| { | |
| "best_metric": 0.5798782706260681, | |
| "best_model_checkpoint": "bert_base_lda_100_v1_mrpc/checkpoint-45", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.1343425512313843, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6665, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_combined_score": 0.7136150234741783, | |
| "eval_f1": 0.76056338028169, | |
| "eval_loss": 0.6143702864646912, | |
| "eval_runtime": 0.2749, | |
| "eval_samples_per_second": 1484.233, | |
| "eval_steps_per_second": 7.276, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.3366286754608154, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.594, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7058823529411765, | |
| "eval_combined_score": 0.7505520979722948, | |
| "eval_f1": 0.795221843003413, | |
| "eval_loss": 0.5914976596832275, | |
| "eval_runtime": 0.2751, | |
| "eval_samples_per_second": 1482.998, | |
| "eval_steps_per_second": 7.27, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.057149052619934, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.575, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6911764705882353, | |
| "eval_combined_score": 0.7486651583710408, | |
| "eval_f1": 0.8061538461538462, | |
| "eval_loss": 0.5798782706260681, | |
| "eval_runtime": 0.2754, | |
| "eval_samples_per_second": 1481.593, | |
| "eval_steps_per_second": 7.263, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.827286958694458, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.5285, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7034313725490197, | |
| "eval_combined_score": 0.7510501288701837, | |
| "eval_f1": 0.7986688851913478, | |
| "eval_loss": 0.5992879867553711, | |
| "eval_runtime": 0.2877, | |
| "eval_samples_per_second": 1417.975, | |
| "eval_steps_per_second": 6.951, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.9875845909118652, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4277, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6764705882352942, | |
| "eval_combined_score": 0.715100965759438, | |
| "eval_f1": 0.753731343283582, | |
| "eval_loss": 0.655691385269165, | |
| "eval_runtime": 0.2814, | |
| "eval_samples_per_second": 1450.083, | |
| "eval_steps_per_second": 7.108, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.7185416221618652, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.3244, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.696078431372549, | |
| "eval_combined_score": 0.749313101036593, | |
| "eval_f1": 0.8025477707006369, | |
| "eval_loss": 0.8220387101173401, | |
| "eval_runtime": 0.2781, | |
| "eval_samples_per_second": 1467.238, | |
| "eval_steps_per_second": 7.192, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 4.188333988189697, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.2387, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6421568627450981, | |
| "eval_combined_score": 0.6838603862597672, | |
| "eval_f1": 0.7255639097744362, | |
| "eval_loss": 0.8551956415176392, | |
| "eval_runtime": 0.2771, | |
| "eval_samples_per_second": 1472.601, | |
| "eval_steps_per_second": 7.219, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.75746488571167, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.1466, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6691176470588235, | |
| "eval_combined_score": 0.7195673414169756, | |
| "eval_f1": 0.7700170357751278, | |
| "eval_loss": 1.0601282119750977, | |
| "eval_runtime": 0.2766, | |
| "eval_samples_per_second": 1475.248, | |
| "eval_steps_per_second": 7.232, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 120, | |
| "total_flos": 3860365404241920.0, | |
| "train_loss": 0.4376857062180837, | |
| "train_runtime": 71.3419, | |
| "train_samples_per_second": 2570.719, | |
| "train_steps_per_second": 10.513 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 750, | |
| "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": 3860365404241920.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |