| { | |
| "best_metric": 1.4602082967758179, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_logit_kd_data_aug_stsb_128/checkpoint-20144", | |
| "epoch": 13.0, | |
| "global_step": 32734, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.5444, | |
| "step": 2518 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_combined_score": 0.16758540917963677, | |
| "eval_loss": 1.4964674711227417, | |
| "eval_pearson": 0.15890705274633266, | |
| "eval_runtime": 3.4685, | |
| "eval_samples_per_second": 432.466, | |
| "eval_spearmanr": 0.17626376561294088, | |
| "eval_steps_per_second": 3.46, | |
| "step": 2518 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.3254, | |
| "step": 5036 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_combined_score": 0.1588004352211329, | |
| "eval_loss": 1.527620792388916, | |
| "eval_pearson": 0.1501969703663422, | |
| "eval_runtime": 3.3223, | |
| "eval_samples_per_second": 451.489, | |
| "eval_spearmanr": 0.1674039000759236, | |
| "eval_steps_per_second": 3.612, | |
| "step": 5036 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.2847, | |
| "step": 7554 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_combined_score": 0.16338999996123976, | |
| "eval_loss": 1.543026328086853, | |
| "eval_pearson": 0.1587361207577816, | |
| "eval_runtime": 3.3209, | |
| "eval_samples_per_second": 451.687, | |
| "eval_spearmanr": 0.16804387916469793, | |
| "eval_steps_per_second": 3.613, | |
| "step": 7554 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.2376, | |
| "step": 10072 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_combined_score": 0.17277920267113558, | |
| "eval_loss": 1.6905953884124756, | |
| "eval_pearson": 0.16692030535267377, | |
| "eval_runtime": 3.3272, | |
| "eval_samples_per_second": 450.836, | |
| "eval_spearmanr": 0.1786380999895974, | |
| "eval_steps_per_second": 3.607, | |
| "step": 10072 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.1741, | |
| "step": 12590 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_combined_score": 0.16936077197686344, | |
| "eval_loss": 1.478766918182373, | |
| "eval_pearson": 0.16624858700331596, | |
| "eval_runtime": 3.3088, | |
| "eval_samples_per_second": 453.334, | |
| "eval_spearmanr": 0.17247295695041093, | |
| "eval_steps_per_second": 3.627, | |
| "step": 12590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.1315, | |
| "step": 15108 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_combined_score": 0.167026990919993, | |
| "eval_loss": 1.5661702156066895, | |
| "eval_pearson": 0.16400657638338206, | |
| "eval_runtime": 3.3102, | |
| "eval_samples_per_second": 453.145, | |
| "eval_spearmanr": 0.17004740545660393, | |
| "eval_steps_per_second": 3.625, | |
| "step": 15108 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.1055, | |
| "step": 17626 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_combined_score": 0.16801853421730611, | |
| "eval_loss": 1.5099858045578003, | |
| "eval_pearson": 0.1662656311397418, | |
| "eval_runtime": 3.4068, | |
| "eval_samples_per_second": 440.294, | |
| "eval_spearmanr": 0.16977143729487043, | |
| "eval_steps_per_second": 3.522, | |
| "step": 17626 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.0879, | |
| "step": 20144 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_combined_score": 0.15889813071988862, | |
| "eval_loss": 1.4602082967758179, | |
| "eval_pearson": 0.1595602474351447, | |
| "eval_runtime": 3.4014, | |
| "eval_samples_per_second": 440.991, | |
| "eval_spearmanr": 0.15823601400463258, | |
| "eval_steps_per_second": 3.528, | |
| "step": 20144 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.0739, | |
| "step": 22662 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_combined_score": 0.16028059024365865, | |
| "eval_loss": 1.6611720323562622, | |
| "eval_pearson": 0.15843700113214274, | |
| "eval_runtime": 3.3106, | |
| "eval_samples_per_second": 453.089, | |
| "eval_spearmanr": 0.16212417935517454, | |
| "eval_steps_per_second": 3.625, | |
| "step": 22662 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0632, | |
| "step": 25180 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_combined_score": 0.15180517392430187, | |
| "eval_loss": 1.5825084447860718, | |
| "eval_pearson": 0.1489430437396317, | |
| "eval_runtime": 3.335, | |
| "eval_samples_per_second": 449.777, | |
| "eval_spearmanr": 0.1546673041089721, | |
| "eval_steps_per_second": 3.598, | |
| "step": 25180 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.0548, | |
| "step": 27698 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_combined_score": 0.1440623890943714, | |
| "eval_loss": 1.5946108102798462, | |
| "eval_pearson": 0.1420620866419895, | |
| "eval_runtime": 3.3218, | |
| "eval_samples_per_second": 451.567, | |
| "eval_spearmanr": 0.1460626915467533, | |
| "eval_steps_per_second": 3.613, | |
| "step": 27698 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.0473, | |
| "step": 30216 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_combined_score": 0.15371309009042938, | |
| "eval_loss": 1.6514661312103271, | |
| "eval_pearson": 0.1526314720694863, | |
| "eval_runtime": 3.4209, | |
| "eval_samples_per_second": 438.481, | |
| "eval_spearmanr": 0.15479470811137244, | |
| "eval_steps_per_second": 3.508, | |
| "step": 30216 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.0415, | |
| "step": 32734 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_combined_score": 0.14916750439198037, | |
| "eval_loss": 1.6544198989868164, | |
| "eval_pearson": 0.15058122723353334, | |
| "eval_runtime": 3.3126, | |
| "eval_samples_per_second": 452.819, | |
| "eval_spearmanr": 0.1477537815504274, | |
| "eval_steps_per_second": 3.623, | |
| "step": 32734 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "step": 32734, | |
| "total_flos": 9.984731610139853e+16, | |
| "train_loss": 0.16706499188415525, | |
| "train_runtime": 23134.1323, | |
| "train_samples_per_second": 696.475, | |
| "train_steps_per_second": 5.442 | |
| } | |
| ], | |
| "max_steps": 125900, | |
| "num_train_epochs": 50, | |
| "total_flos": 9.984731610139853e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |