| { | |
| "best_metric": 0.8875182540327614, | |
| "best_model_checkpoint": "cls_comment-phobert-base-v2-v2.2/checkpoint-2100", | |
| "epoch": 27.36842105263158, | |
| "eval_steps": 100, | |
| "global_step": 2600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 1.7893420457839966, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.663, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_accuracy": 0.5041240514681623, | |
| "eval_f1_score": 0.11172040652189807, | |
| "eval_loss": 1.4870415925979614, | |
| "eval_precision": 0.08402067524469371, | |
| "eval_recall": 0.16666666666666666, | |
| "eval_runtime": 5.0768, | |
| "eval_samples_per_second": 597.028, | |
| "eval_steps_per_second": 9.455, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 2.4307305812835693, | |
| "learning_rate": 5e-06, | |
| "loss": 1.294, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_accuracy": 0.6974595842956121, | |
| "eval_f1_score": 0.39113317224801986, | |
| "eval_loss": 0.9955642223358154, | |
| "eval_precision": 0.49018024013865347, | |
| "eval_recall": 0.3899919974680958, | |
| "eval_runtime": 5.1701, | |
| "eval_samples_per_second": 586.259, | |
| "eval_steps_per_second": 9.284, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 3.8080785274505615, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.898, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "eval_accuracy": 0.8231606730451996, | |
| "eval_f1_score": 0.5499484227307703, | |
| "eval_loss": 0.6778652667999268, | |
| "eval_precision": 0.535367810981718, | |
| "eval_recall": 0.5697240354177526, | |
| "eval_runtime": 5.2797, | |
| "eval_samples_per_second": 574.084, | |
| "eval_steps_per_second": 9.091, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "grad_norm": 4.932918548583984, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6411, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "eval_accuracy": 0.8568129330254042, | |
| "eval_f1_score": 0.5739966477156212, | |
| "eval_loss": 0.5163660645484924, | |
| "eval_precision": 0.5613296135162277, | |
| "eval_recall": 0.5895128268802295, | |
| "eval_runtime": 5.1719, | |
| "eval_samples_per_second": 586.047, | |
| "eval_steps_per_second": 9.281, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "grad_norm": 3.761213541030884, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.5031, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "eval_accuracy": 0.8937644341801386, | |
| "eval_f1_score": 0.7180655496783576, | |
| "eval_loss": 0.4105552136898041, | |
| "eval_precision": 0.7319397223704595, | |
| "eval_recall": 0.7114016304859992, | |
| "eval_runtime": 5.1958, | |
| "eval_samples_per_second": 583.36, | |
| "eval_steps_per_second": 9.238, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "grad_norm": 5.329339504241943, | |
| "learning_rate": 9.444444444444445e-06, | |
| "loss": 0.38, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "eval_accuracy": 0.9096007918178819, | |
| "eval_f1_score": 0.8325922485904789, | |
| "eval_loss": 0.3474419116973877, | |
| "eval_precision": 0.8738654107630431, | |
| "eval_recall": 0.8144519804951473, | |
| "eval_runtime": 5.1729, | |
| "eval_samples_per_second": 585.934, | |
| "eval_steps_per_second": 9.279, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "grad_norm": 5.044618129730225, | |
| "learning_rate": 9.166666666666666e-06, | |
| "loss": 0.2927, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "eval_accuracy": 0.9142197294622237, | |
| "eval_f1_score": 0.8598244657605173, | |
| "eval_loss": 0.3109701871871948, | |
| "eval_precision": 0.8809676600022426, | |
| "eval_recall": 0.8455076124995076, | |
| "eval_runtime": 6.0257, | |
| "eval_samples_per_second": 503.009, | |
| "eval_steps_per_second": 7.966, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "grad_norm": 4.748721599578857, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 0.2532, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "eval_accuracy": 0.9188386671065655, | |
| "eval_f1_score": 0.8702202017241888, | |
| "eval_loss": 0.3045769929885864, | |
| "eval_precision": 0.8880655151226781, | |
| "eval_recall": 0.85512094950349, | |
| "eval_runtime": 5.1742, | |
| "eval_samples_per_second": 585.795, | |
| "eval_steps_per_second": 9.277, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "grad_norm": 4.976996898651123, | |
| "learning_rate": 8.611111111111112e-06, | |
| "loss": 0.2049, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "eval_accuracy": 0.9218079841636424, | |
| "eval_f1_score": 0.8688677581871923, | |
| "eval_loss": 0.28507497906684875, | |
| "eval_precision": 0.890238230846033, | |
| "eval_recall": 0.8539333664213857, | |
| "eval_runtime": 5.1661, | |
| "eval_samples_per_second": 586.706, | |
