| { | |
| "best_metric": 0.8919400441736002, | |
| "best_model_checkpoint": "cls_comment-phobert-base-v2-v2.4/checkpoint-1500", | |
| "epoch": 19.138755980861244, | |
| "eval_steps": 100, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.4111480712890625, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.7008, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.46234623462346236, | |
| "eval_f1_score": 0.10914902653284753, | |
| "eval_loss": 1.5259455442428589, | |
| "eval_precision": 0.22956187895212285, | |
| "eval_recall": 0.16870144284128744, | |
| "eval_runtime": 5.8062, | |
| "eval_samples_per_second": 574.041, | |
| "eval_steps_per_second": 9.128, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 3.172757148742676, | |
| "learning_rate": 5e-06, | |
| "loss": 1.4089, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "eval_accuracy": 0.6567656765676567, | |
| "eval_f1_score": 0.24987557712552566, | |
| "eval_loss": 1.1875011920928955, | |
| "eval_precision": 0.216976056542887, | |
| "eval_recall": 0.2948041271015744, | |
| "eval_runtime": 6.0549, | |
| "eval_samples_per_second": 550.46, | |
| "eval_steps_per_second": 8.753, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 4.23789644241333, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 1.0776, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "eval_accuracy": 0.8037803780378038, | |
| "eval_f1_score": 0.5304195101025111, | |
| "eval_loss": 0.9009036421775818, | |
| "eval_precision": 0.5333440517242575, | |
| "eval_recall": 0.5308669916547205, | |
| "eval_runtime": 6.0211, | |
| "eval_samples_per_second": 553.553, | |
| "eval_steps_per_second": 8.802, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "grad_norm": 2.9036450386047363, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8625, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "eval_accuracy": 0.8574857485748575, | |
| "eval_f1_score": 0.6321452494594356, | |
| "eval_loss": 0.7617138624191284, | |
| "eval_precision": 0.7107045568063487, | |
| "eval_recall": 0.6372368264036259, | |
| "eval_runtime": 6.0254, | |
| "eval_samples_per_second": 553.154, | |
| "eval_steps_per_second": 8.796, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "grad_norm": 4.648315906524658, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.7245, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "eval_accuracy": 0.8817881788178817, | |
| "eval_f1_score": 0.7282115691268598, | |
| "eval_loss": 0.689365804195404, | |
| "eval_precision": 0.8802547295767839, | |
| "eval_recall": 0.7213609117839314, | |
| "eval_runtime": 6.0362, | |
| "eval_samples_per_second": 552.166, | |
| "eval_steps_per_second": 8.78, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "grad_norm": 5.482232093811035, | |
| "learning_rate": 9.444444444444445e-06, | |
| "loss": 0.6573, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "eval_accuracy": 0.8967896789678967, | |
| "eval_f1_score": 0.8406229534596467, | |
| "eval_loss": 0.6651212573051453, | |
| "eval_precision": 0.8770203982052687, | |
| "eval_recall": 0.8212623271516084, | |
| "eval_runtime": 6.089, | |
| "eval_samples_per_second": 547.376, | |
| "eval_steps_per_second": 8.704, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "grad_norm": 4.107127666473389, | |
| "learning_rate": 9.166666666666666e-06, | |
| "loss": 0.6082, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "eval_accuracy": 0.9078907890789079, | |
| "eval_f1_score": 0.8629840378859187, | |
| "eval_loss": 0.6335443258285522, | |
| "eval_precision": 0.8594850910270425, | |
| "eval_recall": 0.866740275628739, | |
| "eval_runtime": 6.012, | |
| "eval_samples_per_second": 554.389, | |
| "eval_steps_per_second": 8.816, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "grad_norm": 4.067190647125244, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 0.5674, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "eval_accuracy": 0.9105910591059105, | |
| "eval_f1_score": 0.8692050710637084, | |
| "eval_loss": 0.636346697807312, | |
| "eval_precision": 0.8621114850690574, | |
| "eval_recall": 0.8795216165515357, | |
| "eval_runtime": 6.0275, | |
| "eval_samples_per_second": 552.965, | |
| "eval_steps_per_second": 8.793, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "grad_norm": 3.8194425106048584, | |
| "learning_rate": 8.611111111111112e-06, | |
| "loss": 0.5477, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "eval_accuracy": 0.9150915091509151, | |
| "eval_f1_score": 0.8776204392865118, | |
| "eval_loss": 0.6268996596336365, | |
| "eval_precision": 0.8876632968569648, | |
| "eval_recall": 0.8693279668307071, | |
| "eval_runtime": 5.982, | |
| "eval_samples_per_second": 557.174, | |
| "eval_steps_per_second": 8.86, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "grad_norm": 4.406583309173584, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.5256, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "eval_accuracy": 0.9204920492049204, | |
| "eval_f1_score": 0.8834661339438045, | |
| "eval_loss": 0.6178370714187622, | |
| "eval_precision": 0.8826304228202296, | |
| "eval_recall": 0.8848781731721993, | |
| "eval_runtime": 6.0125, | |
| "eval_samples_per_second": 554.349, | |
| "eval_steps_per_second": 8.815, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "grad_norm": 3.69624400138855, | |
| "learning_rate": 8.055555555555557e-06, | |
| "loss": 0.5148, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "eval_accuracy": 0.9198919891989199, | |
| "eval_f1_score": 0.879559155366957, | |
| "eval_loss": 0.6214434504508972, | |
| "eval_precision": 0.8761821574730276, | |
| "eval_recall": 0.8838583582269935, | |
| "eval_runtime": 6.0361, | |
| "eval_samples_per_second": 552.174, | |
| "eval_steps_per_second": 8.78, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 11.48, | |
| "grad_norm": 4.3708343505859375, | |
| "learning_rate": 7.77777777777778e-06, | |
| "loss": 0.4999, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.48, | |
| "eval_accuracy": 0.9228922892289229, | |
| "eval_f1_score": 0.8856181382103513, | |
| "eval_loss": 0.6158125400543213, | |
| "eval_precision": 0.8862286567723959, | |
| "eval_recall": 0.8853146392819923, | |
| "eval_runtime": 5.9971, | |
| "eval_samples_per_second": 555.768, | |
| "eval_steps_per_second": 8.838, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "grad_norm": 5.073043346405029, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.4916, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "eval_accuracy": 0.9231923192319232, | |
| "eval_f1_score": 0.8839178584078228, | |
| "eval_loss": 0.6185785531997681, | |
| "eval_precision": 0.8887592838927109, | |
| "eval_recall": 0.8794527947878894, | |
| "eval_runtime": 6.0868, | |
| "eval_samples_per_second": 547.575, | |
| "eval_steps_per_second": 8.707, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "grad_norm": 3.3584494590759277, | |
| "learning_rate": 7.222222222222223e-06, | |
| "loss": 0.479, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "eval_accuracy": 0.9201920192019202, | |
| "eval_f1_score": 0.8847465693125924, | |
| "eval_loss": 0.6285073161125183, | |
| "eval_precision": 0.8863862882902076, | |
| "eval_recall": 0.8833493957517122, | |
| "eval_runtime": 6.0466, | |
| "eval_samples_per_second": 551.218, | |
| "eval_steps_per_second": 8.765, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 14.35, | |
| "grad_norm": 4.605256080627441, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.4812, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.35, | |
| "eval_accuracy": 0.9240924092409241, | |
| "eval_f1_score": 0.8919400441736002, | |
| "eval_loss": 0.6176608204841614, | |
| "eval_precision": 0.8911083019322857, | |
| "eval_recall": 0.8930303467825517, | |
| "eval_runtime": 6.0144, | |
| "eval_samples_per_second": 554.172, | |
| "eval_steps_per_second": 8.812, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "grad_norm": 5.669193744659424, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.4667, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "eval_accuracy": 0.9255925592559255, | |
| "eval_f1_score": 0.8847926962066609, | |
| "eval_loss": 0.6205594539642334, | |
| "eval_precision": 0.8843481979823253, | |
| "eval_recall": 0.8852982243831954, | |
| "eval_runtime": 6.1133, | |
| "eval_samples_per_second": 545.203, | |
| "eval_steps_per_second": 8.67, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 16.27, | |
| "grad_norm": 3.3217363357543945, | |
| "learning_rate": 6.3888888888888885e-06, | |
| "loss": 0.4668, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 16.27, | |
| "eval_accuracy": 0.9264926492649265, | |
| "eval_f1_score": 0.8854482169609333, | |
| "eval_loss": 0.6200627684593201, | |
| "eval_precision": 0.8837445827045557, | |
| "eval_recall": 0.8875863206208373, | |
| "eval_runtime": 6.0633, | |
| "eval_samples_per_second": 549.698, | |
| "eval_steps_per_second": 8.741, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 17.22, | |
| "grad_norm": 3.535698652267456, | |
| "learning_rate": 6.111111111111112e-06, | |
| "loss": 0.4635, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 17.22, | |
| "eval_accuracy": 0.9252925292529253, | |
| "eval_f1_score": 0.8901467035784716, | |
| "eval_loss": 0.6252107620239258, | |
| "eval_precision": 0.8926952965988483, | |
| "eval_recall": 0.8877172530901484, | |
| "eval_runtime": 6.0706, | |
| "eval_samples_per_second": 549.043, | |
| "eval_steps_per_second": 8.731, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "grad_norm": 3.0793726444244385, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 0.4593, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "eval_accuracy": 0.9273927392739274, | |
| "eval_f1_score": 0.88911901951679, | |
| "eval_loss": 0.626419723033905, | |
| "eval_precision": 0.8887473645082232, | |
| "eval_recall": 0.8899116044774821, | |
| "eval_runtime": 6.0336, | |
| "eval_samples_per_second": 552.41, | |
| "eval_steps_per_second": 8.784, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 19.14, | |
| "grad_norm": 0.5877706408500671, | |
| "learning_rate": 5.555555555555557e-06, | |
| "loss": 0.4538, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 19.14, | |
| "eval_accuracy": 0.9264926492649265, | |
| "eval_f1_score": 0.8890948519955958, | |
| "eval_loss": 0.6227503418922424, | |
| "eval_precision": 0.8870090750160463, | |
| "eval_recall": 0.8913327030956307, | |
| "eval_runtime": 6.0553, | |
| "eval_samples_per_second": 550.427, | |
| "eval_steps_per_second": 8.753, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 19.14, | |
| "step": 2000, | |
| "total_flos": 4678479916291584.0, | |
| "train_loss": 0.6728471851348877, | |
| "train_runtime": 1885.5017, | |
| "train_samples_per_second": 271.546, | |
| "train_steps_per_second": 2.121 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 39, | |
| "save_steps": 100, | |
| "total_flos": 4678479916291584.0, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |