{ "best_metric": 0.18365982174873352, "best_model_checkpoint": "thermo-predictor-thermo-evotuning-prot_bert/checkpoint-58", "epoch": 28.677248677248677, "global_step": 58, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.68, "learning_rate": 1.9600000000000002e-05, "loss": 0.4734, "step": 2 }, { "epoch": 0.68, "eval_loss": 0.31455937027931213, "eval_runtime": 2.9218, "eval_samples_per_second": 1835.142, "eval_spearmanr": 0.3358911230016707, "eval_steps_per_second": 7.187, "step": 2 }, { "epoch": 1.68, "learning_rate": 1.9200000000000003e-05, "loss": 0.4392, "step": 4 }, { "epoch": 1.68, "eval_loss": 0.29363012313842773, "eval_runtime": 2.904, "eval_samples_per_second": 1846.417, "eval_spearmanr": 0.34070735481419273, "eval_steps_per_second": 7.231, "step": 4 }, { "epoch": 2.68, "learning_rate": 1.88e-05, "loss": 0.4034, "step": 6 }, { "epoch": 2.68, "eval_loss": 0.2633129358291626, "eval_runtime": 3.0574, "eval_samples_per_second": 1753.755, "eval_spearmanr": 0.36959161893964504, "eval_steps_per_second": 6.868, "step": 6 }, { "epoch": 3.68, "learning_rate": 1.8400000000000003e-05, "loss": 0.3669, "step": 8 }, { "epoch": 3.68, "eval_loss": 0.24365590512752533, "eval_runtime": 2.9265, "eval_samples_per_second": 1832.235, "eval_spearmanr": 0.390325257804677, "eval_steps_per_second": 7.176, "step": 8 }, { "epoch": 4.68, "learning_rate": 1.8e-05, "loss": 0.3496, "step": 10 }, { "epoch": 4.68, "eval_loss": 0.23773197829723358, "eval_runtime": 2.9224, "eval_samples_per_second": 1834.793, "eval_spearmanr": 0.41020718084494817, "eval_steps_per_second": 7.186, "step": 10 }, { "epoch": 5.68, "learning_rate": 1.76e-05, "loss": 0.3351, "step": 12 }, { "epoch": 5.68, "eval_loss": 0.22854185104370117, "eval_runtime": 3.4499, "eval_samples_per_second": 1554.251, "eval_spearmanr": 0.42040641918816646, "eval_steps_per_second": 6.087, "step": 12 }, { "epoch": 6.68, "learning_rate": 1.72e-05, "loss": 0.3289, "step": 14 }, { "epoch": 6.68, "eval_loss": 0.2267082780599594, "eval_runtime": 3.0431, "eval_samples_per_second": 1762.044, "eval_spearmanr": 0.41796639021828647, "eval_steps_per_second": 6.901, "step": 14 }, { "epoch": 7.68, "learning_rate": 1.6800000000000002e-05, "loss": 0.3267, "step": 16 }, { "epoch": 7.68, "eval_loss": 0.22581231594085693, "eval_runtime": 3.1089, "eval_samples_per_second": 1724.753, "eval_spearmanr": 0.4242454004002921, "eval_steps_per_second": 6.755, "step": 16 }, { "epoch": 8.68, "learning_rate": 1.64e-05, "loss": 0.3177, "step": 18 }, { "epoch": 8.68, "eval_loss": 0.2206145077943802, "eval_runtime": 2.9131, "eval_samples_per_second": 1840.623, "eval_spearmanr": 0.42951271566156707, "eval_steps_per_second": 7.209, "step": 18 }, { "epoch": 9.68, "learning_rate": 1.6000000000000003e-05, "loss": 0.3116, "step": 20 }, { "epoch": 9.68, "eval_loss": 0.21502549946308136, "eval_runtime": 2.8807, "eval_samples_per_second": 1861.38, "eval_spearmanr": 0.43652182698224073, "eval_steps_per_second": 7.29, "step": 20 }, { "epoch": 10.68, "learning_rate": 1.5600000000000003e-05, "loss": 0.3039, "step": 22 }, { "epoch": 10.68, "eval_loss": 0.21152722835540771, "eval_runtime": 2.8687, "eval_samples_per_second": 1869.16, "eval_spearmanr": 0.4364973861642812, "eval_steps_per_second": 7.32, "step": 22 }, { "epoch": 11.68, "learning_rate": 1.5200000000000002e-05, "loss": 0.2985, "step": 24 }, { "epoch": 11.68, "eval_loss": 0.20623379945755005, "eval_runtime": 2.9943, "eval_samples_per_second": 1790.717, "eval_spearmanr": 0.4469415528786085, "eval_steps_per_second": 7.013, "step": 24 }, { "epoch": 12.68, "learning_rate": 1.48e-05, "loss": 0.2927, "step": 26 }, { "epoch": 12.68, "eval_loss": 0.20452716946601868, "eval_runtime": 3.0404, "eval_samples_per_second": 1763.602, "eval_spearmanr": 0.45307896545734583, "eval_steps_per_second": 6.907, "step": 26 }, { "epoch": 13.68, "learning_rate": 1.4400000000000001e-05, "loss": 0.2885, "step": 28 }, { "epoch": 13.68, "eval_loss": 0.2004762440919876, "eval_runtime": 2.8984, "eval_samples_per_second": 1849.957, "eval_spearmanr": 0.4602718574665676, "eval_steps_per_second": 7.245, "step": 28 }, { "epoch": 14.68, "learning_rate": 1.4e-05, "loss": 0.2838, "step": 30 }, { "epoch": 14.68, "eval_loss": 0.1986912339925766, "eval_runtime": 3.0267, "eval_samples_per_second": 1771.545, "eval_spearmanr": 0.46902799622360686, "eval_steps_per_second": 6.938, "step": 30 }, { "epoch": 15.68, "learning_rate": 1.3600000000000002e-05, "loss": 0.2806, "step": 32 }, { "epoch": 15.68, "eval_loss": 0.19752489030361176, "eval_runtime": 2.9976, "eval_samples_per_second": 1788.761, "eval_spearmanr": 0.47439420407286187, "eval_steps_per_second": 7.006, "step": 32 }, { "epoch": 16.68, "learning_rate": 1.3200000000000002e-05, "loss": 0.2772, "step": 34 }, { "epoch": 16.68, "eval_loss": 0.19701921939849854, "eval_runtime": 2.9895, "eval_samples_per_second": 1793.634, "eval_spearmanr": 0.47652315821607377, "eval_steps_per_second": 7.025, "step": 34 }, { "epoch": 17.68, "learning_rate": 1.2800000000000001e-05, "loss": 0.2728, "step": 36 }, { "epoch": 17.68, "eval_loss": 0.19386564195156097, "eval_runtime": 3.2955, "eval_samples_per_second": 1627.048, "eval_spearmanr": 0.48447898122969435, "eval_steps_per_second": 6.372, "step": 36 }, { "epoch": 18.68, "learning_rate": 1.2400000000000002e-05, "loss": 0.2684, "step": 38 }, { "epoch": 18.68, "eval_loss": 0.19311943650245667, "eval_runtime": 2.9977, "eval_samples_per_second": 1788.676, "eval_spearmanr": 0.4858414916440366, "eval_steps_per_second": 7.005, "step": 38 }, { "epoch": 19.68, "learning_rate": 1.2e-05, "loss": 0.2641, "step": 40 }, { "epoch": 19.68, "eval_loss": 0.19249317049980164, "eval_runtime": 3.1202, "eval_samples_per_second": 1718.506, "eval_spearmanr": 0.4935711457073984, "eval_steps_per_second": 6.73, "step": 40 }, { "epoch": 20.68, "learning_rate": 1.16e-05, "loss": 0.2608, "step": 42 }, { "epoch": 20.68, "eval_loss": 0.19051019847393036, "eval_runtime": 3.0142, "eval_samples_per_second": 1778.905, "eval_spearmanr": 0.4929029010849903, "eval_steps_per_second": 6.967, "step": 42 }, { "epoch": 21.68, "learning_rate": 1.1200000000000001e-05, "loss": 0.2566, "step": 44 }, { "epoch": 21.68, "eval_loss": 0.18856459856033325, "eval_runtime": 2.9879, "eval_samples_per_second": 1794.551, "eval_spearmanr": 0.5048595718357143, "eval_steps_per_second": 7.028, "step": 44 }, { "epoch": 22.68, "learning_rate": 1.0800000000000002e-05, "loss": 0.2518, "step": 46 }, { "epoch": 22.68, "eval_loss": 0.18753479421138763, "eval_runtime": 3.2412, "eval_samples_per_second": 1654.33, "eval_spearmanr": 0.5095133606919371, "eval_steps_per_second": 6.479, "step": 46 }, { "epoch": 23.68, "learning_rate": 1.04e-05, "loss": 0.2467, "step": 48 }, { "epoch": 23.68, "eval_loss": 0.18694807589054108, "eval_runtime": 3.6439, "eval_samples_per_second": 1471.507, "eval_spearmanr": 0.5140557788640083, "eval_steps_per_second": 5.763, "step": 48 }, { "epoch": 24.68, "learning_rate": 1e-05, "loss": 0.2424, "step": 50 }, { "epoch": 24.68, "eval_loss": 0.18594887852668762, "eval_runtime": 3.0425, "eval_samples_per_second": 1762.352, "eval_spearmanr": 0.5161498654457386, "eval_steps_per_second": 6.902, "step": 50 }, { "epoch": 25.68, "learning_rate": 9.600000000000001e-06, "loss": 0.2375, "step": 52 }, { "epoch": 25.68, "eval_loss": 0.1849762201309204, "eval_runtime": 2.8745, "eval_samples_per_second": 1865.365, "eval_spearmanr": 0.5223331654258908, "eval_steps_per_second": 7.306, "step": 52 }, { "epoch": 26.68, "learning_rate": 9.200000000000002e-06, "loss": 0.2329, "step": 54 }, { "epoch": 26.68, "eval_loss": 0.18514755368232727, "eval_runtime": 2.8748, "eval_samples_per_second": 1865.157, "eval_spearmanr": 0.5210178115492534, "eval_steps_per_second": 7.305, "step": 54 }, { "epoch": 27.68, "learning_rate": 8.8e-06, "loss": 0.2279, "step": 56 }, { "epoch": 27.68, "eval_loss": 0.18495012819766998, "eval_runtime": 2.9179, "eval_samples_per_second": 1837.612, "eval_spearmanr": 0.5294199372441676, "eval_steps_per_second": 7.197, "step": 56 }, { "epoch": 28.68, "learning_rate": 8.400000000000001e-06, "loss": 0.2226, "step": 58 }, { "epoch": 28.68, "eval_loss": 0.18365982174873352, "eval_runtime": 2.8753, "eval_samples_per_second": 1864.829, "eval_spearmanr": 0.5309868099654046, "eval_steps_per_second": 7.304, "step": 58 } ], "max_steps": 100, "num_train_epochs": 50, "total_flos": 1.6543891233169373e+17, "trial_name": null, "trial_params": null }