| { | |
| "best_metric": 0.0374138318002224, | |
| "best_model_checkpoint": "doc-topic-model_eval-02_train-03/checkpoint-11000", | |
| "epoch": 7.889546351084813, | |
| "eval_steps": 1000, | |
| "global_step": 16000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2465483234714004, | |
| "grad_norm": 0.3881610631942749, | |
| "learning_rate": 1.9950690335305722e-05, | |
| "loss": 0.1669, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4930966469428008, | |
| "grad_norm": 0.3564796447753906, | |
| "learning_rate": 1.9901380670611442e-05, | |
| "loss": 0.0935, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4930966469428008, | |
| "eval_accuracy": 0.9813654208653996, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.08986422419548035, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 12.0916, | |
| "eval_samples_per_second": 670.714, | |
| "eval_steps_per_second": 2.646, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7396449704142012, | |
| "grad_norm": 0.4097365140914917, | |
| "learning_rate": 1.9852071005917162e-05, | |
| "loss": 0.0869, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9861932938856016, | |
| "grad_norm": 0.3858756422996521, | |
| "learning_rate": 1.980276134122288e-05, | |
| "loss": 0.0764, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9861932938856016, | |
| "eval_accuracy": 0.9813654208653996, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.07013221085071564, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 12.0099, | |
| "eval_samples_per_second": 675.274, | |
| "eval_steps_per_second": 2.664, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.232741617357002, | |
| "grad_norm": 0.3765203356742859, | |
| "learning_rate": 1.9753451676528602e-05, | |
| "loss": 0.068, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4792899408284024, | |
| "grad_norm": 0.3380969762802124, | |
| "learning_rate": 1.9704142011834322e-05, | |
| "loss": 0.0621, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4792899408284024, | |
| "eval_accuracy": 0.9820258797851382, | |
| "eval_f1": 0.07457676590776416, | |
| "eval_loss": 0.05689988657832146, | |
| "eval_precision": 0.9190647482014388, | |
| "eval_recall": 0.038865226650441134, | |
| "eval_runtime": 14.181, | |
| "eval_samples_per_second": 571.892, | |
| "eval_steps_per_second": 2.257, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.725838264299803, | |
| "grad_norm": 0.3490855395793915, | |
| "learning_rate": 1.965483234714004e-05, | |
| "loss": 0.056, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.972386587771203, | |
| "grad_norm": 0.4068734347820282, | |
| "learning_rate": 1.9605522682445763e-05, | |
| "loss": 0.0542, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.972386587771203, | |
| "eval_accuracy": 0.9839845798432473, | |
| "eval_f1": 0.2899333919819027, | |
| "eval_loss": 0.04995810240507126, | |
| "eval_precision": 0.834056399132321, | |
| "eval_recall": 0.17546394888956496, | |
| "eval_runtime": 14.1676, | |
| "eval_samples_per_second": 572.431, | |
| "eval_steps_per_second": 2.259, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.2189349112426036, | |
| "grad_norm": 0.39391544461250305, | |
| "learning_rate": 1.9556213017751483e-05, | |
| "loss": 0.0488, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.465483234714004, | |
| "grad_norm": 0.2698710858821869, | |
| "learning_rate": 1.95069033530572e-05, | |
| "loss": 0.0468, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.465483234714004, | |
| "eval_accuracy": 0.9852105389968394, | |
| "eval_f1": 0.42338509145162184, | |
| "eval_loss": 0.046836402267217636, | |
| "eval_precision": 0.7740957769246313, | |
| "eval_recall": 0.29137511408579253, | |
| "eval_runtime": 14.2587, | |
| "eval_samples_per_second": 568.774, | |
| "eval_steps_per_second": 2.244, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.712031558185404, | |
| "grad_norm": 0.39123860001564026, | |
| "learning_rate": 1.9457593688362923e-05, | |
| "loss": 0.0456, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.9585798816568047, | |
| "grad_norm": 0.4525609016418457, | |
| "learning_rate": 1.940828402366864e-05, | |
| "loss": 0.0441, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.9585798816568047, | |
| "eval_accuracy": 0.9860779228141786, | |
| "eval_f1": 0.4908516042087804, | |
| "eval_loss": 0.04367026314139366, | |
| "eval_precision": 0.7705451586655818, | |
| "eval_recall": 0.3601308183754183, | |
| "eval_runtime": 14.4536, | |
| "eval_samples_per_second": 561.105, | |
| "eval_steps_per_second": 2.214, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.2051282051282053, | |
| "grad_norm": 0.47004055976867676, | |
| "learning_rate": 1.935897435897436e-05, | |
| "loss": 0.0409, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.4516765285996054, | |
| "grad_norm": 0.3331545293331146, | |
| "learning_rate": 1.930966469428008e-05, | |
| "loss": 0.0395, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.4516765285996054, | |
| "eval_accuracy": 0.9860495769377949, | |
| "eval_f1": 0.5308165308165308, | |
| "eval_loss": 0.04203569516539574, | |
| "eval_precision": 0.7110203039203167, | |
| "eval_recall": 0.4234864618192881, | |
| "eval_runtime": 14.3002, | |
| "eval_samples_per_second": 567.125, | |
| "eval_steps_per_second": 2.238, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.698224852071006, | |
| "grad_norm": 0.46907347440719604, | |
| "learning_rate": 1.92603550295858e-05, | |
| "loss": 0.0382, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.9447731755424065, | |
| "grad_norm": 0.3860171437263489, | |
| "learning_rate": 1.921104536489152e-05, | |
| "loss": 0.0384, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.9447731755424065, | |
| "eval_accuracy": 0.9867086185637144, | |
| "eval_f1": 0.5639761949042217, | |
| "eval_loss": 0.039942242205142975, | |
| "eval_precision": 0.7254784688995215, | |
| "eval_recall": 0.46128688773958015, | |
| "eval_runtime": 14.235, | |
| "eval_samples_per_second": 569.723, | |
| "eval_steps_per_second": 2.248, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.191321499013807, | |
| "grad_norm": 0.39796024560928345, | |
| "learning_rate": 1.916173570019724e-05, | |
| "loss": 0.0348, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.437869822485207, | |
| "grad_norm": 0.3643665313720703, | |
| "learning_rate": 1.911242603550296e-05, | |
| "loss": 0.0343, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.437869822485207, | |
| "eval_accuracy": 0.986822002069249, | |
| "eval_f1": 0.577286779414439, | |
| "eval_loss": 0.03921860456466675, | |
| "eval_precision": 0.7175632911392406, | |
| "eval_recall": 0.4828871311226042, | |
| "eval_runtime": 14.1836, | |
| "eval_samples_per_second": 571.788, | |
| "eval_steps_per_second": 2.256, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.684418145956608, | |
| "grad_norm": 0.2764923572540283, | |
| "learning_rate": 1.906311637080868e-05, | |
| "loss": 0.0345, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.930966469428008, | |
| "grad_norm": 0.432156503200531, | |
| "learning_rate": 1.90138067061144e-05, | |
| "loss": 0.0337, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.930966469428008, | |
| "eval_accuracy": 0.9873166376121434, | |
| "eval_f1": 0.5935782733094146, | |
| "eval_loss": 0.03798520192503929, | |
| "eval_precision": 0.736670048472551, | |
| "eval_recall": 0.4970337693945847, | |
| "eval_runtime": 14.1326, | |
| "eval_samples_per_second": 573.852, | |
| "eval_steps_per_second": 2.264, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.177514792899408, | |
| "grad_norm": 0.27451589703559875, | |
| "learning_rate": 1.896459566074951e-05, | |
| "loss": 0.0309, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.424063116370808, | |
| "grad_norm": 0.49061715602874756, | |
| "learning_rate": 1.891528599605523e-05, | |
| "loss": 0.0305, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.424063116370808, | |
| "eval_accuracy": 0.9874583669940615, | |
| "eval_f1": 0.596470427288066, | |
| "eval_loss": 0.0374138318002224, | |
| "eval_precision": 0.7447898872565767, | |
| "eval_recall": 0.49741405536963795, | |
| "eval_runtime": 14.1978, | |
| "eval_samples_per_second": 571.214, | |
| "eval_steps_per_second": 2.254, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.670611439842209, | |
| "grad_norm": 0.42573484778404236, | |
| "learning_rate": 1.886597633136095e-05, | |
| "loss": 0.0299, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.9171597633136095, | |
| "grad_norm": 0.4632101058959961, | |
| "learning_rate": 1.881666666666667e-05, | |
| "loss": 0.0295, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.9171597633136095, | |
| "eval_accuracy": 0.9874172654733052, | |
| "eval_f1": 0.607688908528502, | |
| "eval_loss": 0.03787928447127342, | |
| "eval_precision": 0.7251634676228643, | |
| "eval_recall": 0.5229692728932157, | |
| "eval_runtime": 14.2477, | |
| "eval_samples_per_second": 569.215, | |
| "eval_steps_per_second": 2.246, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.16370808678501, | |
| "grad_norm": 0.5974611043930054, | |
| "learning_rate": 1.876735700197239e-05, | |
| "loss": 0.0271, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.410256410256411, | |
| "grad_norm": 0.42065271735191345, | |
| "learning_rate": 1.8718145956607497e-05, | |
| "loss": 0.0271, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.410256410256411, | |
| "eval_accuracy": 0.9876397806029168, | |
| "eval_f1": 0.6051523520622991, | |
| "eval_loss": 0.03748546913266182, | |
| "eval_precision": 0.7476227765969348, | |
| "eval_recall": 0.5082902342561606, | |
| "eval_runtime": 14.3656, | |
| "eval_samples_per_second": 564.544, | |
| "eval_steps_per_second": 2.228, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.65680473372781, | |
| "grad_norm": 0.7562135457992554, | |
| "learning_rate": 1.8668836291913217e-05, | |
| "loss": 0.0265, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.903353057199211, | |
| "grad_norm": 0.5237112641334534, | |
| "learning_rate": 1.8619526627218937e-05, | |
| "loss": 0.0257, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.903353057199211, | |
| "eval_accuracy": 0.9876737956545771, | |
| "eval_f1": 0.6152280670707428, | |
| "eval_loss": 0.037624698132276535, | |
| "eval_precision": 0.7353781068217874, | |
| "eval_recall": 0.5288256769090356, | |
| "eval_runtime": 14.2425, | |
| "eval_samples_per_second": 569.424, | |
| "eval_steps_per_second": 2.247, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.149901380670611, | |
| "grad_norm": 0.38859695196151733, | |
| "learning_rate": 1.8570216962524657e-05, | |
| "loss": 0.0239, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 7.396449704142012, | |
| "grad_norm": 0.276382178068161, | |
| "learning_rate": 1.8520907297830377e-05, | |
| "loss": 0.0234, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.396449704142012, | |
| "eval_accuracy": 0.9876780475360347, | |
| "eval_f1": 0.6280800821355236, | |
| "eval_loss": 0.03744862973690033, | |
| "eval_precision": 0.7177356276886977, | |
| "eval_recall": 0.558335868573167, | |
| "eval_runtime": 14.1538, | |
| "eval_samples_per_second": 572.992, | |
| "eval_steps_per_second": 2.261, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.642998027613412, | |
| "grad_norm": 0.23438788950443268, | |
| "learning_rate": 1.8471696252465485e-05, | |
| "loss": 0.0234, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 7.889546351084813, | |
| "grad_norm": 0.3130985498428345, | |
| "learning_rate": 1.8422386587771205e-05, | |
| "loss": 0.0241, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 7.889546351084813, | |
| "eval_accuracy": 0.9877262355258869, | |
| "eval_f1": 0.6217679944095038, | |
| "eval_loss": 0.03812398761510849, | |
| "eval_precision": 0.730201066885515, | |
| "eval_recall": 0.5413751140857925, | |
| "eval_runtime": 14.1384, | |
| "eval_samples_per_second": 573.614, | |
| "eval_steps_per_second": 2.263, | |
| "step": 16000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 202800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 360703978994412.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |