{ "best_metric": 62.53450232086453, "best_model_checkpoint": "/home/jcanete/ft-data/all_results/mlqa/albeto_base_4/epochs_2_bs_16_lr_5e-5/checkpoint-9600", "epoch": 2.0, "global_step": 10258, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.06, "eval_exact_match": 27.6, "eval_f1": 48.50197462974654, "step": 300 }, { "epoch": 0.1, "learning_rate": 4.757262624293235e-05, "loss": 3.1102, "step": 500 }, { "epoch": 0.12, "eval_exact_match": 30.4, "eval_f1": 52.22443824209194, "step": 600 }, { "epoch": 0.18, "eval_exact_match": 31.2, "eval_f1": 53.685799458846745, "step": 900 }, { "epoch": 0.19, "learning_rate": 4.5135503996880485e-05, "loss": 2.4551, "step": 1000 }, { "epoch": 0.23, "eval_exact_match": 33.6, "eval_f1": 57.14425867596316, "step": 1200 }, { "epoch": 0.29, "learning_rate": 4.2698381750828625e-05, "loss": 2.2732, "step": 1500 }, { "epoch": 0.29, "eval_exact_match": 32.4, "eval_f1": 55.85916754141139, "step": 1500 }, { "epoch": 0.35, "eval_exact_match": 33.0, "eval_f1": 57.430255458548, "step": 1800 }, { "epoch": 0.39, "learning_rate": 4.0261259504776764e-05, "loss": 2.2134, "step": 2000 }, { "epoch": 0.41, "eval_exact_match": 35.6, "eval_f1": 59.02849906136428, "step": 2100 }, { "epoch": 0.47, "eval_exact_match": 36.4, "eval_f1": 59.5327821091872, "step": 2400 }, { "epoch": 0.49, "learning_rate": 3.78241372587249e-05, "loss": 2.1589, "step": 2500 }, { "epoch": 0.53, "eval_exact_match": 36.2, "eval_f1": 59.46185724138048, "step": 2700 }, { "epoch": 0.58, "learning_rate": 3.5387015012673035e-05, "loss": 2.1618, "step": 3000 }, { "epoch": 0.58, "eval_exact_match": 35.8, "eval_f1": 59.952077287364006, "step": 3000 }, { "epoch": 0.64, "eval_exact_match": 35.0, "eval_f1": 58.35846229550928, "step": 3300 }, { "epoch": 0.68, "learning_rate": 3.2949892766621175e-05, "loss": 2.0937, "step": 3500 }, { "epoch": 0.7, "eval_exact_match": 37.6, "eval_f1": 60.19368918942722, "step": 3600 }, { "epoch": 0.76, "eval_exact_match": 35.6, "eval_f1": 60.702544015732805, "step": 3900 }, { "epoch": 0.78, "learning_rate": 3.051277052056931e-05, "loss": 2.0803, "step": 4000 }, { "epoch": 0.82, "eval_exact_match": 36.8, "eval_f1": 61.22453549563325, "step": 4200 }, { "epoch": 0.88, "learning_rate": 2.8080522519009555e-05, "loss": 2.0461, "step": 4500 }, { "epoch": 0.88, "eval_exact_match": 36.8, "eval_f1": 60.611829979451855, "step": 4500 }, { "epoch": 0.94, "eval_exact_match": 36.4, "eval_f1": 59.12225712872684, "step": 4800 }, { "epoch": 0.97, "learning_rate": 2.564340027295769e-05, "loss": 2.0302, "step": 5000 }, { "epoch": 0.99, "eval_exact_match": 38.2, "eval_f1": 60.23383357244435, "step": 5100 }, { "epoch": 1.05, "eval_exact_match": 37.4, "eval_f1": 59.583070357215874, "step": 5400 }, { "epoch": 1.07, "learning_rate": 2.320627802690583e-05, "loss": 1.7546, "step": 5500 }, { "epoch": 1.11, "eval_exact_match": 35.6, "eval_f1": 58.85244864490982, "step": 5700 }, { "epoch": 1.17, "learning_rate": 2.076915578085397e-05, "loss": 1.6915, "step": 6000 }, { "epoch": 1.17, "eval_exact_match": 35.2, "eval_f1": 59.16502836102447, "step": 6000 }, { "epoch": 1.23, "eval_exact_match": 36.0, "eval_f1": 59.28728478827406, "step": 6300 }, { "epoch": 1.27, "learning_rate": 1.833690777929421e-05, "loss": 1.7288, "step": 6500 }, { "epoch": 1.29, "eval_exact_match": 36.2, "eval_f1": 59.70995993203818, "step": 6600 }, { "epoch": 1.35, "eval_exact_match": 36.2, "eval_f1": 59.92556989591485, "step": 6900 }, { "epoch": 1.36, "learning_rate": 1.5899785533242347e-05, "loss": 1.6793, "step": 7000 }, { "epoch": 1.4, "eval_exact_match": 38.4, "eval_f1": 61.975454931168336, "step": 7200 }, { "epoch": 1.46, "learning_rate": 1.3462663287190486e-05, "loss": 1.6725, "step": 7500 }, { "epoch": 1.46, "eval_exact_match": 38.0, "eval_f1": 61.930039530596105, "step": 7500 }, { "epoch": 1.52, "eval_exact_match": 38.8, "eval_f1": 62.09105990298747, "step": 7800 }, { "epoch": 1.56, "learning_rate": 1.1025541041138624e-05, "loss": 1.6617, "step": 8000 }, { "epoch": 1.58, "eval_exact_match": 37.8, "eval_f1": 61.672659509788964, "step": 8100 }, { "epoch": 1.64, "eval_exact_match": 37.6, "eval_f1": 61.65640556094907, "step": 8400 }, { "epoch": 1.66, "learning_rate": 8.593293039578867e-06, "loss": 1.6424, "step": 8500 }, { "epoch": 1.7, "eval_exact_match": 38.6, "eval_f1": 61.961397592697885, "step": 8700 }, { "epoch": 1.75, "learning_rate": 6.1561707935270034e-06, "loss": 1.6578, "step": 9000 }, { "epoch": 1.75, "eval_exact_match": 39.2, "eval_f1": 62.377446998405304, "step": 9000 }, { "epoch": 1.81, "eval_exact_match": 39.2, "eval_f1": 62.48747456683246, "step": 9300 }, { "epoch": 1.85, "learning_rate": 3.723922791967245e-06, "loss": 1.6741, "step": 9500 }, { "epoch": 1.87, "eval_exact_match": 39.0, "eval_f1": 62.53450232086453, "step": 9600 }, { "epoch": 1.93, "eval_exact_match": 38.2, "eval_f1": 61.8539243537157, "step": 9900 }, { "epoch": 1.95, "learning_rate": 1.2868005459153833e-06, "loss": 1.6349, "step": 10000 }, { "epoch": 1.99, "eval_exact_match": 39.2, "eval_f1": 62.42564444648843, "step": 10200 }, { "epoch": 2.0, "step": 10258, "total_flos": 2506534774917408.0, "train_loss": 1.963296373723938, "train_runtime": 350.9907, "train_samples_per_second": 467.539, "train_steps_per_second": 29.226 } ], "max_steps": 10258, "num_train_epochs": 2, "total_flos": 2506534774917408.0, "trial_name": null, "trial_params": null }