| "eval_steps_per_second": 9.291, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "grad_norm": 5.230645656585693, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.1785, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "eval_accuracy": 0.9251072253381722, | |
| "eval_f1_score": 0.8768747602558463, | |
| "eval_loss": 0.2801537811756134, | |
| "eval_precision": 0.9044956542270928, | |
| "eval_recall": 0.8561455297131971, | |
| "eval_runtime": 5.2425, | |
| "eval_samples_per_second": 578.155, | |
| "eval_steps_per_second": 9.156, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "grad_norm": 5.533747673034668, | |
| "learning_rate": 8.055555555555557e-06, | |
| "loss": 0.1511, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "eval_accuracy": 0.9231276806334543, | |
| "eval_f1_score": 0.8744252567641876, | |
| "eval_loss": 0.28748995065689087, | |
| "eval_precision": 0.8770259775941726, | |
| "eval_recall": 0.8748474059729406, | |
| "eval_runtime": 5.2089, | |
| "eval_samples_per_second": 581.891, | |
| "eval_steps_per_second": 9.215, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 12.63, | |
| "grad_norm": 5.897303104400635, | |
| "learning_rate": 7.77777777777778e-06, | |
| "loss": 0.1392, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 12.63, | |
| "eval_accuracy": 0.9264269218079841, | |
| "eval_f1_score": 0.8774620123424183, | |
| "eval_loss": 0.2811359465122223, | |
| "eval_precision": 0.9005448213619175, | |
| "eval_recall": 0.8596724578451488, | |
| "eval_runtime": 5.1974, | |
| "eval_samples_per_second": 583.175, | |
| "eval_steps_per_second": 9.235, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "grad_norm": 3.5150811672210693, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.1166, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "eval_accuracy": 0.9247773012207192, | |
| "eval_f1_score": 0.875133246925271, | |
| "eval_loss": 0.27574026584625244, | |
| "eval_precision": 0.8785572499470087, | |
| "eval_recall": 0.8746042908646392, | |
| "eval_runtime": 5.198, | |
| "eval_samples_per_second": 583.104, | |
| "eval_steps_per_second": 9.234, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "grad_norm": 5.184170722961426, | |
| "learning_rate": 7.222222222222223e-06, | |
| "loss": 0.1087, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "eval_accuracy": 0.9257670735730782, | |
| "eval_f1_score": 0.8803669741747352, | |
| "eval_loss": 0.272718608379364, | |
| "eval_precision": 0.8857591375966419, | |
| "eval_recall": 0.8761427115927698, | |
| "eval_runtime": 5.2326, | |
| "eval_samples_per_second": 579.256, | |
| "eval_steps_per_second": 9.173, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "grad_norm": 2.6620497703552246, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.0918, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "eval_accuracy": 0.9284064665127021, | |
| "eval_f1_score": 0.8829680164086238, | |
| "eval_loss": 0.286248117685318, | |
| "eval_precision": 0.8987862948246464, | |
| "eval_recall": 0.8711650603773539, | |
| "eval_runtime": 5.184, | |
| "eval_samples_per_second": 584.68, | |
| "eval_steps_per_second": 9.259, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "grad_norm": 1.70383882522583, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.0824, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "eval_accuracy": 0.929066314747608, | |
| "eval_f1_score": 0.8832576833606276, | |
| "eval_loss": 0.2915453612804413, | |
| "eval_precision": 0.9008601442216951, | |
| "eval_recall": 0.868938029158805, | |
| "eval_runtime": 5.1614, | |
| "eval_samples_per_second": 587.245, | |
| "eval_steps_per_second": 9.3, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "grad_norm": 7.806049823760986, | |
| "learning_rate": 6.3888888888888885e-06, | |
| "loss": 0.0745, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "eval_accuracy": 0.929066314747608, | |
| "eval_f1_score": 0.8796971687211418, | |
| "eval_loss": 0.2994127869606018, | |
| "eval_precision": 0.8847125167911893, | |
| "eval_recall": 0.8796360078693516, | |
| "eval_runtime": 5.1775, | |
| "eval_samples_per_second": 585.421, | |
| "eval_steps_per_second": 9.271, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 18.95, | |
| "grad_norm": 9.288145065307617, | |
| "learning_rate": 6.111111111111112e-06, | |
| "loss": 0.0743, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 18.95, | |
| "eval_accuracy": 0.9254371494556252, | |
| "eval_f1_score": 0.8782515232850638, | |
| "eval_loss": 0.3092294931411743, | |
| "eval_precision": 0.8909710349230974, | |
| "eval_recall": 0.8686312796266131, | |
| "eval_runtime": 5.2161, | |
| "eval_samples_per_second": 581.085, | |
| "eval_steps_per_second": 9.202, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.29120656847953796, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 0.0636, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.929066314747608, | |
| "eval_f1_score": 0.8810930807565414, | |
| "eval_loss": 0.3142439126968384, | |
| "eval_precision": 0.8915600488637722, | |
| "eval_recall": 0.8743221351307482, | |
| "eval_runtime": 5.1856, | |
| "eval_samples_per_second": 584.505, | |
| "eval_steps_per_second": 9.256, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 21.05, | |
| "grad_norm": 3.0424692630767822, | |
| "learning_rate": 5.555555555555557e-06, | |
| "loss": 0.0605, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 21.05, | |
| "eval_accuracy": 0.929066314747608, | |
| "eval_f1_score": 0.8823037485745876, | |
| "eval_loss": 0.30989953875541687, | |
| "eval_precision": 0.8974412044321375, | |
| "eval_recall": 0.8700202850817763, | |
| "eval_runtime": 5.2587, | |
| "eval_samples_per_second": 576.383, | |
| "eval_steps_per_second": 9.128, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 22.11, | |
| "grad_norm": 0.18823295831680298, | |
| "learning_rate": 5.2777777777777785e-06, | |
| "loss": 0.0501, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 22.11, | |
| "eval_accuracy": 0.9317057076872319, | |
| "eval_f1_score": 0.8875182540327614, | |
| "eval_loss": 0.3163454234600067, | |
| "eval_precision": 0.9014604415235737, | |
| "eval_recall": 0.8777053165164634, | |
| "eval_runtime": 5.1733, | |
| "eval_samples_per_second": 585.895, | |
| "eval_steps_per_second": 9.278, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 23.16, | |
| "grad_norm": 0.22511476278305054, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0519, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 23.16, | |
| "eval_accuracy": 0.929726162982514, | |
| "eval_f1_score": 0.883696747826972, | |
| "eval_loss": 0.3289995491504669, | |
| "eval_precision": 0.9010520871368185, | |
| "eval_recall": 0.8692223543133784, | |
| "eval_runtime": 5.2183, | |
| "eval_samples_per_second": 580.844, | |
| "eval_steps_per_second": 9.198, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 24.21, | |
| "grad_norm": 8.104951858520508, | |
| "learning_rate": 4.722222222222222e-06, | |
| "loss": 0.0464, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 24.21, | |
| "eval_accuracy": 0.9274166941603431, | |
| "eval_f1_score": 0.8804834286728226, | |
| "eval_loss": 0.34061628580093384, | |
| "eval_precision": 0.8871850911314553, | |
| "eval_recall": 0.8771793667200236, | |
| "eval_runtime": 5.1789, | |
| "eval_samples_per_second": 585.258, | |
| "eval_steps_per_second": 9.268, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 25.26, | |
| "grad_norm": 2.2458958625793457, | |
| "learning_rate": 4.444444444444444e-06, | |
| "loss": 0.0432, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 25.26, | |
| "eval_accuracy": 0.9284064665127021, | |
| "eval_f1_score": 0.8809944439022924, | |
| "eval_loss": 0.3304508328437805, | |
| "eval_precision": 0.8875581681302606, | |
| "eval_recall": 0.8774637238579275, | |
| "eval_runtime": 5.2785, | |
| "eval_samples_per_second": 574.218, | |
| "eval_steps_per_second": 9.094, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "grad_norm": 8.245588302612305, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.0404, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "eval_accuracy": 0.9293962388650611, | |
| "eval_f1_score": 0.8825762014276358, | |
| "eval_loss": 0.3377806544303894, | |
| "eval_precision": 0.890131895776857, | |
| "eval_recall": 0.8785119382363503, | |
| "eval_runtime": 5.1758, | |
| "eval_samples_per_second": 585.606, | |
| "eval_steps_per_second": 9.274, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 27.37, | |
| "grad_norm": 5.161190032958984, | |
| "learning_rate": 3.88888888888889e-06, | |
| "loss": 0.0416, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 27.37, | |
| "eval_accuracy": 0.9284064665127021, | |
| "eval_f1_score": 0.8829688694414259, | |
| "eval_loss": 0.343578040599823, | |
| "eval_precision": 0.897684321914049, | |
| "eval_recall": 0.8726474609617948, | |
| "eval_runtime": 5.2526, | |
| "eval_samples_per_second": 577.048, | |
| "eval_steps_per_second": 9.138, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 27.37, | |
| "step": 2600, | |
| "total_flos": 5414987099379456.0, | |
| "train_loss": 0.2901941172893231, | |
| "train_runtime": 2142.8815, | |
| "train_samples_per_second": 238.931, | |
| "train_steps_per_second": 1.867 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 43, | |
| "save_steps": 100, | |
| "total_flos": 5414987099379456.0, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |