080177182, | |
| "train_runtime": 1909.167, | |
| "train_samples_per_second": 38.184, | |
| "train_steps_per_second": 1.278 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 8460375995160000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |