| { | |
| "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": 4.3501811027526855, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5459, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7468671679197995, | |
| "eval_f1": 0.6572777139941319, | |
| "eval_loss": 0.46394461393356323, | |
| "eval_precision": 0.6922477833692786, | |
| "eval_recall": 0.6458901618476087, | |
| "eval_runtime": 5.2607, | |
| "eval_samples_per_second": 75.846, | |
| "eval_steps_per_second": 9.504, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.848093271255493, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4335, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7844611528822055, | |
| "eval_f1": 0.7634169884169884, | |
| "eval_loss": 0.41080254316329956, | |
| "eval_precision": 0.7551731309140064, | |
| "eval_recall": 0.7975086379341698, | |
| "eval_runtime": 5.5475, | |
| "eval_samples_per_second": 71.924, | |
| "eval_steps_per_second": 9.013, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.800747394561768, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3375, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8271551457392166, | |
| "eval_loss": 0.3282613754272461, | |
| "eval_precision": 0.8347358430876305, | |
| "eval_recall": 0.8206946717585015, | |
| "eval_runtime": 5.6194, | |
| "eval_samples_per_second": 71.004, | |
| "eval_steps_per_second": 8.898, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.052788019180298, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2801, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8347043853938569, | |
| "eval_loss": 0.32024893164634705, | |
| "eval_precision": 0.8277993283927745, | |
| "eval_recall": 0.8431987634115294, | |
| "eval_runtime": 5.5699, | |
| "eval_samples_per_second": 71.636, | |
| "eval_steps_per_second": 8.977, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.35927698016166687, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2572, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8550061050061051, | |
| "eval_loss": 0.31085968017578125, | |
| "eval_precision": 0.8437710437710437, | |
| "eval_recall": 0.8713402436806692, | |
| "eval_runtime": 5.607, | |
| "eval_samples_per_second": 71.161, | |
| "eval_steps_per_second": 8.917, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.9262036085128784, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2339, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8472754847795472, | |
| "eval_loss": 0.3074397146701813, | |
| "eval_precision": 0.8352906879893387, | |
| "eval_recall": 0.8660210947444991, | |
| "eval_runtime": 5.5884, | |
| "eval_samples_per_second": 71.398, | |
| "eval_steps_per_second": 8.947, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.7879945635795593, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2249, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8472754847795472, | |
| "eval_loss": 0.2915370464324951, | |
| "eval_precision": 0.8352906879893387, | |
| "eval_recall": 0.8660210947444991, | |
| "eval_runtime": 5.5808, | |
| "eval_samples_per_second": 71.495, | |
| "eval_steps_per_second": 8.959, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.2202210426330566, | |
| "learning_rate": 3e-05, | |
| "loss": 0.193, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8751002084335417, | |
| "eval_loss": 0.25404733419418335, | |
| "eval_precision": 0.8780701754385964, | |
| "eval_recall": 0.8722949627204946, | |
| "eval_runtime": 5.6126, | |
| "eval_samples_per_second": 71.09, | |
| "eval_steps_per_second": 8.909, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 8.847825050354004, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.1899, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8628173897776901, | |
| "eval_loss": 0.26359453797340393, | |
| "eval_precision": 0.8526132107592781, | |
| "eval_recall": 0.8766593926168395, | |
| "eval_runtime": 5.6712, | |
| "eval_samples_per_second": 70.355, | |
| "eval_steps_per_second": 8.816, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.7180150747299194, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1801, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9072681704260651, | |
| "eval_f1": 0.8900228699985846, | |
| "eval_loss": 0.23711469769477844, | |
| "eval_precision": 0.8840175953079179, | |
| "eval_recall": 0.8968903436988543, | |
| "eval_runtime": 5.6014, | |
| "eval_samples_per_second": 71.232, | |
| "eval_steps_per_second": 8.926, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.6990593671798706, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.157, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.899749373433584, | |
| "eval_f1": 0.882467302933899, | |
| "eval_loss": 0.2566535174846649, | |
| "eval_precision": 0.8732988802756245, | |
| "eval_recall": 0.8940716493907983, | |
| "eval_runtime": 5.5662, | |
| "eval_samples_per_second": 71.683, | |
| "eval_steps_per_second": 8.983, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 3.7353861331939697, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1553, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8792560061999484, | |
| "eval_loss": 0.25933051109313965, | |
| "eval_precision": 0.8707622232472325, | |
| "eval_recall": 0.889798145117294, | |
| "eval_runtime": 5.5701, | |
| "eval_samples_per_second": 71.632, | |
| "eval_steps_per_second": 8.976, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.2111871242523193, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1381, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9172932330827067, | |
| "eval_f1": 0.8999810095332144, | |
| "eval_loss": 0.24901245534420013, | |
| "eval_precision": 0.9009991470695747, | |
| "eval_recall": 0.8989816330241862, | |
| "eval_runtime": 5.6002, | |
| "eval_samples_per_second": 71.248, | |
| "eval_steps_per_second": 8.928, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 9.943241119384766, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1476, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.899749373433584, | |
| "eval_f1": 0.881931703852755, | |
| "eval_loss": 0.27014100551605225, | |
| "eval_precision": 0.8740012737378415, | |
| "eval_recall": 0.8915711947626841, | |
| "eval_runtime": 5.5922, | |
| "eval_samples_per_second": 71.349, | |
| "eval_steps_per_second": 8.941, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.20924903452396393, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1447, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9122807017543859, | |
| "eval_f1": 0.8959675797283908, | |
| "eval_loss": 0.2611282765865326, | |
| "eval_precision": 0.8898680351906159, | |
| "eval_recall": 0.9029368976177486, | |
| "eval_runtime": 5.5841, | |
| "eval_samples_per_second": 71.453, | |
| "eval_steps_per_second": 8.954, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.40175938606262207, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1336, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.899749373433584, | |
| "eval_f1": 0.884004884004884, | |
| "eval_loss": 0.31003570556640625, | |
| "eval_precision": 0.8717592592592593, | |
| "eval_recall": 0.9015730132751409, | |
| "eval_runtime": 5.5685, | |
| "eval_samples_per_second": 71.653, | |
| "eval_steps_per_second": 8.979, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 7.287301540374756, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1192, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8803300634230913, | |
| "eval_loss": 0.29346275329589844, | |
| "eval_precision": 0.8695948246510044, | |
| "eval_recall": 0.8947990543735225, | |
| "eval_runtime": 5.5988, | |
| "eval_samples_per_second": 71.266, | |
| "eval_steps_per_second": 8.931, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 6.602230072021484, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1247, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9022556390977443, | |
| "eval_f1": 0.885145957117024, | |
| "eval_loss": 0.28685662150382996, | |
| "eval_precision": 0.876513491697417, | |
| "eval_recall": 0.8958446990361884, | |
| "eval_runtime": 5.5796, | |
| "eval_samples_per_second": 71.51, | |
| "eval_steps_per_second": 8.961, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 7.744582653045654, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.117, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9022556390977443, | |
| "eval_f1": 0.884617951284618, | |
| "eval_loss": 0.2761416733264923, | |
| "eval_precision": 0.8772893772893773, | |
| "eval_recall": 0.8933442444080741, | |
| "eval_runtime": 5.559, | |
| "eval_samples_per_second": 71.775, | |
| "eval_steps_per_second": 8.994, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.1699504405260086, | |
| "learning_rate": 0.0, | |
| "loss": 0.1092, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9022556390977443, | |
| "eval_f1": 0.884617951284618, | |
| "eval_loss": 0.2810901701450348, | |
| "eval_precision": 0.8772893772893773, | |
| "eval_recall": 0.8933442444080741, | |
| "eval_runtime": 5.6476, | |
| "eval_samples_per_second": 70.649, | |
| "eval_steps_per_second": 8.853, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8551203605328000.0, | |
| "train_loss": 0.21111928674041247, | |
| "train_runtime": 2276.8554, | |
| "train_samples_per_second": 31.956, | |
| "train_steps_per_second": 1.072 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 8551203605328000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |