| { | |
| "best_metric": 0.6140850782394409, | |
| "best_model_checkpoint": "bert_base_lda_5_v1_cola/checkpoint-102", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 272, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.8962923884391785, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6165, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6255748867988586, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.6921, | |
| "eval_samples_per_second": 1507.045, | |
| "eval_steps_per_second": 7.225, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.1500860452651978, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6085, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6922339200973511, | |
| "eval_loss": 0.6206546425819397, | |
| "eval_matthews_correlation": 0.0437601222642778, | |
| "eval_runtime": 0.6888, | |
| "eval_samples_per_second": 1514.269, | |
| "eval_steps_per_second": 7.259, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.3318846225738525, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.5767, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6788111329078674, | |
| "eval_loss": 0.6140850782394409, | |
| "eval_matthews_correlation": 0.014335296446004425, | |
| "eval_runtime": 0.6907, | |
| "eval_samples_per_second": 1510.046, | |
| "eval_steps_per_second": 7.239, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.0688295364379883, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.5249, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6625120043754578, | |
| "eval_loss": 0.6585533022880554, | |
| "eval_matthews_correlation": 0.05998312692220892, | |
| "eval_runtime": 0.7395, | |
| "eval_samples_per_second": 1410.335, | |
| "eval_steps_per_second": 6.761, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.6873955726623535, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4572, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6519654989242554, | |
| "eval_loss": 0.7236359715461731, | |
| "eval_matthews_correlation": 0.13356139258109112, | |
| "eval_runtime": 0.6908, | |
| "eval_samples_per_second": 1509.857, | |
| "eval_steps_per_second": 7.238, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.0232486724853516, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.3844, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.640460193157196, | |
| "eval_loss": 0.8147491812705994, | |
| "eval_matthews_correlation": 0.08177474609901102, | |
| "eval_runtime": 0.6894, | |
| "eval_samples_per_second": 1513.007, | |
| "eval_steps_per_second": 7.253, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.7721943855285645, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.327, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6126558184623718, | |
| "eval_loss": 0.9067552089691162, | |
| "eval_matthews_correlation": 0.08937795209450178, | |
| "eval_runtime": 0.684, | |
| "eval_samples_per_second": 1524.939, | |
| "eval_steps_per_second": 7.31, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.930617332458496, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.2756, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6423777341842651, | |
| "eval_loss": 0.9654455184936523, | |
| "eval_matthews_correlation": 0.12007934236570234, | |
| "eval_runtime": 0.6807, | |
| "eval_samples_per_second": 1532.247, | |
| "eval_steps_per_second": 7.345, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 272, | |
| "total_flos": 8999450537533440.0, | |
| "train_loss": 0.47134509156731996, | |
| "train_runtime": 138.7221, | |
| "train_samples_per_second": 3082.061, | |
| "train_steps_per_second": 12.255 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 1700, | |
| "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": 8999450537533440.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |