| { | |
| "best_metric": 2.8341832160949707, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_data_aug_stsb/checkpoint-7554", | |
| "epoch": 8.0, | |
| "global_step": 20144, | |
| "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": 1.0254, | |
| "step": 2518 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_combined_score": 0.16585328643275277, | |
| "eval_loss": 2.877626657485962, | |
| "eval_pearson": 0.15748718573093243, | |
| "eval_runtime": 1.5831, | |
| "eval_samples_per_second": 947.527, | |
| "eval_spearmanr": 0.17421938713457308, | |
| "eval_steps_per_second": 7.58, | |
| "step": 2518 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.5854, | |
| "step": 5036 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_combined_score": 0.1635100194118946, | |
| "eval_loss": 3.1463847160339355, | |
| "eval_pearson": 0.15907317751955238, | |
| "eval_runtime": 1.5769, | |
| "eval_samples_per_second": 951.23, | |
| "eval_spearmanr": 0.16794686130423683, | |
| "eval_steps_per_second": 7.61, | |
| "step": 5036 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.4255, | |
| "step": 7554 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_combined_score": 0.17824953346799088, | |
| "eval_loss": 2.8341832160949707, | |
| "eval_pearson": 0.17650415284397877, | |
| "eval_runtime": 1.6928, | |
| "eval_samples_per_second": 886.104, | |
| "eval_spearmanr": 0.17999491409200302, | |
| "eval_steps_per_second": 7.089, | |
| "step": 7554 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.2765, | |
| "step": 10072 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_combined_score": 0.18267513587644282, | |
| "eval_loss": 2.852358818054199, | |
| "eval_pearson": 0.18152000982174782, | |
| "eval_runtime": 1.5758, | |
| "eval_samples_per_second": 951.922, | |
| "eval_spearmanr": 0.18383026193113783, | |
| "eval_steps_per_second": 7.615, | |
| "step": 10072 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.1862, | |
| "step": 12590 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_combined_score": 0.17521614805651298, | |
| "eval_loss": 2.9183528423309326, | |
| "eval_pearson": 0.17358261728487295, | |
| "eval_runtime": 1.5754, | |
| "eval_samples_per_second": 952.151, | |
| "eval_spearmanr": 0.17684967882815297, | |
| "eval_steps_per_second": 7.617, | |
| "step": 12590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.1339, | |
| "step": 15108 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_combined_score": 0.17076120867812122, | |
| "eval_loss": 2.981703519821167, | |
| "eval_pearson": 0.16876644953461767, | |
| "eval_runtime": 1.5872, | |
| "eval_samples_per_second": 945.039, | |
| "eval_spearmanr": 0.17275596782162475, | |
| "eval_steps_per_second": 7.56, | |
| "step": 15108 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.1029, | |
| "step": 17626 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_combined_score": 0.16307581221558015, | |
| "eval_loss": 2.970160484313965, | |
| "eval_pearson": 0.1618052318006157, | |
| "eval_runtime": 1.5744, | |
| "eval_samples_per_second": 952.757, | |
| "eval_spearmanr": 0.16434639263054457, | |
| "eval_steps_per_second": 7.622, | |
| "step": 17626 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.0806, | |
| "step": 20144 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_combined_score": 0.1605832354040539, | |
| "eval_loss": 3.0033271312713623, | |
| "eval_pearson": 0.1587626578082464, | |
| "eval_runtime": 1.5753, | |
| "eval_samples_per_second": 952.217, | |
| "eval_spearmanr": 0.1624038129998614, | |
| "eval_steps_per_second": 7.618, | |
| "step": 20144 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 20144, | |
| "total_flos": 8.08285013719122e+16, | |
| "train_loss": 0.3520539346623743, | |
| "train_runtime": 9383.655, | |
| "train_samples_per_second": 1717.065, | |
| "train_steps_per_second": 13.417 | |
| } | |
| ], | |
| "max_steps": 125900, | |
| "num_train_epochs": 50, | |
| "total_flos": 8.08285013719122e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |