| { |
| "best_metric": 0.6658675670623779, |
| "best_model_checkpoint": "bert_uncased_L-4_H-128_A-2_rte/checkpoint-130", |
| "epoch": 18.0, |
| "eval_steps": 500, |
| "global_step": 180, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.7593374848365784, |
| "learning_rate": 4.9e-05, |
| "loss": 0.6967, |
| "step": 10 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.5523465703971119, |
| "eval_loss": 0.6906870603561401, |
| "eval_runtime": 0.1071, |
| "eval_samples_per_second": 2585.314, |
| "eval_steps_per_second": 18.667, |
| "step": 10 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.3941654860973358, |
| "learning_rate": 4.8e-05, |
| "loss": 0.689, |
| "step": 20 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.5740072202166066, |
| "eval_loss": 0.6874576807022095, |
| "eval_runtime": 0.1068, |
| "eval_samples_per_second": 2594.199, |
| "eval_steps_per_second": 18.731, |
| "step": 20 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.5720370411872864, |
| "learning_rate": 4.7e-05, |
| "loss": 0.6832, |
| "step": 30 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.5379061371841155, |
| "eval_loss": 0.6854128837585449, |
| "eval_runtime": 0.1131, |
| "eval_samples_per_second": 2448.71, |
| "eval_steps_per_second": 17.68, |
| "step": 30 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.5994385480880737, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.6776, |
| "step": 40 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.5631768953068592, |
| "eval_loss": 0.6808509230613708, |
| "eval_runtime": 0.1037, |
| "eval_samples_per_second": 2671.058, |
| "eval_steps_per_second": 19.286, |
| "step": 40 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 1.6622623205184937, |
| "learning_rate": 4.5e-05, |
| "loss": 0.6694, |
| "step": 50 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.5812274368231047, |
| "eval_loss": 0.6779599785804749, |
| "eval_runtime": 0.1085, |
| "eval_samples_per_second": 2553.422, |
| "eval_steps_per_second": 18.436, |
| "step": 50 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 0.6583753824234009, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.6599, |
| "step": 60 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.5740072202166066, |
| "eval_loss": 0.674864649772644, |
| "eval_runtime": 0.1088, |
| "eval_samples_per_second": 2545.065, |
| "eval_steps_per_second": 18.376, |
| "step": 60 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 1.0308715105056763, |
| "learning_rate": 4.3e-05, |
| "loss": 0.6469, |
| "step": 70 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.5992779783393501, |
| "eval_loss": 0.6725589632987976, |
| "eval_runtime": 0.1112, |
| "eval_samples_per_second": 2491.491, |
| "eval_steps_per_second": 17.989, |
| "step": 70 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 1.4242850542068481, |
| "learning_rate": 4.2e-05, |
| "loss": 0.6392, |
| "step": 80 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.5776173285198556, |
| "eval_loss": 0.6711769700050354, |
| "eval_runtime": 0.1088, |
| "eval_samples_per_second": 2545.528, |
| "eval_steps_per_second": 18.379, |
| "step": 80 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 1.1413791179656982, |
| "learning_rate": 4.1e-05, |
| "loss": 0.6221, |
| "step": 90 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.5884476534296029, |
| "eval_loss": 0.6682437658309937, |
| "eval_runtime": 0.1046, |
| "eval_samples_per_second": 2648.022, |
| "eval_steps_per_second": 19.119, |
| "step": 90 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 1.3455125093460083, |
| "learning_rate": 4e-05, |
| "loss": 0.6034, |
| "step": 100 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.5956678700361011, |
| "eval_loss": 0.6683706641197205, |
| "eval_runtime": 0.1082, |
| "eval_samples_per_second": 2559.176, |
| "eval_steps_per_second": 18.478, |
| "step": 100 |
| }, |
| { |
| "epoch": 11.0, |
| "grad_norm": 1.3709012269973755, |
| "learning_rate": 3.9000000000000006e-05, |
| "loss": 0.5867, |
| "step": 110 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_accuracy": 0.5992779783393501, |
| "eval_loss": 0.6717198491096497, |
| "eval_runtime": 0.1056, |
| "eval_samples_per_second": 2623.778, |
| "eval_steps_per_second": 18.944, |
| "step": 110 |
| }, |
| { |
| "epoch": 12.0, |
| "grad_norm": 1.4418270587921143, |
| "learning_rate": 3.8e-05, |
| "loss": 0.5686, |
| "step": 120 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_accuracy": 0.6064981949458483, |
| "eval_loss": 0.6689558625221252, |
| "eval_runtime": 0.1095, |
| "eval_samples_per_second": 2530.503, |
| "eval_steps_per_second": 18.271, |
| "step": 120 |
| }, |
| { |
| "epoch": 13.0, |
| "grad_norm": 1.5649785995483398, |
| "learning_rate": 3.7e-05, |
| "loss": 0.5596, |
| "step": 130 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_accuracy": 0.6173285198555957, |
| "eval_loss": 0.6658675670623779, |
| "eval_runtime": 0.104, |
| "eval_samples_per_second": 2663.27, |
| "eval_steps_per_second": 19.229, |
| "step": 130 |
| }, |
| { |
| "epoch": 14.0, |
| "grad_norm": 1.9856057167053223, |
| "learning_rate": 3.6e-05, |
| "loss": 0.5377, |
| "step": 140 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_accuracy": 0.6101083032490975, |
| "eval_loss": 0.6719772219657898, |
| "eval_runtime": 0.1022, |
| "eval_samples_per_second": 2709.77, |
| "eval_steps_per_second": 19.565, |
| "step": 140 |
| }, |
| { |
| "epoch": 15.0, |
| "grad_norm": 2.440310001373291, |
| "learning_rate": 3.5e-05, |
| "loss": 0.5249, |
| "step": 150 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_accuracy": 0.6028880866425993, |
| "eval_loss": 0.6820036768913269, |
| "eval_runtime": 0.1077, |
| "eval_samples_per_second": 2573.09, |
| "eval_steps_per_second": 18.578, |
| "step": 150 |
| }, |
| { |
| "epoch": 16.0, |
| "grad_norm": 2.5033390522003174, |
| "learning_rate": 3.4000000000000007e-05, |
| "loss": 0.5019, |
| "step": 160 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_accuracy": 0.6064981949458483, |
| "eval_loss": 0.6896188259124756, |
| "eval_runtime": 0.1034, |
| "eval_samples_per_second": 2679.436, |
| "eval_steps_per_second": 19.346, |
| "step": 160 |
| }, |
| { |
| "epoch": 17.0, |
| "grad_norm": 2.3062896728515625, |
| "learning_rate": 3.3e-05, |
| "loss": 0.492, |
| "step": 170 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_accuracy": 0.6028880866425993, |
| "eval_loss": 0.6977380514144897, |
| "eval_runtime": 0.1051, |
| "eval_samples_per_second": 2635.223, |
| "eval_steps_per_second": 19.027, |
| "step": 170 |
| }, |
| { |
| "epoch": 18.0, |
| "grad_norm": 2.3697409629821777, |
| "learning_rate": 3.2000000000000005e-05, |
| "loss": 0.4786, |
| "step": 180 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_accuracy": 0.6173285198555957, |
| "eval_loss": 0.702680766582489, |
| "eval_runtime": 0.1101, |
| "eval_samples_per_second": 2514.93, |
| "eval_steps_per_second": 18.158, |
| "step": 180 |
| }, |
| { |
| "epoch": 18.0, |
| "step": 180, |
| "total_flos": 55771099361280.0, |
| "train_loss": 0.6020836697684394, |
| "train_runtime": 23.8597, |
| "train_samples_per_second": 5218.001, |
| "train_steps_per_second": 20.956 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 500, |
| "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": 55771099361280.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
