{ "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0358227079538556, "global_step": 25000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.12, "learning_rate": 2.0234722784297853e-05, "loss": 3.0626, "step": 1000 }, { "epoch": 0.12, "eval_HasAns_exact": 49.527665317139004, "eval_HasAns_f1": 57.18183857248376, "eval_HasAns_total": 5928, "eval_NoAns_exact": 49.201009251471824, "eval_NoAns_f1": 49.201009251471824, "eval_NoAns_total": 5945, "eval_best_exact": 51.04017518739998, "eval_best_exact_thresh": 0.0, "eval_best_f1": 53.577488916632994, "eval_best_f1_thresh": 0.0, "eval_exact": 49.36410342794576, "eval_f1": 53.18571035607559, "eval_total": 11873, "step": 1000 }, { "epoch": 0.24, "learning_rate": 4.0469445568595707e-05, "loss": 1.5868, "step": 2000 }, { "epoch": 0.24, "eval_HasAns_exact": 63.41093117408907, "eval_HasAns_f1": 71.54881846166067, "eval_HasAns_total": 5928, "eval_NoAns_exact": 49.991589571068126, "eval_NoAns_f1": 49.991589571068126, "eval_NoAns_total": 5945, "eval_best_exact": 56.73376568685252, "eval_best_exact_thresh": 0.0, "eval_best_f1": 60.78846086420696, "eval_best_f1_thresh": 0.0, "eval_exact": 56.69165333108734, "eval_f1": 60.75477097959467, "eval_total": 11873, "step": 2000 }, { "epoch": 0.36, "learning_rate": 4.931660810252171e-05, "loss": 1.4132, "step": 3000 }, { "epoch": 0.36, "eval_HasAns_exact": 62.7361673414305, "eval_HasAns_f1": 70.43865494687375, "eval_HasAns_total": 5928, "eval_NoAns_exact": 60.26913372582002, "eval_NoAns_f1": 60.26913372582002, "eval_NoAns_total": 5945, "eval_best_exact": 61.509306830624105, "eval_best_exact_thresh": 0.0, "eval_best_f1": 65.35503634507485, "eval_best_f1_thresh": 0.0, "eval_exact": 61.500884359471065, "eval_f1": 65.34661387392151, "eval_total": 11873, "step": 3000 }, { "epoch": 0.49, "learning_rate": 4.8024751963621334e-05, "loss": 1.2797, "step": 4000 }, { "epoch": 0.49, "eval_HasAns_exact": 64.1531713900135, "eval_HasAns_f1": 70.62491669164416, "eval_HasAns_total": 5928, "eval_NoAns_exact": 69.7560975609756, "eval_NoAns_f1": 69.7560975609756, "eval_NoAns_total": 5945, "eval_best_exact": 66.95864566663859, "eval_best_exact_thresh": 0.0, "eval_best_f1": 70.18988512996461, "eval_best_f1_thresh": 0.0, "eval_exact": 66.95864566663859, "eval_f1": 70.18988512996451, "eval_total": 11873, "step": 4000 }, { "epoch": 0.61, "learning_rate": 4.673289582472096e-05, "loss": 1.2075, "step": 5000 }, { "epoch": 0.61, "eval_HasAns_exact": 67.25708502024291, "eval_HasAns_f1": 75.0100721049388, "eval_HasAns_total": 5928, "eval_NoAns_exact": 63.65012615643398, "eval_NoAns_f1": 63.65012615643398, "eval_NoAns_total": 5945, "eval_best_exact": 65.45102333024509, "eval_best_exact_thresh": 0.0, "eval_best_f1": 69.32196643123734, "eval_best_f1_thresh": 0.0, "eval_exact": 65.45102333024509, "eval_f1": 69.32196643123719, "eval_total": 11873, "step": 5000 }, { "epoch": 0.73, "learning_rate": 4.544103968582059e-05, "loss": 1.125, "step": 6000 }, { "epoch": 0.73, "eval_HasAns_exact": 68.47165991902834, "eval_HasAns_f1": 75.39561673194235, "eval_HasAns_total": 5928, "eval_NoAns_exact": 67.48528174936922, "eval_NoAns_f1": 67.48528174936922, "eval_NoAns_total": 5945, "eval_best_exact": 67.98618714730902, "eval_best_exact_thresh": 0.0, "eval_best_f1": 71.44320862351178, "eval_best_f1_thresh": 0.0, "eval_exact": 67.97776467615599, "eval_f1": 71.43478615235871, "eval_total": 11873, "step": 6000 }, { "epoch": 0.85, "learning_rate": 4.414918354692022e-05, "loss": 1.1063, "step": 7000 }, { "epoch": 0.85, "eval_HasAns_exact": 66.07624831309042, "eval_HasAns_f1": 72.59462670438361, "eval_HasAns_total": 5928, "eval_NoAns_exact": 72.95206055508831, "eval_NoAns_f1": 72.95206055508831, "eval_NoAns_total": 5945, "eval_best_exact": 69.51065442600859, "eval_best_exact_thresh": 0.0, "eval_best_f1": 72.76517704906819, "eval_best_f1_thresh": 0.0, "eval_exact": 69.51907689716163, "eval_f1": 72.77359952022128, "eval_total": 11873, "step": 7000 }, { "epoch": 0.97, "learning_rate": 4.2857327408019844e-05, "loss": 1.0754, "step": 8000 }, { "epoch": 0.97, "eval_HasAns_exact": 68.55600539811066, "eval_HasAns_f1": 75.38705540287036, "eval_HasAns_total": 5928, "eval_NoAns_exact": 71.8587047939445, "eval_NoAns_f1": 71.8587047939445, "eval_NoAns_total": 5945, "eval_best_exact": 70.2097195317106, "eval_best_exact_thresh": 0.0, "eval_best_f1": 73.62035411675363, "eval_best_f1_thresh": 0.0, "eval_exact": 70.2097195317106, "eval_f1": 73.62035411675373, "eval_total": 11873, "step": 8000 }, { "epoch": 1.09, "learning_rate": 4.156547126911947e-05, "loss": 0.8572, "step": 9000 }, { "epoch": 1.09, "eval_HasAns_exact": 67.66194331983806, "eval_HasAns_f1": 75.07304484460943, "eval_HasAns_total": 5928, "eval_NoAns_exact": 72.44743481917578, "eval_NoAns_f1": 72.44743481917578, "eval_NoAns_total": 5945, "eval_best_exact": 70.04969257980291, "eval_best_exact_thresh": 0.0, "eval_best_f1": 73.74993766014028, "eval_best_f1_thresh": 0.0, "eval_exact": 70.05811505095595, "eval_f1": 73.75836013129344, "eval_total": 11873, "step": 9000 }, { "epoch": 1.21, "learning_rate": 4.02736151302191e-05, "loss": 0.7951, "step": 10000 }, { "epoch": 1.21, "eval_HasAns_exact": 66.75101214574899, "eval_HasAns_f1": 72.62656614039999, "eval_HasAns_total": 5928, "eval_NoAns_exact": 77.49369217830109, "eval_NoAns_f1": 77.49369217830109, "eval_NoAns_total": 5945, "eval_best_exact": 72.13004295460289, "eval_best_exact_thresh": 0.0, "eval_best_f1": 75.063613583786, "eval_best_f1_thresh": 0.0, "eval_exact": 72.13004295460289, "eval_f1": 75.0636135837861, "eval_total": 11873, "step": 10000 }, { "epoch": 1.34, "learning_rate": 3.898175899131873e-05, "loss": 0.7932, "step": 11000 }, { "epoch": 1.34, "eval_HasAns_exact": 69.29824561403508, "eval_HasAns_f1": 75.49290590630278, "eval_HasAns_total": 5928, "eval_NoAns_exact": 74.68460891505467, "eval_NoAns_f1": 74.68460891505467, "eval_NoAns_total": 5945, "eval_best_exact": 71.9952834161543, "eval_best_exact_thresh": 0.0, "eval_best_f1": 75.0881787427409, "eval_best_f1_thresh": 0.0, "eval_exact": 71.9952834161543, "eval_f1": 75.08817874274101, "eval_total": 11873, "step": 11000 }, { "epoch": 1.46, "learning_rate": 3.7689902852418354e-05, "loss": 0.7958, "step": 12000 }, { "epoch": 1.46, "eval_HasAns_exact": 70.78272604588395, "eval_HasAns_f1": 77.6996709947113, "eval_HasAns_total": 5928, "eval_NoAns_exact": 72.17830109335576, "eval_NoAns_f1": 72.17830109335576, "eval_NoAns_total": 5945, "eval_best_exact": 71.48151267581909, "eval_best_exact_thresh": 0.0, "eval_best_f1": 74.93503323984241, "eval_best_f1_thresh": 0.0, "eval_exact": 71.48151267581909, "eval_f1": 74.93503323984244, "eval_total": 11873, "step": 12000 }, { "epoch": 1.58, "learning_rate": 3.639804671351799e-05, "loss": 0.7954, "step": 13000 }, { "epoch": 1.58, "eval_HasAns_exact": 68.08367071524967, "eval_HasAns_f1": 74.95631750408432, "eval_HasAns_total": 5928, "eval_NoAns_exact": 74.90328006728343, "eval_NoAns_f1": 74.90328006728343, "eval_NoAns_total": 5945, "eval_best_exact": 71.49835761812515, "eval_best_exact_thresh": 0.0, "eval_best_f1": 74.92976081565001, "eval_best_f1_thresh": 0.0, "eval_exact": 71.49835761812515, "eval_f1": 74.92976081565006, "eval_total": 11873, "step": 13000 }, { "epoch": 1.7, "learning_rate": 3.5106190574617616e-05, "loss": 0.8068, "step": 14000 }, { "epoch": 1.7, "eval_HasAns_exact": 69.41632928475033, "eval_HasAns_f1": 76.45120101125195, "eval_HasAns_total": 5928, "eval_NoAns_exact": 73.75946173254836, "eval_NoAns_f1": 73.75946173254836, "eval_NoAns_total": 5945, "eval_best_exact": 71.59100480080856, "eval_best_exact_thresh": 0.0, "eval_best_f1": 75.10340432870397, "eval_best_f1_thresh": 0.0, "eval_exact": 71.59100480080856, "eval_f1": 75.10340432870404, "eval_total": 11873, "step": 14000 }, { "epoch": 1.82, "learning_rate": 3.381433443571724e-05, "loss": 0.7992, "step": 15000 }, { "epoch": 1.82, "eval_HasAns_exact": 70.6646423751687, "eval_HasAns_f1": 77.88066795671843, "eval_HasAns_total": 5928, "eval_NoAns_exact": 72.6829268292683, "eval_NoAns_f1": 72.6829268292683, "eval_NoAns_total": 5945, "eval_best_exact": 71.67522951233892, "eval_best_exact_thresh": 0.0, "eval_best_f1": 75.27807627789335, "eval_best_f1_thresh": 0.0, "eval_exact": 71.67522951233892, "eval_f1": 75.2780762778934, "eval_total": 11873, "step": 15000 }, { "epoch": 1.94, "learning_rate": 3.2522478296816864e-05, "loss": 0.7675, "step": 16000 }, { "epoch": 1.94, "eval_HasAns_exact": 73.29622132253711, "eval_HasAns_f1": 80.6750111110637, "eval_HasAns_total": 5928, "eval_NoAns_exact": 68.29268292682927, "eval_NoAns_f1": 68.29268292682927, "eval_NoAns_total": 5945, "eval_best_exact": 70.79087004127011, "eval_best_exact_thresh": 0.0, "eval_best_f1": 74.47498238578171, "eval_best_f1_thresh": 0.0, "eval_exact": 70.79087004127011, "eval_f1": 74.47498238578176, "eval_total": 11873, "step": 16000 }, { "epoch": 2.06, "learning_rate": 3.123062215791649e-05, "loss": 0.6262, "step": 17000 }, { "epoch": 2.06, "eval_HasAns_exact": 74.07219973009447, "eval_HasAns_f1": 81.81573652565241, "eval_HasAns_total": 5928, "eval_NoAns_exact": 65.36585365853658, "eval_NoAns_f1": 65.36585365853658, "eval_NoAns_total": 5945, "eval_best_exact": 69.71279373368147, "eval_best_exact_thresh": 0.0, "eval_best_f1": 73.57901845566141, "eval_best_f1_thresh": 0.0, "eval_exact": 69.71279373368147, "eval_f1": 73.57901845566144, "eval_total": 11873, "step": 17000 }, { "epoch": 2.19, "learning_rate": 2.9938766019016122e-05, "loss": 0.4898, "step": 18000 }, { "epoch": 2.19, "eval_HasAns_exact": 70.17543859649123, "eval_HasAns_f1": 77.1654297749236, "eval_HasAns_total": 5928, "eval_NoAns_exact": 73.4735071488646, "eval_NoAns_f1": 73.4735071488646, "eval_NoAns_total": 5945, "eval_best_exact": 71.82683399309357, "eval_best_exact_thresh": 0.0, "eval_best_f1": 75.31682537738969, "eval_best_f1_thresh": 0.0, "eval_exact": 71.82683399309357, "eval_f1": 75.31682537738969, "eval_total": 11873, "step": 18000 }, { "epoch": 2.31, "learning_rate": 2.8646909880115753e-05, "loss": 0.4974, "step": 19000 }, { "epoch": 2.31, "eval_HasAns_exact": 71.22132253711202, "eval_HasAns_f1": 78.71079807483557, "eval_HasAns_total": 5928, "eval_NoAns_exact": 71.50546677880573, "eval_NoAns_f1": 71.50546677880573, "eval_NoAns_total": 5945, "eval_best_exact": 71.36359807967658, "eval_best_exact_thresh": 0.0, "eval_best_f1": 75.10297405774665, "eval_best_f1_thresh": 0.0, "eval_exact": 71.36359807967658, "eval_f1": 75.10297405774665, "eval_total": 11873, "step": 19000 }, { "epoch": 2.43, "learning_rate": 2.735505374121538e-05, "loss": 0.4995, "step": 20000 }, { "epoch": 2.43, "eval_HasAns_exact": 71.54183535762483, "eval_HasAns_f1": 79.38541316849339, "eval_HasAns_total": 5928, "eval_NoAns_exact": 69.33557611438184, "eval_NoAns_f1": 69.33557611438184, "eval_NoAns_total": 5945, "eval_best_exact": 70.43712625284259, "eval_best_exact_thresh": 0.0, "eval_best_f1": 74.3532998621098, "eval_best_f1_thresh": 0.0, "eval_exact": 70.43712625284259, "eval_f1": 74.3532998621098, "eval_total": 11873, "step": 20000 }, { "epoch": 2.55, "learning_rate": 2.606319760231501e-05, "loss": 0.4876, "step": 21000 }, { "epoch": 2.55, "eval_HasAns_exact": 73.0600539811066, "eval_HasAns_f1": 80.20732398479694, "eval_HasAns_total": 5928, "eval_NoAns_exact": 69.30193439865432, "eval_NoAns_f1": 69.30193439865432, "eval_NoAns_total": 5945, "eval_best_exact": 71.17830371430978, "eval_best_exact_thresh": 0.0, "eval_best_f1": 74.7468219137435, "eval_best_f1_thresh": 0.0, "eval_exact": 71.17830371430978, "eval_f1": 74.7468219137435, "eval_total": 11873, "step": 21000 }, { "epoch": 2.67, "learning_rate": 2.4771341463414636e-05, "loss": 0.4949, "step": 22000 }, { "epoch": 2.67, "eval_HasAns_exact": 70.95141700404858, "eval_HasAns_f1": 77.97020701423389, "eval_HasAns_total": 5928, "eval_NoAns_exact": 73.62489486963835, "eval_NoAns_f1": 73.62489486963835, "eval_NoAns_total": 5945, "eval_best_exact": 72.29006990651057, "eval_best_exact_thresh": 0.0, "eval_best_f1": 75.79444008930999, "eval_best_f1_thresh": 0.0, "eval_exact": 72.29006990651057, "eval_f1": 75.79444008930999, "eval_total": 11873, "step": 22000 }, { "epoch": 2.79, "learning_rate": 2.3479485324514263e-05, "loss": 0.492, "step": 23000 }, { "epoch": 2.79, "eval_HasAns_exact": 69.98987854251013, "eval_HasAns_f1": 76.7639044657501, "eval_HasAns_total": 5928, "eval_NoAns_exact": 76.56854499579478, "eval_NoAns_f1": 76.56854499579478, "eval_NoAns_total": 5945, "eval_best_exact": 73.28392150256886, "eval_best_exact_thresh": 0.0, "eval_best_f1": 76.66608487096481, "eval_best_f1_thresh": 0.0, "eval_exact": 73.28392150256886, "eval_f1": 76.66608487096487, "eval_total": 11873, "step": 23000 }, { "epoch": 2.91, "learning_rate": 2.218762918561389e-05, "loss": 0.4969, "step": 24000 }, { "epoch": 2.91, "eval_HasAns_exact": 72.36842105263158, "eval_HasAns_f1": 79.80006191408404, "eval_HasAns_total": 5928, "eval_NoAns_exact": 71.62321278385198, "eval_NoAns_f1": 71.62321278385198, "eval_NoAns_total": 5945, "eval_best_exact": 71.9952834161543, "eval_best_exact_thresh": 0.0, "eval_best_f1": 75.70578346051465, "eval_best_f1_thresh": 0.0, "eval_exact": 71.9952834161543, "eval_f1": 75.70578346051462, "eval_total": 11873, "step": 24000 }, { "epoch": 3.04, "learning_rate": 2.0895773046713518e-05, "loss": 0.4351, "step": 25000 }, { "epoch": 3.04, "eval_HasAns_exact": 70.41160593792172, "eval_HasAns_f1": 77.89997827092157, "eval_HasAns_total": 5928, "eval_NoAns_exact": 76.78721614802355, "eval_NoAns_f1": 76.78721614802355, "eval_NoAns_total": 5945, "eval_best_exact": 73.60397540638424, "eval_best_exact_thresh": 0.0, "eval_best_f1": 77.34280057188766, "eval_best_f1_thresh": 0.0, "eval_exact": 73.60397540638424, "eval_f1": 77.34280057188766, "eval_total": 11873, "step": 25000 } ], "max_steps": 41175, "num_train_epochs": 5, "total_flos": 7.838549951458406e+16, "trial_name": null, "trial_params": null }