| { | |
| "best_metric": 0.03718467801809311, | |
| "best_model_checkpoint": "doc-topic-model_eval-00_train-02/checkpoint-13000", | |
| "epoch": 8.875739644970414, | |
| "eval_steps": 1000, | |
| "global_step": 18000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2465483234714004, | |
| "grad_norm": 0.3767470121383667, | |
| "learning_rate": 1.9950690335305722e-05, | |
| "loss": 0.1663, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4930966469428008, | |
| "grad_norm": 0.4007192850112915, | |
| "learning_rate": 1.9901380670611442e-05, | |
| "loss": 0.0932, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4930966469428008, | |
| "eval_accuracy": 0.9814660487265615, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.0863451287150383, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 11.8139, | |
| "eval_samples_per_second": 686.478, | |
| "eval_steps_per_second": 2.709, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7396449704142012, | |
| "grad_norm": 0.4121333062648773, | |
| "learning_rate": 1.9852071005917162e-05, | |
| "loss": 0.0831, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9861932938856016, | |
| "grad_norm": 0.35336044430732727, | |
| "learning_rate": 1.980276134122288e-05, | |
| "loss": 0.0742, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9861932938856016, | |
| "eval_accuracy": 0.9814660487265615, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.06615250557661057, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 11.8646, | |
| "eval_samples_per_second": 683.543, | |
| "eval_steps_per_second": 2.697, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.232741617357002, | |
| "grad_norm": 0.4058854281902313, | |
| "learning_rate": 1.9753451676528602e-05, | |
| "loss": 0.0654, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4792899408284024, | |
| "grad_norm": 0.4402715861797333, | |
| "learning_rate": 1.9704142011834322e-05, | |
| "loss": 0.0596, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4792899408284024, | |
| "eval_accuracy": 0.9825927973128109, | |
| "eval_f1": 0.1370151770657673, | |
| "eval_loss": 0.05494347959756851, | |
| "eval_precision": 0.8441558441558441, | |
| "eval_recall": 0.07455838495067676, | |
| "eval_runtime": 14.1809, | |
| "eval_samples_per_second": 571.895, | |
| "eval_steps_per_second": 2.257, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.725838264299803, | |
| "grad_norm": 0.39516812562942505, | |
| "learning_rate": 1.965483234714004e-05, | |
| "loss": 0.0558, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.972386587771203, | |
| "grad_norm": 0.3793538510799408, | |
| "learning_rate": 1.9605522682445763e-05, | |
| "loss": 0.053, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.972386587771203, | |
| "eval_accuracy": 0.9848675538925975, | |
| "eval_f1": 0.38985084861992114, | |
| "eval_loss": 0.049108367413282394, | |
| "eval_precision": 0.771370420624152, | |
| "eval_recall": 0.26083964211975225, | |
| "eval_runtime": 14.0996, | |
| "eval_samples_per_second": 575.194, | |
| "eval_steps_per_second": 2.27, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.2189349112426036, | |
| "grad_norm": 0.33267995715141296, | |
| "learning_rate": 1.9556213017751483e-05, | |
| "loss": 0.0474, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.465483234714004, | |
| "grad_norm": 0.3178996443748474, | |
| "learning_rate": 1.95069033530572e-05, | |
| "loss": 0.0467, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.465483234714004, | |
| "eval_accuracy": 0.9857207647717449, | |
| "eval_f1": 0.45265388167544957, | |
| "eval_loss": 0.045239612460136414, | |
| "eval_precision": 0.7816135084427768, | |
| "eval_recall": 0.31857459662001986, | |
| "eval_runtime": 14.2568, | |
| "eval_samples_per_second": 568.853, | |
| "eval_steps_per_second": 2.245, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.712031558185404, | |
| "grad_norm": 0.41016682982444763, | |
| "learning_rate": 1.9457593688362923e-05, | |
| "loss": 0.0453, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.9585798816568047, | |
| "grad_norm": 0.30701619386672974, | |
| "learning_rate": 1.940828402366864e-05, | |
| "loss": 0.044, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.9585798816568047, | |
| "eval_accuracy": 0.986354295108919, | |
| "eval_f1": 0.5022231413504291, | |
| "eval_loss": 0.04268515482544899, | |
| "eval_precision": 0.7752593774940144, | |
| "eval_recall": 0.3714154622619867, | |
| "eval_runtime": 14.3323, | |
| "eval_samples_per_second": 565.856, | |
| "eval_steps_per_second": 2.233, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.2051282051282053, | |
| "grad_norm": 0.3377997875213623, | |
| "learning_rate": 1.935897435897436e-05, | |
| "loss": 0.0391, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.4516765285996054, | |
| "grad_norm": 0.3832477629184723, | |
| "learning_rate": 1.930966469428008e-05, | |
| "loss": 0.039, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.4516765285996054, | |
| "eval_accuracy": 0.9867454682030131, | |
| "eval_f1": 0.5505142747284437, | |
| "eval_loss": 0.04092588648200035, | |
| "eval_precision": 0.7409755466425152, | |
| "eval_recall": 0.43794448267951364, | |
| "eval_runtime": 14.2213, | |
| "eval_samples_per_second": 570.271, | |
| "eval_steps_per_second": 2.25, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.698224852071006, | |
| "grad_norm": 0.5340074896812439, | |
| "learning_rate": 1.92603550295858e-05, | |
| "loss": 0.0388, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.9447731755424065, | |
| "grad_norm": 0.40983426570892334, | |
| "learning_rate": 1.921104536489152e-05, | |
| "loss": 0.037, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.9447731755424065, | |
| "eval_accuracy": 0.986977904389359, | |
| "eval_f1": 0.5589477726574501, | |
| "eval_loss": 0.039010029286146164, | |
| "eval_precision": 0.7507414571244359, | |
| "eval_recall": 0.445209145828554, | |
| "eval_runtime": 14.193, | |
| "eval_samples_per_second": 571.408, | |
| "eval_steps_per_second": 2.255, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.191321499013807, | |
| "grad_norm": 0.5406789183616638, | |
| "learning_rate": 1.916173570019724e-05, | |
| "loss": 0.0354, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.437869822485207, | |
| "grad_norm": 0.3440360724925995, | |
| "learning_rate": 1.911242603550296e-05, | |
| "loss": 0.0337, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.437869822485207, | |
| "eval_accuracy": 0.9875023031024561, | |
| "eval_f1": 0.5771960107403146, | |
| "eval_loss": 0.03828894719481468, | |
| "eval_precision": 0.7737498393109654, | |
| "eval_recall": 0.4602737630955112, | |
| "eval_runtime": 14.3368, | |
| "eval_samples_per_second": 565.678, | |
| "eval_steps_per_second": 2.232, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.684418145956608, | |
| "grad_norm": 0.44961240887641907, | |
| "learning_rate": 1.906311637080868e-05, | |
| "loss": 0.0328, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.930966469428008, | |
| "grad_norm": 0.4392942488193512, | |
| "learning_rate": 1.90138067061144e-05, | |
| "loss": 0.0337, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.930966469428008, | |
| "eval_accuracy": 0.9875349008602974, | |
| "eval_f1": 0.5916709225126515, | |
| "eval_loss": 0.037467557936906815, | |
| "eval_precision": 0.7530134719924367, | |
| "eval_recall": 0.48726772195457674, | |
| "eval_runtime": 14.0557, | |
| "eval_samples_per_second": 576.991, | |
| "eval_steps_per_second": 2.277, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.177514792899408, | |
| "grad_norm": 0.302545428276062, | |
| "learning_rate": 1.896459566074951e-05, | |
| "loss": 0.0308, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.424063116370808, | |
| "grad_norm": 0.3933587968349457, | |
| "learning_rate": 1.891528599605523e-05, | |
| "loss": 0.0293, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.424063116370808, | |
| "eval_accuracy": 0.9876893858865882, | |
| "eval_f1": 0.6104932735426009, | |
| "eval_loss": 0.03747553750872612, | |
| "eval_precision": 0.738046188875637, | |
| "eval_recall": 0.5205322321633402, | |
| "eval_runtime": 14.3104, | |
| "eval_samples_per_second": 566.721, | |
| "eval_steps_per_second": 2.236, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.670611439842209, | |
| "grad_norm": 0.46541067957878113, | |
| "learning_rate": 1.886597633136095e-05, | |
| "loss": 0.0297, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.9171597633136095, | |
| "grad_norm": 0.41115203499794006, | |
| "learning_rate": 1.881666666666667e-05, | |
| "loss": 0.0297, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.9171597633136095, | |
| "eval_accuracy": 0.9876057655512565, | |
| "eval_f1": 0.6050313897294611, | |
| "eval_loss": 0.03745245933532715, | |
| "eval_precision": 0.7389673433362753, | |
| "eval_recall": 0.5121969870765466, | |
| "eval_runtime": 14.1164, | |
| "eval_samples_per_second": 574.507, | |
| "eval_steps_per_second": 2.267, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.16370808678501, | |
| "grad_norm": 0.3442145884037018, | |
| "learning_rate": 1.8767455621301777e-05, | |
| "loss": 0.0268, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.410256410256411, | |
| "grad_norm": 0.13360050320625305, | |
| "learning_rate": 1.8718145956607497e-05, | |
| "loss": 0.0263, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.410256410256411, | |
| "eval_accuracy": 0.9878920589027311, | |
| "eval_f1": 0.6160276866376017, | |
| "eval_loss": 0.03718467801809311, | |
| "eval_precision": 0.7471652856519843, | |
| "eval_recall": 0.524049858530244, | |
| "eval_runtime": 14.154, | |
| "eval_samples_per_second": 572.983, | |
| "eval_steps_per_second": 2.261, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.65680473372781, | |
| "grad_norm": 0.31057894229888916, | |
| "learning_rate": 1.8668836291913217e-05, | |
| "loss": 0.0264, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.903353057199211, | |
| "grad_norm": 0.2787815034389496, | |
| "learning_rate": 1.8619526627218937e-05, | |
| "loss": 0.0265, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.903353057199211, | |
| "eval_accuracy": 0.9876043482574373, | |
| "eval_f1": 0.6178449707244603, | |
| "eval_loss": 0.03766901046037674, | |
| "eval_precision": 0.7207666428789887, | |
| "eval_recall": 0.5406438785654202, | |
| "eval_runtime": 14.3038, | |
| "eval_samples_per_second": 566.983, | |
| "eval_steps_per_second": 2.237, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.149901380670611, | |
| "grad_norm": 0.20837119221687317, | |
| "learning_rate": 1.8570315581854045e-05, | |
| "loss": 0.0249, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 7.396449704142012, | |
| "grad_norm": 0.23010149598121643, | |
| "learning_rate": 1.8521005917159765e-05, | |
| "loss": 0.0235, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.396449704142012, | |
| "eval_accuracy": 0.9878296979746871, | |
| "eval_f1": 0.6237897042716319, | |
| "eval_loss": 0.03783150017261505, | |
| "eval_precision": 0.7303036520311859, | |
| "eval_recall": 0.5443909153475568, | |
| "eval_runtime": 14.1679, | |
| "eval_samples_per_second": 572.422, | |
| "eval_steps_per_second": 2.259, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.642998027613412, | |
| "grad_norm": 0.3158400058746338, | |
| "learning_rate": 1.8471696252465485e-05, | |
| "loss": 0.0226, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 7.889546351084813, | |
| "grad_norm": 0.5647494196891785, | |
| "learning_rate": 1.8422386587771205e-05, | |
| "loss": 0.0237, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 7.889546351084813, | |
| "eval_accuracy": 0.9877843445724733, | |
| "eval_f1": 0.6255376460876744, | |
| "eval_loss": 0.03789234161376953, | |
| "eval_precision": 0.7242454728370221, | |
| "eval_recall": 0.5505085264204328, | |
| "eval_runtime": 14.1193, | |
| "eval_samples_per_second": 574.389, | |
| "eval_steps_per_second": 2.266, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.136094674556213, | |
| "grad_norm": 0.2036384493112564, | |
| "learning_rate": 1.8373076923076926e-05, | |
| "loss": 0.0216, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 8.382642998027613, | |
| "grad_norm": 0.1960095316171646, | |
| "learning_rate": 1.8323865877712033e-05, | |
| "loss": 0.0205, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 8.382642998027613, | |
| "eval_accuracy": 0.987797100216846, | |
| "eval_f1": 0.6324282786885246, | |
| "eval_loss": 0.038331326097249985, | |
| "eval_precision": 0.7158596694694114, | |
| "eval_recall": 0.5664143152099105, | |
| "eval_runtime": 14.1717, | |
| "eval_samples_per_second": 572.267, | |
| "eval_steps_per_second": 2.258, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 8.629191321499015, | |
| "grad_norm": 0.44110924005508423, | |
| "learning_rate": 1.8274556213017754e-05, | |
| "loss": 0.0209, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 8.875739644970414, | |
| "grad_norm": 0.5276309847831726, | |
| "learning_rate": 1.822524654832347e-05, | |
| "loss": 0.0208, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 8.875739644970414, | |
| "eval_accuracy": 0.9874413594682313, | |
| "eval_f1": 0.631083725384071, | |
| "eval_loss": 0.03931192308664322, | |
| "eval_precision": 0.6926521659660025, | |
| "eval_recall": 0.579567179016594, | |
| "eval_runtime": 14.1285, | |
| "eval_samples_per_second": 574.016, | |
| "eval_steps_per_second": 2.265, | |
| "step": 18000 | |
| } | |
| ], | |
| "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": 406579786604172.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |