| { | |
| "best_metric": 0.3550606966018677, | |
| "best_model_checkpoint": "bert_tiny_lda_100_v1_qqp/checkpoint-8532", | |
| "epoch": 11.0, | |
| "eval_steps": 500, | |
| "global_step": 15642, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.963968276977539, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.4874, | |
| "step": 1422 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.797996537224833, | |
| "eval_combined_score": 0.7552586086455046, | |
| "eval_f1": 0.7125206800661763, | |
| "eval_loss": 0.4273848235607147, | |
| "eval_runtime": 12.4751, | |
| "eval_samples_per_second": 3240.852, | |
| "eval_steps_per_second": 12.665, | |
| "step": 1422 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.7753087282180786, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.388, | |
| "step": 2844 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.822384368043532, | |
| "eval_combined_score": 0.7974788031824944, | |
| "eval_f1": 0.7725732383214569, | |
| "eval_loss": 0.37860846519470215, | |
| "eval_runtime": 12.4162, | |
| "eval_samples_per_second": 3256.242, | |
| "eval_steps_per_second": 12.725, | |
| "step": 2844 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.818178176879883, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.3354, | |
| "step": 4266 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8372248330447687, | |
| "eval_combined_score": 0.8135618147281756, | |
| "eval_f1": 0.7898987964115826, | |
| "eval_loss": 0.3613271117210388, | |
| "eval_runtime": 12.3656, | |
| "eval_samples_per_second": 3269.55, | |
| "eval_steps_per_second": 12.777, | |
| "step": 4266 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.026447296142578, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.2928, | |
| "step": 5688 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8447440019787287, | |
| "eval_combined_score": 0.8138597443078306, | |
| "eval_f1": 0.7829754866369325, | |
| "eval_loss": 0.3564006984233856, | |
| "eval_runtime": 12.2406, | |
| "eval_samples_per_second": 3302.952, | |
| "eval_steps_per_second": 12.908, | |
| "step": 5688 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.733372688293457, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.2583, | |
| "step": 7110 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8509275290625773, | |
| "eval_combined_score": 0.8253042271895796, | |
| "eval_f1": 0.7996809253165819, | |
| "eval_loss": 0.3613673448562622, | |
| "eval_runtime": 12.2989, | |
| "eval_samples_per_second": 3287.292, | |
| "eval_steps_per_second": 12.847, | |
| "step": 7110 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.535404682159424, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.2277, | |
| "step": 8532 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8543408360128617, | |
| "eval_combined_score": 0.83032142284148, | |
| "eval_f1": 0.8063020096700984, | |
| "eval_loss": 0.3550606966018677, | |
| "eval_runtime": 12.5469, | |
| "eval_samples_per_second": 3222.316, | |
| "eval_steps_per_second": 12.593, | |
| "step": 8532 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.6766912937164307, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.2014, | |
| "step": 9954 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8551570615879297, | |
| "eval_combined_score": 0.8322039379600886, | |
| "eval_f1": 0.8092508143322475, | |
| "eval_loss": 0.3854043185710907, | |
| "eval_runtime": 12.1595, | |
| "eval_samples_per_second": 3324.98, | |
| "eval_steps_per_second": 12.994, | |
| "step": 9954 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.472510576248169, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.1784, | |
| "step": 11376 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8544892406628741, | |
| "eval_combined_score": 0.8304560974134679, | |
| "eval_f1": 0.8064229541640617, | |
| "eval_loss": 0.3979368805885315, | |
| "eval_runtime": 12.3624, | |
| "eval_samples_per_second": 3270.399, | |
| "eval_steps_per_second": 12.781, | |
| "step": 11376 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.070095062255859, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.1578, | |
| "step": 12798 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8558496166213208, | |
| "eval_combined_score": 0.8330247887705053, | |
| "eval_f1": 0.8101999609196899, | |
| "eval_loss": 0.4261317253112793, | |
| "eval_runtime": 12.3748, | |
| "eval_samples_per_second": 3267.134, | |
| "eval_steps_per_second": 12.768, | |
| "step": 12798 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.72301983833313, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1403, | |
| "step": 14220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8587929755132327, | |
| "eval_combined_score": 0.8347857688543592, | |
| "eval_f1": 0.8107785621954857, | |
| "eval_loss": 0.4443197548389435, | |
| "eval_runtime": 12.4932, | |
| "eval_samples_per_second": 3236.159, | |
| "eval_steps_per_second": 12.647, | |
| "step": 14220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 4.5569281578063965, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.1246, | |
| "step": 15642 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8566658421963889, | |
| "eval_combined_score": 0.8329733686874654, | |
| "eval_f1": 0.8092808951785421, | |
| "eval_loss": 0.46776074171066284, | |
| "eval_runtime": 12.4239, | |
| "eval_samples_per_second": 3254.203, | |
| "eval_steps_per_second": 12.717, | |
| "step": 15642 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "step": 15642, | |
| "total_flos": 1.0495478401912627e+17, | |
| "train_loss": 0.2538350579257268, | |
| "train_runtime": 2291.1306, | |
| "train_samples_per_second": 7940.316, | |
| "train_steps_per_second": 31.033 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 71100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 5 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0495478401912627e+17, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |