| { | |
| "best_metric": 0.5352208614349365, | |
| "best_model_checkpoint": "bert_uncased_L-2_H-128_A-2_mrpc/checkpoint-180", | |
| "epoch": 17.0, | |
| "eval_steps": 500, | |
| "global_step": 255, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.5705815553665161, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6493, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6226639747619629, | |
| "eval_runtime": 0.1447, | |
| "eval_samples_per_second": 2820.109, | |
| "eval_steps_per_second": 13.824, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.7753280401229858, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6257, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.6134153008460999, | |
| "eval_runtime": 0.1474, | |
| "eval_samples_per_second": 2767.183, | |
| "eval_steps_per_second": 13.565, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.4908242225646973, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6126, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6838235294117647, | |
| "eval_combined_score": 0.7480253018237863, | |
| "eval_f1": 0.8122270742358079, | |
| "eval_loss": 0.605224609375, | |
| "eval_runtime": 0.1429, | |
| "eval_samples_per_second": 2855.034, | |
| "eval_steps_per_second": 13.995, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.7324304580688477, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6036, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.696078431372549, | |
| "eval_combined_score": 0.7568627450980392, | |
| "eval_f1": 0.8176470588235294, | |
| "eval_loss": 0.5954302549362183, | |
| "eval_runtime": 0.1406, | |
| "eval_samples_per_second": 2902.293, | |
| "eval_steps_per_second": 14.227, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.0118967294692993, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5897, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6985294117647058, | |
| "eval_combined_score": 0.7576104584904007, | |
| "eval_f1": 0.8166915052160953, | |
| "eval_loss": 0.5878595113754272, | |
| "eval_runtime": 0.1425, | |
| "eval_samples_per_second": 2863.734, | |
| "eval_steps_per_second": 14.038, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.994917631149292, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5781, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7034313725490197, | |
| "eval_combined_score": 0.7596304503536575, | |
| "eval_f1": 0.8158295281582952, | |
| "eval_loss": 0.5740991234779358, | |
| "eval_runtime": 0.1456, | |
| "eval_samples_per_second": 2801.559, | |
| "eval_steps_per_second": 13.733, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.9900652170181274, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.5635, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7107843137254902, | |
| "eval_combined_score": 0.7654531324725012, | |
| "eval_f1": 0.8201219512195123, | |
| "eval_loss": 0.5711214542388916, | |
| "eval_runtime": 0.1431, | |
| "eval_samples_per_second": 2850.896, | |
| "eval_steps_per_second": 13.975, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.3341139554977417, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.5429, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7132352941176471, | |
| "eval_combined_score": 0.7670311233222232, | |
| "eval_f1": 0.8208269525267994, | |
| "eval_loss": 0.5674187541007996, | |
| "eval_runtime": 0.1449, | |
| "eval_samples_per_second": 2814.798, | |
| "eval_steps_per_second": 13.798, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.9501793384552002, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.5228, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7205882352941176, | |
| "eval_combined_score": 0.7728708047636232, | |
| "eval_f1": 0.8251533742331287, | |
| "eval_loss": 0.5684576034545898, | |
| "eval_runtime": 0.1421, | |
| "eval_samples_per_second": 2872.015, | |
| "eval_steps_per_second": 14.079, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.7963290214538574, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5057, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7303921568627451, | |
| "eval_combined_score": 0.7792585784313726, | |
| "eval_f1": 0.828125, | |
| "eval_loss": 0.5496610403060913, | |
| "eval_runtime": 0.1412, | |
| "eval_samples_per_second": 2890.327, | |
| "eval_steps_per_second": 14.168, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.0463414192199707, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.4856, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7377450980392157, | |
| "eval_combined_score": 0.7835455952716015, | |
| "eval_f1": 0.8293460925039873, | |
| "eval_loss": 0.5438232421875, | |
| "eval_runtime": 0.1389, | |
| "eval_samples_per_second": 2936.684, | |
| "eval_steps_per_second": 14.396, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 2.688190221786499, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.4657, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7328431372549019, | |
| "eval_combined_score": 0.7780909365366891, | |
| "eval_f1": 0.8233387358184764, | |
| "eval_loss": 0.5352208614349365, | |
| "eval_runtime": 0.1404, | |
| "eval_samples_per_second": 2906.764, | |
| "eval_steps_per_second": 14.249, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 2.253540515899658, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.4447, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7401960784313726, | |
| "eval_combined_score": 0.7862372797220154, | |
| "eval_f1": 0.8322784810126583, | |
| "eval_loss": 0.5435168743133545, | |
| "eval_runtime": 0.1362, | |
| "eval_samples_per_second": 2995.973, | |
| "eval_steps_per_second": 14.686, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 4.5999555587768555, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.4175, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7401960784313726, | |
| "eval_combined_score": 0.7865018247046454, | |
| "eval_f1": 0.832807570977918, | |
| "eval_loss": 0.5561667084693909, | |
| "eval_runtime": 0.1498, | |
| "eval_samples_per_second": 2723.141, | |
| "eval_steps_per_second": 13.349, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 3.1501963138580322, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.4039, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.7426470588235294, | |
| "eval_combined_score": 0.7891639049986192, | |
| "eval_f1": 0.8356807511737089, | |
| "eval_loss": 0.5759110450744629, | |
| "eval_runtime": 0.1413, | |
| "eval_samples_per_second": 2886.588, | |
| "eval_steps_per_second": 14.15, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 4.6777119636535645, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.3964, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7377450980392157, | |
| "eval_combined_score": 0.7838169051404346, | |
| "eval_f1": 0.8298887122416534, | |
| "eval_loss": 0.5610219836235046, | |
| "eval_runtime": 0.1417, | |
| "eval_samples_per_second": 2880.029, | |
| "eval_steps_per_second": 14.118, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 4.6045122146606445, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.3735, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.7377450980392157, | |
| "eval_combined_score": 0.7829977496616625, | |
| "eval_f1": 0.8282504012841091, | |
| "eval_loss": 0.5587014555931091, | |
| "eval_runtime": 0.1469, | |
| "eval_samples_per_second": 2777.401, | |
| "eval_steps_per_second": 13.615, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "step": 255, | |
| "total_flos": 39611244933120.0, | |
| "train_loss": 0.5165351082296933, | |
| "train_runtime": 28.2797, | |
| "train_samples_per_second": 6485.214, | |
| "train_steps_per_second": 26.521 | |
| } | |
| ], | |
| "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": 39611244933120.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |