| { | |
| "best_metric": 2.2965381145477295, | |
| "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_stsb_256/checkpoint-135", | |
| "epoch": 8.0, | |
| "global_step": 360, | |
| "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": 4.3289, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_combined_score": 0.016869338769549977, | |
| "eval_loss": 2.5185155868530273, | |
| "eval_pearson": 0.015152034772122189, | |
| "eval_runtime": 2.6668, | |
| "eval_samples_per_second": 562.482, | |
| "eval_spearmanr": 0.018586642766977766, | |
| "eval_steps_per_second": 4.5, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 2.1532, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_combined_score": 0.031155825147051168, | |
| "eval_loss": 2.5569276809692383, | |
| "eval_pearson": 0.03878201425375371, | |
| "eval_runtime": 2.6592, | |
| "eval_samples_per_second": 564.09, | |
| "eval_spearmanr": 0.023529636040348627, | |
| "eval_steps_per_second": 4.513, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 2.1161, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_combined_score": 0.030822169467473876, | |
| "eval_loss": 2.2965381145477295, | |
| "eval_pearson": 0.039909131733008196, | |
| "eval_runtime": 2.7354, | |
| "eval_samples_per_second": 548.375, | |
| "eval_spearmanr": 0.02173520720193955, | |
| "eval_steps_per_second": 4.387, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 2.1083, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_combined_score": 0.034664425266159364, | |
| "eval_loss": 2.4191055297851562, | |
| "eval_pearson": 0.0409414178434408, | |
| "eval_runtime": 2.6813, | |
| "eval_samples_per_second": 559.437, | |
| "eval_spearmanr": 0.028387432688877935, | |
| "eval_steps_per_second": 4.475, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 2.0885, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_combined_score": 0.04204090753651958, | |
| "eval_loss": 2.5610482692718506, | |
| "eval_pearson": 0.0458438458890056, | |
| "eval_runtime": 2.7067, | |
| "eval_samples_per_second": 554.172, | |
| "eval_spearmanr": 0.03823796918403356, | |
| "eval_steps_per_second": 4.433, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 2.0602, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_combined_score": 0.0458476409316318, | |
| "eval_loss": 2.378797769546509, | |
| "eval_pearson": 0.047272793574367096, | |
| "eval_runtime": 2.6789, | |
| "eval_samples_per_second": 559.941, | |
| "eval_spearmanr": 0.04442248828889651, | |
| "eval_steps_per_second": 4.48, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 2.0283, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_combined_score": 0.05492669237451248, | |
| "eval_loss": 2.4458703994750977, | |
| "eval_pearson": 0.05710597845528931, | |
| "eval_runtime": 2.6878, | |
| "eval_samples_per_second": 558.076, | |
| "eval_spearmanr": 0.05274740629373566, | |
| "eval_steps_per_second": 4.465, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 1.9677, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_combined_score": 0.06091007949651277, | |
| "eval_loss": 2.4995479583740234, | |
| "eval_pearson": 0.06144666895354363, | |
| "eval_runtime": 2.6805, | |
| "eval_samples_per_second": 559.594, | |
| "eval_spearmanr": 0.060373490039481906, | |
| "eval_steps_per_second": 4.477, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 360, | |
| "total_flos": 1185435401322496.0, | |
| "train_loss": 2.3563860363430447, | |
| "train_runtime": 347.9343, | |
| "train_samples_per_second": 826.162, | |
| "train_steps_per_second": 6.467 | |
| } | |
| ], | |
| "max_steps": 2250, | |
| "num_train_epochs": 50, | |
| "total_flos": 1185435401322496.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |