| { |
| "best_global_step": 4906, |
| "best_metric": 0.8757205622114191, |
| "best_model_checkpoint": "outputs/mbert-spam-binary/checkpoint-4906", |
| "epoch": 11.0, |
| "eval_steps": 500, |
| "global_step": 4906, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.5657051801681519, |
| "learning_rate": 4.998771926094749e-05, |
| "loss": 0.3358, |
| "step": 446 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8972904851921865, |
| "eval_f1": 0.8565157558602947, |
| "eval_loss": 0.3799394369125366, |
| "eval_precision": 0.9093794413813263, |
| "eval_recall": 0.8265391650419087, |
| "eval_runtime": 1.8927, |
| "eval_samples_per_second": 838.483, |
| "eval_steps_per_second": 26.417, |
| "step": 446 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 9.706869125366211, |
| "learning_rate": 4.995077872180951e-05, |
| "loss": 0.239, |
| "step": 892 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8853182104599874, |
| "eval_f1": 0.8486677201394943, |
| "eval_loss": 0.3240993022918701, |
| "eval_precision": 0.86617355096017, |
| "eval_recall": 0.8351329621515146, |
| "eval_runtime": 1.8802, |
| "eval_samples_per_second": 844.062, |
| "eval_steps_per_second": 26.593, |
| "step": 892 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 5.252721309661865, |
| "learning_rate": 4.988921477641713e-05, |
| "loss": 0.1874, |
| "step": 1338 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8859483301827347, |
| "eval_f1": 0.8551266047827633, |
| "eval_loss": 0.31036534905433655, |
| "eval_precision": 0.8564533765313355, |
| "eval_recall": 0.853831078772286, |
| "eval_runtime": 1.8827, |
| "eval_samples_per_second": 842.955, |
| "eval_steps_per_second": 26.558, |
| "step": 1338 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 7.976099967956543, |
| "learning_rate": 4.980308818095173e-05, |
| "loss": 0.1425, |
| "step": 1784 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8645242596093258, |
| "eval_f1": 0.833393799625288, |
| "eval_loss": 0.3800713121891022, |
| "eval_precision": 0.8251404013423738, |
| "eval_recall": 0.8435217382565174, |
| "eval_runtime": 1.8889, |
| "eval_samples_per_second": 840.172, |
| "eval_steps_per_second": 26.47, |
| "step": 1784 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 19.17702293395996, |
| "learning_rate": 4.9692483931964865e-05, |
| "loss": 0.1182, |
| "step": 2230 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8928796471329553, |
| "eval_f1": 0.8623913575713062, |
| "eval_loss": 0.38209661841392517, |
| "eval_precision": 0.8681797459435936, |
| "eval_recall": 0.857123474905027, |
| "eval_runtime": 1.891, |
| "eval_samples_per_second": 839.244, |
| "eval_steps_per_second": 26.441, |
| "step": 2230 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 0.38656169176101685, |
| "learning_rate": 4.9557511182496865e-05, |
| "loss": 0.0966, |
| "step": 2676 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.9042218021424071, |
| "eval_f1": 0.8754265647593472, |
| "eval_loss": 0.439815878868103, |
| "eval_precision": 0.8869994596227571, |
| "eval_recall": 0.8656328516009728, |
| "eval_runtime": 1.889, |
| "eval_samples_per_second": 840.135, |
| "eval_steps_per_second": 26.469, |
| "step": 2676 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 15.262068748474121, |
| "learning_rate": 4.939830313435599e-05, |
| "loss": 0.0787, |
| "step": 3122 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8985507246376812, |
| "eval_f1": 0.8709438719717468, |
| "eval_loss": 0.500139594078064, |
| "eval_precision": 0.8729009773098388, |
| "eval_recall": 0.8690498683443548, |
| "eval_runtime": 1.8913, |
| "eval_samples_per_second": 839.098, |
| "eval_steps_per_second": 26.437, |
| "step": 3122 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 2.0876944065093994, |
| "learning_rate": 4.9215016906664344e-05, |
| "loss": 0.0827, |
| "step": 3568 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8979206049149339, |
| "eval_f1": 0.8670210785043945, |
| "eval_loss": 0.4308617115020752, |
| "eval_precision": 0.8790773631628384, |
| "eval_recall": 0.8569274989447448, |
| "eval_runtime": 1.8901, |
| "eval_samples_per_second": 839.645, |
| "eval_steps_per_second": 26.454, |
| "step": 3568 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 0.03602065145969391, |
| "learning_rate": 4.900783338080024e-05, |
| "loss": 0.0695, |
| "step": 4014 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.9010712035286704, |
| "eval_f1": 0.8671410632063625, |
| "eval_loss": 0.650521457195282, |
| "eval_precision": 0.8951755990488386, |
| "eval_recall": 0.847398042250407, |
| "eval_runtime": 1.8893, |
| "eval_samples_per_second": 840.014, |
| "eval_steps_per_second": 26.465, |
| "step": 4014 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.15067948400974274, |
| "learning_rate": 4.877695702189014e-05, |
| "loss": 0.0551, |
| "step": 4460 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.8821676118462508, |
| "eval_f1": 0.8511983345308572, |
| "eval_loss": 0.53841632604599, |
| "eval_precision": 0.8504388791271249, |
| "eval_recall": 0.8519688046471428, |
| "eval_runtime": 1.8908, |
| "eval_samples_per_second": 839.349, |
| "eval_steps_per_second": 26.445, |
| "step": 4460 |
| }, |
| { |
| "epoch": 11.0, |
| "grad_norm": 0.37308648228645325, |
| "learning_rate": 4.852261567702625e-05, |
| "loss": 0.0544, |
| "step": 4906 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_accuracy": 0.9029615626969124, |
| "eval_f1": 0.8757205622114191, |
| "eval_loss": 0.5222230553627014, |
| "eval_precision": 0.8804347826086957, |
| "eval_recall": 0.8713442945870435, |
| "eval_runtime": 1.8866, |
| "eval_samples_per_second": 841.185, |
| "eval_steps_per_second": 26.502, |
| "step": 4906 |
| } |
| ], |
| "logging_steps": 446, |
| "max_steps": 44600, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 100, |
| "save_steps": 500, |
| "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": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
