| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 14.999789127409219, | |
| "global_step": 59265, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 6.994913523145705e-06, | |
| "loss": 0.3404, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_accuracy": 0.887192844025166, | |
| "eval_f1": 0.886488607488543, | |
| "eval_loss": 0.4329048991203308, | |
| "eval_runtime": 42.8766, | |
| "eval_samples_per_second": 3340.027, | |
| "eval_steps_per_second": 69.595, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.954310370186513e-06, | |
| "loss": 0.3433, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_accuracy": 0.8883170750441662, | |
| "eval_f1": 0.8876090017723502, | |
| "eval_loss": 0.4280129075050354, | |
| "eval_runtime": 46.562, | |
| "eval_samples_per_second": 3075.664, | |
| "eval_steps_per_second": 64.087, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 6.873575782958537e-06, | |
| "loss": 0.3281, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_accuracy": 0.8889804411733899, | |
| "eval_f1": 0.8882523897529724, | |
| "eval_loss": 0.43023422360420227, | |
| "eval_runtime": 45.3226, | |
| "eval_samples_per_second": 3159.767, | |
| "eval_steps_per_second": 65.839, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 6.753647718516397e-06, | |
| "loss": 0.331, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_accuracy": 0.8891340628033154, | |
| "eval_f1": 0.8885291871869458, | |
| "eval_loss": 0.42653217911720276, | |
| "eval_runtime": 54.726, | |
| "eval_samples_per_second": 2616.839, | |
| "eval_steps_per_second": 54.526, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 6.5959194752957715e-06, | |
| "loss": 0.3224, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_accuracy": 0.8881355222087998, | |
| "eval_f1": 0.8873919365623191, | |
| "eval_loss": 0.43002641201019287, | |
| "eval_runtime": 54.91, | |
| "eval_samples_per_second": 2608.066, | |
| "eval_steps_per_second": 54.343, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 6.402223506072121e-06, | |
| "loss": 0.3361, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "eval_accuracy": 0.8889175959611477, | |
| "eval_f1": 0.8882468792028413, | |
| "eval_loss": 0.42911502718925476, | |
| "eval_runtime": 50.9856, | |
| "eval_samples_per_second": 2808.81, | |
| "eval_steps_per_second": 58.526, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 6.1748101289188055e-06, | |
| "loss": 0.3323, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "eval_accuracy": 0.8877584509353462, | |
| "eval_f1": 0.8871109456457335, | |
| "eval_loss": 0.4336757957935333, | |
| "eval_runtime": 51.7236, | |
| "eval_samples_per_second": 2768.739, | |
| "eval_steps_per_second": 57.691, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 5.916321383496139e-06, | |
| "loss": 0.3556, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "eval_accuracy": 0.8856845589313521, | |
| "eval_f1": 0.8851065217265167, | |
| "eval_loss": 0.4344990849494934, | |
| "eval_runtime": 50.059, | |
| "eval_samples_per_second": 2860.803, | |
| "eval_steps_per_second": 59.61, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 5.629760336403372e-06, | |
| "loss": 0.3663, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "eval_accuracy": 0.8836455809341591, | |
| "eval_f1": 0.8828440694403326, | |
| "eval_loss": 0.44171223044395447, | |
| "eval_runtime": 52.3029, | |
| "eval_samples_per_second": 2738.072, | |
| "eval_steps_per_second": 57.052, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 5.318456192197413e-06, | |
| "loss": 0.3902, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "eval_accuracy": 0.8789252072146304, | |
| "eval_f1": 0.8780871328644805, | |
| "eval_loss": 0.4555477797985077, | |
| "eval_runtime": 50.7366, | |
| "eval_samples_per_second": 2822.599, | |
| "eval_steps_per_second": 58.814, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 4.986025615410962e-06, | |
| "loss": 0.4036, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "eval_accuracy": 0.8788483963996676, | |
| "eval_f1": 0.877917228234661, | |
| "eval_loss": 0.4555710554122925, | |
| "eval_runtime": 49.8006, | |
| "eval_samples_per_second": 2875.647, | |
| "eval_steps_per_second": 59.919, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 4.636330712922542e-06, | |
| "loss": 0.4305, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "eval_accuracy": 0.8750637180624122, | |
| "eval_f1": 0.8741648169470942, | |
| "eval_loss": 0.46971216797828674, | |
| "eval_runtime": 50.8439, | |
| "eval_samples_per_second": 2816.642, | |
| "eval_steps_per_second": 58.689, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 4.273434164830173e-06, | |
| "loss": 0.4501, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "eval_accuracy": 0.8737788826121263, | |
| "eval_f1": 0.8725169321502271, | |
| "eval_loss": 0.4762924611568451, | |
| "eval_runtime": 50.5449, | |
| "eval_samples_per_second": 2833.301, | |
| "eval_steps_per_second": 59.037, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 3.901552025108571e-06, | |
| "loss": 0.4733, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "eval_accuracy": 0.8710486072802687, | |
| "eval_f1": 0.8700483733487925, | |
| "eval_loss": 0.48572415113449097, | |
| "eval_runtime": 51.8675, | |
| "eval_samples_per_second": 2761.054, | |
| "eval_steps_per_second": 57.531, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 3.5250047404016715e-06, | |
| "loss": 0.4851, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "eval_accuracy": 0.8705109315755295, | |
| "eval_f1": 0.8695447653344123, | |
| "eval_loss": 0.4862508177757263, | |
| "eval_runtime": 50.7033, | |
| "eval_samples_per_second": 2824.454, | |
| "eval_steps_per_second": 58.852, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 3.1481669560035742e-06, | |
| "loss": 0.4846, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "eval_accuracy": 0.8708251576367407, | |
| "eval_f1": 0.869757227158951, | |
| "eval_loss": 0.48489654064178467, | |
| "eval_runtime": 50.2252, | |
| "eval_samples_per_second": 2851.338, | |
| "eval_steps_per_second": 59.412, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 2.775416692171258e-06, | |
| "loss": 0.4856, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "eval_accuracy": 0.870664553205455, | |
| "eval_f1": 0.8694665590028052, | |
| "eval_loss": 0.4835050404071808, | |
| "eval_runtime": 50.112, | |
| "eval_samples_per_second": 2857.776, | |
| "eval_steps_per_second": 59.547, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "learning_rate": 2.411084481227692e-06, | |
| "loss": 0.4774, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "eval_accuracy": 0.8719214574502999, | |
| "eval_f1": 0.870818507712314, | |
| "eval_loss": 0.47967976331710815, | |
| "eval_runtime": 51.643, | |
| "eval_samples_per_second": 2773.058, | |
| "eval_steps_per_second": 57.781, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "learning_rate": 2.059403056369544e-06, | |
| "loss": 0.4635, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "eval_accuracy": 0.8728222388257721, | |
| "eval_f1": 0.8716593868274884, | |
| "eval_loss": 0.47762736678123474, | |
| "eval_runtime": 49.924, | |
| "eval_samples_per_second": 2868.539, | |
| "eval_steps_per_second": 59.771, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "learning_rate": 1.7244581766840846e-06, | |
| "loss": 0.4561, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "eval_accuracy": 0.8738906074338904, | |
| "eval_f1": 0.8728714900247343, | |
| "eval_loss": 0.4745886027812958, | |
| "eval_runtime": 52.1859, | |
| "eval_samples_per_second": 2744.206, | |
| "eval_steps_per_second": 57.18, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 10.63, | |
| "learning_rate": 1.4101411596796607e-06, | |
| "loss": 0.4475, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 10.63, | |
| "eval_accuracy": 0.8749380276379278, | |
| "eval_f1": 0.8739688992879336, | |
| "eval_loss": 0.4705161154270172, | |
| "eval_runtime": 52.4489, | |
| "eval_samples_per_second": 2730.449, | |
| "eval_steps_per_second": 56.893, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 11.14, | |
| "learning_rate": 1.1201036727965609e-06, | |
| "loss": 0.4413, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 11.14, | |
| "eval_accuracy": 0.8754268237331453, | |
| "eval_f1": 0.8743869003587098, | |
| "eval_loss": 0.46908074617385864, | |
| "eval_runtime": 50.2149, | |
| "eval_samples_per_second": 2851.921, | |
| "eval_steps_per_second": 59.425, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 11.64, | |
| "learning_rate": 8.577153091208042e-07, | |
| "loss": 0.4389, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 11.64, | |
| "eval_accuracy": 0.8759644994378845, | |
| "eval_f1": 0.875008769768229, | |
| "eval_loss": 0.46789219975471497, | |
| "eval_runtime": 50.3984, | |
| "eval_samples_per_second": 2841.539, | |
| "eval_steps_per_second": 59.208, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 6.260244401770287e-07, | |
| "loss": 0.4361, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "eval_accuracy": 0.875894671424282, | |
| "eval_f1": 0.8749081220424403, | |
| "eval_loss": 0.46769237518310547, | |
| "eval_runtime": 50.042, | |
| "eval_samples_per_second": 2861.778, | |
| "eval_steps_per_second": 59.63, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 12.65, | |
| "learning_rate": 4.2772280060435426e-07, | |
| "loss": 0.4362, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 12.65, | |
| "eval_accuracy": 0.876257777095015, | |
| "eval_f1": 0.875265867499827, | |
| "eval_loss": 0.4671792984008789, | |
| "eval_runtime": 52.5275, | |
| "eval_samples_per_second": 2726.361, | |
| "eval_steps_per_second": 56.808, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "learning_rate": 2.6511421616278837e-07, | |
| "loss": 0.4309, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "eval_accuracy": 0.87611812106781, | |
| "eval_f1": 0.8751424993393537, | |
| "eval_loss": 0.4671061038970947, | |
| "eval_runtime": 51.1386, | |
| "eval_samples_per_second": 2800.408, | |
| "eval_steps_per_second": 58.351, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 13.67, | |
| "learning_rate": 1.400878383814987e-07, | |
| "loss": 0.4316, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 13.67, | |
| "eval_accuracy": 0.8764183815263007, | |
| "eval_f1": 0.8754265388560385, | |
| "eval_loss": 0.4670025110244751, | |
| "eval_runtime": 51.6718, | |
| "eval_samples_per_second": 2771.511, | |
| "eval_steps_per_second": 57.749, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 14.17, | |
| "learning_rate": 5.409619680302491e-08, | |
| "loss": 0.4321, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 14.17, | |
| "eval_accuracy": 0.8764393299303814, | |
| "eval_f1": 0.8754605862048482, | |
| "eval_loss": 0.4668178856372833, | |
| "eval_runtime": 50.8951, | |
| "eval_samples_per_second": 2813.807, | |
| "eval_steps_per_second": 58.63, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "learning_rate": 8.138323807738212e-09, | |
| "loss": 0.4311, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "eval_accuracy": 0.8763904503208597, | |
| "eval_f1": 0.8754165344722714, | |
| "eval_loss": 0.46675533056259155, | |
| "eval_runtime": 50.6992, | |
| "eval_samples_per_second": 2824.68, | |
| "eval_steps_per_second": 58.857, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 59265, | |
| "total_flos": 3.822729571998278e+17, | |
| "train_loss": 0.4135634679527994, | |
| "train_runtime": 19096.9194, | |
| "train_samples_per_second": 893.953, | |
| "train_steps_per_second": 3.103 | |
| } | |
| ], | |
| "max_steps": 59265, | |
| "num_train_epochs": 15, | |
| "total_flos": 3.822729571998278e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |