| { | |
| "best_metric": 0.8919460068914353, | |
| "best_model_checkpoint": "cls_comment-phobert-base-v2-v2.4.1/checkpoint-2300", | |
| "epoch": 26.794258373205743, | |
| "eval_steps": 100, | |
| "global_step": 2800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.6857236623764038, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.7049, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.46144614461446143, | |
| "eval_f1_score": 0.10801682151937281, | |
| "eval_loss": 1.4949835538864136, | |
| "eval_precision": 0.21007122078443174, | |
| "eval_recall": 0.16814650388457272, | |
| "eval_runtime": 6.4303, | |
| "eval_samples_per_second": 518.327, | |
| "eval_steps_per_second": 8.242, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 2.8904168605804443, | |
| "learning_rate": 5e-06, | |
| "loss": 1.3066, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "eval_accuracy": 0.6597659765976598, | |
| "eval_f1_score": 0.24932737263473323, | |
| "eval_loss": 1.0450588464736938, | |
| "eval_precision": 0.21503708017040588, | |
| "eval_recall": 0.29703415957578, | |
| "eval_runtime": 6.2935, | |
| "eval_samples_per_second": 529.594, | |
| "eval_steps_per_second": 8.421, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 4.716286659240723, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.9457, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "eval_accuracy": 0.7971797179717972, | |
| "eval_f1_score": 0.5218513395163904, | |
| "eval_loss": 0.7491273283958435, | |
| "eval_precision": 0.5237982031957714, | |
| "eval_recall": 0.5230368404187595, | |
| "eval_runtime": 6.3214, | |
| "eval_samples_per_second": 527.257, | |
| "eval_steps_per_second": 8.384, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "grad_norm": 4.255033016204834, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6975, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "eval_accuracy": 0.8496849684968497, | |
| "eval_f1_score": 0.569984189542809, | |
| "eval_loss": 0.5573880672454834, | |
| "eval_precision": 0.7142864553492309, | |
| "eval_recall": 0.5934564957773306, | |
| "eval_runtime": 6.2462, | |
| "eval_samples_per_second": 533.607, | |
| "eval_steps_per_second": 8.485, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "grad_norm": 9.066823959350586, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.5187, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "eval_accuracy": 0.8664866486648665, | |
| "eval_f1_score": 0.6685229055473728, | |
| "eval_loss": 0.46813032031059265, | |
| "eval_precision": 0.7076541248010603, | |
| "eval_recall": 0.6591572058707156, | |
| "eval_runtime": 6.2485, | |
| "eval_samples_per_second": 533.408, | |
| "eval_steps_per_second": 8.482, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "grad_norm": 5.799560070037842, | |
| "learning_rate": 9.444444444444445e-06, | |
| "loss": 0.4183, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "eval_accuracy": 0.8820882088208821, | |
| "eval_f1_score": 0.774744299508991, | |
| "eval_loss": 0.4121437668800354, | |
| "eval_precision": 0.8760939894193666, | |
| "eval_recall": 0.7478297020405164, | |
| "eval_runtime": 6.289, | |
| "eval_samples_per_second": 529.974, | |
| "eval_steps_per_second": 8.427, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "grad_norm": 5.299034595489502, | |
| "learning_rate": 9.166666666666666e-06, | |
| "loss": 0.3323, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "eval_accuracy": 0.903990399039904, | |
| "eval_f1_score": 0.850497109426447, | |
| "eval_loss": 0.34878164529800415, | |
| "eval_precision": 0.8647331643827426, | |
| "eval_recall": 0.839118025290519, | |
| "eval_runtime": 6.1906, | |
| "eval_samples_per_second": 538.4, | |
| "eval_steps_per_second": 8.561, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "grad_norm": 5.325865268707275, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 0.2705, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "eval_accuracy": 0.9123912391239124, | |
| "eval_f1_score": 0.8679529123789608, | |
| "eval_loss": 0.3178523778915405, | |
| "eval_precision": 0.8683198068711735, | |
| "eval_recall": 0.8693736512924785, | |
| "eval_runtime": 6.2028, | |
| "eval_samples_per_second": 537.341, | |
| "eval_steps_per_second": 8.545, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "grad_norm": 4.130426406860352, | |
| "learning_rate": 8.611111111111112e-06, | |
| "loss": 0.229, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "eval_accuracy": 0.915991599159916, | |
| "eval_f1_score": 0.8739115802564847, | |
| "eval_loss": 0.3108988404273987, | |
| "eval_precision": 0.8704138816567074, | |
| "eval_recall": 0.8778451486716982, | |
| "eval_runtime": 6.2053, | |
| "eval_samples_per_second": 537.123, | |
| "eval_steps_per_second": 8.541, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "grad_norm": 4.939031600952148, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.1964, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "eval_accuracy": 0.9174917491749175, | |
| "eval_f1_score": 0.8775661449676492, | |
| "eval_loss": 0.3028014600276947, | |
| "eval_precision": 0.8740691720600018, | |
| "eval_recall": 0.8812896711088255, | |
| "eval_runtime": 6.2156, | |
| "eval_samples_per_second": 536.232, | |
| "eval_steps_per_second": 8.527, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "grad_norm": 7.8568878173828125, | |
| "learning_rate": 8.055555555555557e-06, | |
| "loss": 0.1771, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "eval_accuracy": 0.918091809180918, | |
| "eval_f1_score": 0.8807318795078937, | |
| "eval_loss": 0.30322423577308655, | |
| "eval_precision": 0.8743317066593773, | |
| "eval_recall": 0.8877139802470034, | |
| "eval_runtime": 6.2526, | |
| "eval_samples_per_second": 533.058, | |
| "eval_steps_per_second": 8.476, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 11.48, | |
| "grad_norm": 8.988609313964844, | |
| "learning_rate": 7.77777777777778e-06, | |
| "loss": 0.1518, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.48, | |
| "eval_accuracy": 0.9165916591659166, | |
| "eval_f1_score": 0.8762082418137522, | |
| "eval_loss": 0.3150691092014313, | |
| "eval_precision": 0.8828474311875153, | |
| "eval_recall": 0.8702178521387077, | |
| "eval_runtime": 6.2783, | |
| "eval_samples_per_second": 530.878, | |
| "eval_steps_per_second": 8.442, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "grad_norm": 6.535445690155029, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.1368, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "eval_accuracy": 0.9213921392139214, | |
| "eval_f1_score": 0.8794257945528638, | |
| "eval_loss": 0.29376497864723206, | |
| "eval_precision": 0.8789472587468637, | |
| "eval_recall": 0.8800344348044677, | |
| "eval_runtime": 6.2765, | |
| "eval_samples_per_second": 531.025, | |
| "eval_steps_per_second": 8.444, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "grad_norm": 6.200416564941406, | |
| "learning_rate": 7.222222222222223e-06, | |
| "loss": 0.1116, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "eval_accuracy": 0.9204920492049204, | |
| "eval_f1_score": 0.8794630237866472, | |
| "eval_loss": 0.29709574580192566, | |
| "eval_precision": 0.8776282905981169, | |
| "eval_recall": 0.881459841742562, | |
| "eval_runtime": 6.1898, | |
| "eval_samples_per_second": 538.463, | |
| "eval_steps_per_second": 8.562, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 14.35, | |
| "grad_norm": 5.921875, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.1136, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.35, | |
| "eval_accuracy": 0.9234923492349235, | |
| "eval_f1_score": 0.8857825928642494, | |
| "eval_loss": 0.30112606287002563, | |
| "eval_precision": 0.8894251615389424, | |
| "eval_recall": 0.8825136450024287, | |
| "eval_runtime": 6.2125, | |
| "eval_samples_per_second": 536.5, | |
| "eval_steps_per_second": 8.531, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "grad_norm": 13.020410537719727, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.094, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "eval_accuracy": 0.9267926792679267, | |
| "eval_f1_score": 0.8891204106050078, | |
| "eval_loss": 0.2937377691268921, | |
| "eval_precision": 0.8855321931614877, | |
| "eval_recall": 0.8933184630720276, | |
| "eval_runtime": 6.2824, | |
| "eval_samples_per_second": 530.532, | |
| "eval_steps_per_second": 8.436, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 16.27, | |
| "grad_norm": 6.572815418243408, | |
| "learning_rate": 6.3888888888888885e-06, | |
| "loss": 0.0905, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 16.27, | |
| "eval_accuracy": 0.9264926492649265, | |
| "eval_f1_score": 0.8849818091608542, | |
| "eval_loss": 0.30491721630096436, | |
| "eval_precision": 0.8885868021143595, | |
| "eval_recall": 0.8819482912765212, | |
| "eval_runtime": 6.1862, | |
| "eval_samples_per_second": 538.781, | |
| "eval_steps_per_second": 8.567, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 17.22, | |
| "grad_norm": 11.92620849609375, | |
| "learning_rate": 6.111111111111112e-06, | |
| "loss": 0.0838, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 17.22, | |
| "eval_accuracy": 0.9243924392439244, | |
| "eval_f1_score": 0.8823178893108493, | |
| "eval_loss": 0.30605384707450867, | |
| "eval_precision": 0.8783548672468386, | |
| "eval_recall": 0.8869437658317699, | |
| "eval_runtime": 6.213, | |
| "eval_samples_per_second": 536.453, | |
| "eval_steps_per_second": 8.53, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "grad_norm": 6.095840930938721, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 0.0749, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "eval_accuracy": 0.9204920492049204, | |
| "eval_f1_score": 0.8771352609895953, | |
| "eval_loss": 0.3274901509284973, | |
| "eval_precision": 0.8717396476070715, | |
| "eval_recall": 0.883927539576653, | |
| "eval_runtime": 6.221, | |
| "eval_samples_per_second": 535.769, | |
| "eval_steps_per_second": 8.52, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 19.14, | |
| "grad_norm": 6.83023738861084, | |
| "learning_rate": 5.555555555555557e-06, | |
| "loss": 0.0686, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 19.14, | |
| "eval_accuracy": 0.9294929492949295, | |
| "eval_f1_score": 0.8915069438815207, | |
| "eval_loss": 0.3092004358768463, | |
| "eval_precision": 0.8845623861216612, | |
| "eval_recall": 0.8989552312613528, | |
| "eval_runtime": 6.2372, | |
| "eval_samples_per_second": 534.375, | |
| "eval_steps_per_second": 8.497, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 20.1, | |
| "grad_norm": 4.03262186050415, | |
| "learning_rate": 5.2777777777777785e-06, | |
| "loss": 0.0669, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 20.1, | |
| "eval_accuracy": 0.924992499249925, | |
| "eval_f1_score": 0.8835986677013299, | |
| "eval_loss": 0.3168080151081085, | |
| "eval_precision": 0.8824571821499484, | |
| "eval_recall": 0.8848697486050657, | |
| "eval_runtime": 6.2215, | |
| "eval_samples_per_second": 535.72, | |
| "eval_steps_per_second": 8.519, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 21.05, | |
| "grad_norm": 1.4791430234909058, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0582, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 21.05, | |
| "eval_accuracy": 0.9234923492349235, | |
| "eval_f1_score": 0.8762756882804612, | |
| "eval_loss": 0.333869606256485, | |
| "eval_precision": 0.8630877156327875, | |
| "eval_recall": 0.8925977028354047, | |
| "eval_runtime": 6.2137, | |
| "eval_samples_per_second": 536.399, | |
| "eval_steps_per_second": 8.53, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 22.01, | |
| "grad_norm": 1.3378651142120361, | |
| "learning_rate": 4.722222222222222e-06, | |
| "loss": 0.0516, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 22.01, | |
| "eval_accuracy": 0.9267926792679267, | |
| "eval_f1_score": 0.8919460068914353, | |
| "eval_loss": 0.3273985981941223, | |
| "eval_precision": 0.8896937401913322, | |
| "eval_recall": 0.8944103789209331, | |
| "eval_runtime": 6.2539, | |
| "eval_samples_per_second": 532.947, | |
| "eval_steps_per_second": 8.475, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "grad_norm": 2.3011834621429443, | |
| "learning_rate": 4.444444444444444e-06, | |
| "loss": 0.0543, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "eval_accuracy": 0.9294929492949295, | |
| "eval_f1_score": 0.8912698081733191, | |
| "eval_loss": 0.32296448945999146, | |
| "eval_precision": 0.8946222329092709, | |
| "eval_recall": 0.8882296547805383, | |
| "eval_runtime": 6.2569, | |
| "eval_samples_per_second": 532.688, | |
| "eval_steps_per_second": 8.471, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 23.92, | |
| "grad_norm": 0.30257853865623474, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.0435, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 23.92, | |
| "eval_accuracy": 0.9252925292529253, | |
| "eval_f1_score": 0.8806373103770997, | |
| "eval_loss": 0.3364492952823639, | |
| "eval_precision": 0.8917652898214161, | |
| "eval_recall": 0.8705348522415428, | |
| "eval_runtime": 6.2476, | |
| "eval_samples_per_second": 533.488, | |
| "eval_steps_per_second": 8.483, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 24.88, | |
| "grad_norm": 4.352519512176514, | |
| "learning_rate": 3.88888888888889e-06, | |
| "loss": 0.0405, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 24.88, | |
| "eval_accuracy": 0.9240924092409241, | |
| "eval_f1_score": 0.8816233559879275, | |
| "eval_loss": 0.34915250539779663, | |
| "eval_precision": 0.8818560471877731, | |
| "eval_recall": 0.8820867223989518, | |
| "eval_runtime": 6.2343, | |
| "eval_samples_per_second": 534.62, | |
| "eval_steps_per_second": 8.501, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 25.84, | |
| "grad_norm": 1.6296316385269165, | |
| "learning_rate": 3.6111111111111115e-06, | |
| "loss": 0.0398, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 25.84, | |
| "eval_accuracy": 0.9237923792379238, | |
| "eval_f1_score": 0.8799317994806395, | |
| "eval_loss": 0.3558264374732971, | |
| "eval_precision": 0.8807466782442214, | |
| "eval_recall": 0.879644875034168, | |
| "eval_runtime": 6.2137, | |
| "eval_samples_per_second": 536.396, | |
| "eval_steps_per_second": 8.53, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 26.79, | |
| "grad_norm": 0.5989682078361511, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.0363, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 26.79, | |
| "eval_accuracy": 0.9222922292229223, | |
| "eval_f1_score": 0.8741787015976087, | |
| "eval_loss": 0.36046338081359863, | |
| "eval_precision": 0.8698137895606535, | |
| "eval_recall": 0.8795311139961707, | |
| "eval_runtime": 6.2828, | |
| "eval_samples_per_second": 530.494, | |
| "eval_steps_per_second": 8.436, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 26.79, | |
| "step": 2800, | |
| "total_flos": 6555293809710912.0, | |
| "train_loss": 0.28978255237851824, | |
| "train_runtime": 2707.5005, | |
| "train_samples_per_second": 189.104, | |
| "train_steps_per_second": 1.477 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 39, | |
| "save_steps": 100, | |
| "total_flos": 6555293809710912.0, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |