| { | |
| "best_metric": 0.9651834647200833, | |
| "best_model_checkpoint": "./results/checkpoint-3000", | |
| "epoch": 0.9363295880149812, | |
| "eval_steps": 500, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0, | |
| "eval_accuracy": 0.6045787545787545, | |
| "eval_ap": 0.6438655017635215, | |
| "eval_auc": 0.6252253659085975, | |
| "eval_eer": 0.41503637833468066, | |
| "eval_f1": 0.673324254804055, | |
| "eval_loss": 0.6694703102111816, | |
| "eval_precision": 0.6141319348606128, | |
| "eval_recall": 0.7451440053583389, | |
| "eval_runtime": 1161.2986, | |
| "eval_samples_per_second": 70.524, | |
| "eval_steps_per_second": 0.276, | |
| "step": 0 | |
| }, | |
| { | |
| "epoch": 0.031210986267166042, | |
| "grad_norm": 1570057.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3999, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.062421972534332085, | |
| "grad_norm": 1873799.375, | |
| "learning_rate": 2e-05, | |
| "loss": 0.183, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.09363295880149813, | |
| "grad_norm": 937146.625, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1499, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.12484394506866417, | |
| "grad_norm": 1296423.5, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1939, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1560549313358302, | |
| "grad_norm": 698965.0, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1542, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1560549313358302, | |
| "eval_accuracy": 0.7873870573870574, | |
| "eval_ap": 0.9824733762472688, | |
| "eval_auc": 0.9823328646050201, | |
| "eval_eer": 0.06329830234438157, | |
| "eval_f1": 0.7611024983193624, | |
| "eval_loss": 0.586156964302063, | |
| "eval_precision": 0.9871525677070359, | |
| "eval_recall": 0.6192900200937709, | |
| "eval_runtime": 896.8541, | |
| "eval_samples_per_second": 91.319, | |
| "eval_steps_per_second": 0.357, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18726591760299627, | |
| "grad_norm": 854220.375, | |
| "learning_rate": 4.81508875739645e-05, | |
| "loss": 0.1201, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.2184769038701623, | |
| "grad_norm": 1128798.0, | |
| "learning_rate": 4.6301775147928994e-05, | |
| "loss": 0.1162, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.24968789013732834, | |
| "grad_norm": 622432.0, | |
| "learning_rate": 4.44526627218935e-05, | |
| "loss": 0.1071, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.2808988764044944, | |
| "grad_norm": 790906.0625, | |
| "learning_rate": 4.260355029585799e-05, | |
| "loss": 0.0874, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.3121098626716604, | |
| "grad_norm": 1022445.375, | |
| "learning_rate": 4.075443786982249e-05, | |
| "loss": 0.0748, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3121098626716604, | |
| "eval_accuracy": 0.9135164835164835, | |
| "eval_ap": 0.9891846925529121, | |
| "eval_auc": 0.9885782446290119, | |
| "eval_eer": 0.05095661546752897, | |
| "eval_f1": 0.9160672599509415, | |
| "eval_loss": 0.23737847805023193, | |
| "eval_precision": 0.9761105078411071, | |
| "eval_recall": 0.8629828086626479, | |
| "eval_runtime": 899.3174, | |
| "eval_samples_per_second": 91.069, | |
| "eval_steps_per_second": 0.356, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3433208489388265, | |
| "grad_norm": 857955.5625, | |
| "learning_rate": 3.8905325443786985e-05, | |
| "loss": 0.0764, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.37453183520599254, | |
| "grad_norm": 874027.0, | |
| "learning_rate": 3.705621301775148e-05, | |
| "loss": 0.0709, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.40574282147315854, | |
| "grad_norm": 582799.875, | |
| "learning_rate": 3.520710059171598e-05, | |
| "loss": 0.0636, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.4369538077403246, | |
| "grad_norm": 521187.625, | |
| "learning_rate": 3.335798816568047e-05, | |
| "loss": 0.0533, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.4681647940074906, | |
| "grad_norm": 522330.4375, | |
| "learning_rate": 3.150887573964497e-05, | |
| "loss": 0.0465, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4681647940074906, | |
| "eval_accuracy": 0.9357875457875457, | |
| "eval_ap": 0.9924125253776173, | |
| "eval_auc": 0.9918219498411971, | |
| "eval_eer": 0.042576125033683646, | |
| "eval_f1": 0.9388722931898225, | |
| "eval_loss": 0.18004347383975983, | |
| "eval_precision": 0.9792449627815629, | |
| "eval_recall": 0.9016968073230632, | |
| "eval_runtime": 904.8241, | |
| "eval_samples_per_second": 90.515, | |
| "eval_steps_per_second": 0.354, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4993757802746567, | |
| "grad_norm": 1439895.375, | |
| "learning_rate": 2.965976331360947e-05, | |
| "loss": 0.0479, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.5305867665418227, | |
| "grad_norm": 522678.71875, | |
| "learning_rate": 2.7810650887573965e-05, | |
| "loss": 0.0429, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.5617977528089888, | |
| "grad_norm": 1075267.0, | |
| "learning_rate": 2.5961538461538464e-05, | |
| "loss": 0.0419, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.5930087390761548, | |
| "grad_norm": 621287.5625, | |
| "learning_rate": 2.411242603550296e-05, | |
| "loss": 0.0449, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.6242197253433208, | |
| "grad_norm": 431768.9375, | |
| "learning_rate": 2.2263313609467456e-05, | |
| "loss": 0.0373, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6242197253433208, | |
| "eval_accuracy": 0.9467277167277167, | |
| "eval_ap": 0.9953183185775565, | |
| "eval_auc": 0.9948050755617595, | |
| "eval_eer": 0.03290218270008084, | |
| "eval_f1": 0.9495787636800684, | |
| "eval_loss": 0.16710303723812103, | |
| "eval_precision": 0.9842600800172492, | |
| "eval_recall": 0.9172583165885242, | |
| "eval_runtime": 883.8932, | |
| "eval_samples_per_second": 92.658, | |
| "eval_steps_per_second": 0.362, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6554307116104869, | |
| "grad_norm": 300561.40625, | |
| "learning_rate": 2.0414201183431952e-05, | |
| "loss": 0.0367, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.686641697877653, | |
| "grad_norm": 759234.625, | |
| "learning_rate": 1.856508875739645e-05, | |
| "loss": 0.0385, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.717852684144819, | |
| "grad_norm": 110611.8984375, | |
| "learning_rate": 1.6715976331360947e-05, | |
| "loss": 0.029, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.7490636704119851, | |
| "grad_norm": 890414.25, | |
| "learning_rate": 1.4866863905325445e-05, | |
| "loss": 0.0243, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.7802746566791511, | |
| "grad_norm": 110954.5390625, | |
| "learning_rate": 1.3017751479289941e-05, | |
| "loss": 0.028, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.7802746566791511, | |
| "eval_accuracy": 0.9541514041514042, | |
| "eval_ap": 0.9963480520212278, | |
| "eval_auc": 0.9958883442351322, | |
| "eval_eer": 0.02915656157369981, | |
| "eval_f1": 0.9567799633982114, | |
| "eval_loss": 0.1855895221233368, | |
| "eval_precision": 0.9874557506355278, | |
| "eval_recall": 0.9279526680062514, | |
| "eval_runtime": 916.7443, | |
| "eval_samples_per_second": 89.338, | |
| "eval_steps_per_second": 0.349, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.8114856429463171, | |
| "grad_norm": 969664.1875, | |
| "learning_rate": 1.1168639053254439e-05, | |
| "loss": 0.0268, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.8426966292134831, | |
| "grad_norm": 21713.740234375, | |
| "learning_rate": 9.319526627218936e-06, | |
| "loss": 0.0259, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.8739076154806492, | |
| "grad_norm": 625.3981323242188, | |
| "learning_rate": 7.470414201183433e-06, | |
| "loss": 0.0272, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.9051186017478152, | |
| "grad_norm": 51626.33203125, | |
| "learning_rate": 5.621301775147929e-06, | |
| "loss": 0.0255, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.9363295880149812, | |
| "grad_norm": 330527.75, | |
| "learning_rate": 3.772189349112426e-06, | |
| "loss": 0.0196, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9363295880149812, | |
| "eval_accuracy": 0.9628449328449329, | |
| "eval_ap": 0.9971936987222482, | |
| "eval_auc": 0.996572647864952, | |
| "eval_eer": 0.02357855025599569, | |
| "eval_f1": 0.9651834647200833, | |
| "eval_loss": 0.2491898238658905, | |
| "eval_precision": 0.9898617727816761, | |
| "eval_recall": 0.9417057378879214, | |
| "eval_runtime": 887.8512, | |
| "eval_samples_per_second": 92.245, | |
| "eval_steps_per_second": 0.36, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3204, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.6506508689408e+18, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |