{"canonical_length": 14, "canonical_nll": 1.9497860499790736, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.3157894736842105, "prompt_id": "54ef1334aadf1dfd", "reading_canonical_log_prob": -27.29700469970703, "reading_distances": [37, 36, 33, 38], "reading_log_probs": [-30.131423950195312, -31.792617797851562, -30.130577087402344, -31.724775314331055], "reading_logprob_std": 0.9401588606349203, "tokenization_path_logprob_std": 35.42418200284078, "trajectory_balance_rmse": 252.34524507633725} {"canonical_length": 23, "canonical_nll": 0.585747594418733, "generation_rouge_l_f1": 0.15384615384615385, "generation_token_f1": 0.2051282051282051, "prompt_id": "d109cb9b45bb2b84", "reading_canonical_log_prob": -13.47219467163086, "reading_distances": [12, 12, 9, 10], "reading_log_probs": [-20.349119186401367, -16.827327728271484, -17.873336791992188, -16.198217391967773], "reading_logprob_std": 1.827075053809213, "tokenization_path_logprob_std": 49.206924834885726, "trajectory_balance_rmse": 216.76323617175117} {"canonical_length": 79, "canonical_nll": 1.7142867076246044, "generation_rouge_l_f1": 0.35374149659863946, "generation_token_f1": 0.5306122448979591, "prompt_id": "f461f81672c6f974", "reading_canonical_log_prob": -135.42864990234375, "reading_distances": [24, 27, 34, 27], "reading_log_probs": [-131.45204162597656, -132.30120849609375, -135.8720245361328, -130.16302490234375], "reading_logprob_std": 2.4466666184860673, "tokenization_path_logprob_std": 169.09541892221063, "trajectory_balance_rmse": 1891.391975956774} {"canonical_length": 22, "canonical_nll": 0.7279287685047496, "generation_rouge_l_f1": 0.14423076923076922, "generation_token_f1": 0.14423076923076922, "prompt_id": "db2b92b024f2b9fd", "reading_canonical_log_prob": -16.014432907104492, "reading_distances": [20, 13, 17, 14], "reading_log_probs": [-23.62776756286621, -18.14665985107422, -23.78668785095215, -19.875160217285156], "reading_logprob_std": 2.802492002377508, "tokenization_path_logprob_std": 58.104934064858675, "trajectory_balance_rmse": 181.65991212671057} {"canonical_length": 22, "canonical_nll": 2.2437083504416724, "generation_rouge_l_f1": 0.44444444444444436, "generation_token_f1": 0.44444444444444436, "prompt_id": "a3ca3e528a4c44fb", "reading_canonical_log_prob": -49.3615837097168, "reading_distances": [27, 30, 28, 27], "reading_log_probs": [-50.08561325073242, -51.33319854736328, -52.262977600097656, -51.677608489990234], "reading_logprob_std": 0.920036127670522, "tokenization_path_logprob_std": 75.48698727608443, "trajectory_balance_rmse": 655.3405011600305} {"canonical_length": 46, "canonical_nll": 1.3379868217136548, "generation_rouge_l_f1": 0.24324324324324326, "generation_token_f1": 0.35135135135135126, "prompt_id": "bea2699cff758312", "reading_canonical_log_prob": -61.547393798828125, "reading_distances": [21, 15, 20, 25], "reading_log_probs": [-59.93962860107422, -61.71258544921875, -61.139259338378906, -58.488853454589844], "reading_logprob_std": 1.4269151040009944, "tokenization_path_logprob_std": 104.08720425569213, "trajectory_balance_rmse": 1208.0705007922395} {"canonical_length": 41, "canonical_nll": 1.4509633692299448, "generation_rouge_l_f1": 0.2018348623853211, "generation_token_f1": 0.2110091743119266, "prompt_id": "cf9306b8e22601b5", "reading_canonical_log_prob": -59.489498138427734, "reading_distances": [16, 21, 21, 18], "reading_log_probs": [-60.593387603759766, -61.17976760864258, -62.26262283325195, -62.15650939941406], "reading_logprob_std": 0.8016336202664475, "tokenization_path_logprob_std": 27.656336351048985, "trajectory_balance_rmse": 355.02941918607024} {"canonical_length": 231, "canonical_nll": 1.22139446043865, "generation_rouge_l_f1": 0.2336065573770492, "generation_token_f1": 0.5081967213114754, "prompt_id": "ff6c469935e35f5e", "reading_canonical_log_prob": -282.1421203613281, "reading_distances": [17, 21, 23, 17], "reading_log_probs": [-285.5497741699219, -283.59710693359375, -285.81842041015625, -279.8642578125], "reading_logprob_std": 2.7466741214007393, "tokenization_path_logprob_std": 327.7292653238587, "trajectory_balance_rmse": 5198.63038742627} {"canonical_length": 203, "canonical_nll": 1.4411894699622845, "generation_rouge_l_f1": 0.2990033222591362, "generation_token_f1": 0.5448504983388703, "prompt_id": "41c1bf9aacb45501", "reading_canonical_log_prob": -292.56146240234375, "reading_distances": [23, 22, 21, 22], "reading_log_probs": [-308.26318359375, -299.7464599609375, -303.59967041015625, -305.3548583984375], "reading_logprob_std": 3.560469781349985, "tokenization_path_logprob_std": 130.68999970835327, "trajectory_balance_rmse": 5103.771975252342} {"canonical_length": 33, "canonical_nll": 1.3438383160215435, "generation_rouge_l_f1": 0.1176470588235294, "generation_token_f1": 0.1176470588235294, "prompt_id": "aff22e4e1c81f916", "reading_canonical_log_prob": -44.34666442871094, "reading_distances": [25, 29, 34, 25], "reading_log_probs": [-43.83816146850586, -42.36849594116211, -44.00465393066406, -44.3736572265625], "reading_logprob_std": 0.880730775576844, "tokenization_path_logprob_std": 157.69646457136005, "trajectory_balance_rmse": 842.1142957208738} {"canonical_length": 10, "canonical_nll": 1.3745299339294434, "generation_rouge_l_f1": 0.07179487179487179, "generation_token_f1": 0.08205128205128205, "prompt_id": "f599bb889055c49d", "reading_canonical_log_prob": -13.745299339294434, "reading_distances": [63, 53, 60, 54], "reading_log_probs": [-16.73047637939453, -23.28733253479004, -23.449031829833984, -18.390525817871094], "reading_logprob_std": 3.4215062457845096, "tokenization_path_logprob_std": 52.21980017097139, "trajectory_balance_rmse": 190.30158021330953} {"canonical_length": 90, "canonical_nll": 1.4008102416992188, "generation_rouge_l_f1": 0.14594594594594595, "generation_token_f1": 0.2756756756756757, "prompt_id": "7fbc599a6c4c6b54", "reading_canonical_log_prob": -126.07292175292969, "reading_distances": [21, 34, 28, 33], "reading_log_probs": [-137.07778930664062, -141.08282470703125, -136.9124755859375, -135.23550415039062], "reading_logprob_std": 2.480874416765412, "tokenization_path_logprob_std": 152.63952075619554, "trajectory_balance_rmse": 1949.403003493134} {"canonical_length": 294, "canonical_nll": 1.6962220068691538, "generation_rouge_l_f1": 0.2162162162162162, "generation_token_f1": 0.3675675675675676, "prompt_id": "9e624d36638d2bbd", "reading_canonical_log_prob": -498.68927001953125, "reading_distances": [20, 13, 22, 16], "reading_log_probs": [-491.0934753417969, -493.2012023925781, -489.84619140625, -491.46453857421875], "reading_logprob_std": 1.3852516603905454, "tokenization_path_logprob_std": 226.15448614864366, "trajectory_balance_rmse": 6454.177480840409} {"canonical_length": 27, "canonical_nll": 0.7758024710196035, "generation_rouge_l_f1": 0.07352941176470588, "generation_token_f1": 0.08088235294117646, "prompt_id": "3982276a331d29f5", "reading_canonical_log_prob": -20.946666717529297, "reading_distances": [26, 23, 24, 25], "reading_log_probs": [-16.079570770263672, -17.354854583740234, -17.343957901000977, -17.31911849975586], "reading_logprob_std": 0.6300472863355938, "tokenization_path_logprob_std": 34.677652439108854, "trajectory_balance_rmse": 335.39319908421663} {"canonical_length": 10, "canonical_nll": 1.2277457237243652, "generation_rouge_l_f1": 0.03053435114503817, "generation_token_f1": 0.03816793893129771, "prompt_id": "3b79dbc85baf9416", "reading_canonical_log_prob": -12.277457237243652, "reading_distances": [21, 25, 21, 19], "reading_log_probs": [-10.123486518859863, -11.248552322387695, -9.640289306640625, -11.270515441894531], "reading_logprob_std": 0.8195302270684031, "tokenization_path_logprob_std": 17.176371363568755, "trajectory_balance_rmse": 189.51031111859012} {"canonical_length": 174, "canonical_nll": 1.5014560743309986, "generation_rouge_l_f1": 0.24713958810068645, "generation_token_f1": 0.448512585812357, "prompt_id": "bc017f1a688cb1fb", "reading_canonical_log_prob": -261.25335693359375, "reading_distances": [23, 20, 25, 28], "reading_log_probs": [-264.40093994140625, -269.5585021972656, -266.40911865234375, -265.134033203125], "reading_logprob_std": 2.2783590839397823, "tokenization_path_logprob_std": 268.8466015076654, "trajectory_balance_rmse": 4505.631366394592} {"canonical_length": 64, "canonical_nll": 0.845356285572052, "generation_rouge_l_f1": 0.14634146341463417, "generation_token_f1": 0.20209059233449478, "prompt_id": "2060bf62bca6703a", "reading_canonical_log_prob": -54.10280227661133, "reading_distances": [17, 14, 23, 18], "reading_log_probs": [-58.909828186035156, -54.94184494018555, -56.85252380371094, -56.3802490234375], "reading_logprob_std": 1.6411100561064045, "tokenization_path_logprob_std": 70.67329766601227, "trajectory_balance_rmse": 1307.717189149067} {"canonical_length": 12, "canonical_nll": 0.7714955806732178, "generation_rouge_l_f1": 0.42857142857142855, "generation_token_f1": 0.42857142857142855, "prompt_id": "8300e5ea4c06c29e", "reading_canonical_log_prob": -9.257946968078613, "reading_distances": [27, 29, 30, 27], "reading_log_probs": [-16.552738189697266, -16.943790435791016, -14.881085395812988, -9.917574882507324], "reading_logprob_std": 3.23047507773506, "tokenization_path_logprob_std": 16.01189973498564, "trajectory_balance_rmse": 134.04585344628944} {"canonical_length": 142, "canonical_nll": 1.3487664880886885, "generation_rouge_l_f1": 0.14423076923076922, "generation_token_f1": 0.33653846153846156, "prompt_id": "5c25240178b5a183", "reading_canonical_log_prob": -191.52484130859375, "reading_distances": [20, 19, 20, 21], "reading_log_probs": [-211.66073608398438, -214.79019165039062, -219.22576904296875, -217.4642791748047], "reading_logprob_std": 3.299397238752619, "tokenization_path_logprob_std": 43.141548145940206, "trajectory_balance_rmse": 2830.348005627447} {"canonical_length": 14, "canonical_nll": 1.2294213431222099, "generation_rouge_l_f1": 0.6363636363636365, "generation_token_f1": 0.6363636363636365, "prompt_id": "571e075c79645e81", "reading_canonical_log_prob": -17.211898803710938, "reading_distances": [28, 15, 24, 19], "reading_log_probs": [-20.391035079956055, -17.317710876464844, -19.563589096069336, -18.96645164489746], "reading_logprob_std": 1.299945983295593, "tokenization_path_logprob_std": 18.593561204100716, "trajectory_balance_rmse": 318.4728569548606} {"canonical_length": 153, "canonical_nll": 1.5208706325954862, "generation_rouge_l_f1": 0.21052631578947367, "generation_token_f1": 0.4976076555023922, "prompt_id": "fc69046f95700385", "reading_canonical_log_prob": -232.69320678710938, "reading_distances": [39, 41, 39, 38], "reading_log_probs": [-244.11441040039062, -240.1039276123047, -239.96826171875, -240.55917358398438], "reading_logprob_std": 1.9682704096577965, "tokenization_path_logprob_std": 298.0368377478985, "trajectory_balance_rmse": 3658.718313421244} {"canonical_length": 56, "canonical_nll": 1.7218502589634486, "generation_rouge_l_f1": 0.03703703703703703, "generation_token_f1": 0.07407407407407406, "prompt_id": "fcbb32fd544b170a", "reading_canonical_log_prob": -96.42361450195312, "reading_distances": [24, 22, 24, 25], "reading_log_probs": [-101.10171508789062, -99.61028289794922, -100.95167541503906, -98.7996826171875], "reading_logprob_std": 1.10429898875844, "tokenization_path_logprob_std": 65.67758114209133, "trajectory_balance_rmse": 1129.05928123398} {"canonical_length": 83, "canonical_nll": 0.8529988484210279, "generation_rouge_l_f1": 0.3225806451612903, "generation_token_f1": 0.4331797235023041, "prompt_id": "f6d789446e38536d", "reading_canonical_log_prob": -70.79890441894531, "reading_distances": [8, 7, 8, 9], "reading_log_probs": [-72.28863525390625, -70.06280517578125, -70.4365463256836, -69.06256866455078], "reading_logprob_std": 1.3484567942808838, "tokenization_path_logprob_std": 157.62826792129223, "trajectory_balance_rmse": 1771.7513881017837} {"canonical_length": 270, "canonical_nll": 1.332774070457176, "generation_rouge_l_f1": 0.14382022471910114, "generation_token_f1": 0.2831460674157303, "prompt_id": "86891b80d6ccb140", "reading_canonical_log_prob": -359.8489990234375, "reading_distances": [20, 24, 27, 23], "reading_log_probs": [-361.638427734375, -367.2666015625, -360.9367980957031, -370.4036865234375], "reading_logprob_std": 4.550898051596304, "tokenization_path_logprob_std": 105.7928460822157, "trajectory_balance_rmse": 3543.5057797313793} {"canonical_length": 27, "canonical_nll": 1.6250678168402777, "generation_rouge_l_f1": 0.35789473684210527, "generation_token_f1": 0.4210526315789474, "prompt_id": "01a2b8a844748c41", "reading_canonical_log_prob": -43.8768310546875, "reading_distances": [33, 37, 33, 36], "reading_log_probs": [-41.667518615722656, -42.79216003417969, -42.820396423339844, -44.52349090576172], "reading_logprob_std": 1.177899188438816, "tokenization_path_logprob_std": 23.91034541576919, "trajectory_balance_rmse": 660.0470592876661} {"canonical_length": 66, "canonical_nll": 1.1378740252870503, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "fc8d39645d6796a9", "reading_canonical_log_prob": -75.09968566894531, "reading_distances": [28, 32, 30, 26], "reading_log_probs": [-84.13359069824219, -82.9926528930664, -84.29202270507812, -78.40328979492188], "reading_logprob_std": 2.762712722741883, "tokenization_path_logprob_std": 61.33859127518875, "trajectory_balance_rmse": 1574.0843312028092} {"canonical_length": 340, "canonical_nll": 1.227778715245864, "generation_rouge_l_f1": 0.22185970636215335, "generation_token_f1": 0.4469820554649266, "prompt_id": "7a3c4e3465a24fae", "reading_canonical_log_prob": -417.44476318359375, "reading_distances": [28, 28, 21, 20], "reading_log_probs": [-434.563720703125, -433.4266662597656, -432.45672607421875, -431.85205078125], "reading_logprob_std": 1.1857242167149433, "tokenization_path_logprob_std": 270.48772320895137, "trajectory_balance_rmse": 7164.407659105661} {"canonical_length": 22, "canonical_nll": 0.6750792590054598, "generation_rouge_l_f1": 0.8333333333333334, "generation_token_f1": 0.8333333333333334, "prompt_id": "9652191d0c5b1125", "reading_canonical_log_prob": -14.851743698120117, "reading_distances": [20, 17, 15, 23], "reading_log_probs": [-17.177465438842773, -24.283308029174805, -22.751739501953125, -23.871257781982422], "reading_logprob_std": 3.2931898786412677, "tokenization_path_logprob_std": 46.424743745904486, "trajectory_balance_rmse": 418.57524946722225} {"canonical_length": 229, "canonical_nll": 1.0042640652719024, "generation_rouge_l_f1": 0.29661016949152547, "generation_token_f1": 0.5127118644067797, "prompt_id": "1fb3dd895fb36ca1", "reading_canonical_log_prob": -229.97647094726562, "reading_distances": [46, 47, 47, 53], "reading_log_probs": [-228.78228759765625, -227.33395385742188, -230.16900634765625, -230.35150146484375], "reading_logprob_std": 1.4041460135646893, "tokenization_path_logprob_std": 205.74387005847728, "trajectory_balance_rmse": 4936.589344213509} {"canonical_length": 84, "canonical_nll": 0.24569852011544363, "generation_rouge_l_f1": 0.5217391304347826, "generation_token_f1": 0.6521739130434783, "prompt_id": "fba6cc8156910498", "reading_canonical_log_prob": -20.638675689697266, "reading_distances": [22, 14, 24, 27], "reading_log_probs": [-24.706499099731445, -21.638282775878906, -22.54901885986328, -24.702205657958984], "reading_logprob_std": 1.5524702857735513, "tokenization_path_logprob_std": 195.84893710736944, "trajectory_balance_rmse": 1476.6136564014707} {"canonical_length": 24, "canonical_nll": 0.5704621076583862, "generation_rouge_l_f1": 0.13793103448275862, "generation_token_f1": 0.13793103448275862, "prompt_id": "02b01e693f3f83a0", "reading_canonical_log_prob": -13.69109058380127, "reading_distances": [34, 32, 25, 24], "reading_log_probs": [-11.166200637817383, -11.215022087097168, -10.601247787475586, -11.135313034057617], "reading_logprob_std": 0.28734523554780333, "tokenization_path_logprob_std": 56.136266458109425, "trajectory_balance_rmse": 372.015656283551} {"canonical_length": 331, "canonical_nll": 0.8866763388642371, "generation_rouge_l_f1": 0.2596491228070175, "generation_token_f1": 0.5473684210526316, "prompt_id": "40bc9c44485ded59", "reading_canonical_log_prob": -293.4898681640625, "reading_distances": [27, 34, 28, 33], "reading_log_probs": [-304.820556640625, -305.47393798828125, -305.040771484375, -305.97601318359375], "reading_logprob_std": 0.5102998582022023, "tokenization_path_logprob_std": 197.3292615416616, "trajectory_balance_rmse": 4693.584315825606} {"canonical_length": 11, "canonical_nll": 1.8179811997847124, "generation_rouge_l_f1": 0.3, "generation_token_f1": 0.3, "prompt_id": "f0c9ccf7c1a3a5c6", "reading_canonical_log_prob": -19.997793197631836, "reading_distances": [22, 26, 27, 28], "reading_log_probs": [-18.434232711791992, -20.299884796142578, -20.90508460998535, -20.831409454345703], "reading_logprob_std": 1.1542110019113936, "tokenization_path_logprob_std": 39.75180634155779, "trajectory_balance_rmse": 263.1295894079302} {"canonical_length": 149, "canonical_nll": 1.9166034468068371, "generation_rouge_l_f1": 0.23940149625935164, "generation_token_f1": 0.4139650872817956, "prompt_id": "7b4b39937f27934d", "reading_canonical_log_prob": -285.57391357421875, "reading_distances": [27, 27, 23, 30], "reading_log_probs": [-282.19049072265625, -284.7328186035156, -282.2171630859375, -283.1026306152344], "reading_logprob_std": 1.1925543299324357, "tokenization_path_logprob_std": 126.5330486718361, "trajectory_balance_rmse": 3403.248526333784} {"canonical_length": 10, "canonical_nll": 0.9396727561950684, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.7777777777777778, "prompt_id": "177155c261ac8893", "reading_canonical_log_prob": -9.396727561950684, "reading_distances": [18, 15, 11, 15], "reading_log_probs": [-12.289041519165039, -10.631338119506836, -8.5715913772583, -10.48171329498291], "reading_logprob_std": 1.520618333650693, "tokenization_path_logprob_std": 45.334042071787565, "trajectory_balance_rmse": 205.0905158262058} {"canonical_length": 9, "canonical_nll": 0.4697124163309733, "generation_rouge_l_f1": 0.6666666666666665, "generation_token_f1": 0.7777777777777777, "prompt_id": "316f7da66f20e71c", "reading_canonical_log_prob": -4.22741174697876, "reading_distances": [31, 31, 30, 26], "reading_log_probs": [-4.425214767456055, -5.053736209869385, -4.718759536743164, -4.704432964324951], "reading_logprob_std": 0.2571636826639979, "tokenization_path_logprob_std": 39.97814127796114, "trajectory_balance_rmse": 204.53128934722557} {"canonical_length": 10, "canonical_nll": 1.5975896835327148, "generation_rouge_l_f1": 0.17977528089887643, "generation_token_f1": 0.20224719101123595, "prompt_id": "4c771fd22dee11bd", "reading_canonical_log_prob": -15.975896835327148, "reading_distances": [25, 31, 29, 30], "reading_log_probs": [-15.922562599182129, -18.94124984741211, -17.85403060913086, -19.243030548095703], "reading_logprob_std": 1.5019705631181715, "tokenization_path_logprob_std": 42.64066578607948, "trajectory_balance_rmse": 287.3071747815683} {"canonical_length": 150, "canonical_nll": 1.1588053385416666, "generation_rouge_l_f1": 0.1848341232227488, "generation_token_f1": 0.3744075829383886, "prompt_id": "fb7e151d45f5c282", "reading_canonical_log_prob": -173.82080078125, "reading_distances": [37, 33, 38, 39], "reading_log_probs": [-182.91925048828125, -184.49581909179688, -184.72935485839844, -182.03643798828125], "reading_logprob_std": 1.2876431939282993, "tokenization_path_logprob_std": 123.00500859381785, "trajectory_balance_rmse": 2947.655309016205} {"canonical_length": 186, "canonical_nll": 1.1706890803511425, "generation_rouge_l_f1": 0.2583732057416268, "generation_token_f1": 0.4736842105263158, "prompt_id": "c6ce7ec89de605a5", "reading_canonical_log_prob": -217.7481689453125, "reading_distances": [17, 19, 20, 19], "reading_log_probs": [-221.7207794189453, -215.6871795654297, -213.72265625, -218.21986389160156], "reading_logprob_std": 3.4536144022379314, "tokenization_path_logprob_std": 272.7170111869327, "trajectory_balance_rmse": 4391.604296496903} {"canonical_length": 220, "canonical_nll": 1.005317549272017, "generation_rouge_l_f1": 0.33472803347280333, "generation_token_f1": 0.5564853556485355, "prompt_id": "320fa6e957218c02", "reading_canonical_log_prob": -221.16986083984375, "reading_distances": [14, 13, 14, 13], "reading_log_probs": [-229.19802856445312, -220.642822265625, -227.8655548095703, -231.6480255126953], "reading_logprob_std": 4.730741438541889, "tokenization_path_logprob_std": 192.23185377336532, "trajectory_balance_rmse": 4264.228768831172} {"canonical_length": 58, "canonical_nll": 1.1937336099558864, "generation_rouge_l_f1": 0.04545454545454545, "generation_token_f1": 0.04545454545454545, "prompt_id": "d3de64d71666f0d5", "reading_canonical_log_prob": -69.2365493774414, "reading_distances": [22, 20, 23, 23], "reading_log_probs": [-75.10354614257812, -77.43204498291016, -86.99222564697266, -75.92497253417969], "reading_logprob_std": 5.504466078273172, "tokenization_path_logprob_std": 63.0700815330257, "trajectory_balance_rmse": 930.0715923227201} {"canonical_length": 11, "canonical_nll": 0.11106746846979315, "generation_rouge_l_f1": 0.6, "generation_token_f1": 0.9, "prompt_id": "bd4efb1cb33c7494", "reading_canonical_log_prob": -1.2217421531677246, "reading_distances": [44, 47, 47, 43], "reading_log_probs": [-2.4241766929626465, -2.865818738937378, -3.492772340774536, -3.059509515762329], "reading_logprob_std": 0.4433673949111005, "tokenization_path_logprob_std": 36.40830884216114, "trajectory_balance_rmse": 188.49520761358417} {"canonical_length": 38, "canonical_nll": 0.6213798522949219, "generation_rouge_l_f1": 0.18357487922705315, "generation_token_f1": 0.27053140096618356, "prompt_id": "d19ca6fde647262d", "reading_canonical_log_prob": -23.61243438720703, "reading_distances": [17, 15, 15, 17], "reading_log_probs": [-25.39507293701172, -24.19778060913086, -25.550437927246094, -24.12137222290039], "reading_logprob_std": 0.7614519878054649, "tokenization_path_logprob_std": 68.95746257396517, "trajectory_balance_rmse": 966.0253862748135} {"canonical_length": 88, "canonical_nll": 1.3089749596335671, "generation_rouge_l_f1": 0.2751322751322752, "generation_token_f1": 0.5291005291005292, "prompt_id": "b0e33c086e3feda9", "reading_canonical_log_prob": -115.1897964477539, "reading_distances": [14, 17, 19, 20], "reading_log_probs": [-119.43267059326172, -122.50727081298828, -120.7280044555664, -122.02059936523438], "reading_logprob_std": 1.3814808215580192, "tokenization_path_logprob_std": 88.85382649781239, "trajectory_balance_rmse": 1814.0977992990863} {"canonical_length": 53, "canonical_nll": 0.9078232387326798, "generation_rouge_l_f1": 0.46511627906976744, "generation_token_f1": 0.7674418604651162, "prompt_id": "85224f0bc07faa4c", "reading_canonical_log_prob": -48.11463165283203, "reading_distances": [18, 20, 17, 19], "reading_log_probs": [-49.24110412597656, -54.89569091796875, -49.67790985107422, -51.797706604003906], "reading_logprob_std": 2.582273733321932, "tokenization_path_logprob_std": 100.03850354511165, "trajectory_balance_rmse": 1062.7700177608617} {"canonical_length": 26, "canonical_nll": 1.357811267559345, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4912280701754386, "prompt_id": "7327a63351d4e2c9", "reading_canonical_log_prob": -35.30309295654297, "reading_distances": [40, 29, 31, 36], "reading_log_probs": [-43.78435516357422, -38.98894500732422, -37.598968505859375, -41.54953384399414], "reading_logprob_std": 2.7438569168011027, "tokenization_path_logprob_std": 93.98915077608285, "trajectory_balance_rmse": 614.3653482881413} {"canonical_length": 84, "canonical_nll": 1.395102092197963, "generation_rouge_l_f1": 0.31724137931034485, "generation_token_f1": 0.5379310344827586, "prompt_id": "7fabfcdc4a3bff58", "reading_canonical_log_prob": -117.1885757446289, "reading_distances": [29, 25, 32, 33], "reading_log_probs": [-118.19733428955078, -118.39522552490234, -118.77702331542969, -119.51531219482422], "reading_logprob_std": 0.581499515403082, "tokenization_path_logprob_std": 135.86680451922382, "trajectory_balance_rmse": 1846.088944593068} {"canonical_length": 13, "canonical_nll": 0.6232903553889348, "generation_rouge_l_f1": 0.14285714285714285, "generation_token_f1": 0.14285714285714285, "prompt_id": "173c8d85c343f33f", "reading_canonical_log_prob": -8.102774620056152, "reading_distances": [24, 32, 28, 32], "reading_log_probs": [-7.360267639160156, -6.12272834777832, -5.301791191101074, -8.886127471923828], "reading_logprob_std": 1.5613790945489274, "tokenization_path_logprob_std": 54.275459024437104, "trajectory_balance_rmse": 211.45364951429073} {"canonical_length": 118, "canonical_nll": 1.4928612143306408, "generation_rouge_l_f1": 0.26063829787234044, "generation_token_f1": 0.48936170212765956, "prompt_id": "e1f635b187b10d1b", "reading_canonical_log_prob": -176.15762329101562, "reading_distances": [29, 26, 23, 20], "reading_log_probs": [-177.0896453857422, -174.4681396484375, -176.765869140625, -175.94068908691406], "reading_logprob_std": 1.1699753826936081, "tokenization_path_logprob_std": 78.79068290000521, "trajectory_balance_rmse": 2348.703663968562} {"canonical_length": 242, "canonical_nll": 0.8136358339924458, "generation_rouge_l_f1": 0.39917695473251025, "generation_token_f1": 0.5679012345679012, "prompt_id": "5ff715a6459080e1", "reading_canonical_log_prob": -196.89987182617188, "reading_distances": [20, 30, 28, 26], "reading_log_probs": [-187.7091064453125, -193.48721313476562, -189.03094482421875, -191.59136962890625], "reading_logprob_std": 2.5854410661125184, "tokenization_path_logprob_std": 146.54608657862562, "trajectory_balance_rmse": 2983.6055446860587} {"canonical_length": 147, "canonical_nll": 0.7271813632679635, "generation_rouge_l_f1": 0.22023809523809526, "generation_token_f1": 0.3214285714285714, "prompt_id": "3e82cf9b74d0c14b", "reading_canonical_log_prob": -106.89566040039062, "reading_distances": [24, 26, 25, 23], "reading_log_probs": [-110.21884155273438, -108.0694351196289, -109.56341552734375, -109.22599029541016], "reading_logprob_std": 0.8999415735207692, "tokenization_path_logprob_std": 136.83071151463213, "trajectory_balance_rmse": 1657.0238988452584} {"canonical_length": 42, "canonical_nll": 1.1982054937453497, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.5, "prompt_id": "b1354ebe20eb5cb8", "reading_canonical_log_prob": -50.32463073730469, "reading_distances": [46, 54, 47, 50], "reading_log_probs": [-66.8611831665039, -64.5502700805664, -63.392032623291016, -66.43626403808594], "reading_logprob_std": 1.6258770241008185, "tokenization_path_logprob_std": 42.173977833990804, "trajectory_balance_rmse": 864.7905741268034} {"canonical_length": 124, "canonical_nll": 1.1890359693957913, "generation_rouge_l_f1": 0.18592964824120603, "generation_token_f1": 0.37185929648241206, "prompt_id": "751bafc67843f928", "reading_canonical_log_prob": -147.44046020507812, "reading_distances": [31, 31, 34, 36], "reading_log_probs": [-151.4251708984375, -155.15557861328125, -157.4653778076172, -155.28384399414062], "reading_logprob_std": 2.5066596862091606, "tokenization_path_logprob_std": 67.12314533327023, "trajectory_balance_rmse": 2733.134758166058} {"canonical_length": 283, "canonical_nll": 1.1704819750027606, "generation_rouge_l_f1": 0.17482517482517484, "generation_token_f1": 0.3286713286713287, "prompt_id": "2c0f185491268c7a", "reading_canonical_log_prob": -331.24639892578125, "reading_distances": [23, 30, 38, 21], "reading_log_probs": [-327.4717712402344, -328.46441650390625, -325.13885498046875, -329.3976745605469], "reading_logprob_std": 1.8304102626612873, "tokenization_path_logprob_std": 164.0094983574784, "trajectory_balance_rmse": 4476.261145969822} {"canonical_length": 384, "canonical_nll": 1.4805256525675456, "generation_rouge_l_f1": 0.22259136212624586, "generation_token_f1": 0.3853820598006645, "prompt_id": "d93394062daf791b", "reading_canonical_log_prob": -568.5218505859375, "reading_distances": [25, 28, 29, 25], "reading_log_probs": [-583.126220703125, -579.2742919921875, -572.9810791015625, -572.6715698242188], "reading_logprob_std": 5.085577994163089, "tokenization_path_logprob_std": 161.19802467356953, "trajectory_balance_rmse": 7176.613365851401} {"canonical_length": 13, "canonical_nll": 1.5513904278094952, "generation_rouge_l_f1": 0.24999999999999994, "generation_token_f1": 0.41666666666666663, "prompt_id": "1bc766d97da83b42", "reading_canonical_log_prob": -20.168075561523438, "reading_distances": [20, 20, 19, 19], "reading_log_probs": [-23.757640838623047, -22.195987701416016, -22.63919448852539, -20.44833755493164], "reading_logprob_std": 1.3751284296557096, "tokenization_path_logprob_std": 38.24004572358623, "trajectory_balance_rmse": 295.01103371759683} {"canonical_length": 5, "canonical_nll": 0.49666194915771483, "generation_rouge_l_f1": 0.9090909090909091, "generation_token_f1": 0.9090909090909091, "prompt_id": "03cbf10fd71003f1", "reading_canonical_log_prob": -2.483309745788574, "reading_distances": [27, 20, 23, 23], "reading_log_probs": [-3.065438747406006, -3.670248508453369, -3.6780107021331787, -3.263700485229492], "reading_logprob_std": 0.3051422436244518, "tokenization_path_logprob_std": 24.325359091211247, "trajectory_balance_rmse": 103.29419923624981} {"canonical_length": 218, "canonical_nll": 1.336617006074398, "generation_rouge_l_f1": 0.2292134831460674, "generation_token_f1": 0.4629213483146068, "prompt_id": "3b33c4e6381082ec", "reading_canonical_log_prob": -291.38250732421875, "reading_distances": [23, 20, 19, 25], "reading_log_probs": [-293.674072265625, -296.6055908203125, -294.0398864746094, -294.10089111328125], "reading_logprob_std": 1.346906219780117, "tokenization_path_logprob_std": 158.0049547447675, "trajectory_balance_rmse": 3739.1784889533415} {"canonical_length": 14, "canonical_nll": 0.8792092459542411, "generation_rouge_l_f1": 0.5185185185185186, "generation_token_f1": 0.6666666666666666, "prompt_id": "275fffbea930b52f", "reading_canonical_log_prob": -12.308929443359375, "reading_distances": [12, 17, 19, 17], "reading_log_probs": [-25.22675895690918, -28.32659912109375, -30.61043930053711, -30.13938331604004], "reading_logprob_std": 2.4401316122803856, "tokenization_path_logprob_std": 13.955170831601494, "trajectory_balance_rmse": 287.5199774571916} {"canonical_length": 181, "canonical_nll": 0.9791215085192938, "generation_rouge_l_f1": 0.2127659574468085, "generation_token_f1": 0.5295508274231678, "prompt_id": "a7f0f685867c31e2", "reading_canonical_log_prob": -177.2209930419922, "reading_distances": [81, 74, 72, 75], "reading_log_probs": [-197.41781616210938, -205.66775512695312, -202.90750122070312, -204.4287872314453], "reading_logprob_std": 3.6379962949038243, "tokenization_path_logprob_std": 116.50228580197404, "trajectory_balance_rmse": 2829.579150823272} {"canonical_length": 128, "canonical_nll": 1.1345092058181763, "generation_rouge_l_f1": 0.22572178477690288, "generation_token_f1": 0.42519685039370075, "prompt_id": "858a5416234d0d2a", "reading_canonical_log_prob": -145.21717834472656, "reading_distances": [19, 21, 21, 19], "reading_log_probs": [-163.4263458251953, -165.78042602539062, -159.54144287109375, -158.85894775390625], "reading_logprob_std": 3.2760858446802157, "tokenization_path_logprob_std": 39.087014621253424, "trajectory_balance_rmse": 2796.5945285144326} {"canonical_length": 205, "canonical_nll": 0.8387918611852134, "generation_rouge_l_f1": 0.3378016085790884, "generation_token_f1": 0.5308310991957105, "prompt_id": "17197512e608a44f", "reading_canonical_log_prob": -171.95233154296875, "reading_distances": [16, 14, 12, 13], "reading_log_probs": [-196.851318359375, -183.16036987304688, -177.20553588867188, -195.59117126464844], "reading_logprob_std": 9.587332220344615, "tokenization_path_logprob_std": 200.1304381327198, "trajectory_balance_rmse": 3790.2487714978506} {"canonical_length": 160, "canonical_nll": 1.540947151184082, "generation_rouge_l_f1": 0.2099737532808399, "generation_token_f1": 0.39370078740157477, "prompt_id": "7c90d4dc6f1c3493", "reading_canonical_log_prob": -246.55154418945312, "reading_distances": [22, 24, 15, 17], "reading_log_probs": [-251.52842712402344, -253.31747436523438, -246.74337768554688, -248.58580017089844], "reading_logprob_std": 2.9404995323209815, "tokenization_path_logprob_std": 65.21103653714212, "trajectory_balance_rmse": 3315.540905336944} {"canonical_length": 23, "canonical_nll": 4.225556083347486, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "d1e684b980513357", "reading_canonical_log_prob": -97.18778991699219, "reading_distances": [9, 19, 23, 13], "reading_log_probs": [-100.49637603759766, -95.93827819824219, -93.23497009277344, -95.8656997680664], "reading_logprob_std": 3.0163625429750125, "tokenization_path_logprob_std": 46.65136401557521, "trajectory_balance_rmse": 379.5198730685012} {"canonical_length": 392, "canonical_nll": 1.0270040862414302, "generation_rouge_l_f1": 0.2385321100917431, "generation_token_f1": 0.47094801223241595, "prompt_id": "5b2aa008c5db43fd", "reading_canonical_log_prob": -402.5856018066406, "reading_distances": [7, 12, 10, 9], "reading_log_probs": [-399.9477233886719, -397.9456787109375, -397.9828796386719, -394.0704345703125], "reading_logprob_std": 2.462001983132376, "tokenization_path_logprob_std": 264.42522258033716, "trajectory_balance_rmse": 7205.265904800515} {"canonical_length": 82, "canonical_nll": 0.9839157476657774, "generation_rouge_l_f1": 0.20886075949367086, "generation_token_f1": 0.310126582278481, "prompt_id": "180245122ca6704c", "reading_canonical_log_prob": -80.68109130859375, "reading_distances": [29, 24, 35, 31], "reading_log_probs": [-97.88654327392578, -95.59281921386719, -94.1202163696289, -88.53141784667969], "reading_logprob_std": 3.981534225947791, "tokenization_path_logprob_std": 70.20459837039209, "trajectory_balance_rmse": 1178.0327010758144} {"canonical_length": 99, "canonical_nll": 2.084275833284012, "generation_rouge_l_f1": 0.16201117318435754, "generation_token_f1": 0.2681564245810056, "prompt_id": "0280ead696ae0c9e", "reading_canonical_log_prob": -206.3433074951172, "reading_distances": [34, 24, 24, 32], "reading_log_probs": [-206.44024658203125, -208.23870849609375, -206.15835571289062, -209.02517700195312], "reading_logprob_std": 1.3892701234517393, "tokenization_path_logprob_std": 173.79026768904498, "trajectory_balance_rmse": 2685.395655797215} {"canonical_length": 148, "canonical_nll": 1.5668788600612331, "generation_rouge_l_f1": 0.2032967032967033, "generation_token_f1": 0.3131868131868132, "prompt_id": "1d5af645718b1db0", "reading_canonical_log_prob": -231.8980712890625, "reading_distances": [58, 46, 51, 52], "reading_log_probs": [-236.31564331054688, -229.95330810546875, -239.28570556640625, -240.66769409179688], "reading_logprob_std": 4.761324454964223, "tokenization_path_logprob_std": 110.68328778424484, "trajectory_balance_rmse": 2543.1963900569044} {"canonical_length": 120, "canonical_nll": 1.4862744649251303, "generation_rouge_l_f1": 0.321285140562249, "generation_token_f1": 0.4417670682730923, "prompt_id": "0211ac7bc9c6651a", "reading_canonical_log_prob": -178.35293579101562, "reading_distances": [17, 21, 20, 24], "reading_log_probs": [-208.4502716064453, -203.91168212890625, -207.02938842773438, -213.82931518554688], "reading_logprob_std": 4.141963656715089, "tokenization_path_logprob_std": 129.25618098781686, "trajectory_balance_rmse": 2482.5786580563017} {"canonical_length": 115, "canonical_nll": 0.6067611694335937, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.4651162790697674, "prompt_id": "0da3d3229c8dfac7", "reading_canonical_log_prob": -69.77753448486328, "reading_distances": [14, 15, 19, 15], "reading_log_probs": [-73.39289855957031, -75.1070785522461, -74.85179901123047, -71.59677124023438], "reading_logprob_std": 1.6144015346558234, "tokenization_path_logprob_std": 148.0888114054629, "trajectory_balance_rmse": 1495.7746400872397} {"canonical_length": 196, "canonical_nll": 1.6112988530373087, "generation_rouge_l_f1": 0.22675736961451246, "generation_token_f1": 0.46258503401360546, "prompt_id": "dd26708537c260a2", "reading_canonical_log_prob": -315.8145751953125, "reading_distances": [35, 38, 37, 36], "reading_log_probs": [-318.1626892089844, -318.0522155761719, -315.6744689941406, -315.1344299316406], "reading_logprob_std": 1.5767210854793505, "tokenization_path_logprob_std": 229.45270775720354, "trajectory_balance_rmse": 4927.868856595305} {"canonical_length": 95, "canonical_nll": 1.6347175999691612, "generation_rouge_l_f1": 0.2613065326633166, "generation_token_f1": 0.4824120603015075, "prompt_id": "9fc2e86bc1fbb199", "reading_canonical_log_prob": -155.2981719970703, "reading_distances": [13, 14, 14, 17], "reading_log_probs": [-158.4520721435547, -158.71730041503906, -157.61068725585938, -155.92788696289062], "reading_logprob_std": 1.2578759315022796, "tokenization_path_logprob_std": 114.69322209042274, "trajectory_balance_rmse": 2488.778435723449} {"canonical_length": 178, "canonical_nll": 1.185450693194786, "generation_rouge_l_f1": 0.26146788990825687, "generation_token_f1": 0.4862385321100918, "prompt_id": "0d0deb69605ca531", "reading_canonical_log_prob": -211.01022338867188, "reading_distances": [17, 19, 18, 14], "reading_log_probs": [-224.35423278808594, -219.05361938476562, -219.63449096679688, -227.93212890625], "reading_logprob_std": 4.195135520612033, "tokenization_path_logprob_std": 104.69314772217625, "trajectory_balance_rmse": 4206.438360112152} {"canonical_length": 20, "canonical_nll": 0.6773427963256836, "generation_rouge_l_f1": 0.42857142857142855, "generation_token_f1": 0.46428571428571425, "prompt_id": "5b98bbd785876818", "reading_canonical_log_prob": -13.546855926513672, "reading_distances": [25, 35, 29, 29], "reading_log_probs": [-14.239872932434082, -15.337894439697266, -14.023865699768066, -14.055354118347168], "reading_logprob_std": 0.623092440538984, "tokenization_path_logprob_std": 6.014026538279708, "trajectory_balance_rmse": 246.93956302206226} {"canonical_length": 322, "canonical_nll": 1.0143533078780085, "generation_rouge_l_f1": 0.1657848324514991, "generation_token_f1": 0.3492063492063492, "prompt_id": "9d8ebc1fd91528b6", "reading_canonical_log_prob": -326.62176513671875, "reading_distances": [21, 16, 23, 24], "reading_log_probs": [-335.62872314453125, -331.0585632324219, -334.64306640625, -336.54925537109375], "reading_logprob_std": 2.4037328599260643, "tokenization_path_logprob_std": 193.67718099009556, "trajectory_balance_rmse": 5829.973256754019} {"canonical_length": 202, "canonical_nll": 1.1862169020246751, "generation_rouge_l_f1": 0.26914153132250584, "generation_token_f1": 0.4965197215777263, "prompt_id": "5f17b0be639454cf", "reading_canonical_log_prob": -239.61581420898438, "reading_distances": [20, 15, 22, 24], "reading_log_probs": [-233.71585083007812, -237.51718139648438, -238.7593994140625, -234.63101196289062], "reading_logprob_std": 2.3741959839546367, "tokenization_path_logprob_std": 255.10166249544721, "trajectory_balance_rmse": 4274.687750337663} {"canonical_length": 11, "canonical_nll": 0.38014936447143555, "generation_rouge_l_f1": 0.9, "generation_token_f1": 0.9, "prompt_id": "1932b722766373ae", "reading_canonical_log_prob": -4.181643009185791, "reading_distances": [39, 43, 40, 40], "reading_log_probs": [-5.512370586395264, -7.392113208770752, -6.305584907531738, -7.202442646026611], "reading_logprob_std": 0.8679399617888697, "tokenization_path_logprob_std": 9.482893794193625, "trajectory_balance_rmse": 180.35993970348247} {"canonical_length": 31, "canonical_nll": 1.7752243780320691, "generation_rouge_l_f1": 0.08843537414965986, "generation_token_f1": 0.16326530612244897, "prompt_id": "3f8935161866e686", "reading_canonical_log_prob": -55.03195571899414, "reading_distances": [29, 21, 32, 27], "reading_log_probs": [-61.71436309814453, -62.52757263183594, -55.15277099609375, -71.20614624023438], "reading_logprob_std": 6.590561076569817, "tokenization_path_logprob_std": 71.73365829409585, "trajectory_balance_rmse": 556.1202112054387} {"canonical_length": 45, "canonical_nll": 1.067165035671658, "generation_rouge_l_f1": 0.4473684210526316, "generation_token_f1": 0.6315789473684211, "prompt_id": "88f54467af0c5001", "reading_canonical_log_prob": -48.02242660522461, "reading_distances": [23, 28, 28, 26], "reading_log_probs": [-50.729270935058594, -55.408531188964844, -50.14244079589844, -49.703857421875], "reading_logprob_std": 2.6419482268752628, "tokenization_path_logprob_std": 40.85013061901521, "trajectory_balance_rmse": 811.062279352168} {"canonical_length": 272, "canonical_nll": 1.3916435241699219, "generation_rouge_l_f1": 0.2433460076045627, "generation_token_f1": 0.5171102661596959, "prompt_id": "05940c01ebeedfd9", "reading_canonical_log_prob": -378.52703857421875, "reading_distances": [21, 29, 30, 24], "reading_log_probs": [-381.76226806640625, -385.24176025390625, -388.5406494140625, -385.7702331542969], "reading_logprob_std": 2.7831975506704403, "tokenization_path_logprob_std": 240.64523000992148, "trajectory_balance_rmse": 6764.68129008645} {"canonical_length": 26, "canonical_nll": 1.1543677403376653, "generation_rouge_l_f1": 0.16216216216216214, "generation_token_f1": 0.16216216216216214, "prompt_id": "cc3b11b7124977c4", "reading_canonical_log_prob": -30.013561248779297, "reading_distances": [22, 25, 18, 24], "reading_log_probs": [-36.12590789794922, -34.66185760498047, -34.12157440185547, -30.603363037109375], "reading_logprob_std": 2.3416588327497774, "tokenization_path_logprob_std": 75.30868284857581, "trajectory_balance_rmse": 540.7789191322325} {"canonical_length": 25, "canonical_nll": 0.9942060852050781, "generation_rouge_l_f1": 0.7741935483870969, "generation_token_f1": 0.7741935483870969, "prompt_id": "9335f10e0202a2ca", "reading_canonical_log_prob": -24.855152130126953, "reading_distances": [21, 24, 28, 25], "reading_log_probs": [-28.384748458862305, -27.661962509155273, -27.434612274169922, -28.467510223388672], "reading_logprob_std": 0.5163575336736211, "tokenization_path_logprob_std": 15.48587627223928, "trajectory_balance_rmse": 259.6699108439691} {"canonical_length": 160, "canonical_nll": 1.262875461578369, "generation_rouge_l_f1": 0.27338129496402874, "generation_token_f1": 0.43165467625899284, "prompt_id": "ee0b55cb338c65fa", "reading_canonical_log_prob": -202.06007385253906, "reading_distances": [31, 27, 24, 29], "reading_log_probs": [-220.13034057617188, -226.61351013183594, -231.57156372070312, -217.68624877929688], "reading_logprob_std": 6.298061347951014, "tokenization_path_logprob_std": 208.66881313788807, "trajectory_balance_rmse": 3212.0805910124695} {"canonical_length": 205, "canonical_nll": 0.47595977783203125, "generation_rouge_l_f1": 0.37280701754385964, "generation_token_f1": 0.5833333333333333, "prompt_id": "9ed2cebe3982e4a6", "reading_canonical_log_prob": -97.5717544555664, "reading_distances": [11, 9, 12, 13], "reading_log_probs": [-94.90589904785156, -95.7464828491211, -102.68312072753906, -99.24523162841797], "reading_logprob_std": 3.5613531482840557, "tokenization_path_logprob_std": 236.12776391427755, "trajectory_balance_rmse": 2020.159938430683} {"canonical_length": 67, "canonical_nll": 1.5109899720149254, "generation_rouge_l_f1": 0.2938388625592417, "generation_token_f1": 0.4360189573459715, "prompt_id": "57c5e7bb82e63751", "reading_canonical_log_prob": -101.236328125, "reading_distances": [25, 31, 31, 31], "reading_log_probs": [-100.09224700927734, -99.27998352050781, -103.1185302734375, -102.28221893310547], "reading_logprob_std": 1.8041945271674047, "tokenization_path_logprob_std": 139.02338005817077, "trajectory_balance_rmse": 1371.0814768683476} {"canonical_length": 233, "canonical_nll": 1.2846656111688572, "generation_rouge_l_f1": 0.27642276422764234, "generation_token_f1": 0.5447154471544715, "prompt_id": "2049d32643760e75", "reading_canonical_log_prob": -299.32708740234375, "reading_distances": [33, 23, 35, 30], "reading_log_probs": [-304.79461669921875, -309.26898193359375, -306.9161376953125, -308.32171630859375], "reading_logprob_std": 1.9444379362242683, "tokenization_path_logprob_std": 263.26287052025134, "trajectory_balance_rmse": 4608.354092720017} {"canonical_length": 65, "canonical_nll": 1.2572606013371395, "generation_rouge_l_f1": 0.2649006622516556, "generation_token_f1": 0.43708609271523186, "prompt_id": "9327dc9ed2ef5314", "reading_canonical_log_prob": -81.72193908691406, "reading_distances": [14, 17, 18, 19], "reading_log_probs": [-73.07252502441406, -76.3521957397461, -77.89594268798828, -73.01271057128906], "reading_logprob_std": 2.4393723144694968, "tokenization_path_logprob_std": 84.79276333804067, "trajectory_balance_rmse": 1340.3009508306375} {"canonical_length": 27, "canonical_nll": 1.22370359632704, "generation_rouge_l_f1": 0.4680851063829788, "generation_token_f1": 0.5106382978723405, "prompt_id": "fb2aa3ee59bdbf8a", "reading_canonical_log_prob": -33.03999710083008, "reading_distances": [103, 96, 95, 98], "reading_log_probs": [-35.25102233886719, -37.890724182128906, -36.5616340637207, -40.47060775756836], "reading_logprob_std": 2.229203071402915, "tokenization_path_logprob_std": 101.82202979967529, "trajectory_balance_rmse": 682.1172670145515} {"canonical_length": 176, "canonical_nll": 0.7155953320589933, "generation_rouge_l_f1": 0.4586666666666667, "generation_token_f1": 0.6346666666666666, "prompt_id": "24055c0763c8403a", "reading_canonical_log_prob": -125.94477844238281, "reading_distances": [140, 136, 147, 136], "reading_log_probs": [-127.96427917480469, -129.44105529785156, -131.93641662597656, -136.19970703125], "reading_logprob_std": 3.6039661268095435, "tokenization_path_logprob_std": 51.89816498415684, "trajectory_balance_rmse": 3193.0339489762528} {"canonical_length": 73, "canonical_nll": 0.49964622602070846, "generation_rouge_l_f1": 0.3687150837988827, "generation_token_f1": 0.4245810055865921, "prompt_id": "c59968b8c3d9c69d", "reading_canonical_log_prob": -36.47417449951172, "reading_distances": [15, 15, 12, 15], "reading_log_probs": [-38.8802375793457, -38.228973388671875, -37.31223678588867, -38.70320129394531], "reading_logprob_std": 0.70203239172539, "tokenization_path_logprob_std": 123.38105581026515, "trajectory_balance_rmse": 1099.9111878302851} {"canonical_length": 283, "canonical_nll": 1.5044065010421268, "generation_rouge_l_f1": 0.28516377649325625, "generation_token_f1": 0.4778420038535646, "prompt_id": "c3213521900f0885", "reading_canonical_log_prob": -425.7470397949219, "reading_distances": [39, 42, 35, 42], "reading_log_probs": [-429.5526123046875, -429.36279296875, -429.2222900390625, -426.76611328125], "reading_logprob_std": 1.3135515383057783, "tokenization_path_logprob_std": 208.1398127062026, "trajectory_balance_rmse": 5297.20853895582} {"canonical_length": 70, "canonical_nll": 1.4394677298409597, "generation_rouge_l_f1": 0.27312775330396477, "generation_token_f1": 0.36123348017621143, "prompt_id": "48bb22e3d5050293", "reading_canonical_log_prob": -100.76274108886719, "reading_distances": [16, 17, 20, 23], "reading_log_probs": [-106.39407348632812, -108.12251281738281, -110.49842834472656, -115.08838653564453], "reading_logprob_std": 3.7711669079402843, "tokenization_path_logprob_std": 49.92046277435872, "trajectory_balance_rmse": 1125.445920318238} {"canonical_length": 183, "canonical_nll": 1.2718916095671107, "generation_rouge_l_f1": 0.3066666666666667, "generation_token_f1": 0.5777777777777777, "prompt_id": "5a342c18ce379992", "reading_canonical_log_prob": -232.75616455078125, "reading_distances": [30, 32, 26, 26], "reading_log_probs": [-235.36378479003906, -233.38327026367188, -233.47572326660156, -231.94911193847656], "reading_logprob_std": 1.400687467973967, "tokenization_path_logprob_std": 232.13446719272363, "trajectory_balance_rmse": 3928.026044803697} {"canonical_length": 280, "canonical_nll": 1.5723279680524553, "generation_rouge_l_f1": 0.29545454545454547, "generation_token_f1": 0.5318181818181817, "prompt_id": "66f2043754e3c9e5", "reading_canonical_log_prob": -440.2518310546875, "reading_distances": [21, 15, 28, 21], "reading_log_probs": [-450.9353942871094, -443.9150695800781, -448.598876953125, -449.15313720703125], "reading_logprob_std": 2.9944855003969977, "tokenization_path_logprob_std": 251.9063796980962, "trajectory_balance_rmse": 5871.7494975464115} {"canonical_length": 61, "canonical_nll": 1.8064682757268187, "generation_rouge_l_f1": 0.21951219512195122, "generation_token_f1": 0.3536585365853659, "prompt_id": "b6ca714dbce654ce", "reading_canonical_log_prob": -110.19456481933594, "reading_distances": [44, 40, 44, 45], "reading_log_probs": [-127.99491119384766, -123.2599105834961, -122.73528289794922, -131.44737243652344], "reading_logprob_std": 4.135352455299086, "tokenization_path_logprob_std": 132.2593993447681, "trajectory_balance_rmse": 1728.3014619201228} {"canonical_length": 38, "canonical_nll": 1.2158171001233553, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "4112128a24dc114d", "reading_canonical_log_prob": -46.2010498046875, "reading_distances": [28, 33, 29, 29], "reading_log_probs": [-50.84478759765625, -50.14234161376953, -50.59428405761719, -51.07964324951172], "reading_logprob_std": 0.4010090850105128, "tokenization_path_logprob_std": 158.64868775995424, "trajectory_balance_rmse": 804.4822938745144} {"canonical_length": 24, "canonical_nll": 0.6962072054545084, "generation_rouge_l_f1": 0.5384615384615385, "generation_token_f1": 0.576923076923077, "prompt_id": "07948a71ef97c7bb", "reading_canonical_log_prob": -16.708972930908203, "reading_distances": [22, 17, 20, 21], "reading_log_probs": [-18.61897087097168, -17.05960464477539, -17.40824317932129, -17.635265350341797], "reading_logprob_std": 0.6689316272912873, "tokenization_path_logprob_std": 86.58042410225825, "trajectory_balance_rmse": 635.0072124646224} {"canonical_length": 130, "canonical_nll": 1.6172778789813702, "generation_rouge_l_f1": 0.18556701030927833, "generation_token_f1": 0.3608247422680413, "prompt_id": "d894cb33b4a83ecb", "reading_canonical_log_prob": -210.24612426757812, "reading_distances": [22, 32, 25, 33], "reading_log_probs": [-218.9940185546875, -222.00265502929688, -228.24462890625, -225.48568725585938], "reading_logprob_std": 4.036014517654948, "tokenization_path_logprob_std": 146.48999911336108, "trajectory_balance_rmse": 3087.2993540741695} {"canonical_length": 49, "canonical_nll": 0.4983270606216119, "generation_rouge_l_f1": 0.13924050632911392, "generation_token_f1": 0.1518987341772152, "prompt_id": "98f5a9670dc63974", "reading_canonical_log_prob": -24.418025970458984, "reading_distances": [14, 19, 21, 21], "reading_log_probs": [-25.350229263305664, -24.578529357910156, -26.350418090820312, -26.87905502319336], "reading_logprob_std": 1.0265099310884358, "tokenization_path_logprob_std": 60.606018860799814, "trajectory_balance_rmse": 753.8084287448397} {"canonical_length": 155, "canonical_nll": 1.3793449155745967, "generation_rouge_l_f1": 0.21649484536082472, "generation_token_f1": 0.37628865979381443, "prompt_id": "9866aa5bc569efc5", "reading_canonical_log_prob": -213.7984619140625, "reading_distances": [11, 8, 10, 13], "reading_log_probs": [-223.9603271484375, -223.5155792236328, -223.23211669921875, -223.97015380859375], "reading_logprob_std": 0.36054074770561456, "tokenization_path_logprob_std": 187.43450563794076, "trajectory_balance_rmse": 3351.6824376371433} {"canonical_length": 173, "canonical_nll": 1.873979689757948, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "63dbcd016e93c526", "reading_canonical_log_prob": -324.198486328125, "reading_distances": [23, 18, 27, 28], "reading_log_probs": [-323.632080078125, -328.01251220703125, -323.42681884765625, -321.11138916015625], "reading_logprob_std": 2.880964798533925, "tokenization_path_logprob_std": 126.07984412154264, "trajectory_balance_rmse": 3694.856188266734} {"canonical_length": 45, "canonical_nll": 0.972894287109375, "generation_rouge_l_f1": 0.326530612244898, "generation_token_f1": 0.38775510204081637, "prompt_id": "bb113220673ce1f4", "reading_canonical_log_prob": -43.780242919921875, "reading_distances": [24, 30, 27, 22], "reading_log_probs": [-45.83067321777344, -44.659263610839844, -44.892494201660156, -47.59858322143555], "reading_logprob_std": 1.3352527906485212, "tokenization_path_logprob_std": 25.6255619928763, "trajectory_balance_rmse": 584.2122565517343} {"canonical_length": 16, "canonical_nll": 2.2273945808410645, "generation_rouge_l_f1": 0.4444444444444445, "generation_token_f1": 0.5925925925925927, "prompt_id": "5d38e8c442eebce3", "reading_canonical_log_prob": -35.63831329345703, "reading_distances": [37, 40, 42, 45], "reading_log_probs": [-39.40292739868164, -45.146095275878906, -41.80894088745117, -44.37932586669922], "reading_logprob_std": 2.611972970235645, "tokenization_path_logprob_std": 46.98050513363756, "trajectory_balance_rmse": 348.0554752056059} {"canonical_length": 7, "canonical_nll": 1.6335875647408622, "generation_rouge_l_f1": 0.6, "generation_token_f1": 1.0, "prompt_id": "8ee1f7d9e89653dc", "reading_canonical_log_prob": -11.435112953186035, "reading_distances": [21, 23, 24, 19], "reading_log_probs": [-10.19539737701416, -9.536157608032227, -9.22426986694336, -11.38166618347168], "reading_logprob_std": 0.9549194611165669, "tokenization_path_logprob_std": 40.91803236802802, "trajectory_balance_rmse": 128.33844469545107} {"canonical_length": 220, "canonical_nll": 1.7617155595259233, "generation_rouge_l_f1": 0.2181818181818182, "generation_token_f1": 0.3717171717171717, "prompt_id": "8768fc8d898485e9", "reading_canonical_log_prob": -387.5774230957031, "reading_distances": [16, 15, 12, 16], "reading_log_probs": [-395.45355224609375, -401.8929443359375, -389.03717041015625, -399.13641357421875], "reading_logprob_std": 5.560750035377808, "tokenization_path_logprob_std": 119.3483570660833, "trajectory_balance_rmse": 4794.528838310525} {"canonical_length": 238, "canonical_nll": 1.1914749786633403, "generation_rouge_l_f1": 0.32057416267942584, "generation_token_f1": 0.5454545454545455, "prompt_id": "f7a53c2bee8bfe30", "reading_canonical_log_prob": -283.571044921875, "reading_distances": [14, 15, 19, 24], "reading_log_probs": [-284.2467346191406, -283.57391357421875, -289.5729064941406, -287.49273681640625], "reading_logprob_std": 2.814094330480817, "tokenization_path_logprob_std": 84.28695117834087, "trajectory_balance_rmse": 4383.77561936986} {"canonical_length": 218, "canonical_nll": 1.2288188409367833, "generation_rouge_l_f1": 0.25052192066805845, "generation_token_f1": 0.49686847599164935, "prompt_id": "90b0a660ae2019c3", "reading_canonical_log_prob": -267.88250732421875, "reading_distances": [31, 29, 40, 26], "reading_log_probs": [-271.6253662109375, -270.27557373046875, -264.40704345703125, -272.7942199707031], "reading_logprob_std": 3.724024464617067, "tokenization_path_logprob_std": 205.2797532108577, "trajectory_balance_rmse": 3817.2696306040802} {"canonical_length": 153, "canonical_nll": 1.2038778666577308, "generation_rouge_l_f1": 0.23493975903614456, "generation_token_f1": 0.3433734939759036, "prompt_id": "e11b424a8c471248", "reading_canonical_log_prob": -184.1933135986328, "reading_distances": [38, 51, 52, 57], "reading_log_probs": [-190.853759765625, -186.42410278320312, -187.40338134765625, -195.02967834472656], "reading_logprob_std": 3.8959443798198605, "tokenization_path_logprob_std": 271.5117057294532, "trajectory_balance_rmse": 2685.5905757300666} {"canonical_length": 69, "canonical_nll": 1.7637957144474639, "generation_rouge_l_f1": 0.25757575757575757, "generation_token_f1": 0.3333333333333333, "prompt_id": "cc633ec6fcbbe3de", "reading_canonical_log_prob": -121.701904296875, "reading_distances": [18, 20, 20, 22], "reading_log_probs": [-123.7466049194336, -122.79342651367188, -122.46684265136719, -121.23414611816406], "reading_logprob_std": 1.0374786294167504, "tokenization_path_logprob_std": 139.26353544153886, "trajectory_balance_rmse": 1273.3918973911505} {"canonical_length": 51, "canonical_nll": 1.7523778279622395, "generation_rouge_l_f1": 0.1554054054054054, "generation_token_f1": 0.23648648648648649, "prompt_id": "66e12e5aa4182ef3", "reading_canonical_log_prob": -89.37126922607422, "reading_distances": [33, 39, 39, 49], "reading_log_probs": [-90.10105895996094, -87.1590347290039, -95.96902465820312, -91.01496124267578], "reading_logprob_std": 3.6623146318293083, "tokenization_path_logprob_std": 92.17348869003027, "trajectory_balance_rmse": 891.2909668245221} {"canonical_length": 147, "canonical_nll": 1.5330451394424958, "generation_rouge_l_f1": 0.16786570743405274, "generation_token_f1": 0.3021582733812949, "prompt_id": "831cd880037811e2", "reading_canonical_log_prob": -225.35763549804688, "reading_distances": [19, 19, 28, 19], "reading_log_probs": [-218.6829376220703, -221.54971313476562, -219.07614135742188, -220.3008575439453], "reading_logprob_std": 1.2964311028489932, "tokenization_path_logprob_std": 229.35461201522256, "trajectory_balance_rmse": 3547.515564525693} {"canonical_length": 239, "canonical_nll": 1.012785588348261, "generation_rouge_l_f1": 0.22088353413654618, "generation_token_f1": 0.4899598393574297, "prompt_id": "61ee3d334ec189e4", "reading_canonical_log_prob": -242.05575561523438, "reading_distances": [22, 26, 31, 29], "reading_log_probs": [-262.58026123046875, -261.9789123535156, -257.24822998046875, -260.752685546875], "reading_logprob_std": 2.3856586671705005, "tokenization_path_logprob_std": 247.60003831124837, "trajectory_balance_rmse": 5680.192074705777} {"canonical_length": 239, "canonical_nll": 1.6740068890559623, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.5373134328358209, "prompt_id": "978f2098f6623381", "reading_canonical_log_prob": -400.087646484375, "reading_distances": [16, 19, 23, 20], "reading_log_probs": [-398.16925048828125, -398.34515380859375, -397.9270935058594, -397.7976989746094], "reading_logprob_std": 0.24475441198561942, "tokenization_path_logprob_std": 314.23121614350976, "trajectory_balance_rmse": 4976.78139402694} {"canonical_length": 366, "canonical_nll": 1.946721478238132, "generation_rouge_l_f1": 0.22585924713584288, "generation_token_f1": 0.45171849427168576, "prompt_id": "f4a2dbf6cf78f4f4", "reading_canonical_log_prob": -712.5000610351562, "reading_distances": [41, 37, 37, 36], "reading_log_probs": [-725.3699340820312, -722.5311279296875, -722.2054443359375, -722.6027221679688], "reading_logprob_std": 1.4719404062626282, "tokenization_path_logprob_std": 157.68403907427162, "trajectory_balance_rmse": 8145.741434863139} {"canonical_length": 323, "canonical_nll": 1.623653447295859, "generation_rouge_l_f1": 0.21071428571428572, "generation_token_f1": 0.4357142857142857, "prompt_id": "607d603874554f5a", "reading_canonical_log_prob": -524.4400634765625, "reading_distances": [21, 20, 22, 23], "reading_log_probs": [-522.29150390625, -524.783935546875, -528.4216918945312, -522.4332885742188], "reading_logprob_std": 2.86401975535302, "tokenization_path_logprob_std": 184.9807989805545, "trajectory_balance_rmse": 5994.893689497195} {"canonical_length": 16, "canonical_nll": 1.776570439338684, "generation_rouge_l_f1": 0.13333333333333333, "generation_token_f1": 0.13333333333333333, "prompt_id": "65ee5494d4aae49e", "reading_canonical_log_prob": -28.425127029418945, "reading_distances": [15, 17, 22, 18], "reading_log_probs": [-31.84292221069336, -36.61409378051758, -33.1880989074707, -34.323421478271484], "reading_logprob_std": 2.0207266208985715, "tokenization_path_logprob_std": 50.84548029673221, "trajectory_balance_rmse": 320.9502805502754} {"canonical_length": 120, "canonical_nll": 1.9149976094563803, "generation_rouge_l_f1": 0.24719101123595508, "generation_token_f1": 0.4419475655430712, "prompt_id": "bc654a75738c95db", "reading_canonical_log_prob": -229.79971313476562, "reading_distances": [27, 29, 26, 27], "reading_log_probs": [-226.0516357421875, -228.89772033691406, -227.16624450683594, -229.0208740234375], "reading_logprob_std": 1.4321240697506572, "tokenization_path_logprob_std": 146.1317987431496, "trajectory_balance_rmse": 2401.5364186874926} {"canonical_length": 166, "canonical_nll": 1.059880543904132, "generation_rouge_l_f1": 0.2512315270935961, "generation_token_f1": 0.5172413793103449, "prompt_id": "1277e399b9abf4ff", "reading_canonical_log_prob": -175.94017028808594, "reading_distances": [28, 32, 24, 24], "reading_log_probs": [-187.19786071777344, -189.51959228515625, -184.64535522460938, -184.40394592285156], "reading_logprob_std": 2.4100135670856524, "tokenization_path_logprob_std": 70.08444306106972, "trajectory_balance_rmse": 4031.9261764103053} {"canonical_length": 13, "canonical_nll": 1.82622190622183, "generation_rouge_l_f1": 0.17391304347826086, "generation_token_f1": 0.17391304347826086, "prompt_id": "888761fd7cf1d9af", "reading_canonical_log_prob": -23.74088478088379, "reading_distances": [29, 23, 26, 29], "reading_log_probs": [-24.676197052001953, -24.733196258544922, -23.751136779785156, -25.035037994384766], "reading_logprob_std": 0.5546548341744915, "tokenization_path_logprob_std": 58.836374213134995, "trajectory_balance_rmse": 361.1851272454957} {"canonical_length": 196, "canonical_nll": 1.3558875881895727, "generation_rouge_l_f1": 0.1894736842105263, "generation_token_f1": 0.4126315789473684, "prompt_id": "4049144029e8d577", "reading_canonical_log_prob": -265.75396728515625, "reading_distances": [20, 24, 26, 23], "reading_log_probs": [-265.62652587890625, -270.2095947265625, -266.3697509765625, -267.04510498046875], "reading_logprob_std": 2.0162623978472305, "tokenization_path_logprob_std": 187.99369343386954, "trajectory_balance_rmse": 4803.887532426119} {"canonical_length": 105, "canonical_nll": 1.1616864885602678, "generation_rouge_l_f1": 0.3173076923076923, "generation_token_f1": 0.38461538461538464, "prompt_id": "35fc3a98ce123749", "reading_canonical_log_prob": -121.97708129882812, "reading_distances": [37, 34, 41, 33], "reading_log_probs": [-124.44586944580078, -119.11756134033203, -119.99779510498047, -119.5045394897461], "reading_logprob_std": 2.479261298138735, "tokenization_path_logprob_std": 237.16037300936708, "trajectory_balance_rmse": 2095.254695115631} {"canonical_length": 17, "canonical_nll": 3.879818187040441, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "30594b15ee7b3362", "reading_canonical_log_prob": -65.9569091796875, "reading_distances": [20, 22, 21, 20], "reading_log_probs": [-64.02786254882812, -64.41773223876953, -67.56487274169922, -64.11534881591797], "reading_logprob_std": 1.697184868163345, "tokenization_path_logprob_std": 81.09657464925445, "trajectory_balance_rmse": 388.7505151934599} {"canonical_length": 89, "canonical_nll": 1.1611084670163272, "generation_rouge_l_f1": 0.21369863013698628, "generation_token_f1": 0.3013698630136986, "prompt_id": "55622d0a348ebb7e", "reading_canonical_log_prob": -103.33865356445312, "reading_distances": [18, 15, 16, 18], "reading_log_probs": [-106.99649047851562, -108.00550079345703, -107.1644287109375, -108.37901306152344], "reading_logprob_std": 0.6633125561133711, "tokenization_path_logprob_std": 169.1223337041733, "trajectory_balance_rmse": 1675.0943496073892} {"canonical_length": 181, "canonical_nll": 1.235990071164969, "generation_rouge_l_f1": 0.2571428571428571, "generation_token_f1": 0.49523809523809526, "prompt_id": "96bafc4191db5d29", "reading_canonical_log_prob": -223.71420288085938, "reading_distances": [35, 26, 23, 33], "reading_log_probs": [-250.69244384765625, -245.19931030273438, -242.95294189453125, -249.5555419921875], "reading_logprob_std": 3.6398741514517474, "tokenization_path_logprob_std": 112.58967492257536, "trajectory_balance_rmse": 3767.452140441261} {"canonical_length": 186, "canonical_nll": 1.3218177877446657, "generation_rouge_l_f1": 0.31279620853080564, "generation_token_f1": 0.4644549763033175, "prompt_id": "a69ab9afafd56708", "reading_canonical_log_prob": -245.8581085205078, "reading_distances": [26, 30, 32, 31], "reading_log_probs": [-251.13795471191406, -249.38177490234375, -251.13731384277344, -252.5281219482422], "reading_logprob_std": 1.288813949918433, "tokenization_path_logprob_std": 145.27162483459267, "trajectory_balance_rmse": 3660.561342376146} {"canonical_length": 7, "canonical_nll": 1.5508989606584822, "generation_rouge_l_f1": 0.03755868544600939, "generation_token_f1": 0.03755868544600939, "prompt_id": "7d277a000cf626ea", "reading_canonical_log_prob": -10.856292724609375, "reading_distances": [35, 27, 28, 28], "reading_log_probs": [-11.044538497924805, -8.265727043151855, -8.148492813110352, -9.0921630859375], "reading_logprob_std": 1.3387773476935543, "tokenization_path_logprob_std": 55.169745633065624, "trajectory_balance_rmse": 179.09354695345192} {"canonical_length": 8, "canonical_nll": 0.4528108239173889, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "4b1173c764ce0a81", "reading_canonical_log_prob": -3.6224865913391113, "reading_distances": [45, 57, 50, 50], "reading_log_probs": [-6.628847599029541, -8.869915008544922, -6.209334850311279, -8.802679061889648], "reading_logprob_std": 1.406311930781061, "tokenization_path_logprob_std": 26.76160031604973, "trajectory_balance_rmse": 127.46120548170839} {"canonical_length": 48, "canonical_nll": 1.0377055009206135, "generation_rouge_l_f1": 0.2694300518134715, "generation_token_f1": 0.3212435233160622, "prompt_id": "c86574795b337627", "reading_canonical_log_prob": -49.80986404418945, "reading_distances": [8, 10, 9, 8], "reading_log_probs": [-55.872802734375, -52.570030212402344, -51.54332733154297, -53.03675842285156], "reading_logprob_std": 1.8528829200516803, "tokenization_path_logprob_std": 13.604061830211856, "trajectory_balance_rmse": 509.12836041516215} {"canonical_length": 251, "canonical_nll": 1.2328564161323456, "generation_rouge_l_f1": 0.2256637168141593, "generation_token_f1": 0.4070796460176991, "prompt_id": "f0455cfad098aced", "reading_canonical_log_prob": -309.44696044921875, "reading_distances": [40, 40, 42, 42], "reading_log_probs": [-330.107177734375, -336.12054443359375, -333.59771728515625, -334.482421875], "reading_logprob_std": 2.5383395368835147, "tokenization_path_logprob_std": 217.36270396052856, "trajectory_balance_rmse": 5235.77945683031} {"canonical_length": 92, "canonical_nll": 1.342961021091627, "generation_rouge_l_f1": 0.2657342657342658, "generation_token_f1": 0.34965034965034963, "prompt_id": "765c339341bf24b5", "reading_canonical_log_prob": -123.55241394042969, "reading_distances": [73, 64, 58, 62], "reading_log_probs": [-134.4619598388672, -132.66012573242188, -136.42800903320312, -132.74488830566406], "reading_logprob_std": 1.7755233450918027, "tokenization_path_logprob_std": 146.26096286652563, "trajectory_balance_rmse": 1358.7366383867197} {"canonical_length": 25, "canonical_nll": 2.181556396484375, "generation_rouge_l_f1": 0.27118644067796616, "generation_token_f1": 0.4406779661016949, "prompt_id": "ef02d7967a69b355", "reading_canonical_log_prob": -54.538909912109375, "reading_distances": [34, 29, 29, 42], "reading_log_probs": [-52.324806213378906, -52.544273376464844, -52.187294006347656, -50.88588333129883], "reading_logprob_std": 0.7477151898202552, "tokenization_path_logprob_std": 18.457304243082195, "trajectory_balance_rmse": 486.7297063327723} {"canonical_length": 26, "canonical_nll": 1.141248996441181, "generation_rouge_l_f1": 0.48148148148148145, "generation_token_f1": 0.5555555555555556, "prompt_id": "27392a7bee3d725d", "reading_canonical_log_prob": -29.672473907470703, "reading_distances": [30, 32, 35, 30], "reading_log_probs": [-33.82945251464844, -35.629825592041016, -35.96915817260742, -40.25110626220703], "reading_logprob_std": 2.7212708650056734, "tokenization_path_logprob_std": 43.83541179036776, "trajectory_balance_rmse": 594.0170403013653} {"canonical_length": 53, "canonical_nll": 1.6254633777546432, "generation_rouge_l_f1": 0.04166666666666667, "generation_token_f1": 0.04166666666666667, "prompt_id": "deecb5d2b0490e50", "reading_canonical_log_prob": -86.1495590209961, "reading_distances": [15, 23, 21, 17], "reading_log_probs": [-89.67301940917969, -91.0243911743164, -91.66471862792969, -90.902587890625], "reading_logprob_std": 0.8321963222228719, "tokenization_path_logprob_std": 80.6054514872991, "trajectory_balance_rmse": 1107.9522759792146} {"canonical_length": 124, "canonical_nll": 1.2365506079889113, "generation_rouge_l_f1": 0.20526315789473687, "generation_token_f1": 0.4368421052631579, "prompt_id": "8c1c19dea7da1533", "reading_canonical_log_prob": -153.332275390625, "reading_distances": [26, 29, 26, 20], "reading_log_probs": [-156.48751831054688, -156.17160034179688, -152.72061157226562, -158.26715087890625], "reading_logprob_std": 2.3187862564528032, "tokenization_path_logprob_std": 211.07024667640027, "trajectory_balance_rmse": 2151.8891947028196} {"canonical_length": 81, "canonical_nll": 1.4417977980625483, "generation_rouge_l_f1": 0.21621621621621626, "generation_token_f1": 0.3783783783783784, "prompt_id": "8ddcb00e61289cea", "reading_canonical_log_prob": -116.7856216430664, "reading_distances": [17, 20, 20, 22], "reading_log_probs": [-130.05490112304688, -127.02366638183594, -123.48745727539062, -126.89889526367188], "reading_logprob_std": 2.6838766668939678, "tokenization_path_logprob_std": 56.39809067020346, "trajectory_balance_rmse": 1678.7267026548225} {"canonical_length": 78, "canonical_nll": 0.8153157845521585, "generation_rouge_l_f1": 0.35514018691588783, "generation_token_f1": 0.41121495327102814, "prompt_id": "e509fb17129966c3", "reading_canonical_log_prob": -63.59463119506836, "reading_distances": [48, 40, 43, 52], "reading_log_probs": [-69.7365493774414, -71.84226989746094, -69.90609741210938, -71.17782592773438], "reading_logprob_std": 1.0143810361622707, "tokenization_path_logprob_std": 43.015150609826, "trajectory_balance_rmse": 1177.4758057986662} {"canonical_length": 170, "canonical_nll": 1.1468864889705883, "generation_rouge_l_f1": 0.23017902813299235, "generation_token_f1": 0.4245524296675192, "prompt_id": "07a3f091145dd3c4", "reading_canonical_log_prob": -194.970703125, "reading_distances": [27, 25, 30, 29], "reading_log_probs": [-186.9697723388672, -189.69662475585938, -191.95677185058594, -190.73892211914062], "reading_logprob_std": 2.1250526454608107, "tokenization_path_logprob_std": 119.25785539220011, "trajectory_balance_rmse": 3861.377383766603} {"canonical_length": 159, "canonical_nll": 1.4578095442094143, "generation_rouge_l_f1": 0.24697336561743344, "generation_token_f1": 0.5230024213075061, "prompt_id": "1cfb5bc2275fa797", "reading_canonical_log_prob": -231.79171752929688, "reading_distances": [24, 27, 18, 22], "reading_log_probs": [-242.98391723632812, -243.468994140625, -248.98985290527344, -234.00389099121094], "reading_logprob_std": 6.202113536553222, "tokenization_path_logprob_std": 86.7244433398197, "trajectory_balance_rmse": 3176.9902182021947} {"canonical_length": 11, "canonical_nll": 0.35730177705938165, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "c9f832251ee1421d", "reading_canonical_log_prob": -3.9303195476531982, "reading_distances": [20, 22, 20, 18], "reading_log_probs": [-2.950313091278076, -3.2896859645843506, -3.6884512901306152, -3.824432373046875], "reading_logprob_std": 0.39660694518566847, "tokenization_path_logprob_std": 28.545258395019893, "trajectory_balance_rmse": 125.34856319881996} {"canonical_length": 174, "canonical_nll": 1.0525912580818966, "generation_rouge_l_f1": 0.2871046228710462, "generation_token_f1": 0.4914841849148418, "prompt_id": "13d2c1dd5e0685fa", "reading_canonical_log_prob": -183.15087890625, "reading_distances": [54, 61, 58, 50], "reading_log_probs": [-177.26177978515625, -176.7695770263672, -172.48985290527344, -170.92323303222656], "reading_logprob_std": 3.137684909808828, "tokenization_path_logprob_std": 104.0161707009056, "trajectory_balance_rmse": 1871.2520256338767} {"canonical_length": 266, "canonical_nll": 1.2487221624618186, "generation_rouge_l_f1": 0.22348484848484848, "generation_token_f1": 0.48484848484848486, "prompt_id": "e2b16f584507a85d", "reading_canonical_log_prob": -332.16009521484375, "reading_distances": [32, 28, 27, 23], "reading_log_probs": [-333.614990234375, -336.7932434082031, -338.5971984863281, -336.1915588378906], "reading_logprob_std": 2.060861034365257, "tokenization_path_logprob_std": 472.41879628772574, "trajectory_balance_rmse": 6418.357109641469} {"canonical_length": 279, "canonical_nll": 1.2234976163474462, "generation_rouge_l_f1": 0.3236363636363636, "generation_token_f1": 0.5745454545454546, "prompt_id": "76e96744ab3b125b", "reading_canonical_log_prob": -341.3558349609375, "reading_distances": [18, 25, 16, 24], "reading_log_probs": [-347.51629638671875, -349.68157958984375, -349.41058349609375, -349.9233093261719], "reading_logprob_std": 1.0979240341912189, "tokenization_path_logprob_std": 161.39062854761795, "trajectory_balance_rmse": 6796.737961264032} {"canonical_length": 157, "canonical_nll": 1.5452710777331309, "generation_rouge_l_f1": 0.2992874109263658, "generation_token_f1": 0.5035629453681709, "prompt_id": "2fad87f3c29a8aa9", "reading_canonical_log_prob": -242.60755920410156, "reading_distances": [27, 26, 28, 22], "reading_log_probs": [-239.92330932617188, -241.6875762939453, -241.18466186523438, -243.00486755371094], "reading_logprob_std": 1.2747868189463607, "tokenization_path_logprob_std": 69.70379670201507, "trajectory_balance_rmse": 3624.9523797177253} {"canonical_length": 135, "canonical_nll": 2.1193827311197917, "generation_rouge_l_f1": 0.16296296296296295, "generation_token_f1": 0.2765432098765432, "prompt_id": "057edeee1c342070", "reading_canonical_log_prob": -286.1166687011719, "reading_distances": [19, 19, 23, 21], "reading_log_probs": [-282.47698974609375, -282.7742004394531, -284.8822937011719, -284.1727294921875], "reading_logprob_std": 1.1421024719481703, "tokenization_path_logprob_std": 125.69842170837163, "trajectory_balance_rmse": 3301.719638751714} {"canonical_length": 273, "canonical_nll": 1.0304464822287087, "generation_rouge_l_f1": 0.24390243902439027, "generation_token_f1": 0.4840525328330207, "prompt_id": "b42b291887a21049", "reading_canonical_log_prob": -281.3118896484375, "reading_distances": [31, 28, 27, 29], "reading_log_probs": [-305.4306640625, -303.91680908203125, -300.83062744140625, -305.8578796386719], "reading_logprob_std": 2.2766961377823627, "tokenization_path_logprob_std": 124.15450251869994, "trajectory_balance_rmse": 5911.963957346555} {"canonical_length": 8, "canonical_nll": 0.5802167654037476, "generation_rouge_l_f1": 0.7058823529411764, "generation_token_f1": 0.7058823529411764, "prompt_id": "751d7c04030e09ea", "reading_canonical_log_prob": -4.6417341232299805, "reading_distances": [22, 26, 26, 24], "reading_log_probs": [-4.885215759277344, -3.0522265434265137, -2.3966727256774902, -4.017339706420898], "reading_logprob_std": 1.091392857256565, "tokenization_path_logprob_std": 76.66809048662431, "trajectory_balance_rmse": 212.1323492799328} {"canonical_length": 45, "canonical_nll": 1.190960693359375, "generation_rouge_l_f1": 0.6373626373626373, "generation_token_f1": 0.7692307692307692, "prompt_id": "bdda8d0e92726300", "reading_canonical_log_prob": -53.593231201171875, "reading_distances": [18, 23, 23, 24], "reading_log_probs": [-55.772769927978516, -55.88298797607422, -52.56964874267578, -55.562198638916016], "reading_logprob_std": 1.5904127129338117, "tokenization_path_logprob_std": 45.1985240584316, "trajectory_balance_rmse": 1035.387681791469} {"canonical_length": 225, "canonical_nll": 0.758565673828125, "generation_rouge_l_f1": 0.3613963039014374, "generation_token_f1": 0.5010266940451745, "prompt_id": "ce2f11428dfe632d", "reading_canonical_log_prob": -170.67727661132812, "reading_distances": [35, 33, 26, 26], "reading_log_probs": [-169.89820861816406, -171.84518432617188, -169.2128448486328, -172.79515075683594], "reading_logprob_std": 1.6662651505517172, "tokenization_path_logprob_std": 50.84236004415784, "trajectory_balance_rmse": 3423.0792510610027} {"canonical_length": 371, "canonical_nll": 1.1597685698229026, "generation_rouge_l_f1": 0.2464285714285714, "generation_token_f1": 0.4285714285714285, "prompt_id": "3ce75f08c1522ce9", "reading_canonical_log_prob": -430.2741394042969, "reading_distances": [20, 17, 24, 20], "reading_log_probs": [-438.50860595703125, -446.51556396484375, -435.2967834472656, -446.9334716796875], "reading_logprob_std": 5.822755211210871, "tokenization_path_logprob_std": 267.2731317844713, "trajectory_balance_rmse": 6302.060546083426} {"canonical_length": 3, "canonical_nll": 1.6111644109090169, "generation_rouge_l_f1": 0.03418803418803419, "generation_token_f1": 0.03418803418803419, "prompt_id": "cc1031d1c53d6dff", "reading_canonical_log_prob": -4.833493232727051, "reading_distances": [40, 29, 40, 40], "reading_log_probs": [-4.176474571228027, -4.282686710357666, -4.590517044067383, -4.032347202301025], "reading_logprob_std": 0.236725551745511, "tokenization_path_logprob_std": 12.56495697359529, "trajectory_balance_rmse": 32.003501919197966} {"canonical_length": 14, "canonical_nll": 3.7210426330566406, "generation_rouge_l_f1": 0.07142857142857142, "generation_token_f1": 0.07142857142857142, "prompt_id": "6b9afce682e6088d", "reading_canonical_log_prob": -52.09459686279297, "reading_distances": [55, 50, 45, 53], "reading_log_probs": [-55.46272277832031, -55.27147674560547, -51.44103240966797, -58.65870666503906], "reading_logprob_std": 2.953322048487298, "tokenization_path_logprob_std": 84.182577366367, "trajectory_balance_rmse": 370.21508336564125} {"canonical_length": 15, "canonical_nll": 0.4247511545817057, "generation_rouge_l_f1": 0.7142857142857143, "generation_token_f1": 0.7142857142857143, "prompt_id": "16fc84421f512267", "reading_canonical_log_prob": -6.371267318725586, "reading_distances": [24, 30, 36, 24], "reading_log_probs": [-8.49226188659668, -4.659757137298584, -8.232742309570312, -5.986100673675537], "reading_logprob_std": 1.839589989356806, "tokenization_path_logprob_std": 22.076144410398676, "trajectory_balance_rmse": 268.81774191477115} {"canonical_length": 12, "canonical_nll": 0.7075645128885905, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "7550a792cc616b59", "reading_canonical_log_prob": -8.490774154663086, "reading_distances": [24, 37, 26, 31], "reading_log_probs": [-10.864445686340332, -14.308194160461426, -11.336352348327637, -14.462608337402344], "reading_logprob_std": 1.9073987067414928, "tokenization_path_logprob_std": 23.439265992972924, "trajectory_balance_rmse": 181.5608871082561} {"canonical_length": 242, "canonical_nll": 1.5986977569327867, "generation_rouge_l_f1": 0.22519083969465647, "generation_token_f1": 0.4427480916030534, "prompt_id": "d8ba10df8798bb3c", "reading_canonical_log_prob": -386.8848571777344, "reading_distances": [33, 38, 35, 38], "reading_log_probs": [-393.70819091796875, -391.3500061035156, -394.95367431640625, -395.95989990234375], "reading_logprob_std": 1.988141787785842, "tokenization_path_logprob_std": 467.63379615208504, "trajectory_balance_rmse": 5739.0312287176475} {"canonical_length": 15, "canonical_nll": 1.0250547409057618, "generation_rouge_l_f1": 0.3636363636363637, "generation_token_f1": 0.3636363636363637, "prompt_id": "c62abdde12be0560", "reading_canonical_log_prob": -15.375821113586426, "reading_distances": [25, 28, 29, 29], "reading_log_probs": [-20.233850479125977, -17.99492645263672, -20.963199615478516, -19.23859214782715], "reading_logprob_std": 1.286691200755773, "tokenization_path_logprob_std": 60.12849634986213, "trajectory_balance_rmse": 280.07093925320567} {"canonical_length": 318, "canonical_nll": 1.537877784585053, "generation_rouge_l_f1": 0.22145328719723184, "generation_token_f1": 0.4013840830449827, "prompt_id": "e65b179e41f3017d", "reading_canonical_log_prob": -489.0451354980469, "reading_distances": [21, 27, 27, 30], "reading_log_probs": [-495.76165771484375, -497.1871643066406, -496.0802001953125, -503.5509033203125], "reading_logprob_std": 3.6553594718032008, "tokenization_path_logprob_std": 191.14256871137235, "trajectory_balance_rmse": 6947.6591146129085} {"canonical_length": 83, "canonical_nll": 1.1428651924592903, "generation_rouge_l_f1": 0.18435754189944137, "generation_token_f1": 0.3575418994413408, "prompt_id": "688f5dc10c3a0a47", "reading_canonical_log_prob": -94.8578109741211, "reading_distances": [33, 34, 24, 33], "reading_log_probs": [-125.51163482666016, -121.26952362060547, -107.80577087402344, -120.62548828125], "reading_logprob_std": 7.6452606659159565, "tokenization_path_logprob_std": 35.65651524306081, "trajectory_balance_rmse": 1679.0852888320987} {"canonical_length": 324, "canonical_nll": 1.6087047435619213, "generation_rouge_l_f1": 0.34615384615384615, "generation_token_f1": 0.5314685314685315, "prompt_id": "d7836a3fff052985", "reading_canonical_log_prob": -521.2203369140625, "reading_distances": [22, 21, 22, 22], "reading_log_probs": [-528.8226928710938, -524.5599365234375, -526.5816650390625, -524.575927734375], "reading_logprob_std": 2.027702632599386, "tokenization_path_logprob_std": 250.22447988613007, "trajectory_balance_rmse": 6042.894639750076} {"canonical_length": 18, "canonical_nll": 1.4257148106892903, "generation_rouge_l_f1": 0.4210526315789473, "generation_token_f1": 0.4210526315789473, "prompt_id": "97d1bc52cae659b0", "reading_canonical_log_prob": -25.662866592407227, "reading_distances": [9, 12, 13, 10], "reading_log_probs": [-27.69719696044922, -27.293071746826172, -27.4661922454834, -27.04833221435547], "reading_logprob_std": 0.27419290301555327, "tokenization_path_logprob_std": 23.94798566769602, "trajectory_balance_rmse": 220.34596891846743} {"canonical_length": 311, "canonical_nll": 0.9545270423030546, "generation_rouge_l_f1": 0.29360967184801384, "generation_token_f1": 0.5492227979274611, "prompt_id": "d655055a06f3b252", "reading_canonical_log_prob": -296.85791015625, "reading_distances": [30, 34, 35, 31], "reading_log_probs": [-305.40399169921875, -302.9452209472656, -305.96722412109375, -307.3623352050781], "reading_logprob_std": 1.8436293496006433, "tokenization_path_logprob_std": 359.79244647273805, "trajectory_balance_rmse": 6153.058026144095} {"canonical_length": 270, "canonical_nll": 1.432325122974537, "generation_rouge_l_f1": 0.29328621908127206, "generation_token_f1": 0.46289752650176674, "prompt_id": "62d551bf467199ab", "reading_canonical_log_prob": -386.727783203125, "reading_distances": [18, 19, 24, 20], "reading_log_probs": [-387.2098388671875, -386.6791687011719, -388.84295654296875, -384.66851806640625], "reading_logprob_std": 1.7213760477819406, "tokenization_path_logprob_std": 208.84911650420628, "trajectory_balance_rmse": 4947.756050502163} {"canonical_length": 11, "canonical_nll": 0.8358870419588956, "generation_rouge_l_f1": 0.7407407407407407, "generation_token_f1": 0.7407407407407407, "prompt_id": "1237b2c3cf7851d2", "reading_canonical_log_prob": -9.194757461547852, "reading_distances": [20, 20, 16, 18], "reading_log_probs": [-12.49440860748291, -11.721536636352539, -11.733133316040039, -14.959097862243652], "reading_logprob_std": 1.5313487710563485, "tokenization_path_logprob_std": 67.15753231217808, "trajectory_balance_rmse": 294.8373374506349} {"canonical_length": 3, "canonical_nll": 1.94081449508667, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "9af597ad9c807de8", "reading_canonical_log_prob": -5.82244348526001, "reading_distances": [17, 19, 21, 18], "reading_log_probs": [-5.596325874328613, -6.900830268859863, -7.209153175354004, -10.542126655578613], "reading_logprob_std": 2.106076597146494, "tokenization_path_logprob_std": 14.380339314602542, "trajectory_balance_rmse": 36.170628253476266} {"canonical_length": 158, "canonical_nll": 1.2686240280730814, "generation_rouge_l_f1": 0.32634032634032634, "generation_token_f1": 0.5174825174825175, "prompt_id": "0366eeefae975865", "reading_canonical_log_prob": -200.44259643554688, "reading_distances": [31, 34, 31, 32], "reading_log_probs": [-199.49205017089844, -201.46273803710938, -204.74722290039062, -198.43865966796875], "reading_logprob_std": 2.7740004730200396, "tokenization_path_logprob_std": 157.0944020028253, "trajectory_balance_rmse": 2760.52610564238} {"canonical_length": 94, "canonical_nll": 1.7816454298952793, "generation_rouge_l_f1": 0.1756373937677054, "generation_token_f1": 0.31728045325779036, "prompt_id": "e99bd769bc8f61be", "reading_canonical_log_prob": -167.47467041015625, "reading_distances": [18, 16, 21, 22], "reading_log_probs": [-168.34442138671875, -173.59457397460938, -173.77706909179688, -171.2227783203125], "reading_logprob_std": 2.5420762898272335, "tokenization_path_logprob_std": 317.0219145822422, "trajectory_balance_rmse": 2267.9430313083867} {"canonical_length": 287, "canonical_nll": 0.3326626555014155, "generation_rouge_l_f1": 0.596401028277635, "generation_token_f1": 0.6940874035989717, "prompt_id": "e1ea57924db21e38", "reading_canonical_log_prob": -95.47418212890625, "reading_distances": [20, 19, 16, 21], "reading_log_probs": [-97.17064666748047, -95.32450866699219, -96.02246856689453, -96.05879211425781], "reading_logprob_std": 0.7632386377540004, "tokenization_path_logprob_std": 117.35618705084428, "trajectory_balance_rmse": 2507.0717229389043} {"canonical_length": 248, "canonical_nll": 0.8499796467442666, "generation_rouge_l_f1": 0.4007285974499089, "generation_token_f1": 0.6302367941712205, "prompt_id": "d13dad8f3ef63531", "reading_canonical_log_prob": -210.79495239257812, "reading_distances": [29, 34, 32, 32], "reading_log_probs": [-215.68955993652344, -217.67654418945312, -223.22760009765625, -222.66372680664062], "reading_logprob_std": 3.7127407291870953, "tokenization_path_logprob_std": 218.36919498588787, "trajectory_balance_rmse": 3411.235134620288} {"canonical_length": 21, "canonical_nll": 1.4473465510777064, "generation_rouge_l_f1": 0.5294117647058824, "generation_token_f1": 0.5294117647058824, "prompt_id": "146de30da91fe96b", "reading_canonical_log_prob": -30.394277572631836, "reading_distances": [19, 19, 18, 14], "reading_log_probs": [-29.251296997070312, -29.343189239501953, -27.3831787109375, -31.427648544311523], "reading_logprob_std": 1.652754374545346, "tokenization_path_logprob_std": 55.66468320308638, "trajectory_balance_rmse": 452.1812740496736} {"canonical_length": 141, "canonical_nll": 1.077667236328125, "generation_rouge_l_f1": 0.31976744186046513, "generation_token_f1": 0.48837209302325585, "prompt_id": "e8d40d5d31440017", "reading_canonical_log_prob": -151.95108032226562, "reading_distances": [76, 71, 76, 68], "reading_log_probs": [-168.0336151123047, -167.21139526367188, -167.434326171875, -164.91763305664062], "reading_logprob_std": 1.365934490051066, "tokenization_path_logprob_std": 106.12514241421594, "trajectory_balance_rmse": 2673.582021215148} {"canonical_length": 7, "canonical_nll": 0.18645589692252024, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "e77edca13c4c71bd", "reading_canonical_log_prob": -1.3051912784576416, "reading_distances": [13, 20, 17, 13], "reading_log_probs": [-4.4195380210876465, -2.9704198837280273, -4.866613388061523, -3.5542070865631104], "reading_logprob_std": 0.851830626274371, "tokenization_path_logprob_std": 29.57360995870941, "trajectory_balance_rmse": 212.851735631348} {"canonical_length": 255, "canonical_nll": 1.383279718137255, "generation_rouge_l_f1": 0.16015625, "generation_token_f1": 0.34765625, "prompt_id": "5dc83250d862853a", "reading_canonical_log_prob": -352.736328125, "reading_distances": [31, 29, 31, 27], "reading_log_probs": [-370.9789123535156, -371.2861022949219, -366.09765625, -365.0389404296875], "reading_logprob_std": 3.2438682693116068, "tokenization_path_logprob_std": 323.5801175891341, "trajectory_balance_rmse": 5640.452886915203} {"canonical_length": 25, "canonical_nll": 1.3657568359375, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.3272727272727273, "prompt_id": "a9f2944f6ac15039", "reading_canonical_log_prob": -34.1439208984375, "reading_distances": [26, 27, 34, 28], "reading_log_probs": [-36.79048156738281, -37.9655647277832, -37.28938293457031, -36.23051452636719], "reading_logprob_std": 0.7377956706669592, "tokenization_path_logprob_std": 63.56909509459799, "trajectory_balance_rmse": 424.8990933263347} {"canonical_length": 51, "canonical_nll": 1.460252649643842, "generation_rouge_l_f1": 0.16597510373443983, "generation_token_f1": 0.2987551867219917, "prompt_id": "0e9e30fa66e7e5e1", "reading_canonical_log_prob": -74.47288513183594, "reading_distances": [14, 18, 21, 19], "reading_log_probs": [-91.57363891601562, -86.79145812988281, -89.86898803710938, -88.72236633300781], "reading_logprob_std": 2.0087139580652065, "tokenization_path_logprob_std": 73.81440902177852, "trajectory_balance_rmse": 921.6389758368086} {"canonical_length": 102, "canonical_nll": 2.3627813002642464, "generation_rouge_l_f1": 0.15160349854227406, "generation_token_f1": 0.2099125364431487, "prompt_id": "f6e26ca6d4af6e82", "reading_canonical_log_prob": -241.00369262695312, "reading_distances": [27, 34, 25, 22], "reading_log_probs": [-254.53152465820312, -272.5957946777344, -272.01470947265625, -254.6605224609375], "reading_logprob_std": 10.227315138767052, "tokenization_path_logprob_std": 51.659319972172284, "trajectory_balance_rmse": 2063.2446700679366} {"canonical_length": 92, "canonical_nll": 1.8865313322647759, "generation_rouge_l_f1": 0.1949860724233983, "generation_token_f1": 0.350974930362117, "prompt_id": "e5d1110bf220108e", "reading_canonical_log_prob": -173.56088256835938, "reading_distances": [30, 29, 24, 27], "reading_log_probs": [-184.5202178955078, -187.25653076171875, -183.93243408203125, -185.49667358398438], "reading_logprob_std": 1.454301804593818, "tokenization_path_logprob_std": 126.54297989163568, "trajectory_balance_rmse": 1864.9711907848018} {"canonical_length": 10, "canonical_nll": 1.117439079284668, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.5555555555555556, "prompt_id": "dd04d981a73d4900", "reading_canonical_log_prob": -11.17439079284668, "reading_distances": [20, 14, 16, 15], "reading_log_probs": [-15.922994613647461, -14.990853309631348, -16.202701568603516, -13.836228370666504], "reading_logprob_std": 1.0686302405866812, "tokenization_path_logprob_std": 23.823092775543504, "trajectory_balance_rmse": 132.3181764591026} {"canonical_length": 299, "canonical_nll": 1.6451598712831836, "generation_rouge_l_f1": 0.26642984014209586, "generation_token_f1": 0.5186500888099467, "prompt_id": "d9940a473d55b366", "reading_canonical_log_prob": -491.9028015136719, "reading_distances": [26, 27, 16, 26], "reading_log_probs": [-495.6156311035156, -491.7284240722656, -493.9034729003906, -494.64569091796875], "reading_logprob_std": 1.6526448252031896, "tokenization_path_logprob_std": 264.72468586568414, "trajectory_balance_rmse": 6018.04596896303} {"canonical_length": 416, "canonical_nll": 1.1316114572378306, "generation_rouge_l_f1": 0.23076923076923075, "generation_token_f1": 0.38141025641025644, "prompt_id": "7c0396dbe5594b7c", "reading_canonical_log_prob": -470.7503662109375, "reading_distances": [25, 21, 23, 25], "reading_log_probs": [-476.1591796875, -476.1550598144531, -477.8043518066406, -475.21722412109375], "reading_logprob_std": 1.075748502539476, "tokenization_path_logprob_std": 91.40725257215028, "trajectory_balance_rmse": 6269.093614758333} {"canonical_length": 138, "canonical_nll": 1.4942968617314878, "generation_rouge_l_f1": 0.20197044334975367, "generation_token_f1": 0.40394088669950734, "prompt_id": "5ee117416302abcd", "reading_canonical_log_prob": -206.2129669189453, "reading_distances": [20, 17, 19, 22], "reading_log_probs": [-228.3773651123047, -224.24879455566406, -229.89906311035156, -227.791748046875], "reading_logprob_std": 2.3913461680169394, "tokenization_path_logprob_std": 81.78823773672138, "trajectory_balance_rmse": 2696.782866383991} {"canonical_length": 326, "canonical_nll": 1.5321578511431173, "generation_rouge_l_f1": 0.22984562607204115, "generation_token_f1": 0.48370497427101206, "prompt_id": "7d6d0adc840aad0a", "reading_canonical_log_prob": -499.48345947265625, "reading_distances": [26, 24, 27, 19], "reading_log_probs": [-506.56317138671875, -508.1313171386719, -508.0253601074219, -508.8442687988281], "reading_logprob_std": 0.9570188418949634, "tokenization_path_logprob_std": 398.9852874823108, "trajectory_balance_rmse": 7240.31189957735} {"canonical_length": 17, "canonical_nll": 1.576725679285386, "generation_rouge_l_f1": 0.2580645161290323, "generation_token_f1": 0.38709677419354843, "prompt_id": "305ea9936db72d06", "reading_canonical_log_prob": -26.804336547851562, "reading_distances": [62, 54, 57, 53], "reading_log_probs": [-27.788713455200195, -22.536026000976562, -22.529050827026367, -23.14691162109375], "reading_logprob_std": 2.542244457427398, "tokenization_path_logprob_std": 14.757533584867845, "trajectory_balance_rmse": 331.3343045240164} {"canonical_length": 8, "canonical_nll": 1.3999344110488892, "generation_rouge_l_f1": 0.5454545454545454, "generation_token_f1": 0.5454545454545454, "prompt_id": "f9a917228bf6b5fe", "reading_canonical_log_prob": -11.199475288391113, "reading_distances": [32, 38, 38, 37], "reading_log_probs": [-12.235350608825684, -12.521687507629395, -12.625722885131836, -11.77822494506836], "reading_logprob_std": 0.3791609054109131, "tokenization_path_logprob_std": 24.772457416756435, "trajectory_balance_rmse": 118.69863798161961} {"canonical_length": 206, "canonical_nll": 1.2632992938884253, "generation_rouge_l_f1": 0.3661417322834646, "generation_token_f1": 0.5196850393700788, "prompt_id": "6fb0440b40107b60", "reading_canonical_log_prob": -260.2396545410156, "reading_distances": [13, 11, 13, 12], "reading_log_probs": [-265.54217529296875, -266.25799560546875, -264.78277587890625, -267.29205322265625], "reading_logprob_std": 1.0682220486956024, "tokenization_path_logprob_std": 166.70171928574365, "trajectory_balance_rmse": 2909.2914329258588} {"canonical_length": 94, "canonical_nll": 1.8769083225980718, "generation_rouge_l_f1": 0.22478386167146977, "generation_token_f1": 0.3688760806916426, "prompt_id": "16cf88f25ee512b8", "reading_canonical_log_prob": -176.42938232421875, "reading_distances": [17, 17, 16, 22], "reading_log_probs": [-180.2920684814453, -180.52491760253906, -180.25802612304688, -175.504638671875], "reading_logprob_std": 2.429745968226843, "tokenization_path_logprob_std": 51.073205549500166, "trajectory_balance_rmse": 2364.2631487435524} {"canonical_length": 12, "canonical_nll": 1.0896989504496257, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.28571428571428575, "prompt_id": "024090bc27e03da4", "reading_canonical_log_prob": -13.076387405395508, "reading_distances": [26, 19, 17, 29], "reading_log_probs": [-13.06888198852539, -11.524077415466309, -10.431750297546387, -14.176523208618164], "reading_logprob_std": 1.6537766165811676, "tokenization_path_logprob_std": 62.63412387500427, "trajectory_balance_rmse": 308.57779438179847} {"canonical_length": 58, "canonical_nll": 1.9525925208782327, "generation_rouge_l_f1": 0.12322274881516586, "generation_token_f1": 0.1990521327014218, "prompt_id": "d97dcaaab25ed5d3", "reading_canonical_log_prob": -113.2503662109375, "reading_distances": [11, 17, 10, 15], "reading_log_probs": [-137.5779571533203, -133.84109497070312, -135.99058532714844, -137.17236328125], "reading_logprob_std": 1.6773652742114495, "tokenization_path_logprob_std": 79.54705404113824, "trajectory_balance_rmse": 1314.9675624520044} {"canonical_length": 386, "canonical_nll": 0.8933710286036675, "generation_rouge_l_f1": 0.2901023890784983, "generation_token_f1": 0.4709897610921502, "prompt_id": "5a59813799d04978", "reading_canonical_log_prob": -344.8412170410156, "reading_distances": [9, 11, 12, 14], "reading_log_probs": [-345.63946533203125, -349.9737854003906, -345.4590759277344, -347.6976013183594], "reading_logprob_std": 2.1140286589952995, "tokenization_path_logprob_std": 319.06589787072215, "trajectory_balance_rmse": 5892.416463927329} {"canonical_length": 120, "canonical_nll": 1.1458994547526042, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.4266666666666667, "prompt_id": "84eecbd6c7c38a00", "reading_canonical_log_prob": -137.5079345703125, "reading_distances": [20, 24, 27, 23], "reading_log_probs": [-135.450439453125, -135.99533081054688, -136.2484130859375, -137.37628173828125], "reading_logprob_std": 0.8106427911036562, "tokenization_path_logprob_std": 98.2077873910674, "trajectory_balance_rmse": 2265.0855331993166} {"canonical_length": 12, "canonical_nll": 0.3581600983937581, "generation_rouge_l_f1": 0.8000000000000002, "generation_token_f1": 0.8000000000000002, "prompt_id": "1914d368729c6a5a", "reading_canonical_log_prob": -4.297921180725098, "reading_distances": [23, 22, 23, 27], "reading_log_probs": [-5.356057167053223, -4.17108678817749, -6.462409019470215, -4.639513969421387], "reading_logprob_std": 0.9972516954723096, "tokenization_path_logprob_std": 35.17624319733041, "trajectory_balance_rmse": 250.0974209810788} {"canonical_length": 313, "canonical_nll": 1.1933443599615614, "generation_rouge_l_f1": 0.20350877192982456, "generation_token_f1": 0.4035087719298246, "prompt_id": "dc582c1cebaf17e1", "reading_canonical_log_prob": -373.51678466796875, "reading_distances": [43, 44, 34, 50], "reading_log_probs": [-396.2080078125, -397.46148681640625, -387.88812255859375, -398.82745361328125], "reading_logprob_std": 4.923052313534834, "tokenization_path_logprob_std": 257.889722532804, "trajectory_balance_rmse": 7307.137536042447} {"canonical_length": 34, "canonical_nll": 1.350354138542624, "generation_rouge_l_f1": 0.14583333333333334, "generation_token_f1": 0.16666666666666666, "prompt_id": "c32c06c0fe76b404", "reading_canonical_log_prob": -45.91204071044922, "reading_distances": [19, 19, 20, 22], "reading_log_probs": [-46.362701416015625, -55.712860107421875, -46.229164123535156, -45.474342346191406], "reading_logprob_std": 4.861155323731138, "tokenization_path_logprob_std": 42.35751494696103, "trajectory_balance_rmse": 522.6316942763283} {"canonical_length": 124, "canonical_nll": 1.8177786796323714, "generation_rouge_l_f1": 0.2198952879581152, "generation_token_f1": 0.39267015706806285, "prompt_id": "3415a33aecea8511", "reading_canonical_log_prob": -225.40455627441406, "reading_distances": [22, 20, 19, 23], "reading_log_probs": [-229.77435302734375, -230.0027618408203, -227.15028381347656, -227.71697998046875], "reading_logprob_std": 1.4391336466349534, "tokenization_path_logprob_std": 44.34482054418409, "trajectory_balance_rmse": 3201.1107787608216} {"canonical_length": 333, "canonical_nll": 1.449696950368337, "generation_rouge_l_f1": 0.22628951747088186, "generation_token_f1": 0.48252911813643923, "prompt_id": "f7e2c3707f2870f0", "reading_canonical_log_prob": -482.74908447265625, "reading_distances": [58, 53, 59, 53], "reading_log_probs": [-487.7235107421875, -483.6065368652344, -482.7237548828125, -487.34796142578125], "reading_logprob_std": 2.5535745823046656, "tokenization_path_logprob_std": 190.2006374626729, "trajectory_balance_rmse": 6859.339873477022} {"canonical_length": 39, "canonical_nll": 1.7264196933844151, "generation_rouge_l_f1": 0.11501597444089456, "generation_token_f1": 0.16613418530351437, "prompt_id": "ec9f4061e1384314", "reading_canonical_log_prob": -67.33036804199219, "reading_distances": [19, 18, 19, 16], "reading_log_probs": [-72.27804565429688, -73.33887481689453, -73.52194213867188, -71.72428131103516], "reading_logprob_std": 0.858841380805741, "tokenization_path_logprob_std": 67.20041398439304, "trajectory_balance_rmse": 832.0600743557645} {"canonical_length": 14, "canonical_nll": 1.3416384288242884, "generation_rouge_l_f1": 0.3870967741935484, "generation_token_f1": 0.3870967741935484, "prompt_id": "a05868d3246d849c", "reading_canonical_log_prob": -18.78293800354004, "reading_distances": [10, 9, 8, 9], "reading_log_probs": [-18.62823486328125, -18.60723304748535, -19.01409149169922, -18.860361099243164], "reading_logprob_std": 0.19503197198625757, "tokenization_path_logprob_std": 78.20637031700414, "trajectory_balance_rmse": 391.041063626299} {"canonical_length": 127, "canonical_nll": 1.4371631051611713, "generation_rouge_l_f1": 0.23728813559322035, "generation_token_f1": 0.41162227602905566, "prompt_id": "420ee5406fba3002", "reading_canonical_log_prob": -182.51971435546875, "reading_distances": [21, 24, 21, 25], "reading_log_probs": [-188.307861328125, -189.379638671875, -186.75987243652344, -186.51812744140625], "reading_logprob_std": 1.3496291453895193, "tokenization_path_logprob_std": 134.82762509135554, "trajectory_balance_rmse": 2937.2783251592246} {"canonical_length": 9, "canonical_nll": 3.0420989990234375, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "0a4e72c4105acd5f", "reading_canonical_log_prob": -27.378890991210938, "reading_distances": [16, 8, 15, 16], "reading_log_probs": [-26.754789352416992, -27.448244094848633, -24.874956130981445, -27.171524047851562], "reading_logprob_std": 1.1604930472570545, "tokenization_path_logprob_std": 26.091260591488798, "trajectory_balance_rmse": 151.61543158127367} {"canonical_length": 16, "canonical_nll": 0.7182347774505615, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "0ce850d0751da7a2", "reading_canonical_log_prob": -11.491756439208984, "reading_distances": [61, 60, 73, 64], "reading_log_probs": [-23.667892456054688, -18.773284912109375, -20.84900665283203, -25.730281829833984], "reading_logprob_std": 3.0644741843923815, "tokenization_path_logprob_std": 26.175614047298023, "trajectory_balance_rmse": 338.5418633722003} {"canonical_length": 9, "canonical_nll": 3.8234613206651478, "generation_rouge_l_f1": 0.30769230769230765, "generation_token_f1": 0.30769230769230765, "prompt_id": "eeee24292d24cae3", "reading_canonical_log_prob": -34.41115188598633, "reading_distances": [19, 19, 26, 23], "reading_log_probs": [-42.794551849365234, -39.9150390625, -43.84022521972656, -42.279876708984375], "reading_logprob_std": 1.660425350259621, "tokenization_path_logprob_std": 33.554867365362924, "trajectory_balance_rmse": 182.474901403933} {"canonical_length": 20, "canonical_nll": 2.2813703536987306, "generation_rouge_l_f1": 0.325, "generation_token_f1": 0.375, "prompt_id": "aa417b16bf366ab6", "reading_canonical_log_prob": -45.62740707397461, "reading_distances": [31, 33, 27, 38], "reading_log_probs": [-35.899513244628906, -37.109962463378906, -38.703128814697266, -37.45338821411133], "reading_logprob_std": 1.1531820770762342, "tokenization_path_logprob_std": 56.810095335457085, "trajectory_balance_rmse": 420.72870125422224} {"canonical_length": 277, "canonical_nll": 1.4702505394009477, "generation_rouge_l_f1": 0.23300970873786409, "generation_token_f1": 0.46601941747572817, "prompt_id": "31ffef5f718912de", "reading_canonical_log_prob": -407.2593994140625, "reading_distances": [15, 18, 20, 15], "reading_log_probs": [-415.7868347167969, -420.38818359375, -420.3815612792969, -414.98193359375], "reading_logprob_std": 2.905674786080048, "tokenization_path_logprob_std": 266.0358857446212, "trajectory_balance_rmse": 6012.366388210549} {"canonical_length": 170, "canonical_nll": 1.0882064819335937, "generation_rouge_l_f1": 0.18952618453865339, "generation_token_f1": 0.2793017456359102, "prompt_id": "fe5c1f86ece4d243", "reading_canonical_log_prob": -184.99510192871094, "reading_distances": [18, 14, 18, 19], "reading_log_probs": [-185.30398559570312, -176.71597290039062, -181.05148315429688, -171.91445922851562], "reading_logprob_std": 5.747851369061398, "tokenization_path_logprob_std": 56.178271510666356, "trajectory_balance_rmse": 3143.3020110921857} {"canonical_length": 166, "canonical_nll": 1.4711618078760353, "generation_rouge_l_f1": 0.24875621890547267, "generation_token_f1": 0.4626865671641791, "prompt_id": "8db7426fa0df6c62", "reading_canonical_log_prob": -244.21286010742188, "reading_distances": [18, 17, 24, 22], "reading_log_probs": [-256.47064208984375, -254.68003845214844, -254.86720275878906, -258.5567321777344], "reading_logprob_std": 1.7982737783644274, "tokenization_path_logprob_std": 94.09638648972678, "trajectory_balance_rmse": 2158.560247484685} {"canonical_length": 39, "canonical_nll": 0.7905854934301132, "generation_rouge_l_f1": 0.34710743801652894, "generation_token_f1": 0.42975206611570255, "prompt_id": "6c3aa6c41d3147aa", "reading_canonical_log_prob": -30.832834243774414, "reading_distances": [49, 49, 41, 43], "reading_log_probs": [-37.907203674316406, -38.28547668457031, -37.822139739990234, -35.902034759521484], "reading_logprob_std": 1.0705648193976707, "tokenization_path_logprob_std": 92.91355468565575, "trajectory_balance_rmse": 737.7146532811097} {"canonical_length": 8, "canonical_nll": 2.2708425521850586, "generation_rouge_l_f1": 0.0390625, "generation_token_f1": 0.0390625, "prompt_id": "e07ed69174530f07", "reading_canonical_log_prob": -18.16674041748047, "reading_distances": [35, 38, 33, 38], "reading_log_probs": [-16.574121475219727, -17.342222213745117, -18.13024139404297, -17.464126586914062], "reading_logprob_std": 0.6379094035803689, "tokenization_path_logprob_std": 48.587411739039865, "trajectory_balance_rmse": 118.65712406651407} {"canonical_length": 57, "canonical_nll": 1.9452858640436541, "generation_rouge_l_f1": 0.12751677852348994, "generation_token_f1": 0.18120805369127518, "prompt_id": "93c96f6a3b9968fc", "reading_canonical_log_prob": -110.88129425048828, "reading_distances": [25, 18, 29, 22], "reading_log_probs": [-115.6377944946289, -115.27538299560547, -117.4475326538086, -118.35078430175781], "reading_logprob_std": 1.4651222606560663, "tokenization_path_logprob_std": 165.14142074139315, "trajectory_balance_rmse": 1256.214799977173} {"canonical_length": 264, "canonical_nll": 1.3240487069794626, "generation_rouge_l_f1": 0.2598425196850394, "generation_token_f1": 0.515748031496063, "prompt_id": "0a9cdb872ed6cdad", "reading_canonical_log_prob": -349.5488586425781, "reading_distances": [23, 20, 21, 19], "reading_log_probs": [-347.91748046875, -345.5214538574219, -345.69354248046875, -345.3865661621094], "reading_logprob_std": 1.198416127298637, "tokenization_path_logprob_std": 245.04807841764307, "trajectory_balance_rmse": 5000.443423470675} {"canonical_length": 141, "canonical_nll": 1.5484667838887964, "generation_rouge_l_f1": 0.2123287671232877, "generation_token_f1": 0.4109589041095891, "prompt_id": "c6291fd5f11c978c", "reading_canonical_log_prob": -218.3338165283203, "reading_distances": [23, 17, 21, 19], "reading_log_probs": [-222.14283752441406, -225.71539306640625, -228.91488647460938, -228.85260009765625], "reading_logprob_std": 3.21101514961333, "tokenization_path_logprob_std": 186.26609207453743, "trajectory_balance_rmse": 3204.918977142485} {"canonical_length": 9, "canonical_nll": 0.3405393759409587, "generation_rouge_l_f1": 0.6956521739130436, "generation_token_f1": 0.6956521739130436, "prompt_id": "c5b68b2a56c5748c", "reading_canonical_log_prob": -3.064854383468628, "reading_distances": [48, 43, 38, 41], "reading_log_probs": [-2.071141481399536, -2.494964599609375, -2.3323965072631836, -2.723292589187622], "reading_logprob_std": 0.27455161837001846, "tokenization_path_logprob_std": 49.08680685042078, "trajectory_balance_rmse": 120.54389516239215} {"canonical_length": 101, "canonical_nll": 1.924111545676052, "generation_rouge_l_f1": 0.234375, "generation_token_f1": 0.359375, "prompt_id": "4478e3a9a8e3ebd2", "reading_canonical_log_prob": -194.33526611328125, "reading_distances": [28, 21, 23, 19], "reading_log_probs": [-196.10623168945312, -195.9689483642578, -192.75547790527344, -190.67845153808594], "reading_logprob_std": 2.635289082713679, "tokenization_path_logprob_std": 95.85275723265573, "trajectory_balance_rmse": 1187.108095931738} {"canonical_length": 153, "canonical_nll": 1.6524331086601307, "generation_rouge_l_f1": 0.24056603773584903, "generation_token_f1": 0.4245283018867924, "prompt_id": "1679ec014dbdb2b7", "reading_canonical_log_prob": -252.822265625, "reading_distances": [25, 25, 20, 28], "reading_log_probs": [-266.4521789550781, -263.2154541015625, -259.601318359375, -268.8554992675781], "reading_logprob_std": 4.017654561671142, "tokenization_path_logprob_std": 130.65483709059652, "trajectory_balance_rmse": 3223.83124074605} {"canonical_length": 6, "canonical_nll": 0.8120741844177246, "generation_rouge_l_f1": 0.20833333333333334, "generation_token_f1": 0.20833333333333334, "prompt_id": "d608b8ecb5eb0aa8", "reading_canonical_log_prob": -4.872445106506348, "reading_distances": [6, 12, 17, 13], "reading_log_probs": [-5.190810203552246, -4.534075736999512, -5.322995662689209, -8.725790977478027], "reading_logprob_std": 1.8867244430482621, "tokenization_path_logprob_std": 34.88985982472092, "trajectory_balance_rmse": 133.62545949285678} {"canonical_length": 8, "canonical_nll": 0.3435807526111603, "generation_rouge_l_f1": 0.923076923076923, "generation_token_f1": 0.923076923076923, "prompt_id": "c7c5da8841670598", "reading_canonical_log_prob": -2.7486460208892822, "reading_distances": [23, 23, 28, 23], "reading_log_probs": [-2.3532485961914062, -2.513376235961914, -2.68441104888916, -1.927518367767334], "reading_logprob_std": 0.32428440940317865, "tokenization_path_logprob_std": 42.692184463621416, "trajectory_balance_rmse": 214.59238870010438} {"canonical_length": 72, "canonical_nll": 1.6951235665215387, "generation_rouge_l_f1": 0.2763157894736842, "generation_token_f1": 0.4342105263157895, "prompt_id": "8b986ced31f57ea9", "reading_canonical_log_prob": -122.04889678955078, "reading_distances": [31, 31, 35, 33], "reading_log_probs": [-138.27511596679688, -127.1943588256836, -155.0845947265625, -136.41799926757812], "reading_logprob_std": 11.619571223684291, "tokenization_path_logprob_std": 151.7731281328934, "trajectory_balance_rmse": 1488.4813853184826} {"canonical_length": 288, "canonical_nll": 1.2767437828911676, "generation_rouge_l_f1": 0.20493358633776093, "generation_token_f1": 0.4478178368121442, "prompt_id": "3d6847202e1bf0e3", "reading_canonical_log_prob": -367.70220947265625, "reading_distances": [24, 20, 25, 23], "reading_log_probs": [-375.49517822265625, -371.0196228027344, -369.73931884765625, -371.701416015625], "reading_logprob_std": 2.4749914249708267, "tokenization_path_logprob_std": 315.1708910952582, "trajectory_balance_rmse": 5922.777364284756} {"canonical_length": 6, "canonical_nll": 0.6992433071136475, "generation_rouge_l_f1": 0.8000000000000002, "generation_token_f1": 0.8000000000000002, "prompt_id": "81930576a5952ad4", "reading_canonical_log_prob": -4.195459842681885, "reading_distances": [48, 43, 45, 39], "reading_log_probs": [-1.9793037176132202, -3.2920403480529785, -2.2023940086364746, -2.0455081462860107], "reading_logprob_std": 0.6153057921387748, "tokenization_path_logprob_std": 18.890995787452606, "trajectory_balance_rmse": 80.4340645319547} {"canonical_length": 271, "canonical_nll": 1.1150502560323454, "generation_rouge_l_f1": 0.260707635009311, "generation_token_f1": 0.5363128491620113, "prompt_id": "52dc7c85e884efbf", "reading_canonical_log_prob": -302.1786193847656, "reading_distances": [16, 11, 11, 19], "reading_log_probs": [-308.7040710449219, -306.25421142578125, -307.54595947265625, -310.36517333984375], "reading_logprob_std": 1.7468753796881213, "tokenization_path_logprob_std": 129.8533334493668, "trajectory_balance_rmse": 5853.490305134392} {"canonical_length": 220, "canonical_nll": 1.128641718084162, "generation_rouge_l_f1": 0.2979683972911964, "generation_token_f1": 0.4966139954853273, "prompt_id": "02a68089e544097d", "reading_canonical_log_prob": -248.30117797851562, "reading_distances": [27, 21, 23, 25], "reading_log_probs": [-246.69076538085938, -244.91583251953125, -246.73655700683594, -244.37283325195312], "reading_logprob_std": 1.215262959956489, "tokenization_path_logprob_std": 369.57774585333544, "trajectory_balance_rmse": 3792.047278500309} {"canonical_length": 116, "canonical_nll": 1.7027141965668777, "generation_rouge_l_f1": 0.27705627705627706, "generation_token_f1": 0.45021645021645024, "prompt_id": "4fcd7cfb63238642", "reading_canonical_log_prob": -197.5148468017578, "reading_distances": [28, 26, 36, 35], "reading_log_probs": [-199.23382568359375, -199.5247802734375, -201.82937622070312, -203.30520629882812], "reading_logprob_std": 1.9403293344010895, "tokenization_path_logprob_std": 60.91163101629825, "trajectory_balance_rmse": 3047.1566467320295} {"canonical_length": 260, "canonical_nll": 1.5273456280048077, "generation_rouge_l_f1": 0.16296296296296295, "generation_token_f1": 0.3851851851851852, "prompt_id": "0b49b7e13d1bd9b0", "reading_canonical_log_prob": -397.10986328125, "reading_distances": [35, 29, 27, 24], "reading_log_probs": [-407.38720703125, -403.7064208984375, -403.5939636230469, -402.99212646484375], "reading_logprob_std": 2.0028872167496194, "tokenization_path_logprob_std": 184.41498947745538, "trajectory_balance_rmse": 4974.254579075872} {"canonical_length": 169, "canonical_nll": 1.2326342340051775, "generation_rouge_l_f1": 0.20737327188940094, "generation_token_f1": 0.4193548387096774, "prompt_id": "112cdc2afc089b69", "reading_canonical_log_prob": -208.315185546875, "reading_distances": [23, 27, 25, 31], "reading_log_probs": [-221.88006591796875, -218.67877197265625, -221.676513671875, -221.99281311035156], "reading_logprob_std": 1.5909063890050967, "tokenization_path_logprob_std": 86.9836626856821, "trajectory_balance_rmse": 3040.0027705040548} {"canonical_length": 250, "canonical_nll": 1.14650048828125, "generation_rouge_l_f1": 0.30452674897119336, "generation_token_f1": 0.5020576131687243, "prompt_id": "845212ac338f2658", "reading_canonical_log_prob": -286.6251220703125, "reading_distances": [41, 42, 28, 40], "reading_log_probs": [-291.2968444824219, -299.10772705078125, -293.2535400390625, -295.329345703125], "reading_logprob_std": 3.3411107238025006, "tokenization_path_logprob_std": 118.98718920105834, "trajectory_balance_rmse": 5292.754779800335} {"canonical_length": 30, "canonical_nll": 0.5365698496500652, "generation_rouge_l_f1": 0.4193548387096774, "generation_token_f1": 0.5806451612903226, "prompt_id": "9ca4c748ffa8cd48", "reading_canonical_log_prob": -16.097095489501953, "reading_distances": [46, 53, 45, 52], "reading_log_probs": [-15.197861671447754, -14.448274612426758, -15.166574478149414, -15.021445274353027], "reading_logprob_std": 0.3487506491349843, "tokenization_path_logprob_std": 114.2359498286305, "trajectory_balance_rmse": 594.4337987383811} {"canonical_length": 501, "canonical_nll": 0.5833335769866517, "generation_rouge_l_f1": 0.29009433962264153, "generation_token_f1": 0.4528301886792453, "prompt_id": "585bee96b6b0ece7", "reading_canonical_log_prob": -292.2501220703125, "reading_distances": [19, 22, 22, 17], "reading_log_probs": [-302.2265625, -301.1133117675781, -303.6290283203125, -302.18707275390625], "reading_logprob_std": 1.0315370931842947, "tokenization_path_logprob_std": 154.1228823105096, "trajectory_balance_rmse": 8101.397981956908} {"canonical_length": 168, "canonical_nll": 1.2066795712425595, "generation_rouge_l_f1": 0.13106796116504857, "generation_token_f1": 0.24271844660194175, "prompt_id": "b3cb4c0e99b9a63e", "reading_canonical_log_prob": -202.72216796875, "reading_distances": [21, 25, 25, 26], "reading_log_probs": [-208.54617309570312, -209.91122436523438, -209.0460205078125, -210.22113037109375], "reading_logprob_std": 0.7715891024573671, "tokenization_path_logprob_std": 214.98294537526118, "trajectory_balance_rmse": 3513.0000089831715} {"canonical_length": 430, "canonical_nll": 0.9874715405841207, "generation_rouge_l_f1": 0.21693907875185736, "generation_token_f1": 0.424962852897474, "prompt_id": "2fa618fc412d053e", "reading_canonical_log_prob": -424.6127624511719, "reading_distances": [27, 29, 26, 25], "reading_log_probs": [-459.38702392578125, -459.14459228515625, -456.9351806640625, -453.6020812988281], "reading_logprob_std": 2.6808945770810184, "tokenization_path_logprob_std": 168.77711010781957, "trajectory_balance_rmse": 7002.357056989572} {"canonical_length": 160, "canonical_nll": 1.085921859741211, "generation_rouge_l_f1": 0.26107226107226106, "generation_token_f1": 0.5314685314685315, "prompt_id": "72f894b5c3eed051", "reading_canonical_log_prob": -173.74749755859375, "reading_distances": [27, 26, 28, 25], "reading_log_probs": [-183.71107482910156, -181.74148559570312, -185.94473266601562, -184.1089324951172], "reading_logprob_std": 1.724071074188471, "tokenization_path_logprob_std": 271.5862328335781, "trajectory_balance_rmse": 3344.7274039694257} {"canonical_length": 98, "canonical_nll": 1.3949510224011479, "generation_rouge_l_f1": 0.22284122562674094, "generation_token_f1": 0.3231197771587743, "prompt_id": "9c00b56f3aa02b5b", "reading_canonical_log_prob": -136.7052001953125, "reading_distances": [37, 31, 42, 34], "reading_log_probs": [-142.61355590820312, -135.6832733154297, -137.62815856933594, -138.5054931640625], "reading_logprob_std": 2.9194234942858976, "tokenization_path_logprob_std": 172.97366727329612, "trajectory_balance_rmse": 2165.939666217033} {"canonical_length": 10, "canonical_nll": 0.6755030632019043, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.4444444444444444, "prompt_id": "363eae7c1280c1a6", "reading_canonical_log_prob": -6.755030632019043, "reading_distances": [65, 53, 54, 43], "reading_log_probs": [-2.553571939468384, -4.5313849449157715, -3.0991857051849365, -4.176835060119629], "reading_logprob_std": 0.9211700096439104, "tokenization_path_logprob_std": 51.68933061938301, "trajectory_balance_rmse": 210.48886449414695} {"canonical_length": 7, "canonical_nll": 1.803654670715332, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "86c615edc3ae2a5f", "reading_canonical_log_prob": -12.625582695007324, "reading_distances": [42, 39, 44, 43], "reading_log_probs": [-20.823503494262695, -25.72819709777832, -24.23175811767578, -22.777664184570312], "reading_logprob_std": 2.0926523478056005, "tokenization_path_logprob_std": 22.774349478257463, "trajectory_balance_rmse": 123.34294935252794} {"canonical_length": 28, "canonical_nll": 0.9224012919834682, "generation_rouge_l_f1": 0.27118644067796616, "generation_token_f1": 0.27118644067796616, "prompt_id": "b424089c85e11874", "reading_canonical_log_prob": -25.82723617553711, "reading_distances": [23, 24, 20, 16], "reading_log_probs": [-27.790895462036133, -27.551193237304688, -26.780292510986328, -28.637008666992188], "reading_logprob_std": 0.7646001677331964, "tokenization_path_logprob_std": 66.8185098157577, "trajectory_balance_rmse": 523.7601534462649} {"canonical_length": 7, "canonical_nll": 0.18448638916015625, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "69e0fa630dc3c290", "reading_canonical_log_prob": -1.2914047241210938, "reading_distances": [41, 32, 25, 23], "reading_log_probs": [-3.6350131034851074, -4.451851844787598, -3.775564432144165, -3.5716054439544678], "reading_logprob_std": 0.4046395923236925, "tokenization_path_logprob_std": 13.774925354394831, "trajectory_balance_rmse": 101.33553880089637} {"canonical_length": 158, "canonical_nll": 1.3048841259147548, "generation_rouge_l_f1": 0.20652173913043478, "generation_token_f1": 0.3804347826086957, "prompt_id": "edfb4a427bd0504a", "reading_canonical_log_prob": -206.17169189453125, "reading_distances": [31, 25, 27, 23], "reading_log_probs": [-213.89691162109375, -211.21102905273438, -208.45193481445312, -208.1778106689453], "reading_logprob_std": 2.684182248583967, "tokenization_path_logprob_std": 127.15538163627677, "trajectory_balance_rmse": 3263.6341915528924} {"canonical_length": 9, "canonical_nll": 0.38617833455403644, "generation_rouge_l_f1": 0.8421052631578948, "generation_token_f1": 0.8421052631578948, "prompt_id": "9a5bfd989c2d5e52", "reading_canonical_log_prob": -3.475605010986328, "reading_distances": [37, 40, 36, 44], "reading_log_probs": [-3.891145706176758, -5.418067932128906, -4.371074199676514, -3.4802913665771484], "reading_logprob_std": 0.8354286365337721, "tokenization_path_logprob_std": 35.95963969650777, "trajectory_balance_rmse": 126.29549619939462} {"canonical_length": 272, "canonical_nll": 0.8588823430678424, "generation_rouge_l_f1": 0.41637010676156583, "generation_token_f1": 0.5836298932384343, "prompt_id": "a2cdcfce41950902", "reading_canonical_log_prob": -233.61599731445312, "reading_distances": [29, 18, 27, 24], "reading_log_probs": [-243.7224884033203, -241.18978881835938, -244.618408203125, -244.13763427734375], "reading_logprob_std": 1.5293231856886504, "tokenization_path_logprob_std": 233.67469214366602, "trajectory_balance_rmse": 3519.0503116203795} {"canonical_length": 72, "canonical_nll": 1.3114681243896484, "generation_rouge_l_f1": 0.20261437908496732, "generation_token_f1": 0.30065359477124187, "prompt_id": "41dfa00e84da9369", "reading_canonical_log_prob": -94.42570495605469, "reading_distances": [44, 37, 39, 41], "reading_log_probs": [-112.98055267333984, -108.263916015625, -118.80311584472656, -116.8988037109375], "reading_logprob_std": 4.661580719836471, "tokenization_path_logprob_std": 90.6006739321975, "trajectory_balance_rmse": 831.2270000902117} {"canonical_length": 13, "canonical_nll": 0.979535176203801, "generation_rouge_l_f1": 0.06324110671936758, "generation_token_f1": 0.07905138339920949, "prompt_id": "398d73f7e39625a8", "reading_canonical_log_prob": -12.733957290649414, "reading_distances": [27, 27, 24, 31], "reading_log_probs": [-18.91150665283203, -19.644580841064453, -22.35198974609375, -17.570968627929688], "reading_logprob_std": 2.013691829255116, "tokenization_path_logprob_std": 34.08411525331541, "trajectory_balance_rmse": 143.22407829753988} {"canonical_length": 89, "canonical_nll": 1.393503767720769, "generation_rouge_l_f1": 0.28820960698689957, "generation_token_f1": 0.445414847161572, "prompt_id": "d769bfcaa2a347d7", "reading_canonical_log_prob": -124.02183532714844, "reading_distances": [67, 74, 63, 79], "reading_log_probs": [-125.32975006103516, -127.72976684570312, -124.21145629882812, -125.16958618164062], "reading_logprob_std": 1.4968678683273813, "tokenization_path_logprob_std": 98.33681434617124, "trajectory_balance_rmse": 1966.4915044072688} {"canonical_length": 351, "canonical_nll": 1.3342682493378295, "generation_rouge_l_f1": 0.2288261515601783, "generation_token_f1": 0.3833580980683507, "prompt_id": "026d0e5d489e80b2", "reading_canonical_log_prob": -468.3281555175781, "reading_distances": [7, 12, 10, 9], "reading_log_probs": [-474.92578125, -481.90008544921875, -481.54949951171875, -475.6282958984375], "reading_logprob_std": 3.7363863603301315, "tokenization_path_logprob_std": 56.61418313853281, "trajectory_balance_rmse": 7210.613331329534} {"canonical_length": 55, "canonical_nll": 1.1180191733620384, "generation_rouge_l_f1": 0.3220338983050847, "generation_token_f1": 0.38983050847457623, "prompt_id": "4c12c24de5859788", "reading_canonical_log_prob": -61.49105453491211, "reading_distances": [20, 25, 24, 27], "reading_log_probs": [-62.79865264892578, -62.19993591308594, -62.35779571533203, -66.1134262084961], "reading_logprob_std": 1.8480987240836917, "tokenization_path_logprob_std": 86.44583270540849, "trajectory_balance_rmse": 1269.3256956448315} {"canonical_length": 298, "canonical_nll": 1.7559603492685611, "generation_rouge_l_f1": 0.18114602587800369, "generation_token_f1": 0.36598890942698714, "prompt_id": "1259351bdd54ab2c", "reading_canonical_log_prob": -523.2761840820312, "reading_distances": [18, 17, 17, 15], "reading_log_probs": [-528.4362182617188, -528.07275390625, -528.690185546875, -530.44677734375], "reading_logprob_std": 1.0544262814226286, "tokenization_path_logprob_std": 121.81680599292014, "trajectory_balance_rmse": 5409.13328838983} {"canonical_length": 17, "canonical_nll": 1.4025582706227022, "generation_rouge_l_f1": 0.07168458781362008, "generation_token_f1": 0.07168458781362008, "prompt_id": "1e2afb46510ad456", "reading_canonical_log_prob": -23.843490600585938, "reading_distances": [26, 30, 36, 28], "reading_log_probs": [-36.6720085144043, -35.32721710205078, -43.84822463989258, -47.387611389160156], "reading_logprob_std": 5.764247747779135, "tokenization_path_logprob_std": 58.31003183864337, "trajectory_balance_rmse": 290.0399950392798} {"canonical_length": 23, "canonical_nll": 1.305503098861031, "generation_rouge_l_f1": 0.10144927536231883, "generation_token_f1": 0.10144927536231883, "prompt_id": "dfa52b61eae824fc", "reading_canonical_log_prob": -30.02657127380371, "reading_distances": [14, 16, 20, 16], "reading_log_probs": [-26.327220916748047, -27.305757522583008, -27.023014068603516, -25.083667755126953], "reading_logprob_std": 0.9902347105569669, "tokenization_path_logprob_std": 27.873599019120068, "trajectory_balance_rmse": 257.5940968403463} {"canonical_length": 5, "canonical_nll": 1.1991461753845214, "generation_rouge_l_f1": 0.4444444444444445, "generation_token_f1": 0.4444444444444445, "prompt_id": "30fd5bfb1635c7a5", "reading_canonical_log_prob": -5.995730876922607, "reading_distances": [19, 19, 28, 23], "reading_log_probs": [-11.692852973937988, -9.757084846496582, -11.098726272583008, -9.798137664794922], "reading_logprob_std": 0.9653733586878737, "tokenization_path_logprob_std": 9.680748675399009, "trajectory_balance_rmse": 14.851865305718434} {"canonical_length": 11, "canonical_nll": 0.26299166679382324, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "bd172f5e24a2f430", "reading_canonical_log_prob": -2.8929083347320557, "reading_distances": [36, 40, 44, 39], "reading_log_probs": [-2.8635149002075195, -3.204739570617676, -2.233527898788452, -2.6895804405212402], "reading_logprob_std": 0.4041654009148393, "tokenization_path_logprob_std": 25.947922200985857, "trajectory_balance_rmse": 171.71778256386503} {"canonical_length": 8, "canonical_nll": 0.13502609729766846, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "74eb19d7e1933dab", "reading_canonical_log_prob": -1.0802087783813477, "reading_distances": [27, 23, 22, 23], "reading_log_probs": [-1.953102946281433, -2.5129873752593994, -1.8272545337677002, -1.4709663391113281], "reading_logprob_std": 0.4325065271226731, "tokenization_path_logprob_std": 37.23015743900947, "trajectory_balance_rmse": 161.6600158889415} {"canonical_length": 2, "canonical_nll": 1.7751214504241943, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "69770982bd8468eb", "reading_canonical_log_prob": -3.5502429008483887, "reading_distances": [27, 24, 27, 30], "reading_log_probs": [-3.0115928649902344, -3.2603137493133545, -3.262455940246582, -2.5286781787872314], "reading_logprob_std": 0.345413514890663, "tokenization_path_logprob_std": 17.6030092427689, "trajectory_balance_rmse": 26.671744666215773} {"canonical_length": 33, "canonical_nll": 0.7175994641853102, "generation_rouge_l_f1": 0.16546762589928057, "generation_token_f1": 0.2014388489208633, "prompt_id": "decae069198bb200", "reading_canonical_log_prob": -23.680782318115234, "reading_distances": [16, 19, 18, 15], "reading_log_probs": [-22.633872985839844, -24.65536880493164, -22.88011360168457, -22.118995666503906], "reading_logprob_std": 1.1021266823249416, "tokenization_path_logprob_std": 68.107369671343, "trajectory_balance_rmse": 483.31132922083214} {"canonical_length": 296, "canonical_nll": 1.506135064202386, "generation_rouge_l_f1": 0.20614828209764918, "generation_token_f1": 0.4954792043399638, "prompt_id": "78378b5209c8e413", "reading_canonical_log_prob": -445.81597900390625, "reading_distances": [15, 20, 20, 20], "reading_log_probs": [-453.4945068359375, -451.93817138671875, -455.6890869140625, -456.0087890625], "reading_logprob_std": 1.9214031739845077, "tokenization_path_logprob_std": 295.6867393095692, "trajectory_balance_rmse": 5944.754121910845} {"canonical_length": 46, "canonical_nll": 0.945492288340693, "generation_rouge_l_f1": 0.19178082191780824, "generation_token_f1": 0.29223744292237447, "prompt_id": "6c87b1ea5d43b53e", "reading_canonical_log_prob": -43.492645263671875, "reading_distances": [22, 18, 15, 22], "reading_log_probs": [-49.58728790283203, -48.269866943359375, -46.78033447265625, -47.20465850830078], "reading_logprob_std": 1.25249535108726, "tokenization_path_logprob_std": 59.83210802268152, "trajectory_balance_rmse": 595.152863348236} {"canonical_length": 120, "canonical_nll": 2.1858037312825522, "generation_rouge_l_f1": 0.15625000000000003, "generation_token_f1": 0.2760416666666667, "prompt_id": "c823f34b5de71522", "reading_canonical_log_prob": -262.29644775390625, "reading_distances": [21, 26, 20, 22], "reading_log_probs": [-258.04443359375, -259.1336669921875, -261.27752685546875, -261.5220947265625], "reading_logprob_std": 1.685575667964532, "tokenization_path_logprob_std": 118.46398473614698, "trajectory_balance_rmse": 2776.953261850539} {"canonical_length": 23, "canonical_nll": 1.1952715334684954, "generation_rouge_l_f1": 0.15384615384615383, "generation_token_f1": 0.23076923076923075, "prompt_id": "95ee759ef85d853e", "reading_canonical_log_prob": -27.49124526977539, "reading_distances": [29, 27, 29, 26], "reading_log_probs": [-29.060749053955078, -29.964523315429688, -30.240962982177734, -31.703062057495117], "reading_logprob_std": 1.0965172286915847, "tokenization_path_logprob_std": 45.03446405544156, "trajectory_balance_rmse": 226.26337364970686} {"canonical_length": 11, "canonical_nll": 0.752770337191495, "generation_rouge_l_f1": 0.7058823529411765, "generation_token_f1": 0.7058823529411765, "prompt_id": "036acca12e6542a8", "reading_canonical_log_prob": -8.280473709106445, "reading_distances": [18, 20, 25, 21], "reading_log_probs": [-8.756172180175781, -8.552827835083008, -9.01192569732666, -9.193500518798828], "reading_logprob_std": 0.28169377737595674, "tokenization_path_logprob_std": 37.43312699843526, "trajectory_balance_rmse": 186.86105121555875} {"canonical_length": 13, "canonical_nll": 1.225680791414701, "generation_rouge_l_f1": 0.4615384615384615, "generation_token_f1": 0.4615384615384615, "prompt_id": "5e5a1ac1eccda1ac", "reading_canonical_log_prob": -15.933850288391113, "reading_distances": [20, 36, 27, 27], "reading_log_probs": [-16.56709861755371, -17.693471908569336, -19.61092758178711, -17.05092430114746], "reading_logprob_std": 1.3357525759463855, "tokenization_path_logprob_std": 106.78686011842738, "trajectory_balance_rmse": 366.9988078275745} {"canonical_length": 8, "canonical_nll": 0.11748841404914856, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "05b4960bf40b9d90", "reading_canonical_log_prob": -0.9399073123931885, "reading_distances": [24, 30, 24, 25], "reading_log_probs": [-2.447437047958374, -1.1648701429367065, -2.2321739196777344, -1.4836642742156982], "reading_logprob_std": 0.6069876288549166, "tokenization_path_logprob_std": 35.31979262481461, "trajectory_balance_rmse": 140.3246404964965} {"canonical_length": 199, "canonical_nll": 1.2776384976641018, "generation_rouge_l_f1": 0.31947483588621445, "generation_token_f1": 0.5689277899343546, "prompt_id": "55f01821168c15c2", "reading_canonical_log_prob": -254.25006103515625, "reading_distances": [29, 33, 37, 34], "reading_log_probs": [-255.8157958984375, -255.5681915283203, -255.8306427001953, -258.1424560546875], "reading_logprob_std": 1.208134763786871, "tokenization_path_logprob_std": 72.69302472397857, "trajectory_balance_rmse": 3780.5686860466744} {"canonical_length": 181, "canonical_nll": 0.9364586087221599, "generation_rouge_l_f1": 0.31890660592255127, "generation_token_f1": 0.469248291571754, "prompt_id": "28aed3c6f636dfb4", "reading_canonical_log_prob": -169.49900817871094, "reading_distances": [13, 12, 11, 15], "reading_log_probs": [-170.25790405273438, -167.82896423339844, -170.38621520996094, -168.51734924316406], "reading_logprob_std": 1.2731787621538186, "tokenization_path_logprob_std": 146.58316592176118, "trajectory_balance_rmse": 3271.5961736347244} {"canonical_length": 102, "canonical_nll": 2.0386003980449603, "generation_rouge_l_f1": 0.1890243902439024, "generation_token_f1": 0.426829268292683, "prompt_id": "33e9d73a4126f71a", "reading_canonical_log_prob": -207.93724060058594, "reading_distances": [18, 20, 12, 20], "reading_log_probs": [-205.3733673095703, -208.05445861816406, -207.9438934326172, -208.33541870117188], "reading_logprob_std": 1.3788297835842445, "tokenization_path_logprob_std": 187.46724296606754, "trajectory_balance_rmse": 2418.098209139256} {"canonical_length": 169, "canonical_nll": 1.9017584986940643, "generation_rouge_l_f1": 0.23908045977011494, "generation_token_f1": 0.45977011494252873, "prompt_id": "ba9c16c03160cc50", "reading_canonical_log_prob": -321.3971862792969, "reading_distances": [23, 21, 23, 16], "reading_log_probs": [-327.2271423339844, -326.92169189453125, -324.7744140625, -322.59716796875], "reading_logprob_std": 2.152489806262452, "tokenization_path_logprob_std": 264.1484916486938, "trajectory_balance_rmse": 3169.5354866536863} {"canonical_length": 278, "canonical_nll": 0.7183388360112691, "generation_rouge_l_f1": 0.25447316103379725, "generation_token_f1": 0.3618290258449304, "prompt_id": "d04a82b0b7e2a434", "reading_canonical_log_prob": -199.6981964111328, "reading_distances": [23, 27, 24, 28], "reading_log_probs": [-187.03488159179688, -191.01564025878906, -193.22171020507812, -190.49441528320312], "reading_logprob_std": 2.560407176129471, "tokenization_path_logprob_std": 103.55258071511382, "trajectory_balance_rmse": 3258.8658063627668} {"canonical_length": 38, "canonical_nll": 0.6808747743305407, "generation_rouge_l_f1": 0.3409090909090909, "generation_token_f1": 0.38636363636363635, "prompt_id": "e1a492712f46f15d", "reading_canonical_log_prob": -25.873241424560547, "reading_distances": [19, 21, 19, 17], "reading_log_probs": [-27.049175262451172, -30.38935089111328, -27.34511947631836, -29.46578598022461], "reading_logprob_std": 1.6253693693415836, "tokenization_path_logprob_std": 67.18427235303281, "trajectory_balance_rmse": 554.1182321453048} {"canonical_length": 25, "canonical_nll": 0.5811145782470704, "generation_rouge_l_f1": 0.14046822742474915, "generation_token_f1": 0.14046822742474915, "prompt_id": "fcc6cd8702ae37ce", "reading_canonical_log_prob": -14.527864456176758, "reading_distances": [23, 26, 26, 22], "reading_log_probs": [-14.715049743652344, -17.653926849365234, -16.133237838745117, -17.203289031982422], "reading_logprob_std": 1.3070376791674285, "tokenization_path_logprob_std": 22.86668554804116, "trajectory_balance_rmse": 285.4447794271348} {"canonical_length": 342, "canonical_nll": 1.2200527971948099, "generation_rouge_l_f1": 0.24999999999999994, "generation_token_f1": 0.46381578947368424, "prompt_id": "38014139c1d9865d", "reading_canonical_log_prob": -417.258056640625, "reading_distances": [32, 30, 25, 38], "reading_log_probs": [-430.8861083984375, -431.77484130859375, -434.0836181640625, -437.4408264160156], "reading_logprob_std": 2.9252197670956015, "tokenization_path_logprob_std": 156.58493781583124, "trajectory_balance_rmse": 5714.899093487363} {"canonical_length": 6, "canonical_nll": 0.5518355369567871, "generation_rouge_l_f1": 0.8000000000000002, "generation_token_f1": 0.8000000000000002, "prompt_id": "bd70ea68bde60a24", "reading_canonical_log_prob": -3.3110132217407227, "reading_distances": [25, 25, 19, 25], "reading_log_probs": [-2.716811418533325, -3.321553945541382, -2.431436777114868, -3.027616262435913], "reading_logprob_std": 0.38491230329396964, "tokenization_path_logprob_std": 15.9407547310517, "trajectory_balance_rmse": 106.73781328984641} {"canonical_length": 11, "canonical_nll": 2.3421573638916016, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "57765c15fcf7839d", "reading_canonical_log_prob": -25.763731002807617, "reading_distances": [10, 12, 12, 14], "reading_log_probs": [-24.38497543334961, -25.47216796875, -24.682708740234375, -24.99078369140625], "reading_logprob_std": 0.46435647657899387, "tokenization_path_logprob_std": 40.39360181994491, "trajectory_balance_rmse": 200.74718532529275} {"canonical_length": 50, "canonical_nll": 1.4618154907226562, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.29545454545454547, "prompt_id": "253987951dace9e0", "reading_canonical_log_prob": -73.09077453613281, "reading_distances": [32, 23, 29, 26], "reading_log_probs": [-72.21365356445312, -69.9797134399414, -71.19098663330078, -71.0291748046875], "reading_logprob_std": 0.9144242741217387, "tokenization_path_logprob_std": 135.25178804575125, "trajectory_balance_rmse": 1127.8391897811427} {"canonical_length": 38, "canonical_nll": 1.2579662925318669, "generation_rouge_l_f1": 0.30656934306569344, "generation_token_f1": 0.3503649635036496, "prompt_id": "e17273b705c2c377", "reading_canonical_log_prob": -47.80271911621094, "reading_distances": [29, 22, 28, 29], "reading_log_probs": [-56.80362319946289, -51.66608428955078, -51.92521667480469, -53.94892120361328], "reading_logprob_std": 2.375506152118087, "tokenization_path_logprob_std": 48.693883969690354, "trajectory_balance_rmse": 737.7124351686963} {"canonical_length": 344, "canonical_nll": 1.5900037898573764, "generation_rouge_l_f1": 0.23240589198036007, "generation_token_f1": 0.46808510638297873, "prompt_id": "3e31d044adc4af42", "reading_canonical_log_prob": -546.9613037109375, "reading_distances": [23, 27, 31, 26], "reading_log_probs": [-542.5718994140625, -545.3155517578125, -545.4965209960938, -546.1376953125], "reading_logprob_std": 1.5789161462983305, "tokenization_path_logprob_std": 182.32294445963535, "trajectory_balance_rmse": 8062.456273713472} {"canonical_length": 23, "canonical_nll": 0.8669013977050781, "generation_rouge_l_f1": 0.41379310344827586, "generation_token_f1": 0.4482758620689656, "prompt_id": "31f5a769dd1b152b", "reading_canonical_log_prob": -19.938732147216797, "reading_distances": [24, 19, 24, 19], "reading_log_probs": [-21.260570526123047, -21.46466064453125, -21.091442108154297, -21.136289596557617], "reading_logprob_std": 0.1670413785834907, "tokenization_path_logprob_std": 31.66998600754786, "trajectory_balance_rmse": 495.03028563704333} {"canonical_length": 22, "canonical_nll": 1.12218562039462, "generation_rouge_l_f1": 0.7333333333333333, "generation_token_f1": 0.7333333333333333, "prompt_id": "74fb912570f58841", "reading_canonical_log_prob": -24.68808364868164, "reading_distances": [26, 24, 17, 29], "reading_log_probs": [-26.032453536987305, -23.573192596435547, -24.28256607055664, -25.512304306030273], "reading_logprob_std": 1.1238418786028743, "tokenization_path_logprob_std": 45.59534642102105, "trajectory_balance_rmse": 221.18042950467503} {"canonical_length": 10, "canonical_nll": 0.4536423683166504, "generation_rouge_l_f1": 0.17391304347826086, "generation_token_f1": 0.17391304347826086, "prompt_id": "434a6990c55f121a", "reading_canonical_log_prob": -4.536423683166504, "reading_distances": [14, 20, 19, 15], "reading_log_probs": [-6.1137166023254395, -6.093165397644043, -7.930542945861816, -4.673203945159912], "reading_logprob_std": 1.3347553201731077, "tokenization_path_logprob_std": 36.106860696973115, "trajectory_balance_rmse": 228.70209170896197} {"canonical_length": 222, "canonical_nll": 1.1682493192655545, "generation_rouge_l_f1": 0.225130890052356, "generation_token_f1": 0.4450261780104712, "prompt_id": "2e39c9a4bb908fa5", "reading_canonical_log_prob": -259.3513488769531, "reading_distances": [16, 23, 18, 23], "reading_log_probs": [-260.7913513183594, -262.21728515625, -259.97308349609375, -260.2344665527344], "reading_logprob_std": 1.0020521439100423, "tokenization_path_logprob_std": 104.00969393295522, "trajectory_balance_rmse": 5483.1025147710225} {"canonical_length": 226, "canonical_nll": 1.5365533069171737, "generation_rouge_l_f1": 0.1714285714285714, "generation_token_f1": 0.29890109890109895, "prompt_id": "1a576b185bace9d2", "reading_canonical_log_prob": -347.26104736328125, "reading_distances": [10, 16, 12, 16], "reading_log_probs": [-349.5121765136719, -347.06170654296875, -345.53472900390625, -357.88031005859375], "reading_logprob_std": 5.504829431257455, "tokenization_path_logprob_std": 185.89483272982253, "trajectory_balance_rmse": 4099.275086487588} {"canonical_length": 270, "canonical_nll": 1.1776294849537037, "generation_rouge_l_f1": 0.2681564245810056, "generation_token_f1": 0.5400372439478586, "prompt_id": "c2df8f1e4778ecd3", "reading_canonical_log_prob": -317.9599609375, "reading_distances": [35, 28, 35, 35], "reading_log_probs": [-319.71759033203125, -321.6524658203125, -320.58245849609375, -322.0745544433594], "reading_logprob_std": 1.0644421353414466, "tokenization_path_logprob_std": 150.74038751434804, "trajectory_balance_rmse": 5685.180332086187} {"canonical_length": 2, "canonical_nll": 1.3475195169448853, "generation_rouge_l_f1": 0.022988505747126436, "generation_token_f1": 0.022988505747126436, "prompt_id": "42c831675952234b", "reading_canonical_log_prob": -2.6950390338897705, "reading_distances": [68, 67, 67, 49], "reading_log_probs": [-0.7835102677345276, -2.2077293395996094, -1.4160490036010742, -3.2224955558776855], "reading_logprob_std": 1.052651882436005, "tokenization_path_logprob_std": 7.321127908001274, "trajectory_balance_rmse": 24.547909330179653} {"canonical_length": 16, "canonical_nll": 2.6162567138671875, "generation_rouge_l_f1": 0.21621621621621623, "generation_token_f1": 0.27027027027027023, "prompt_id": "a9f4c15e6cb406fd", "reading_canonical_log_prob": -41.860107421875, "reading_distances": [21, 18, 20, 19], "reading_log_probs": [-45.03369140625, -51.295555114746094, -54.982791900634766, -55.88299560546875], "reading_logprob_std": 4.927420145130567, "tokenization_path_logprob_std": 50.476305133618126, "trajectory_balance_rmse": 430.8240966832067} {"canonical_length": 161, "canonical_nll": 1.7674983243764557, "generation_rouge_l_f1": 0.165, "generation_token_f1": 0.33999999999999997, "prompt_id": "25e2da0cacffb5d9", "reading_canonical_log_prob": -284.5672302246094, "reading_distances": [23, 21, 19, 23], "reading_log_probs": [-282.2342224121094, -284.07159423828125, -285.7535705566406, -284.8739318847656], "reading_logprob_std": 1.4993433817731325, "tokenization_path_logprob_std": 195.85199209564036, "trajectory_balance_rmse": 3337.515791067373} {"canonical_length": 116, "canonical_nll": 1.1804333390860722, "generation_rouge_l_f1": 0.19117647058823525, "generation_token_f1": 0.29411764705882354, "prompt_id": "d71e3178e4fc08e6", "reading_canonical_log_prob": -136.93026733398438, "reading_distances": [36, 35, 42, 35], "reading_log_probs": [-141.82850646972656, -140.83641052246094, -142.2989501953125, -143.77178955078125], "reading_logprob_std": 1.2215648741887908, "tokenization_path_logprob_std": 48.36860856240383, "trajectory_balance_rmse": 3050.7894906810347} {"canonical_length": 9, "canonical_nll": 1.0147751702202692, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.2105263157894737, "prompt_id": "c69244953386b7ea", "reading_canonical_log_prob": -9.132976531982422, "reading_distances": [16, 19, 15, 5], "reading_log_probs": [-10.384641647338867, -9.780542373657227, -10.41486644744873, -8.031624794006348], "reading_logprob_std": 1.1196524123129077, "tokenization_path_logprob_std": 21.554142599635796, "trajectory_balance_rmse": 212.0970907370001} {"canonical_length": 16, "canonical_nll": 1.9754126071929932, "generation_rouge_l_f1": 0.23728813559322035, "generation_token_f1": 0.2711864406779661, "prompt_id": "44b99fb81cd83d68", "reading_canonical_log_prob": -31.60660171508789, "reading_distances": [25, 21, 24, 30], "reading_log_probs": [-31.442838668823242, -30.381996154785156, -32.71379852294922, -29.52735137939453], "reading_logprob_std": 1.3763170274716072, "tokenization_path_logprob_std": 76.91240700107433, "trajectory_balance_rmse": 396.9519197826757} {"canonical_length": 23, "canonical_nll": 0.27188929267551587, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "f3f998d8d70ee428", "reading_canonical_log_prob": -6.253453731536865, "reading_distances": [51, 41, 40, 42], "reading_log_probs": [-8.737499237060547, -10.437334060668945, -8.469767570495605, -9.296992301940918], "reading_logprob_std": 0.8722647313530739, "tokenization_path_logprob_std": 20.977186730495195, "trajectory_balance_rmse": 298.6243653038239} {"canonical_length": 45, "canonical_nll": 0.817822265625, "generation_rouge_l_f1": 0.2463768115942029, "generation_token_f1": 0.34782608695652173, "prompt_id": "8565123aabdca8e0", "reading_canonical_log_prob": -36.802001953125, "reading_distances": [21, 22, 24, 22], "reading_log_probs": [-45.79608154296875, -42.84437942504883, -44.22321701049805, -43.58203125], "reading_logprob_std": 1.256480183276512, "tokenization_path_logprob_std": 39.68382901274896, "trajectory_balance_rmse": 967.6732217136275} {"canonical_length": 293, "canonical_nll": 1.6397434273677474, "generation_rouge_l_f1": 0.21532846715328469, "generation_token_f1": 0.4416058394160584, "prompt_id": "c754674f366c507b", "reading_canonical_log_prob": -480.44482421875, "reading_distances": [27, 30, 36, 32], "reading_log_probs": [-476.1412048339844, -481.3006591796875, -478.39190673828125, -481.1693420410156], "reading_logprob_std": 2.4691433028026717, "tokenization_path_logprob_std": 116.92867290795527, "trajectory_balance_rmse": 6515.567303399202} {"canonical_length": 15, "canonical_nll": 1.2183638254801432, "generation_rouge_l_f1": 0.10596026490066225, "generation_token_f1": 0.10596026490066225, "prompt_id": "f7c03b54285f64b4", "reading_canonical_log_prob": -18.27545738220215, "reading_distances": [26, 26, 29, 25], "reading_log_probs": [-16.43271255493164, -19.660913467407227, -15.056312561035156, -16.05518341064453], "reading_logprob_std": 1.992897138437246, "tokenization_path_logprob_std": 82.5968371967089, "trajectory_balance_rmse": 270.1925179558276} {"canonical_length": 33, "canonical_nll": 0.36456657178474194, "generation_rouge_l_f1": 0.1282051282051282, "generation_token_f1": 0.15384615384615385, "prompt_id": "278cabb1c777cfb5", "reading_canonical_log_prob": -12.030696868896484, "reading_distances": [58, 44, 50, 56], "reading_log_probs": [-20.82574462890625, -28.653667449951172, -22.63113021850586, -24.304866790771484], "reading_logprob_std": 3.3494340296448817, "tokenization_path_logprob_std": 78.00203052693165, "trajectory_balance_rmse": 438.92489135841384} {"canonical_length": 82, "canonical_nll": 0.494830945642983, "generation_rouge_l_f1": 0.4088888888888889, "generation_token_f1": 0.4533333333333333, "prompt_id": "26496139eab8348f", "reading_canonical_log_prob": -40.57613754272461, "reading_distances": [17, 20, 18, 17], "reading_log_probs": [-44.98204803466797, -45.753318786621094, -45.05589294433594, -44.83495330810547], "reading_logprob_std": 0.4083055210826355, "tokenization_path_logprob_std": 44.24902957658725, "trajectory_balance_rmse": 801.1782963620309} {"canonical_length": 130, "canonical_nll": 1.9943026029146635, "generation_rouge_l_f1": 0.16666666666666669, "generation_token_f1": 0.3585858585858586, "prompt_id": "5e25b104005e6f72", "reading_canonical_log_prob": -259.25933837890625, "reading_distances": [25, 23, 23, 19], "reading_log_probs": [-268.2918701171875, -262.57342529296875, -261.0420227050781, -250.60891723632812], "reading_logprob_std": 7.372765525561365, "tokenization_path_logprob_std": 229.4139162245407, "trajectory_balance_rmse": 1725.1017147150278} {"canonical_length": 107, "canonical_nll": 2.410878689489632, "generation_rouge_l_f1": 0.24669603524229075, "generation_token_f1": 0.4581497797356828, "prompt_id": "14f8273006e150ca", "reading_canonical_log_prob": -257.9640197753906, "reading_distances": [25, 25, 34, 30], "reading_log_probs": [-262.88739013671875, -258.6769104003906, -262.11309814453125, -260.0794372558594], "reading_logprob_std": 1.9175181764399019, "tokenization_path_logprob_std": 118.99371959655102, "trajectory_balance_rmse": 2041.8238048699177} {"canonical_length": 185, "canonical_nll": 1.4674786541913007, "generation_rouge_l_f1": 0.3112391930835735, "generation_token_f1": 0.478386167146974, "prompt_id": "c0e6a5cad32d62b6", "reading_canonical_log_prob": -271.4835510253906, "reading_distances": [13, 17, 17, 12], "reading_log_probs": [-279.75531005859375, -279.0694274902344, -275.00360107421875, -277.07574462890625], "reading_logprob_std": 2.141431350944632, "tokenization_path_logprob_std": 243.75389080955674, "trajectory_balance_rmse": 3117.825572816133} {"canonical_length": 16, "canonical_nll": 1.0834076404571533, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4571428571428572, "prompt_id": "937bfa51bfbd0542", "reading_canonical_log_prob": -17.334522247314453, "reading_distances": [31, 34, 37, 39], "reading_log_probs": [-22.574331283569336, -22.833839416503906, -21.635696411132812, -18.855396270751953], "reading_logprob_std": 1.8205421106608681, "tokenization_path_logprob_std": 18.302943668380138, "trajectory_balance_rmse": 429.3900528285655} {"canonical_length": 75, "canonical_nll": 0.4320518493652344, "generation_rouge_l_f1": 0.28664495114006516, "generation_token_f1": 0.3127035830618892, "prompt_id": "19b52394b9bc5537", "reading_canonical_log_prob": -32.40388870239258, "reading_distances": [13, 13, 17, 12], "reading_log_probs": [-34.341590881347656, -34.520263671875, -34.51499938964844, -32.93827438354492], "reading_logprob_std": 0.7648568077186007, "tokenization_path_logprob_std": 58.642113332163746, "trajectory_balance_rmse": 538.2035967015161} {"canonical_length": 339, "canonical_nll": 1.4618659005404222, "generation_rouge_l_f1": 0.22463768115942032, "generation_token_f1": 0.36231884057971014, "prompt_id": "c03437022c1ce69c", "reading_canonical_log_prob": -495.5725402832031, "reading_distances": [29, 29, 24, 28], "reading_log_probs": [-504.65228271484375, -504.0814208984375, -502.28564453125, -507.03680419921875], "reading_logprob_std": 1.9609823988020914, "tokenization_path_logprob_std": 213.72586811957459, "trajectory_balance_rmse": 6036.010432100742} {"canonical_length": 22, "canonical_nll": 0.7633348811756481, "generation_rouge_l_f1": 0.0842911877394636, "generation_token_f1": 0.10727969348659006, "prompt_id": "01ef5d8df6df3761", "reading_canonical_log_prob": -16.793367385864258, "reading_distances": [21, 20, 21, 26], "reading_log_probs": [-21.09416389465332, -21.3870849609375, -20.009525299072266, -24.920150756835938], "reading_logprob_std": 2.1290625094603155, "tokenization_path_logprob_std": 49.10548299112536, "trajectory_balance_rmse": 359.4974837433676} {"canonical_length": 110, "canonical_nll": 1.6866174871271307, "generation_rouge_l_f1": 0.23218997361477572, "generation_token_f1": 0.33245382585751976, "prompt_id": "1275798cfa3be44f", "reading_canonical_log_prob": -185.52792358398438, "reading_distances": [36, 34, 37, 38], "reading_log_probs": [-190.3336181640625, -193.10720825195312, -190.46395874023438, -195.1759033203125], "reading_logprob_std": 2.3206716909956704, "tokenization_path_logprob_std": 121.32189684967484, "trajectory_balance_rmse": 2678.2944789039666} {"canonical_length": 4, "canonical_nll": 0.30223894119262695, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1f2fdcd7eec76cc5", "reading_canonical_log_prob": -1.2089557647705078, "reading_distances": [29, 29, 25, 33], "reading_log_probs": [-2.6569883823394775, -1.3799457550048828, -0.7706301212310791, -1.2795040607452393], "reading_logprob_std": 0.8024418555957323, "tokenization_path_logprob_std": 29.811001605970493, "trajectory_balance_rmse": 52.012149516197596} {"canonical_length": 128, "canonical_nll": 1.3304717540740967, "generation_rouge_l_f1": 0.11285266457680251, "generation_token_f1": 0.13793103448275862, "prompt_id": "b8b97c6b0ef87a98", "reading_canonical_log_prob": -170.30038452148438, "reading_distances": [32, 28, 24, 26], "reading_log_probs": [-164.25997924804688, -166.36660766601562, -166.74581909179688, -170.6531219482422], "reading_logprob_std": 2.665748459851226, "tokenization_path_logprob_std": 29.83867002328512, "trajectory_balance_rmse": 1602.8842795567468} {"canonical_length": 168, "canonical_nll": 1.1785697937011719, "generation_rouge_l_f1": 0.22686567164179103, "generation_token_f1": 0.47164179104477616, "prompt_id": "7ec56cfda55797c0", "reading_canonical_log_prob": -197.99972534179688, "reading_distances": [27, 35, 24, 27], "reading_log_probs": [-209.979736328125, -220.7691192626953, -213.97386169433594, -207.11093139648438], "reading_logprob_std": 5.919003975669093, "tokenization_path_logprob_std": 76.8120497491383, "trajectory_balance_rmse": 3323.5961826520224} {"canonical_length": 32, "canonical_nll": 1.4721683263778687, "generation_rouge_l_f1": 0.5789473684210527, "generation_token_f1": 0.6315789473684211, "prompt_id": "c6a2889afcb5222e", "reading_canonical_log_prob": -47.1093864440918, "reading_distances": [15, 16, 13, 9], "reading_log_probs": [-46.629676818847656, -46.56907653808594, -49.274410247802734, -48.3275032043457], "reading_logprob_std": 1.3287973732554268, "tokenization_path_logprob_std": 41.59882303225125, "trajectory_balance_rmse": 474.2134943012681} {"canonical_length": 32, "canonical_nll": 1.512040138244629, "generation_rouge_l_f1": 0.35714285714285715, "generation_token_f1": 0.3928571428571428, "prompt_id": "5371fa87b29b339c", "reading_canonical_log_prob": -48.385284423828125, "reading_distances": [12, 17, 19, 16], "reading_log_probs": [-47.124671936035156, -44.95265197753906, -47.2088623046875, -46.731204986572266], "reading_logprob_std": 1.0552040607713509, "tokenization_path_logprob_std": 100.75479419510545, "trajectory_balance_rmse": 695.6127195504781} {"canonical_length": 63, "canonical_nll": 1.2324752807617188, "generation_rouge_l_f1": 0.3270440251572327, "generation_token_f1": 0.4779874213836478, "prompt_id": "c89f55288eb64d0c", "reading_canonical_log_prob": -77.64594268798828, "reading_distances": [22, 29, 25, 31], "reading_log_probs": [-80.14511108398438, -78.89424896240234, -79.91160583496094, -82.55094909667969], "reading_logprob_std": 1.5486510534226834, "tokenization_path_logprob_std": 66.5556769733046, "trajectory_balance_rmse": 1546.5235235431978} {"canonical_length": 221, "canonical_nll": 1.490878696355345, "generation_rouge_l_f1": 0.21118012422360252, "generation_token_f1": 0.3312629399585922, "prompt_id": "0e1e524655779cac", "reading_canonical_log_prob": -329.48419189453125, "reading_distances": [18, 24, 18, 17], "reading_log_probs": [-333.6396789550781, -331.9216613769531, -328.8470153808594, -328.8038330078125], "reading_logprob_std": 2.3889115551182596, "tokenization_path_logprob_std": 135.82672925355064, "trajectory_balance_rmse": 4598.898596437147} {"canonical_length": 9, "canonical_nll": 0.4927203920152452, "generation_rouge_l_f1": 0.8235294117647058, "generation_token_f1": 0.8235294117647058, "prompt_id": "5f85bd9588d13639", "reading_canonical_log_prob": -4.434483528137207, "reading_distances": [21, 13, 20, 14], "reading_log_probs": [-5.789300918579102, -4.959333419799805, -4.898360729217529, -3.674121379852295], "reading_logprob_std": 0.8713425748165508, "tokenization_path_logprob_std": 45.036118055547135, "trajectory_balance_rmse": 205.90148114682367} {"canonical_length": 26, "canonical_nll": 0.8778312389667218, "generation_rouge_l_f1": 0.1299638989169675, "generation_token_f1": 0.15884476534296027, "prompt_id": "570dbf2f8a32567a", "reading_canonical_log_prob": -22.823612213134766, "reading_distances": [20, 14, 21, 23], "reading_log_probs": [-23.171558380126953, -21.39267349243164, -21.726123809814453, -23.272350311279297], "reading_logprob_std": 0.9703548322400256, "tokenization_path_logprob_std": 114.41691365921052, "trajectory_balance_rmse": 498.65394849515747} {"canonical_length": 26, "canonical_nll": 1.037784796494704, "generation_rouge_l_f1": 0.14046822742474915, "generation_token_f1": 0.15384615384615385, "prompt_id": "b4dad9a8bbd784f6", "reading_canonical_log_prob": -26.982404708862305, "reading_distances": [40, 29, 38, 37], "reading_log_probs": [-29.2044677734375, -32.65895462036133, -30.215106964111328, -28.31130027770996], "reading_logprob_std": 1.8764236522802327, "tokenization_path_logprob_std": 39.63234267247943, "trajectory_balance_rmse": 599.6202952301321} {"canonical_length": 10, "canonical_nll": 2.2278654098510744, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.39999999999999997, "prompt_id": "9e76d45f4e1b2541", "reading_canonical_log_prob": -22.278654098510742, "reading_distances": [28, 41, 31, 40], "reading_log_probs": [-25.577659606933594, -26.279645919799805, -24.411712646484375, -29.04827308654785], "reading_logprob_std": 1.9695502625237462, "tokenization_path_logprob_std": 44.71142568906065, "trajectory_balance_rmse": 316.5312776571182} {"canonical_length": 256, "canonical_nll": 1.3891205787658691, "generation_rouge_l_f1": 0.2149712092130518, "generation_token_f1": 0.44145873320537427, "prompt_id": "6a9973c4375e58ef", "reading_canonical_log_prob": -355.6148681640625, "reading_distances": [32, 35, 29, 33], "reading_log_probs": [-357.7562561035156, -360.03045654296875, -358.82244873046875, -357.7474365234375], "reading_logprob_std": 1.0853550368227614, "tokenization_path_logprob_std": 338.7979226051527, "trajectory_balance_rmse": 5990.153497214603} {"canonical_length": 277, "canonical_nll": 1.136051549808213, "generation_rouge_l_f1": 0.2169117647058824, "generation_token_f1": 0.4448529411764706, "prompt_id": "4a7ad2175af88b68", "reading_canonical_log_prob": -314.686279296875, "reading_distances": [30, 31, 28, 36], "reading_log_probs": [-320.26885986328125, -318.6861877441406, -316.8348388671875, -324.2000732421875], "reading_logprob_std": 3.1335402398519654, "tokenization_path_logprob_std": 235.05590496715126, "trajectory_balance_rmse": 5381.306092653906} {"canonical_length": 24, "canonical_nll": 0.34895992279052734, "generation_rouge_l_f1": 0.37037037037037035, "generation_token_f1": 0.37037037037037035, "prompt_id": "7271cafab8ca2bf4", "reading_canonical_log_prob": -8.375038146972656, "reading_distances": [28, 34, 34, 26], "reading_log_probs": [-9.141197204589844, -11.706948280334473, -17.505094528198242, -13.343385696411133], "reading_logprob_std": 3.5096592055861935, "tokenization_path_logprob_std": 85.06863701988586, "trajectory_balance_rmse": 506.4814929098489} {"canonical_length": 11, "canonical_nll": 0.5947467197071422, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "7798c57a62239065", "reading_canonical_log_prob": -6.5422139167785645, "reading_distances": [14, 14, 17, 15], "reading_log_probs": [-9.172545433044434, -8.258142471313477, -8.61686897277832, -10.000656127929688], "reading_logprob_std": 0.7588687779083131, "tokenization_path_logprob_std": 56.090315863813586, "trajectory_balance_rmse": 233.85448746885913} {"canonical_length": 97, "canonical_nll": 1.027299468050298, "generation_rouge_l_f1": 0.2904109589041096, "generation_token_f1": 0.4054794520547945, "prompt_id": "fd06dfbeb5280c45", "reading_canonical_log_prob": -99.6480484008789, "reading_distances": [39, 44, 39, 45], "reading_log_probs": [-100.14124298095703, -106.44834899902344, -98.27680969238281, -101.19160461425781], "reading_logprob_std": 3.503122452121695, "tokenization_path_logprob_std": 236.77931078081943, "trajectory_balance_rmse": 2008.5610062755675} {"canonical_length": 210, "canonical_nll": 0.9984141031901042, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.5024875621890547, "prompt_id": "3efd48165ac7ab7c", "reading_canonical_log_prob": -209.66696166992188, "reading_distances": [16, 19, 17, 19], "reading_log_probs": [-209.6393280029297, -212.21499633789062, -214.31398010253906, -211.91400146484375], "reading_logprob_std": 1.9130431081066228, "tokenization_path_logprob_std": 131.7204752823774, "trajectory_balance_rmse": 2139.8839390588105} {"canonical_length": 69, "canonical_nll": 1.479774364526721, "generation_rouge_l_f1": 0.26190476190476186, "generation_token_f1": 0.4642857142857143, "prompt_id": "18e11e5d4d3c48a7", "reading_canonical_log_prob": -102.10443115234375, "reading_distances": [32, 40, 35, 39], "reading_log_probs": [-106.18336486816406, -107.93204498291016, -102.38762664794922, -102.85296630859375], "reading_logprob_std": 2.66632114217342, "tokenization_path_logprob_std": 119.56553475699398, "trajectory_balance_rmse": 1767.6052559789257} {"canonical_length": 17, "canonical_nll": 0.3987610760857077, "generation_rouge_l_f1": 0.08450704225352113, "generation_token_f1": 0.09154929577464789, "prompt_id": "b05a564949c24a61", "reading_canonical_log_prob": -6.778938293457031, "reading_distances": [70, 64, 51, 67], "reading_log_probs": [-3.975156307220459, -6.821446895599365, -5.433920860290527, -5.988356590270996], "reading_logprob_std": 1.1975324090476553, "tokenization_path_logprob_std": 53.45421828677862, "trajectory_balance_rmse": 142.6457826126644} {"canonical_length": 129, "canonical_nll": 0.7668585962103319, "generation_rouge_l_f1": 0.4086021505376344, "generation_token_f1": 0.5232974910394265, "prompt_id": "977aa631a6e3fb74", "reading_canonical_log_prob": -98.92475891113281, "reading_distances": [9, 15, 14, 11], "reading_log_probs": [-101.84632873535156, -103.1265869140625, -102.79967498779297, -100.29816436767578], "reading_logprob_std": 1.2685012114547602, "tokenization_path_logprob_std": 80.12704747802925, "trajectory_balance_rmse": 1267.878236139785} {"canonical_length": 85, "canonical_nll": 1.2229063146254595, "generation_rouge_l_f1": 0.28444444444444444, "generation_token_f1": 0.4711111111111111, "prompt_id": "308d6723913890bf", "reading_canonical_log_prob": -103.94703674316406, "reading_distances": [24, 23, 23, 21], "reading_log_probs": [-106.0806884765625, -107.71539306640625, -105.48213195800781, -105.67994689941406], "reading_logprob_std": 1.014919941081878, "tokenization_path_logprob_std": 55.93266230363385, "trajectory_balance_rmse": 1730.2873004677335} {"canonical_length": 189, "canonical_nll": 1.159871257802166, "generation_rouge_l_f1": 0.27790432801822323, "generation_token_f1": 0.47835990888382685, "prompt_id": "ff140f7a02786106", "reading_canonical_log_prob": -219.21566772460938, "reading_distances": [15, 15, 20, 21], "reading_log_probs": [-224.50262451171875, -222.20449829101562, -224.8125, -225.10260009765625], "reading_logprob_std": 1.3235747904387904, "tokenization_path_logprob_std": 228.66783995426982, "trajectory_balance_rmse": 3507.3343408924075} {"canonical_length": 27, "canonical_nll": 1.8854350337275751, "generation_rouge_l_f1": 0.17142857142857143, "generation_token_f1": 0.17142857142857143, "prompt_id": "536f5c5bf3146493", "reading_canonical_log_prob": -50.90674591064453, "reading_distances": [31, 42, 32, 33], "reading_log_probs": [-59.732452392578125, -60.327476501464844, -60.7141227722168, -53.60608673095703], "reading_logprob_std": 3.350383545132404, "tokenization_path_logprob_std": 53.23990376782177, "trajectory_balance_rmse": 808.1039818456921} {"canonical_length": 210, "canonical_nll": 1.5986900692894346, "generation_rouge_l_f1": 0.22633744855967078, "generation_token_f1": 0.4897119341563786, "prompt_id": "18e930a2b5a84e85", "reading_canonical_log_prob": -335.72491455078125, "reading_distances": [14, 19, 17, 14], "reading_log_probs": [-341.4771423339844, -344.8926086425781, -344.5156555175781, -344.2242126464844], "reading_logprob_std": 1.5577268049678372, "tokenization_path_logprob_std": 166.73188510432578, "trajectory_balance_rmse": 4328.8602297194075} {"canonical_length": 21, "canonical_nll": 1.5565685090564547, "generation_rouge_l_f1": 0.19178082191780824, "generation_token_f1": 0.273972602739726, "prompt_id": "2631f3bcf5083818", "reading_canonical_log_prob": -32.68793869018555, "reading_distances": [32, 21, 35, 28], "reading_log_probs": [-30.618881225585938, -29.04625129699707, -27.086524963378906, -30.468921661376953], "reading_logprob_std": 1.640079246493234, "tokenization_path_logprob_std": 62.45314382962191, "trajectory_balance_rmse": 297.9262082048582} {"canonical_length": 169, "canonical_nll": 0.8697605471639238, "generation_rouge_l_f1": 0.3535353535353536, "generation_token_f1": 0.5202020202020202, "prompt_id": "62ad42126bc04505", "reading_canonical_log_prob": -146.98953247070312, "reading_distances": [43, 45, 43, 38], "reading_log_probs": [-148.33535766601562, -148.28378295898438, -151.71499633789062, -147.7393035888672], "reading_logprob_std": 1.8178676573674086, "tokenization_path_logprob_std": 176.6248944373444, "trajectory_balance_rmse": 2859.9266363886945} {"canonical_length": 221, "canonical_nll": 1.3023277041059813, "generation_rouge_l_f1": 0.2809224318658281, "generation_token_f1": 0.5157232704402515, "prompt_id": "44a0d1ef8b9a47d0", "reading_canonical_log_prob": -287.8144226074219, "reading_distances": [15, 7, 13, 19], "reading_log_probs": [-297.5082702636719, -289.8904113769531, -291.29730224609375, -294.6443786621094], "reading_logprob_std": 3.4228672220753458, "tokenization_path_logprob_std": 191.5373372913855, "trajectory_balance_rmse": 5322.683650773419} {"canonical_length": 84, "canonical_nll": 1.1725309462774367, "generation_rouge_l_f1": 0.024390243902439022, "generation_token_f1": 0.024390243902439022, "prompt_id": "1bc849b3b651f6ed", "reading_canonical_log_prob": -98.49259948730469, "reading_distances": [23, 20, 21, 26], "reading_log_probs": [-108.37522888183594, -105.63133239746094, -104.4886703491211, -107.10520935058594], "reading_logprob_std": 1.6973391168432919, "tokenization_path_logprob_std": 184.4537623910664, "trajectory_balance_rmse": 1959.5234922342058} {"canonical_length": 225, "canonical_nll": 1.2557892523871528, "generation_rouge_l_f1": 0.1963927855711423, "generation_token_f1": 0.42885771543086165, "prompt_id": "a8febb0397bb60c3", "reading_canonical_log_prob": -282.5525817871094, "reading_distances": [16, 15, 18, 12], "reading_log_probs": [-285.2103271484375, -287.3564758300781, -286.19342041015625, -285.41357421875], "reading_logprob_std": 0.9725165191354989, "tokenization_path_logprob_std": 62.67492194921525, "trajectory_balance_rmse": 4421.871691365713} {"canonical_length": 7, "canonical_nll": 2.85873658316476, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "37af528e072ecf7b", "reading_canonical_log_prob": -20.01115608215332, "reading_distances": [9, 15, 22, 20], "reading_log_probs": [-19.961559295654297, -20.693849563598633, -19.579174041748047, -20.820335388183594], "reading_logprob_std": 0.5929408041526528, "tokenization_path_logprob_std": 58.80688673932306, "trajectory_balance_rmse": 144.52790100232815} {"canonical_length": 252, "canonical_nll": 1.5981951516772073, "generation_rouge_l_f1": 0.21285140562248997, "generation_token_f1": 0.42570281124497994, "prompt_id": "192d126d4e011ab4", "reading_canonical_log_prob": -402.74517822265625, "reading_distances": [21, 25, 16, 20], "reading_log_probs": [-412.55352783203125, -411.09881591796875, -405.50408935546875, -411.302001953125], "reading_logprob_std": 3.140264412665143, "tokenization_path_logprob_std": 231.25977416818628, "trajectory_balance_rmse": 6140.611869166205} {"canonical_length": 182, "canonical_nll": 1.3383456219683636, "generation_rouge_l_f1": 0.29032258064516125, "generation_token_f1": 0.52073732718894, "prompt_id": "55712d2518889f49", "reading_canonical_log_prob": -243.5789031982422, "reading_distances": [23, 25, 23, 19], "reading_log_probs": [-282.9347839355469, -278.9729919433594, -279.1080017089844, -277.49365234375], "reading_logprob_std": 2.323050583445627, "tokenization_path_logprob_std": 106.82253310062463, "trajectory_balance_rmse": 3195.4862835698664} {"canonical_length": 12, "canonical_nll": 1.0116907755533855, "generation_rouge_l_f1": 0.3225806451612903, "generation_token_f1": 0.3225806451612903, "prompt_id": "587eb61a8e257f2b", "reading_canonical_log_prob": -12.140289306640625, "reading_distances": [21, 25, 23, 29], "reading_log_probs": [-13.135499954223633, -14.05309009552002, -10.65376091003418, -12.117817878723145], "reading_logprob_std": 1.4571907170311795, "tokenization_path_logprob_std": 53.73126413432721, "trajectory_balance_rmse": 368.05754402698494} {"canonical_length": 109, "canonical_nll": 0.8415711429140983, "generation_rouge_l_f1": 0.13293051359516614, "generation_token_f1": 0.21752265861027192, "prompt_id": "fff5e81d18ab40f8", "reading_canonical_log_prob": -91.73125457763672, "reading_distances": [15, 15, 14, 16], "reading_log_probs": [-93.00189208984375, -95.04917907714844, -92.47992706298828, -95.09943389892578], "reading_logprob_std": 1.364090376447361, "tokenization_path_logprob_std": 143.9065969626025, "trajectory_balance_rmse": 1260.6799061780673} {"canonical_length": 15, "canonical_nll": 0.40414098103841145, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.9285714285714286, "prompt_id": "e02d75efb09e53b0", "reading_canonical_log_prob": -6.062114715576172, "reading_distances": [25, 28, 20, 18], "reading_log_probs": [-7.365386009216309, -10.72225570678711, -7.704286098480225, -7.577710151672363], "reading_logprob_std": 1.5927135585118064, "tokenization_path_logprob_std": 39.840976880442696, "trajectory_balance_rmse": 240.14351812735694} {"canonical_length": 27, "canonical_nll": 1.2876653318051938, "generation_rouge_l_f1": 0.10600706713780919, "generation_token_f1": 0.127208480565371, "prompt_id": "93316a155b82ba8e", "reading_canonical_log_prob": -34.766963958740234, "reading_distances": [27, 27, 28, 24], "reading_log_probs": [-42.23338317871094, -38.206016540527344, -37.449703216552734, -41.30898666381836], "reading_logprob_std": 2.3283092415572417, "tokenization_path_logprob_std": 128.68795483154892, "trajectory_balance_rmse": 571.129078356283} {"canonical_length": 110, "canonical_nll": 1.3480331420898437, "generation_rouge_l_f1": 0.38211382113821135, "generation_token_f1": 0.5203252032520325, "prompt_id": "37f1255c888facef", "reading_canonical_log_prob": -148.2836456298828, "reading_distances": [14, 19, 18, 12], "reading_log_probs": [-147.327880859375, -147.83050537109375, -147.57620239257812, -147.35369873046875], "reading_logprob_std": 0.23389527627743564, "tokenization_path_logprob_std": 181.75078626695844, "trajectory_balance_rmse": 2478.758821589076} {"canonical_length": 150, "canonical_nll": 0.9391184488932292, "generation_rouge_l_f1": 0.4021739130434783, "generation_token_f1": 0.5706521739130435, "prompt_id": "cf1ad5289e6ad557", "reading_canonical_log_prob": -140.86776733398438, "reading_distances": [19, 21, 21, 19], "reading_log_probs": [-140.75424194335938, -141.13211059570312, -142.54254150390625, -141.29571533203125], "reading_logprob_std": 0.7748468035011924, "tokenization_path_logprob_std": 109.85599007354116, "trajectory_balance_rmse": 2853.539151977751} {"canonical_length": 314, "canonical_nll": 0.9491549935310509, "generation_rouge_l_f1": 0.32051282051282054, "generation_token_f1": 0.4903846153846154, "prompt_id": "ff9bcfc5d56ada93", "reading_canonical_log_prob": -298.03466796875, "reading_distances": [41, 34, 36, 40], "reading_log_probs": [-302.83367919921875, -304.4349365234375, -306.34515380859375, -305.73675537109375], "reading_logprob_std": 1.5555315612820029, "tokenization_path_logprob_std": 156.597102103461, "trajectory_balance_rmse": 5225.022740785851} {"canonical_length": 12, "canonical_nll": 1.6536814371744792, "generation_rouge_l_f1": 0.10526315789473685, "generation_token_f1": 0.10526315789473685, "prompt_id": "6f26ffa7e9fb10c0", "reading_canonical_log_prob": -19.84417724609375, "reading_distances": [22, 21, 20, 22], "reading_log_probs": [-18.985368728637695, -19.658889770507812, -19.96495246887207, -20.226289749145508], "reading_logprob_std": 0.5351802003578103, "tokenization_path_logprob_std": 45.35923963404357, "trajectory_balance_rmse": 253.29845048196927} {"canonical_length": 41, "canonical_nll": 0.9469495633753334, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "07e67bead15cc6c8", "reading_canonical_log_prob": -38.82493209838867, "reading_distances": [31, 38, 35, 24], "reading_log_probs": [-37.57271957397461, -36.77773666381836, -39.497474670410156, -39.942623138427734], "reading_logprob_std": 1.5156066248026017, "tokenization_path_logprob_std": 93.97283483570095, "trajectory_balance_rmse": 767.4534105527542} {"canonical_length": 413, "canonical_nll": 1.2226831468485169, "generation_rouge_l_f1": 0.23283582089552238, "generation_token_f1": 0.4447761194029851, "prompt_id": "52cb3482c2bf1893", "reading_canonical_log_prob": -504.9681396484375, "reading_distances": [26, 24, 22, 33], "reading_log_probs": [-514.2741088867188, -507.5346374511719, -510.4327392578125, -511.82867431640625], "reading_logprob_std": 2.8128148336907954, "tokenization_path_logprob_std": 153.30790799652445, "trajectory_balance_rmse": 7586.5326105539625} {"canonical_length": 265, "canonical_nll": 1.695814370209316, "generation_rouge_l_f1": 0.2248062015503876, "generation_token_f1": 0.4728682170542636, "prompt_id": "3b670d936478ce43", "reading_canonical_log_prob": -449.39080810546875, "reading_distances": [31, 26, 30, 29], "reading_log_probs": [-456.80462646484375, -455.5249328613281, -455.0368957519531, -454.286865234375], "reading_logprob_std": 1.0581094245619589, "tokenization_path_logprob_std": 379.3074772515009, "trajectory_balance_rmse": 6433.778983419708} {"canonical_length": 78, "canonical_nll": 2.305433615660056, "generation_rouge_l_f1": 0.1533923303834808, "generation_token_f1": 0.2831858407079646, "prompt_id": "d940482038c5842a", "reading_canonical_log_prob": -179.82382202148438, "reading_distances": [20, 30, 28, 26], "reading_log_probs": [-177.736572265625, -179.31625366210938, -175.06204223632812, -178.68724060058594], "reading_logprob_std": 1.8750179986278201, "tokenization_path_logprob_std": 195.9400349641723, "trajectory_balance_rmse": 2001.9478098065713} {"canonical_length": 291, "canonical_nll": 0.9788302392074743, "generation_rouge_l_f1": 0.29338103756708406, "generation_token_f1": 0.5152057245080501, "prompt_id": "07d10dd03fee840c", "reading_canonical_log_prob": -284.839599609375, "reading_distances": [26, 27, 19, 24], "reading_log_probs": [-294.2298889160156, -297.55950927734375, -294.85943603515625, -294.16351318359375], "reading_logprob_std": 1.6019415475424579, "tokenization_path_logprob_std": 347.9867695456292, "trajectory_balance_rmse": 5725.291022954588} {"canonical_length": 4, "canonical_nll": 1.6795105934143066, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "7ef3a945dcbdd086", "reading_canonical_log_prob": -6.718042373657227, "reading_distances": [21, 23, 17, 24], "reading_log_probs": [-7.368317127227783, -9.610788345336914, -7.87083101272583, -12.711002349853516], "reading_logprob_std": 2.413357269501864, "tokenization_path_logprob_std": 17.85651646662951, "trajectory_balance_rmse": 45.64551465728045} {"canonical_length": 156, "canonical_nll": 1.0265819842998798, "generation_rouge_l_f1": 0.23409669211195927, "generation_token_f1": 0.3969465648854961, "prompt_id": "ef63cfbdbb1633a2", "reading_canonical_log_prob": -160.14678955078125, "reading_distances": [25, 27, 34, 30], "reading_log_probs": [-167.60198974609375, -161.93499755859375, -170.39810180664062, -169.855712890625], "reading_logprob_std": 3.8694138417112365, "tokenization_path_logprob_std": 154.55373477891217, "trajectory_balance_rmse": 2959.695970310861} {"canonical_length": 152, "canonical_nll": 1.5223972922877262, "generation_rouge_l_f1": 0.22950819672131145, "generation_token_f1": 0.43559718969555034, "prompt_id": "1d1550150b11ce18", "reading_canonical_log_prob": -231.40438842773438, "reading_distances": [24, 30, 29, 27], "reading_log_probs": [-238.59449768066406, -241.22018432617188, -237.68771362304688, -239.25900268554688], "reading_logprob_std": 1.4986545800304294, "tokenization_path_logprob_std": 244.52214981300975, "trajectory_balance_rmse": 3787.43784703109} {"canonical_length": 436, "canonical_nll": 0.8003895610844324, "generation_rouge_l_f1": 0.3304347826086957, "generation_token_f1": 0.47652173913043483, "prompt_id": "becce7dcacc67b08", "reading_canonical_log_prob": -348.9698486328125, "reading_distances": [22, 25, 20, 29], "reading_log_probs": [-371.56829833984375, -370.08905029296875, -359.79693603515625, -366.30059814453125], "reading_logprob_std": 5.25261377038363, "tokenization_path_logprob_std": 112.52856819737691, "trajectory_balance_rmse": 5534.69942319988} {"canonical_length": 289, "canonical_nll": 1.067862857171821, "generation_rouge_l_f1": 0.2003454231433506, "generation_token_f1": 0.3799654576856649, "prompt_id": "11042f17f722f10a", "reading_canonical_log_prob": -308.61236572265625, "reading_distances": [23, 29, 28, 26], "reading_log_probs": [-321.3117370605469, -322.9449157714844, -318.2641906738281, -323.1906433105469], "reading_logprob_std": 2.26797643694526, "tokenization_path_logprob_std": 150.96181681394887, "trajectory_balance_rmse": 6346.871439179859} {"canonical_length": 31, "canonical_nll": 1.1003267841954385, "generation_rouge_l_f1": 0.11643835616438354, "generation_token_f1": 0.15753424657534248, "prompt_id": "b29cd52aed559d3c", "reading_canonical_log_prob": -34.110130310058594, "reading_distances": [21, 19, 19, 18], "reading_log_probs": [-41.08506393432617, -41.30901336669922, -43.20124816894531, -41.49608612060547], "reading_logprob_std": 0.9669735722822924, "tokenization_path_logprob_std": 54.8831448195251, "trajectory_balance_rmse": 542.3235353722555} {"canonical_length": 192, "canonical_nll": 0.8761210441589355, "generation_rouge_l_f1": 0.2577777777777778, "generation_token_f1": 0.4355555555555556, "prompt_id": "3d6b176b83ba1bb0", "reading_canonical_log_prob": -168.21524047851562, "reading_distances": [20, 22, 20, 16], "reading_log_probs": [-180.96591186523438, -182.2447509765625, -178.89913940429688, -176.8134765625], "reading_logprob_std": 2.3838264908952356, "tokenization_path_logprob_std": 42.32870699033084, "trajectory_balance_rmse": 2113.8690843464774} {"canonical_length": 146, "canonical_nll": 1.9193328439372859, "generation_rouge_l_f1": 0.18735362997658078, "generation_token_f1": 0.3372365339578454, "prompt_id": "ed74cc90058602c2", "reading_canonical_log_prob": -280.22259521484375, "reading_distances": [28, 29, 22, 30], "reading_log_probs": [-277.1527404785156, -280.6053466796875, -276.8936462402344, -277.4167175292969], "reading_logprob_std": 1.7386532832535229, "tokenization_path_logprob_std": 254.71560175622304, "trajectory_balance_rmse": 3550.5583920215595} {"canonical_length": 217, "canonical_nll": 1.686430900327621, "generation_rouge_l_f1": 0.22997946611909653, "generation_token_f1": 0.4681724845995893, "prompt_id": "b2e99fb7bff1b5b7", "reading_canonical_log_prob": -365.95550537109375, "reading_distances": [26, 37, 39, 36], "reading_log_probs": [-378.62725830078125, -381.75042724609375, -383.38848876953125, -383.47039794921875], "reading_logprob_std": 2.2643570717328134, "tokenization_path_logprob_std": 256.67732288361077, "trajectory_balance_rmse": 5135.1027417398} {"canonical_length": 318, "canonical_nll": 0.7288065976316824, "generation_rouge_l_f1": 0.21428571428571425, "generation_token_f1": 0.21978021978021975, "prompt_id": "5a32b0efbf18f769", "reading_canonical_log_prob": -231.760498046875, "reading_distances": [26, 29, 29, 25], "reading_log_probs": [-231.97653198242188, -231.9650421142578, -231.00270080566406, -231.28375244140625], "reading_logprob_std": 0.4913984385454802, "tokenization_path_logprob_std": 175.87992066441896, "trajectory_balance_rmse": 5280.332489791671} {"canonical_length": 25, "canonical_nll": 0.564500961303711, "generation_rouge_l_f1": 0.09999999999999999, "generation_token_f1": 0.1090909090909091, "prompt_id": "3697c04353d3a1f1", "reading_canonical_log_prob": -14.112524032592773, "reading_distances": [29, 29, 32, 26], "reading_log_probs": [-16.507225036621094, -15.49989128112793, -21.0084228515625, -18.669160842895508], "reading_logprob_std": 2.446252335332801, "tokenization_path_logprob_std": 61.11180201093262, "trajectory_balance_rmse": 301.7575947444175} {"canonical_length": 83, "canonical_nll": 1.7381922710372741, "generation_rouge_l_f1": 0.4203821656050955, "generation_token_f1": 0.5987261146496815, "prompt_id": "4983a8fb02a849d4", "reading_canonical_log_prob": -144.26995849609375, "reading_distances": [18, 19, 13, 10], "reading_log_probs": [-140.94664001464844, -139.79623413085938, -140.2790985107422, -143.26173400878906], "reading_logprob_std": 1.5348090947003672, "tokenization_path_logprob_std": 105.50437928987323, "trajectory_balance_rmse": 1423.9843655714476} {"canonical_length": 244, "canonical_nll": 1.8038266291383838, "generation_rouge_l_f1": 0.2304147465437788, "generation_token_f1": 0.4792626728110599, "prompt_id": "fe89f27185bbc3b2", "reading_canonical_log_prob": -440.1336975097656, "reading_distances": [43, 39, 29, 41], "reading_log_probs": [-448.313232421875, -449.80279541015625, -450.066162109375, -447.923583984375], "reading_logprob_std": 1.0659434363118234, "tokenization_path_logprob_std": 157.08922670477097, "trajectory_balance_rmse": 6479.227465452336} {"canonical_length": 87, "canonical_nll": 1.5456069420123923, "generation_rouge_l_f1": 0.1875, "generation_token_f1": 0.35714285714285715, "prompt_id": "6b767f6a353fd065", "reading_canonical_log_prob": -134.46780395507812, "reading_distances": [20, 17, 23, 18], "reading_log_probs": [-144.44004821777344, -144.54684448242188, -140.52879333496094, -151.2041015625], "reading_logprob_std": 4.429896071557433, "tokenization_path_logprob_std": 139.32588564213953, "trajectory_balance_rmse": 1679.3564009397996} {"canonical_length": 136, "canonical_nll": 1.359443888944738, "generation_rouge_l_f1": 0.3488372093023256, "generation_token_f1": 0.5058139534883721, "prompt_id": "e77b6aefa4b6596d", "reading_canonical_log_prob": -184.88436889648438, "reading_distances": [31, 25, 29, 28], "reading_log_probs": [-182.26541137695312, -186.98129272460938, -179.50758361816406, -186.48968505859375], "reading_logprob_std": 3.5653118774194743, "tokenization_path_logprob_std": 145.19045800654737, "trajectory_balance_rmse": 2556.693052755231} {"canonical_length": 9, "canonical_nll": 1.474744372897678, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.8571428571428571, "prompt_id": "383239282305a169", "reading_canonical_log_prob": -13.272699356079102, "reading_distances": [31, 34, 35, 32], "reading_log_probs": [-14.230514526367188, -14.805618286132812, -14.5974702835083, -13.69361686706543], "reading_logprob_std": 0.4873810713205811, "tokenization_path_logprob_std": 29.611701429503743, "trajectory_balance_rmse": 120.36285870797941} {"canonical_length": 10, "canonical_nll": 1.6265445709228517, "generation_rouge_l_f1": 0.5714285714285714, "generation_token_f1": 0.5714285714285714, "prompt_id": "29653e40ee0ea781", "reading_canonical_log_prob": -16.265445709228516, "reading_distances": [30, 24, 20, 25], "reading_log_probs": [-14.081696510314941, -16.366357803344727, -14.364483833312988, -14.88750171661377], "reading_logprob_std": 1.0172278648291924, "tokenization_path_logprob_std": 30.5501388481815, "trajectory_balance_rmse": 180.33602602106197} {"canonical_length": 251, "canonical_nll": 1.6846349951755477, "generation_rouge_l_f1": 0.2012072434607646, "generation_token_f1": 0.4225352112676056, "prompt_id": "b8e0545bd0cc139e", "reading_canonical_log_prob": -422.8433837890625, "reading_distances": [48, 52, 34, 45], "reading_log_probs": [-427.4844665527344, -430.415283203125, -432.2012023925781, -438.7467346191406], "reading_logprob_std": 4.770766425134796, "tokenization_path_logprob_std": 88.12269996579236, "trajectory_balance_rmse": 6034.354207167745} {"canonical_length": 430, "canonical_nll": 1.2203850324763807, "generation_rouge_l_f1": 0.19844961240310077, "generation_token_f1": 0.3503875968992247, "prompt_id": "a7d20d598bd1e5dd", "reading_canonical_log_prob": -524.7655639648438, "reading_distances": [36, 35, 38, 29], "reading_log_probs": [-607.803955078125, -607.8934936523438, -611.75390625, -609.142822265625], "reading_logprob_std": 1.841287225100055, "tokenization_path_logprob_std": 369.7758678257468, "trajectory_balance_rmse": 8244.97587093843} {"canonical_length": 47, "canonical_nll": 2.942827265313331, "generation_rouge_l_f1": 0.20437956204379562, "generation_token_f1": 0.3211678832116789, "prompt_id": "ecbe9c8ff4f546fd", "reading_canonical_log_prob": -138.31288146972656, "reading_distances": [12, 14, 10, 13], "reading_log_probs": [-138.3053436279297, -163.94699096679688, -156.11862182617188, -155.57345581054688], "reading_logprob_std": 10.819316510951197, "tokenization_path_logprob_std": 112.93102974796486, "trajectory_balance_rmse": 857.5447691600737} {"canonical_length": 12, "canonical_nll": 1.2401835123697917, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "641a36ce0ce7d08c", "reading_canonical_log_prob": -14.8822021484375, "reading_distances": [25, 23, 26, 29], "reading_log_probs": [-15.46908187866211, -17.720569610595703, -18.83487892150879, -16.169309616088867], "reading_logprob_std": 1.5177136905525177, "tokenization_path_logprob_std": 43.63386761364857, "trajectory_balance_rmse": 293.2221669000599} {"canonical_length": 105, "canonical_nll": 1.6729096912202381, "generation_rouge_l_f1": 0.19607843137254904, "generation_token_f1": 0.3081232492997199, "prompt_id": "32d37daf5bb72f67", "reading_canonical_log_prob": -175.655517578125, "reading_distances": [19, 20, 24, 20], "reading_log_probs": [-177.51751708984375, -177.96310424804688, -187.89816284179688, -180.52239990234375], "reading_logprob_std": 4.801414178016676, "tokenization_path_logprob_std": 147.51094082346208, "trajectory_balance_rmse": 2565.744270294446} {"canonical_length": 19, "canonical_nll": 2.9871842233758223, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.33333333333333337, "prompt_id": "b21a0da1ea5ee3b7", "reading_canonical_log_prob": -56.756500244140625, "reading_distances": [20, 26, 23, 24], "reading_log_probs": [-55.38643264770508, -58.42890167236328, -55.69841766357422, -58.45707702636719], "reading_logprob_std": 1.6795174795431589, "tokenization_path_logprob_std": 66.4783604368206, "trajectory_balance_rmse": 418.4309211081814} {"canonical_length": 281, "canonical_nll": 1.4870874805382561, "generation_rouge_l_f1": 0.21923076923076923, "generation_token_f1": 0.47692307692307695, "prompt_id": "11a80e6e30ed7f8f", "reading_canonical_log_prob": -417.87158203125, "reading_distances": [19, 17, 19, 21], "reading_log_probs": [-419.7655334472656, -425.878662109375, -421.292236328125, -421.06744384765625], "reading_logprob_std": 2.67129476541275, "tokenization_path_logprob_std": 114.44540217872606, "trajectory_balance_rmse": 6361.991396816223} {"canonical_length": 119, "canonical_nll": 1.9088421989889706, "generation_rouge_l_f1": 0.21390374331550804, "generation_token_f1": 0.4117647058823529, "prompt_id": "e2dff42d7f8d6988", "reading_canonical_log_prob": -227.1522216796875, "reading_distances": [23, 21, 26, 21], "reading_log_probs": [-226.82186889648438, -224.8249969482422, -225.82513427734375, -225.75823974609375], "reading_logprob_std": 0.8158827593138007, "tokenization_path_logprob_std": 172.27005586219678, "trajectory_balance_rmse": 2499.7227090979295} {"canonical_length": 338, "canonical_nll": 1.736863175792807, "generation_rouge_l_f1": 0.19931271477663232, "generation_token_f1": 0.43986254295532645, "prompt_id": "bd63a203fd3cb2b2", "reading_canonical_log_prob": -587.0597534179688, "reading_distances": [29, 29, 31, 30], "reading_log_probs": [-589.7119140625, -595.198974609375, -592.9389038085938, -596.9820556640625], "reading_logprob_std": 3.135955864003643, "tokenization_path_logprob_std": 127.79690746340076, "trajectory_balance_rmse": 5963.615900223177} {"canonical_length": 17, "canonical_nll": 0.28997886882108803, "generation_rouge_l_f1": 0.14444444444444446, "generation_token_f1": 0.14444444444444446, "prompt_id": "f5560d8544cd62d8", "reading_canonical_log_prob": -4.929640769958496, "reading_distances": [25, 21, 22, 18], "reading_log_probs": [-6.9208502769470215, -5.945829391479492, -5.1912641525268555, -6.675077438354492], "reading_logprob_std": 0.7802462996804141, "tokenization_path_logprob_std": 59.19343783672161, "trajectory_balance_rmse": 410.9457339969162} {"canonical_length": 143, "canonical_nll": 1.5466895470252404, "generation_rouge_l_f1": 0.2702702702702703, "generation_token_f1": 0.4668304668304668, "prompt_id": "d5ad06fe63008db3", "reading_canonical_log_prob": -221.17660522460938, "reading_distances": [21, 24, 21, 26], "reading_log_probs": [-223.61993408203125, -231.34793090820312, -227.6561737060547, -228.0998077392578], "reading_logprob_std": 3.168314743259185, "tokenization_path_logprob_std": 266.8935505849011, "trajectory_balance_rmse": 3466.077629030523} {"canonical_length": 68, "canonical_nll": 1.3337964450611788, "generation_rouge_l_f1": 0.21021021021021022, "generation_token_f1": 0.31231231231231227, "prompt_id": "32cf7dbccdbc97bc", "reading_canonical_log_prob": -90.69815826416016, "reading_distances": [16, 20, 16, 23], "reading_log_probs": [-94.20989990234375, -95.7437515258789, -91.53021240234375, -100.02537536621094], "reading_logprob_std": 3.5544243463820404, "tokenization_path_logprob_std": 33.13016291521256, "trajectory_balance_rmse": 1845.936334831786} {"canonical_length": 47, "canonical_nll": 1.8509002036236701, "generation_rouge_l_f1": 0.1527272727272727, "generation_token_f1": 0.17454545454545456, "prompt_id": "c180433ec97360b0", "reading_canonical_log_prob": -86.9923095703125, "reading_distances": [32, 24, 28, 36], "reading_log_probs": [-90.35628509521484, -87.33595275878906, -93.52543640136719, -102.04197692871094], "reading_logprob_std": 6.343169806000918, "tokenization_path_logprob_std": 12.7049721119778, "trajectory_balance_rmse": 856.9329826579326} {"canonical_length": 122, "canonical_nll": 0.9525388498775295, "generation_rouge_l_f1": 0.311614730878187, "generation_token_f1": 0.48725212464589235, "prompt_id": "aaf9ee6d37011f66", "reading_canonical_log_prob": -116.2097396850586, "reading_distances": [35, 34, 46, 32], "reading_log_probs": [-142.8516845703125, -135.0140380859375, -122.35284423828125, -134.9126434326172], "reading_logprob_std": 8.479014279796214, "tokenization_path_logprob_std": 105.57423459001862, "trajectory_balance_rmse": 2115.028636355806} {"canonical_length": 19, "canonical_nll": 0.2161637356406764, "generation_rouge_l_f1": 0.08695652173913043, "generation_token_f1": 0.08695652173913043, "prompt_id": "f459f20b3852c4a3", "reading_canonical_log_prob": -4.107110977172852, "reading_distances": [15, 21, 20, 22], "reading_log_probs": [-0.7202613949775696, -0.8531240820884705, -0.9828231334686279, -0.7317484021186829], "reading_logprob_std": 0.12292122539060134, "tokenization_path_logprob_std": 71.15029453306873, "trajectory_balance_rmse": 242.64030378324256} {"canonical_length": 19, "canonical_nll": 1.4983951167056435, "generation_rouge_l_f1": 0.34285714285714286, "generation_token_f1": 0.34285714285714286, "prompt_id": "a05d7928d7de0d6c", "reading_canonical_log_prob": -28.469507217407227, "reading_distances": [26, 28, 28, 20], "reading_log_probs": [-32.34461212158203, -32.583900451660156, -34.8013916015625, -28.088897705078125], "reading_logprob_std": 2.804520465769194, "tokenization_path_logprob_std": 20.032814200580507, "trajectory_balance_rmse": 362.4853184665956} {"canonical_length": 15, "canonical_nll": 1.5440205891927083, "generation_rouge_l_f1": 0.14285714285714288, "generation_token_f1": 0.14285714285714288, "prompt_id": "2c22dccbca7a08c7", "reading_canonical_log_prob": -23.160308837890625, "reading_distances": [26, 28, 26, 28], "reading_log_probs": [-24.23041534423828, -19.511266708374023, -22.36921501159668, -21.26633644104004], "reading_logprob_std": 1.9787343109884719, "tokenization_path_logprob_std": 45.3142451792725, "trajectory_balance_rmse": 359.6157687703864} {"canonical_length": 166, "canonical_nll": 1.892740272613893, "generation_rouge_l_f1": 0.3359375, "generation_token_f1": 0.5078125, "prompt_id": "6e818f0ad7674362", "reading_canonical_log_prob": -314.19488525390625, "reading_distances": [25, 24, 28, 23], "reading_log_probs": [-313.7330322265625, -311.269775390625, -311.4764404296875, -309.2030029296875], "reading_logprob_std": 1.8521110279473163, "tokenization_path_logprob_std": 246.2579777808618, "trajectory_balance_rmse": 3181.400477588779} {"canonical_length": 70, "canonical_nll": 1.129691423688616, "generation_rouge_l_f1": 0.3366336633663366, "generation_token_f1": 0.4752475247524752, "prompt_id": "8ebd94ecdb15c9f4", "reading_canonical_log_prob": -79.07839965820312, "reading_distances": [22, 16, 20, 23], "reading_log_probs": [-80.0999526977539, -88.26864624023438, -83.24171447753906, -84.8144760131836], "reading_logprob_std": 3.3973011974685265, "tokenization_path_logprob_std": 113.68633303714871, "trajectory_balance_rmse": 1789.9552427154922} {"canonical_length": 3, "canonical_nll": 0.5935000578562418, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "2148a4bb4107090b", "reading_canonical_log_prob": -1.7805001735687256, "reading_distances": [15, 22, 21, 16], "reading_log_probs": [-1.5220491886138916, -2.4057726860046387, -1.5290918350219727, -1.607964277267456], "reading_logprob_std": 0.4281439258439198, "tokenization_path_logprob_std": 20.94996750604537, "trajectory_balance_rmse": 49.564128738314885} {"canonical_length": 368, "canonical_nll": 0.952703393023947, "generation_rouge_l_f1": 0.20408163265306123, "generation_token_f1": 0.43956043956043955, "prompt_id": "5d2176dd2ade1ce9", "reading_canonical_log_prob": -350.5948486328125, "reading_distances": [25, 27, 20, 22], "reading_log_probs": [-348.9720153808594, -347.06878662109375, -353.24237060546875, -347.0079345703125], "reading_logprob_std": 2.9254751999657627, "tokenization_path_logprob_std": 176.68672976389485, "trajectory_balance_rmse": 7526.662715840852} {"canonical_length": 12, "canonical_nll": 0.6787544091542562, "generation_rouge_l_f1": 0.06896551724137931, "generation_token_f1": 0.08429118773946359, "prompt_id": "9d6ee475e283101d", "reading_canonical_log_prob": -8.145052909851074, "reading_distances": [30, 28, 33, 33], "reading_log_probs": [-7.03036642074585, -7.017210960388184, -6.705418109893799, -7.614633560180664], "reading_logprob_std": 0.37946605474982115, "tokenization_path_logprob_std": 15.477248419472591, "trajectory_balance_rmse": 264.1725672524799} {"canonical_length": 12, "canonical_nll": 0.5585266749064127, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "a41c0de97fa957ce", "reading_canonical_log_prob": -6.702320098876953, "reading_distances": [20, 17, 24, 23], "reading_log_probs": [-7.950334072113037, -7.437896728515625, -8.030607223510742, -7.496982574462891], "reading_logprob_std": 0.3047014282471484, "tokenization_path_logprob_std": 18.291129587356867, "trajectory_balance_rmse": 199.99519568229294} {"canonical_length": 9, "canonical_nll": 0.16862399048275417, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "916ba9265d9ad9c6", "reading_canonical_log_prob": -1.5176159143447876, "reading_distances": [28, 36, 39, 32], "reading_log_probs": [-1.380875825881958, -0.9586178660392761, -1.0371030569076538, -0.768051266670227], "reading_logprob_std": 0.25607613801408596, "tokenization_path_logprob_std": 28.473144834889595, "trajectory_balance_rmse": 166.30111276801767} {"canonical_length": 99, "canonical_nll": 0.8416353476167929, "generation_rouge_l_f1": 0.5504587155963303, "generation_token_f1": 0.6880733944954128, "prompt_id": "5baddab4b656da75", "reading_canonical_log_prob": -83.3218994140625, "reading_distances": [20, 25, 26, 30], "reading_log_probs": [-83.85621643066406, -83.71109008789062, -84.72248840332031, -85.05862426757812], "reading_logprob_std": 0.6563177348770257, "tokenization_path_logprob_std": 159.32912011192897, "trajectory_balance_rmse": 2072.109039194174} {"canonical_length": 54, "canonical_nll": 0.6192998532895688, "generation_rouge_l_f1": 0.39688715953307385, "generation_token_f1": 0.4046692607003891, "prompt_id": "22a09297862230a9", "reading_canonical_log_prob": -33.44219207763672, "reading_distances": [31, 27, 27, 34], "reading_log_probs": [-35.593650817871094, -37.37334060668945, -36.27288818359375, -35.58229446411133], "reading_logprob_std": 0.842839922322079, "tokenization_path_logprob_std": 73.65669120079616, "trajectory_balance_rmse": 870.8005951039622} {"canonical_length": 40, "canonical_nll": 0.2922965526580811, "generation_rouge_l_f1": 0.1865671641791045, "generation_token_f1": 0.20149253731343286, "prompt_id": "c6508f3625213115", "reading_canonical_log_prob": -11.691862106323242, "reading_distances": [10, 15, 15, 12], "reading_log_probs": [-9.455276489257812, -8.658224105834961, -9.350266456604004, -8.909402847290039], "reading_logprob_std": 0.37424053531868595, "tokenization_path_logprob_std": 27.671022932500488, "trajectory_balance_rmse": 212.1250751125649} {"canonical_length": 272, "canonical_nll": 1.533188314998851, "generation_rouge_l_f1": 0.27547169811320754, "generation_token_f1": 0.5056603773584906, "prompt_id": "a4b74ba3880275a6", "reading_canonical_log_prob": -417.0272216796875, "reading_distances": [20, 15, 19, 14], "reading_log_probs": [-427.6695556640625, -426.5563049316406, -425.325927734375, -419.8209533691406], "reading_logprob_std": 3.482288621603759, "tokenization_path_logprob_std": 162.54563295872035, "trajectory_balance_rmse": 6187.77659584335} {"canonical_length": 17, "canonical_nll": 0.4011688793406767, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.2222222222222222, "prompt_id": "5f38035dc4f5627c", "reading_canonical_log_prob": -6.819870948791504, "reading_distances": [20, 17, 20, 22], "reading_log_probs": [-7.839777946472168, -6.652517318725586, -7.83181619644165, -8.376341819763184], "reading_logprob_std": 0.7278035956215231, "tokenization_path_logprob_std": 31.333860814778554, "trajectory_balance_rmse": 379.02037679427764} {"canonical_length": 34, "canonical_nll": 0.954249662511489, "generation_rouge_l_f1": 0.16438356164383564, "generation_token_f1": 0.19863013698630136, "prompt_id": "0e027c7a776ba5aa", "reading_canonical_log_prob": -32.444488525390625, "reading_distances": [18, 25, 24, 27], "reading_log_probs": [-35.660400390625, -41.939125061035156, -44.73474884033203, -39.50645446777344], "reading_logprob_std": 3.8473668031382835, "tokenization_path_logprob_std": 76.06572514575123, "trajectory_balance_rmse": 544.6742515622417} {"canonical_length": 23, "canonical_nll": 2.5470362124235733, "generation_rouge_l_f1": 0.27027027027027023, "generation_token_f1": 0.37837837837837834, "prompt_id": "9d315888010aca18", "reading_canonical_log_prob": -58.58183288574219, "reading_distances": [19, 24, 26, 19], "reading_log_probs": [-57.77538299560547, -66.3636245727539, -55.46567916870117, -55.00309753417969], "reading_logprob_std": 5.282194637234893, "tokenization_path_logprob_std": 69.40950463828646, "trajectory_balance_rmse": 411.1335280299716} {"canonical_length": 390, "canonical_nll": 0.9680700057592148, "generation_rouge_l_f1": 0.2523076923076923, "generation_token_f1": 0.47384615384615386, "prompt_id": "b3017655183fdef3", "reading_canonical_log_prob": -377.54730224609375, "reading_distances": [23, 22, 20, 23], "reading_log_probs": [-394.9715576171875, -390.18328857421875, -387.0314025878906, -387.90338134765625], "reading_logprob_std": 3.556981483462159, "tokenization_path_logprob_std": 241.64474086173036, "trajectory_balance_rmse": 7559.872926262616} {"canonical_length": 24, "canonical_nll": 0.9711410999298096, "generation_rouge_l_f1": 0.47058823529411764, "generation_token_f1": 0.47058823529411764, "prompt_id": "70f7bc293fb9b6e2", "reading_canonical_log_prob": -23.30738639831543, "reading_distances": [36, 38, 32, 32], "reading_log_probs": [-25.18489646911621, -27.005340576171875, -26.258729934692383, -25.321163177490234], "reading_logprob_std": 0.8543328810114136, "tokenization_path_logprob_std": 52.31694072404289, "trajectory_balance_rmse": 407.16011701677775} {"canonical_length": 85, "canonical_nll": 1.3632431927849264, "generation_rouge_l_f1": 0.19847328244274812, "generation_token_f1": 0.33587786259541985, "prompt_id": "6a1786b00d1d1907", "reading_canonical_log_prob": -115.87567138671875, "reading_distances": [30, 26, 28, 25], "reading_log_probs": [-117.73941040039062, -116.08768463134766, -114.87942504882812, -122.25216674804688], "reading_logprob_std": 3.2286572961247915, "tokenization_path_logprob_std": 152.465252649941, "trajectory_balance_rmse": 1241.6959854291283} {"canonical_length": 108, "canonical_nll": 1.5481089839228876, "generation_rouge_l_f1": 0.20915032679738563, "generation_token_f1": 0.35947712418300654, "prompt_id": "7988aedbe67fdb66", "reading_canonical_log_prob": -167.19577026367188, "reading_distances": [15, 15, 17, 16], "reading_log_probs": [-169.99710083007812, -169.87252807617188, -173.1154022216797, -170.81527709960938], "reading_logprob_std": 1.5028949389394692, "tokenization_path_logprob_std": 59.81771657564745, "trajectory_balance_rmse": 2230.8354202274863} {"canonical_length": 98, "canonical_nll": 2.2681377177335778, "generation_rouge_l_f1": 0.21917808219178084, "generation_token_f1": 0.2904109589041096, "prompt_id": "0b1fe757c0710416", "reading_canonical_log_prob": -222.27749633789062, "reading_distances": [12, 14, 16, 18], "reading_log_probs": [-221.0269775390625, -216.8408203125, -220.53570556640625, -216.68991088867188], "reading_logprob_std": 2.328098204607744, "tokenization_path_logprob_std": 139.5816820436514, "trajectory_balance_rmse": 2060.9704117798806} {"canonical_length": 13, "canonical_nll": 1.0355417545025165, "generation_rouge_l_f1": 0.0885608856088561, "generation_token_f1": 0.09594095940959409, "prompt_id": "78fa2b118c48dd30", "reading_canonical_log_prob": -13.462042808532715, "reading_distances": [22, 33, 29, 27], "reading_log_probs": [-19.906930923461914, -12.719850540161133, -11.658503532409668, -13.309852600097656], "reading_logprob_std": 3.7351219849673445, "tokenization_path_logprob_std": 34.19438800502191, "trajectory_balance_rmse": 264.8969001914141} {"canonical_length": 424, "canonical_nll": 1.154949260207842, "generation_rouge_l_f1": 0.1758957654723127, "generation_token_f1": 0.42996742671009774, "prompt_id": "995817411c15ddee", "reading_canonical_log_prob": -489.698486328125, "reading_distances": [18, 16, 10, 10], "reading_log_probs": [-488.78009033203125, -491.04443359375, -493.0199890136719, -494.1326599121094], "reading_logprob_std": 2.3528704080666136, "tokenization_path_logprob_std": 153.8209581554189, "trajectory_balance_rmse": 7960.183838519671} {"canonical_length": 64, "canonical_nll": 1.226598858833313, "generation_rouge_l_f1": 0.22, "generation_token_f1": 0.42, "prompt_id": "fd32715308ec2a93", "reading_canonical_log_prob": -78.50232696533203, "reading_distances": [37, 47, 42, 46], "reading_log_probs": [-81.48258972167969, -83.5191650390625, -84.32918548583984, -80.91765594482422], "reading_logprob_std": 1.6235869020208455, "tokenization_path_logprob_std": 179.87358687340281, "trajectory_balance_rmse": 1450.2037148684446} {"canonical_length": 333, "canonical_nll": 1.2576271034217812, "generation_rouge_l_f1": 0.1992619926199262, "generation_token_f1": 0.3837638376383764, "prompt_id": "6c1da32736122296", "reading_canonical_log_prob": -418.7898254394531, "reading_distances": [35, 35, 36, 41], "reading_log_probs": [-440.78387451171875, -442.00177001953125, -438.28302001953125, -440.83831787109375], "reading_logprob_std": 1.5666518522214836, "tokenization_path_logprob_std": 211.8144884373544, "trajectory_balance_rmse": 5512.367196924553} {"canonical_length": 53, "canonical_nll": 0.8698786609577682, "generation_rouge_l_f1": 0.29885057471264365, "generation_token_f1": 0.33333333333333337, "prompt_id": "ffb0593302e5d193", "reading_canonical_log_prob": -46.10356903076172, "reading_distances": [40, 37, 35, 40], "reading_log_probs": [-54.68674850463867, -56.74581527709961, -55.343658447265625, -56.04718780517578], "reading_logprob_std": 0.8884049102201369, "tokenization_path_logprob_std": 92.20662307846209, "trajectory_balance_rmse": 840.1049776669003} {"canonical_length": 142, "canonical_nll": 1.3711628443758253, "generation_rouge_l_f1": 0.27, "generation_token_f1": 0.47500000000000003, "prompt_id": "780311bcf262bb9d", "reading_canonical_log_prob": -194.7051239013672, "reading_distances": [35, 31, 31, 37], "reading_log_probs": [-206.6183624267578, -205.41683959960938, -215.19679260253906, -212.17576599121094], "reading_logprob_std": 4.622184045954031, "tokenization_path_logprob_std": 217.93780734951304, "trajectory_balance_rmse": 3315.0743035139635} {"canonical_length": 32, "canonical_nll": 1.6802161931991577, "generation_rouge_l_f1": 0.16513761467889906, "generation_token_f1": 0.1743119266055046, "prompt_id": "ddc773449a2312dd", "reading_canonical_log_prob": -53.76691818237305, "reading_distances": [34, 28, 33, 31], "reading_log_probs": [-56.39167785644531, -56.31227493286133, -61.79896926879883, -57.543914794921875], "reading_logprob_std": 2.5868094129361534, "tokenization_path_logprob_std": 123.67756796279244, "trajectory_balance_rmse": 833.8891196922827} {"canonical_length": 259, "canonical_nll": 0.6937935140602377, "generation_rouge_l_f1": 0.42125984251968507, "generation_token_f1": 0.5984251968503937, "prompt_id": "fa8dcc0f054c9984", "reading_canonical_log_prob": -179.69252014160156, "reading_distances": [21, 28, 27, 21], "reading_log_probs": [-180.43031311035156, -181.9407958984375, -183.1328582763672, -180.508544921875], "reading_logprob_std": 1.2894057093320868, "tokenization_path_logprob_std": 175.37713971416483, "trajectory_balance_rmse": 3062.8807527910285} {"canonical_length": 79, "canonical_nll": 1.1975532242014437, "generation_rouge_l_f1": 0.24203821656050956, "generation_token_f1": 0.356687898089172, "prompt_id": "2d9359e9dd847731", "reading_canonical_log_prob": -94.60670471191406, "reading_distances": [21, 13, 17, 16], "reading_log_probs": [-98.6854248046875, -101.30604553222656, -103.56187438964844, -104.89466857910156], "reading_logprob_std": 2.7225209114647244, "tokenization_path_logprob_std": 137.93797568468372, "trajectory_balance_rmse": 1544.7977341847504} {"canonical_length": 91, "canonical_nll": 1.8248128366994334, "generation_rouge_l_f1": 0.15254237288135594, "generation_token_f1": 0.2711864406779661, "prompt_id": "5cb4ccbc96ab0eb7", "reading_canonical_log_prob": -166.05796813964844, "reading_distances": [27, 35, 32, 31], "reading_log_probs": [-165.23590087890625, -168.80691528320312, -165.7862091064453, -168.65428161621094], "reading_logprob_std": 1.8733684601550462, "tokenization_path_logprob_std": 139.86929792645398, "trajectory_balance_rmse": 2046.183896731873} {"canonical_length": 76, "canonical_nll": 0.5185963982029965, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.671875, "prompt_id": "23716ea232c0bbd4", "reading_canonical_log_prob": -39.413326263427734, "reading_distances": [4, 4, 2, 5], "reading_log_probs": [-42.489219665527344, -41.457515716552734, -41.719642639160156, -43.0225715637207], "reading_logprob_std": 0.7162902226004028, "tokenization_path_logprob_std": 22.97786989767884, "trajectory_balance_rmse": 898.9405766856223} {"canonical_length": 7, "canonical_nll": 1.828620365687779, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "d26a8398a22a7919", "reading_canonical_log_prob": -12.800342559814453, "reading_distances": [39, 35, 40, 40], "reading_log_probs": [-16.118366241455078, -14.92773723602295, -14.693953514099121, -18.111234664916992], "reading_logprob_std": 1.5621884961971395, "tokenization_path_logprob_std": 26.416623095467454, "trajectory_balance_rmse": 187.6549864074895} {"canonical_length": 177, "canonical_nll": 2.0149432683395125, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.3545454545454546, "prompt_id": "cf319e0486bfdfdc", "reading_canonical_log_prob": -356.64495849609375, "reading_distances": [31, 27, 24, 18], "reading_log_probs": [-361.73944091796875, -366.71307373046875, -362.7591552734375, -359.95428466796875], "reading_logprob_std": 2.859860477757677, "tokenization_path_logprob_std": 208.20541018798733, "trajectory_balance_rmse": 4827.670492102212} {"canonical_length": 8, "canonical_nll": 1.4560474157333374, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.23529411764705882, "prompt_id": "2145da07c0ca4001", "reading_canonical_log_prob": -11.6483793258667, "reading_distances": [22, 22, 21, 25], "reading_log_probs": [-14.199138641357422, -18.612945556640625, -16.501550674438477, -14.797028541564941], "reading_logprob_std": 1.9804223393053837, "tokenization_path_logprob_std": 18.065679313569234, "trajectory_balance_rmse": 206.1316210974182} {"canonical_length": 272, "canonical_nll": 1.2727304346421187, "generation_rouge_l_f1": 0.21172022684310018, "generation_token_f1": 0.4423440453686201, "prompt_id": "924cae405c1c753d", "reading_canonical_log_prob": -346.18267822265625, "reading_distances": [27, 22, 24, 21], "reading_log_probs": [-356.0998229980469, -354.5130615234375, -354.97796630859375, -354.1609191894531], "reading_logprob_std": 0.8437727313232184, "tokenization_path_logprob_std": 116.31918742122373, "trajectory_balance_rmse": 6195.003651690809} {"canonical_length": 147, "canonical_nll": 1.3542856229405824, "generation_rouge_l_f1": 0.23821339950372208, "generation_token_f1": 0.4665012406947891, "prompt_id": "eb989912613f5027", "reading_canonical_log_prob": -199.07998657226562, "reading_distances": [30, 30, 34, 36], "reading_log_probs": [-201.70069885253906, -206.35906982421875, -201.6322479248047, -211.59413146972656], "reading_logprob_std": 4.730864562052036, "tokenization_path_logprob_std": 92.28016283523274, "trajectory_balance_rmse": 3173.448167389912} {"canonical_length": 7, "canonical_nll": 1.5643997192382812, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.888888888888889, "prompt_id": "1a61687e59394e25", "reading_canonical_log_prob": -10.950798034667969, "reading_distances": [19, 19, 20, 19], "reading_log_probs": [-13.478920936584473, -13.861639022827148, -14.018777847290039, -12.006739616394043], "reading_logprob_std": 0.9182809445651698, "tokenization_path_logprob_std": 16.238318627129956, "trajectory_balance_rmse": 69.42804630532325} {"canonical_length": 12, "canonical_nll": 1.2177095413208008, "generation_rouge_l_f1": 0.3529411764705882, "generation_token_f1": 0.3529411764705882, "prompt_id": "687a259eb6971a8a", "reading_canonical_log_prob": -14.61251449584961, "reading_distances": [22, 21, 20, 22], "reading_log_probs": [-14.865496635437012, -15.758440017700195, -13.935758590698242, -15.60689926147461], "reading_logprob_std": 0.8341303803121306, "tokenization_path_logprob_std": 30.339476519780312, "trajectory_balance_rmse": 238.15359403469716} {"canonical_length": 177, "canonical_nll": 1.2648030943789725, "generation_rouge_l_f1": 0.22277227722772278, "generation_token_f1": 0.43069306930693074, "prompt_id": "ae453e9c5b7b5add", "reading_canonical_log_prob": -223.87014770507812, "reading_distances": [27, 28, 24, 25], "reading_log_probs": [-232.10850524902344, -223.7992401123047, -225.55328369140625, -226.0338897705078], "reading_logprob_std": 3.6195888367916935, "tokenization_path_logprob_std": 187.07856693766283, "trajectory_balance_rmse": 3293.33240922311} {"canonical_length": 5, "canonical_nll": 2.082530975341797, "generation_rouge_l_f1": 0.06451612903225806, "generation_token_f1": 0.06451612903225806, "prompt_id": "ffdd99b122d4c9bc", "reading_canonical_log_prob": -10.412654876708984, "reading_distances": [42, 43, 43, 40], "reading_log_probs": [-10.129598617553711, -12.31428337097168, -11.96825122833252, -15.256508827209473], "reading_logprob_std": 2.1218656513482483, "tokenization_path_logprob_std": 35.28547692848821, "trajectory_balance_rmse": 75.22854632950946} {"canonical_length": 350, "canonical_nll": 0.95015869140625, "generation_rouge_l_f1": 0.2916666666666667, "generation_token_f1": 0.5381944444444444, "prompt_id": "56f7657e17072dd3", "reading_canonical_log_prob": -332.5555419921875, "reading_distances": [16, 12, 11, 15], "reading_log_probs": [-334.8297424316406, -333.821533203125, -333.98760986328125, -337.7784729003906], "reading_logprob_std": 1.8365792541084265, "tokenization_path_logprob_std": 223.39990318472547, "trajectory_balance_rmse": 7193.7686816275855} {"canonical_length": 172, "canonical_nll": 1.2124674597451852, "generation_rouge_l_f1": 0.26327944572748263, "generation_token_f1": 0.48036951501154734, "prompt_id": "f0c44899ac8f3454", "reading_canonical_log_prob": -208.54440307617188, "reading_distances": [26, 32, 27, 32], "reading_log_probs": [-212.52716064453125, -214.69375610351562, -211.64013671875, -214.432861328125], "reading_logprob_std": 1.4805576257472792, "tokenization_path_logprob_std": 264.5044263630924, "trajectory_balance_rmse": 3953.7609178948437} {"canonical_length": 316, "canonical_nll": 1.199454681782783, "generation_rouge_l_f1": 0.22654867256637168, "generation_token_f1": 0.37168141592920345, "prompt_id": "fb86a83720d2950d", "reading_canonical_log_prob": -379.0276794433594, "reading_distances": [25, 22, 27, 26], "reading_log_probs": [-395.44000244140625, -386.4945983886719, -395.403564453125, -395.2755126953125], "reading_logprob_std": 4.4397744680056555, "tokenization_path_logprob_std": 143.46794958403046, "trajectory_balance_rmse": 5952.261934435302} {"canonical_length": 60, "canonical_nll": 1.522460683186849, "generation_rouge_l_f1": 0.13496932515337423, "generation_token_f1": 0.21472392638036808, "prompt_id": "4f164749bc77ca14", "reading_canonical_log_prob": -91.34764099121094, "reading_distances": [26, 25, 19, 23], "reading_log_probs": [-93.162109375, -92.49610900878906, -92.46174621582031, -97.38688659667969], "reading_logprob_std": 2.362214737668132, "tokenization_path_logprob_std": 96.55433776859495, "trajectory_balance_rmse": 1533.9472871883106} {"canonical_length": 161, "canonical_nll": 1.4557332104037266, "generation_rouge_l_f1": 0.2946859903381642, "generation_token_f1": 0.5072463768115942, "prompt_id": "8d15f97380395bd9", "reading_canonical_log_prob": -234.373046875, "reading_distances": [26, 31, 27, 30], "reading_log_probs": [-246.29531860351562, -244.67605590820312, -249.3922576904297, -244.37294006347656], "reading_logprob_std": 2.2992130998440383, "tokenization_path_logprob_std": 181.13056332166138, "trajectory_balance_rmse": 2947.6903861139035} {"canonical_length": 9, "canonical_nll": 1.0154796176486545, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.8571428571428571, "prompt_id": "a75869a5f1810362", "reading_canonical_log_prob": -9.13931655883789, "reading_distances": [44, 44, 46, 48], "reading_log_probs": [-8.813225746154785, -8.861950874328613, -9.703242301940918, -8.9580659866333], "reading_logprob_std": 0.4171110388567184, "tokenization_path_logprob_std": 15.247612289827648, "trajectory_balance_rmse": 146.69811531686068} {"canonical_length": 228, "canonical_nll": 1.287009791324013, "generation_rouge_l_f1": 0.24680851063829787, "generation_token_f1": 0.5702127659574467, "prompt_id": "f3ee3068b8bfca4b", "reading_canonical_log_prob": -293.438232421875, "reading_distances": [22, 24, 28, 26], "reading_log_probs": [-305.23065185546875, -309.18695068359375, -307.4166259765625, -304.7054443359375], "reading_logprob_std": 2.0671050774565063, "tokenization_path_logprob_std": 109.67893529449665, "trajectory_balance_rmse": 4779.225244046551} {"canonical_length": 126, "canonical_nll": 1.421876211015005, "generation_rouge_l_f1": 0.0851063829787234, "generation_token_f1": 0.1276595744680851, "prompt_id": "6a31e531aca13031", "reading_canonical_log_prob": -179.15640258789062, "reading_distances": [22, 30, 22, 21], "reading_log_probs": [-174.74270629882812, -176.03208923339844, -175.02252197265625, -174.36953735351562], "reading_logprob_std": 0.7123759509045106, "tokenization_path_logprob_std": 135.66957609220265, "trajectory_balance_rmse": 2430.009143953337} {"canonical_length": 11, "canonical_nll": 0.8036344701593573, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "01a0ba7b191885ca", "reading_canonical_log_prob": -8.83997917175293, "reading_distances": [12, 6, 7, 9], "reading_log_probs": [-8.675495147705078, -9.21739387512207, -12.99514102935791, -14.539042472839355], "reading_logprob_std": 2.862241706452205, "tokenization_path_logprob_std": 28.279263182242808, "trajectory_balance_rmse": 153.92515977326707} {"canonical_length": 9, "canonical_nll": 2.0081808302137585, "generation_rouge_l_f1": 0.06106870229007635, "generation_token_f1": 0.06106870229007635, "prompt_id": "8b612a049e9d2b42", "reading_canonical_log_prob": -18.073627471923828, "reading_distances": [16, 11, 14, 13], "reading_log_probs": [-16.783618927001953, -15.88621711730957, -17.141433715820312, -16.65938949584961], "reading_logprob_std": 0.5287189903296164, "tokenization_path_logprob_std": 41.70310591800446, "trajectory_balance_rmse": 159.89029211091807} {"canonical_length": 168, "canonical_nll": 0.635583105541411, "generation_rouge_l_f1": 0.30071599045346065, "generation_token_f1": 0.5011933174224344, "prompt_id": "1792c7fe245165d5", "reading_canonical_log_prob": -106.77796173095703, "reading_distances": [53, 51, 52, 52], "reading_log_probs": [-112.46949005126953, -110.92605590820312, -111.9970474243164, -112.57698059082031], "reading_logprob_std": 0.7542017202927738, "tokenization_path_logprob_std": 65.63094349222024, "trajectory_balance_rmse": 2298.183799666539} {"canonical_length": 240, "canonical_nll": 1.526435089111328, "generation_rouge_l_f1": 0.31643002028397565, "generation_token_f1": 0.5760649087221095, "prompt_id": "9148af04e8535f5f", "reading_canonical_log_prob": -366.34442138671875, "reading_distances": [24, 22, 23, 22], "reading_log_probs": [-362.0837707519531, -367.4881591796875, -367.52044677734375, -366.0819091796875], "reading_logprob_std": 2.5625173713483673, "tokenization_path_logprob_std": 120.9593193453343, "trajectory_balance_rmse": 4388.835788951643} {"canonical_length": 11, "canonical_nll": 1.072273687882857, "generation_rouge_l_f1": 0.07111111111111111, "generation_token_f1": 0.08, "prompt_id": "4f964e48306df1f7", "reading_canonical_log_prob": -11.795010566711426, "reading_distances": [19, 16, 11, 12], "reading_log_probs": [-12.742400169372559, -12.66817569732666, -13.03221607208252, -10.866519927978516], "reading_logprob_std": 0.9864565332555817, "tokenization_path_logprob_std": 24.154980118107826, "trajectory_balance_rmse": 156.06749560119758} {"canonical_length": 13, "canonical_nll": 1.1911512521597056, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.4444444444444444, "prompt_id": "65511a559eb975a4", "reading_canonical_log_prob": -15.484966278076172, "reading_distances": [22, 21, 18, 18], "reading_log_probs": [-16.463462829589844, -20.108642578125, -16.79416275024414, -20.23331069946289], "reading_logprob_std": 2.0501525976773967, "tokenization_path_logprob_std": 7.71784970956862, "trajectory_balance_rmse": 320.44410551060855} {"canonical_length": 273, "canonical_nll": 1.149850586831788, "generation_rouge_l_f1": 0.2380106571936057, "generation_token_f1": 0.5079928952042629, "prompt_id": "dfd43bd8f8a6af69", "reading_canonical_log_prob": -313.9092102050781, "reading_distances": [16, 9, 16, 19], "reading_log_probs": [-312.97772216796875, -314.853515625, -313.8534240722656, -317.45709228515625], "reading_logprob_std": 1.9389737554769817, "tokenization_path_logprob_std": 227.40179365533257, "trajectory_balance_rmse": 4677.631504618023} {"canonical_length": 137, "canonical_nll": 1.8882914076756387, "generation_rouge_l_f1": 0.19746835443037974, "generation_token_f1": 0.3848101265822785, "prompt_id": "9ebcabae9c8c462a", "reading_canonical_log_prob": -258.6959228515625, "reading_distances": [13, 16, 16, 21], "reading_log_probs": [-266.5311279296875, -257.57611083984375, -264.031005859375, -260.822509765625], "reading_logprob_std": 3.889413898337957, "tokenization_path_logprob_std": 178.15416759017953, "trajectory_balance_rmse": 3300.636360815767} {"canonical_length": 224, "canonical_nll": 1.0538294655936105, "generation_rouge_l_f1": 0.25274725274725274, "generation_token_f1": 0.4468864468864469, "prompt_id": "4e90bb5af9e629db", "reading_canonical_log_prob": -236.05780029296875, "reading_distances": [57, 54, 44, 47], "reading_log_probs": [-237.9176788330078, -240.0740966796875, -234.8072052001953, -239.2624053955078], "reading_logprob_std": 2.316269829123015, "tokenization_path_logprob_std": 102.31237911224214, "trajectory_balance_rmse": 4178.649492603931} {"canonical_length": 17, "canonical_nll": 0.7255789812873391, "generation_rouge_l_f1": 0.1016949152542373, "generation_token_f1": 0.1016949152542373, "prompt_id": "3779413f4fba856c", "reading_canonical_log_prob": -12.334842681884766, "reading_distances": [12, 16, 21, 18], "reading_log_probs": [-11.902875900268555, -14.050712585449219, -13.825162887573242, -18.813152313232422], "reading_logprob_std": 2.9392750399942877, "tokenization_path_logprob_std": 75.17128812571416, "trajectory_balance_rmse": 416.27789071911} {"canonical_length": 125, "canonical_nll": 1.220606689453125, "generation_rouge_l_f1": 0.1907216494845361, "generation_token_f1": 0.34536082474226804, "prompt_id": "f749a091b4c13176", "reading_canonical_log_prob": -152.57583618164062, "reading_distances": [18, 22, 15, 19], "reading_log_probs": [-158.82644653320312, -158.65147399902344, -160.07366943359375, -158.32888793945312], "reading_logprob_std": 0.7640194689615724, "tokenization_path_logprob_std": 101.68036024320739, "trajectory_balance_rmse": 2785.403267080521} {"canonical_length": 45, "canonical_nll": 1.65579104953342, "generation_rouge_l_f1": 0.25287356321839083, "generation_token_f1": 0.4367816091954023, "prompt_id": "6378420685c9b069", "reading_canonical_log_prob": -74.5105972290039, "reading_distances": [23, 19, 22, 21], "reading_log_probs": [-76.66085815429688, -75.52236938476562, -76.15478515625, -74.16487121582031], "reading_logprob_std": 1.0795340751058222, "tokenization_path_logprob_std": 156.07630395914921, "trajectory_balance_rmse": 1224.194868811541} {"canonical_length": 119, "canonical_nll": 0.9417804108948267, "generation_rouge_l_f1": 0.30769230769230765, "generation_token_f1": 0.41687344913151364, "prompt_id": "634524f69adfb3f0", "reading_canonical_log_prob": -112.07186889648438, "reading_distances": [28, 25, 28, 25], "reading_log_probs": [-115.97245788574219, -117.4068603515625, -117.49293518066406, -117.1717529296875], "reading_logprob_std": 0.7055442730689117, "tokenization_path_logprob_std": 86.09569934558145, "trajectory_balance_rmse": 1939.6824200754545} {"canonical_length": 162, "canonical_nll": 1.601744098427855, "generation_rouge_l_f1": 0.18097447795823665, "generation_token_f1": 0.3990719257540603, "prompt_id": "c5c8a284983e4726", "reading_canonical_log_prob": -259.4825439453125, "reading_distances": [20, 22, 23, 19], "reading_log_probs": [-270.94610595703125, -272.44775390625, -271.0970764160156, -275.5394592285156], "reading_logprob_std": 2.131008553374862, "tokenization_path_logprob_std": 46.664765615215245, "trajectory_balance_rmse": 2886.0998836296317} {"canonical_length": 88, "canonical_nll": 1.406823071566495, "generation_rouge_l_f1": 0.23602484472049687, "generation_token_f1": 0.31677018633540377, "prompt_id": "626ba72af7ac67f2", "reading_canonical_log_prob": -123.80043029785156, "reading_distances": [31, 26, 29, 27], "reading_log_probs": [-118.77208709716797, -120.87725830078125, -118.7196044921875, -120.1730728149414], "reading_logprob_std": 1.066973075386978, "tokenization_path_logprob_std": 27.100401143441314, "trajectory_balance_rmse": 1340.5673122143746} {"canonical_length": 297, "canonical_nll": 0.9811882250236742, "generation_rouge_l_f1": 0.30601092896174864, "generation_token_f1": 0.5063752276867031, "prompt_id": "c0257d516383b045", "reading_canonical_log_prob": -291.41290283203125, "reading_distances": [26, 29, 34, 23], "reading_log_probs": [-292.7803649902344, -294.20220947265625, -295.5119323730469, -293.3603515625], "reading_logprob_std": 1.1857798180965455, "tokenization_path_logprob_std": 233.60802189515869, "trajectory_balance_rmse": 6470.208585792425} {"canonical_length": 38, "canonical_nll": 0.5403120141280325, "generation_rouge_l_f1": 0.19178082191780818, "generation_token_f1": 0.27397260273972607, "prompt_id": "dae9d23ce12cb989", "reading_canonical_log_prob": -20.531856536865234, "reading_distances": [21, 14, 18, 13], "reading_log_probs": [-22.993539810180664, -19.683027267456055, -18.75877571105957, -19.90225601196289], "reading_logprob_std": 1.8407110304702652, "tokenization_path_logprob_std": 29.741541319485773, "trajectory_balance_rmse": 272.6075256811076} {"canonical_length": 10, "canonical_nll": 0.4533041477203369, "generation_rouge_l_f1": 0.10191082802547771, "generation_token_f1": 0.10191082802547771, "prompt_id": "cde09f27bb3893ac", "reading_canonical_log_prob": -4.533041477203369, "reading_distances": [29, 28, 22, 26], "reading_log_probs": [-6.487037181854248, -7.982783794403076, -5.77309513092041, -7.327946662902832], "reading_logprob_std": 0.9653668940349256, "tokenization_path_logprob_std": 29.930037522571848, "trajectory_balance_rmse": 204.4629089725426} {"canonical_length": 174, "canonical_nll": 1.6286675464147808, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.4449760765550239, "prompt_id": "64bb51344f196fcb", "reading_canonical_log_prob": -283.3881530761719, "reading_distances": [27, 18, 20, 18], "reading_log_probs": [-302.3888244628906, -297.1694030761719, -293.0627746582031, -295.41363525390625], "reading_logprob_std": 3.9617388040920285, "tokenization_path_logprob_std": 111.79118158727897, "trajectory_balance_rmse": 3447.344602607872} {"canonical_length": 312, "canonical_nll": 1.3740428044245794, "generation_rouge_l_f1": 0.2146690518783542, "generation_token_f1": 0.42218246869409654, "prompt_id": "bc3436553628cc88", "reading_canonical_log_prob": -428.70135498046875, "reading_distances": [37, 30, 30, 31], "reading_log_probs": [-434.0008239746094, -437.2213439941406, -430.4400329589844, -444.0113525390625], "reading_logprob_std": 5.770129797306304, "tokenization_path_logprob_std": 429.394966901521, "trajectory_balance_rmse": 7080.0468603161125} {"canonical_length": 39, "canonical_nll": 1.3790797698192108, "generation_rouge_l_f1": 0.16058394160583941, "generation_token_f1": 0.2116788321167883, "prompt_id": "b3b9c2d52d21969e", "reading_canonical_log_prob": -53.78411102294922, "reading_distances": [27, 30, 29, 26], "reading_log_probs": [-47.688926696777344, -49.15877914428711, -45.810691833496094, -50.142417907714844], "reading_logprob_std": 1.8852262503677373, "tokenization_path_logprob_std": 62.96241863028577, "trajectory_balance_rmse": 853.3207139254423} {"canonical_length": 21, "canonical_nll": 0.44665027800060453, "generation_rouge_l_f1": 0.3125, "generation_token_f1": 0.375, "prompt_id": "cc11564e8f134b4b", "reading_canonical_log_prob": -9.379655838012695, "reading_distances": [15, 14, 14, 14], "reading_log_probs": [-10.397066116333008, -10.067957878112793, -10.780561447143555, -12.74848747253418], "reading_logprob_std": 1.2024398422301934, "tokenization_path_logprob_std": 15.021306672920716, "trajectory_balance_rmse": 167.78748072232764} {"canonical_length": 20, "canonical_nll": 1.1066983222961426, "generation_rouge_l_f1": 0.17687074829931973, "generation_token_f1": 0.21768707482993196, "prompt_id": "e07efb6f1797853d", "reading_canonical_log_prob": -22.13396644592285, "reading_distances": [27, 28, 25, 24], "reading_log_probs": [-25.45095443725586, -25.115036010742188, -26.20442008972168, -26.49037742614746], "reading_logprob_std": 0.640380427771403, "tokenization_path_logprob_std": 60.89318772598331, "trajectory_balance_rmse": 342.80729298852174} {"canonical_length": 10, "canonical_nll": 0.4856422901153564, "generation_rouge_l_f1": 0.72, "generation_token_f1": 0.72, "prompt_id": "302fe191c6bb7b5b", "reading_canonical_log_prob": -4.8564229011535645, "reading_distances": [24, 30, 26, 22], "reading_log_probs": [-3.394963502883911, -4.072648525238037, -3.2775416374206543, -2.847193956375122], "reading_logprob_std": 0.5076275989080096, "tokenization_path_logprob_std": 29.91239020872504, "trajectory_balance_rmse": 175.5198659283155} {"canonical_length": 22, "canonical_nll": 1.7806927074085583, "generation_rouge_l_f1": 0.6451612903225806, "generation_token_f1": 0.6451612903225806, "prompt_id": "de0e448f061c6724", "reading_canonical_log_prob": -39.17523956298828, "reading_distances": [32, 25, 34, 28], "reading_log_probs": [-42.751556396484375, -38.50579071044922, -42.388206481933594, -40.63836669921875], "reading_logprob_std": 1.9430880052437518, "tokenization_path_logprob_std": 60.53024149999271, "trajectory_balance_rmse": 671.6319781392443} {"canonical_length": 79, "canonical_nll": 1.0847809224189082, "generation_rouge_l_f1": 0.4204545454545454, "generation_token_f1": 0.5454545454545455, "prompt_id": "f979fa68d4ceeb01", "reading_canonical_log_prob": -85.69769287109375, "reading_distances": [26, 27, 25, 24], "reading_log_probs": [-91.84272766113281, -94.74813079833984, -91.2036361694336, -93.29007720947266], "reading_logprob_std": 1.580802135400658, "tokenization_path_logprob_std": 107.30373161099268, "trajectory_balance_rmse": 1600.0810973715124} {"canonical_length": 193, "canonical_nll": 1.4736173165276878, "generation_rouge_l_f1": 0.20734341252699787, "generation_token_f1": 0.4319654427645788, "prompt_id": "e2f73d0c042e6f14", "reading_canonical_log_prob": -284.40814208984375, "reading_distances": [12, 13, 12, 16], "reading_log_probs": [-283.0965576171875, -284.97198486328125, -284.9179992675781, -275.4324035644531], "reading_logprob_std": 4.532817244182064, "tokenization_path_logprob_std": 273.6799200809332, "trajectory_balance_rmse": 4521.17724502577} {"canonical_length": 216, "canonical_nll": 1.9129702250162761, "generation_rouge_l_f1": 0.21757322175732216, "generation_token_f1": 0.44769874476987453, "prompt_id": "409538e18660e490", "reading_canonical_log_prob": -413.2015686035156, "reading_distances": [28, 38, 30, 31], "reading_log_probs": [-417.87274169921875, -417.4302978515625, -417.89532470703125, -415.48785400390625], "reading_logprob_std": 1.142701810543534, "tokenization_path_logprob_std": 105.80963258299863, "trajectory_balance_rmse": 4974.494620913845} {"canonical_length": 64, "canonical_nll": 2.094329833984375, "generation_rouge_l_f1": 0.28235294117647064, "generation_token_f1": 0.4, "prompt_id": "cc38da16fec412a8", "reading_canonical_log_prob": -134.037109375, "reading_distances": [32, 31, 26, 30], "reading_log_probs": [-137.23764038085938, -139.24740600585938, -135.27137756347656, -139.82675170898438], "reading_logprob_std": 2.0717287875066757, "tokenization_path_logprob_std": 81.48937713387564, "trajectory_balance_rmse": 1133.465868351297} {"canonical_length": 53, "canonical_nll": 0.7230063384433962, "generation_rouge_l_f1": 0.3132530120481928, "generation_token_f1": 0.3373493975903614, "prompt_id": "0e93241c78cc22bd", "reading_canonical_log_prob": -38.3193359375, "reading_distances": [32, 28, 26, 35], "reading_log_probs": [-41.96589660644531, -42.27337646484375, -40.932411193847656, -41.085609436035156], "reading_logprob_std": 0.6563782914545387, "tokenization_path_logprob_std": 141.08975866925144, "trajectory_balance_rmse": 886.6182907605396} {"canonical_length": 19, "canonical_nll": 0.262618667200992, "generation_rouge_l_f1": 0.21052631578947367, "generation_token_f1": 0.21052631578947367, "prompt_id": "32e3b4bcc84e208d", "reading_canonical_log_prob": -4.989754676818848, "reading_distances": [31, 28, 36, 29], "reading_log_probs": [-5.224601745605469, -4.248859405517578, -6.037315368652344, -4.792425632476807], "reading_logprob_std": 0.7551568363858465, "tokenization_path_logprob_std": 24.037412173864414, "trajectory_balance_rmse": 375.7036482893923} {"canonical_length": 28, "canonical_nll": 1.6701654706682478, "generation_rouge_l_f1": 0.24390243902439027, "generation_token_f1": 0.34146341463414637, "prompt_id": "08748791c0916706", "reading_canonical_log_prob": -46.76463317871094, "reading_distances": [22, 19, 23, 21], "reading_log_probs": [-61.09075927734375, -66.70602416992188, -58.751949310302734, -59.68964767456055], "reading_logprob_std": 3.5630098444365412, "tokenization_path_logprob_std": 113.03800626249817, "trajectory_balance_rmse": 883.3063945550324} {"canonical_length": 268, "canonical_nll": 1.5313757142024254, "generation_rouge_l_f1": 0.2085661080074488, "generation_token_f1": 0.4022346368715084, "prompt_id": "340800ab7642e87c", "reading_canonical_log_prob": -410.40869140625, "reading_distances": [20, 20, 25, 20], "reading_log_probs": [-421.4994812011719, -417.80828857421875, -425.68670654296875, -414.42388916015625], "reading_logprob_std": 4.844209999972094, "tokenization_path_logprob_std": 291.3080075962173, "trajectory_balance_rmse": 4808.825084112349} {"canonical_length": 335, "canonical_nll": 1.3183905302588619, "generation_rouge_l_f1": 0.26513911620294595, "generation_token_f1": 0.5040916530278233, "prompt_id": "ba9377f6ac227fec", "reading_canonical_log_prob": -441.66082763671875, "reading_distances": [21, 25, 30, 26], "reading_log_probs": [-442.04742431640625, -443.58856201171875, -442.7471008300781, -443.2300109863281], "reading_logprob_std": 0.6666446972588527, "tokenization_path_logprob_std": 140.32440054821126, "trajectory_balance_rmse": 7146.905193428687} {"canonical_length": 12, "canonical_nll": 0.13283525904019675, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "39897df6fcbea057", "reading_canonical_log_prob": -1.5940231084823608, "reading_distances": [22, 20, 26, 18], "reading_log_probs": [-2.6171364784240723, -3.0727105140686035, -4.429746150970459, -3.7065258026123047], "reading_logprob_std": 0.7877280536829799, "tokenization_path_logprob_std": 67.97425683918271, "trajectory_balance_rmse": 216.5630521213075} {"canonical_length": 55, "canonical_nll": 0.6142262545498934, "generation_rouge_l_f1": 0.07407407407407407, "generation_token_f1": 0.07407407407407407, "prompt_id": "d9f80153d939b659", "reading_canonical_log_prob": -33.78244400024414, "reading_distances": [21, 17, 22, 21], "reading_log_probs": [-38.625099182128906, -38.01087951660156, -41.87369918823242, -39.161468505859375], "reading_logprob_std": 1.703422680975981, "tokenization_path_logprob_std": 84.74238732340827, "trajectory_balance_rmse": 1135.8671199106561} {"canonical_length": 105, "canonical_nll": 1.5314878917875745, "generation_rouge_l_f1": 0.2569832402234637, "generation_token_f1": 0.36312849162011174, "prompt_id": "ea105e2e33c793e5", "reading_canonical_log_prob": -160.8062286376953, "reading_distances": [34, 44, 32, 30], "reading_log_probs": [-167.46926879882812, -170.61509704589844, -169.10720825195312, -165.695556640625], "reading_logprob_std": 2.1181774587962026, "tokenization_path_logprob_std": 111.7531280155133, "trajectory_balance_rmse": 2290.047860399186} {"canonical_length": 311, "canonical_nll": 1.4810010903901225, "generation_rouge_l_f1": 0.22141560798548093, "generation_token_f1": 0.4355716878402904, "prompt_id": "2fb44fa5d32cf514", "reading_canonical_log_prob": -460.5913391113281, "reading_distances": [33, 30, 30, 31], "reading_log_probs": [-470.1924743652344, -475.5543518066406, -472.46099853515625, -474.8463134765625], "reading_logprob_std": 2.437793233457039, "tokenization_path_logprob_std": 83.40479140318233, "trajectory_balance_rmse": 6510.200236838438} {"canonical_length": 134, "canonical_nll": 1.0879507491837686, "generation_rouge_l_f1": 0.23404255319148934, "generation_token_f1": 0.36170212765957444, "prompt_id": "f67931e5043d993b", "reading_canonical_log_prob": -145.785400390625, "reading_distances": [27, 26, 23, 29], "reading_log_probs": [-149.60708618164062, -151.29493713378906, -149.0601806640625, -151.93682861328125], "reading_logprob_std": 1.3618893388043194, "tokenization_path_logprob_std": 222.5264791984602, "trajectory_balance_rmse": 2626.40018546209} {"canonical_length": 42, "canonical_nll": 1.5037651062011719, "generation_rouge_l_f1": 0.184, "generation_token_f1": 0.26399999999999996, "prompt_id": "f4dd2f85181ef8ef", "reading_canonical_log_prob": -63.15813446044922, "reading_distances": [23, 21, 17, 16], "reading_log_probs": [-70.77220153808594, -64.49278259277344, -67.48384094238281, -66.9707260131836], "reading_logprob_std": 2.5827239244746227, "tokenization_path_logprob_std": 73.0951918670028, "trajectory_balance_rmse": 967.0356042399493} {"canonical_length": 386, "canonical_nll": 0.9010006603181671, "generation_rouge_l_f1": 0.22087745839636916, "generation_token_f1": 0.36006051437216335, "prompt_id": "044b1de06d0cdd9b", "reading_canonical_log_prob": -347.7862548828125, "reading_distances": [44, 47, 39, 43], "reading_log_probs": [-358.032470703125, -355.3140869140625, -356.61968994140625, -358.4769287109375], "reading_logprob_std": 1.4358680931219407, "tokenization_path_logprob_std": 349.00137800087634, "trajectory_balance_rmse": 6566.449618428324} {"canonical_length": 120, "canonical_nll": 0.8130342483520507, "generation_rouge_l_f1": 0.254335260115607, "generation_token_f1": 0.3815028901734104, "prompt_id": "c7dd76b068a7ae00", "reading_canonical_log_prob": -97.5641098022461, "reading_distances": [41, 43, 39, 47], "reading_log_probs": [-106.33161926269531, -106.6903305053711, -104.8701400756836, -106.54872131347656], "reading_logprob_std": 0.839767319240347, "tokenization_path_logprob_std": 227.14267505410456, "trajectory_balance_rmse": 2189.365969965909} {"canonical_length": 35, "canonical_nll": 2.624527849469866, "generation_rouge_l_f1": 0.47457627118644063, "generation_token_f1": 0.47457627118644063, "prompt_id": "2fa90c11cfdc9fd0", "reading_canonical_log_prob": -91.85847473144531, "reading_distances": [20, 19, 24, 27], "reading_log_probs": [-96.81599426269531, -97.5782241821289, -99.71719360351562, -101.49478149414062], "reading_logprob_std": 2.120601960434587, "tokenization_path_logprob_std": 17.014336412633238, "trajectory_balance_rmse": 704.7686791812469} {"canonical_length": 33, "canonical_nll": 1.2507958845658735, "generation_rouge_l_f1": 0.2282608695652174, "generation_token_f1": 0.2934782608695652, "prompt_id": "acac7a68437d400e", "reading_canonical_log_prob": -41.27626419067383, "reading_distances": [24, 27, 34, 27], "reading_log_probs": [-40.82102584838867, -40.25585174560547, -44.45432662963867, -43.378623962402344], "reading_logprob_std": 2.0124122826525546, "tokenization_path_logprob_std": 67.68673660859952, "trajectory_balance_rmse": 522.7966391946442} {"canonical_length": 41, "canonical_nll": 0.7476232575207222, "generation_rouge_l_f1": 0.2531645569620253, "generation_token_f1": 0.2531645569620253, "prompt_id": "98a669e9058b6ddc", "reading_canonical_log_prob": -30.65255355834961, "reading_distances": [12, 12, 9, 12], "reading_log_probs": [-28.665794372558594, -33.35615539550781, -29.542098999023438, -33.36289596557617], "reading_logprob_std": 2.482869992393883, "tokenization_path_logprob_std": 66.84393364532727, "trajectory_balance_rmse": 537.2777433795804} {"canonical_length": 133, "canonical_nll": 1.3770439893679511, "generation_rouge_l_f1": 0.20565552699228792, "generation_token_f1": 0.38046272493573263, "prompt_id": "610adc3864696d51", "reading_canonical_log_prob": -183.1468505859375, "reading_distances": [21, 19, 21, 21], "reading_log_probs": [-188.90208435058594, -190.26113891601562, -191.7925567626953, -189.04769897460938], "reading_logprob_std": 1.3408636857468408, "tokenization_path_logprob_std": 161.44894524543784, "trajectory_balance_rmse": 3383.7186961309267} {"canonical_length": 34, "canonical_nll": 0.76397424585679, "generation_rouge_l_f1": 0.31304347826086953, "generation_token_f1": 0.4695652173913043, "prompt_id": "74b66fda46123ad6", "reading_canonical_log_prob": -25.97512435913086, "reading_distances": [38, 42, 44, 39], "reading_log_probs": [-30.18025016784668, -26.557661056518555, -28.61836814880371, -29.233095169067383], "reading_logprob_std": 1.5341143107903763, "tokenization_path_logprob_std": 72.94176510352287, "trajectory_balance_rmse": 612.0763801056185} {"canonical_length": 23, "canonical_nll": 0.6720493980076002, "generation_rouge_l_f1": 0.47727272727272724, "generation_token_f1": 0.5, "prompt_id": "9ce2e67b60ca246c", "reading_canonical_log_prob": -15.457136154174805, "reading_distances": [19, 18, 17, 19], "reading_log_probs": [-16.455381393432617, -16.244495391845703, -16.34274673461914, -15.566496849060059], "reading_logprob_std": 0.39991399496876034, "tokenization_path_logprob_std": 74.60361738996957, "trajectory_balance_rmse": 433.0897708178684} {"canonical_length": 9, "canonical_nll": 0.8121350606282552, "generation_rouge_l_f1": 0.4615384615384615, "generation_token_f1": 0.4615384615384615, "prompt_id": "f37797ee8ed6a36e", "reading_canonical_log_prob": -7.309215545654297, "reading_distances": [42, 32, 42, 48], "reading_log_probs": [-8.101794242858887, -7.4295172691345215, -7.933881759643555, -7.355274677276611], "reading_logprob_std": 0.36879579384736333, "tokenization_path_logprob_std": 28.78139134012331, "trajectory_balance_rmse": 172.2497367389381} {"canonical_length": 25, "canonical_nll": 1.2963720703125, "generation_rouge_l_f1": 0.2342342342342342, "generation_token_f1": 0.32432432432432434, "prompt_id": "b9be7345d7f0f09d", "reading_canonical_log_prob": -32.4093017578125, "reading_distances": [29, 23, 24, 28], "reading_log_probs": [-33.5026741027832, -32.49449920654297, -33.61565399169922, -33.75472640991211], "reading_logprob_std": 0.5742540985295208, "tokenization_path_logprob_std": 22.264623979596347, "trajectory_balance_rmse": 598.8446254882932} {"canonical_length": 102, "canonical_nll": 0.843317892037186, "generation_rouge_l_f1": 0.44881889763779526, "generation_token_f1": 0.6141732283464567, "prompt_id": "101280c93f2c70da", "reading_canonical_log_prob": -86.01842498779297, "reading_distances": [18, 16, 10, 18], "reading_log_probs": [-90.13233947753906, -88.6663589477539, -90.51823425292969, -91.38253021240234], "reading_logprob_std": 1.1333955824315227, "tokenization_path_logprob_std": 83.2417633996584, "trajectory_balance_rmse": 844.0723613928112} {"canonical_length": 309, "canonical_nll": 1.1472667706437096, "generation_rouge_l_f1": 0.3079584775086505, "generation_token_f1": 0.5570934256055362, "prompt_id": "bc039a3de3397314", "reading_canonical_log_prob": -354.50543212890625, "reading_distances": [35, 38, 27, 35], "reading_log_probs": [-357.4788818359375, -361.71466064453125, -358.12646484375, -357.35577392578125], "reading_logprob_std": 2.0584247303776024, "tokenization_path_logprob_std": 274.16761708139535, "trajectory_balance_rmse": 7036.429305840145} {"canonical_length": 144, "canonical_nll": 1.0009341769748263, "generation_rouge_l_f1": 0.30319148936170215, "generation_token_f1": 0.4042553191489362, "prompt_id": "3bf14cf65dcd8fce", "reading_canonical_log_prob": -144.134521484375, "reading_distances": [35, 36, 35, 39], "reading_log_probs": [-148.44955444335938, -148.1936492919922, -147.87136840820312, -148.66323852539062], "reading_logprob_std": 0.3411847890481103, "tokenization_path_logprob_std": 159.13775204665342, "trajectory_balance_rmse": 2228.8527436535633} {"canonical_length": 38, "canonical_nll": 1.768587212813528, "generation_rouge_l_f1": 0.07860262008733623, "generation_token_f1": 0.11353711790393012, "prompt_id": "ab60c5be348f9273", "reading_canonical_log_prob": -67.20631408691406, "reading_distances": [15, 20, 25, 22], "reading_log_probs": [-71.50709533691406, -74.06808471679688, -75.06143188476562, -73.79474639892578], "reading_logprob_std": 1.5025330510218047, "tokenization_path_logprob_std": 80.77179384749951, "trajectory_balance_rmse": 965.0802446656214} {"canonical_length": 12, "canonical_nll": 2.6842727661132812, "generation_rouge_l_f1": 0.4705882352941177, "generation_token_f1": 0.4705882352941177, "prompt_id": "8392d28ac48cc769", "reading_canonical_log_prob": -32.211273193359375, "reading_distances": [38, 35, 33, 32], "reading_log_probs": [-42.30015182495117, -38.82861328125, -37.77113342285156, -39.93864440917969], "reading_logprob_std": 1.9405506403563186, "tokenization_path_logprob_std": 66.39971939720753, "trajectory_balance_rmse": 354.415232654713} {"canonical_length": 90, "canonical_nll": 1.8579216851128473, "generation_rouge_l_f1": 0.1756373937677054, "generation_token_f1": 0.3002832861189802, "prompt_id": "2a39144255d482e3", "reading_canonical_log_prob": -167.21295166015625, "reading_distances": [29, 29, 25, 23], "reading_log_probs": [-185.22898864746094, -186.06015014648438, -187.71826171875, -200.93710327148438], "reading_logprob_std": 7.373618600389884, "tokenization_path_logprob_std": 157.76798533648912, "trajectory_balance_rmse": 1759.4135486332027} {"canonical_length": 292, "canonical_nll": 1.4357329068118578, "generation_rouge_l_f1": 0.21428571428571427, "generation_token_f1": 0.4642857142857143, "prompt_id": "7938846720b7f660", "reading_canonical_log_prob": -419.2340087890625, "reading_distances": [20, 13, 22, 17], "reading_log_probs": [-435.8888854980469, -432.77587890625, -437.81787109375, -435.06195068359375], "reading_logprob_std": 2.0884308972976577, "tokenization_path_logprob_std": 315.0597538608927, "trajectory_balance_rmse": 6018.966703989607} {"canonical_length": 196, "canonical_nll": 1.4962759212571748, "generation_rouge_l_f1": 0.29310344827586204, "generation_token_f1": 0.5431034482758621, "prompt_id": "2a7c37e2ef87098f", "reading_canonical_log_prob": -293.27008056640625, "reading_distances": [27, 30, 28, 31], "reading_log_probs": [-304.1692810058594, -302.710205078125, -301.85552978515625, -305.9185485839844], "reading_logprob_std": 1.7812529298621615, "tokenization_path_logprob_std": 211.1605143974339, "trajectory_balance_rmse": 4663.353800867616} {"canonical_length": 284, "canonical_nll": 1.2908374625192562, "generation_rouge_l_f1": 0.2340036563071298, "generation_token_f1": 0.45338208409506403, "prompt_id": "f9f1a4c90d1c6936", "reading_canonical_log_prob": -366.59783935546875, "reading_distances": [27, 25, 28, 28], "reading_log_probs": [-377.5123291015625, -377.4571533203125, -382.6075134277344, -377.55767822265625], "reading_logprob_std": 2.5495613073273407, "tokenization_path_logprob_std": 277.819351910311, "trajectory_balance_rmse": 6488.997189180762} {"canonical_length": 149, "canonical_nll": 1.3334452993917785, "generation_rouge_l_f1": 0.22748815165876776, "generation_token_f1": 0.4170616113744076, "prompt_id": "4c7fd497c60d26bd", "reading_canonical_log_prob": -198.683349609375, "reading_distances": [20, 22, 27, 23], "reading_log_probs": [-217.7579803466797, -214.26889038085938, -219.86676025390625, -222.1246795654297], "reading_logprob_std": 3.3396174181966627, "tokenization_path_logprob_std": 155.23869082699355, "trajectory_balance_rmse": 3105.7145056818563} {"canonical_length": 25, "canonical_nll": 1.1624423217773439, "generation_rouge_l_f1": 0.36363636363636365, "generation_token_f1": 0.5194805194805194, "prompt_id": "c2e4f5be5a2ad0b2", "reading_canonical_log_prob": -29.061058044433594, "reading_distances": [13, 17, 20, 16], "reading_log_probs": [-30.69232177734375, -34.34397506713867, -35.568565368652344, -33.54328155517578], "reading_logprob_std": 2.0712878642134553, "tokenization_path_logprob_std": 105.67301773532088, "trajectory_balance_rmse": 496.4865662528749} {"canonical_length": 14, "canonical_nll": 1.6785718372889928, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.22222222222222224, "prompt_id": "619e2fb6adbffc57", "reading_canonical_log_prob": -23.5000057220459, "reading_distances": [36, 30, 30, 30], "reading_log_probs": [-32.11142349243164, -27.97895050048828, -35.15129089355469, -29.007062911987305], "reading_logprob_std": 3.2430207213997067, "tokenization_path_logprob_std": 63.737878489233296, "trajectory_balance_rmse": 367.213831137916} {"canonical_length": 136, "canonical_nll": 1.444884692921358, "generation_rouge_l_f1": 0.19597989949748745, "generation_token_f1": 0.42211055276381915, "prompt_id": "237ba288f6603bf1", "reading_canonical_log_prob": -196.5043182373047, "reading_distances": [18, 17, 18, 21], "reading_log_probs": [-198.91807556152344, -194.83901977539062, -195.85227966308594, -196.4698486328125], "reading_logprob_std": 1.7344380811064466, "tokenization_path_logprob_std": 215.60616746496612, "trajectory_balance_rmse": 2853.0923973247072} {"canonical_length": 9, "canonical_nll": 0.7298073238796658, "generation_rouge_l_f1": 0.7142857142857143, "generation_token_f1": 0.7142857142857143, "prompt_id": "52afc3f6a37f4218", "reading_canonical_log_prob": -6.568265914916992, "reading_distances": [27, 26, 24, 22], "reading_log_probs": [-6.9712066650390625, -7.048493385314941, -6.648263454437256, -6.411247730255127], "reading_logprob_std": 0.29527699411198527, "tokenization_path_logprob_std": 26.30645772873089, "trajectory_balance_rmse": 208.13145530139087} {"canonical_length": 115, "canonical_nll": 1.6382422405740489, "generation_rouge_l_f1": 0.2124352331606217, "generation_token_f1": 0.33678756476683935, "prompt_id": "0096b48ae77be6ff", "reading_canonical_log_prob": -188.39785766601562, "reading_distances": [35, 31, 30, 25], "reading_log_probs": [-206.78240966796875, -202.76815795898438, -205.95448303222656, -213.82177734375], "reading_logprob_std": 4.659952300648729, "tokenization_path_logprob_std": 133.70009388089585, "trajectory_balance_rmse": 2927.8016659837945} {"canonical_length": 252, "canonical_nll": 1.0745876251705109, "generation_rouge_l_f1": 0.32193158953722334, "generation_token_f1": 0.5392354124748491, "prompt_id": "2061c66b7116786e", "reading_canonical_log_prob": -270.79608154296875, "reading_distances": [31, 38, 33, 29], "reading_log_probs": [-274.17425537109375, -277.5560607910156, -282.865234375, -275.3291931152344], "reading_logprob_std": 3.854021566240901, "tokenization_path_logprob_std": 183.67930930731222, "trajectory_balance_rmse": 4845.009358418934} {"canonical_length": 94, "canonical_nll": 1.463074379778923, "generation_rouge_l_f1": 0.2376237623762376, "generation_token_f1": 0.4158415841584158, "prompt_id": "5622ae2789dbdaf4", "reading_canonical_log_prob": -137.52899169921875, "reading_distances": [18, 18, 16, 17], "reading_log_probs": [-139.47918701171875, -137.99563598632812, -140.32147216796875, -136.25228881835938], "reading_logprob_std": 1.7872293805476802, "tokenization_path_logprob_std": 125.32891302529224, "trajectory_balance_rmse": 1983.5029013430592} {"canonical_length": 279, "canonical_nll": 1.2267677006328406, "generation_rouge_l_f1": 0.2734082397003746, "generation_token_f1": 0.4981273408239701, "prompt_id": "531c2e72ca7ad121", "reading_canonical_log_prob": -342.2681884765625, "reading_distances": [27, 16, 22, 25], "reading_log_probs": [-347.0062255859375, -345.9197692871094, -344.7030029296875, -346.58355712890625], "reading_logprob_std": 1.005049305663082, "tokenization_path_logprob_std": 207.78444983338775, "trajectory_balance_rmse": 5823.343242079168} {"canonical_length": 33, "canonical_nll": 0.7458625562263258, "generation_rouge_l_f1": 0.6557377049180327, "generation_token_f1": 0.7868852459016394, "prompt_id": "50acd3d956ee080e", "reading_canonical_log_prob": -24.61346435546875, "reading_distances": [26, 23, 37, 36], "reading_log_probs": [-22.774341583251953, -24.90546417236328, -21.449359893798828, -22.785728454589844], "reading_logprob_std": 1.4294872873351052, "tokenization_path_logprob_std": 59.704915743231936, "trajectory_balance_rmse": 675.304732076572} {"canonical_length": 26, "canonical_nll": 1.3701597360464244, "generation_rouge_l_f1": 0.27586206896551724, "generation_token_f1": 0.31034482758620685, "prompt_id": "3fcc8e674911ba73", "reading_canonical_log_prob": -35.62415313720703, "reading_distances": [16, 15, 15, 15], "reading_log_probs": [-36.22119903564453, -36.13423156738281, -37.36806106567383, -34.083892822265625], "reading_logprob_std": 1.3663491183258996, "tokenization_path_logprob_std": 41.01854271142664, "trajectory_balance_rmse": 496.36864545146307} {"canonical_length": 347, "canonical_nll": 0.9631698564424982, "generation_rouge_l_f1": 0.19354838709677422, "generation_token_f1": 0.4161290322580645, "prompt_id": "915d77693c098234", "reading_canonical_log_prob": -334.2199401855469, "reading_distances": [17, 13, 11, 17], "reading_log_probs": [-340.56732177734375, -340.8631591796875, -340.18865966796875, -337.990478515625], "reading_logprob_std": 1.304168640829406, "tokenization_path_logprob_std": 225.7371750316421, "trajectory_balance_rmse": 5973.995820715071} {"canonical_length": 16, "canonical_nll": 0.18422217667102814, "generation_rouge_l_f1": 0.9333333333333333, "generation_token_f1": 0.9333333333333333, "prompt_id": "d4ba094ba07a6bba", "reading_canonical_log_prob": -2.94755482673645, "reading_distances": [21, 21, 28, 21], "reading_log_probs": [-2.192138671875, -1.6134285926818848, -1.9873449802398682, -1.6639785766601562], "reading_logprob_std": 0.27427685143532327, "tokenization_path_logprob_std": 41.99690457532884, "trajectory_balance_rmse": 251.11902424578767} {"canonical_length": 35, "canonical_nll": 0.8547585623604911, "generation_rouge_l_f1": 0.14765100671140938, "generation_token_f1": 0.20805369127516776, "prompt_id": "40f13ac2314179d4", "reading_canonical_log_prob": -29.916549682617188, "reading_distances": [20, 25, 28, 30], "reading_log_probs": [-30.164011001586914, -31.576152801513672, -31.11124610900879, -28.59696388244629], "reading_logprob_std": 1.3153059038514463, "tokenization_path_logprob_std": 93.99020490617562, "trajectory_balance_rmse": 767.2853649859229} {"canonical_length": 228, "canonical_nll": 1.1480026245117188, "generation_rouge_l_f1": 0.2063157894736842, "generation_token_f1": 0.37894736842105264, "prompt_id": "c60f4d0722caa17f", "reading_canonical_log_prob": -261.7445983886719, "reading_distances": [62, 53, 67, 59], "reading_log_probs": [-257.4425964355469, -259.5035095214844, -253.00753784179688, -256.6296691894531], "reading_logprob_std": 2.7103874262149827, "tokenization_path_logprob_std": 222.96401596777807, "trajectory_balance_rmse": 4871.850454854969} {"canonical_length": 20, "canonical_nll": 1.856924819946289, "generation_rouge_l_f1": 0.45614035087719296, "generation_token_f1": 0.5263157894736843, "prompt_id": "2ca8b957d3b2b5bc", "reading_canonical_log_prob": -37.13849639892578, "reading_distances": [24, 23, 30, 30], "reading_log_probs": [-45.36573791503906, -42.21501159667969, -42.221397399902344, -43.55662536621094], "reading_logprob_std": 1.4907951969153714, "tokenization_path_logprob_std": 68.97314752561189, "trajectory_balance_rmse": 474.0506511026531} {"canonical_length": 159, "canonical_nll": 1.2718592229879127, "generation_rouge_l_f1": 0.2795180722891566, "generation_token_f1": 0.5397590361445783, "prompt_id": "762bfb8c99f50dd6", "reading_canonical_log_prob": -202.22561645507812, "reading_distances": [22, 20, 31, 20], "reading_log_probs": [-224.27301025390625, -224.7239990234375, -221.96066284179688, -226.97366333007812], "reading_logprob_std": 2.054893726027657, "tokenization_path_logprob_std": 88.18625027462478, "trajectory_balance_rmse": 2904.9905430122676} {"canonical_length": 231, "canonical_nll": 1.1060879529812635, "generation_rouge_l_f1": 0.26640159045725653, "generation_token_f1": 0.5208747514910537, "prompt_id": "9f2ef447478c22f4", "reading_canonical_log_prob": -255.50631713867188, "reading_distances": [29, 27, 19, 24], "reading_log_probs": [-258.28033447265625, -256.23236083984375, -256.6250305175781, -258.201904296875], "reading_logprob_std": 1.0590955377760711, "tokenization_path_logprob_std": 242.32761282447237, "trajectory_balance_rmse": 4947.2207670783} {"canonical_length": 361, "canonical_nll": 1.2217618567131232, "generation_rouge_l_f1": 0.26870748299319724, "generation_token_f1": 0.46938775510204084, "prompt_id": "dc0376ea25a8c1bc", "reading_canonical_log_prob": -441.0560302734375, "reading_distances": [27, 26, 30, 30], "reading_log_probs": [-441.4711608886719, -449.47027587890625, -443.85906982421875, -449.50640869140625], "reading_logprob_std": 4.058248105319502, "tokenization_path_logprob_std": 177.26023119872497, "trajectory_balance_rmse": 6356.144663261003} {"canonical_length": 18, "canonical_nll": 3.1914776696099176, "generation_rouge_l_f1": 0.16666666666666666, "generation_token_f1": 0.25, "prompt_id": "9cf805580809e94d", "reading_canonical_log_prob": -57.446598052978516, "reading_distances": [31, 22, 23, 20], "reading_log_probs": [-66.4383316040039, -69.1578369140625, -65.52198028564453, -63.044036865234375], "reading_logprob_std": 2.52479130652334, "tokenization_path_logprob_std": 42.112332046540466, "trajectory_balance_rmse": 406.7745726026968} {"canonical_length": 280, "canonical_nll": 1.6038649422781808, "generation_rouge_l_f1": 0.25136612021857924, "generation_token_f1": 0.49544626593806923, "prompt_id": "d3ad116b6d046506", "reading_canonical_log_prob": -449.0821838378906, "reading_distances": [24, 22, 22, 24], "reading_log_probs": [-451.6078796386719, -450.74249267578125, -452.28106689453125, -451.01953125], "reading_logprob_std": 0.6821328489912806, "tokenization_path_logprob_std": 144.60130208022719, "trajectory_balance_rmse": 6988.715636500745} {"canonical_length": 141, "canonical_nll": 1.3806570066627881, "generation_rouge_l_f1": 0.31521739130434784, "generation_token_f1": 0.5054347826086957, "prompt_id": "d4c38137aba16167", "reading_canonical_log_prob": -194.67263793945312, "reading_distances": [12, 14, 9, 16], "reading_log_probs": [-212.72943115234375, -230.49798583984375, -208.43763732910156, -214.19049072265625], "reading_logprob_std": 9.669386487570504, "tokenization_path_logprob_std": 22.928518015998208, "trajectory_balance_rmse": 1344.6668965534716} {"canonical_length": 41, "canonical_nll": 1.492904104837557, "generation_rouge_l_f1": 0.23333333333333334, "generation_token_f1": 0.26666666666666666, "prompt_id": "84547b634d24f63a", "reading_canonical_log_prob": -61.209068298339844, "reading_distances": [24, 28, 29, 24], "reading_log_probs": [-66.34956359863281, -61.8855094909668, -65.79124450683594, -64.57820129394531], "reading_logprob_std": 1.9864971829675584, "tokenization_path_logprob_std": 146.6855796618001, "trajectory_balance_rmse": 977.4235555886478} {"canonical_length": 6, "canonical_nll": 0.28261741002400714, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "58c9afa198f60472", "reading_canonical_log_prob": -1.695704460144043, "reading_distances": [43, 40, 52, 40], "reading_log_probs": [-2.9544925689697266, -4.636956691741943, -4.03112268447876, -4.326112270355225], "reading_logprob_std": 0.731541550978987, "tokenization_path_logprob_std": 13.700173307291184, "trajectory_balance_rmse": 61.574946722902844} {"canonical_length": 238, "canonical_nll": 1.566974030823267, "generation_rouge_l_f1": 0.20491803278688522, "generation_token_f1": 0.38524590163934425, "prompt_id": "74608c6aad72d41d", "reading_canonical_log_prob": -372.9398193359375, "reading_distances": [37, 39, 36, 32], "reading_log_probs": [-387.5786437988281, -390.0888671875, -384.40875244140625, -385.98321533203125], "reading_logprob_std": 2.4237340123698243, "tokenization_path_logprob_std": 197.2544197274615, "trajectory_balance_rmse": 5502.125293812229} {"canonical_length": 296, "canonical_nll": 1.5081440693623311, "generation_rouge_l_f1": 0.23290203327171904, "generation_token_f1": 0.4658040665434381, "prompt_id": "5c242c11a7bd0e44", "reading_canonical_log_prob": -446.41064453125, "reading_distances": [26, 35, 28, 36], "reading_log_probs": [-447.58624267578125, -446.2974853515625, -448.89532470703125, -447.14385986328125], "reading_logprob_std": 1.0840910934920769, "tokenization_path_logprob_std": 282.42208825275253, "trajectory_balance_rmse": 7235.993046339855} {"canonical_length": 42, "canonical_nll": 0.6803854533604213, "generation_rouge_l_f1": 0.7391304347826086, "generation_token_f1": 0.7391304347826086, "prompt_id": "c3513f19426dd701", "reading_canonical_log_prob": -28.576189041137695, "reading_distances": [23, 27, 26, 33], "reading_log_probs": [-27.61651611328125, -27.425813674926758, -25.63974380493164, -30.289443969726562], "reading_logprob_std": 1.9170001559099885, "tokenization_path_logprob_std": 28.589321865529545, "trajectory_balance_rmse": 330.5497928656245} {"canonical_length": 356, "canonical_nll": 1.0775454231862271, "generation_rouge_l_f1": 0.23723228995057663, "generation_token_f1": 0.4316309719934102, "prompt_id": "80a15e1bafbe7e14", "reading_canonical_log_prob": -383.6061706542969, "reading_distances": [20, 20, 17, 16], "reading_log_probs": [-386.8230895996094, -383.41168212890625, -382.0411376953125, -385.4070739746094], "reading_logprob_std": 2.115407228231757, "tokenization_path_logprob_std": 375.0872726835657, "trajectory_balance_rmse": 7345.277647668734} {"canonical_length": 51, "canonical_nll": 1.455877565870098, "generation_rouge_l_f1": 0.42372881355932207, "generation_token_f1": 0.5932203389830508, "prompt_id": "74f44ff0ad2c7451", "reading_canonical_log_prob": -74.249755859375, "reading_distances": [23, 25, 25, 22], "reading_log_probs": [-79.43572235107422, -75.97425842285156, -75.63771057128906, -79.24729919433594], "reading_logprob_std": 2.0473014691618747, "tokenization_path_logprob_std": 57.27556666445529, "trajectory_balance_rmse": 1262.5322291314774} {"canonical_length": 21, "canonical_nll": 1.522920426868257, "generation_rouge_l_f1": 0.42857142857142855, "generation_token_f1": 0.42857142857142855, "prompt_id": "0b30260636e7066b", "reading_canonical_log_prob": -31.9813289642334, "reading_distances": [32, 34, 23, 31], "reading_log_probs": [-35.43981170654297, -34.33555603027344, -33.90818786621094, -30.824800491333008], "reading_logprob_std": 1.9765058902174715, "tokenization_path_logprob_std": 94.29886198289628, "trajectory_balance_rmse": 545.8459541378468} {"canonical_length": 142, "canonical_nll": 1.260041465221996, "generation_rouge_l_f1": 0.3206521739130435, "generation_token_f1": 0.4782608695652174, "prompt_id": "270a3b8ee410411a", "reading_canonical_log_prob": -178.92588806152344, "reading_distances": [24, 31, 31, 29], "reading_log_probs": [-182.241455078125, -181.06454467773438, -181.67462158203125, -183.90481567382812], "reading_logprob_std": 1.2208701002998117, "tokenization_path_logprob_std": 139.13368040157542, "trajectory_balance_rmse": 2452.5985852785866} {"canonical_length": 2, "canonical_nll": 2.0206985473632812, "generation_rouge_l_f1": 0.09523809523809523, "generation_token_f1": 0.09523809523809523, "prompt_id": "aa51f5831004ecdf", "reading_canonical_log_prob": -4.0413970947265625, "reading_distances": [35, 33, 39, 28], "reading_log_probs": [-2.7577199935913086, -2.635730504989624, -2.442366600036621, -3.028972625732422], "reading_logprob_std": 0.24563570303175083, "tokenization_path_logprob_std": 25.507922400862387, "trajectory_balance_rmse": 44.448525168120014} {"canonical_length": 53, "canonical_nll": 0.8317480986973025, "generation_rouge_l_f1": 0.22754491017964076, "generation_token_f1": 0.2994011976047904, "prompt_id": "d23e5c49d3b87636", "reading_canonical_log_prob": -44.08264923095703, "reading_distances": [20, 16, 18, 13], "reading_log_probs": [-39.881675720214844, -39.68144607543945, -40.696563720703125, -41.36808776855469], "reading_logprob_std": 0.7767305261457451, "tokenization_path_logprob_std": 135.5072456426632, "trajectory_balance_rmse": 883.2246899117914} {"canonical_length": 78, "canonical_nll": 1.648043705866887, "generation_rouge_l_f1": 0.18627450980392157, "generation_token_f1": 0.29411764705882354, "prompt_id": "1cf9ae29e6e4ee3d", "reading_canonical_log_prob": -128.5474090576172, "reading_distances": [19, 22, 27, 21], "reading_log_probs": [-135.13223266601562, -140.9285430908203, -134.74647521972656, -141.25997924804688], "reading_logprob_std": 3.5595979330271543, "tokenization_path_logprob_std": 74.60312336435912, "trajectory_balance_rmse": 1716.7110534983306} {"canonical_length": 113, "canonical_nll": 1.3522650794645326, "generation_rouge_l_f1": 0.25263157894736843, "generation_token_f1": 0.39473684210526316, "prompt_id": "4b26a788bce312c3", "reading_canonical_log_prob": -152.8059539794922, "reading_distances": [27, 37, 31, 34], "reading_log_probs": [-167.42271423339844, -173.77603149414062, -171.5370635986328, -167.88430786132812], "reading_logprob_std": 3.0355338447457463, "tokenization_path_logprob_std": 256.9828005432629, "trajectory_balance_rmse": 2383.4541577335463} {"canonical_length": 320, "canonical_nll": 1.5254377365112304, "generation_rouge_l_f1": 0.2747826086956522, "generation_token_f1": 0.5217391304347826, "prompt_id": "c59752dc85c30f5b", "reading_canonical_log_prob": -488.14007568359375, "reading_distances": [35, 38, 41, 35], "reading_log_probs": [-493.74945068359375, -491.06951904296875, -494.5927734375, -494.52801513671875], "reading_logprob_std": 1.6552467054273385, "tokenization_path_logprob_std": 337.2181815196017, "trajectory_balance_rmse": 8135.746287011251} {"canonical_length": 9, "canonical_nll": 1.7420507007175021, "generation_rouge_l_f1": 0.5454545454545454, "generation_token_f1": 0.5454545454545454, "prompt_id": "b4e868a234198084", "reading_canonical_log_prob": -15.67845630645752, "reading_distances": [25, 32, 30, 31], "reading_log_probs": [-14.703886032104492, -15.436767578125, -15.079291343688965, -17.495201110839844], "reading_logprob_std": 1.2473652982401573, "tokenization_path_logprob_std": 33.58975684704447, "trajectory_balance_rmse": 247.81142597901436} {"canonical_length": 68, "canonical_nll": 1.7076667336856617, "generation_rouge_l_f1": 0.23214285714285715, "generation_token_f1": 0.3660714285714286, "prompt_id": "466e7627e6013e86", "reading_canonical_log_prob": -116.121337890625, "reading_distances": [33, 36, 37, 37], "reading_log_probs": [-112.8511962890625, -114.12803649902344, -114.45323181152344, -111.87359619140625], "reading_logprob_std": 1.1900740072363751, "tokenization_path_logprob_std": 96.52013366405907, "trajectory_balance_rmse": 1618.2871377394356} {"canonical_length": 14, "canonical_nll": 0.9596988133021763, "generation_rouge_l_f1": 0.4827586206896552, "generation_token_f1": 0.4827586206896552, "prompt_id": "ca45cac53e43d02f", "reading_canonical_log_prob": -13.435783386230469, "reading_distances": [24, 24, 23, 29], "reading_log_probs": [-15.832087516784668, -14.18946647644043, -15.689343452453613, -15.773165702819824], "reading_logprob_std": 0.789874034404107, "tokenization_path_logprob_std": 36.2137540491147, "trajectory_balance_rmse": 178.52545880930202} {"canonical_length": 20, "canonical_nll": 2.200551223754883, "generation_rouge_l_f1": 0.10309278350515463, "generation_token_f1": 0.11340206185567009, "prompt_id": "685a9c22a80a27a9", "reading_canonical_log_prob": -44.011024475097656, "reading_distances": [24, 21, 20, 16], "reading_log_probs": [-46.7094612121582, -44.565757751464844, -47.7338752746582, -51.06361770629883], "reading_logprob_std": 2.707241904730892, "tokenization_path_logprob_std": 42.28608524073632, "trajectory_balance_rmse": 403.4796587144569} {"canonical_length": 12, "canonical_nll": 1.092390775680542, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.24242424242424246, "prompt_id": "426d94bf331d4f8d", "reading_canonical_log_prob": -13.108689308166504, "reading_distances": [23, 26, 23, 31], "reading_log_probs": [-13.418874740600586, -11.714008331298828, -12.915202140808105, -9.194031715393066], "reading_logprob_std": 1.8852456116885365, "tokenization_path_logprob_std": 48.405934926332975, "trajectory_balance_rmse": 154.4719693866629} {"canonical_length": 123, "canonical_nll": 1.2456552148834477, "generation_rouge_l_f1": 0.1763085399449036, "generation_token_f1": 0.28099173553719003, "prompt_id": "efba508a45047679", "reading_canonical_log_prob": -153.21559143066406, "reading_distances": [29, 34, 36, 30], "reading_log_probs": [-159.383056640625, -156.40768432617188, -159.44580078125, -160.56741333007812], "reading_logprob_std": 1.7807065473872388, "tokenization_path_logprob_std": 108.6294457274989, "trajectory_balance_rmse": 2529.978118733691} {"canonical_length": 138, "canonical_nll": 1.228504346764606, "generation_rouge_l_f1": 0.233502538071066, "generation_token_f1": 0.4416243654822335, "prompt_id": "79eb49a82dc5be50", "reading_canonical_log_prob": -169.53359985351562, "reading_distances": [20, 28, 20, 15], "reading_log_probs": [-174.39503479003906, -170.68133544921875, -177.81802368164062, -172.98953247070312], "reading_logprob_std": 2.9868918112671143, "tokenization_path_logprob_std": 214.59913199023953, "trajectory_balance_rmse": 2908.452790464147} {"canonical_length": 355, "canonical_nll": 1.1600440656635123, "generation_rouge_l_f1": 0.2469135802469136, "generation_token_f1": 0.4722222222222222, "prompt_id": "144d91c0bdbf405c", "reading_canonical_log_prob": -411.8156433105469, "reading_distances": [30, 28, 30, 35], "reading_log_probs": [-422.4185485839844, -419.675048828125, -419.50390625, -422.6962585449219], "reading_logprob_std": 1.718700206010013, "tokenization_path_logprob_std": 263.352467444773, "trajectory_balance_rmse": 6025.720822154606} {"canonical_length": 194, "canonical_nll": 1.1434537749929525, "generation_rouge_l_f1": 0.24255319148936172, "generation_token_f1": 0.4127659574468085, "prompt_id": "6bebc48cdc1b16ab", "reading_canonical_log_prob": -221.8300323486328, "reading_distances": [58, 61, 53, 54], "reading_log_probs": [-221.71873474121094, -221.82916259765625, -224.1741485595703, -223.68038940429688], "reading_logprob_std": 1.2602624826694926, "tokenization_path_logprob_std": 157.7770332606063, "trajectory_balance_rmse": 3472.313153906613} {"canonical_length": 136, "canonical_nll": 2.7720673504997704, "generation_rouge_l_f1": 0.234375, "generation_token_f1": 0.4453125, "prompt_id": "f19548b23e650d6d", "reading_canonical_log_prob": -377.00115966796875, "reading_distances": [20, 12, 22, 19], "reading_log_probs": [-406.369873046875, -382.933837890625, -401.7843017578125, -395.2781982421875], "reading_logprob_std": 10.179035077543768, "tokenization_path_logprob_std": 160.11788306884372, "trajectory_balance_rmse": 2871.5303302353204} {"canonical_length": 283, "canonical_nll": 1.1575533054742713, "generation_rouge_l_f1": 0.19607843137254902, "generation_token_f1": 0.46345811051693403, "prompt_id": "b85ad6ca4cfd648f", "reading_canonical_log_prob": -327.58758544921875, "reading_distances": [25, 23, 24, 23], "reading_log_probs": [-319.50225830078125, -319.41796875, -315.3514099121094, -320.834228515625], "reading_logprob_std": 2.373723939161019, "tokenization_path_logprob_std": 335.63843466022416, "trajectory_balance_rmse": 5314.86373469354} {"canonical_length": 146, "canonical_nll": 1.450117241846372, "generation_rouge_l_f1": 0.2430379746835443, "generation_token_f1": 0.4658227848101266, "prompt_id": "6b7b364a0fc8e714", "reading_canonical_log_prob": -211.7171173095703, "reading_distances": [33, 22, 34, 27], "reading_log_probs": [-214.25772094726562, -231.58840942382812, -211.44107055664062, -216.50238037109375], "reading_logprob_std": 9.002060743105593, "tokenization_path_logprob_std": 82.41095442887972, "trajectory_balance_rmse": 3180.067444810603} {"canonical_length": 112, "canonical_nll": 2.6390348161969865, "generation_rouge_l_f1": 0.14395886889460155, "generation_token_f1": 0.2570694087403599, "prompt_id": "1d91ffc1f9553c07", "reading_canonical_log_prob": -295.5718994140625, "reading_distances": [12, 16, 19, 16], "reading_log_probs": [-299.0094909667969, -294.6256103515625, -297.2107849121094, -297.4510803222656], "reading_logprob_std": 1.8167403594710387, "tokenization_path_logprob_std": 116.58067101997142, "trajectory_balance_rmse": 2271.764962934459} {"canonical_length": 37, "canonical_nll": 0.9509100011877112, "generation_rouge_l_f1": 0.19889502762430938, "generation_token_f1": 0.2541436464088398, "prompt_id": "1ac7a2e92a59b9ce", "reading_canonical_log_prob": -35.18367004394531, "reading_distances": [13, 20, 20, 13], "reading_log_probs": [-36.98466110229492, -39.26052474975586, -37.71028137207031, -38.35227584838867], "reading_logprob_std": 0.9668151200372656, "tokenization_path_logprob_std": 144.54899743817475, "trajectory_balance_rmse": 940.5744604709579} {"canonical_length": 43, "canonical_nll": 1.9926750715388808, "generation_rouge_l_f1": 0.3209876543209877, "generation_token_f1": 0.4444444444444445, "prompt_id": "66c94013abca2b3a", "reading_canonical_log_prob": -85.68502807617188, "reading_distances": [36, 24, 28, 26], "reading_log_probs": [-89.74918365478516, -92.42298889160156, -89.93660736083984, -89.62535095214844], "reading_logprob_std": 1.332461642255679, "tokenization_path_logprob_std": 143.2954113502702, "trajectory_balance_rmse": 863.129100685976} {"canonical_length": 15, "canonical_nll": 1.0494525909423829, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4333333333333333, "prompt_id": "785090c8f7a0ea91", "reading_canonical_log_prob": -15.741788864135742, "reading_distances": [17, 16, 18, 13], "reading_log_probs": [-15.916172981262207, -14.743136405944824, -16.0114688873291, -15.577088356018066], "reading_logprob_std": 0.5768379382583568, "tokenization_path_logprob_std": 64.68056673636944, "trajectory_balance_rmse": 359.2104886526354} {"canonical_length": 217, "canonical_nll": 1.1622438211045507, "generation_rouge_l_f1": 0.1735159817351598, "generation_token_f1": 0.3013698630136986, "prompt_id": "30ed32c8494397b3", "reading_canonical_log_prob": -252.2069091796875, "reading_distances": [23, 23, 22, 15], "reading_log_probs": [-261.4447937011719, -260.16778564453125, -261.8519592285156, -258.585205078125], "reading_logprob_std": 1.471576960998336, "tokenization_path_logprob_std": 90.78978642233416, "trajectory_balance_rmse": 3464.085287980777} {"canonical_length": 16, "canonical_nll": 1.4244794845581055, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.3555555555555555, "prompt_id": "e8d7fb5246616f5e", "reading_canonical_log_prob": -22.791671752929688, "reading_distances": [79, 78, 81, 78], "reading_log_probs": [-14.032383918762207, -16.28727149963379, -15.554600715637207, -15.671107292175293], "reading_logprob_std": 0.9581716471262423, "tokenization_path_logprob_std": 33.14207026133653, "trajectory_balance_rmse": 299.6448790115227} {"canonical_length": 6, "canonical_nll": 1.647186279296875, "generation_rouge_l_f1": 0.3636363636363636, "generation_token_f1": 0.3636363636363636, "prompt_id": "96f83a49dbeabc97", "reading_canonical_log_prob": -9.88311767578125, "reading_distances": [31, 28, 32, 30], "reading_log_probs": [-10.677366256713867, -9.460308074951172, -8.702747344970703, -9.328128814697266], "reading_logprob_std": 0.8257957646252464, "tokenization_path_logprob_std": 7.810740335019535, "trajectory_balance_rmse": 108.24962651059298} {"canonical_length": 80, "canonical_nll": 1.2439531326293944, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "dea31d77a681f384", "reading_canonical_log_prob": -99.51625061035156, "reading_distances": [13, 18, 15, 14], "reading_log_probs": [-101.32166290283203, -100.25662231445312, -100.60781860351562, -100.5847396850586], "reading_logprob_std": 0.4489314578600352, "tokenization_path_logprob_std": 88.34531390888607, "trajectory_balance_rmse": 1459.62154762257} {"canonical_length": 9, "canonical_nll": 1.4242497550116644, "generation_rouge_l_f1": 0.14285714285714285, "generation_token_f1": 0.14285714285714285, "prompt_id": "428bb97daf254741", "reading_canonical_log_prob": -12.81824779510498, "reading_distances": [21, 22, 25, 22], "reading_log_probs": [-13.787065505981445, -14.75721549987793, -14.205289840698242, -14.871932983398438], "reading_logprob_std": 0.5045819360585526, "tokenization_path_logprob_std": 76.06536309222064, "trajectory_balance_rmse": 273.37556075495456} {"canonical_length": 249, "canonical_nll": 1.3681227595930598, "generation_rouge_l_f1": 0.2345679012345679, "generation_token_f1": 0.46502057613168724, "prompt_id": "ed06fac728bea8f5", "reading_canonical_log_prob": -340.6625671386719, "reading_distances": [21, 18, 19, 17], "reading_log_probs": [-343.103271484375, -346.4327392578125, -346.82415771484375, -344.2516784667969], "reading_logprob_std": 1.7742855993321267, "tokenization_path_logprob_std": 142.68288539285555, "trajectory_balance_rmse": 5546.593877110266} {"canonical_length": 42, "canonical_nll": 2.054227556501116, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.3333333333333333, "prompt_id": "2455341860ece852", "reading_canonical_log_prob": -86.27755737304688, "reading_distances": [18, 16, 24, 16], "reading_log_probs": [-90.54824829101562, -87.37312316894531, -88.19829559326172, -86.79998779296875], "reading_logprob_std": 1.6486816040175267, "tokenization_path_logprob_std": 104.3432964259588, "trajectory_balance_rmse": 893.7503231275508} {"canonical_length": 56, "canonical_nll": 1.6726012911115373, "generation_rouge_l_f1": 0.391304347826087, "generation_token_f1": 0.4565217391304348, "prompt_id": "2b713b1aba655ea9", "reading_canonical_log_prob": -93.6656723022461, "reading_distances": [19, 18, 10, 18], "reading_log_probs": [-98.0113525390625, -102.074951171875, -97.43927001953125, -99.44462585449219], "reading_logprob_std": 2.068090659294695, "tokenization_path_logprob_std": 149.40580933201332, "trajectory_balance_rmse": 1176.212356774365} {"canonical_length": 21, "canonical_nll": 2.4101504371279763, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.2222222222222222, "prompt_id": "12ac8fa81a314c1b", "reading_canonical_log_prob": -50.6131591796875, "reading_distances": [27, 26, 32, 32], "reading_log_probs": [-52.8082275390625, -57.928428649902344, -57.751068115234375, -59.812618255615234], "reading_logprob_std": 2.993621523309537, "tokenization_path_logprob_std": 28.634942187741743, "trajectory_balance_rmse": 437.5005760898367} {"canonical_length": 174, "canonical_nll": 1.2590439018161816, "generation_rouge_l_f1": 0.23580786026200876, "generation_token_f1": 0.4148471615720524, "prompt_id": "46239be323c13eb4", "reading_canonical_log_prob": -219.07363891601562, "reading_distances": [21, 15, 22, 21], "reading_log_probs": [-261.98236083984375, -257.9957275390625, -262.96820068359375, -258.721435546875], "reading_logprob_std": 2.4287499262460766, "tokenization_path_logprob_std": 202.0347138924754, "trajectory_balance_rmse": 3151.5597901484775} {"canonical_length": 298, "canonical_nll": 1.2630390960898175, "generation_rouge_l_f1": 0.16727272727272727, "generation_token_f1": 0.35272727272727267, "prompt_id": "3fde994857b8ea58", "reading_canonical_log_prob": -376.3856506347656, "reading_distances": [39, 36, 35, 35], "reading_log_probs": [-386.9288330078125, -380.5010070800781, -387.72991943359375, -384.16986083984375], "reading_logprob_std": 3.265505218169947, "tokenization_path_logprob_std": 166.1723928262031, "trajectory_balance_rmse": 6799.766251392165} {"canonical_length": 111, "canonical_nll": 1.1386746243313626, "generation_rouge_l_f1": 0.2936288088642659, "generation_token_f1": 0.45983379501385047, "prompt_id": "366758412bdbb007", "reading_canonical_log_prob": -126.39288330078125, "reading_distances": [39, 36, 33, 39], "reading_log_probs": [-134.5443878173828, -133.3060302734375, -135.44989013671875, -134.8837890625], "reading_logprob_std": 0.9071285005554622, "tokenization_path_logprob_std": 101.32189503637686, "trajectory_balance_rmse": 2446.665006589736} {"canonical_length": 17, "canonical_nll": 1.0540203767664291, "generation_rouge_l_f1": 0.17307692307692307, "generation_token_f1": 0.1923076923076923, "prompt_id": "f77e57c7031a2b6c", "reading_canonical_log_prob": -17.918346405029297, "reading_distances": [10, 15, 16, 13], "reading_log_probs": [-18.29625129699707, -17.276473999023438, -17.038604736328125, -16.6370849609375], "reading_logprob_std": 0.7071699346649042, "tokenization_path_logprob_std": 24.705974131618614, "trajectory_balance_rmse": 518.0319093656318} {"canonical_length": 8, "canonical_nll": 1.7676820755004883, "generation_rouge_l_f1": 0.16, "generation_token_f1": 0.16, "prompt_id": "1f37960a14b0dc02", "reading_canonical_log_prob": -14.141456604003906, "reading_distances": [24, 23, 21, 30], "reading_log_probs": [-31.69003677368164, -33.10376739501953, -26.05426025390625, -30.542505264282227], "reading_logprob_std": 3.0479145564870187, "tokenization_path_logprob_std": 52.81061930983032, "trajectory_balance_rmse": 179.36681623052633} {"canonical_length": 28, "canonical_nll": 2.6310653686523438, "generation_rouge_l_f1": 0.13953488372093023, "generation_token_f1": 0.1627906976744186, "prompt_id": "8d8a8351f2dceabb", "reading_canonical_log_prob": -73.66983032226562, "reading_distances": [20, 20, 21, 19], "reading_log_probs": [-75.75115966796875, -73.89763641357422, -77.36091613769531, -71.73916625976562], "reading_logprob_std": 2.421781833627642, "tokenization_path_logprob_std": 93.7080162529253, "trajectory_balance_rmse": 673.2989406866047} {"canonical_length": 2, "canonical_nll": 0.685929000377655, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "7a9b64f9112cd0c5", "reading_canonical_log_prob": -1.37185800075531, "reading_distances": [45, 42, 52, 41], "reading_log_probs": [-1.63774573802948, -1.3755491971969604, -1.9379358291625977, -1.7785377502441406], "reading_logprob_std": 0.23853192532763642, "tokenization_path_logprob_std": 0.0, "trajectory_balance_rmse": 1.3932312726974487} {"canonical_length": 5, "canonical_nll": 1.1178220748901366, "generation_rouge_l_f1": 0.8, "generation_token_f1": 0.8, "prompt_id": "2797e2b7991dcf3e", "reading_canonical_log_prob": -5.589110374450684, "reading_distances": [28, 20, 20, 23], "reading_log_probs": [-8.475947380065918, -7.8756303787231445, -8.436339378356934, -6.221312046051025], "reading_logprob_std": 1.056836458344613, "tokenization_path_logprob_std": 48.50873629575173, "trajectory_balance_rmse": 75.7304900967065} {"canonical_length": 313, "canonical_nll": 1.695204859724441, "generation_rouge_l_f1": 0.2041522491349481, "generation_token_f1": 0.5224913494809689, "prompt_id": "d353ae8508e0bb05", "reading_canonical_log_prob": -530.59912109375, "reading_distances": [42, 41, 40, 45], "reading_log_probs": [-549.2760009765625, -543.599853515625, -545.4051513671875, -546.0079345703125], "reading_logprob_std": 2.3682568663670525, "tokenization_path_logprob_std": 308.0601745096098, "trajectory_balance_rmse": 6806.708940973058} {"canonical_length": 7, "canonical_nll": 0.02801424264907837, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1099a8369d87bee2", "reading_canonical_log_prob": -0.19609969854354858, "reading_distances": [20, 19, 25, 19], "reading_log_probs": [-1.8269903659820557, -2.8332037925720215, -2.2390167713165283, -3.173032522201538], "reading_logprob_std": 0.6010399258420177, "tokenization_path_logprob_std": 26.622939037749777, "trajectory_balance_rmse": 163.7939222958913} {"canonical_length": 264, "canonical_nll": 1.34537529222893, "generation_rouge_l_f1": 0.27099236641221375, "generation_token_f1": 0.5267175572519084, "prompt_id": "d8d494771b6c2d81", "reading_canonical_log_prob": -355.1790771484375, "reading_distances": [25, 25, 31, 25], "reading_log_probs": [-365.9305114746094, -365.30963134765625, -362.9358825683594, -361.5097351074219], "reading_logprob_std": 2.0616407177895986, "tokenization_path_logprob_std": 351.93133452139284, "trajectory_balance_rmse": 6311.361500140354} {"canonical_length": 372, "canonical_nll": 1.217383764123404, "generation_rouge_l_f1": 0.24960998439937598, "generation_token_f1": 0.4149765990639626, "prompt_id": "27da9cdd8cf02e41", "reading_canonical_log_prob": -452.86676025390625, "reading_distances": [33, 34, 34, 31], "reading_log_probs": [-482.0646667480469, -475.8299255371094, -478.80438232421875, -477.40411376953125], "reading_logprob_std": 2.653746099885542, "tokenization_path_logprob_std": 296.1545633012609, "trajectory_balance_rmse": 7110.610670987103} {"canonical_length": 51, "canonical_nll": 1.296866024241728, "generation_rouge_l_f1": 0.23076923076923075, "generation_token_f1": 0.2769230769230769, "prompt_id": "47574188e8957081", "reading_canonical_log_prob": -66.14016723632812, "reading_distances": [33, 28, 28, 33], "reading_log_probs": [-64.57081604003906, -61.798431396484375, -58.40520477294922, -60.54924774169922], "reading_logprob_std": 2.574640063206111, "tokenization_path_logprob_std": 110.18215869493612, "trajectory_balance_rmse": 993.1117875606583} {"canonical_length": 238, "canonical_nll": 2.0841933819426206, "generation_rouge_l_f1": 0.2015810276679842, "generation_token_f1": 0.32411067193675885, "prompt_id": "b7c16abbc7061a65", "reading_canonical_log_prob": -496.03802490234375, "reading_distances": [27, 30, 34, 27], "reading_log_probs": [-508.4760437011719, -519.716064453125, -510.53436279296875, -517.367919921875], "reading_logprob_std": 5.37087157624408, "tokenization_path_logprob_std": 130.1524289407179, "trajectory_balance_rmse": 5467.853713529507} {"canonical_length": 334, "canonical_nll": 1.3116197414740831, "generation_rouge_l_f1": 0.22857142857142856, "generation_token_f1": 0.48739495798319327, "prompt_id": "01fef5f51abd900f", "reading_canonical_log_prob": -438.08099365234375, "reading_distances": [20, 31, 34, 33], "reading_log_probs": [-443.8563232421875, -448.1868896484375, -443.95501708984375, -442.7982177734375], "reading_logprob_std": 2.3834122535512274, "tokenization_path_logprob_std": 379.124560612646, "trajectory_balance_rmse": 7244.182104375077} {"canonical_length": 60, "canonical_nll": 0.9624937057495118, "generation_rouge_l_f1": 0.2830188679245283, "generation_token_f1": 0.3113207547169811, "prompt_id": "1d0839cb7b1b8564", "reading_canonical_log_prob": -57.7496223449707, "reading_distances": [29, 34, 29, 30], "reading_log_probs": [-65.09310913085938, -60.24584197998047, -63.540924072265625, -58.986812591552734], "reading_logprob_std": 2.833942881814482, "tokenization_path_logprob_std": 112.33907819472684, "trajectory_balance_rmse": 1027.6469429086692} {"canonical_length": 198, "canonical_nll": 1.4735579057173296, "generation_rouge_l_f1": 0.35159817351598166, "generation_token_f1": 0.5251141552511416, "prompt_id": "b900859423a57fd5", "reading_canonical_log_prob": -291.76446533203125, "reading_distances": [19, 22, 22, 25], "reading_log_probs": [-295.53936767578125, -303.78790283203125, -297.9642333984375, -297.41094970703125], "reading_logprob_std": 3.562629003160223, "tokenization_path_logprob_std": 274.76037252288666, "trajectory_balance_rmse": 4339.966596763634} {"canonical_length": 48, "canonical_nll": 1.4566109975179036, "generation_rouge_l_f1": 0.1647058823529412, "generation_token_f1": 0.24313725490196078, "prompt_id": "5fa2bbf4741331df", "reading_canonical_log_prob": -69.91732788085938, "reading_distances": [18, 16, 18, 16], "reading_log_probs": [-68.92947387695312, -70.10079193115234, -69.83271789550781, -71.97080993652344], "reading_logprob_std": 1.277296277730286, "tokenization_path_logprob_std": 44.33176425898593, "trajectory_balance_rmse": 1153.1647827241695} {"canonical_length": 501, "canonical_nll": 0.5966182868637725, "generation_rouge_l_f1": 0.2777777777777778, "generation_token_f1": 0.48055555555555557, "prompt_id": "97c9524265f7d241", "reading_canonical_log_prob": -298.90576171875, "reading_distances": [22, 18, 21, 19], "reading_log_probs": [-311.7008056640625, -311.4873046875, -309.6994323730469, -305.7588806152344], "reading_logprob_std": 2.7522237815291355, "tokenization_path_logprob_std": 181.0017807841185, "trajectory_balance_rmse": 6690.257687402177} {"canonical_length": 11, "canonical_nll": 1.3115141608498313, "generation_rouge_l_f1": 0.588235294117647, "generation_token_f1": 0.588235294117647, "prompt_id": "990e2b768a18fc67", "reading_canonical_log_prob": -14.426655769348145, "reading_distances": [25, 26, 23, 31], "reading_log_probs": [-16.812057495117188, -16.554012298583984, -15.344621658325195, -18.30451774597168], "reading_logprob_std": 1.2157053079961597, "tokenization_path_logprob_std": 51.423165908770244, "trajectory_balance_rmse": 212.18783430654264} {"canonical_length": 30, "canonical_nll": 1.3624001820882161, "generation_rouge_l_f1": 0.5116279069767442, "generation_token_f1": 0.6976744186046512, "prompt_id": "4a59564b57b1ad24", "reading_canonical_log_prob": -40.872005462646484, "reading_distances": [31, 30, 32, 32], "reading_log_probs": [-38.857147216796875, -36.89848327636719, -39.348876953125, -37.12933349609375], "reading_logprob_std": 1.226363084193515, "tokenization_path_logprob_std": 27.85655004146212, "trajectory_balance_rmse": 578.712546049616} {"canonical_length": 118, "canonical_nll": 1.6313598115565413, "generation_rouge_l_f1": 0.18384401114206125, "generation_token_f1": 0.30083565459610023, "prompt_id": "fc01869b83f5b227", "reading_canonical_log_prob": -192.50045776367188, "reading_distances": [26, 26, 28, 33], "reading_log_probs": [-214.60606384277344, -209.979248046875, -212.58847045898438, -211.70669555664062], "reading_logprob_std": 1.9247096621605484, "tokenization_path_logprob_std": 218.87370445892705, "trajectory_balance_rmse": 2767.87233902422} {"canonical_length": 10, "canonical_nll": 0.37330658435821534, "generation_rouge_l_f1": 0.3636363636363636, "generation_token_f1": 0.3636363636363636, "prompt_id": "e35706a135002934", "reading_canonical_log_prob": -3.7330658435821533, "reading_distances": [32, 30, 33, 36], "reading_log_probs": [-2.4643397331237793, -2.428572416305542, -2.494415760040283, -2.9529876708984375], "reading_logprob_std": 0.24674473618682768, "tokenization_path_logprob_std": 9.566440570794063, "trajectory_balance_rmse": 196.52276370455368} {"canonical_length": 123, "canonical_nll": 1.0658604071392277, "generation_rouge_l_f1": 0.212987012987013, "generation_token_f1": 0.37922077922077924, "prompt_id": "5df9638e97e9c3b6", "reading_canonical_log_prob": -131.100830078125, "reading_distances": [11, 11, 10, 14], "reading_log_probs": [-132.65533447265625, -136.5913543701172, -136.27333068847656, -133.28192138671875], "reading_logprob_std": 2.0202470095697413, "tokenization_path_logprob_std": 125.15126770290618, "trajectory_balance_rmse": 2301.1743062632654} {"canonical_length": 25, "canonical_nll": 0.5485408782958985, "generation_rouge_l_f1": 0.13076923076923075, "generation_token_f1": 0.14615384615384616, "prompt_id": "6f6b5ba97837e4b7", "reading_canonical_log_prob": -13.713521957397461, "reading_distances": [27, 27, 33, 26], "reading_log_probs": [-12.58556842803955, -10.641364097595215, -11.515645980834961, -12.507814407348633], "reading_logprob_std": 0.9202859141771672, "tokenization_path_logprob_std": 52.21502359791924, "trajectory_balance_rmse": 454.75213666714336} {"canonical_length": 211, "canonical_nll": 0.753381955115151, "generation_rouge_l_f1": 0.47357293868921774, "generation_token_f1": 0.6046511627906976, "prompt_id": "111777e66227bbb0", "reading_canonical_log_prob": -158.96359252929688, "reading_distances": [40, 43, 44, 36], "reading_log_probs": [-172.14968872070312, -174.19058227539062, -176.2589111328125, -168.58041381835938], "reading_logprob_std": 3.2723841213125775, "tokenization_path_logprob_std": 73.39375560446132, "trajectory_balance_rmse": 3649.6161749835383} {"canonical_length": 264, "canonical_nll": 1.1135794899680398, "generation_rouge_l_f1": 0.23300970873786409, "generation_token_f1": 0.4893203883495146, "prompt_id": "7eae8a9735d3d72c", "reading_canonical_log_prob": -293.9849853515625, "reading_distances": [31, 30, 35, 31], "reading_log_probs": [-314.0224609375, -309.6321105957031, -316.0758361816406, -311.64599609375], "reading_logprob_std": 2.8038653171550094, "tokenization_path_logprob_std": 82.18702686630522, "trajectory_balance_rmse": 5989.641989013794} {"canonical_length": 6, "canonical_nll": 0.05202531814575195, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "88814d5eacafd109", "reading_canonical_log_prob": -0.3121519088745117, "reading_distances": [38, 42, 36, 41], "reading_log_probs": [-2.171250581741333, -1.8860113620758057, -0.7582056522369385, -1.1791558265686035], "reading_logprob_std": 0.646213258463418, "tokenization_path_logprob_std": 12.97758685685304, "trajectory_balance_rmse": 101.56385543831252} {"canonical_length": 150, "canonical_nll": 1.3457236735026041, "generation_rouge_l_f1": 0.17682926829268295, "generation_token_f1": 0.3170731707317073, "prompt_id": "c6b7c5355ef042f7", "reading_canonical_log_prob": -201.85855102539062, "reading_distances": [40, 40, 38, 41], "reading_log_probs": [-210.417724609375, -204.32296752929688, -205.7998046875, -205.2477569580078], "reading_logprob_std": 2.7163218691387634, "tokenization_path_logprob_std": 112.58292779331342, "trajectory_balance_rmse": 3110.5696792921017} {"canonical_length": 21, "canonical_nll": 1.590982891264416, "generation_rouge_l_f1": 0.1111111111111111, "generation_token_f1": 0.16666666666666669, "prompt_id": "62bbc871380d57bb", "reading_canonical_log_prob": -33.410640716552734, "reading_distances": [13, 17, 17, 13], "reading_log_probs": [-33.513607025146484, -29.341264724731445, -32.95329666137695, -28.72622299194336], "reading_logprob_std": 2.4483776381545885, "tokenization_path_logprob_std": 31.27454412136362, "trajectory_balance_rmse": 379.0862711063914} {"canonical_length": 28, "canonical_nll": 1.7369469233921595, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.2857142857142857, "prompt_id": "085c601e3edc073f", "reading_canonical_log_prob": -48.63451385498047, "reading_distances": [15, 11, 13, 11], "reading_log_probs": [-53.88550567626953, -55.394561767578125, -57.45441436767578, -54.85753631591797], "reading_logprob_std": 1.5064961495770939, "tokenization_path_logprob_std": 50.066245180111345, "trajectory_balance_rmse": 555.798420997865} {"canonical_length": 314, "canonical_nll": 1.420926622524383, "generation_rouge_l_f1": 0.27303754266211605, "generation_token_f1": 0.5085324232081911, "prompt_id": "575d828d632e226f", "reading_canonical_log_prob": -446.17095947265625, "reading_distances": [37, 32, 28, 34], "reading_log_probs": [-458.3624572753906, -457.24505615234375, -456.80133056640625, -460.24127197265625], "reading_logprob_std": 1.53359721607652, "tokenization_path_logprob_std": 309.67538097187037, "trajectory_balance_rmse": 7230.3333542171695} {"canonical_length": 382, "canonical_nll": 1.0185204950302684, "generation_rouge_l_f1": 0.011799410029498525, "generation_token_f1": 0.011799410029498525, "prompt_id": "acf1157605893744", "reading_canonical_log_prob": -389.0748291015625, "reading_distances": [20, 17, 14, 14], "reading_log_probs": [-397.32879638671875, -400.36541748046875, -396.615966796875, -397.6417236328125], "reading_logprob_std": 1.6420562019270923, "tokenization_path_logprob_std": 237.80600833245077, "trajectory_balance_rmse": 7814.719599765904} {"canonical_length": 19, "canonical_nll": 1.7900053325452303, "generation_rouge_l_f1": 0.3225806451612903, "generation_token_f1": 0.3225806451612903, "prompt_id": "c0621d0076f956e9", "reading_canonical_log_prob": -34.010101318359375, "reading_distances": [34, 34, 32, 34], "reading_log_probs": [-40.129859924316406, -42.783653259277344, -44.28599548339844, -41.884193420410156], "reading_logprob_std": 1.7375385169724198, "tokenization_path_logprob_std": 34.719084046350986, "trajectory_balance_rmse": 316.7469105996817} {"canonical_length": 252, "canonical_nll": 1.1630188472687253, "generation_rouge_l_f1": 0.34285714285714286, "generation_token_f1": 0.5676190476190476, "prompt_id": "71ec6348f5f044b8", "reading_canonical_log_prob": -293.08074951171875, "reading_distances": [19, 20, 17, 19], "reading_log_probs": [-306.96282958984375, -298.4658203125, -297.72674560546875, -310.9787292480469], "reading_logprob_std": 6.4959349853419255, "tokenization_path_logprob_std": 139.2292556403041, "trajectory_balance_rmse": 5645.0941926251035} {"canonical_length": 13, "canonical_nll": 0.700949082007775, "generation_rouge_l_f1": 0.1138211382113821, "generation_token_f1": 0.13008130081300812, "prompt_id": "73a9260fbda0ad11", "reading_canonical_log_prob": -9.112338066101074, "reading_distances": [40, 30, 36, 38], "reading_log_probs": [-7.8369035720825195, -8.169059753417969, -9.171812057495117, -9.691399574279785], "reading_logprob_std": 0.86238360163868, "tokenization_path_logprob_std": 39.136513054337, "trajectory_balance_rmse": 156.47381679025034} {"canonical_length": 142, "canonical_nll": 1.005886453977773, "generation_rouge_l_f1": 0.3163538873994638, "generation_token_f1": 0.4772117962466488, "prompt_id": "0e246b28f0fca81c", "reading_canonical_log_prob": -142.83587646484375, "reading_distances": [10, 7, 10, 8], "reading_log_probs": [-144.2518310546875, -145.09072875976562, -149.16770935058594, -146.40652465820312], "reading_logprob_std": 2.1503787223498847, "tokenization_path_logprob_std": 103.69451177555607, "trajectory_balance_rmse": 2905.5198729359354} {"canonical_length": 38, "canonical_nll": 1.348841014661287, "generation_rouge_l_f1": 0.05957446808510638, "generation_token_f1": 0.07659574468085106, "prompt_id": "7dd4910bb043bf96", "reading_canonical_log_prob": -51.255958557128906, "reading_distances": [14, 9, 11, 12], "reading_log_probs": [-53.98908996582031, -52.56848907470703, -56.69135665893555, -51.6143798828125], "reading_logprob_std": 2.210649504921639, "tokenization_path_logprob_std": 83.9876247318649, "trajectory_balance_rmse": 584.839155725075} {"canonical_length": 187, "canonical_nll": 1.29362495952749, "generation_rouge_l_f1": 0.20785219399538107, "generation_token_f1": 0.4387990762124711, "prompt_id": "b7fe1fa68d367af3", "reading_canonical_log_prob": -241.90786743164062, "reading_distances": [17, 13, 15, 15], "reading_log_probs": [-247.52731323242188, -246.736083984375, -249.19284057617188, -246.10716247558594], "reading_logprob_std": 1.334461744920525, "tokenization_path_logprob_std": 247.07228000103748, "trajectory_balance_rmse": 4004.9401310528497} {"canonical_length": 96, "canonical_nll": 1.4730661710103352, "generation_rouge_l_f1": 0.2184873949579832, "generation_token_f1": 0.3865546218487395, "prompt_id": "b435d26419c9a382", "reading_canonical_log_prob": -141.4143524169922, "reading_distances": [41, 46, 48, 52], "reading_log_probs": [-137.82916259765625, -141.91171264648438, -139.07879638671875, -141.48863220214844], "reading_logprob_std": 1.9500478102837002, "tokenization_path_logprob_std": 176.38536135234614, "trajectory_balance_rmse": 2248.397473180315} {"canonical_length": 113, "canonical_nll": 1.3907860950031112, "generation_rouge_l_f1": 0.17692307692307696, "generation_token_f1": 0.36923076923076925, "prompt_id": "fc2765baa4652639", "reading_canonical_log_prob": -157.15882873535156, "reading_distances": [19, 26, 24, 21], "reading_log_probs": [-170.63412475585938, -167.59814453125, -168.56105041503906, -168.54672241210938], "reading_logprob_std": 1.2812511697633349, "tokenization_path_logprob_std": 216.89556344052474, "trajectory_balance_rmse": 2208.2971696952554} {"canonical_length": 216, "canonical_nll": 1.4324326691804108, "generation_rouge_l_f1": 0.25053078556263264, "generation_token_f1": 0.47983014861995754, "prompt_id": "6389a5bbf9eaba60", "reading_canonical_log_prob": -309.40545654296875, "reading_distances": [34, 29, 31, 26], "reading_log_probs": [-310.48297119140625, -308.1460876464844, -308.6387939453125, -311.1037902832031], "reading_logprob_std": 1.4234465828944958, "tokenization_path_logprob_std": 140.42611594463867, "trajectory_balance_rmse": 4552.051280140208} {"canonical_length": 13, "canonical_nll": 0.1549972937657283, "generation_rouge_l_f1": 0.888888888888889, "generation_token_f1": 0.888888888888889, "prompt_id": "286cf1906a688fcb", "reading_canonical_log_prob": -2.0149648189544678, "reading_distances": [25, 25, 24, 24], "reading_log_probs": [-3.929429531097412, -4.444279670715332, -3.98948073387146, -3.957287073135376], "reading_logprob_std": 0.24401044217942425, "tokenization_path_logprob_std": 18.96035343211193, "trajectory_balance_rmse": 244.87809908248383} {"canonical_length": 93, "canonical_nll": 1.352346686906712, "generation_rouge_l_f1": 0.2515723270440251, "generation_token_f1": 0.4150943396226416, "prompt_id": "33333486aa163f12", "reading_canonical_log_prob": -125.76824188232422, "reading_distances": [34, 26, 32, 28], "reading_log_probs": [-129.20046997070312, -134.7128143310547, -131.842041015625, -137.3461151123047], "reading_logprob_std": 3.525925608973973, "tokenization_path_logprob_std": 168.21785608483205, "trajectory_balance_rmse": 2325.376836748184} {"canonical_length": 60, "canonical_nll": 2.1996388753255207, "generation_rouge_l_f1": 0.21359223300970875, "generation_token_f1": 0.2459546925566343, "prompt_id": "c5556f6cb2c9f8dd", "reading_canonical_log_prob": -131.97833251953125, "reading_distances": [25, 23, 27, 23], "reading_log_probs": [-130.8897247314453, -136.62765502929688, -126.45147705078125, -133.4180908203125], "reading_logprob_std": 4.295386378944843, "tokenization_path_logprob_std": 35.63973890072076, "trajectory_balance_rmse": 375.7148643914324} {"canonical_length": 11, "canonical_nll": 3.220491582697088, "generation_rouge_l_f1": 0.18181818181818185, "generation_token_f1": 0.18181818181818185, "prompt_id": "b1950a5b21116b51", "reading_canonical_log_prob": -35.42540740966797, "reading_distances": [17, 18, 12, 19], "reading_log_probs": [-38.40334701538086, -35.724571228027344, -37.80607604980469, -37.64240264892578], "reading_logprob_std": 1.1600695854952832, "tokenization_path_logprob_std": 37.7846230566521, "trajectory_balance_rmse": 239.3837807975457} {"canonical_length": 263, "canonical_nll": 1.476326481924311, "generation_rouge_l_f1": 0.1954022988505747, "generation_token_f1": 0.4750957854406131, "prompt_id": "355011c0ec753a64", "reading_canonical_log_prob": -388.27386474609375, "reading_distances": [18, 21, 20, 18], "reading_log_probs": [-399.738525390625, -404.5162048339844, -403.5853271484375, -404.704833984375], "reading_logprob_std": 2.3173932095729923, "tokenization_path_logprob_std": 303.9758019271018, "trajectory_balance_rmse": 6253.37004662441} {"canonical_length": 29, "canonical_nll": 0.984411963101091, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "c2e4ff98064c8812", "reading_canonical_log_prob": -28.54794692993164, "reading_distances": [25, 22, 22, 29], "reading_log_probs": [-29.183555603027344, -26.857086181640625, -26.221006393432617, -27.179162979125977], "reading_logprob_std": 1.279099376573895, "tokenization_path_logprob_std": 100.97230815858657, "trajectory_balance_rmse": 606.7444976982719} {"canonical_length": 57, "canonical_nll": 1.9742938928436815, "generation_rouge_l_f1": 0.13953488372093026, "generation_token_f1": 0.2059800664451827, "prompt_id": "e95bb93274990ecf", "reading_canonical_log_prob": -112.53475189208984, "reading_distances": [28, 30, 27, 28], "reading_log_probs": [-114.28002166748047, -113.38314056396484, -112.66490173339844, -113.70406341552734], "reading_logprob_std": 0.6735138899395984, "tokenization_path_logprob_std": 140.52426078231653, "trajectory_balance_rmse": 1032.6566477270933} {"canonical_length": 40, "canonical_nll": 1.9163951873779297, "generation_rouge_l_f1": 0.3376623376623376, "generation_token_f1": 0.3896103896103896, "prompt_id": "0eb7b4a395752869", "reading_canonical_log_prob": -76.65580749511719, "reading_distances": [30, 29, 36, 29], "reading_log_probs": [-94.14556884765625, -80.55047607421875, -88.55122375488281, -93.2432632446289], "reading_logprob_std": 6.218731409439429, "tokenization_path_logprob_std": 95.14638808941602, "trajectory_balance_rmse": 704.2284490158112} {"canonical_length": 124, "canonical_nll": 1.8465953949959046, "generation_rouge_l_f1": 0.3549488054607508, "generation_token_f1": 0.5324232081911262, "prompt_id": "5d63d37b660c138f", "reading_canonical_log_prob": -228.9778289794922, "reading_distances": [38, 28, 37, 39], "reading_log_probs": [-239.10226440429688, -238.79351806640625, -244.29986572265625, -241.45980834960938], "reading_logprob_std": 2.552177036176546, "tokenization_path_logprob_std": 74.06515497617646, "trajectory_balance_rmse": 2951.6488355218885} {"canonical_length": 241, "canonical_nll": 1.213645460200013, "generation_rouge_l_f1": 0.23300970873786409, "generation_token_f1": 0.45825242718446596, "prompt_id": "17f90ddfcc762d15", "reading_canonical_log_prob": -292.4885559082031, "reading_distances": [23, 25, 31, 31], "reading_log_probs": [-292.45849609375, -289.7715148925781, -291.939697265625, -282.84442138671875], "reading_logprob_std": 4.428403918898782, "tokenization_path_logprob_std": 95.17158892754553, "trajectory_balance_rmse": 5266.270045326833} {"canonical_length": 61, "canonical_nll": 1.4294213466956966, "generation_rouge_l_f1": 0.30303030303030304, "generation_token_f1": 0.40404040404040403, "prompt_id": "9e46c646f8e1ae11", "reading_canonical_log_prob": -87.1947021484375, "reading_distances": [21, 19, 23, 26], "reading_log_probs": [-88.69444274902344, -90.74790954589844, -91.7992172241211, -90.54729461669922], "reading_logprob_std": 1.2910659313695387, "tokenization_path_logprob_std": 43.97894735223644, "trajectory_balance_rmse": 1247.91563120905} {"canonical_length": 317, "canonical_nll": 1.0879706253388703, "generation_rouge_l_f1": 0.2517241379310345, "generation_token_f1": 0.5206896551724137, "prompt_id": "c405755d9e3e8c9b", "reading_canonical_log_prob": -344.8866882324219, "reading_distances": [23, 26, 23, 21], "reading_log_probs": [-352.1580505371094, -352.23455810546875, -354.3860168457031, -351.616943359375], "reading_logprob_std": 1.2227181099313635, "tokenization_path_logprob_std": 214.86219375525167, "trajectory_balance_rmse": 6209.270551507829} {"canonical_length": 225, "canonical_nll": 0.5766236029730902, "generation_rouge_l_f1": 0.24778761061946902, "generation_token_f1": 0.4557522123893806, "prompt_id": "d5c04061128f24af", "reading_canonical_log_prob": -129.7403106689453, "reading_distances": [27, 26, 23, 22], "reading_log_probs": [-129.3516387939453, -127.94256591796875, -130.3804931640625, -130.218994140625], "reading_logprob_std": 1.116095486762496, "tokenization_path_logprob_std": 209.7281939578012, "trajectory_balance_rmse": 3203.3437197978924} {"canonical_length": 61, "canonical_nll": 0.9991213063724705, "generation_rouge_l_f1": 0.5137614678899083, "generation_token_f1": 0.5871559633027522, "prompt_id": "1419b413999c869f", "reading_canonical_log_prob": -60.9463996887207, "reading_distances": [17, 13, 12, 12], "reading_log_probs": [-60.0377197265625, -68.23643493652344, -61.2508544921875, -60.07994842529297], "reading_logprob_std": 3.930543194927314, "tokenization_path_logprob_std": 35.61081817747015, "trajectory_balance_rmse": 698.7397829683506} {"canonical_length": 116, "canonical_nll": 1.4660482735469424, "generation_rouge_l_f1": 0.2065217391304348, "generation_token_f1": 0.40217391304347827, "prompt_id": "3501affcfd13e96c", "reading_canonical_log_prob": -170.0615997314453, "reading_distances": [24, 21, 25, 18], "reading_log_probs": [-189.83108520507812, -184.51272583007812, -187.94622802734375, -188.91720581054688], "reading_logprob_std": 2.323861507317578, "tokenization_path_logprob_std": 90.44839257785763, "trajectory_balance_rmse": 2415.159706519565} {"canonical_length": 26, "canonical_nll": 1.9241975637582631, "generation_rouge_l_f1": 0.2033898305084746, "generation_token_f1": 0.3050847457627119, "prompt_id": "6fdd3868e470d9e0", "reading_canonical_log_prob": -50.029136657714844, "reading_distances": [57, 60, 47, 39], "reading_log_probs": [-63.83183288574219, -63.87525177001953, -66.25203704833984, -70.72050476074219], "reading_logprob_std": 3.2376286853781684, "tokenization_path_logprob_std": 61.91984692599404, "trajectory_balance_rmse": 605.1082469269325} {"canonical_length": 304, "canonical_nll": 0.8293057993838662, "generation_rouge_l_f1": 0.368888888888889, "generation_token_f1": 0.5822222222222222, "prompt_id": "1d8b10c0b2c4c1d8", "reading_canonical_log_prob": -252.1089630126953, "reading_distances": [27, 19, 19, 23], "reading_log_probs": [-262.26495361328125, -259.15618896484375, -255.79949951171875, -259.1845397949219], "reading_logprob_std": 2.6407405723319375, "tokenization_path_logprob_std": 257.92269524342, "trajectory_balance_rmse": 4747.887515135564} {"canonical_length": 31, "canonical_nll": 0.8372014568698022, "generation_rouge_l_f1": 0.07692307692307693, "generation_token_f1": 0.07692307692307693, "prompt_id": "6f83d657c17ec17d", "reading_canonical_log_prob": -25.953245162963867, "reading_distances": [16, 19, 23, 24], "reading_log_probs": [-22.601646423339844, -22.261470794677734, -22.379518508911133, -24.757064819335938], "reading_logprob_std": 1.1798848809475209, "tokenization_path_logprob_std": 79.89698117425739, "trajectory_balance_rmse": 502.9907149448188} {"canonical_length": 39, "canonical_nll": 1.5331037472455928, "generation_rouge_l_f1": 0.09756097560975609, "generation_token_f1": 0.09756097560975609, "prompt_id": "6587efe9dc00b7ac", "reading_canonical_log_prob": -59.791046142578125, "reading_distances": [24, 22, 21, 23], "reading_log_probs": [-64.33503723144531, -71.9206314086914, -66.12963104248047, -64.37442779541016], "reading_logprob_std": 3.586141964076034, "tokenization_path_logprob_std": 121.4252212867948, "trajectory_balance_rmse": 946.7784232602986} {"canonical_length": 15, "canonical_nll": 0.816888173421224, "generation_rouge_l_f1": 0.6153846153846153, "generation_token_f1": 0.6153846153846153, "prompt_id": "ea95653fa930225e", "reading_canonical_log_prob": -12.25332260131836, "reading_distances": [15, 23, 15, 16], "reading_log_probs": [-13.351081848144531, -14.162079811096191, -13.852507591247559, -14.507847785949707], "reading_logprob_std": 0.49092719497166915, "tokenization_path_logprob_std": 61.92765766420515, "trajectory_balance_rmse": 444.8580380063919} {"canonical_length": 218, "canonical_nll": 1.4986852243405964, "generation_rouge_l_f1": 0.2911392405063291, "generation_token_f1": 0.48523206751054854, "prompt_id": "dbfaa5f0bb7fb12c", "reading_canonical_log_prob": -326.71337890625, "reading_distances": [38, 42, 40, 45], "reading_log_probs": [-330.7364807128906, -338.5401611328125, -333.2481689453125, -337.128173828125], "reading_logprob_std": 3.5720334323778573, "tokenization_path_logprob_std": 100.25803942513357, "trajectory_balance_rmse": 5151.713164478147} {"canonical_length": 236, "canonical_nll": 1.2549481149447166, "generation_rouge_l_f1": 0.24497991967871488, "generation_token_f1": 0.5220883534136546, "prompt_id": "6ef66ee37f69a6e9", "reading_canonical_log_prob": -296.1677551269531, "reading_distances": [37, 45, 46, 48], "reading_log_probs": [-296.8299560546875, -298.6761474609375, -295.90960693359375, -295.3232421875], "reading_logprob_std": 1.4652984840504488, "tokenization_path_logprob_std": 240.08758838476533, "trajectory_balance_rmse": 5670.627341022409} {"canonical_length": 254, "canonical_nll": 0.3875891317532757, "generation_rouge_l_f1": 0.2426614481409002, "generation_token_f1": 0.4540117416829746, "prompt_id": "e2bfc09d77b30452", "reading_canonical_log_prob": -98.44763946533203, "reading_distances": [21, 26, 24, 23], "reading_log_probs": [-98.7232666015625, -98.63441467285156, -101.86810302734375, -99.3814697265625], "reading_logprob_std": 1.5146313898031347, "tokenization_path_logprob_std": 73.9131636441769, "trajectory_balance_rmse": 3355.031948015234} {"canonical_length": 315, "canonical_nll": 0.8550653366815476, "generation_rouge_l_f1": 0.17317487266553483, "generation_token_f1": 0.3870967741935484, "prompt_id": "3024b6f11a619ed8", "reading_canonical_log_prob": -269.3455810546875, "reading_distances": [16, 13, 15, 12], "reading_log_probs": [-272.31439208984375, -270.87042236328125, -271.85137939453125, -271.9912109375], "reading_logprob_std": 0.6219506927232318, "tokenization_path_logprob_std": 125.0317836600873, "trajectory_balance_rmse": 5850.5273200502625} {"canonical_length": 147, "canonical_nll": 1.550362308009141, "generation_rouge_l_f1": 0.16826923076923075, "generation_token_f1": 0.3365384615384615, "prompt_id": "2027b774c4dd5ef1", "reading_canonical_log_prob": -227.90325927734375, "reading_distances": [79, 81, 68, 82], "reading_log_probs": [-236.03250122070312, -235.77403259277344, -232.49037170410156, -237.36517333984375], "reading_logprob_std": 2.070977048354918, "tokenization_path_logprob_std": 66.82888295636688, "trajectory_balance_rmse": 3203.248528881607} {"canonical_length": 320, "canonical_nll": 1.0204129219055176, "generation_rouge_l_f1": 0.24542124542124544, "generation_token_f1": 0.45054945054945056, "prompt_id": "ea4a6ac8af4f585f", "reading_canonical_log_prob": -326.5321350097656, "reading_distances": [29, 33, 37, 34], "reading_log_probs": [-326.2271728515625, -332.7806091308594, -334.17315673828125, -334.599609375], "reading_logprob_std": 3.8903051820334213, "tokenization_path_logprob_std": 163.9351035262409, "trajectory_balance_rmse": 6745.5742439954065} {"canonical_length": 142, "canonical_nll": 1.207552305409606, "generation_rouge_l_f1": 0.26815642458100564, "generation_token_f1": 0.4357541899441341, "prompt_id": "e9a730206e7c88d2", "reading_canonical_log_prob": -171.47242736816406, "reading_distances": [33, 22, 21, 22], "reading_log_probs": [-179.41937255859375, -178.35572814941406, -176.79258728027344, -179.626220703125], "reading_logprob_std": 1.296185607959937, "tokenization_path_logprob_std": 115.32962309394476, "trajectory_balance_rmse": 2728.72290633459} {"canonical_length": 324, "canonical_nll": 1.049944559733073, "generation_rouge_l_f1": 0.1966604823747681, "generation_token_f1": 0.40074211502782936, "prompt_id": "21922a55d95c3a47", "reading_canonical_log_prob": -340.1820373535156, "reading_distances": [18, 19, 18, 19], "reading_log_probs": [-347.521728515625, -346.51019287109375, -345.6522216796875, -343.88031005859375], "reading_logprob_std": 1.543001877151461, "tokenization_path_logprob_std": 124.48435826096171, "trajectory_balance_rmse": 6813.137569826408} {"canonical_length": 17, "canonical_nll": 0.5540488187004539, "generation_rouge_l_f1": 0.8666666666666666, "generation_token_f1": 0.8666666666666666, "prompt_id": "732a8e545760e4c1", "reading_canonical_log_prob": -9.418829917907715, "reading_distances": [24, 22, 16, 17], "reading_log_probs": [-15.945259094238281, -12.19761848449707, -11.151209831237793, -11.032313346862793], "reading_logprob_std": 2.30274831994151, "tokenization_path_logprob_std": 40.92405376646405, "trajectory_balance_rmse": 291.6607620270277} {"canonical_length": 8, "canonical_nll": 0.588773250579834, "generation_rouge_l_f1": 0.875, "generation_token_f1": 0.875, "prompt_id": "530dfa24f61035c2", "reading_canonical_log_prob": -4.710186004638672, "reading_distances": [34, 32, 26, 35], "reading_log_probs": [-4.233458995819092, -5.0284271240234375, -3.920172929763794, -4.20265531539917], "reading_logprob_std": 0.47618222072188193, "tokenization_path_logprob_std": 29.963275428463163, "trajectory_balance_rmse": 111.21129389203045} {"canonical_length": 8, "canonical_nll": 0.8330545425415039, "generation_rouge_l_f1": 0.7142857142857143, "generation_token_f1": 0.7142857142857143, "prompt_id": "ab024353313a6e65", "reading_canonical_log_prob": -6.664436340332031, "reading_distances": [27, 29, 36, 33], "reading_log_probs": [-6.113619327545166, -8.244169235229492, -7.174607276916504, -7.940580368041992], "reading_logprob_std": 0.9498056266335485, "tokenization_path_logprob_std": 10.559186594987409, "trajectory_balance_rmse": 134.9270029871085} {"canonical_length": 12, "canonical_nll": 0.9634536107381185, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "2321d8df80c63fef", "reading_canonical_log_prob": -11.561443328857422, "reading_distances": [20, 20, 20, 16], "reading_log_probs": [-12.844955444335938, -13.509904861450195, -13.502254486083984, -13.360779762268066], "reading_logprob_std": 0.31392493867984744, "tokenization_path_logprob_std": 61.2602651048439, "trajectory_balance_rmse": 234.91778856198843} {"canonical_length": 16, "canonical_nll": 0.5935214757919312, "generation_rouge_l_f1": 0.2777777777777778, "generation_token_f1": 0.3055555555555556, "prompt_id": "f3f8000f2b6cf3b2", "reading_canonical_log_prob": -9.496343612670898, "reading_distances": [5, 7, 4, 12], "reading_log_probs": [-9.275550842285156, -8.619678497314453, -9.478434562683105, -9.879039764404297], "reading_logprob_std": 0.5259488666939708, "tokenization_path_logprob_std": 11.64437552606575, "trajectory_balance_rmse": 167.83774411166536} {"canonical_length": 8, "canonical_nll": 0.1274949461221695, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "30a8ec8b58b46182", "reading_canonical_log_prob": -1.019959568977356, "reading_distances": [28, 24, 26, 26], "reading_log_probs": [-1.208789348602295, -1.6718144416809082, -1.3481662273406982, -1.4263540506362915], "reading_logprob_std": 0.1941358752368096, "tokenization_path_logprob_std": 22.93064434570944, "trajectory_balance_rmse": 48.49226468778034} {"canonical_length": 267, "canonical_nll": 1.455431305960323, "generation_rouge_l_f1": 0.22180451127819548, "generation_token_f1": 0.4699248120300752, "prompt_id": "f1f8d8a792fc708b", "reading_canonical_log_prob": -388.60015869140625, "reading_distances": [24, 28, 29, 30], "reading_log_probs": [-401.4442138671875, -423.9420166015625, -403.8633728027344, -399.5810852050781], "reading_logprob_std": 11.293138066304316, "tokenization_path_logprob_std": 246.562410582803, "trajectory_balance_rmse": 5739.609238240769} {"canonical_length": 194, "canonical_nll": 0.9341663478576031, "generation_rouge_l_f1": 0.23194748358862144, "generation_token_f1": 0.4814004376367615, "prompt_id": "1d19c88736fcd9d4", "reading_canonical_log_prob": -181.228271484375, "reading_distances": [52, 48, 29, 47], "reading_log_probs": [-188.73194885253906, -186.9727783203125, -185.87252807617188, -186.3923797607422], "reading_logprob_std": 1.2437245988930206, "tokenization_path_logprob_std": 166.5930989043474, "trajectory_balance_rmse": 3585.704298714392} {"canonical_length": 15, "canonical_nll": 2.3578478495279946, "generation_rouge_l_f1": 0.0606060606060606, "generation_token_f1": 0.0606060606060606, "prompt_id": "1dc2cb66ddb80668", "reading_canonical_log_prob": -35.36771774291992, "reading_distances": [15, 15, 15, 13], "reading_log_probs": [-39.56227111816406, -38.8720588684082, -39.8592529296875, -41.031497955322266], "reading_logprob_std": 0.9006990956084006, "tokenization_path_logprob_std": 36.91021326773666, "trajectory_balance_rmse": 312.7893400060905} {"canonical_length": 16, "canonical_nll": 1.753328800201416, "generation_rouge_l_f1": 0.09137055837563451, "generation_token_f1": 0.11167512690355329, "prompt_id": "bf66528e05d6425b", "reading_canonical_log_prob": -28.053260803222656, "reading_distances": [29, 37, 31, 36], "reading_log_probs": [-29.022377014160156, -32.65779495239258, -29.83681869506836, -32.20791244506836], "reading_logprob_std": 1.775049335109368, "tokenization_path_logprob_std": 76.08449005655066, "trajectory_balance_rmse": 515.7878544059292} {"canonical_length": 118, "canonical_nll": 1.2646114543332891, "generation_rouge_l_f1": 0.26332288401253917, "generation_token_f1": 0.42006269592476486, "prompt_id": "471e2be2592f5add", "reading_canonical_log_prob": -149.22415161132812, "reading_distances": [20, 22, 21, 20], "reading_log_probs": [-153.96902465820312, -145.57247924804688, -152.0489501953125, -154.20599365234375], "reading_logprob_std": 4.035053680431272, "tokenization_path_logprob_std": 168.35979004097015, "trajectory_balance_rmse": 2782.027276930456} {"canonical_length": 19, "canonical_nll": 4.583344308953536, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "57103ee2e4cc89e4", "reading_canonical_log_prob": -87.08354187011719, "reading_distances": [36, 32, 40, 39], "reading_log_probs": [-91.74303436279297, -88.22164154052734, -83.63468170166016, -89.71975708007812], "reading_logprob_std": 3.4466400830253754, "tokenization_path_logprob_std": 16.620540264174732, "trajectory_balance_rmse": 251.52855139336566} {"canonical_length": 23, "canonical_nll": 2.2500918844471807, "generation_rouge_l_f1": 0.12403100775193798, "generation_token_f1": 0.15503875968992248, "prompt_id": "aa0821e4d154fd43", "reading_canonical_log_prob": -51.752113342285156, "reading_distances": [20, 22, 25, 21], "reading_log_probs": [-54.74601745605469, -52.161460876464844, -51.954429626464844, -53.125213623046875], "reading_logprob_std": 1.2728680492684512, "tokenization_path_logprob_std": 24.29409776121416, "trajectory_balance_rmse": 511.23269122664954} {"canonical_length": 24, "canonical_nll": 1.3427181243896484, "generation_rouge_l_f1": 0.3, "generation_token_f1": 0.39999999999999997, "prompt_id": "2aaa42d342162266", "reading_canonical_log_prob": -32.22523498535156, "reading_distances": [31, 32, 27, 28], "reading_log_probs": [-36.267093658447266, -38.113861083984375, -39.383792877197266, -36.83338165283203], "reading_logprob_std": 1.3905034395004643, "tokenization_path_logprob_std": 106.58289010868288, "trajectory_balance_rmse": 702.4337815463634} {"canonical_length": 10, "canonical_nll": 0.37697687149047854, "generation_rouge_l_f1": 0.07207207207207207, "generation_token_f1": 0.07207207207207207, "prompt_id": "b5604eaaebf6e52c", "reading_canonical_log_prob": -3.769768714904785, "reading_distances": [19, 20, 12, 15], "reading_log_probs": [-6.37447452545166, -4.34709358215332, -5.025496482849121, -4.778307914733887], "reading_logprob_std": 0.8748829246602822, "tokenization_path_logprob_std": 54.763919768907314, "trajectory_balance_rmse": 233.92631239349947} {"canonical_length": 223, "canonical_nll": 1.3949092300483463, "generation_rouge_l_f1": 0.20140515222482436, "generation_token_f1": 0.3700234192037471, "prompt_id": "5dbcfc1c200ae9f7", "reading_canonical_log_prob": -311.06475830078125, "reading_distances": [23, 24, 25, 24], "reading_log_probs": [-308.6935729980469, -318.706787109375, -317.89434814453125, -307.7208251953125], "reading_logprob_std": 5.850334348495951, "tokenization_path_logprob_std": 203.08911535172757, "trajectory_balance_rmse": 5101.997937733006} {"canonical_length": 163, "canonical_nll": 1.376762530555023, "generation_rouge_l_f1": 0.1783132530120482, "generation_token_f1": 0.3710843373493976, "prompt_id": "d526cee0379fafdd", "reading_canonical_log_prob": -224.41229248046875, "reading_distances": [34, 29, 22, 30], "reading_log_probs": [-225.97593688964844, -223.3191375732422, -220.60452270507812, -222.7720947265625], "reading_logprob_std": 2.208733343319758, "tokenization_path_logprob_std": 214.7919720047841, "trajectory_balance_rmse": 4081.180530606745} {"canonical_length": 85, "canonical_nll": 1.4661962172564338, "generation_rouge_l_f1": 0.25974025974025977, "generation_token_f1": 0.44155844155844154, "prompt_id": "b641694c0de50f89", "reading_canonical_log_prob": -124.62667846679688, "reading_distances": [35, 25, 31, 37], "reading_log_probs": [-148.50222778320312, -143.75704956054688, -140.98219299316406, -144.9931640625], "reading_logprob_std": 3.1184513351269416, "tokenization_path_logprob_std": 63.1447038282181, "trajectory_balance_rmse": 1845.1891884812223} {"canonical_length": 13, "canonical_nll": 0.39441735927875227, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.2857142857142857, "prompt_id": "2e5bfbd760499e1d", "reading_canonical_log_prob": -5.127425670623779, "reading_distances": [22, 24, 25, 19], "reading_log_probs": [-4.6389265060424805, -3.687152624130249, -3.424534797668457, -4.257866859436035], "reading_logprob_std": 0.5488588452610618, "tokenization_path_logprob_std": 17.13065941680172, "trajectory_balance_rmse": 359.8209991178775} {"canonical_length": 363, "canonical_nll": 1.380949330395575, "generation_rouge_l_f1": 0.2360655737704918, "generation_token_f1": 0.4590163934426229, "prompt_id": "1d0a6d1714584f14", "reading_canonical_log_prob": -501.28460693359375, "reading_distances": [36, 39, 33, 28], "reading_log_probs": [-500.3889465332031, -498.9797058105469, -501.5426025390625, -506.86279296875], "reading_logprob_std": 3.4429093590824786, "tokenization_path_logprob_std": 101.25014796422813, "trajectory_balance_rmse": 7299.147006420668} {"canonical_length": 37, "canonical_nll": 1.978715432656778, "generation_rouge_l_f1": 0.05405405405405406, "generation_token_f1": 0.05405405405405406, "prompt_id": "c7b4cbd5453e19a7", "reading_canonical_log_prob": -73.21247100830078, "reading_distances": [24, 21, 16, 21], "reading_log_probs": [-69.87948608398438, -74.912841796875, -72.61984252929688, -71.74185180664062], "reading_logprob_std": 2.089587681280892, "tokenization_path_logprob_std": 54.294605723842245, "trajectory_balance_rmse": 733.1152100697511} {"canonical_length": 10, "canonical_nll": 0.6394563674926758, "generation_rouge_l_f1": 0.5833333333333334, "generation_token_f1": 0.6666666666666666, "prompt_id": "88beef2f88591814", "reading_canonical_log_prob": -6.394563674926758, "reading_distances": [62, 63, 57, 70], "reading_log_probs": [-9.429936408996582, -9.157651901245117, -9.25576400756836, -8.878621101379395], "reading_logprob_std": 0.23060507210440104, "tokenization_path_logprob_std": 33.14028549468664, "trajectory_balance_rmse": 268.5822891812413} {"canonical_length": 65, "canonical_nll": 0.8661203824556791, "generation_rouge_l_f1": 0.7746478873239436, "generation_token_f1": 0.7887323943661972, "prompt_id": "cfc50376a0f13c13", "reading_canonical_log_prob": -56.29782485961914, "reading_distances": [28, 24, 30, 17], "reading_log_probs": [-59.560096740722656, -60.401336669921875, -59.4659538269043, -59.17596435546875], "reading_logprob_std": 0.5263615251331142, "tokenization_path_logprob_std": 66.64533473377651, "trajectory_balance_rmse": 708.5663138794527} {"canonical_length": 90, "canonical_nll": 1.5453548855251735, "generation_rouge_l_f1": 0.26022304832713755, "generation_token_f1": 0.39405204460966536, "prompt_id": "6a7a68d7cf446159", "reading_canonical_log_prob": -139.08193969726562, "reading_distances": [21, 18, 14, 18], "reading_log_probs": [-156.81521606445312, -153.08326721191406, -149.3349151611328, -153.218994140625], "reading_logprob_std": 3.054638418077099, "tokenization_path_logprob_std": 79.33829345169502, "trajectory_balance_rmse": 2072.1735761512427} {"canonical_length": 216, "canonical_nll": 1.6467501322428386, "generation_rouge_l_f1": 0.2208067940552017, "generation_token_f1": 0.42462845010615713, "prompt_id": "a929bd728f13cca9", "reading_canonical_log_prob": -355.6980285644531, "reading_distances": [35, 36, 29, 38], "reading_log_probs": [-361.11627197265625, -366.558837890625, -364.03704833984375, -365.48583984375], "reading_logprob_std": 2.3603553035413403, "tokenization_path_logprob_std": 256.6633658023985, "trajectory_balance_rmse": 5856.588834630319} {"canonical_length": 321, "canonical_nll": 0.9285406665267232, "generation_rouge_l_f1": 0.2273476112026359, "generation_token_f1": 0.46787479406919275, "prompt_id": "167bbbea26df8184", "reading_canonical_log_prob": -298.0615539550781, "reading_distances": [31, 34, 34, 37], "reading_log_probs": [-310.5487976074219, -314.5894775390625, -307.4354248046875, -309.5858459472656], "reading_logprob_std": 2.9970620494864106, "tokenization_path_logprob_std": 278.302581439145, "trajectory_balance_rmse": 6203.89353097579} {"canonical_length": 195, "canonical_nll": 1.6795835837339743, "generation_rouge_l_f1": 0.2403846153846154, "generation_token_f1": 0.4278846153846154, "prompt_id": "00b0a2b88466f41c", "reading_canonical_log_prob": -327.518798828125, "reading_distances": [21, 26, 16, 25], "reading_log_probs": [-339.9766845703125, -340.30078125, -338.25677490234375, -342.37335205078125], "reading_logprob_std": 1.688857158658905, "tokenization_path_logprob_std": 104.8102261582482, "trajectory_balance_rmse": 4291.961230250935} {"canonical_length": 89, "canonical_nll": 1.3383590183900984, "generation_rouge_l_f1": 0.17901234567901234, "generation_token_f1": 0.2901234567901234, "prompt_id": "a26fac121253e753", "reading_canonical_log_prob": -119.11395263671875, "reading_distances": [29, 27, 29, 31], "reading_log_probs": [-121.15413665771484, -118.25807189941406, -119.18081665039062, -121.43389892578125], "reading_logprob_std": 1.5376704452609995, "tokenization_path_logprob_std": 69.79278549768378, "trajectory_balance_rmse": 2140.8481252760143} {"canonical_length": 223, "canonical_nll": 0.9883807400416901, "generation_rouge_l_f1": 0.20140515222482436, "generation_token_f1": 0.28103044496487123, "prompt_id": "109439f51e6651d1", "reading_canonical_log_prob": -220.40890502929688, "reading_distances": [38, 33, 38, 41], "reading_log_probs": [-228.2572021484375, -230.22994995117188, -228.42636108398438, -226.64187622070312], "reading_logprob_std": 1.4674586483635037, "tokenization_path_logprob_std": 260.96046175448606, "trajectory_balance_rmse": 3629.516839245051} {"canonical_length": 174, "canonical_nll": 1.0217946370442708, "generation_rouge_l_f1": 0.2818181818181818, "generation_token_f1": 0.5409090909090909, "prompt_id": "25f04ac4bbf2d3c3", "reading_canonical_log_prob": -177.79226684570312, "reading_distances": [22, 20, 16, 20], "reading_log_probs": [-180.73764038085938, -180.0430908203125, -178.9196319580078, -181.40650939941406], "reading_logprob_std": 1.0622526742455927, "tokenization_path_logprob_std": 97.162472562456, "trajectory_balance_rmse": 3740.5976551442445} {"canonical_length": 13, "canonical_nll": 1.484188960148738, "generation_rouge_l_f1": 0.06796116504854369, "generation_token_f1": 0.06796116504854369, "prompt_id": "68cd67de83285a28", "reading_canonical_log_prob": -19.294456481933594, "reading_distances": [10, 9, 11, 10], "reading_log_probs": [-18.332916259765625, -16.368274688720703, -14.996641159057617, -15.36241626739502], "reading_logprob_std": 1.4955862671062838, "tokenization_path_logprob_std": 15.689847059043576, "trajectory_balance_rmse": 178.7387951166794} {"canonical_length": 12, "canonical_nll": 0.33476706345876056, "generation_rouge_l_f1": 0.145985401459854, "generation_token_f1": 0.145985401459854, "prompt_id": "41752c3a8cc3cb46", "reading_canonical_log_prob": -4.017204761505127, "reading_distances": [20, 21, 21, 19], "reading_log_probs": [-3.0427355766296387, -2.8204784393310547, -2.8336753845214844, -3.191840171813965], "reading_logprob_std": 0.17834944394850177, "tokenization_path_logprob_std": 26.593502376423636, "trajectory_balance_rmse": 201.13358925173375} {"canonical_length": 33, "canonical_nll": 0.41971374280524976, "generation_rouge_l_f1": 0.22950819672131148, "generation_token_f1": 0.3770491803278689, "prompt_id": "6595c3b5c5d331ea", "reading_canonical_log_prob": -13.850553512573242, "reading_distances": [24, 21, 22, 23], "reading_log_probs": [-17.41798210144043, -17.379091262817383, -19.609493255615234, -21.488372802734375], "reading_logprob_std": 1.9740691957169634, "tokenization_path_logprob_std": 53.78306755159017, "trajectory_balance_rmse": 332.9118755485722} {"canonical_length": 7, "canonical_nll": 0.3842164788927351, "generation_rouge_l_f1": 0.4615384615384615, "generation_token_f1": 0.7692307692307692, "prompt_id": "77f51b75ebf88292", "reading_canonical_log_prob": -2.6895153522491455, "reading_distances": [19, 22, 19, 26], "reading_log_probs": [-4.327537536621094, -3.7768797874450684, -2.9673187732696533, -3.5260848999023438], "reading_logprob_std": 0.5646718859484358, "tokenization_path_logprob_std": 38.50850163029403, "trajectory_balance_rmse": 131.28609530522792} {"canonical_length": 250, "canonical_nll": 1.0323345947265625, "generation_rouge_l_f1": 0.3076923076923077, "generation_token_f1": 0.5256410256410255, "prompt_id": "3d42fc04af88620e", "reading_canonical_log_prob": -258.0836486816406, "reading_distances": [29, 25, 33, 29], "reading_log_probs": [-270.6935729980469, -269.32012939453125, -272.2176513671875, -270.89678955078125], "reading_logprob_std": 1.1859112510005978, "tokenization_path_logprob_std": 59.15667949479624, "trajectory_balance_rmse": 5067.980331117482} {"canonical_length": 311, "canonical_nll": 0.6012236665682777, "generation_rouge_l_f1": 0.3961165048543689, "generation_token_f1": 0.5631067961165048, "prompt_id": "391687d0204ea44e", "reading_canonical_log_prob": -186.98056030273438, "reading_distances": [21, 24, 23, 26], "reading_log_probs": [-199.38998413085938, -194.42575073242188, -190.9603729248047, -198.5046844482422], "reading_logprob_std": 3.8949635622782193, "tokenization_path_logprob_std": 175.3854250291541, "trajectory_balance_rmse": 4060.944355821429} {"canonical_length": 45, "canonical_nll": 1.5734771728515624, "generation_rouge_l_f1": 0.13215859030837004, "generation_token_f1": 0.1762114537444934, "prompt_id": "9043379228e84740", "reading_canonical_log_prob": -70.80647277832031, "reading_distances": [32, 31, 36, 33], "reading_log_probs": [-76.9529800415039, -79.878662109375, -75.92924499511719, -76.80668640136719], "reading_logprob_std": 1.7183782216330203, "tokenization_path_logprob_std": 61.79537592775453, "trajectory_balance_rmse": 985.2187613619909} {"canonical_length": 14, "canonical_nll": 1.4326340811593192, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.5263157894736842, "prompt_id": "82a1a337beb39341", "reading_canonical_log_prob": -20.05687713623047, "reading_distances": [26, 23, 25, 34], "reading_log_probs": [-19.631364822387695, -27.00426483154297, -20.11553192138672, -25.77166748046875], "reading_logprob_std": 3.799815739585818, "tokenization_path_logprob_std": 69.9270266203691, "trajectory_balance_rmse": 353.08635556849936} {"canonical_length": 15, "canonical_nll": 1.1341517130533854, "generation_rouge_l_f1": 0.056818181818181816, "generation_token_f1": 0.0909090909090909, "prompt_id": "32a6cf563f35ecba", "reading_canonical_log_prob": -17.01227569580078, "reading_distances": [33, 25, 30, 33], "reading_log_probs": [-26.709611892700195, -31.003684997558594, -23.417856216430664, -23.438697814941406], "reading_logprob_std": 3.5910520963939105, "tokenization_path_logprob_std": 27.225887918055165, "trajectory_balance_rmse": 182.7872488325012} {"canonical_length": 147, "canonical_nll": 1.533760745509141, "generation_rouge_l_f1": 0.16986301369863016, "generation_token_f1": 0.27945205479452057, "prompt_id": "2655c987f350c6c7", "reading_canonical_log_prob": -225.46282958984375, "reading_distances": [14, 17, 14, 15], "reading_log_probs": [-228.8407440185547, -228.51467895507812, -231.005615234375, -229.2171630859375], "reading_logprob_std": 1.1117360939042578, "tokenization_path_logprob_std": 89.95321131115624, "trajectory_balance_rmse": 3304.3767276327185} {"canonical_length": 392, "canonical_nll": 1.0269138569734535, "generation_rouge_l_f1": 0.24731182795698922, "generation_token_f1": 0.4336917562724014, "prompt_id": "dd04ff95cd33338f", "reading_canonical_log_prob": -402.55023193359375, "reading_distances": [25, 30, 24, 23], "reading_log_probs": [-415.50006103515625, -438.5897216796875, -418.50732421875, -408.85821533203125], "reading_logprob_std": 12.801889947278232, "tokenization_path_logprob_std": 86.33557273107203, "trajectory_balance_rmse": 4333.319171325524} {"canonical_length": 197, "canonical_nll": 0.6224484516279346, "generation_rouge_l_f1": 0.17177914110429446, "generation_token_f1": 0.3803680981595092, "prompt_id": "f51cb0e119af1f71", "reading_canonical_log_prob": -122.62234497070312, "reading_distances": [38, 44, 38, 39], "reading_log_probs": [-145.75982666015625, -141.9145050048828, -145.21717834472656, -141.67864990234375], "reading_logprob_std": 2.145177380232285, "tokenization_path_logprob_std": 234.0767112770556, "trajectory_balance_rmse": 3605.632799779413} {"canonical_length": 11, "canonical_nll": 1.6177347356622869, "generation_rouge_l_f1": 0.15384615384615383, "generation_token_f1": 0.15384615384615383, "prompt_id": "40f09143e7b23121", "reading_canonical_log_prob": -17.795082092285156, "reading_distances": [42, 47, 49, 42], "reading_log_probs": [-16.18756103515625, -17.463369369506836, -16.261856079101562, -16.91912078857422], "reading_logprob_std": 0.6014028000960242, "tokenization_path_logprob_std": 86.8732402605505, "trajectory_balance_rmse": 291.82022235277714} {"canonical_length": 98, "canonical_nll": 1.3198847868004624, "generation_rouge_l_f1": 0.25958702064896755, "generation_token_f1": 0.40117994100294985, "prompt_id": "09e3575b08c40649", "reading_canonical_log_prob": -129.3487091064453, "reading_distances": [16, 17, 13, 13], "reading_log_probs": [-134.04446411132812, -136.84637451171875, -133.30511474609375, -132.23883056640625], "reading_logprob_std": 1.9698687723151533, "tokenization_path_logprob_std": 52.11255204130168, "trajectory_balance_rmse": 2144.162716057061} {"canonical_length": 11, "canonical_nll": 0.6279218847101385, "generation_rouge_l_f1": 0.06926406926406928, "generation_token_f1": 0.06926406926406928, "prompt_id": "fc90d5db5297a0af", "reading_canonical_log_prob": -6.907140731811523, "reading_distances": [18, 22, 20, 20], "reading_log_probs": [-6.753106117248535, -7.216258525848389, -7.339608192443848, -7.324772834777832], "reading_logprob_std": 0.27575778276410423, "tokenization_path_logprob_std": 43.75186116377477, "trajectory_balance_rmse": 248.97618077729632} {"canonical_length": 244, "canonical_nll": 1.0681304931640625, "generation_rouge_l_f1": 0.33946830265848676, "generation_token_f1": 0.6298568507157464, "prompt_id": "e15ab7087e38fd8b", "reading_canonical_log_prob": -260.62384033203125, "reading_distances": [21, 16, 16, 15], "reading_log_probs": [-270.8970642089844, -268.90167236328125, -268.69256591796875, -265.1257019042969], "reading_logprob_std": 2.4009411607669344, "tokenization_path_logprob_std": 133.97977833113188, "trajectory_balance_rmse": 4760.541334374687} {"canonical_length": 83, "canonical_nll": 0.9577029124799982, "generation_rouge_l_f1": 0.43283582089552236, "generation_token_f1": 0.5671641791044777, "prompt_id": "514750b55e58033f", "reading_canonical_log_prob": -79.48934173583984, "reading_distances": [17, 18, 16, 13], "reading_log_probs": [-88.26336669921875, -79.38911437988281, -85.41803741455078, -79.921630859375], "reading_logprob_std": 4.313494151746569, "tokenization_path_logprob_std": 171.27798746015102, "trajectory_balance_rmse": 1711.4711704282324} {"canonical_length": 145, "canonical_nll": 1.3381583378232758, "generation_rouge_l_f1": 0.16346153846153846, "generation_token_f1": 0.2932692307692308, "prompt_id": "612827ad6442cf24", "reading_canonical_log_prob": -194.032958984375, "reading_distances": [27, 26, 28, 29], "reading_log_probs": [-205.2562713623047, -201.22589111328125, -199.14210510253906, -202.605224609375], "reading_logprob_std": 2.564061957140016, "tokenization_path_logprob_std": 245.25927737299426, "trajectory_balance_rmse": 3588.093636828092} {"canonical_length": 184, "canonical_nll": 0.7947608284328295, "generation_rouge_l_f1": 0.31529411764705884, "generation_token_f1": 0.4423529411764706, "prompt_id": "637d5437d0c071a7", "reading_canonical_log_prob": -146.23599243164062, "reading_distances": [249, 215, 215, 227], "reading_log_probs": [-164.65899658203125, -165.4076385498047, -158.52976989746094, -157.632080078125], "reading_logprob_std": 4.042231993350255, "tokenization_path_logprob_std": 94.76820730885204, "trajectory_balance_rmse": 2962.910266770765} {"canonical_length": 65, "canonical_nll": 1.0580731905423677, "generation_rouge_l_f1": 0.14285714285714288, "generation_token_f1": 0.14285714285714288, "prompt_id": "c97d2c584d8df586", "reading_canonical_log_prob": -68.7747573852539, "reading_distances": [15, 19, 16, 17], "reading_log_probs": [-71.36539459228516, -72.28498840332031, -72.34886169433594, -72.2350082397461], "reading_logprob_std": 0.4644556524523048, "tokenization_path_logprob_std": 38.7343906462762, "trajectory_balance_rmse": 874.8849622107581} {"canonical_length": 17, "canonical_nll": 1.0733529259176815, "generation_rouge_l_f1": 0.6923076923076923, "generation_token_f1": 0.6923076923076923, "prompt_id": "a17a07dc87573da6", "reading_canonical_log_prob": -18.246999740600586, "reading_distances": [29, 25, 21, 24], "reading_log_probs": [-19.824642181396484, -19.980958938598633, -21.85655975341797, -19.79990577697754], "reading_logprob_std": 0.9972554036148262, "tokenization_path_logprob_std": 116.91472483226923, "trajectory_balance_rmse": 488.0745428397508} {"canonical_length": 13, "canonical_nll": 0.4029342211209811, "generation_rouge_l_f1": 0.7, "generation_token_f1": 0.7, "prompt_id": "4b95b246a8b6c8d6", "reading_canonical_log_prob": -5.238144874572754, "reading_distances": [10, 12, 14, 12], "reading_log_probs": [-4.565258502960205, -3.772770404815674, -8.36904525756836, -4.9940104484558105], "reading_logprob_std": 2.0266698674582257, "tokenization_path_logprob_std": 20.326674021944537, "trajectory_balance_rmse": 326.0634881555477} {"canonical_length": 99, "canonical_nll": 1.229902325254498, "generation_rouge_l_f1": 0.32773109243697485, "generation_token_f1": 0.5546218487394958, "prompt_id": "0c93e5e51c22b30d", "reading_canonical_log_prob": -121.76033020019531, "reading_distances": [17, 18, 18, 15], "reading_log_probs": [-127.68283081054688, -126.42900848388672, -127.92005157470703, -125.73341369628906], "reading_logprob_std": 1.0375057438270614, "tokenization_path_logprob_std": 123.23210447162555, "trajectory_balance_rmse": 2066.1069515310883} {"canonical_length": 306, "canonical_nll": 1.1934272915709252, "generation_rouge_l_f1": 0.2097902097902098, "generation_token_f1": 0.4615384615384615, "prompt_id": "6c10022ee752e034", "reading_canonical_log_prob": -365.1887512207031, "reading_distances": [29, 34, 30, 36], "reading_log_probs": [-382.15301513671875, -383.4436950683594, -382.0418701171875, -388.07781982421875], "reading_logprob_std": 2.8380514559396763, "tokenization_path_logprob_std": 253.94382232192018, "trajectory_balance_rmse": 6571.119360450844} {"canonical_length": 69, "canonical_nll": 1.1289042597231658, "generation_rouge_l_f1": 0.23134328358208955, "generation_token_f1": 0.3731343283582089, "prompt_id": "7884c89879ae1c04", "reading_canonical_log_prob": -77.89439392089844, "reading_distances": [35, 28, 26, 37], "reading_log_probs": [-83.47035217285156, -82.6593017578125, -80.75751495361328, -84.0531234741211], "reading_logprob_std": 1.436937672312382, "tokenization_path_logprob_std": 129.90854478722719, "trajectory_balance_rmse": 1263.810665021518} {"canonical_length": 46, "canonical_nll": 0.953792489093283, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.5185185185185185, "prompt_id": "803eb754dd6450de", "reading_canonical_log_prob": -43.874454498291016, "reading_distances": [14, 15, 17, 13], "reading_log_probs": [-49.46244812011719, -45.5025634765625, -52.2197265625, -47.36164855957031], "reading_logprob_std": 2.8849325345112207, "tokenization_path_logprob_std": 50.938606345510976, "trajectory_balance_rmse": 1141.9344122950733} {"canonical_length": 12, "canonical_nll": 0.9686092535654703, "generation_rouge_l_f1": 0.37499999999999994, "generation_token_f1": 0.37499999999999994, "prompt_id": "7f58078c29e97e0d", "reading_canonical_log_prob": -11.623311042785645, "reading_distances": [39, 46, 39, 39], "reading_log_probs": [-26.68197250366211, -17.64406967163086, -17.775184631347656, -24.919658660888672], "reading_logprob_std": 4.7268312440447735, "tokenization_path_logprob_std": 16.040308989321986, "trajectory_balance_rmse": 186.7209838555478} {"canonical_length": 135, "canonical_nll": 1.4648118760850695, "generation_rouge_l_f1": 0.20069204152249134, "generation_token_f1": 0.46366782006920415, "prompt_id": "979bfe699ee39dca", "reading_canonical_log_prob": -197.74960327148438, "reading_distances": [22, 21, 20, 29], "reading_log_probs": [-196.91375732421875, -198.2593994140625, -196.0160675048828, -198.55966186523438], "reading_logprob_std": 1.1873699672025178, "tokenization_path_logprob_std": 200.76876698810815, "trajectory_balance_rmse": 2555.35216278159} {"canonical_length": 13, "canonical_nll": 1.514238797701322, "generation_rouge_l_f1": 0.07881773399014778, "generation_token_f1": 0.09852216748768473, "prompt_id": "f30d075da952702b", "reading_canonical_log_prob": -19.685104370117188, "reading_distances": [34, 26, 31, 29], "reading_log_probs": [-34.707210540771484, -31.26356315612793, -27.914390563964844, -32.13501739501953], "reading_logprob_std": 2.8048673247843703, "tokenization_path_logprob_std": 78.6027615760875, "trajectory_balance_rmse": 203.22335552421063} {"canonical_length": 117, "canonical_nll": 1.126392592731704, "generation_rouge_l_f1": 0.22857142857142856, "generation_token_f1": 0.37402597402597404, "prompt_id": "ad627413d7298d7b", "reading_canonical_log_prob": -131.78793334960938, "reading_distances": [24, 17, 22, 29], "reading_log_probs": [-136.36361694335938, -137.7444610595703, -134.2756805419922, -138.78802490234375], "reading_logprob_std": 1.9499290209407942, "tokenization_path_logprob_std": 145.53330737516328, "trajectory_balance_rmse": 2250.873348763687} {"canonical_length": 314, "canonical_nll": 1.561489129522044, "generation_rouge_l_f1": 0.28835489833641403, "generation_token_f1": 0.5360443622920519, "prompt_id": "802f028e684a8238", "reading_canonical_log_prob": -490.3075866699219, "reading_distances": [24, 22, 23, 19], "reading_log_probs": [-493.48626708984375, -492.62969970703125, -495.2522277832031, -493.72357177734375], "reading_logprob_std": 1.0923841344839698, "tokenization_path_logprob_std": 307.4095015578933, "trajectory_balance_rmse": 6133.560237653461} {"canonical_length": 190, "canonical_nll": 1.4865819027549343, "generation_rouge_l_f1": 0.24516129032258063, "generation_token_f1": 0.421505376344086, "prompt_id": "2e2423f516c40a85", "reading_canonical_log_prob": -282.4505615234375, "reading_distances": [8, 16, 21, 16], "reading_log_probs": [-285.5587463378906, -285.17327880859375, -282.6826477050781, -280.02984619140625], "reading_logprob_std": 2.560676402240409, "tokenization_path_logprob_std": 199.4472137266515, "trajectory_balance_rmse": 4308.7868867016} {"canonical_length": 297, "canonical_nll": 1.574747824106955, "generation_rouge_l_f1": 0.23344947735191635, "generation_token_f1": 0.421602787456446, "prompt_id": "8d1b09dca5de1a98", "reading_canonical_log_prob": -467.7001037597656, "reading_distances": [26, 21, 28, 23], "reading_log_probs": [-463.3004150390625, -462.8702392578125, -461.2901611328125, -461.13824462890625], "reading_logprob_std": 1.0962315658938653, "tokenization_path_logprob_std": 185.83721667396196, "trajectory_balance_rmse": 7385.541287498474} {"canonical_length": 82, "canonical_nll": 0.6367000021585604, "generation_rouge_l_f1": 0.23052959501557632, "generation_token_f1": 0.3613707165109034, "prompt_id": "0609bdeeadeb9baf", "reading_canonical_log_prob": -52.20940017700195, "reading_distances": [35, 36, 39, 34], "reading_log_probs": [-55.81857681274414, -55.13807678222656, -54.7835693359375, -55.69906997680664], "reading_logprob_std": 0.4853806899131879, "tokenization_path_logprob_std": 85.5790353186114, "trajectory_balance_rmse": 987.8274712437241} {"canonical_length": 222, "canonical_nll": 1.089935302734375, "generation_rouge_l_f1": 0.3976143141153082, "generation_token_f1": 0.6242544731610338, "prompt_id": "e9f992cd65ad5461", "reading_canonical_log_prob": -241.96563720703125, "reading_distances": [25, 32, 29, 24], "reading_log_probs": [-241.8640594482422, -240.9583282470703, -246.92735290527344, -241.48611450195312], "reading_logprob_std": 2.770604944262551, "tokenization_path_logprob_std": 117.92715025846178, "trajectory_balance_rmse": 3236.6159801725353} {"canonical_length": 6, "canonical_nll": 1.322067419687907, "generation_rouge_l_f1": 0.06153846153846154, "generation_token_f1": 0.06153846153846154, "prompt_id": "15eeee68ceb20913", "reading_canonical_log_prob": -7.932404518127441, "reading_distances": [32, 38, 36, 31], "reading_log_probs": [-6.417309761047363, -8.176240921020508, -8.986620903015137, -8.636940002441406], "reading_logprob_std": 1.1406574518430488, "tokenization_path_logprob_std": 49.060299399907834, "trajectory_balance_rmse": 152.7357999684061} {"canonical_length": 105, "canonical_nll": 1.3476004464285714, "generation_rouge_l_f1": 0.2076502732240437, "generation_token_f1": 0.32786885245901637, "prompt_id": "728a7ca9e481ba81", "reading_canonical_log_prob": -141.498046875, "reading_distances": [29, 30, 30, 24], "reading_log_probs": [-141.82452392578125, -140.52830505371094, -136.43753051757812, -141.91940307617188], "reading_logprob_std": 2.5727644042894235, "tokenization_path_logprob_std": 60.64120179927275, "trajectory_balance_rmse": 1969.716512065201} {"canonical_length": 123, "canonical_nll": 1.4166231232930004, "generation_rouge_l_f1": 0.12994350282485875, "generation_token_f1": 0.22033898305084745, "prompt_id": "92cf4663b97aed32", "reading_canonical_log_prob": -174.24464416503906, "reading_distances": [26, 24, 30, 22], "reading_log_probs": [-176.00527954101562, -175.9342041015625, -177.5452423095703, -171.59385681152344], "reading_logprob_std": 2.560765438904962, "tokenization_path_logprob_std": 139.44829184330266, "trajectory_balance_rmse": 1817.6260211496228} {"canonical_length": 66, "canonical_nll": 1.0237234866980351, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.4505494505494505, "prompt_id": "3226fde653ee5231", "reading_canonical_log_prob": -67.56575012207031, "reading_distances": [8, 13, 12, 12], "reading_log_probs": [-62.58056640625, -67.21707153320312, -66.59766387939453, -64.49691772460938], "reading_logprob_std": 2.1115296838081012, "tokenization_path_logprob_std": 69.33278027129403, "trajectory_balance_rmse": 1559.3597158098687} {"canonical_length": 259, "canonical_nll": 1.2596417872601955, "generation_rouge_l_f1": 0.3202099737532808, "generation_token_f1": 0.4881889763779528, "prompt_id": "d40a7f2ec7ebb044", "reading_canonical_log_prob": -326.2472229003906, "reading_distances": [22, 24, 26, 20], "reading_log_probs": [-341.02490234375, -342.7364807128906, -339.12274169921875, -344.58831787109375], "reading_logprob_std": 2.3382076337412796, "tokenization_path_logprob_std": 71.33971298741346, "trajectory_balance_rmse": 4738.8717069773365} {"canonical_length": 13, "canonical_nll": 1.3694592989408052, "generation_rouge_l_f1": 0.21428571428571427, "generation_token_f1": 0.21428571428571427, "prompt_id": "6844823da97b3fe4", "reading_canonical_log_prob": -17.80297088623047, "reading_distances": [20, 24, 22, 21], "reading_log_probs": [-20.155590057373047, -21.232736587524414, -23.452102661132812, -25.99329376220703], "reading_logprob_std": 2.5844433174586725, "tokenization_path_logprob_std": 72.45786509582554, "trajectory_balance_rmse": 280.39895259864204} {"canonical_length": 118, "canonical_nll": 0.9292996293407375, "generation_rouge_l_f1": 0.185929648241206, "generation_token_f1": 0.3165829145728643, "prompt_id": "49597fd7a90f1c14", "reading_canonical_log_prob": -109.65735626220703, "reading_distances": [24, 25, 31, 24], "reading_log_probs": [-117.4195556640625, -119.83979797363281, -119.97705078125, -122.56255340576172], "reading_logprob_std": 2.100909310678328, "tokenization_path_logprob_std": 129.39176615442156, "trajectory_balance_rmse": 1815.601138314494} {"canonical_length": 258, "canonical_nll": 1.6452310251635174, "generation_rouge_l_f1": 0.17509727626459143, "generation_token_f1": 0.32684824902723736, "prompt_id": "06dc86d1f57bb10a", "reading_canonical_log_prob": -424.4696044921875, "reading_distances": [29, 34, 39, 30], "reading_log_probs": [-432.6266174316406, -431.4878845214844, -426.61468505859375, -434.80322265625], "reading_logprob_std": 3.46373283399737, "tokenization_path_logprob_std": 158.9500252378858, "trajectory_balance_rmse": 5303.248256707006} {"canonical_length": 250, "canonical_nll": 1.3201622314453125, "generation_rouge_l_f1": 0.3089108910891089, "generation_token_f1": 0.499009900990099, "prompt_id": "104414445caa5ddc", "reading_canonical_log_prob": -330.0405578613281, "reading_distances": [37, 32, 36, 35], "reading_log_probs": [-330.5101623535156, -330.85113525390625, -331.070068359375, -331.20684814453125], "reading_logprob_std": 0.3039051963545587, "tokenization_path_logprob_std": 322.0614173043568, "trajectory_balance_rmse": 5736.279590553715} {"canonical_length": 274, "canonical_nll": 0.8694618949054802, "generation_rouge_l_f1": 0.3578154425612053, "generation_token_f1": 0.5725047080979284, "prompt_id": "96cee7006ba2cc41", "reading_canonical_log_prob": -238.23255920410156, "reading_distances": [20, 20, 21, 18], "reading_log_probs": [-247.53038024902344, -249.85494995117188, -249.9408416748047, -245.6497802734375], "reading_logprob_std": 2.058612846282019, "tokenization_path_logprob_std": 115.85481651317455, "trajectory_balance_rmse": 4647.693594006642} {"canonical_length": 150, "canonical_nll": 2.0009311930338542, "generation_rouge_l_f1": 0.21052631578947367, "generation_token_f1": 0.39234449760765544, "prompt_id": "c61662a19e060e22", "reading_canonical_log_prob": -300.1396789550781, "reading_distances": [28, 29, 33, 33], "reading_log_probs": [-304.54296875, -309.95758056640625, -307.7597351074219, -311.710205078125], "reading_logprob_std": 3.0895430330105036, "tokenization_path_logprob_std": 230.86274779327576, "trajectory_balance_rmse": 3169.9905829489517} {"canonical_length": 80, "canonical_nll": 1.324944019317627, "generation_rouge_l_f1": 0.32926829268292684, "generation_token_f1": 0.4878048780487805, "prompt_id": "448e404116b3a283", "reading_canonical_log_prob": -105.99552154541016, "reading_distances": [18, 27, 26, 24], "reading_log_probs": [-112.64912414550781, -121.94136047363281, -116.9033203125, -118.99832153320312], "reading_logprob_std": 3.907135624487951, "tokenization_path_logprob_std": 87.12052106931338, "trajectory_balance_rmse": 1415.5717074222314} {"canonical_length": 40, "canonical_nll": 2.188016891479492, "generation_rouge_l_f1": 0.3703703703703704, "generation_token_f1": 0.5432098765432098, "prompt_id": "62e94a0db3b86cbd", "reading_canonical_log_prob": -87.52067565917969, "reading_distances": [16, 17, 15, 16], "reading_log_probs": [-91.61722564697266, -91.06156921386719, -91.70140075683594, -91.28230285644531], "reading_logprob_std": 0.2974563820744732, "tokenization_path_logprob_std": 53.44485130872228, "trajectory_balance_rmse": 1086.871359115219} {"canonical_length": 203, "canonical_nll": 1.0125751965151633, "generation_rouge_l_f1": 0.2745098039215686, "generation_token_f1": 0.5343137254901961, "prompt_id": "6996ebbe576a8626", "reading_canonical_log_prob": -205.55276489257812, "reading_distances": [33, 32, 37, 29], "reading_log_probs": [-209.2108154296875, -206.32144165039062, -209.07843017578125, -204.86492919921875], "reading_logprob_std": 2.135586151956987, "tokenization_path_logprob_std": 352.3376391907335, "trajectory_balance_rmse": 4427.282146004138} {"canonical_length": 97, "canonical_nll": 1.2851588455672116, "generation_rouge_l_f1": 0.2722513089005235, "generation_token_f1": 0.4712041884816754, "prompt_id": "539a3dd889cf333b", "reading_canonical_log_prob": -124.66040802001953, "reading_distances": [39, 25, 33, 33], "reading_log_probs": [-123.9139404296875, -125.84386444091797, -129.65554809570312, -127.7822494506836], "reading_logprob_std": 2.4740317382962416, "tokenization_path_logprob_std": 162.1420914416631, "trajectory_balance_rmse": 2555.0493547034725} {"canonical_length": 83, "canonical_nll": 1.8684756727103728, "generation_rouge_l_f1": 0.06382978723404255, "generation_token_f1": 0.08510638297872342, "prompt_id": "22dd62a830e20aab", "reading_canonical_log_prob": -155.08348083496094, "reading_distances": [24, 37, 29, 27], "reading_log_probs": [-159.15206909179688, -158.4122772216797, -155.12071228027344, -163.68914794921875], "reading_logprob_std": 3.5294247473075098, "tokenization_path_logprob_std": 160.25390226108036, "trajectory_balance_rmse": 1845.5215898191643} {"canonical_length": 26, "canonical_nll": 1.1219711303710938, "generation_rouge_l_f1": 0.23255813953488372, "generation_token_f1": 0.2790697674418604, "prompt_id": "23f859fe909ddbe9", "reading_canonical_log_prob": -29.171249389648438, "reading_distances": [17, 14, 14, 16], "reading_log_probs": [-43.714752197265625, -33.1866455078125, -39.82737350463867, -40.434940338134766], "reading_logprob_std": 4.413200776533635, "tokenization_path_logprob_std": 101.03265957695243, "trajectory_balance_rmse": 526.8589693553826} {"canonical_length": 97, "canonical_nll": 0.68323469653572, "generation_rouge_l_f1": 0.25555555555555554, "generation_token_f1": 0.38333333333333336, "prompt_id": "9658518adf176971", "reading_canonical_log_prob": -66.27376556396484, "reading_distances": [16, 13, 18, 14], "reading_log_probs": [-58.021060943603516, -60.81249237060547, -59.88477325439453, -62.43540954589844], "reading_logprob_std": 1.842829685104138, "tokenization_path_logprob_std": 84.91718883739132, "trajectory_balance_rmse": 1092.8963815276777} {"canonical_length": 243, "canonical_nll": 1.3437910668643904, "generation_rouge_l_f1": 0.23762376237623764, "generation_token_f1": 0.40792079207920795, "prompt_id": "c496b50e51f7e82e", "reading_canonical_log_prob": -326.5412292480469, "reading_distances": [30, 37, 26, 33], "reading_log_probs": [-326.28778076171875, -323.24273681640625, -326.781982421875, -325.6956787109375], "reading_logprob_std": 1.570306549013543, "tokenization_path_logprob_std": 128.72726258730384, "trajectory_balance_rmse": 5554.152267937754} {"canonical_length": 157, "canonical_nll": 1.0639274257003881, "generation_rouge_l_f1": 0.23448275862068965, "generation_token_f1": 0.4689655172413793, "prompt_id": "9433be6b41e84522", "reading_canonical_log_prob": -167.03660583496094, "reading_distances": [41, 34, 38, 35], "reading_log_probs": [-177.86663818359375, -178.55953979492188, -181.55958557128906, -173.50804138183594], "reading_logprob_std": 3.3224064301344503, "tokenization_path_logprob_std": 182.70167617762507, "trajectory_balance_rmse": 3655.4002227814694} {"canonical_length": 14, "canonical_nll": 0.8172859464372907, "generation_rouge_l_f1": 0.6666666666666667, "generation_token_f1": 0.6666666666666667, "prompt_id": "22dd55d64faca0d8", "reading_canonical_log_prob": -11.44200325012207, "reading_distances": [40, 42, 49, 49], "reading_log_probs": [-16.7855224609375, -17.663820266723633, -18.218780517578125, -18.270343780517578], "reading_logprob_std": 0.689736692909688, "tokenization_path_logprob_std": 41.37648617546838, "trajectory_balance_rmse": 414.1323922151088} {"canonical_length": 52, "canonical_nll": 1.2744770050048828, "generation_rouge_l_f1": 0.13636363636363638, "generation_token_f1": 0.2077922077922078, "prompt_id": "9ce142b4fe68303a", "reading_canonical_log_prob": -66.2728042602539, "reading_distances": [31, 27, 33, 38], "reading_log_probs": [-72.13319396972656, -75.74485778808594, -74.6055908203125, -74.69722747802734], "reading_logprob_std": 1.5312028311856931, "tokenization_path_logprob_std": 78.00203871497628, "trajectory_balance_rmse": 1265.9608324457945} {"canonical_length": 28, "canonical_nll": 1.1275969914027624, "generation_rouge_l_f1": 0.4800000000000001, "generation_token_f1": 0.5333333333333333, "prompt_id": "e62a412929959b77", "reading_canonical_log_prob": -31.572715759277344, "reading_distances": [27, 24, 28, 24], "reading_log_probs": [-38.38939666748047, -31.69305419921875, -30.209686279296875, -35.52544403076172], "reading_logprob_std": 3.7091751314723846, "tokenization_path_logprob_std": 48.613500078652486, "trajectory_balance_rmse": 776.6919943041125} {"canonical_length": 21, "canonical_nll": 1.2547046116420202, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4210526315789474, "prompt_id": "273ee8648fccd69e", "reading_canonical_log_prob": -26.348796844482422, "reading_distances": [31, 32, 30, 28], "reading_log_probs": [-43.21636962890625, -37.39649200439453, -37.43067932128906, -32.82877731323242], "reading_logprob_std": 4.255290338516319, "tokenization_path_logprob_std": 43.861952306636965, "trajectory_balance_rmse": 396.9683644522459} {"canonical_length": 109, "canonical_nll": 1.3102277003297018, "generation_rouge_l_f1": 0.3146067415730337, "generation_token_f1": 0.46067415730337075, "prompt_id": "0ffd3304b685d298", "reading_canonical_log_prob": -142.8148193359375, "reading_distances": [22, 24, 24, 19], "reading_log_probs": [-146.7559051513672, -163.693603515625, -163.04476928710938, -156.71856689453125], "reading_logprob_std": 7.855785066611252, "tokenization_path_logprob_std": 146.27135784760728, "trajectory_balance_rmse": 2403.5065630768936} {"canonical_length": 288, "canonical_nll": 1.1458083258734808, "generation_rouge_l_f1": 0.39080459770114945, "generation_token_f1": 0.5900383141762452, "prompt_id": "6b90af69594f1bfc", "reading_canonical_log_prob": -329.9927978515625, "reading_distances": [30, 33, 33, 33], "reading_log_probs": [-335.2255859375, -337.51800537109375, -336.8992614746094, -337.56427001953125], "reading_logprob_std": 1.093657498584978, "tokenization_path_logprob_std": 217.90446825136496, "trajectory_balance_rmse": 6983.049962584422} {"canonical_length": 256, "canonical_nll": 1.2957913875579834, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.6240601503759399, "prompt_id": "7021b5d065291a9c", "reading_canonical_log_prob": -331.72259521484375, "reading_distances": [18, 24, 19, 23], "reading_log_probs": [-327.726318359375, -331.0401916503906, -329.1961364746094, -332.20489501953125], "reading_logprob_std": 1.9792564216713582, "tokenization_path_logprob_std": 140.71564869312064, "trajectory_balance_rmse": 5728.646753469314} {"canonical_length": 68, "canonical_nll": 0.734378365909352, "generation_rouge_l_f1": 0.3756906077348066, "generation_token_f1": 0.4861878453038674, "prompt_id": "56589cd90846fa45", "reading_canonical_log_prob": -49.93772888183594, "reading_distances": [39, 45, 32, 27], "reading_log_probs": [-66.63157653808594, -64.85526275634766, -58.20077896118164, -56.17434310913086], "reading_logprob_std": 5.060750374085043, "tokenization_path_logprob_std": 22.544211105025788, "trajectory_balance_rmse": 953.691725532039} {"canonical_length": 187, "canonical_nll": 1.5871023902281083, "generation_rouge_l_f1": 0.2329113924050633, "generation_token_f1": 0.49620253164556966, "prompt_id": "b2517684dcd5d425", "reading_canonical_log_prob": -296.78814697265625, "reading_distances": [47, 36, 38, 36], "reading_log_probs": [-303.79254150390625, -300.53253173828125, -298.6169128417969, -301.1263122558594], "reading_logprob_std": 2.137811002756473, "tokenization_path_logprob_std": 73.33342951385374, "trajectory_balance_rmse": 4505.187465536014} {"canonical_length": 2, "canonical_nll": 7.353148460388184, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "44fca97732ef92a8", "reading_canonical_log_prob": -14.706296920776367, "reading_distances": [22, 24, 22, 18], "reading_log_probs": [-10.105454444885254, -10.180644989013672, -13.394423484802246, -8.648943901062012], "reading_logprob_std": 2.002882315777437, "tokenization_path_logprob_std": 21.57331284182077, "trajectory_balance_rmse": 39.24637006684672} {"canonical_length": 14, "canonical_nll": 0.6937011310032436, "generation_rouge_l_f1": 0.37499999999999994, "generation_token_f1": 0.37499999999999994, "prompt_id": "907693271b754ffe", "reading_canonical_log_prob": -9.71181583404541, "reading_distances": [5, 6, 7, 8], "reading_log_probs": [-10.130441665649414, -10.966141700744629, -10.511618614196777, -10.792024612426758], "reading_logprob_std": 0.3647965107107942, "tokenization_path_logprob_std": 19.600296745140657, "trajectory_balance_rmse": 175.9603101287057} {"canonical_length": 220, "canonical_nll": 1.1987394853071733, "generation_rouge_l_f1": 0.21900826446280994, "generation_token_f1": 0.41735537190082644, "prompt_id": "3d885c9990d3f7a8", "reading_canonical_log_prob": -263.7226867675781, "reading_distances": [43, 52, 46, 50], "reading_log_probs": [-273.36328125, -279.24005126953125, -273.0422058105469, -291.0233154296875], "reading_logprob_std": 8.401861045154616, "tokenization_path_logprob_std": 100.78872508239999, "trajectory_balance_rmse": 3981.3107313633673} {"canonical_length": 12, "canonical_nll": 1.1556708017985027, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.22222222222222224, "prompt_id": "946f59f054cd3b7a", "reading_canonical_log_prob": -13.868049621582031, "reading_distances": [27, 25, 33, 27], "reading_log_probs": [-14.323083877563477, -14.32302188873291, -13.614836692810059, -14.522661209106445], "reading_logprob_std": 0.3986406769857445, "tokenization_path_logprob_std": 44.71473085388957, "trajectory_balance_rmse": 260.465843595447} {"canonical_length": 114, "canonical_nll": 1.0690850374991434, "generation_rouge_l_f1": 0.19767441860465118, "generation_token_f1": 0.36046511627906974, "prompt_id": "1e094a0368e082c2", "reading_canonical_log_prob": -121.87569427490234, "reading_distances": [30, 31, 29, 21], "reading_log_probs": [-132.74830627441406, -128.73190307617188, -129.36898803710938, -131.64013671875], "reading_logprob_std": 1.8885872732752016, "tokenization_path_logprob_std": 171.40071134788903, "trajectory_balance_rmse": 2681.449234884013} {"canonical_length": 8, "canonical_nll": 0.13641682267189026, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "f01addec53ef2695", "reading_canonical_log_prob": -1.091334581375122, "reading_distances": [24, 25, 23, 24], "reading_log_probs": [-1.5159099102020264, -0.8188730478286743, -0.7126538753509521, -1.7799417972564697], "reading_logprob_std": 0.5224009191852752, "tokenization_path_logprob_std": 46.18178616133577, "trajectory_balance_rmse": 141.49946143085344} {"canonical_length": 43, "canonical_nll": 1.0626777826353562, "generation_rouge_l_f1": 0.37209302325581395, "generation_token_f1": 0.5116279069767442, "prompt_id": "bd70996cafa5e523", "reading_canonical_log_prob": -45.69514465332031, "reading_distances": [32, 29, 25, 18], "reading_log_probs": [-52.37285614013672, -56.23794174194336, -58.68858337402344, -53.22639465332031], "reading_logprob_std": 2.893475290662504, "tokenization_path_logprob_std": 81.53335692475189, "trajectory_balance_rmse": 821.094424984841} {"canonical_length": 280, "canonical_nll": 1.0499823434012276, "generation_rouge_l_f1": 0.2513863216266174, "generation_token_f1": 0.5175600739371534, "prompt_id": "40b3b0fca502cfb8", "reading_canonical_log_prob": -293.99505615234375, "reading_distances": [29, 29, 26, 34], "reading_log_probs": [-291.36572265625, -295.16534423828125, -293.2851867675781, -291.1662292480469], "reading_logprob_std": 1.8748146141811575, "tokenization_path_logprob_std": 147.0337212963293, "trajectory_balance_rmse": 5463.534585431631} {"canonical_length": 74, "canonical_nll": 1.2139390481484902, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "88f331fe7934b488", "reading_canonical_log_prob": -89.83148956298828, "reading_distances": [15, 10, 13, 16], "reading_log_probs": [-91.48065185546875, -92.88520812988281, -91.77847290039062, -92.64724731445312], "reading_logprob_std": 0.6744552041046388, "tokenization_path_logprob_std": 97.62833897553746, "trajectory_balance_rmse": 1489.4067390953203} {"canonical_length": 33, "canonical_nll": 0.6797325249874231, "generation_rouge_l_f1": 0.24590163934426232, "generation_token_f1": 0.3770491803278688, "prompt_id": "8b4c34e2f1db7af4", "reading_canonical_log_prob": -22.43117332458496, "reading_distances": [18, 17, 20, 21], "reading_log_probs": [-21.307405471801758, -21.233732223510742, -23.45065689086914, -23.135263442993164], "reading_logprob_std": 1.175091065338754, "tokenization_path_logprob_std": 61.11427536838884, "trajectory_balance_rmse": 844.0638022007186} {"canonical_length": 155, "canonical_nll": 2.0442621046496976, "generation_rouge_l_f1": 0.10722610722610723, "generation_token_f1": 0.1678321678321678, "prompt_id": "7a99289ed340c72d", "reading_canonical_log_prob": -316.8606262207031, "reading_distances": [13, 21, 17, 17], "reading_log_probs": [-312.9859619140625, -313.42742919921875, -319.08489990234375, -317.0125732421875], "reading_logprob_std": 2.9263217396658883, "tokenization_path_logprob_std": 79.8977308697131, "trajectory_balance_rmse": 2459.0820007832695} {"canonical_length": 21, "canonical_nll": 1.224790300641741, "generation_rouge_l_f1": 0.380952380952381, "generation_token_f1": 0.4285714285714286, "prompt_id": "321ea8e8622b4791", "reading_canonical_log_prob": -25.720596313476562, "reading_distances": [26, 33, 35, 27], "reading_log_probs": [-21.886180877685547, -35.68284606933594, -36.34856033325195, -32.08246994018555], "reading_logprob_std": 6.677562626485896, "tokenization_path_logprob_std": 63.62395987585002, "trajectory_balance_rmse": 464.79776768067956} {"canonical_length": 176, "canonical_nll": 1.8915415677157315, "generation_rouge_l_f1": 0.19277108433734938, "generation_token_f1": 0.4289156626506025, "prompt_id": "e9ba47f9f768bdc0", "reading_canonical_log_prob": -332.91131591796875, "reading_distances": [25, 29, 29, 20], "reading_log_probs": [-359.93670654296875, -360.34320068359375, -355.9232177734375, -352.00982666015625], "reading_logprob_std": 3.909441091896836, "tokenization_path_logprob_std": 150.45561594691543, "trajectory_balance_rmse": 4508.161637351647} {"canonical_length": 7, "canonical_nll": 0.8350035122462681, "generation_rouge_l_f1": 0.05853658536585366, "generation_token_f1": 0.05853658536585366, "prompt_id": "cd4e165565532ef7", "reading_canonical_log_prob": -5.845024585723877, "reading_distances": [31, 32, 36, 37], "reading_log_probs": [-9.56353759765625, -13.218168258666992, -10.975590705871582, -7.7035040855407715], "reading_logprob_std": 2.326606430487415, "tokenization_path_logprob_std": 14.596286877444081, "trajectory_balance_rmse": 156.4007261592743} {"canonical_length": 270, "canonical_nll": 0.9826551649305556, "generation_rouge_l_f1": 0.29225352112676056, "generation_token_f1": 0.5422535211267606, "prompt_id": "df9cf174d34cff3e", "reading_canonical_log_prob": -265.31689453125, "reading_distances": [32, 36, 42, 30], "reading_log_probs": [-269.70440673828125, -271.25860595703125, -272.1864013671875, -268.67486572265625], "reading_logprob_std": 1.5679920215832686, "tokenization_path_logprob_std": 176.5947241473597, "trajectory_balance_rmse": 4134.816752281963} {"canonical_length": 143, "canonical_nll": 1.5959461985768137, "generation_rouge_l_f1": 0.43729903536977494, "generation_token_f1": 0.5466237942122187, "prompt_id": "1fba2562017bea7b", "reading_canonical_log_prob": -228.22030639648438, "reading_distances": [29, 34, 33, 29], "reading_log_probs": [-222.89883422851562, -232.5050506591797, -225.75985717773438, -226.12008666992188], "reading_logprob_std": 4.054179728515533, "tokenization_path_logprob_std": 57.142286394379, "trajectory_balance_rmse": 3081.2933058099516} {"canonical_length": 28, "canonical_nll": 0.9793647357395717, "generation_rouge_l_f1": 0.23529411764705885, "generation_token_f1": 0.2823529411764706, "prompt_id": "56997da805bc232c", "reading_canonical_log_prob": -27.422212600708008, "reading_distances": [25, 26, 29, 36], "reading_log_probs": [-26.239730834960938, -23.971525192260742, -24.42751693725586, -26.43392562866211], "reading_logprob_std": 1.2504535138686335, "tokenization_path_logprob_std": 17.212008568650823, "trajectory_balance_rmse": 597.6815444932103} {"canonical_length": 110, "canonical_nll": 1.5965897993607954, "generation_rouge_l_f1": 0.4385026737967914, "generation_token_f1": 0.5454545454545455, "prompt_id": "02d66047fd43a021", "reading_canonical_log_prob": -175.6248779296875, "reading_distances": [22, 19, 16, 18], "reading_log_probs": [-178.89089965820312, -173.7450714111328, -180.3872528076172, -177.5194549560547], "reading_logprob_std": 2.845878467664038, "tokenization_path_logprob_std": 100.51537468829783, "trajectory_balance_rmse": 2188.8477508501132} {"canonical_length": 79, "canonical_nll": 1.5108573044402689, "generation_rouge_l_f1": 0.1329479768786127, "generation_token_f1": 0.20231213872832368, "prompt_id": "83a9bce5dad053f6", "reading_canonical_log_prob": -119.35772705078125, "reading_distances": [20, 21, 22, 24], "reading_log_probs": [-147.2223358154297, -150.84432983398438, -151.84555053710938, -148.53048706054688], "reading_logprob_std": 2.112466406902468, "tokenization_path_logprob_std": 114.6678821136939, "trajectory_balance_rmse": 1432.989997729893} {"canonical_length": 21, "canonical_nll": 6.2388494582403276, "generation_rouge_l_f1": 0.02390438247011952, "generation_token_f1": 0.02390438247011952, "prompt_id": "35f4df3b6373154b", "reading_canonical_log_prob": -131.01583862304688, "reading_distances": [20, 17, 23, 20], "reading_log_probs": [-133.0187225341797, -125.10787200927734, -129.4127197265625, -124.69119262695312], "reading_logprob_std": 3.936270658468692, "tokenization_path_logprob_std": 22.000066876279366, "trajectory_balance_rmse": 239.45502060464096} {"canonical_length": 228, "canonical_nll": 1.2164507414165295, "generation_rouge_l_f1": 0.19999999999999998, "generation_token_f1": 0.43529411764705883, "prompt_id": "7c0757fa3b0cff4b", "reading_canonical_log_prob": -277.35076904296875, "reading_distances": [21, 29, 32, 26], "reading_log_probs": [-273.2645568847656, -273.50927734375, -275.2358093261719, -273.6033935546875], "reading_logprob_std": 0.8997724887509592, "tokenization_path_logprob_std": 217.66484661469258, "trajectory_balance_rmse": 3088.07779520301} {"canonical_length": 211, "canonical_nll": 1.7017113563573756, "generation_rouge_l_f1": 0.25165562913907286, "generation_token_f1": 0.49448123620309054, "prompt_id": "5fa3ea0df830456f", "reading_canonical_log_prob": -359.06109619140625, "reading_distances": [20, 27, 23, 19], "reading_log_probs": [-364.9305419921875, -362.61273193359375, -362.3762512207031, -369.55499267578125], "reading_logprob_std": 3.3300078222748906, "tokenization_path_logprob_std": 139.0129022621403, "trajectory_balance_rmse": 4635.472946961422} {"canonical_length": 145, "canonical_nll": 1.1993761786099137, "generation_rouge_l_f1": 0.3404255319148936, "generation_token_f1": 0.4302600472813239, "prompt_id": "70669892361b488c", "reading_canonical_log_prob": -173.9095458984375, "reading_distances": [35, 36, 29, 32], "reading_log_probs": [-189.3448486328125, -194.01028442382812, -191.15176391601562, -198.9019775390625], "reading_logprob_std": 4.168688434043327, "tokenization_path_logprob_std": 135.3600696130654, "trajectory_balance_rmse": 2495.748908007007} {"canonical_length": 12, "canonical_nll": 1.041111707687378, "generation_rouge_l_f1": 0.7272727272727273, "generation_token_f1": 0.9090909090909091, "prompt_id": "e90b400c44335d9a", "reading_canonical_log_prob": -12.493340492248535, "reading_distances": [19, 18, 21, 25], "reading_log_probs": [-15.076462745666504, -14.00628662109375, -16.59127426147461, -18.17074966430664], "reading_logprob_std": 1.815080312462258, "tokenization_path_logprob_std": 35.60589548782557, "trajectory_balance_rmse": 191.93108568883085} {"canonical_length": 232, "canonical_nll": 1.409936049888874, "generation_rouge_l_f1": 0.3130081300813008, "generation_token_f1": 0.540650406504065, "prompt_id": "9f0a3bc3aaf621a1", "reading_canonical_log_prob": -327.10516357421875, "reading_distances": [20, 31, 25, 26], "reading_log_probs": [-327.915283203125, -327.863037109375, -325.67779541015625, -328.9070739746094], "reading_logprob_std": 1.3627871110096146, "tokenization_path_logprob_std": 318.353590042071, "trajectory_balance_rmse": 5328.337434312005} {"canonical_length": 93, "canonical_nll": 0.5373414562594506, "generation_rouge_l_f1": 0.3647798742138365, "generation_token_f1": 0.5408805031446541, "prompt_id": "f2be334fa1e04283", "reading_canonical_log_prob": -49.972755432128906, "reading_distances": [25, 15, 23, 24], "reading_log_probs": [-50.26274490356445, -53.052330017089844, -52.58436584472656, -54.002838134765625], "reading_logprob_std": 1.5888813167370723, "tokenization_path_logprob_std": 28.125578806426983, "trajectory_balance_rmse": 1675.6904354412782} {"canonical_length": 54, "canonical_nll": 1.5391529224537037, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "1a42eb844c9543a9", "reading_canonical_log_prob": -83.1142578125, "reading_distances": [45, 31, 40, 36], "reading_log_probs": [-84.1326675415039, -83.49754333496094, -82.7173080444336, -84.99696350097656], "reading_logprob_std": 0.9664147536396863, "tokenization_path_logprob_std": 37.13205299887194, "trajectory_balance_rmse": 1061.0616868769746} {"canonical_length": 40, "canonical_nll": 0.538700008392334, "generation_rouge_l_f1": 0.11726384364820848, "generation_token_f1": 0.1758957654723127, "prompt_id": "3afed29700596cb5", "reading_canonical_log_prob": -21.54800033569336, "reading_distances": [11, 9, 16, 14], "reading_log_probs": [-24.444419860839844, -27.12662696838379, -25.80370330810547, -25.341896057128906], "reading_logprob_std": 1.1178851817249416, "tokenization_path_logprob_std": 43.76350621175208, "trajectory_balance_rmse": 438.8358979554244} {"canonical_length": 28, "canonical_nll": 0.7047033309936523, "generation_rouge_l_f1": 0.3928571428571428, "generation_token_f1": 0.5357142857142857, "prompt_id": "e24f8c298fcc6b27", "reading_canonical_log_prob": -19.731693267822266, "reading_distances": [13, 12, 13, 12], "reading_log_probs": [-19.310823440551758, -21.716495513916016, -21.13595199584961, -20.92420768737793], "reading_logprob_std": 1.0300059678004572, "tokenization_path_logprob_std": 16.997954893193878, "trajectory_balance_rmse": 561.1670877545465} {"canonical_length": 232, "canonical_nll": 1.5596142472891972, "generation_rouge_l_f1": 0.20907297830374755, "generation_token_f1": 0.44181459566074954, "prompt_id": "d2492eb6d00f3204", "reading_canonical_log_prob": -361.83050537109375, "reading_distances": [31, 27, 20, 34], "reading_log_probs": [-377.79132080078125, -372.9911193847656, -375.6042175292969, -378.71710205078125], "reading_logprob_std": 2.5493076629722546, "tokenization_path_logprob_std": 177.31155788845675, "trajectory_balance_rmse": 5972.357219657684} {"canonical_length": 83, "canonical_nll": 1.5096077976456608, "generation_rouge_l_f1": 0.22857142857142856, "generation_token_f1": 0.35555555555555557, "prompt_id": "d1d32462252d2ba1", "reading_canonical_log_prob": -125.29744720458984, "reading_distances": [18, 20, 14, 14], "reading_log_probs": [-121.448486328125, -121.58560180664062, -121.43827056884766, -121.62442016601562], "reading_logprob_std": 0.09474648748154121, "tokenization_path_logprob_std": 155.85458537322776, "trajectory_balance_rmse": 1661.5240690366709} {"canonical_length": 171, "canonical_nll": 1.3224282069512976, "generation_rouge_l_f1": 0.24311926605504589, "generation_token_f1": 0.41743119266055045, "prompt_id": "981d7ddcc229e43f", "reading_canonical_log_prob": -226.13522338867188, "reading_distances": [22, 19, 20, 21], "reading_log_probs": [-214.64508056640625, -225.3516387939453, -221.05062866210938, -215.33181762695312], "reading_logprob_std": 5.064034518912628, "tokenization_path_logprob_std": 101.7174232332657, "trajectory_balance_rmse": 3152.635293248071} {"canonical_length": 14, "canonical_nll": 1.760364259992327, "generation_rouge_l_f1": 0.18867924528301888, "generation_token_f1": 0.30188679245283023, "prompt_id": "a353a5c04a14d88b", "reading_canonical_log_prob": -24.645099639892578, "reading_distances": [35, 31, 31, 35], "reading_log_probs": [-20.231250762939453, -21.718008041381836, -22.62590789794922, -21.708370208740234], "reading_logprob_std": 0.9913358960223595, "tokenization_path_logprob_std": 49.73032897808985, "trajectory_balance_rmse": 329.89818510107665} {"canonical_length": 37, "canonical_nll": 0.6388453921756229, "generation_rouge_l_f1": 0.26126126126126126, "generation_token_f1": 0.2702702702702703, "prompt_id": "104ca723f309949c", "reading_canonical_log_prob": -23.637279510498047, "reading_distances": [17, 15, 21, 14], "reading_log_probs": [-26.581438064575195, -26.666414260864258, -26.775440216064453, -26.239646911621094], "reading_logprob_std": 0.23143942247911553, "tokenization_path_logprob_std": 72.47036576374376, "trajectory_balance_rmse": 576.2663249266169} {"canonical_length": 270, "canonical_nll": 0.8968722873263889, "generation_rouge_l_f1": 0.3070362473347548, "generation_token_f1": 0.4690831556503198, "prompt_id": "5768cefb5b534d32", "reading_canonical_log_prob": -242.155517578125, "reading_distances": [22, 20, 25, 20], "reading_log_probs": [-241.911376953125, -240.42886352539062, -241.5691680908203, -243.49188232421875], "reading_logprob_std": 1.2646527482585883, "tokenization_path_logprob_std": 219.7711640591679, "trajectory_balance_rmse": 4373.196223881381} {"canonical_length": 460, "canonical_nll": 0.7538102523140285, "generation_rouge_l_f1": 0.34210526315789475, "generation_token_f1": 0.5263157894736843, "prompt_id": "2c20880d9a9f5257", "reading_canonical_log_prob": -346.7527160644531, "reading_distances": [31, 27, 35, 21], "reading_log_probs": [-345.388427734375, -351.0662536621094, -351.2027282714844, -346.12847900390625], "reading_logprob_std": 3.119023931279518, "tokenization_path_logprob_std": 187.61124132518538, "trajectory_balance_rmse": 8048.920883066553} {"canonical_length": 43, "canonical_nll": 1.295146764710892, "generation_rouge_l_f1": 0.2809917355371901, "generation_token_f1": 0.4132231404958678, "prompt_id": "1a2898fe9876d2bc", "reading_canonical_log_prob": -55.69131088256836, "reading_distances": [19, 17, 12, 16], "reading_log_probs": [-60.98249053955078, -65.38202667236328, -71.73664855957031, -70.10774230957031], "reading_logprob_std": 4.861798329443202, "tokenization_path_logprob_std": 66.75801182723268, "trajectory_balance_rmse": 822.9029710945059} {"canonical_length": 85, "canonical_nll": 1.011567777745864, "generation_rouge_l_f1": 0.4878048780487805, "generation_token_f1": 0.646341463414634, "prompt_id": "52568183284a12f5", "reading_canonical_log_prob": -85.98326110839844, "reading_distances": [17, 20, 15, 17], "reading_log_probs": [-88.63740539550781, -88.30722045898438, -92.47531127929688, -93.59648132324219], "reading_logprob_std": 2.677642955507499, "tokenization_path_logprob_std": 117.25132798317617, "trajectory_balance_rmse": 1824.9256279488263} {"canonical_length": 202, "canonical_nll": 1.667326067933942, "generation_rouge_l_f1": 0.2554347826086957, "generation_token_f1": 0.5652173913043478, "prompt_id": "ea255680cda4b618", "reading_canonical_log_prob": -336.79986572265625, "reading_distances": [29, 17, 25, 26], "reading_log_probs": [-339.33953857421875, -336.3436584472656, -336.60699462890625, -345.46832275390625], "reading_logprob_std": 4.241226930005858, "tokenization_path_logprob_std": 267.15356317158086, "trajectory_balance_rmse": 4451.943212649789} {"canonical_length": 19, "canonical_nll": 1.1167055431165194, "generation_rouge_l_f1": 0.18750000000000003, "generation_token_f1": 0.3125, "prompt_id": "559aadd3a6bc5deb", "reading_canonical_log_prob": -21.217405319213867, "reading_distances": [32, 35, 35, 39], "reading_log_probs": [-25.804523468017578, -26.735431671142578, -26.912317276000977, -25.3803653717041], "reading_logprob_std": 0.7353047624669469, "tokenization_path_logprob_std": 74.05510616300475, "trajectory_balance_rmse": 434.4750369282803} {"canonical_length": 24, "canonical_nll": 0.9988199869791666, "generation_rouge_l_f1": 0.3283582089552239, "generation_token_f1": 0.44776119402985076, "prompt_id": "4196e10bb50daa8f", "reading_canonical_log_prob": -23.9716796875, "reading_distances": [25, 23, 20, 18], "reading_log_probs": [-25.234403610229492, -25.793750762939453, -30.671287536621094, -27.45877456665039], "reading_logprob_std": 2.444437751403508, "tokenization_path_logprob_std": 20.13190978787891, "trajectory_balance_rmse": 486.0583302816336} {"canonical_length": 293, "canonical_nll": 1.339147053480962, "generation_rouge_l_f1": 0.2401500938086304, "generation_token_f1": 0.5140712945590995, "prompt_id": "0d2dbbf4ae30ce19", "reading_canonical_log_prob": -392.3700866699219, "reading_distances": [18, 13, 20, 21], "reading_log_probs": [-404.3046875, -397.9330749511719, -392.713623046875, -396.34881591796875], "reading_logprob_std": 4.840913277594246, "tokenization_path_logprob_std": 125.84206738194517, "trajectory_balance_rmse": 6887.638732611314} {"canonical_length": 227, "canonical_nll": 1.0829617000361371, "generation_rouge_l_f1": 0.28832951945080093, "generation_token_f1": 0.5034324942791761, "prompt_id": "cd0a37baf57f0d68", "reading_canonical_log_prob": -245.83230590820312, "reading_distances": [23, 29, 26, 27], "reading_log_probs": [-253.9263916015625, -256.24578857421875, -253.2523193359375, -252.69418334960938], "reading_logprob_std": 1.5609448645899375, "tokenization_path_logprob_std": 284.6846727261882, "trajectory_balance_rmse": 4466.052096644015} {"canonical_length": 27, "canonical_nll": 1.6618982950846355, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.2857142857142857, "prompt_id": "f0dd5c7b12d6ff90", "reading_canonical_log_prob": -44.871253967285156, "reading_distances": [23, 23, 19, 24], "reading_log_probs": [-54.55548095703125, -53.36397933959961, -53.91365051269531, -52.41978454589844], "reading_logprob_std": 0.9045302334123049, "tokenization_path_logprob_std": 83.59059670250213, "trajectory_balance_rmse": 527.1817214510031} {"canonical_length": 29, "canonical_nll": 0.9300013575060614, "generation_rouge_l_f1": 0.5263157894736842, "generation_token_f1": 0.5263157894736842, "prompt_id": "c737e978a37038bd", "reading_canonical_log_prob": -26.97003936767578, "reading_distances": [23, 22, 27, 26], "reading_log_probs": [-29.022411346435547, -28.453235626220703, -32.638343811035156, -28.288650512695312], "reading_logprob_std": 2.049377631501336, "tokenization_path_logprob_std": 66.75626084054723, "trajectory_balance_rmse": 457.4732527639439} {"canonical_length": 90, "canonical_nll": 1.414278835720486, "generation_rouge_l_f1": 0.3439490445859873, "generation_token_f1": 0.535031847133758, "prompt_id": "6fe0c1612b081574", "reading_canonical_log_prob": -127.28509521484375, "reading_distances": [20, 25, 24, 27], "reading_log_probs": [-126.31346130371094, -135.4749755859375, -129.96485900878906, -126.58177185058594], "reading_logprob_std": 4.264522010982897, "tokenization_path_logprob_std": 201.80197385523664, "trajectory_balance_rmse": 2091.658805113953} {"canonical_length": 276, "canonical_nll": 1.3486012997834578, "generation_rouge_l_f1": 0.2631578947368421, "generation_token_f1": 0.5020242914979757, "prompt_id": "a45d6c8378dedb7b", "reading_canonical_log_prob": -372.2139587402344, "reading_distances": [13, 14, 15, 14], "reading_log_probs": [-375.36199951171875, -371.62786865234375, -376.64190673828125, -380.4620666503906], "reading_logprob_std": 3.6442858760041386, "tokenization_path_logprob_std": 197.02482000109666, "trajectory_balance_rmse": 5926.438771577069} {"canonical_length": 103, "canonical_nll": 1.1661217698773134, "generation_rouge_l_f1": 0.23978201634877386, "generation_token_f1": 0.34877384196185285, "prompt_id": "2e7570c8d7724f9e", "reading_canonical_log_prob": -120.11054229736328, "reading_distances": [25, 22, 17, 25], "reading_log_probs": [-127.96102142333984, -130.5218048095703, -128.525390625, -130.81607055664062], "reading_logprob_std": 1.4243977141237492, "tokenization_path_logprob_std": 77.17918934357871, "trajectory_balance_rmse": 1490.2533493257024} {"canonical_length": 30, "canonical_nll": 2.06608034769694, "generation_rouge_l_f1": 0.12631578947368421, "generation_token_f1": 0.1473684210526316, "prompt_id": "7aa76ccd6b778b98", "reading_canonical_log_prob": -61.9824104309082, "reading_distances": [40, 36, 27, 35], "reading_log_probs": [-59.96065139770508, -59.91017532348633, -61.048946380615234, -61.200172424316406], "reading_logprob_std": 0.6896319994693763, "tokenization_path_logprob_std": 92.58919089106774, "trajectory_balance_rmse": 719.8592795629248} {"canonical_length": 252, "canonical_nll": 1.543016585092696, "generation_rouge_l_f1": 0.24379232505643342, "generation_token_f1": 0.38374717832957106, "prompt_id": "0667a89ba9993054", "reading_canonical_log_prob": -388.8401794433594, "reading_distances": [28, 27, 25, 26], "reading_log_probs": [-394.9930114746094, -390.34759521484375, -395.2584228515625, -392.7796936035156], "reading_logprob_std": 2.2862731821109725, "tokenization_path_logprob_std": 123.71518144393531, "trajectory_balance_rmse": 5008.1092074075} {"canonical_length": 248, "canonical_nll": 1.1653332864084551, "generation_rouge_l_f1": 0.2664015904572564, "generation_token_f1": 0.5646123260437377, "prompt_id": "45c2d1c36166f6b6", "reading_canonical_log_prob": -289.0026550292969, "reading_distances": [24, 24, 22, 32], "reading_log_probs": [-294.74468994140625, -297.040283203125, -298.0632629394531, -295.52862548828125], "reading_logprob_std": 1.490335915851865, "tokenization_path_logprob_std": 60.796018019425254, "trajectory_balance_rmse": 4380.721883569176} {"canonical_length": 20, "canonical_nll": 1.3397778511047362, "generation_rouge_l_f1": 0.5454545454545454, "generation_token_f1": 0.7272727272727272, "prompt_id": "e5f7552dff198384", "reading_canonical_log_prob": -26.795557022094727, "reading_distances": [29, 20, 21, 27], "reading_log_probs": [-26.719961166381836, -26.155683517456055, -27.10870361328125, -27.734046936035156], "reading_logprob_std": 0.6638544367793908, "tokenization_path_logprob_std": 85.1524935972831, "trajectory_balance_rmse": 463.02130294977246} {"canonical_length": 117, "canonical_nll": 1.6637748000968215, "generation_rouge_l_f1": 0.2058823529411765, "generation_token_f1": 0.3725490196078432, "prompt_id": "4a930bccec9843cd", "reading_canonical_log_prob": -194.66165161132812, "reading_distances": [34, 32, 26, 27], "reading_log_probs": [-204.55682373046875, -197.06298828125, -197.48837280273438, -208.15093994140625], "reading_logprob_std": 5.4455807892826105, "tokenization_path_logprob_std": 205.7625378017723, "trajectory_balance_rmse": 2787.5818162970772} {"canonical_length": 49, "canonical_nll": 0.9345281951281489, "generation_rouge_l_f1": 0.2626262626262626, "generation_token_f1": 0.3838383838383838, "prompt_id": "35925631dfc295f8", "reading_canonical_log_prob": -45.7918815612793, "reading_distances": [33, 29, 36, 28], "reading_log_probs": [-56.176231384277344, -54.443138122558594, -56.781089782714844, -52.78596878051758], "reading_logprob_std": 1.803620582653425, "tokenization_path_logprob_std": 156.99955132882405, "trajectory_balance_rmse": 982.9062865835592} {"canonical_length": 295, "canonical_nll": 1.169253798662606, "generation_rouge_l_f1": 0.26807760141093484, "generation_token_f1": 0.5185185185185185, "prompt_id": "f6597a51ed7cc783", "reading_canonical_log_prob": -344.92987060546875, "reading_distances": [21, 21, 19, 23], "reading_log_probs": [-359.41400146484375, -353.1893005371094, -356.4183349609375, -353.5068054199219], "reading_logprob_std": 2.910045267182099, "tokenization_path_logprob_std": 258.7543389211125, "trajectory_balance_rmse": 5814.5812162171715} {"canonical_length": 213, "canonical_nll": 1.225994844391872, "generation_rouge_l_f1": 0.22881355932203393, "generation_token_f1": 0.4406779661016949, "prompt_id": "87270c0f1888627a", "reading_canonical_log_prob": -261.13690185546875, "reading_distances": [13, 15, 9, 10], "reading_log_probs": [-256.70465087890625, -261.8726501464844, -257.7978210449219, -260.1383056640625], "reading_logprob_std": 2.323490080824225, "tokenization_path_logprob_std": 367.10046460647385, "trajectory_balance_rmse": 4904.329335146753} {"canonical_length": 14, "canonical_nll": 1.7994296210152763, "generation_rouge_l_f1": 0.15384615384615385, "generation_token_f1": 0.15384615384615385, "prompt_id": "a53e27e32808bf25", "reading_canonical_log_prob": -25.192014694213867, "reading_distances": [25, 20, 27, 27], "reading_log_probs": [-25.036354064941406, -25.5306339263916, -24.41357421875, -25.109817504882812], "reading_logprob_std": 0.46072657833381875, "tokenization_path_logprob_std": 23.575951026300192, "trajectory_balance_rmse": 270.35633683515226} {"canonical_length": 128, "canonical_nll": 1.1767538785934448, "generation_rouge_l_f1": 0.23896103896103896, "generation_token_f1": 0.43636363636363634, "prompt_id": "1e1844cf1025dc97", "reading_canonical_log_prob": -150.62449645996094, "reading_distances": [15, 16, 18, 16], "reading_log_probs": [-156.2438201904297, -156.83033752441406, -157.33119201660156, -161.61849975585938], "reading_logprob_std": 2.449012061535352, "tokenization_path_logprob_std": 177.42147223710896, "trajectory_balance_rmse": 2935.6096008128775} {"canonical_length": 155, "canonical_nll": 1.5611719931325605, "generation_rouge_l_f1": 0.21333333333333332, "generation_token_f1": 0.37866666666666665, "prompt_id": "87981277aeac45ee", "reading_canonical_log_prob": -241.98165893554688, "reading_distances": [21, 21, 23, 18], "reading_log_probs": [-247.3105010986328, -243.0272979736328, -246.19677734375, -247.03750610351562], "reading_logprob_std": 1.9683946388201894, "tokenization_path_logprob_std": 122.07433203961644, "trajectory_balance_rmse": 3098.895632281101} {"canonical_length": 26, "canonical_nll": 0.18903614924504206, "generation_rouge_l_f1": 0.17674418604651163, "generation_token_f1": 0.17674418604651163, "prompt_id": "40b80040395896ae", "reading_canonical_log_prob": -4.914939880371094, "reading_distances": [39, 46, 45, 47], "reading_log_probs": [-4.572225570678711, -5.929042816162109, -5.324554443359375, -4.900821208953857], "reading_logprob_std": 0.5857418788828792, "tokenization_path_logprob_std": 103.44981284740432, "trajectory_balance_rmse": 306.3083021548979} {"canonical_length": 356, "canonical_nll": 1.5343576495567064, "generation_rouge_l_f1": 0.22553897180762852, "generation_token_f1": 0.4709784411276949, "prompt_id": "63d17b19976fa874", "reading_canonical_log_prob": -546.2313232421875, "reading_distances": [29, 29, 27, 20], "reading_log_probs": [-559.8518676757812, -555.6793212890625, -557.5709228515625, -553.4818115234375], "reading_logprob_std": 2.7129090339758344, "tokenization_path_logprob_std": 262.0540719199768, "trajectory_balance_rmse": 8277.194369171419} {"canonical_length": 19, "canonical_nll": 0.5544486798738179, "generation_rouge_l_f1": 0.1217391304347826, "generation_token_f1": 0.1217391304347826, "prompt_id": "53bd5d756df134c2", "reading_canonical_log_prob": -10.534524917602539, "reading_distances": [27, 21, 25, 25], "reading_log_probs": [-5.132636070251465, -5.616946220397949, -5.887775897979736, -5.919641494750977], "reading_logprob_std": 0.3640236460703648, "tokenization_path_logprob_std": 57.14330304872518, "trajectory_balance_rmse": 212.49561299974215} {"canonical_length": 114, "canonical_nll": 1.0456813343784266, "generation_rouge_l_f1": 0.18857142857142858, "generation_token_f1": 0.3485714285714286, "prompt_id": "1cf0a058de57af94", "reading_canonical_log_prob": -119.20767211914062, "reading_distances": [38, 28, 33, 36], "reading_log_probs": [-131.50161743164062, -118.37771606445312, -126.7669448852539, -132.30288696289062], "reading_logprob_std": 6.3916100183774605, "tokenization_path_logprob_std": 83.29909543843026, "trajectory_balance_rmse": 1519.5016601646366} {"canonical_length": 257, "canonical_nll": 1.4660542410171449, "generation_rouge_l_f1": 0.1904761904761905, "generation_token_f1": 0.43478260869565216, "prompt_id": "a5e93d67499f0604", "reading_canonical_log_prob": -376.77593994140625, "reading_distances": [15, 19, 20, 19], "reading_log_probs": [-378.2296142578125, -378.52557373046875, -382.840087890625, -382.57464599609375], "reading_logprob_std": 2.505058848024672, "tokenization_path_logprob_std": 243.13951857560463, "trajectory_balance_rmse": 4649.095418544854} {"canonical_length": 118, "canonical_nll": 1.3174068644895391, "generation_rouge_l_f1": 0.3571428571428571, "generation_token_f1": 0.48412698412698413, "prompt_id": "9493ef83593c3f2d", "reading_canonical_log_prob": -155.45401000976562, "reading_distances": [32, 36, 37, 32], "reading_log_probs": [-165.2293701171875, -160.3260955810547, -165.48623657226562, -161.20510864257812], "reading_logprob_std": 2.677538317969086, "tokenization_path_logprob_std": 102.49411173677726, "trajectory_balance_rmse": 2638.6941833338947} {"canonical_length": 22, "canonical_nll": 2.1381603587757456, "generation_rouge_l_f1": 0.21428571428571427, "generation_token_f1": 0.21428571428571427, "prompt_id": "be453082a5e24aed", "reading_canonical_log_prob": -47.039527893066406, "reading_distances": [35, 24, 25, 37], "reading_log_probs": [-41.93138122558594, -42.73185729980469, -41.762657165527344, -47.7342529296875], "reading_logprob_std": 2.8279235120018025, "tokenization_path_logprob_std": 35.939104899863736, "trajectory_balance_rmse": 413.49330891119615} {"canonical_length": 280, "canonical_nll": 1.119319806780134, "generation_rouge_l_f1": 0.34892086330935246, "generation_token_f1": 0.5683453237410071, "prompt_id": "6b6295e1afd98591", "reading_canonical_log_prob": -313.4095458984375, "reading_distances": [14, 17, 17, 20], "reading_log_probs": [-316.63494873046875, -317.6562805175781, -320.68487548828125, -316.8618469238281], "reading_logprob_std": 1.8689472925782482, "tokenization_path_logprob_std": 309.7680650303401, "trajectory_balance_rmse": 6066.7825597721085} {"canonical_length": 200, "canonical_nll": 1.253603973388672, "generation_rouge_l_f1": 0.2844827586206896, "generation_token_f1": 0.5043103448275862, "prompt_id": "900ff807cdf33dd9", "reading_canonical_log_prob": -250.72079467773438, "reading_distances": [23, 20, 27, 21], "reading_log_probs": [-257.53875732421875, -257.5404968261719, -261.5161437988281, -250.94073486328125], "reading_logprob_std": 4.383257992648104, "tokenization_path_logprob_std": 221.01772970588794, "trajectory_balance_rmse": 3688.00473665084} {"canonical_length": 120, "canonical_nll": 1.5241629282633464, "generation_rouge_l_f1": 0.23140495867768596, "generation_token_f1": 0.39118457300275483, "prompt_id": "6b5f709d21a7a023", "reading_canonical_log_prob": -182.89955139160156, "reading_distances": [13, 19, 16, 18], "reading_log_probs": [-185.30609130859375, -182.760986328125, -182.45773315429688, -184.0924072265625], "reading_logprob_std": 1.310233312763507, "tokenization_path_logprob_std": 56.677328212332654, "trajectory_balance_rmse": 2381.514689322541} {"canonical_length": 52, "canonical_nll": 0.8888505055354192, "generation_rouge_l_f1": 0.15894039735099336, "generation_token_f1": 0.23841059602649006, "prompt_id": "e072285f618dcbd9", "reading_canonical_log_prob": -46.2202262878418, "reading_distances": [28, 34, 21, 27], "reading_log_probs": [-48.35102081298828, -47.38457489013672, -47.87305450439453, -47.4222297668457], "reading_logprob_std": 0.45354120435865297, "tokenization_path_logprob_std": 11.431911222797606, "trajectory_balance_rmse": 1133.7326446246188} {"canonical_length": 147, "canonical_nll": 1.463953446368782, "generation_rouge_l_f1": 0.16625916870415647, "generation_token_f1": 0.30806845965770174, "prompt_id": "21874077bcbf7b19", "reading_canonical_log_prob": -215.20115661621094, "reading_distances": [27, 20, 22, 23], "reading_log_probs": [-213.16421508789062, -215.96185302734375, -215.26986694335938, -216.14309692382812], "reading_logprob_std": 1.3665195893953617, "tokenization_path_logprob_std": 130.8130524065936, "trajectory_balance_rmse": 4082.767871039697} {"canonical_length": 108, "canonical_nll": 1.1453503502739801, "generation_rouge_l_f1": 0.21546961325966849, "generation_token_f1": 0.39779005524861877, "prompt_id": "8ba54ff004576b32", "reading_canonical_log_prob": -123.69783782958984, "reading_distances": [20, 23, 22, 21], "reading_log_probs": [-124.01066589355469, -123.40670776367188, -124.16252899169922, -123.64823913574219], "reading_logprob_std": 0.34318097344055876, "tokenization_path_logprob_std": 28.482576897216788, "trajectory_balance_rmse": 2700.2890507299635} {"canonical_length": 8, "canonical_nll": 0.3828320801258087, "generation_rouge_l_f1": 0.030418250950570342, "generation_token_f1": 0.030418250950570342, "prompt_id": "bccef35e52f5dc4b", "reading_canonical_log_prob": -3.0626566410064697, "reading_distances": [34, 34, 34, 34], "reading_log_probs": [-5.279749393463135, -6.3738813400268555, -7.567453384399414, -3.6549670696258545], "reading_logprob_std": 1.6632120045432035, "tokenization_path_logprob_std": 36.006084482474144, "trajectory_balance_rmse": 149.61188392632963} {"canonical_length": 53, "canonical_nll": 0.9870195928609596, "generation_rouge_l_f1": 0.1377049180327869, "generation_token_f1": 0.22295081967213118, "prompt_id": "492f8c42a0de57b2", "reading_canonical_log_prob": -52.31203842163086, "reading_distances": [47, 39, 37, 45], "reading_log_probs": [-53.17278289794922, -58.74184799194336, -55.579246520996094, -57.5674934387207], "reading_logprob_std": 2.440172746854784, "tokenization_path_logprob_std": 115.89161164654622, "trajectory_balance_rmse": 958.3073928491417} {"canonical_length": 211, "canonical_nll": 1.2776301234819314, "generation_rouge_l_f1": 0.21123595505617979, "generation_token_f1": 0.39101123595505616, "prompt_id": "6aa6f6caad65a47f", "reading_canonical_log_prob": -269.5799560546875, "reading_distances": [21, 21, 29, 27], "reading_log_probs": [-303.40252685546875, -298.858154296875, -292.449951171875, -294.21185302734375], "reading_logprob_std": 4.923042778590142, "tokenization_path_logprob_std": 179.78612727018862, "trajectory_balance_rmse": 4033.941258965207} {"canonical_length": 107, "canonical_nll": 1.5175022588712033, "generation_rouge_l_f1": 0.24031007751937983, "generation_token_f1": 0.4108527131782946, "prompt_id": "83dae6df424d2709", "reading_canonical_log_prob": -162.37274169921875, "reading_distances": [31, 32, 29, 25], "reading_log_probs": [-166.46499633789062, -163.51368713378906, -177.21905517578125, -165.79025268554688], "reading_logprob_std": 6.1131870898227385, "tokenization_path_logprob_std": 171.65762045918737, "trajectory_balance_rmse": 2204.1288609381877} {"canonical_length": 104, "canonical_nll": 1.2185645470252404, "generation_rouge_l_f1": 0.19672131147540983, "generation_token_f1": 0.34972677595628415, "prompt_id": "d7aad76280327bd4", "reading_canonical_log_prob": -126.730712890625, "reading_distances": [28, 32, 36, 32], "reading_log_probs": [-130.57501220703125, -131.61883544921875, -130.7520294189453, -130.54995727539062], "reading_logprob_std": 0.5046629402919585, "tokenization_path_logprob_std": 156.81032937598712, "trajectory_balance_rmse": 2457.456690696018} {"canonical_length": 362, "canonical_nll": 1.5589989087858254, "generation_rouge_l_f1": 0.14446227929373995, "generation_token_f1": 0.32423756019261635, "prompt_id": "117db47d424e536b", "reading_canonical_log_prob": -564.3576049804688, "reading_distances": [27, 15, 15, 23], "reading_log_probs": [-584.3198852539062, -580.96728515625, -586.99951171875, -583.326171875], "reading_logprob_std": 2.4975541560064314, "tokenization_path_logprob_std": 294.19344983899606, "trajectory_balance_rmse": 7142.947192104117} {"canonical_length": 207, "canonical_nll": 1.4463226760643115, "generation_rouge_l_f1": 0.21748400852878466, "generation_token_f1": 0.47334754797441364, "prompt_id": "f593668e2032bc26", "reading_canonical_log_prob": -299.3887939453125, "reading_distances": [32, 31, 31, 33], "reading_log_probs": [-307.1547546386719, -304.1673583984375, -311.2171630859375, -308.5782470703125], "reading_logprob_std": 2.937878942034741, "tokenization_path_logprob_std": 221.97939894359888, "trajectory_balance_rmse": 4632.911131133171} {"canonical_length": 243, "canonical_nll": 1.0464398419415508, "generation_rouge_l_f1": 0.28884026258205686, "generation_token_f1": 0.5514223194748359, "prompt_id": "62a2987fac290e0c", "reading_canonical_log_prob": -254.28488159179688, "reading_distances": [13, 16, 18, 14], "reading_log_probs": [-252.73092651367188, -261.0446472167969, -255.24081420898438, -254.50534057617188], "reading_logprob_std": 3.6003930783678166, "tokenization_path_logprob_std": 234.47141208838127, "trajectory_balance_rmse": 5269.455362239289} {"canonical_length": 28, "canonical_nll": 0.8642869676862445, "generation_rouge_l_f1": 0.1411764705882353, "generation_token_f1": 0.16470588235294117, "prompt_id": "0597c4917add4bdd", "reading_canonical_log_prob": -24.200035095214844, "reading_distances": [17, 16, 17, 20], "reading_log_probs": [-23.688817977905273, -27.285865783691406, -25.069807052612305, -29.096113204956055], "reading_logprob_std": 2.3889289558336517, "tokenization_path_logprob_std": 17.857694307525595, "trajectory_balance_rmse": 321.28517996282443} {"canonical_length": 51, "canonical_nll": 1.6526961981081496, "generation_rouge_l_f1": 0.09433962264150944, "generation_token_f1": 0.16352201257861637, "prompt_id": "ea67625f5161437c", "reading_canonical_log_prob": -84.28750610351562, "reading_distances": [19, 21, 20, 20], "reading_log_probs": [-95.88897705078125, -97.1307373046875, -98.53396606445312, -96.45120239257812], "reading_logprob_std": 1.1410067158535901, "tokenization_path_logprob_std": 105.31671319797871, "trajectory_balance_rmse": 996.9193987154131} {"canonical_length": 98, "canonical_nll": 0.9500726972307477, "generation_rouge_l_f1": 0.30625, "generation_token_f1": 0.41250000000000003, "prompt_id": "f9bf7abf6deb13c3", "reading_canonical_log_prob": -93.10712432861328, "reading_distances": [17, 20, 14, 24], "reading_log_probs": [-97.69297790527344, -94.46056365966797, -94.78704833984375, -96.56810760498047], "reading_logprob_std": 1.5242147574148617, "tokenization_path_logprob_std": 82.91315789398547, "trajectory_balance_rmse": 2136.049079608463} {"canonical_length": 144, "canonical_nll": 1.5034221013387044, "generation_rouge_l_f1": 0.28211586901763225, "generation_token_f1": 0.47355163727959704, "prompt_id": "dd61adf3859cf501", "reading_canonical_log_prob": -216.49278259277344, "reading_distances": [40, 41, 39, 36], "reading_log_probs": [-249.11587524414062, -246.12814331054688, -256.4361877441406, -246.37376403808594], "reading_logprob_std": 4.809722544821358, "tokenization_path_logprob_std": 173.68944926989468, "trajectory_balance_rmse": 2464.7820392387503} {"canonical_length": 38, "canonical_nll": 2.5052446064196134, "generation_rouge_l_f1": 0.19696969696969696, "generation_token_f1": 0.33333333333333337, "prompt_id": "932b65e0c79cc6cd", "reading_canonical_log_prob": -95.19929504394531, "reading_distances": [31, 23, 29, 24], "reading_log_probs": [-100.07051849365234, -98.19245910644531, -101.51792907714844, -101.53324890136719], "reading_logprob_std": 1.5806524597367766, "tokenization_path_logprob_std": 50.55594088806873, "trajectory_balance_rmse": 622.9872910407049} {"canonical_length": 246, "canonical_nll": 1.4364518576521215, "generation_rouge_l_f1": 0.21765913757700206, "generation_token_f1": 0.3819301848049281, "prompt_id": "5a38778847ffa7e8", "reading_canonical_log_prob": -353.3671569824219, "reading_distances": [28, 29, 34, 30], "reading_log_probs": [-364.7378234863281, -363.1058654785156, -367.708740234375, -362.9343566894531], "reading_logprob_std": 2.212704515357343, "tokenization_path_logprob_std": 242.49648364850594, "trajectory_balance_rmse": 5702.484692352706} {"canonical_length": 163, "canonical_nll": 0.8726598821534701, "generation_rouge_l_f1": 0.31666666666666665, "generation_token_f1": 0.5111111111111112, "prompt_id": "bea3ed585e39be34", "reading_canonical_log_prob": -142.24356079101562, "reading_distances": [19, 15, 16, 14], "reading_log_probs": [-153.48355102539062, -152.2627716064453, -150.8661346435547, -152.1920166015625], "reading_logprob_std": 1.0693766142382353, "tokenization_path_logprob_std": 300.1475337161013, "trajectory_balance_rmse": 3088.407125829386} {"canonical_length": 256, "canonical_nll": 0.6479513645172119, "generation_rouge_l_f1": 0.32727272727272727, "generation_token_f1": 0.602020202020202, "prompt_id": "e77970903b119833", "reading_canonical_log_prob": -165.87554931640625, "reading_distances": [21, 26, 26, 33], "reading_log_probs": [-171.10386657714844, -173.70594787597656, -177.16213989257812, -183.85379028320312], "reading_logprob_std": 5.520689954527511, "tokenization_path_logprob_std": 153.47284612267092, "trajectory_balance_rmse": 2354.803442612328} {"canonical_length": 28, "canonical_nll": 1.0143534796578544, "generation_rouge_l_f1": 0.09090909090909091, "generation_token_f1": 0.12121212121212122, "prompt_id": "4ea6cdd31bfaf581", "reading_canonical_log_prob": -28.401897430419922, "reading_distances": [24, 17, 19, 19], "reading_log_probs": [-43.98585510253906, -41.12461853027344, -36.5130615234375, -32.150543212890625], "reading_logprob_std": 5.203652386888138, "tokenization_path_logprob_std": 9.511040933727918, "trajectory_balance_rmse": 255.7398613458556} {"canonical_length": 92, "canonical_nll": 0.7255196778670602, "generation_rouge_l_f1": 0.43601895734597146, "generation_token_f1": 0.5876777251184835, "prompt_id": "c7f44052e9719f7b", "reading_canonical_log_prob": -66.74781036376953, "reading_distances": [30, 36, 40, 40], "reading_log_probs": [-69.33727264404297, -69.7576904296875, -68.1507339477539, -68.42899322509766], "reading_logprob_std": 0.754694768151246, "tokenization_path_logprob_std": 132.00527990896387, "trajectory_balance_rmse": 1797.5351724042782} {"canonical_length": 22, "canonical_nll": 1.8729967637495561, "generation_rouge_l_f1": 0.3673469387755102, "generation_token_f1": 0.4897959183673469, "prompt_id": "4d4bec1ba05bf153", "reading_canonical_log_prob": -41.205928802490234, "reading_distances": [41, 31, 45, 38], "reading_log_probs": [-38.942386627197266, -40.52156066894531, -45.5158805847168, -41.56782150268555], "reading_logprob_std": 2.8021210034782396, "tokenization_path_logprob_std": 54.45089043535473, "trajectory_balance_rmse": 666.991813586226} {"canonical_length": 56, "canonical_nll": 1.9184979030064173, "generation_rouge_l_f1": 0.3838383838383838, "generation_token_f1": 0.3838383838383838, "prompt_id": "2dc7b73ea6e1f7f9", "reading_canonical_log_prob": -107.43588256835938, "reading_distances": [45, 47, 42, 49], "reading_log_probs": [-125.8658676147461, -128.89930725097656, -120.68162536621094, -128.46939086914062], "reading_logprob_std": 3.7773543677809043, "tokenization_path_logprob_std": 139.99449540033723, "trajectory_balance_rmse": 1210.7018888404764} {"canonical_length": 14, "canonical_nll": 2.8491532461983815, "generation_rouge_l_f1": 0.03731343283582089, "generation_token_f1": 0.04477611940298507, "prompt_id": "cc99639a6604cf81", "reading_canonical_log_prob": -39.888145446777344, "reading_distances": [37, 35, 37, 33], "reading_log_probs": [-46.01288986206055, -35.85075378417969, -45.78478240966797, -42.27179718017578], "reading_logprob_std": 4.7396607242991955, "tokenization_path_logprob_std": 25.56910685013247, "trajectory_balance_rmse": 296.70065310119463} {"canonical_length": 41, "canonical_nll": 1.0633038776676829, "generation_rouge_l_f1": 0.1889763779527559, "generation_token_f1": 0.2125984251968504, "prompt_id": "34af9dd76e319507", "reading_canonical_log_prob": -43.595458984375, "reading_distances": [29, 35, 37, 35], "reading_log_probs": [-47.606361389160156, -44.51984405517578, -48.87495422363281, -45.649810791015625], "reading_logprob_std": 1.94955907066652, "tokenization_path_logprob_std": 87.45784792054562, "trajectory_balance_rmse": 821.0966904565186} {"canonical_length": 319, "canonical_nll": 1.3765379328712775, "generation_rouge_l_f1": 0.3103448275862069, "generation_token_f1": 0.5379310344827586, "prompt_id": "761d4b8b4ce8e2b9", "reading_canonical_log_prob": -439.1156005859375, "reading_distances": [29, 23, 27, 31], "reading_log_probs": [-444.59765625, -445.3570556640625, -447.2808837890625, -450.4403991699219], "reading_logprob_std": 2.6050923787163116, "tokenization_path_logprob_std": 223.14184863071083, "trajectory_balance_rmse": 7185.807155801198} {"canonical_length": 30, "canonical_nll": 0.6991831461588541, "generation_rouge_l_f1": 0.5714285714285714, "generation_token_f1": 0.6984126984126985, "prompt_id": "667ddd428394372e", "reading_canonical_log_prob": -20.975494384765625, "reading_distances": [32, 33, 31, 34], "reading_log_probs": [-27.84893798828125, -24.398534774780273, -23.577117919921875, -26.29897117614746], "reading_logprob_std": 1.920308036583469, "tokenization_path_logprob_std": 58.06922619968707, "trajectory_balance_rmse": 612.3337081080202} {"canonical_length": 15, "canonical_nll": 0.5307352066040039, "generation_rouge_l_f1": 0.11353711790393015, "generation_token_f1": 0.12227074235807858, "prompt_id": "eaab92bbea496d12", "reading_canonical_log_prob": -7.961028099060059, "reading_distances": [28, 24, 20, 23], "reading_log_probs": [-11.907941818237305, -10.501361846923828, -11.664210319519043, -10.117084503173828], "reading_logprob_std": 0.8726647944095556, "tokenization_path_logprob_std": 50.9583046009483, "trajectory_balance_rmse": 317.68618391921484} {"canonical_length": 77, "canonical_nll": 0.8912206872717127, "generation_rouge_l_f1": 0.6888888888888889, "generation_token_f1": 0.7555555555555555, "prompt_id": "7cb40862b1b15473", "reading_canonical_log_prob": -68.62399291992188, "reading_distances": [33, 25, 25, 25], "reading_log_probs": [-71.24822998046875, -88.74046325683594, -86.09971618652344, -78.36802673339844], "reading_logprob_std": 7.913998039592973, "tokenization_path_logprob_std": 121.33451016621227, "trajectory_balance_rmse": 769.2190984758879} {"canonical_length": 37, "canonical_nll": 1.4717768076303843, "generation_rouge_l_f1": 0.2545454545454545, "generation_token_f1": 0.32727272727272727, "prompt_id": "b218ec9e88427665", "reading_canonical_log_prob": -54.45574188232422, "reading_distances": [28, 35, 29, 27], "reading_log_probs": [-65.38687133789062, -65.28277587890625, -60.805145263671875, -66.92472839355469], "reading_logprob_std": 2.6388529022696536, "tokenization_path_logprob_std": 114.30864027893553, "trajectory_balance_rmse": 1007.6312109297228} {"canonical_length": 68, "canonical_nll": 1.3993262122659123, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.2823529411764706, "prompt_id": "605c5bbf05c7cb07", "reading_canonical_log_prob": -95.15418243408203, "reading_distances": [34, 31, 35, 28], "reading_log_probs": [-93.50746154785156, -92.1400146484375, -91.337158203125, -90.40834045410156], "reading_logprob_std": 1.3130960736171224, "tokenization_path_logprob_std": 109.44603480070002, "trajectory_balance_rmse": 1925.599492650854} {"canonical_length": 163, "canonical_nll": 2.194194208624904, "generation_rouge_l_f1": 0.20095693779904306, "generation_token_f1": 0.3827751196172249, "prompt_id": "371fd9b264246fc1", "reading_canonical_log_prob": -357.6536560058594, "reading_distances": [18, 15, 18, 17], "reading_log_probs": [-359.6326904296875, -359.5281677246094, -359.150146484375, -358.156005859375], "reading_logprob_std": 0.6732022232076592, "tokenization_path_logprob_std": 154.69244363191842, "trajectory_balance_rmse": 3515.60134633138} {"canonical_length": 9, "canonical_nll": 1.1824671427408855, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.2105263157894737, "prompt_id": "ed3c13f24a96d3b3", "reading_canonical_log_prob": -10.642204284667969, "reading_distances": [28, 24, 28, 29], "reading_log_probs": [-13.244285583496094, -10.934606552124023, -11.665451049804688, -11.155872344970703], "reading_logprob_std": 1.0420959750617562, "tokenization_path_logprob_std": 28.424667566607848, "trajectory_balance_rmse": 168.94620340675417} {"canonical_length": 27, "canonical_nll": 0.8400069342719184, "generation_rouge_l_f1": 0.5652173913043479, "generation_token_f1": 0.5652173913043479, "prompt_id": "9e311e757597dd6a", "reading_canonical_log_prob": -22.680187225341797, "reading_distances": [32, 34, 33, 34], "reading_log_probs": [-27.878740310668945, -24.88071060180664, -24.55006217956543, -24.15873146057129], "reading_logprob_std": 1.7002565510252339, "tokenization_path_logprob_std": 55.04390060808531, "trajectory_balance_rmse": 646.4507854460965} {"canonical_length": 23, "canonical_nll": 0.938609911047894, "generation_rouge_l_f1": 0.24561403508771928, "generation_token_f1": 0.2807017543859649, "prompt_id": "6d2468c671bba59a", "reading_canonical_log_prob": -21.588027954101562, "reading_distances": [38, 41, 45, 45], "reading_log_probs": [-20.69329833984375, -24.31011962890625, -24.72235870361328, -21.846805572509766], "reading_logprob_std": 1.9397586961759412, "tokenization_path_logprob_std": 63.675521483159585, "trajectory_balance_rmse": 427.9747043881981} {"canonical_length": 34, "canonical_nll": 1.7512783723718979, "generation_rouge_l_f1": 0.1111111111111111, "generation_token_f1": 0.1111111111111111, "prompt_id": "d8ebdfdb02245d42", "reading_canonical_log_prob": -59.54346466064453, "reading_distances": [20, 20, 21, 20], "reading_log_probs": [-61.16478729248047, -63.05513000488281, -58.41001892089844, -64.67520141601562], "reading_logprob_std": 2.6916437314858315, "tokenization_path_logprob_std": 103.57525144386715, "trajectory_balance_rmse": 928.0390548839933} {"canonical_length": 282, "canonical_nll": 1.1836179909131206, "generation_rouge_l_f1": 0.25585585585585585, "generation_token_f1": 0.5153153153153153, "prompt_id": "0659b1d0d1c9b272", "reading_canonical_log_prob": -333.7802734375, "reading_distances": [22, 21, 19, 21], "reading_log_probs": [-341.5354309082031, -338.25927734375, -341.46826171875, -337.87408447265625], "reading_logprob_std": 1.9897071799274082, "tokenization_path_logprob_std": 88.35978422185394, "trajectory_balance_rmse": 5247.707290225362} {"canonical_length": 126, "canonical_nll": 1.6179733276367188, "generation_rouge_l_f1": 0.24404761904761907, "generation_token_f1": 0.3988095238095238, "prompt_id": "11e44461a25c9604", "reading_canonical_log_prob": -203.86463928222656, "reading_distances": [13, 16, 12, 13], "reading_log_probs": [-204.22689819335938, -207.91937255859375, -203.95950317382812, -204.47190856933594], "reading_logprob_std": 1.8617650655067959, "tokenization_path_logprob_std": 283.5852139562946, "trajectory_balance_rmse": 3514.1439014324033} {"canonical_length": 4, "canonical_nll": 0.6438968181610107, "generation_rouge_l_f1": 0.017857142857142856, "generation_token_f1": 0.017857142857142856, "prompt_id": "72cbdeacc12e1d9a", "reading_canonical_log_prob": -2.575587272644043, "reading_distances": [20, 24, 19, 21], "reading_log_probs": [-3.2365236282348633, -3.054765462875366, -3.2853362560272217, -2.6569125652313232], "reading_logprob_std": 0.2854431994920232, "tokenization_path_logprob_std": 31.820232585476795, "trajectory_balance_rmse": 77.93822437509519} {"canonical_length": 9, "canonical_nll": 1.6783118777804904, "generation_rouge_l_f1": 0.05303030303030303, "generation_token_f1": 0.05303030303030303, "prompt_id": "e46df451982e05b1", "reading_canonical_log_prob": -15.104806900024414, "reading_distances": [16, 22, 15, 22], "reading_log_probs": [-14.601043701171875, -11.04335880279541, -12.398832321166992, -13.771626472473145], "reading_logprob_std": 1.5641852633212665, "tokenization_path_logprob_std": 27.720119944387633, "trajectory_balance_rmse": 118.58562112290296} {"canonical_length": 112, "canonical_nll": 0.9613617488316127, "generation_rouge_l_f1": 0.35449735449735453, "generation_token_f1": 0.5185185185185185, "prompt_id": "ad4b8c7e2797f458", "reading_canonical_log_prob": -107.67251586914062, "reading_distances": [36, 35, 37, 30], "reading_log_probs": [-106.49899291992188, -112.38934326171875, -122.7408676147461, -115.6265640258789], "reading_logprob_std": 6.770365852421221, "tokenization_path_logprob_std": 119.77867297210148, "trajectory_balance_rmse": 1698.1408142243376} {"canonical_length": 229, "canonical_nll": 1.314555472161572, "generation_rouge_l_f1": 0.2597402597402597, "generation_token_f1": 0.4935064935064935, "prompt_id": "9d2ca51ff7ce1141", "reading_canonical_log_prob": -301.033203125, "reading_distances": [29, 33, 32, 26], "reading_log_probs": [-299.9564208984375, -308.6119689941406, -295.4560241699219, -304.5466613769531], "reading_logprob_std": 5.68981316529652, "tokenization_path_logprob_std": 244.677592494875, "trajectory_balance_rmse": 5349.295790876275} {"canonical_length": 3, "canonical_nll": 1.8078675270080566, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "f22d1f7967fa63cf", "reading_canonical_log_prob": -5.42360258102417, "reading_distances": [24, 31, 24, 32], "reading_log_probs": [-6.477930068969727, -7.323730945587158, -6.760849952697754, -7.519737243652344], "reading_logprob_std": 0.48407543955164756, "tokenization_path_logprob_std": 37.44813491143834, "trajectory_balance_rmse": 59.90185708100575} {"canonical_length": 341, "canonical_nll": 1.4314043598790323, "generation_rouge_l_f1": 0.2006472491909385, "generation_token_f1": 0.3883495145631068, "prompt_id": "a546d753c3fe436f", "reading_canonical_log_prob": -488.10888671875, "reading_distances": [20, 22, 25, 22], "reading_log_probs": [-500.3086853027344, -497.2439270019531, -494.585205078125, -493.4822998046875], "reading_logprob_std": 3.043922219950162, "tokenization_path_logprob_std": 74.503639878332, "trajectory_balance_rmse": 7046.965693857133} {"canonical_length": 175, "canonical_nll": 1.3617583356584821, "generation_rouge_l_f1": 0.2512820512820513, "generation_token_f1": 0.46153846153846156, "prompt_id": "55985253cc61ade7", "reading_canonical_log_prob": -238.30770874023438, "reading_distances": [25, 30, 26, 24], "reading_log_probs": [-239.7417755126953, -242.4608154296875, -240.9292449951172, -240.08706665039062], "reading_logprob_std": 1.2114778660531647, "tokenization_path_logprob_std": 190.18376610865292, "trajectory_balance_rmse": 3337.8791472996695} {"canonical_length": 134, "canonical_nll": 1.5136424961374766, "generation_rouge_l_f1": 0.33187772925764186, "generation_token_f1": 0.5327510917030569, "prompt_id": "e4db8e939e6495ed", "reading_canonical_log_prob": -202.82809448242188, "reading_distances": [17, 15, 20, 15], "reading_log_probs": [-201.1246795654297, -201.78672790527344, -201.01446533203125, -202.10977172851562], "reading_logprob_std": 0.5260944661094593, "tokenization_path_logprob_std": 134.37649045374528, "trajectory_balance_rmse": 3167.4392093216643} {"canonical_length": 36, "canonical_nll": 0.6898475223117404, "generation_rouge_l_f1": 0.223463687150838, "generation_token_f1": 0.24581005586592178, "prompt_id": "8ba984bcebc3168b", "reading_canonical_log_prob": -24.834510803222656, "reading_distances": [16, 17, 15, 15], "reading_log_probs": [-29.666818618774414, -33.732818603515625, -28.090457916259766, -29.337844848632812], "reading_logprob_std": 2.446655800476155, "tokenization_path_logprob_std": 32.905988424396575, "trajectory_balance_rmse": 691.2876366580431} {"canonical_length": 4, "canonical_nll": 2.193131446838379, "generation_rouge_l_f1": 0.015267175572519083, "generation_token_f1": 0.015267175572519083, "prompt_id": "fe48fe173961fdd5", "reading_canonical_log_prob": -8.772525787353516, "reading_distances": [18, 27, 24, 23], "reading_log_probs": [-9.643614768981934, -10.573003768920898, -10.512044906616211, -10.274267196655273], "reading_logprob_std": 0.4247700420566875, "tokenization_path_logprob_std": 18.138146582266973, "trajectory_balance_rmse": 41.2030502299409} {"canonical_length": 267, "canonical_nll": 1.8187271861101357, "generation_rouge_l_f1": 0.14122137404580154, "generation_token_f1": 0.25190839694656486, "prompt_id": "51d00b55119644b0", "reading_canonical_log_prob": -485.60015869140625, "reading_distances": [25, 28, 26, 25], "reading_log_probs": [-505.8533935546875, -519.8877563476562, -509.5389404296875, -507.77337646484375], "reading_logprob_std": 6.266354177342877, "tokenization_path_logprob_std": 98.56474023745216, "trajectory_balance_rmse": 6028.586184642286} {"canonical_length": 7, "canonical_nll": 0.13014593294688634, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "e708a1adc69b9f0e", "reading_canonical_log_prob": -0.9110215306282043, "reading_distances": [22, 28, 24, 25], "reading_log_probs": [-1.2398120164871216, -1.3884541988372803, -1.3293638229370117, -1.3971251249313354], "reading_logprob_std": 0.07246832068334351, "tokenization_path_logprob_std": 9.80763571879683, "trajectory_balance_rmse": 106.71557753660173} {"canonical_length": 22, "canonical_nll": 1.0067856528542258, "generation_rouge_l_f1": 0.11764705882352941, "generation_token_f1": 0.11764705882352941, "prompt_id": "d7a89d7235636e3d", "reading_canonical_log_prob": -22.14928436279297, "reading_distances": [14, 16, 15, 14], "reading_log_probs": [-23.56981658935547, -23.482501983642578, -27.598146438598633, -29.120601654052734], "reading_logprob_std": 2.8590623041085443, "tokenization_path_logprob_std": 59.32606599897191, "trajectory_balance_rmse": 504.16060746937205} {"canonical_length": 12, "canonical_nll": 0.6681664784749349, "generation_rouge_l_f1": 0.8571428571428572, "generation_token_f1": 0.8571428571428572, "prompt_id": "17e2c9ee9612ad40", "reading_canonical_log_prob": -8.017997741699219, "reading_distances": [21, 22, 23, 29], "reading_log_probs": [-8.901680946350098, -9.333252906799316, -8.853804588317871, -8.989473342895508], "reading_logprob_std": 0.21654774246818795, "tokenization_path_logprob_std": 43.48478238973027, "trajectory_balance_rmse": 264.85347372146725} {"canonical_length": 324, "canonical_nll": 1.141529224537037, "generation_rouge_l_f1": 0.26436781609195403, "generation_token_f1": 0.5287356321839081, "prompt_id": "36541d1f10617d25", "reading_canonical_log_prob": -369.85546875, "reading_distances": [11, 11, 15, 17], "reading_log_probs": [-383.68853759765625, -388.9610595703125, -377.5042724609375, -381.48187255859375], "reading_logprob_std": 4.777826437397007, "tokenization_path_logprob_std": 223.92631034401495, "trajectory_balance_rmse": 6689.168488037935} {"canonical_length": 315, "canonical_nll": 1.0976046123201886, "generation_rouge_l_f1": 0.253968253968254, "generation_token_f1": 0.5255731922398589, "prompt_id": "e15e866d282a743f", "reading_canonical_log_prob": -345.7454528808594, "reading_distances": [24, 27, 30, 33], "reading_log_probs": [-350.5931091308594, -351.87164306640625, -351.21319580078125, -352.3629150390625], "reading_logprob_std": 0.7718011798638672, "tokenization_path_logprob_std": 305.11248601956424, "trajectory_balance_rmse": 6006.29193595124} {"canonical_length": 10, "canonical_nll": 1.5683156967163085, "generation_rouge_l_f1": 0.3636363636363636, "generation_token_f1": 0.5454545454545454, "prompt_id": "718643fa581e9e44", "reading_canonical_log_prob": -15.683156967163086, "reading_distances": [18, 21, 22, 21], "reading_log_probs": [-21.220699310302734, -21.597217559814453, -20.156267166137695, -20.145727157592773], "reading_logprob_std": 0.7423846951664179, "tokenization_path_logprob_std": 28.457333628382386, "trajectory_balance_rmse": 135.94451714300354} {"canonical_length": 159, "canonical_nll": 0.9840579242826258, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.5211267605633801, "prompt_id": "482bbea97f7ae0fa", "reading_canonical_log_prob": -156.4652099609375, "reading_distances": [12, 10, 15, 12], "reading_log_probs": [-160.15792846679688, -161.8282012939453, -161.21493530273438, -160.20840454101562], "reading_logprob_std": 0.812535305564277, "tokenization_path_logprob_std": 175.74305161722054, "trajectory_balance_rmse": 3346.6067391789584} {"canonical_length": 11, "canonical_nll": 1.0646774985573508, "generation_rouge_l_f1": 0.20224719101123595, "generation_token_f1": 0.20224719101123595, "prompt_id": "f9bf285092a25c3b", "reading_canonical_log_prob": -11.71145248413086, "reading_distances": [47, 46, 43, 42], "reading_log_probs": [-13.385721206665039, -14.370804786682129, -13.436212539672852, -14.136367797851562], "reading_logprob_std": 0.4962400372337657, "tokenization_path_logprob_std": 9.2156407325003, "trajectory_balance_rmse": 223.89370101593315} {"canonical_length": 150, "canonical_nll": 0.9897731526692708, "generation_rouge_l_f1": 0.34803921568627455, "generation_token_f1": 0.5245098039215687, "prompt_id": "9d4eb4020bd4213f", "reading_canonical_log_prob": -148.46597290039062, "reading_distances": [31, 18, 28, 24], "reading_log_probs": [-150.57952880859375, -147.91574096679688, -151.71363830566406, -151.60458374023438], "reading_logprob_std": 1.7672055561509223, "tokenization_path_logprob_std": 105.29011439636068, "trajectory_balance_rmse": 2819.1132320430484} {"canonical_length": 134, "canonical_nll": 1.0873967640435518, "generation_rouge_l_f1": 0.2464985994397759, "generation_token_f1": 0.39775910364145656, "prompt_id": "13778e3d12464b89", "reading_canonical_log_prob": -145.71116638183594, "reading_distances": [18, 19, 21, 19], "reading_log_probs": [-149.9272003173828, -149.7200164794922, -150.44085693359375, -148.33950805664062], "reading_logprob_std": 0.8976179841385111, "tokenization_path_logprob_std": 99.06321453128608, "trajectory_balance_rmse": 1564.4964999692559} {"canonical_length": 395, "canonical_nll": 1.4018304366099683, "generation_rouge_l_f1": 0.2954545454545454, "generation_token_f1": 0.45454545454545453, "prompt_id": "513a1cb695ee9eb0", "reading_canonical_log_prob": -553.7230224609375, "reading_distances": [25, 24, 16, 27], "reading_log_probs": [-555.7860717773438, -557.2012939453125, -566.1708984375, -557.5628051757812], "reading_logprob_std": 4.72306419634717, "tokenization_path_logprob_std": 376.5329088725188, "trajectory_balance_rmse": 5712.643407105133} {"canonical_length": 193, "canonical_nll": 1.0578632255909974, "generation_rouge_l_f1": 0.3162583518930957, "generation_token_f1": 0.49888641425389757, "prompt_id": "444facd57b3cdbe1", "reading_canonical_log_prob": -204.1676025390625, "reading_distances": [24, 22, 22, 14], "reading_log_probs": [-220.2838897705078, -218.2888641357422, -221.05712890625, -224.63580322265625], "reading_logprob_std": 2.6500159152815126, "tokenization_path_logprob_std": 260.36435293767516, "trajectory_balance_rmse": 3779.3477763296964} {"canonical_length": 226, "canonical_nll": 1.177513055041828, "generation_rouge_l_f1": 0.29508196721311475, "generation_token_f1": 0.6434426229508196, "prompt_id": "5870502ae5fa3ae7", "reading_canonical_log_prob": -266.1179504394531, "reading_distances": [27, 20, 23, 23], "reading_log_probs": [-272.1998291015625, -271.75042724609375, -274.89825439453125, -271.8232421875], "reading_logprob_std": 1.4998637893377382, "tokenization_path_logprob_std": 187.62217589225108, "trajectory_balance_rmse": 2704.823204789533} {"canonical_length": 17, "canonical_nll": 0.9378174052518957, "generation_rouge_l_f1": 0.8387096774193549, "generation_token_f1": 0.9032258064516129, "prompt_id": "8c61a7a56532124f", "reading_canonical_log_prob": -15.942895889282227, "reading_distances": [29, 40, 31, 29], "reading_log_probs": [-14.337903022766113, -16.28921890258789, -12.905845642089844, -12.507465362548828], "reading_logprob_std": 1.7106652053347084, "tokenization_path_logprob_std": 69.15641887970315, "trajectory_balance_rmse": 365.3807598084602} {"canonical_length": 12, "canonical_nll": 0.7215415636698405, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "fb0ad86c22711d8d", "reading_canonical_log_prob": -8.658498764038086, "reading_distances": [9, 12, 14, 16], "reading_log_probs": [-7.894575595855713, -5.668924331665039, -6.937466621398926, -10.403553009033203], "reading_logprob_std": 2.004249899142611, "tokenization_path_logprob_std": 40.887298649797735, "trajectory_balance_rmse": 185.2872496507928} {"canonical_length": 219, "canonical_nll": 1.4475585380101312, "generation_rouge_l_f1": 0.21848739495798317, "generation_token_f1": 0.4327731092436975, "prompt_id": "a18ee8ea214b4b09", "reading_canonical_log_prob": -317.01531982421875, "reading_distances": [36, 42, 52, 45], "reading_log_probs": [-335.6184997558594, -336.5424499511719, -339.2633056640625, -339.70294189453125], "reading_logprob_std": 2.0084406639148096, "tokenization_path_logprob_std": 127.72705820100865, "trajectory_balance_rmse": 5166.572853765273} {"canonical_length": 283, "canonical_nll": 1.3194897115862412, "generation_rouge_l_f1": 0.20357142857142857, "generation_token_f1": 0.43214285714285716, "prompt_id": "6933d659ab69515c", "reading_canonical_log_prob": -373.41558837890625, "reading_distances": [32, 27, 25, 23], "reading_log_probs": [-383.59991455078125, -391.23809814453125, -388.2349853515625, -377.9000244140625], "reading_logprob_std": 5.816992295401424, "tokenization_path_logprob_std": 348.7772447965443, "trajectory_balance_rmse": 5318.692795730042} {"canonical_length": 49, "canonical_nll": 1.3043105456293846, "generation_rouge_l_f1": 0.49438202247191015, "generation_token_f1": 0.7191011235955055, "prompt_id": "e748f45af7778fcd", "reading_canonical_log_prob": -63.911216735839844, "reading_distances": [21, 31, 38, 33], "reading_log_probs": [-68.34943389892578, -64.98450469970703, -67.01899719238281, -66.55636596679688], "reading_logprob_std": 1.388399502776471, "tokenization_path_logprob_std": 57.350781649057275, "trajectory_balance_rmse": 856.8262411885696} {"canonical_length": 8, "canonical_nll": 1.3088672161102295, "generation_rouge_l_f1": 0.761904761904762, "generation_token_f1": 0.761904761904762, "prompt_id": "3f21812c287e05df", "reading_canonical_log_prob": -10.470937728881836, "reading_distances": [41, 27, 42, 32], "reading_log_probs": [-16.283123016357422, -12.988566398620605, -14.994033813476562, -12.153190612792969], "reading_logprob_std": 1.8788812780689361, "tokenization_path_logprob_std": 19.539500642146347, "trajectory_balance_rmse": 179.83941059731075} {"canonical_length": 38, "canonical_nll": 1.278693952058491, "generation_rouge_l_f1": 0.256, "generation_token_f1": 0.304, "prompt_id": "cc2b221229d6b7ba", "reading_canonical_log_prob": -48.590370178222656, "reading_distances": [27, 19, 21, 26], "reading_log_probs": [-59.72818374633789, -60.835655212402344, -63.493473052978516, -65.73988342285156], "reading_logprob_std": 2.7034890908966736, "tokenization_path_logprob_std": 192.52850852646887, "trajectory_balance_rmse": 913.0121549248048} {"canonical_length": 57, "canonical_nll": 1.1338691042180646, "generation_rouge_l_f1": 0.2644628099173554, "generation_token_f1": 0.3305785123966943, "prompt_id": "6c3587030700c60e", "reading_canonical_log_prob": -64.63053894042969, "reading_distances": [33, 25, 30, 23], "reading_log_probs": [-69.17877197265625, -69.37555694580078, -69.42037963867188, -68.00503540039062], "reading_logprob_std": 0.668224876324858, "tokenization_path_logprob_std": 36.98872899063368, "trajectory_balance_rmse": 1033.829572632409} {"canonical_length": 108, "canonical_nll": 1.7251418784812644, "generation_rouge_l_f1": 0.24786324786324784, "generation_token_f1": 0.5042735042735043, "prompt_id": "3f751ca73a73cf7d", "reading_canonical_log_prob": -186.31532287597656, "reading_distances": [26, 21, 13, 23], "reading_log_probs": [-189.55416870117188, -190.0286865234375, -190.04196166992188, -186.99237060546875], "reading_logprob_std": 1.4590326381651744, "tokenization_path_logprob_std": 81.54541134154982, "trajectory_balance_rmse": 1707.3288171136992} {"canonical_length": 77, "canonical_nll": 0.7191560918634589, "generation_rouge_l_f1": 0.2333333333333333, "generation_token_f1": 0.32, "prompt_id": "7c7baba7466fe6f6", "reading_canonical_log_prob": -55.37501907348633, "reading_distances": [8, 7, 7, 11], "reading_log_probs": [-58.08014678955078, -57.040191650390625, -60.563838958740234, -61.86246109008789], "reading_logprob_std": 2.2157189821348005, "tokenization_path_logprob_std": 160.63669289127293, "trajectory_balance_rmse": 1562.6886517909204} {"canonical_length": 294, "canonical_nll": 0.7391870200228529, "generation_rouge_l_f1": 0.2780748663101604, "generation_token_f1": 0.5525846702317291, "prompt_id": "fd3d2e71f4e4243f", "reading_canonical_log_prob": -217.32098388671875, "reading_distances": [15, 19, 17, 20], "reading_log_probs": [-223.82717895507812, -225.5134735107422, -222.4161834716797, -224.03643798828125], "reading_logprob_std": 1.2674893538298837, "tokenization_path_logprob_std": 225.45011464130178, "trajectory_balance_rmse": 4965.719413720546} {"canonical_length": 399, "canonical_nll": 1.033715652045152, "generation_rouge_l_f1": 0.19783616692426587, "generation_token_f1": 0.4482225656877898, "prompt_id": "5bc59064e72854cb", "reading_canonical_log_prob": -412.4525451660156, "reading_distances": [22, 20, 21, 24], "reading_log_probs": [-418.8797302246094, -417.90740966796875, -418.41729736328125, -414.3267822265625], "reading_logprob_std": 2.075688225399166, "tokenization_path_logprob_std": 119.52332950022566, "trajectory_balance_rmse": 7577.2530251044645} {"canonical_length": 280, "canonical_nll": 1.2095714024135045, "generation_rouge_l_f1": 0.21268656716417908, "generation_token_f1": 0.38059701492537307, "prompt_id": "cd25f5705dfa5cbb", "reading_canonical_log_prob": -338.67999267578125, "reading_distances": [22, 18, 25, 28], "reading_log_probs": [-361.80755615234375, -355.31890869140625, -366.91241455078125, -361.13702392578125], "reading_logprob_std": 4.74540759112125, "tokenization_path_logprob_std": 151.12950835189824, "trajectory_balance_rmse": 6346.010432409051} {"canonical_length": 365, "canonical_nll": 0.8490466810252568, "generation_rouge_l_f1": 0.20615384615384616, "generation_token_f1": 0.44, "prompt_id": "ee3200ffbd24e2bc", "reading_canonical_log_prob": -309.90203857421875, "reading_distances": [26, 28, 28, 31], "reading_log_probs": [-328.183349609375, -322.7317199707031, -328.5569152832031, -328.1929626464844], "reading_logprob_std": 2.7950915750459115, "tokenization_path_logprob_std": 194.4921203904816, "trajectory_balance_rmse": 4975.676421598184} {"canonical_length": 115, "canonical_nll": 2.162905485733696, "generation_rouge_l_f1": 0.2588235294117647, "generation_token_f1": 0.38823529411764707, "prompt_id": "4180e393ac001d5f", "reading_canonical_log_prob": -248.734130859375, "reading_distances": [23, 25, 20, 17], "reading_log_probs": [-259.123779296875, -272.8232421875, -301.66107177734375, -263.327392578125], "reading_logprob_std": 19.16165753939406, "tokenization_path_logprob_std": 120.12492312670933, "trajectory_balance_rmse": 2688.6751832963478} {"canonical_length": 9, "canonical_nll": 0.9737368689643012, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.8571428571428571, "prompt_id": "79896777ff1e0459", "reading_canonical_log_prob": -8.763631820678711, "reading_distances": [36, 51, 52, 45], "reading_log_probs": [-7.619384288787842, -10.602462768554688, -9.392345428466797, -6.151922225952148], "reading_logprob_std": 1.9572010404879923, "tokenization_path_logprob_std": 40.39470100646328, "trajectory_balance_rmse": 192.87166615662952} {"canonical_length": 147, "canonical_nll": 1.5321315843231824, "generation_rouge_l_f1": 0.2287104622871046, "generation_token_f1": 0.43795620437956206, "prompt_id": "4be1dda9c6193970", "reading_canonical_log_prob": -225.2233428955078, "reading_distances": [43, 40, 45, 43], "reading_log_probs": [-236.80697631835938, -233.52162170410156, -233.09732055664062, -234.94740295410156], "reading_logprob_std": 1.6745348215479223, "tokenization_path_logprob_std": 276.8610828288928, "trajectory_balance_rmse": 4056.380423102885} {"canonical_length": 9, "canonical_nll": 1.25760989718967, "generation_rouge_l_f1": 0.09090909090909091, "generation_token_f1": 0.09090909090909091, "prompt_id": "e275be5afb765439", "reading_canonical_log_prob": -11.318489074707031, "reading_distances": [32, 31, 30, 24], "reading_log_probs": [-32.66938400268555, -27.178266525268555, -30.576570510864258, -39.58979797363281], "reading_logprob_std": 5.238139150813973, "tokenization_path_logprob_std": 26.244132725553214, "trajectory_balance_rmse": 121.18735492312516} {"canonical_length": 8, "canonical_nll": 0.9828038215637207, "generation_rouge_l_f1": 0.05, "generation_token_f1": 0.05, "prompt_id": "b4100d452af762d4", "reading_canonical_log_prob": -7.862430572509766, "reading_distances": [17, 18, 23, 23], "reading_log_probs": [-9.705924987792969, -13.875401496887207, -10.960836410522461, -18.471500396728516], "reading_logprob_std": 3.8925520499837534, "tokenization_path_logprob_std": 12.985389819976483, "trajectory_balance_rmse": 113.64728320799263} {"canonical_length": 65, "canonical_nll": 0.5099753159743089, "generation_rouge_l_f1": 0.25641025641025644, "generation_token_f1": 0.336996336996337, "prompt_id": "660009a4d953661a", "reading_canonical_log_prob": -33.14839553833008, "reading_distances": [24, 31, 30, 34], "reading_log_probs": [-45.66142654418945, -45.652374267578125, -39.883853912353516, -48.037567138671875], "reading_logprob_std": 3.4698040660652443, "tokenization_path_logprob_std": 73.82041531142234, "trajectory_balance_rmse": 781.123413683086} {"canonical_length": 245, "canonical_nll": 0.9587403589365434, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.6711111111111111, "prompt_id": "9c1c5be112145501", "reading_canonical_log_prob": -234.89138793945312, "reading_distances": [26, 30, 16, 16], "reading_log_probs": [-235.75201416015625, -236.8090057373047, -233.31527709960938, -234.32931518554688], "reading_logprob_std": 1.5400830425343148, "tokenization_path_logprob_std": 72.5930666666251, "trajectory_balance_rmse": 4338.400885264315} {"canonical_length": 41, "canonical_nll": 0.7183907671672541, "generation_rouge_l_f1": 0.1384083044982699, "generation_token_f1": 0.17301038062283738, "prompt_id": "8a1926c471e546ee", "reading_canonical_log_prob": -29.454021453857422, "reading_distances": [21, 15, 27, 24], "reading_log_probs": [-27.864105224609375, -27.490419387817383, -29.284351348876953, -24.256643295288086], "reading_logprob_std": 2.1237519805614844, "tokenization_path_logprob_std": 36.63323141257339, "trajectory_balance_rmse": 458.4265028267761} {"canonical_length": 311, "canonical_nll": 1.1911226622161375, "generation_rouge_l_f1": 0.23488773747841105, "generation_token_f1": 0.5077720207253885, "prompt_id": "a1e93e4a3a2191cd", "reading_canonical_log_prob": -370.43914794921875, "reading_distances": [37, 35, 36, 42], "reading_log_probs": [-381.6045227050781, -383.43463134765625, -378.4156494140625, -381.2262878417969], "reading_logprob_std": 2.07420509237198, "tokenization_path_logprob_std": 232.26652747525404, "trajectory_balance_rmse": 6360.184871421859} {"canonical_length": 130, "canonical_nll": 1.5403663048377403, "generation_rouge_l_f1": 0.21025641025641023, "generation_token_f1": 0.37948717948717947, "prompt_id": "c4f0abd89b25e9b2", "reading_canonical_log_prob": -200.24761962890625, "reading_distances": [22, 21, 18, 21], "reading_log_probs": [-203.114013671875, -201.79367065429688, -204.56982421875, -204.11135864257812], "reading_logprob_std": 1.2297114814740828, "tokenization_path_logprob_std": 203.427994476658, "trajectory_balance_rmse": 2993.5454818325097} {"canonical_length": 226, "canonical_nll": 1.1280028756740874, "generation_rouge_l_f1": 0.26552462526766596, "generation_token_f1": 0.4411134903640257, "prompt_id": "8d709bdbfc4890e5", "reading_canonical_log_prob": -254.92864990234375, "reading_distances": [23, 27, 30, 20], "reading_log_probs": [-253.0482177734375, -253.92379760742188, -253.40103149414062, -249.25308227539062], "reading_logprob_std": 2.1328486401041156, "tokenization_path_logprob_std": 125.50199346523257, "trajectory_balance_rmse": 3872.12348552073} {"canonical_length": 410, "canonical_nll": 1.225685975609756, "generation_rouge_l_f1": 0.2524271844660194, "generation_token_f1": 0.4304207119741101, "prompt_id": "fbcfdeca84611d04", "reading_canonical_log_prob": -502.53125, "reading_distances": [30, 19, 29, 28], "reading_log_probs": [-532.4943237304688, -525.622802734375, -542.24365234375, -537.176513671875], "reading_logprob_std": 7.0687509760782525, "tokenization_path_logprob_std": 222.6698966263762, "trajectory_balance_rmse": 7760.729760695796} {"canonical_length": 45, "canonical_nll": 1.3341722276475694, "generation_rouge_l_f1": 0.23703703703703702, "generation_token_f1": 0.4148148148148148, "prompt_id": "8c6898d598b2732f", "reading_canonical_log_prob": -60.037750244140625, "reading_distances": [31, 31, 33, 26], "reading_log_probs": [-68.78485870361328, -69.61959838867188, -69.13404083251953, -62.320655822753906], "reading_logprob_std": 3.4464617290548216, "tokenization_path_logprob_std": 15.305556877195945, "trajectory_balance_rmse": 964.618390431262} {"canonical_length": 93, "canonical_nll": 0.5513881150112357, "generation_rouge_l_f1": 0.526829268292683, "generation_token_f1": 0.673170731707317, "prompt_id": "f5259298cfe12698", "reading_canonical_log_prob": -51.27909469604492, "reading_distances": [19, 16, 17, 21], "reading_log_probs": [-55.47312927246094, -55.06672668457031, -56.57714080810547, -54.97869110107422], "reading_logprob_std": 0.7344281332012058, "tokenization_path_logprob_std": 30.73671536511653, "trajectory_balance_rmse": 1070.1745475951643} {"canonical_length": 183, "canonical_nll": 0.39290894054975667, "generation_rouge_l_f1": 0.32628398791540786, "generation_token_f1": 0.42900302114803623, "prompt_id": "af8c4385ce2b0480", "reading_canonical_log_prob": -71.90233612060547, "reading_distances": [31, 31, 37, 35], "reading_log_probs": [-81.27145385742188, -77.3173828125, -76.3974609375, -79.36418914794922], "reading_logprob_std": 2.176872607460677, "tokenization_path_logprob_std": 57.109753945689825, "trajectory_balance_rmse": 2653.349577543547} {"canonical_length": 164, "canonical_nll": 1.6390306426257621, "generation_rouge_l_f1": 0.18401937046004843, "generation_token_f1": 0.3922518159806295, "prompt_id": "8db3810d387bcbf4", "reading_canonical_log_prob": -268.801025390625, "reading_distances": [24, 22, 19, 25], "reading_log_probs": [-278.8796081542969, -277.28472900390625, -273.12408447265625, -276.2156982421875], "reading_logprob_std": 2.4286136950770234, "tokenization_path_logprob_std": 353.3839967807177, "trajectory_balance_rmse": 3756.2683147037674} {"canonical_length": 170, "canonical_nll": 0.8208959242876839, "generation_rouge_l_f1": 0.33971291866028713, "generation_token_f1": 0.5454545454545455, "prompt_id": "d92a9462a284aee7", "reading_canonical_log_prob": -139.55230712890625, "reading_distances": [12, 16, 13, 14], "reading_log_probs": [-143.1276092529297, -141.50845336914062, -144.9711456298828, -142.15895080566406], "reading_logprob_std": 1.5077692040867134, "tokenization_path_logprob_std": 95.50525094235712, "trajectory_balance_rmse": 3319.8080594576672} {"canonical_length": 76, "canonical_nll": 0.9126317877518503, "generation_rouge_l_f1": 0.3253968253968254, "generation_token_f1": 0.46031746031746035, "prompt_id": "32c20ba8cc10f798", "reading_canonical_log_prob": -69.36001586914062, "reading_distances": [33, 33, 28, 28], "reading_log_probs": [-72.36796569824219, -73.3801498413086, -71.91822814941406, -69.75914001464844], "reading_logprob_std": 1.5259789432078863, "tokenization_path_logprob_std": 89.0187604756268, "trajectory_balance_rmse": 1783.5392338438783} {"canonical_length": 21, "canonical_nll": 0.7110481262207031, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.26666666666666666, "prompt_id": "2e575375b1e838f0", "reading_canonical_log_prob": -14.932010650634766, "reading_distances": [25, 23, 27, 29], "reading_log_probs": [-16.13966178894043, -15.106914520263672, -15.004173278808594, -15.3798189163208], "reading_logprob_std": 0.5131117881560637, "tokenization_path_logprob_std": 82.89785741174292, "trajectory_balance_rmse": 246.7809156177762} {"canonical_length": 9, "canonical_nll": 0.24837761455112034, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.7142857142857143, "prompt_id": "f19b5cf2d17aa856", "reading_canonical_log_prob": -2.235398530960083, "reading_distances": [37, 36, 30, 42], "reading_log_probs": [-1.9100587368011475, -5.107854843139648, -6.079684257507324, -3.87357234954834], "reading_logprob_std": 1.7981911759491374, "tokenization_path_logprob_std": 21.05950808326975, "trajectory_balance_rmse": 178.9866521355175} {"canonical_length": 3, "canonical_nll": 1.4549781481424968, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "745badf44c39e597", "reading_canonical_log_prob": -4.36493444442749, "reading_distances": [24, 26, 19, 20], "reading_log_probs": [-5.5652570724487305, -5.451928615570068, -4.473116397857666, -5.7043609619140625], "reading_logprob_std": 0.5599645158282606, "tokenization_path_logprob_std": 8.015248978065562, "trajectory_balance_rmse": 39.35558906022294} {"canonical_length": 272, "canonical_nll": 1.300362979664522, "generation_rouge_l_f1": 0.25918762088974856, "generation_token_f1": 0.5145067698259187, "prompt_id": "8f529e5554e9c902", "reading_canonical_log_prob": -353.69873046875, "reading_distances": [21, 22, 18, 19], "reading_log_probs": [-352.05499267578125, -354.66387939453125, -355.78033447265625, -354.33740234375], "reading_logprob_std": 1.5633555695479477, "tokenization_path_logprob_std": 100.743335025795, "trajectory_balance_rmse": 5571.6465905641835} {"canonical_length": 324, "canonical_nll": 0.9853743564935378, "generation_rouge_l_f1": 0.22468354430379747, "generation_token_f1": 0.5316455696202531, "prompt_id": "642f20171ea337c8", "reading_canonical_log_prob": -319.26129150390625, "reading_distances": [26, 21, 25, 27], "reading_log_probs": [-314.447021484375, -318.1177978515625, -318.2789001464844, -318.70574951171875], "reading_logprob_std": 1.9758647709004353, "tokenization_path_logprob_std": 201.92180918098873, "trajectory_balance_rmse": 4805.040453948712} {"canonical_length": 14, "canonical_nll": 1.4311210087367467, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4210526315789474, "prompt_id": "c076cc444fb03f81", "reading_canonical_log_prob": -20.035694122314453, "reading_distances": [196, 181, 192, 186], "reading_log_probs": [-23.67742347717285, -23.087646484375, -23.064655303955078, -22.015056610107422], "reading_logprob_std": 0.691581544176366, "tokenization_path_logprob_std": 57.357119301658834, "trajectory_balance_rmse": 384.6351526668144} {"canonical_length": 183, "canonical_nll": 1.2081358862704918, "generation_rouge_l_f1": 0.2393162393162393, "generation_token_f1": 0.2564102564102564, "prompt_id": "8082ddc02c38399e", "reading_canonical_log_prob": -221.0888671875, "reading_distances": [18, 26, 25, 25], "reading_log_probs": [-223.43797302246094, -220.47535705566406, -220.34934997558594, -221.83758544921875], "reading_logprob_std": 1.442345164492267, "tokenization_path_logprob_std": 153.28731705811623, "trajectory_balance_rmse": 1651.133480452621} {"canonical_length": 7, "canonical_nll": 0.41781394822256906, "generation_rouge_l_f1": 0.08695652173913045, "generation_token_f1": 0.08695652173913045, "prompt_id": "3af881f54016c0d1", "reading_canonical_log_prob": -2.9246976375579834, "reading_distances": [35, 38, 36, 40], "reading_log_probs": [-6.290343284606934, -4.4900360107421875, -6.103565216064453, -6.717667102813721], "reading_logprob_std": 0.9747450897083437, "tokenization_path_logprob_std": 9.16936764259556, "trajectory_balance_rmse": 78.42960915175814} {"canonical_length": 241, "canonical_nll": 0.6713363204259596, "generation_rouge_l_f1": 0.3140495867768595, "generation_token_f1": 0.5041322314049587, "prompt_id": "edd1c37dcb8ffb27", "reading_canonical_log_prob": -161.79205322265625, "reading_distances": [14, 14, 15, 16], "reading_log_probs": [-172.63739013671875, -173.55775451660156, -167.75338745117188, -164.91912841796875], "reading_logprob_std": 4.08882296585079, "tokenization_path_logprob_std": 72.37564053204666, "trajectory_balance_rmse": 3616.1753159035966} {"canonical_length": 277, "canonical_nll": 1.624097693267712, "generation_rouge_l_f1": 0.18691588785046728, "generation_token_f1": 0.3588785046728972, "prompt_id": "b7a747a09567587c", "reading_canonical_log_prob": -449.87506103515625, "reading_distances": [36, 29, 37, 27], "reading_log_probs": [-449.76812744140625, -449.5173034667969, -450.6380920410156, -450.38568115234375], "reading_logprob_std": 0.522420787159817, "tokenization_path_logprob_std": 351.1199418229044, "trajectory_balance_rmse": 5752.612918156292} {"canonical_length": 203, "canonical_nll": 1.5134220217249077, "generation_rouge_l_f1": 0.20085470085470086, "generation_token_f1": 0.4444444444444445, "prompt_id": "85ba85bd1384cec6", "reading_canonical_log_prob": -307.22467041015625, "reading_distances": [19, 22, 16, 16], "reading_log_probs": [-306.62982177734375, -305.5265197753906, -305.51495361328125, -310.0132141113281], "reading_logprob_std": 2.1266651782573147, "tokenization_path_logprob_std": 213.3961695417484, "trajectory_balance_rmse": 4731.57739869219} {"canonical_length": 25, "canonical_nll": 0.54549072265625, "generation_rouge_l_f1": 0.7272727272727272, "generation_token_f1": 0.9090909090909091, "prompt_id": "ff4cb1405a9ba9bb", "reading_canonical_log_prob": -13.63726806640625, "reading_distances": [16, 15, 16, 22], "reading_log_probs": [-15.197519302368164, -13.963701248168945, -14.921001434326172, -15.760229110717773], "reading_logprob_std": 0.7507579809088422, "tokenization_path_logprob_std": 82.63690746949143, "trajectory_balance_rmse": 343.14950617364354} {"canonical_length": 103, "canonical_nll": 1.501685281401699, "generation_rouge_l_f1": 0.3270440251572327, "generation_token_f1": 0.4150943396226416, "prompt_id": "ea60f11b47ae285e", "reading_canonical_log_prob": -154.673583984375, "reading_distances": [22, 29, 23, 29], "reading_log_probs": [-156.52053833007812, -154.03375244140625, -155.1604461669922, -155.42994689941406], "reading_logprob_std": 1.0212236441905373, "tokenization_path_logprob_std": 159.41199403731218, "trajectory_balance_rmse": 2609.8860297938586} {"canonical_length": 22, "canonical_nll": 0.41535429521040484, "generation_rouge_l_f1": 0.22047244094488186, "generation_token_f1": 0.31496062992125984, "prompt_id": "6245bcac7beebc0c", "reading_canonical_log_prob": -9.137794494628906, "reading_distances": [25, 30, 21, 25], "reading_log_probs": [-9.762015342712402, -8.728254318237305, -9.372873306274414, -8.392000198364258], "reading_logprob_std": 0.6183140654689633, "tokenization_path_logprob_std": 76.38195270615446, "trajectory_balance_rmse": 489.48505083169624} {"canonical_length": 156, "canonical_nll": 1.4283115680401142, "generation_rouge_l_f1": 0.1898148148148148, "generation_token_f1": 0.3796296296296296, "prompt_id": "5016da9f4e182301", "reading_canonical_log_prob": -222.8166046142578, "reading_distances": [14, 16, 20, 18], "reading_log_probs": [-223.14976501464844, -225.9716033935547, -225.560302734375, -228.70565795898438], "reading_logprob_std": 2.2763071781539193, "tokenization_path_logprob_std": 161.414449432669, "trajectory_balance_rmse": 3973.2753869909793} {"canonical_length": 103, "canonical_nll": 1.6219171320350425, "generation_rouge_l_f1": 0.23676012461059193, "generation_token_f1": 0.41744548286604366, "prompt_id": "d6b153dd541c60bc", "reading_canonical_log_prob": -167.05746459960938, "reading_distances": [30, 33, 31, 34], "reading_log_probs": [-166.7840576171875, -167.48472595214844, -166.2203369140625, -168.12850952148438], "reading_logprob_std": 0.8301871121023546, "tokenization_path_logprob_std": 164.72750045963738, "trajectory_balance_rmse": 1776.9748340919612} {"canonical_length": 50, "canonical_nll": 1.1491517639160156, "generation_rouge_l_f1": 0.15172413793103445, "generation_token_f1": 0.21379310344827585, "prompt_id": "f50b16ccb1ed30ed", "reading_canonical_log_prob": -57.45758819580078, "reading_distances": [14, 17, 13, 16], "reading_log_probs": [-77.52726745605469, -72.22250366210938, -67.5192642211914, -71.00254821777344], "reading_logprob_std": 4.149444026238135, "tokenization_path_logprob_std": 59.92999354466092, "trajectory_balance_rmse": 926.1529634973518} {"canonical_length": 26, "canonical_nll": 0.5048969342158391, "generation_rouge_l_f1": 0.21428571428571425, "generation_token_f1": 0.21428571428571425, "prompt_id": "ac38f5f5447a0903", "reading_canonical_log_prob": -13.127320289611816, "reading_distances": [35, 43, 36, 36], "reading_log_probs": [-14.847053527832031, -15.637397766113281, -15.490274429321289, -16.08060073852539], "reading_logprob_std": 0.5104389586774526, "tokenization_path_logprob_std": 105.10724148371243, "trajectory_balance_rmse": 548.5776714841809} {"canonical_length": 10, "canonical_nll": 1.4388532638549805, "generation_rouge_l_f1": 0.07633587786259541, "generation_token_f1": 0.07633587786259541, "prompt_id": "0027f65734731d27", "reading_canonical_log_prob": -14.388532638549805, "reading_distances": [25, 23, 18, 21], "reading_log_probs": [-15.82016372680664, -18.276782989501953, -14.555577278137207, -18.472475051879883], "reading_logprob_std": 1.9126044187755578, "tokenization_path_logprob_std": 32.14800692829297, "trajectory_balance_rmse": 264.3766412540554} {"canonical_length": 146, "canonical_nll": 1.2791259974649507, "generation_rouge_l_f1": 0.23295454545454544, "generation_token_f1": 0.38636363636363635, "prompt_id": "488c8f3398b00ac7", "reading_canonical_log_prob": -186.7523956298828, "reading_distances": [15, 14, 19, 12], "reading_log_probs": [-193.038330078125, -193.47415161132812, -192.17913818359375, -192.39712524414062], "reading_logprob_std": 0.5932859549261748, "tokenization_path_logprob_std": 150.89444112340098, "trajectory_balance_rmse": 2525.50325874749} {"canonical_length": 4, "canonical_nll": 4.388460159301758, "generation_rouge_l_f1": 0.05, "generation_token_f1": 0.05, "prompt_id": "91adc07e055d938f", "reading_canonical_log_prob": -17.55384063720703, "reading_distances": [31, 31, 34, 30], "reading_log_probs": [-23.03875732421875, -17.892745971679688, -19.597488403320312, -21.032489776611328], "reading_logprob_std": 2.182739886762637, "tokenization_path_logprob_std": 5.431723994769332, "trajectory_balance_rmse": 30.485671253642067} {"canonical_length": 263, "canonical_nll": 1.2739140615717086, "generation_rouge_l_f1": 0.2702702702702703, "generation_token_f1": 0.4787644787644788, "prompt_id": "5f3070e7c5d93f6b", "reading_canonical_log_prob": -335.0393981933594, "reading_distances": [19, 21, 16, 20], "reading_log_probs": [-340.2715759277344, -340.6298522949219, -339.6258850097656, -338.363037109375], "reading_logprob_std": 0.9970366656843641, "tokenization_path_logprob_std": 152.72402496749697, "trajectory_balance_rmse": 6032.149225207823} {"canonical_length": 17, "canonical_nll": 0.13953241179971135, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.2857142857142857, "prompt_id": "f969cf0286bd838d", "reading_canonical_log_prob": -2.3720510005950928, "reading_distances": [18, 22, 21, 15], "reading_log_probs": [-3.0198426246643066, -3.937525510787964, -2.154365062713623, -2.0867061614990234], "reading_logprob_std": 0.8694686175477884, "tokenization_path_logprob_std": 36.02856530785446, "trajectory_balance_rmse": 239.0466986053857} {"canonical_length": 291, "canonical_nll": 0.6174261348763692, "generation_rouge_l_f1": 0.21729490022172948, "generation_token_f1": 0.4257206208425721, "prompt_id": "60292acfc5a172b6", "reading_canonical_log_prob": -179.67100524902344, "reading_distances": [25, 36, 33, 23], "reading_log_probs": [-185.04116821289062, -186.13525390625, -192.7250213623047, -186.67791748046875], "reading_logprob_std": 3.4545206508823116, "tokenization_path_logprob_std": 235.18244699933365, "trajectory_balance_rmse": 4867.299604325085} {"canonical_length": 174, "canonical_nll": 0.853436261758037, "generation_rouge_l_f1": 0.2071197411003236, "generation_token_f1": 0.34951456310679613, "prompt_id": "2b8f8cb7903f03af", "reading_canonical_log_prob": -148.49790954589844, "reading_distances": [37, 33, 41, 42], "reading_log_probs": [-147.57537841796875, -149.0791015625, -147.2239532470703, -149.6976318359375], "reading_logprob_std": 1.184338185395829, "tokenization_path_logprob_std": 245.07498525237386, "trajectory_balance_rmse": 3428.768364284533} {"canonical_length": 71, "canonical_nll": 1.1708915602992958, "generation_rouge_l_f1": 0.21262458471760795, "generation_token_f1": 0.292358803986711, "prompt_id": "2a8ed34f73d4ba31", "reading_canonical_log_prob": -83.13330078125, "reading_distances": [19, 14, 21, 19], "reading_log_probs": [-86.24940490722656, -85.79452514648438, -85.00385284423828, -84.38418579101562], "reading_logprob_std": 0.8284304906366989, "tokenization_path_logprob_std": 38.04931037027333, "trajectory_balance_rmse": 1173.3949540555855} {"canonical_length": 370, "canonical_nll": 0.753729248046875, "generation_rouge_l_f1": 0.36018957345971564, "generation_token_f1": 0.5624012638230648, "prompt_id": "a2f2bd1ddbd7ac6c", "reading_canonical_log_prob": -278.87982177734375, "reading_distances": [22, 19, 21, 22], "reading_log_probs": [-279.8619384765625, -274.426025390625, -277.95086669921875, -278.7859802246094], "reading_logprob_std": 2.3539003020061435, "tokenization_path_logprob_std": 173.6581382751773, "trajectory_balance_rmse": 7053.046500625551} {"canonical_length": 292, "canonical_nll": 1.4641619120558647, "generation_rouge_l_f1": 0.2018348623853211, "generation_token_f1": 0.3889908256880734, "prompt_id": "439bf3be7d027444", "reading_canonical_log_prob": -427.5352783203125, "reading_distances": [22, 23, 22, 22], "reading_log_probs": [-428.3792724609375, -427.9268798828125, -429.3828430175781, -427.10113525390625], "reading_logprob_std": 0.9510222915219871, "tokenization_path_logprob_std": 127.86448029122276, "trajectory_balance_rmse": 5756.77038703452} {"canonical_length": 152, "canonical_nll": 1.0294713472065173, "generation_rouge_l_f1": 0.38421052631578945, "generation_token_f1": 0.5315789473684212, "prompt_id": "53583cdc2f7841b7", "reading_canonical_log_prob": -156.47964477539062, "reading_distances": [23, 21, 25, 25], "reading_log_probs": [-173.1112060546875, -173.56280517578125, -168.26242065429688, -169.8485107421875], "reading_logprob_std": 2.561991141497581, "tokenization_path_logprob_std": 120.57752856899675, "trajectory_balance_rmse": 2481.3827771125975} {"canonical_length": 73, "canonical_nll": 1.114371939881207, "generation_rouge_l_f1": 0.1453488372093023, "generation_token_f1": 0.2674418604651163, "prompt_id": "11b8b10c9b5e1733", "reading_canonical_log_prob": -81.34915161132812, "reading_distances": [17, 16, 12, 20], "reading_log_probs": [-81.53487396240234, -81.43714904785156, -82.87368774414062, -81.87683868408203], "reading_logprob_std": 0.6563529806599142, "tokenization_path_logprob_std": 118.96740193002374, "trajectory_balance_rmse": 1682.6611562151766} {"canonical_length": 17, "canonical_nll": 0.9638510311351103, "generation_rouge_l_f1": 0.10778443113772454, "generation_token_f1": 0.11976047904191615, "prompt_id": "d62e67048e61fa64", "reading_canonical_log_prob": -16.385467529296875, "reading_distances": [15, 9, 17, 13], "reading_log_probs": [-27.647624969482422, -25.519073486328125, -25.027652740478516, -23.051523208618164], "reading_logprob_std": 1.8875584746478296, "tokenization_path_logprob_std": 78.8250275364158, "trajectory_balance_rmse": 279.1737691063804} {"canonical_length": 17, "canonical_nll": 1.196514466229607, "generation_rouge_l_f1": 0.18823529411764706, "generation_token_f1": 0.21176470588235294, "prompt_id": "15d2843edae95663", "reading_canonical_log_prob": -20.34074592590332, "reading_distances": [21, 22, 23, 26], "reading_log_probs": [-26.349227905273438, -25.975204467773438, -24.84611701965332, -21.900737762451172], "reading_logprob_std": 2.0153510195516526, "tokenization_path_logprob_std": 33.574750925827765, "trajectory_balance_rmse": 326.10425184379176} {"canonical_length": 243, "canonical_nll": 1.3947431148324974, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.48582995951417, "prompt_id": "960951e022fd8e15", "reading_canonical_log_prob": -338.9225769042969, "reading_distances": [33, 34, 33, 30], "reading_log_probs": [-354.37408447265625, -354.1444091796875, -351.4924621582031, -353.59454345703125], "reading_logprob_std": 1.3139667424441401, "tokenization_path_logprob_std": 215.858040670274, "trajectory_balance_rmse": 5974.089831165828} {"canonical_length": 192, "canonical_nll": 1.2980165481567383, "generation_rouge_l_f1": 0.29977628635346754, "generation_token_f1": 0.5145413870246085, "prompt_id": "e17ce1bfa59fa95a", "reading_canonical_log_prob": -249.21917724609375, "reading_distances": [18, 25, 15, 25], "reading_log_probs": [-250.79266357421875, -245.22613525390625, -245.63333129882812, -248.2996826171875], "reading_logprob_std": 2.590717926352637, "tokenization_path_logprob_std": 132.94559320481207, "trajectory_balance_rmse": 4194.790407632609} {"canonical_length": 37, "canonical_nll": 1.2992667120856207, "generation_rouge_l_f1": 0.19999999999999998, "generation_token_f1": 0.28, "prompt_id": "305cbe322d895404", "reading_canonical_log_prob": -48.07286834716797, "reading_distances": [13, 24, 14, 13], "reading_log_probs": [-46.98246765136719, -48.75138854980469, -49.746551513671875, -49.612266540527344], "reading_logprob_std": 1.272614619388831, "tokenization_path_logprob_std": 24.381609237075498, "trajectory_balance_rmse": 533.2752770523638} {"canonical_length": 25, "canonical_nll": 1.3035293579101563, "generation_rouge_l_f1": 0.11718749999999999, "generation_token_f1": 0.1484375, "prompt_id": "fee5a1b64fc319ac", "reading_canonical_log_prob": -32.588233947753906, "reading_distances": [27, 26, 24, 24], "reading_log_probs": [-38.52960968017578, -36.52599334716797, -37.841922760009766, -33.17755889892578], "reading_logprob_std": 2.377530458576105, "tokenization_path_logprob_std": 46.37585626576449, "trajectory_balance_rmse": 562.5622655220646} {"canonical_length": 12, "canonical_nll": 1.2830328146616619, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.2857142857142857, "prompt_id": "dce2143d59fcc65f", "reading_canonical_log_prob": -15.396393775939941, "reading_distances": [11, 11, 13, 20], "reading_log_probs": [-16.97539520263672, -21.43335723876953, -19.700368881225586, -18.23191261291504], "reading_logprob_std": 1.921080494987827, "tokenization_path_logprob_std": 33.217631117669924, "trajectory_balance_rmse": 232.57777919781012} {"canonical_length": 30, "canonical_nll": 0.4184571901957194, "generation_rouge_l_f1": 0.14915254237288136, "generation_token_f1": 0.14915254237288136, "prompt_id": "bc27bf2b83eeb772", "reading_canonical_log_prob": -12.553715705871582, "reading_distances": [35, 42, 39, 37], "reading_log_probs": [-10.187223434448242, -14.688032150268555, -17.43046760559082, -12.3626070022583], "reading_logprob_std": 3.1100097042548227, "tokenization_path_logprob_std": 56.7134820863859, "trajectory_balance_rmse": 364.33797139826294} {"canonical_length": 171, "canonical_nll": 2.024245234260782, "generation_rouge_l_f1": 0.28423772609819126, "generation_token_f1": 0.47545219638242897, "prompt_id": "8e8e4001993842cc", "reading_canonical_log_prob": -346.14593505859375, "reading_distances": [23, 25, 30, 23], "reading_log_probs": [-340.181396484375, -342.03515625, -340.652587890625, -341.8384094238281], "reading_logprob_std": 0.9018737800662014, "tokenization_path_logprob_std": 130.8821269597097, "trajectory_balance_rmse": 2666.7035703011247} {"canonical_length": 93, "canonical_nll": 1.8173299809937835, "generation_rouge_l_f1": 0.26143790849673204, "generation_token_f1": 0.36601307189542487, "prompt_id": "ae4b947d6478d1ef", "reading_canonical_log_prob": -169.01168823242188, "reading_distances": [21, 24, 18, 27], "reading_log_probs": [-175.263916015625, -179.90097045898438, -178.17706298828125, -177.28761291503906], "reading_logprob_std": 1.9295220562897983, "tokenization_path_logprob_std": 87.15434661271787, "trajectory_balance_rmse": 2050.1022405456797} {"canonical_length": 153, "canonical_nll": 2.184887455958946, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.4857142857142857, "prompt_id": "e8d78f3e32f3fc0a", "reading_canonical_log_prob": -334.28778076171875, "reading_distances": [18, 20, 23, 24], "reading_log_probs": [-345.0582275390625, -344.33380126953125, -348.309326171875, -345.1626281738281], "reading_logprob_std": 1.7677385697116803, "tokenization_path_logprob_std": 77.0875751266899, "trajectory_balance_rmse": 4092.819427762184} {"canonical_length": 14, "canonical_nll": 1.3732305254255022, "generation_rouge_l_f1": 0.5333333333333333, "generation_token_f1": 0.5333333333333333, "prompt_id": "0652d72e6ba6a111", "reading_canonical_log_prob": -19.22522735595703, "reading_distances": [35, 27, 25, 28], "reading_log_probs": [-21.8012752532959, -22.496850967407227, -21.74398422241211, -23.012977600097656], "reading_logprob_std": 0.6054522856804225, "tokenization_path_logprob_std": 29.34012187754935, "trajectory_balance_rmse": 195.47321726284878} {"canonical_length": 35, "canonical_nll": 1.6850494384765624, "generation_rouge_l_f1": 0.198019801980198, "generation_token_f1": 0.37623762376237624, "prompt_id": "abb413900ce0e656", "reading_canonical_log_prob": -58.97673034667969, "reading_distances": [28, 24, 23, 27], "reading_log_probs": [-57.63823699951172, -61.651161193847656, -60.343196868896484, -56.45169448852539], "reading_logprob_std": 2.392997894663724, "tokenization_path_logprob_std": 89.97196667116154, "trajectory_balance_rmse": 925.0012476646384} {"canonical_length": 294, "canonical_nll": 0.8864903871704932, "generation_rouge_l_f1": 0.18439716312056736, "generation_token_f1": 0.42907801418439717, "prompt_id": "7607920df7068b2c", "reading_canonical_log_prob": -260.628173828125, "reading_distances": [14, 8, 13, 9], "reading_log_probs": [-269.752685546875, -271.93145751953125, -268.45721435546875, -265.5114440917969], "reading_logprob_std": 2.6829387248955094, "tokenization_path_logprob_std": 214.08735120634358, "trajectory_balance_rmse": 5451.364164533425} {"canonical_length": 50, "canonical_nll": 2.6402703857421876, "generation_rouge_l_f1": 0.30769230769230765, "generation_token_f1": 0.4358974358974359, "prompt_id": "6e3d23fca9ec801f", "reading_canonical_log_prob": -132.01351928710938, "reading_distances": [28, 32, 27, 24], "reading_log_probs": [-141.11859130859375, -143.4976043701172, -141.27816772460938, -139.677490234375], "reading_logprob_std": 1.577003475657373, "tokenization_path_logprob_std": 113.48477070062017, "trajectory_balance_rmse": 1248.4387201862187} {"canonical_length": 18, "canonical_nll": 1.9276771545410156, "generation_rouge_l_f1": 0.4666666666666667, "generation_token_f1": 0.4666666666666667, "prompt_id": "4afc9f7951fa42e2", "reading_canonical_log_prob": -34.69818878173828, "reading_distances": [18, 20, 24, 21], "reading_log_probs": [-32.23305892944336, -32.58955383300781, -31.52487564086914, -32.49953842163086], "reading_logprob_std": 0.48228911363266075, "tokenization_path_logprob_std": 41.404577877357404, "trajectory_balance_rmse": 376.7111534276547} {"canonical_length": 501, "canonical_nll": 0.6360296984156687, "generation_rouge_l_f1": 0.2421052631578947, "generation_token_f1": 0.3763157894736842, "prompt_id": "031d7ff67f65087b", "reading_canonical_log_prob": -318.65087890625, "reading_distances": [34, 27, 22, 24], "reading_log_probs": [-322.5064392089844, -323.2408142089844, -319.81524658203125, -321.135498046875], "reading_logprob_std": 1.5157846621922682, "tokenization_path_logprob_std": 163.93652242013226, "trajectory_balance_rmse": 6064.579233641631} {"canonical_length": 128, "canonical_nll": 1.152758240699768, "generation_rouge_l_f1": 0.2136986301369863, "generation_token_f1": 0.3561643835616438, "prompt_id": "81ff4f6d7865cbd6", "reading_canonical_log_prob": -147.5530548095703, "reading_distances": [20, 19, 18, 19], "reading_log_probs": [-163.55479431152344, -163.99484252929688, -163.07345581054688, -164.94183349609375], "reading_logprob_std": 0.7950779541736086, "tokenization_path_logprob_std": 155.16164011048966, "trajectory_balance_rmse": 2323.009173820879} {"canonical_length": 112, "canonical_nll": 1.394688333783831, "generation_rouge_l_f1": 0.20105820105820105, "generation_token_f1": 0.3280423280423281, "prompt_id": "fa3e1b2e7d03f031", "reading_canonical_log_prob": -156.20509338378906, "reading_distances": [26, 29, 30, 28], "reading_log_probs": [-158.9398193359375, -161.24758911132812, -162.71893310546875, -157.3484649658203], "reading_logprob_std": 2.386592082878725, "tokenization_path_logprob_std": 166.72829332432383, "trajectory_balance_rmse": 2722.255134328017} {"canonical_length": 136, "canonical_nll": 1.175948647891774, "generation_rouge_l_f1": 0.3, "generation_token_f1": 0.46111111111111114, "prompt_id": "c4484ce7220f53d7", "reading_canonical_log_prob": -159.92901611328125, "reading_distances": [19, 15, 13, 15], "reading_log_probs": [-162.4453125, -161.00106811523438, -163.09214782714844, -158.80967712402344], "reading_logprob_std": 1.8981668027224208, "tokenization_path_logprob_std": 107.77347938974158, "trajectory_balance_rmse": 2749.7797056857357} {"canonical_length": 5, "canonical_nll": 0.15211386680603028, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "e265516b7f250f1e", "reading_canonical_log_prob": -0.7605693340301514, "reading_distances": [36, 36, 40, 41], "reading_log_probs": [-2.5922112464904785, -1.795020341873169, -2.3801469802856445, -2.4480109214782715], "reading_logprob_std": 0.35055357646707314, "tokenization_path_logprob_std": 41.18227207734934, "trajectory_balance_rmse": 117.73926614394152} {"canonical_length": 20, "canonical_nll": 0.46318774223327636, "generation_rouge_l_f1": 0.8823529411764706, "generation_token_f1": 0.8823529411764706, "prompt_id": "d3f0dad3a37f19c3", "reading_canonical_log_prob": -9.263754844665527, "reading_distances": [16, 14, 16, 17], "reading_log_probs": [-8.247053146362305, -8.021060943603516, -7.640684604644775, -8.077155113220215], "reading_logprob_std": 0.2559231404345693, "tokenization_path_logprob_std": 56.20376006922601, "trajectory_balance_rmse": 463.49924812851674} {"canonical_length": 23, "canonical_nll": 1.45605817048446, "generation_rouge_l_f1": 0.1904761904761905, "generation_token_f1": 0.1904761904761905, "prompt_id": "448ac2a15956c368", "reading_canonical_log_prob": -33.48933792114258, "reading_distances": [29, 35, 41, 31], "reading_log_probs": [-32.39417266845703, -33.56029510498047, -34.98176956176758, -33.074195861816406], "reading_logprob_std": 1.0959641983579542, "tokenization_path_logprob_std": 31.695803097909547, "trajectory_balance_rmse": 463.08613594323333} {"canonical_length": 165, "canonical_nll": 1.446948612097538, "generation_rouge_l_f1": 0.3130841121495327, "generation_token_f1": 0.46728971962616817, "prompt_id": "e00cf91341eb5691", "reading_canonical_log_prob": -238.74652099609375, "reading_distances": [14, 14, 9, 17], "reading_log_probs": [-241.31076049804688, -242.74908447265625, -242.48773193359375, -242.763916015625], "reading_logprob_std": 0.6898370185416881, "tokenization_path_logprob_std": 103.64450303583402, "trajectory_balance_rmse": 3918.1270701294025} {"canonical_length": 224, "canonical_nll": 1.0358702114650182, "generation_rouge_l_f1": 0.29959514170040485, "generation_token_f1": 0.6356275303643725, "prompt_id": "bf8645b4ff7492c2", "reading_canonical_log_prob": -232.03492736816406, "reading_distances": [52, 55, 60, 48], "reading_log_probs": [-236.21670532226562, -237.80862426757812, -238.10308837890625, -233.01312255859375], "reading_logprob_std": 2.3335753557971053, "tokenization_path_logprob_std": 244.13892452711835, "trajectory_balance_rmse": 4817.803443385328} {"canonical_length": 261, "canonical_nll": 1.3828300388380028, "generation_rouge_l_f1": 0.22304832713754646, "generation_token_f1": 0.5055762081784387, "prompt_id": "93af13c34520a089", "reading_canonical_log_prob": -360.91864013671875, "reading_distances": [25, 28, 26, 22], "reading_log_probs": [-368.52960205078125, -358.216064453125, -357.51495361328125, -362.21282958984375], "reading_logprob_std": 5.050815767133138, "tokenization_path_logprob_std": 255.09824519317118, "trajectory_balance_rmse": 5764.785676562648} {"canonical_length": 107, "canonical_nll": 1.0739164263288552, "generation_rouge_l_f1": 0.2958904109589041, "generation_token_f1": 0.410958904109589, "prompt_id": "8f7942ab9e010ef0", "reading_canonical_log_prob": -114.9090576171875, "reading_distances": [34, 31, 34, 35], "reading_log_probs": [-123.83770751953125, -121.0377197265625, -120.87286376953125, -120.16697692871094], "reading_logprob_std": 1.6173058850985913, "tokenization_path_logprob_std": 131.74841967055798, "trajectory_balance_rmse": 1803.8296032362866} {"canonical_length": 39, "canonical_nll": 2.748377873347356, "generation_rouge_l_f1": 0.2692307692307692, "generation_token_f1": 0.4230769230769231, "prompt_id": "ba9caab6c5bf7b20", "reading_canonical_log_prob": -107.18673706054688, "reading_distances": [38, 37, 40, 36], "reading_log_probs": [-111.89613342285156, -116.83683013916016, -117.68527221679688, -120.79004669189453], "reading_logprob_std": 3.6857123335534836, "tokenization_path_logprob_std": 30.505169000417954, "trajectory_balance_rmse": 755.79325907922} {"canonical_length": 292, "canonical_nll": 1.4304444822546554, "generation_rouge_l_f1": 0.20071684587813618, "generation_token_f1": 0.37992831541218636, "prompt_id": "90e4d685d3873c08", "reading_canonical_log_prob": -417.6897888183594, "reading_distances": [30, 30, 22, 20], "reading_log_probs": [-423.179443359375, -419.99420166015625, -418.98272705078125, -420.78662109375], "reading_logprob_std": 1.7885942807722621, "tokenization_path_logprob_std": 271.8061542316994, "trajectory_balance_rmse": 6464.720206868021} {"canonical_length": 15, "canonical_nll": 2.004952239990234, "generation_rouge_l_f1": 0.0970873786407767, "generation_token_f1": 0.0970873786407767, "prompt_id": "ec455005996bdab4", "reading_canonical_log_prob": -30.074283599853516, "reading_distances": [42, 34, 48, 44], "reading_log_probs": [-33.727718353271484, -32.8824462890625, -34.02927017211914, -34.54839324951172], "reading_logprob_std": 0.6975551776599235, "tokenization_path_logprob_std": 59.933861507741604, "trajectory_balance_rmse": 258.45064980437746} {"canonical_length": 243, "canonical_nll": 1.631054787969393, "generation_rouge_l_f1": 0.19413092550790068, "generation_token_f1": 0.41986455981941306, "prompt_id": "ab90737e4575d4fd", "reading_canonical_log_prob": -396.3463134765625, "reading_distances": [11, 17, 14, 17], "reading_log_probs": [-399.6031799316406, -403.7411804199219, -407.8140869140625, -406.402099609375], "reading_logprob_std": 3.610518391804415, "tokenization_path_logprob_std": 391.792573086596, "trajectory_balance_rmse": 5155.736807990678} {"canonical_length": 240, "canonical_nll": 1.4541946411132813, "generation_rouge_l_f1": 0.19253438113948917, "generation_token_f1": 0.4086444007858547, "prompt_id": "ec5a98de132967a0", "reading_canonical_log_prob": -349.0067138671875, "reading_distances": [33, 32, 29, 37], "reading_log_probs": [-362.9737548828125, -357.5810546875, -365.186279296875, -362.3010559082031], "reading_logprob_std": 3.199887634417912, "tokenization_path_logprob_std": 178.8213868782278, "trajectory_balance_rmse": 5364.0319617590185} {"canonical_length": 37, "canonical_nll": 0.8292725021774704, "generation_rouge_l_f1": 0.06060606060606061, "generation_token_f1": 0.06060606060606061, "prompt_id": "71251a192c44eeb3", "reading_canonical_log_prob": -30.683082580566406, "reading_distances": [22, 24, 25, 22], "reading_log_probs": [-36.53017807006836, -39.6115608215332, -38.9028205871582, -37.51078796386719], "reading_logprob_std": 1.3826081055496247, "tokenization_path_logprob_std": 69.14778397795415, "trajectory_balance_rmse": 652.9821174925908} {"canonical_length": 221, "canonical_nll": 1.8076597187853507, "generation_rouge_l_f1": 0.2707423580786026, "generation_token_f1": 0.5021834061135371, "prompt_id": "646766dc50567beb", "reading_canonical_log_prob": -399.4927978515625, "reading_distances": [44, 41, 39, 54], "reading_log_probs": [-395.6224060058594, -388.610595703125, -393.56365966796875, -395.18621826171875], "reading_logprob_std": 3.2145312127965764, "tokenization_path_logprob_std": 243.23832692809933, "trajectory_balance_rmse": 4155.003107850369} {"canonical_length": 87, "canonical_nll": 1.1432204849418552, "generation_rouge_l_f1": 0.27472527472527475, "generation_token_f1": 0.5494505494505495, "prompt_id": "3175f6135d7796a6", "reading_canonical_log_prob": -99.4601821899414, "reading_distances": [40, 40, 42, 47], "reading_log_probs": [-98.74130249023438, -95.73139953613281, -95.13105010986328, -95.95489501953125], "reading_logprob_std": 1.605890063064253, "tokenization_path_logprob_std": 32.44683611069706, "trajectory_balance_rmse": 1470.9095191239646} {"canonical_length": 315, "canonical_nll": 0.9634448823474703, "generation_rouge_l_f1": 0.24772313296903462, "generation_token_f1": 0.42622950819672134, "prompt_id": "364270b26598f5f6", "reading_canonical_log_prob": -303.4851379394531, "reading_distances": [25, 17, 25, 26], "reading_log_probs": [-309.9404296875, -308.0914306640625, -309.80047607421875, -311.01348876953125], "reading_logprob_std": 1.2083218314662048, "tokenization_path_logprob_std": 252.3841502459774, "trajectory_balance_rmse": 5546.23697610117} {"canonical_length": 132, "canonical_nll": 1.9197305621522847, "generation_rouge_l_f1": 0.1945137157107232, "generation_token_f1": 0.3391521197007481, "prompt_id": "3b49b131c410b62c", "reading_canonical_log_prob": -253.40443420410156, "reading_distances": [18, 13, 10, 23], "reading_log_probs": [-250.92926025390625, -251.32357788085938, -256.4004211425781, -252.74691772460938], "reading_logprob_std": 2.4923430673002653, "tokenization_path_logprob_std": 212.13965225796295, "trajectory_balance_rmse": 2662.7179088218973} {"canonical_length": 9, "canonical_nll": 0.4803614086574978, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.8571428571428571, "prompt_id": "300deaab60201ff0", "reading_canonical_log_prob": -4.3232526779174805, "reading_distances": [31, 19, 23, 20], "reading_log_probs": [-4.658048629760742, -3.1460824012756348, -3.7048749923706055, -4.385225772857666], "reading_logprob_std": 0.681885670405448, "tokenization_path_logprob_std": 55.637550906077585, "trajectory_balance_rmse": 165.14596868178594} {"canonical_length": 239, "canonical_nll": 1.4592088515788442, "generation_rouge_l_f1": 0.24796747967479674, "generation_token_f1": 0.540650406504065, "prompt_id": "166e7bed707f11c1", "reading_canonical_log_prob": -348.75091552734375, "reading_distances": [21, 26, 23, 26], "reading_log_probs": [-357.8338623046875, -355.40753173828125, -360.614501953125, -355.92962646484375], "reading_logprob_std": 2.3554605188115225, "tokenization_path_logprob_std": 321.97982009234977, "trajectory_balance_rmse": 5620.235781947884} {"canonical_length": 278, "canonical_nll": 1.3704316941954249, "generation_rouge_l_f1": 0.2209737827715356, "generation_token_f1": 0.47191011235955066, "prompt_id": "af085ada31a636ce", "reading_canonical_log_prob": -380.9800109863281, "reading_distances": [29, 31, 32, 32], "reading_log_probs": [-381.679931640625, -381.10455322265625, -381.5078125, -376.78619384765625], "reading_logprob_std": 2.33477128329056, "tokenization_path_logprob_std": 66.80964729655896, "trajectory_balance_rmse": 5679.4046284743645} {"canonical_length": 157, "canonical_nll": 1.5365124356215167, "generation_rouge_l_f1": 0.195, "generation_token_f1": 0.405, "prompt_id": "f0f7445e4deb0e6e", "reading_canonical_log_prob": -241.23245239257812, "reading_distances": [28, 29, 13, 35], "reading_log_probs": [-241.73605346679688, -243.29873657226562, -243.3192138671875, -246.86488342285156], "reading_logprob_std": 2.17069242672165, "tokenization_path_logprob_std": 152.55275934533447, "trajectory_balance_rmse": 4097.211018779074} {"canonical_length": 20, "canonical_nll": 2.1304805755615233, "generation_rouge_l_f1": 0.1739130434782609, "generation_token_f1": 0.1739130434782609, "prompt_id": "b5d6efcaf5d67694", "reading_canonical_log_prob": -42.60961151123047, "reading_distances": [41, 35, 40, 28], "reading_log_probs": [-35.65675735473633, -37.68815612792969, -42.91698455810547, -40.49740982055664], "reading_logprob_std": 3.1800981547674256, "tokenization_path_logprob_std": 66.1586085969764, "trajectory_balance_rmse": 484.61380634878975} {"canonical_length": 91, "canonical_nll": 2.0406903277386674, "generation_rouge_l_f1": 0.2389078498293515, "generation_token_f1": 0.3959044368600683, "prompt_id": "b7652a39606177d7", "reading_canonical_log_prob": -185.70281982421875, "reading_distances": [24, 13, 22, 24], "reading_log_probs": [-189.59625244140625, -189.15524291992188, -189.8567657470703, -185.81809997558594], "reading_logprob_std": 1.8814061930099197, "tokenization_path_logprob_std": 133.91459060658042, "trajectory_balance_rmse": 1853.8861987177133} {"canonical_length": 18, "canonical_nll": 0.8484606213039823, "generation_rouge_l_f1": 0.09022556390977444, "generation_token_f1": 0.09022556390977444, "prompt_id": "f3723ab195d4cd31", "reading_canonical_log_prob": -15.27229118347168, "reading_distances": [25, 26, 22, 20], "reading_log_probs": [-13.790154457092285, -12.326101303100586, -13.389537811279297, -12.31911849975586], "reading_logprob_std": 0.7497019180616818, "tokenization_path_logprob_std": 83.3170105754033, "trajectory_balance_rmse": 388.2742079434421} {"canonical_length": 7, "canonical_nll": 1.057474204472133, "generation_rouge_l_f1": 0.15, "generation_token_f1": 0.15, "prompt_id": "ff3fea8e176d412d", "reading_canonical_log_prob": -7.402319431304932, "reading_distances": [15, 23, 26, 19], "reading_log_probs": [-8.499775886535645, -6.714561462402344, -6.979274749755859, -8.505514144897461], "reading_logprob_std": 0.9620264420054246, "tokenization_path_logprob_std": 30.60139358713389, "trajectory_balance_rmse": 132.59057771509129} {"canonical_length": 69, "canonical_nll": 1.1432956197987432, "generation_rouge_l_f1": 0.16296296296296298, "generation_token_f1": 0.2296296296296296, "prompt_id": "f5b5083636173713", "reading_canonical_log_prob": -78.88739776611328, "reading_distances": [16, 25, 19, 24], "reading_log_probs": [-82.37828063964844, -84.92060852050781, -87.44746398925781, -90.73457336425781], "reading_logprob_std": 3.5704802757313225, "tokenization_path_logprob_std": 105.96048596505854, "trajectory_balance_rmse": 1034.6225417409241} {"canonical_length": 291, "canonical_nll": 0.9978612526175902, "generation_rouge_l_f1": 0.26591760299625467, "generation_token_f1": 0.5093632958801498, "prompt_id": "640dfdad1fff517f", "reading_canonical_log_prob": -290.37762451171875, "reading_distances": [22, 25, 24, 22], "reading_log_probs": [-294.94281005859375, -298.0516357421875, -296.8277282714844, -293.87530517578125], "reading_logprob_std": 1.8711353306631502, "tokenization_path_logprob_std": 414.69645604672513, "trajectory_balance_rmse": 6646.265541350625} {"canonical_length": 17, "canonical_nll": 1.803090263815487, "generation_rouge_l_f1": 0.1764705882352941, "generation_token_f1": 0.1764705882352941, "prompt_id": "f39da6ed7768e21d", "reading_canonical_log_prob": -30.65253448486328, "reading_distances": [34, 32, 27, 33], "reading_log_probs": [-32.69873046875, -31.154001235961914, -33.94247055053711, -29.83245849609375], "reading_logprob_std": 1.7926430262372386, "tokenization_path_logprob_std": 43.3681886958574, "trajectory_balance_rmse": 413.66465874230306} {"canonical_length": 16, "canonical_nll": 2.1582088470458984, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.3703703703703704, "prompt_id": "968e82be4a52a789", "reading_canonical_log_prob": -34.531341552734375, "reading_distances": [34, 29, 30, 27], "reading_log_probs": [-35.846458435058594, -36.708213806152344, -36.954833984375, -38.0878791809082], "reading_logprob_std": 0.9239036346687785, "tokenization_path_logprob_std": 33.561155933396805, "trajectory_balance_rmse": 405.9092636204615} {"canonical_length": 15, "canonical_nll": 2.462449900309245, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "e36a9df5c4b484e7", "reading_canonical_log_prob": -36.93674850463867, "reading_distances": [23, 25, 23, 26], "reading_log_probs": [-38.37334442138672, -36.59328079223633, -40.402549743652344, -40.25420379638672], "reading_logprob_std": 1.7971928680322153, "tokenization_path_logprob_std": 67.24916343698939, "trajectory_balance_rmse": 313.55427436914954} {"canonical_length": 243, "canonical_nll": 1.1957823199990354, "generation_rouge_l_f1": 0.2412451361867704, "generation_token_f1": 0.5097276264591439, "prompt_id": "87e83988cc685281", "reading_canonical_log_prob": -290.5751037597656, "reading_distances": [28, 33, 37, 39], "reading_log_probs": [-295.1458740234375, -296.08929443359375, -298.4263610839844, -295.3436584472656], "reading_logprob_std": 1.5058681107014924, "tokenization_path_logprob_std": 268.60716732895264, "trajectory_balance_rmse": 5391.333222587818} {"canonical_length": 106, "canonical_nll": 0.921961802356648, "generation_rouge_l_f1": 0.24332344213649854, "generation_token_f1": 0.4035608308605341, "prompt_id": "753de9bc3063e720", "reading_canonical_log_prob": -97.72795104980469, "reading_distances": [21, 16, 16, 21], "reading_log_probs": [-97.37825012207031, -97.46588897705078, -97.63923645019531, -99.2432632446289], "reading_logprob_std": 0.8811015345597045, "tokenization_path_logprob_std": 115.17022864876277, "trajectory_balance_rmse": 2301.3954684706555} {"canonical_length": 18, "canonical_nll": 1.4631724887424045, "generation_rouge_l_f1": 0.5384615384615384, "generation_token_f1": 0.5384615384615384, "prompt_id": "52785eaaf4af8786", "reading_canonical_log_prob": -26.33710479736328, "reading_distances": [20, 24, 24, 27], "reading_log_probs": [-20.081600189208984, -20.681682586669922, -24.43603515625, -22.47723960876465], "reading_logprob_std": 1.9624890493880387, "tokenization_path_logprob_std": 57.155300409917125, "trajectory_balance_rmse": 370.0186465902879} {"canonical_length": 293, "canonical_nll": 1.6848225772584258, "generation_rouge_l_f1": 0.21428571428571427, "generation_token_f1": 0.38928571428571423, "prompt_id": "9296c91d93887868", "reading_canonical_log_prob": -493.65301513671875, "reading_distances": [19, 19, 23, 17], "reading_log_probs": [-500.71563720703125, -506.68438720703125, -500.2035827636719, -498.4983825683594], "reading_logprob_std": 3.567487448374251, "tokenization_path_logprob_std": 197.67828676718761, "trajectory_balance_rmse": 6055.606580957803} {"canonical_length": 130, "canonical_nll": 1.6522533710186298, "generation_rouge_l_f1": 0.2914572864321608, "generation_token_f1": 0.4321608040201005, "prompt_id": "dd7d0916ee5ef6b2", "reading_canonical_log_prob": -214.79293823242188, "reading_distances": [54, 60, 62, 57], "reading_log_probs": [-217.78402709960938, -215.89305114746094, -216.8459930419922, -216.9727783203125], "reading_logprob_std": 0.7748013753918763, "tokenization_path_logprob_std": 160.30870312182944, "trajectory_balance_rmse": 2595.0138284233626} {"canonical_length": 172, "canonical_nll": 1.8339873912722566, "generation_rouge_l_f1": 0.21226415094339623, "generation_token_f1": 0.4150943396226416, "prompt_id": "0d595f7cde9d737e", "reading_canonical_log_prob": -315.4458312988281, "reading_distances": [17, 18, 19, 20], "reading_log_probs": [-318.77703857421875, -321.1348876953125, -320.83837890625, -322.323486328125], "reading_logprob_std": 1.4745648041572361, "tokenization_path_logprob_std": 307.0773639401986, "trajectory_balance_rmse": 3685.2017555467096} {"canonical_length": 154, "canonical_nll": 0.8693789197252942, "generation_rouge_l_f1": 0.22377622377622378, "generation_token_f1": 0.41958041958041953, "prompt_id": "8c2391cc4e076841", "reading_canonical_log_prob": -133.8843536376953, "reading_distances": [233, 246, 247, 230], "reading_log_probs": [-134.03306579589844, -137.25759887695312, -133.96566772460938, -134.45306396484375], "reading_logprob_std": 1.568394519690787, "tokenization_path_logprob_std": 95.4047004579287, "trajectory_balance_rmse": 2607.3066928636786} {"canonical_length": 395, "canonical_nll": 0.7286645816851266, "generation_rouge_l_f1": 0.27816901408450706, "generation_token_f1": 0.39788732394366194, "prompt_id": "a3d283a16147427b", "reading_canonical_log_prob": -287.822509765625, "reading_distances": [14, 17, 17, 14], "reading_log_probs": [-294.71197509765625, -295.1353759765625, -294.95733642578125, -292.45367431640625], "reading_logprob_std": 1.2526950526045655, "tokenization_path_logprob_std": 261.6943263684734, "trajectory_balance_rmse": 5952.779481731995} {"canonical_length": 11, "canonical_nll": 1.4684443040327593, "generation_rouge_l_f1": 0.25316455696202533, "generation_token_f1": 0.25316455696202533, "prompt_id": "977c41664864ae25", "reading_canonical_log_prob": -16.15288734436035, "reading_distances": [30, 32, 35, 27], "reading_log_probs": [-16.618789672851562, -18.281152725219727, -17.471546173095703, -15.575787544250488], "reading_logprob_std": 1.1599878705306133, "tokenization_path_logprob_std": 35.219432206238025, "trajectory_balance_rmse": 277.2772033203385} {"canonical_length": 20, "canonical_nll": 1.1473117828369142, "generation_rouge_l_f1": 0.11827956989247312, "generation_token_f1": 0.16129032258064516, "prompt_id": "4c8a56abb260b4fb", "reading_canonical_log_prob": -22.94623565673828, "reading_distances": [28, 29, 25, 28], "reading_log_probs": [-25.945817947387695, -28.109580993652344, -25.621469497680664, -29.01453399658203], "reading_logprob_std": 1.6514290800039773, "tokenization_path_logprob_std": 52.43966174245402, "trajectory_balance_rmse": 241.58426894104016} {"canonical_length": 244, "canonical_nll": 0.7427403809594326, "generation_rouge_l_f1": 0.31434184675834975, "generation_token_f1": 0.5893909626719057, "prompt_id": "7a98c8dec132eec9", "reading_canonical_log_prob": -181.22865295410156, "reading_distances": [32, 35, 31, 33], "reading_log_probs": [-185.92568969726562, -183.1800537109375, -186.59603881835938, -182.80612182617188], "reading_logprob_std": 1.9124992098511555, "tokenization_path_logprob_std": 213.90119801145386, "trajectory_balance_rmse": 4647.428781247349} {"canonical_length": 419, "canonical_nll": 1.1579372069147151, "generation_rouge_l_f1": 0.23661971830985915, "generation_token_f1": 0.3887323943661972, "prompt_id": "7f79088a335cbb01", "reading_canonical_log_prob": -485.1756896972656, "reading_distances": [23, 13, 20, 16], "reading_log_probs": [-511.531494140625, -513.6880493164062, -515.63916015625, -510.48492431640625], "reading_logprob_std": 2.295864925737599, "tokenization_path_logprob_std": 172.320184329906, "trajectory_balance_rmse": 7664.206821541106} {"canonical_length": 12, "canonical_nll": 2.24161958694458, "generation_rouge_l_f1": 0.4285714285714285, "generation_token_f1": 0.4285714285714285, "prompt_id": "48b64f1a7aabf6e4", "reading_canonical_log_prob": -26.89943504333496, "reading_distances": [30, 23, 21, 27], "reading_log_probs": [-26.46677017211914, -26.968368530273438, -26.386463165283203, -26.435714721679688], "reading_logprob_std": 0.2713813765830527, "tokenization_path_logprob_std": 45.29882442355132, "trajectory_balance_rmse": 303.35834184001766} {"canonical_length": 319, "canonical_nll": 1.4750834976244123, "generation_rouge_l_f1": 0.21342512908777972, "generation_token_f1": 0.42340791738382094, "prompt_id": "735c6c6b04d8c3fe", "reading_canonical_log_prob": -470.5516357421875, "reading_distances": [35, 38, 37, 39], "reading_log_probs": [-491.60791015625, -492.5673828125, -491.998779296875, -490.170654296875], "reading_logprob_std": 1.0226081613415543, "tokenization_path_logprob_std": 329.8693705837692, "trajectory_balance_rmse": 6642.84170082723} {"canonical_length": 57, "canonical_nll": 1.2228738885176809, "generation_rouge_l_f1": 0.33082706766917297, "generation_token_f1": 0.42105263157894735, "prompt_id": "eab393eed1fb7651", "reading_canonical_log_prob": -69.70381164550781, "reading_distances": [35, 35, 37, 35], "reading_log_probs": [-73.27839660644531, -80.99259948730469, -73.61702728271484, -78.61376953125], "reading_logprob_std": 3.7981918008435978, "tokenization_path_logprob_std": 90.44909989781576, "trajectory_balance_rmse": 1237.399308867034} {"canonical_length": 20, "canonical_nll": 0.02195672392845154, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "691dfdc26548dcb9", "reading_canonical_log_prob": -0.43913447856903076, "reading_distances": [30, 31, 33, 31], "reading_log_probs": [-1.0295456647872925, -0.6387479305267334, -0.608250617980957, -0.7907854318618774], "reading_logprob_std": 0.1924801462828111, "tokenization_path_logprob_std": 42.554801584074596, "trajectory_balance_rmse": 315.98360747370293} {"canonical_length": 230, "canonical_nll": 1.9428148352581522, "generation_rouge_l_f1": 0.2619047619047619, "generation_token_f1": 0.46190476190476193, "prompt_id": "2b073d802d00719d", "reading_canonical_log_prob": -446.847412109375, "reading_distances": [24, 28, 25, 21], "reading_log_probs": [-460.0274963378906, -448.8948669433594, -457.33966064453125, -452.0867614746094], "reading_logprob_std": 5.027514703797078, "tokenization_path_logprob_std": 50.15612113120061, "trajectory_balance_rmse": 4277.492977159043} {"canonical_length": 11, "canonical_nll": 4.296413074840199, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "11a956113e5f3fb8", "reading_canonical_log_prob": -47.26054382324219, "reading_distances": [29, 30, 27, 26], "reading_log_probs": [-38.402549743652344, -42.59178924560547, -43.77184295654297, -38.98084259033203], "reading_logprob_std": 2.647303767423877, "tokenization_path_logprob_std": 0.0, "trajectory_balance_rmse": 47.12293243408203} {"canonical_length": 162, "canonical_nll": 1.227442611882716, "generation_rouge_l_f1": 0.3, "generation_token_f1": 0.4263157894736842, "prompt_id": "b48c6b78ff2012e6", "reading_canonical_log_prob": -198.845703125, "reading_distances": [30, 20, 32, 22], "reading_log_probs": [-202.38882446289062, -200.84457397460938, -198.75750732421875, -197.65634155273438], "reading_logprob_std": 2.1154354843712913, "tokenization_path_logprob_std": 190.30816862917177, "trajectory_balance_rmse": 3006.238092689396} {"canonical_length": 142, "canonical_nll": 0.8904308265363666, "generation_rouge_l_f1": 0.2934782608695652, "generation_token_f1": 0.4673913043478261, "prompt_id": "c057418a3f03722b", "reading_canonical_log_prob": -126.44117736816406, "reading_distances": [34, 36, 34, 30], "reading_log_probs": [-125.81670379638672, -122.88922882080078, -123.20645141601562, -126.41250610351562], "reading_logprob_std": 1.79191255497548, "tokenization_path_logprob_std": 207.66151845513002, "trajectory_balance_rmse": 2794.5671068544475} {"canonical_length": 239, "canonical_nll": 1.2809803216527196, "generation_rouge_l_f1": 0.20338983050847456, "generation_token_f1": 0.461864406779661, "prompt_id": "14f504daadaa6ae8", "reading_canonical_log_prob": -306.154296875, "reading_distances": [23, 17, 20, 23], "reading_log_probs": [-313.483154296875, -314.3140563964844, -314.2359619140625, -316.6964111328125], "reading_logprob_std": 1.3939649773597595, "tokenization_path_logprob_std": 328.91427473159837, "trajectory_balance_rmse": 5803.9988769953725} {"canonical_length": 97, "canonical_nll": 1.509129475072487, "generation_rouge_l_f1": 0.23574144486692014, "generation_token_f1": 0.37262357414448666, "prompt_id": "20abb31273f6c931", "reading_canonical_log_prob": -146.38555908203125, "reading_distances": [20, 23, 30, 29], "reading_log_probs": [-146.87161254882812, -145.15060424804688, -148.7051239013672, -144.3602294921875], "reading_logprob_std": 1.931495274735143, "tokenization_path_logprob_std": 209.53236493405004, "trajectory_balance_rmse": 2316.277401119364} {"canonical_length": 5, "canonical_nll": 0.46824045181274415, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "609097a0b2afbf5d", "reading_canonical_log_prob": -2.3412022590637207, "reading_distances": [26, 26, 23, 33], "reading_log_probs": [-1.9891033172607422, -1.8504395484924316, -2.1703388690948486, -1.864703893661499], "reading_logprob_std": 0.14818371696495622, "tokenization_path_logprob_std": 32.59278132254577, "trajectory_balance_rmse": 74.59762449835642} {"canonical_length": 70, "canonical_nll": 2.1439442225864958, "generation_rouge_l_f1": 0.10591900311526481, "generation_token_f1": 0.18691588785046728, "prompt_id": "b13cf3764c885d75", "reading_canonical_log_prob": -150.0760955810547, "reading_distances": [30, 22, 25, 35], "reading_log_probs": [-158.11087036132812, -156.6298828125, -157.46780395507812, -158.4902801513672], "reading_logprob_std": 0.814425051955366, "tokenization_path_logprob_std": 74.57845929856326, "trajectory_balance_rmse": 1497.0168374948844} {"canonical_length": 42, "canonical_nll": 0.8199640001569476, "generation_rouge_l_f1": 0.07142857142857142, "generation_token_f1": 0.07142857142857142, "prompt_id": "d289e388118f2de0", "reading_canonical_log_prob": -34.4384880065918, "reading_distances": [21, 22, 26, 20], "reading_log_probs": [-36.02216339111328, -37.258487701416016, -39.096458435058594, -38.71445083618164], "reading_logprob_std": 1.4104400725030195, "tokenization_path_logprob_std": 90.33121244850567, "trajectory_balance_rmse": 755.7071334307919} {"canonical_length": 10, "canonical_nll": 2.7958442687988283, "generation_rouge_l_f1": 0.09090909090909093, "generation_token_f1": 0.09090909090909093, "prompt_id": "58e5e1069ca06eb2", "reading_canonical_log_prob": -27.95844268798828, "reading_distances": [23, 12, 20, 22], "reading_log_probs": [-36.45415115356445, -33.88186264038086, -29.00971221923828, -32.718204498291016], "reading_logprob_std": 3.093521478222701, "tokenization_path_logprob_std": 61.607546791625104, "trajectory_balance_rmse": 280.25421467578235} {"canonical_length": 113, "canonical_nll": 1.5215959126970409, "generation_rouge_l_f1": 0.3223684210526316, "generation_token_f1": 0.4539473684210526, "prompt_id": "ec3868510227235c", "reading_canonical_log_prob": -171.94033813476562, "reading_distances": [34, 28, 26, 26], "reading_log_probs": [-174.6104736328125, -174.25167846679688, -172.39987182617188, -172.22528076171875], "reading_logprob_std": 1.2339164235673552, "tokenization_path_logprob_std": 137.9640761313327, "trajectory_balance_rmse": 2518.7581536235743} {"canonical_length": 16, "canonical_nll": 2.354593276977539, "generation_rouge_l_f1": 0.1818181818181818, "generation_token_f1": 0.1818181818181818, "prompt_id": "d625c7ada7581821", "reading_canonical_log_prob": -37.673492431640625, "reading_distances": [11, 10, 11, 10], "reading_log_probs": [-41.25695037841797, -41.26506042480469, -42.10004425048828, -54.54655456542969], "reading_logprob_std": 6.51495303202542, "tokenization_path_logprob_std": 21.284500517769548, "trajectory_balance_rmse": 396.0066764246753} {"canonical_length": 7, "canonical_nll": 2.592205047607422, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "34ca0227975643ca", "reading_canonical_log_prob": -18.145435333251953, "reading_distances": [23, 19, 21, 16], "reading_log_probs": [-17.204221725463867, -14.047368049621582, -16.645023345947266, -13.48835563659668], "reading_logprob_std": 1.850922015564755, "tokenization_path_logprob_std": 20.459531200974208, "trajectory_balance_rmse": 86.9465797734239} {"canonical_length": 92, "canonical_nll": 0.4850777335788893, "generation_rouge_l_f1": 0.37820512820512825, "generation_token_f1": 0.467948717948718, "prompt_id": "03d8c4a193a46caa", "reading_canonical_log_prob": -44.62715148925781, "reading_distances": [29, 29, 32, 23], "reading_log_probs": [-43.81451416015625, -44.30254364013672, -50.09685516357422, -49.354331970214844], "reading_logprob_std": 3.291928445461082, "tokenization_path_logprob_std": 87.03006782037026, "trajectory_balance_rmse": 1294.5560708093253} {"canonical_length": 22, "canonical_nll": 1.1012698086825283, "generation_rouge_l_f1": 0.2608695652173913, "generation_token_f1": 0.4782608695652174, "prompt_id": "cfb9b15522adbd99", "reading_canonical_log_prob": -24.227935791015625, "reading_distances": [22, 14, 16, 16], "reading_log_probs": [-27.264705657958984, -24.631160736083984, -26.983150482177734, -29.233890533447266], "reading_logprob_std": 1.885809494135285, "tokenization_path_logprob_std": 18.40661026768406, "trajectory_balance_rmse": 538.2047564931102} {"canonical_length": 115, "canonical_nll": 1.440857862389606, "generation_rouge_l_f1": 0.3322259136212624, "generation_token_f1": 0.4053156146179402, "prompt_id": "7e152e3b35975991", "reading_canonical_log_prob": -165.6986541748047, "reading_distances": [25, 29, 33, 23], "reading_log_probs": [-176.27651977539062, -174.37417602539062, -182.71737670898438, -172.54104614257812], "reading_logprob_std": 4.430804852153253, "tokenization_path_logprob_std": 73.77905002460972, "trajectory_balance_rmse": 3088.0125046204107} {"canonical_length": 10, "canonical_nll": 2.679336929321289, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.6, "prompt_id": "6df3b4a03782fc55", "reading_canonical_log_prob": -26.79336929321289, "reading_distances": [26, 26, 24, 27], "reading_log_probs": [-22.77791976928711, -24.503883361816406, -23.00358009338379, -23.55160140991211], "reading_logprob_std": 0.7684689519975163, "tokenization_path_logprob_std": 53.0951651246041, "trajectory_balance_rmse": 285.796289394488} {"canonical_length": 150, "canonical_nll": 1.5840472412109374, "generation_rouge_l_f1": 0.1766109785202864, "generation_token_f1": 0.39618138424821003, "prompt_id": "68f61446f4cfbca0", "reading_canonical_log_prob": -237.60708618164062, "reading_distances": [20, 26, 25, 32], "reading_log_probs": [-242.44003295898438, -238.77468872070312, -241.42832946777344, -239.21719360351562], "reading_logprob_std": 1.7552712859275277, "tokenization_path_logprob_std": 216.88814585687604, "trajectory_balance_rmse": 2964.8097451947447} {"canonical_length": 206, "canonical_nll": 1.380712416565534, "generation_rouge_l_f1": 0.24539877300613494, "generation_token_f1": 0.38036809815950917, "prompt_id": "6a128630a7f46740", "reading_canonical_log_prob": -284.4267578125, "reading_distances": [26, 28, 25, 28], "reading_log_probs": [-283.3546447753906, -283.4618835449219, -284.12158203125, -285.49713134765625], "reading_logprob_std": 0.98571096396284, "tokenization_path_logprob_std": 131.41942195751992, "trajectory_balance_rmse": 4234.201505402516} {"canonical_length": 324, "canonical_nll": 1.2371469191562983, "generation_rouge_l_f1": 0.19130434782608693, "generation_token_f1": 0.3652173913043479, "prompt_id": "fc8ef010d8778b0f", "reading_canonical_log_prob": -400.8356018066406, "reading_distances": [41, 46, 39, 39], "reading_log_probs": [-405.92584228515625, -405.77935791015625, -408.5776062011719, -404.79638671875], "reading_logprob_std": 1.6182044074230593, "tokenization_path_logprob_std": 283.8285523481133, "trajectory_balance_rmse": 7262.1711108888285} {"canonical_length": 54, "canonical_nll": 0.4472892196090133, "generation_rouge_l_f1": 0.5217391304347827, "generation_token_f1": 0.5797101449275361, "prompt_id": "68eb421daa21c33e", "reading_canonical_log_prob": -24.15361785888672, "reading_distances": [20, 21, 24, 17], "reading_log_probs": [-25.4620361328125, -21.392616271972656, -24.48172378540039, -27.03857421875], "reading_logprob_std": 2.3798384599154327, "tokenization_path_logprob_std": 48.20344526135814, "trajectory_balance_rmse": 686.2764466593791} {"canonical_length": 133, "canonical_nll": 1.035209254214638, "generation_rouge_l_f1": 0.3557692307692307, "generation_token_f1": 0.5480769230769231, "prompt_id": "a12c87820d75347d", "reading_canonical_log_prob": -137.68283081054688, "reading_distances": [22, 28, 23, 20], "reading_log_probs": [-139.33572387695312, -141.76797485351562, -143.54095458984375, -141.2254180908203], "reading_logprob_std": 1.731335915991782, "tokenization_path_logprob_std": 81.65878301580624, "trajectory_balance_rmse": 2528.6236798923046} {"canonical_length": 220, "canonical_nll": 0.0498377496545965, "generation_rouge_l_f1": 0.8924731182795699, "generation_token_f1": 0.9032258064516129, "prompt_id": "c386599fdaa650e0", "reading_canonical_log_prob": -10.96430492401123, "reading_distances": [14, 10, 11, 12], "reading_log_probs": [-14.03879451751709, -14.13764762878418, -14.15438461303711, -15.91136360168457], "reading_logprob_std": 0.9019872333021989, "tokenization_path_logprob_std": 154.48082532721043, "trajectory_balance_rmse": 2489.65633977571} {"canonical_length": 13, "canonical_nll": 1.0687343157254732, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.3157894736842105, "prompt_id": "28aa44aad6506016", "reading_canonical_log_prob": -13.893546104431152, "reading_distances": [24, 28, 30, 33], "reading_log_probs": [-19.765056610107422, -17.500770568847656, -14.13548469543457, -15.170601844787598], "reading_logprob_std": 2.5125416848109756, "tokenization_path_logprob_std": 38.83506644582159, "trajectory_balance_rmse": 273.90001456878394} {"canonical_length": 150, "canonical_nll": 1.3763034057617187, "generation_rouge_l_f1": 0.26107226107226106, "generation_token_f1": 0.4289044289044289, "prompt_id": "764869f6f55c420d", "reading_canonical_log_prob": -206.4455108642578, "reading_distances": [11, 13, 15, 13], "reading_log_probs": [-210.1085662841797, -206.5556182861328, -206.43124389648438, -203.40347290039062], "reading_logprob_std": 2.7420081362577595, "tokenization_path_logprob_std": 136.65815166591327, "trajectory_balance_rmse": 2435.3614826664248} {"canonical_length": 266, "canonical_nll": 0.7573311597781074, "generation_rouge_l_f1": 0.3431952662721893, "generation_token_f1": 0.5364891518737673, "prompt_id": "8fb6a1df0bc7d753", "reading_canonical_log_prob": -201.45008850097656, "reading_distances": [44, 32, 42, 44], "reading_log_probs": [-213.2329864501953, -213.01451110839844, -215.59774780273438, -215.38772583007812], "reading_logprob_std": 1.373320146209321, "tokenization_path_logprob_std": 173.8889912358208, "trajectory_balance_rmse": 4283.193716321541} {"canonical_length": 81, "canonical_nll": 1.7153218587239583, "generation_rouge_l_f1": 0.1941176470588235, "generation_token_f1": 0.2823529411764706, "prompt_id": "f5aeb1160ca2b326", "reading_canonical_log_prob": -138.94107055664062, "reading_distances": [52, 48, 48, 53], "reading_log_probs": [-147.32296752929688, -143.7086181640625, -145.85177612304688, -145.63137817382812], "reading_logprob_std": 1.4840290704604635, "tokenization_path_logprob_std": 123.51526653493941, "trajectory_balance_rmse": 1721.4780119965392} {"canonical_length": 29, "canonical_nll": 0.8835405152419518, "generation_rouge_l_f1": 0.46938775510204084, "generation_token_f1": 0.48979591836734687, "prompt_id": "7d27494a4e6427f6", "reading_canonical_log_prob": -25.6226749420166, "reading_distances": [38, 36, 37, 27], "reading_log_probs": [-25.16912078857422, -27.638965606689453, -26.95913314819336, -27.541013717651367], "reading_logprob_std": 1.1452985593820024, "tokenization_path_logprob_std": 28.505739069134062, "trajectory_balance_rmse": 528.6257909500721} {"canonical_length": 13, "canonical_nll": 0.3535900115966797, "generation_rouge_l_f1": 0.8421052631578948, "generation_token_f1": 0.8421052631578948, "prompt_id": "b015ce761d144f7f", "reading_canonical_log_prob": -4.596670150756836, "reading_distances": [29, 23, 31, 25], "reading_log_probs": [-6.134666919708252, -5.377636432647705, -6.664711952209473, -6.686061382293701], "reading_logprob_std": 0.6142118112036276, "tokenization_path_logprob_std": 54.34840060421439, "trajectory_balance_rmse": 184.26172852469116} {"canonical_length": 246, "canonical_nll": 1.409233031234121, "generation_rouge_l_f1": 0.26459143968871596, "generation_token_f1": 0.49416342412451364, "prompt_id": "e7d47f51d6085f91", "reading_canonical_log_prob": -346.67132568359375, "reading_distances": [24, 16, 22, 18], "reading_log_probs": [-357.73480224609375, -358.8616638183594, -356.4908752441406, -351.8775329589844], "reading_logprob_std": 3.0660292488206258, "tokenization_path_logprob_std": 194.38023866925928, "trajectory_balance_rmse": 5716.768526970191} {"canonical_length": 205, "canonical_nll": 1.5418461497237044, "generation_rouge_l_f1": 0.28837209302325584, "generation_token_f1": 0.5627906976744186, "prompt_id": "321f893720a9d001", "reading_canonical_log_prob": -316.0784606933594, "reading_distances": [29, 23, 37, 23], "reading_log_probs": [-319.5018615722656, -316.3453063964844, -317.2861022949219, -318.25299072265625], "reading_logprob_std": 1.3506892621428406, "tokenization_path_logprob_std": 387.38530700796963, "trajectory_balance_rmse": 4715.169961245625} {"canonical_length": 159, "canonical_nll": 1.2211730765096795, "generation_rouge_l_f1": 0.30267062314540055, "generation_token_f1": 0.5163204747774481, "prompt_id": "46617b87eb38bf31", "reading_canonical_log_prob": -194.16651916503906, "reading_distances": [34, 38, 37, 34], "reading_log_probs": [-201.6573486328125, -192.08018493652344, -195.66741943359375, -198.44711303710938], "reading_logprob_std": 4.072669580687809, "tokenization_path_logprob_std": 107.06863634051093, "trajectory_balance_rmse": 1831.1870212250813} {"canonical_length": 4, "canonical_nll": 0.7052232623100281, "generation_rouge_l_f1": 0.04081632653061225, "generation_token_f1": 0.04081632653061225, "prompt_id": "03d95aafdc590416", "reading_canonical_log_prob": -2.8208930492401123, "reading_distances": [38, 34, 23, 35], "reading_log_probs": [-4.603021144866943, -4.259190082550049, -4.627546310424805, -3.575974941253662], "reading_logprob_std": 0.49006094620017165, "tokenization_path_logprob_std": 19.482936776841495, "trajectory_balance_rmse": 69.48325201821048} {"canonical_length": 153, "canonical_nll": 0.6095874761444291, "generation_rouge_l_f1": 0.34408602150537637, "generation_token_f1": 0.5483870967741936, "prompt_id": "3e79c37d02563d09", "reading_canonical_log_prob": -93.26688385009766, "reading_distances": [39, 36, 39, 30], "reading_log_probs": [-112.94430541992188, -117.19778442382812, -111.33922576904297, -118.29144287109375], "reading_logprob_std": 3.330570073687869, "tokenization_path_logprob_std": 119.97001609000017, "trajectory_balance_rmse": 2161.8138925013577} {"canonical_length": 385, "canonical_nll": 0.971715040330763, "generation_rouge_l_f1": 0.19047619047619047, "generation_token_f1": 0.4217687074829932, "prompt_id": "8a6a50a6d59a29d8", "reading_canonical_log_prob": -374.11029052734375, "reading_distances": [35, 25, 33, 31], "reading_log_probs": [-402.3468017578125, -401.1480712890625, -396.1358642578125, -390.9732666015625], "reading_logprob_std": 5.201532511587911, "tokenization_path_logprob_std": 201.8928895804661, "trajectory_balance_rmse": 7198.508853301693} {"canonical_length": 28, "canonical_nll": 0.7999698775155204, "generation_rouge_l_f1": 0.2676056338028169, "generation_token_f1": 0.29577464788732394, "prompt_id": "280d0259825e43a6", "reading_canonical_log_prob": -22.39915657043457, "reading_distances": [32, 33, 36, 36], "reading_log_probs": [-23.85266876220703, -25.100730895996094, -24.005046844482422, -24.424297332763672], "reading_logprob_std": 0.5583826548289156, "tokenization_path_logprob_std": 97.97095973912788, "trajectory_balance_rmse": 409.76869004337755} {"canonical_length": 245, "canonical_nll": 1.6818837691326531, "generation_rouge_l_f1": 0.15324165029469547, "generation_token_f1": 0.33398821218074654, "prompt_id": "fcecb04380951f2a", "reading_canonical_log_prob": -412.0615234375, "reading_distances": [26, 23, 25, 26], "reading_log_probs": [-415.69952392578125, -414.77667236328125, -416.2493896484375, -414.6155700683594], "reading_logprob_std": 0.7742297169905112, "tokenization_path_logprob_std": 178.45003493391204, "trajectory_balance_rmse": 5442.7653811761165} {"canonical_length": 126, "canonical_nll": 1.6398107135106648, "generation_rouge_l_f1": 0.21022727272727273, "generation_token_f1": 0.31818181818181823, "prompt_id": "1ae061f91192abf8", "reading_canonical_log_prob": -206.61614990234375, "reading_distances": [29, 32, 30, 21], "reading_log_probs": [-226.31370544433594, -213.54978942871094, -226.6016845703125, -234.0084228515625], "reading_logprob_std": 8.49498175539005, "tokenization_path_logprob_std": 131.13979837684008, "trajectory_balance_rmse": 2458.335920018982} {"canonical_length": 205, "canonical_nll": 1.269740555926067, "generation_rouge_l_f1": 0.184, "generation_token_f1": 0.32, "prompt_id": "eedaf90695c8e20b", "reading_canonical_log_prob": -260.29681396484375, "reading_distances": [16, 17, 16, 21], "reading_log_probs": [-262.2005310058594, -263.95159912109375, -259.31951904296875, -258.81134033203125], "reading_logprob_std": 2.4322444156413057, "tokenization_path_logprob_std": 90.30073591740707, "trajectory_balance_rmse": 4507.938391821997} {"canonical_length": 119, "canonical_nll": 1.1565241453026522, "generation_rouge_l_f1": 0.19169329073482427, "generation_token_f1": 0.35143769968051114, "prompt_id": "a0b38c7913178787", "reading_canonical_log_prob": -137.62637329101562, "reading_distances": [48, 44, 45, 41], "reading_log_probs": [-121.50823974609375, -119.39783477783203, -121.41709899902344, -125.32916259765625], "reading_logprob_std": 2.4769585937930874, "tokenization_path_logprob_std": 98.04276930686918, "trajectory_balance_rmse": 1501.2545014637024} {"canonical_length": 68, "canonical_nll": 1.6770899155560661, "generation_rouge_l_f1": 0.24161073825503357, "generation_token_f1": 0.33557046979865773, "prompt_id": "4101000b96f3bbde", "reading_canonical_log_prob": -114.0421142578125, "reading_distances": [18, 19, 23, 15], "reading_log_probs": [-125.38677215576172, -121.03524780273438, -119.80836486816406, -119.40452575683594], "reading_logprob_std": 2.7411865769031865, "tokenization_path_logprob_std": 77.51909518447216, "trajectory_balance_rmse": 1171.1222738706108} {"canonical_length": 138, "canonical_nll": 1.3326126319774683, "generation_rouge_l_f1": 0.24999999999999994, "generation_token_f1": 0.45918367346938777, "prompt_id": "5539256ae94ad40f", "reading_canonical_log_prob": -183.90054321289062, "reading_distances": [63, 57, 65, 62], "reading_log_probs": [-193.08404541015625, -191.53814697265625, -195.27017211914062, -186.64682006835938], "reading_logprob_std": 3.6608527380729554, "tokenization_path_logprob_std": 281.91015470086893, "trajectory_balance_rmse": 2709.43923608802} {"canonical_length": 86, "canonical_nll": 1.2001984174861464, "generation_rouge_l_f1": 0.23283582089552238, "generation_token_f1": 0.35820895522388063, "prompt_id": "3b0f78eeffb87a9e", "reading_canonical_log_prob": -103.2170639038086, "reading_distances": [32, 33, 33, 35], "reading_log_probs": [-100.01875305175781, -103.98897552490234, -104.69995880126953, -102.9684066772461], "reading_logprob_std": 2.059978580312789, "tokenization_path_logprob_std": 27.378071950097574, "trajectory_balance_rmse": 1656.402122441084} {"canonical_length": 302, "canonical_nll": 1.4727215293227442, "generation_rouge_l_f1": 0.24456521739130435, "generation_token_f1": 0.3315217391304348, "prompt_id": "3a8d8459792a6678", "reading_canonical_log_prob": -444.76190185546875, "reading_distances": [14, 16, 18, 17], "reading_log_probs": [-439.8047180175781, -442.18621826171875, -439.7972717285156, -436.8725280761719], "reading_logprob_std": 2.174968027944168, "tokenization_path_logprob_std": 189.9044360752718, "trajectory_balance_rmse": 6508.597126604883} {"canonical_length": 289, "canonical_nll": 1.1199453809269573, "generation_rouge_l_f1": 0.23394495412844038, "generation_token_f1": 0.44954128440366975, "prompt_id": "a788a16b80a319d9", "reading_canonical_log_prob": -323.6642150878906, "reading_distances": [12, 20, 16, 20], "reading_log_probs": [-327.4328308105469, -329.73480224609375, -330.92010498046875, -329.9862976074219], "reading_logprob_std": 1.4810057114413906, "tokenization_path_logprob_std": 224.99250859593155, "trajectory_balance_rmse": 5714.962962085638} {"canonical_length": 252, "canonical_nll": 1.305753919813368, "generation_rouge_l_f1": 0.1845574387947269, "generation_token_f1": 0.42561205273069674, "prompt_id": "79833cd8d7978d26", "reading_canonical_log_prob": -329.04998779296875, "reading_distances": [30, 30, 26, 31], "reading_log_probs": [-325.6971435546875, -322.97845458984375, -334.2645568847656, -330.82208251953125], "reading_logprob_std": 5.064632876752994, "tokenization_path_logprob_std": 190.1455167816199, "trajectory_balance_rmse": 4237.6892487645755} {"canonical_length": 26, "canonical_nll": 1.674670439500075, "generation_rouge_l_f1": 0.08270676691729323, "generation_token_f1": 0.12030075187969924, "prompt_id": "7c29635f759f46ec", "reading_canonical_log_prob": -43.54143142700195, "reading_distances": [18, 22, 22, 16], "reading_log_probs": [-40.06965637207031, -39.08053207397461, -38.75297546386719, -39.904815673828125], "reading_logprob_std": 0.6359153902054822, "tokenization_path_logprob_std": 23.381595735958896, "trajectory_balance_rmse": 535.6443795121451} {"canonical_length": 346, "canonical_nll": 1.2971622709594022, "generation_rouge_l_f1": 0.34087481146304677, "generation_token_f1": 0.539969834087481, "prompt_id": "88945c2caccedba1", "reading_canonical_log_prob": -448.8181457519531, "reading_distances": [39, 36, 53, 44], "reading_log_probs": [-452.2125244140625, -459.90972900390625, -459.7850646972656, -460.12713623046875], "reading_logprob_std": 3.8666439482276154, "tokenization_path_logprob_std": 100.40065974647902, "trajectory_balance_rmse": 7365.8113973358495} {"canonical_length": 17, "canonical_nll": 1.0379810333251953, "generation_rouge_l_f1": 0.6153846153846153, "generation_token_f1": 0.7692307692307692, "prompt_id": "f3be16dc1ce038fa", "reading_canonical_log_prob": -17.64567756652832, "reading_distances": [34, 49, 48, 47], "reading_log_probs": [-17.130901336669922, -16.792709350585938, -19.9704532623291, -19.357433319091797], "reading_logprob_std": 1.5860467752867382, "tokenization_path_logprob_std": 63.018706506373626, "trajectory_balance_rmse": 448.9218140742737} {"canonical_length": 56, "canonical_nll": 1.297546659197126, "generation_rouge_l_f1": 0.33999999999999997, "generation_token_f1": 0.4, "prompt_id": "08f613a2192f5ff2", "reading_canonical_log_prob": -72.66261291503906, "reading_distances": [23, 24, 20, 20], "reading_log_probs": [-72.68090057373047, -71.22840881347656, -76.32035827636719, -71.14568328857422], "reading_logprob_std": 2.4225388003415347, "tokenization_path_logprob_std": 23.45040886051946, "trajectory_balance_rmse": 1157.4836826699884} {"canonical_length": 180, "canonical_nll": 1.784307861328125, "generation_rouge_l_f1": 0.2534562211981567, "generation_token_f1": 0.4055299539170507, "prompt_id": "ddaf5127b3985530", "reading_canonical_log_prob": -321.1754150390625, "reading_distances": [26, 27, 31, 27], "reading_log_probs": [-327.86102294921875, -338.4233703613281, -329.6527404785156, -330.294921875], "reading_logprob_std": 4.691351251877994, "tokenization_path_logprob_std": 86.28096417309833, "trajectory_balance_rmse": 3650.851444948136} {"canonical_length": 184, "canonical_nll": 1.0852619668711787, "generation_rouge_l_f1": 0.23797468354430382, "generation_token_f1": 0.40506329113924044, "prompt_id": "6fc1da282516cfc1", "reading_canonical_log_prob": -199.68820190429688, "reading_distances": [18, 12, 9, 14], "reading_log_probs": [-215.5947723388672, -211.42083740234375, -206.1622772216797, -210.49908447265625], "reading_logprob_std": 3.869428387900922, "tokenization_path_logprob_std": 204.04886131976096, "trajectory_balance_rmse": 3509.8176858501565} {"canonical_length": 80, "canonical_nll": 1.2141057968139648, "generation_rouge_l_f1": 0.13836477987421383, "generation_token_f1": 0.25786163522012584, "prompt_id": "9d2634370eff9e7e", "reading_canonical_log_prob": -97.12846374511719, "reading_distances": [39, 26, 33, 34], "reading_log_probs": [-127.24267578125, -121.67189025878906, -116.47618865966797, -126.21986389160156], "reading_logprob_std": 4.921175560230517, "tokenization_path_logprob_std": 151.75678764669405, "trajectory_balance_rmse": 1798.977635398718} {"canonical_length": 132, "canonical_nll": 0.782440185546875, "generation_rouge_l_f1": 0.4280442804428044, "generation_token_f1": 0.6420664206642067, "prompt_id": "8a81f9867fa3fdd7", "reading_canonical_log_prob": -103.2821044921875, "reading_distances": [33, 32, 31, 34], "reading_log_probs": [-108.18905639648438, -111.11659240722656, -112.76344299316406, -108.84872436523438], "reading_logprob_std": 2.103786386162294, "tokenization_path_logprob_std": 72.64999368595076, "trajectory_balance_rmse": 1771.692799165525} {"canonical_length": 33, "canonical_nll": 1.0860795685739228, "generation_rouge_l_f1": 0.71875, "generation_token_f1": 0.75, "prompt_id": "32094cae1c9e7173", "reading_canonical_log_prob": -35.84062576293945, "reading_distances": [44, 46, 39, 44], "reading_log_probs": [-39.136314392089844, -39.072017669677734, -39.998878479003906, -36.17654800415039], "reading_logprob_std": 1.6673680294246849, "tokenization_path_logprob_std": 57.01990875758484, "trajectory_balance_rmse": 639.638560921968} {"canonical_length": 23, "canonical_nll": 1.7080831113068953, "generation_rouge_l_f1": 0.10309278350515463, "generation_token_f1": 0.11340206185567009, "prompt_id": "896dee39b103c9ac", "reading_canonical_log_prob": -39.285911560058594, "reading_distances": [27, 29, 30, 26], "reading_log_probs": [-36.11298370361328, -51.51982879638672, -51.14982986450195, -48.467132568359375], "reading_logprob_std": 7.261515742851692, "tokenization_path_logprob_std": 87.92491619977568, "trajectory_balance_rmse": 616.1459011950775} {"canonical_length": 105, "canonical_nll": 1.7438334147135417, "generation_rouge_l_f1": 0.18617021276595744, "generation_token_f1": 0.33510638297872336, "prompt_id": "1057782952469d89", "reading_canonical_log_prob": -183.10250854492188, "reading_distances": [11, 10, 10, 9], "reading_log_probs": [-190.3383331298828, -188.279052734375, -186.85043334960938, -183.5902557373047], "reading_logprob_std": 2.83721045234597, "tokenization_path_logprob_std": 110.13940399704174, "trajectory_balance_rmse": 2573.428813566016} {"canonical_length": 25, "canonical_nll": 0.723299789428711, "generation_rouge_l_f1": 0.10561056105610563, "generation_token_f1": 0.1254125412541254, "prompt_id": "03db11358469fd8f", "reading_canonical_log_prob": -18.082494735717773, "reading_distances": [19, 20, 25, 24], "reading_log_probs": [-21.748138427734375, -21.08578872680664, -22.298431396484375, -21.687313079833984], "reading_logprob_std": 0.4959022576136405, "tokenization_path_logprob_std": 31.450999792561305, "trajectory_balance_rmse": 232.07486534854456} {"canonical_length": 240, "canonical_nll": 1.2088943481445313, "generation_rouge_l_f1": 0.21074380165289258, "generation_token_f1": 0.5247933884297521, "prompt_id": "85339fcc838b71e5", "reading_canonical_log_prob": -290.1346435546875, "reading_distances": [21, 19, 22, 25], "reading_log_probs": [-289.04901123046875, -281.95428466796875, -285.04595947265625, -287.93792724609375], "reading_logprob_std": 3.179622425121081, "tokenization_path_logprob_std": 175.66262689006234, "trajectory_balance_rmse": 5883.977894227989} {"canonical_length": 28, "canonical_nll": 0.47996207645961214, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.275, "prompt_id": "5af4615607f58b8b", "reading_canonical_log_prob": -13.43893814086914, "reading_distances": [47, 45, 38, 39], "reading_log_probs": [-18.093000411987305, -17.239721298217773, -17.182279586791992, -16.713199615478516], "reading_logprob_std": 0.5745946801572095, "tokenization_path_logprob_std": 89.47365611118524, "trajectory_balance_rmse": 618.0415886082958} {"canonical_length": 65, "canonical_nll": 2.0002483661358172, "generation_rouge_l_f1": 0.2649006622516556, "generation_token_f1": 0.4105960264900662, "prompt_id": "13f18463e2c1c5f7", "reading_canonical_log_prob": -130.01614379882812, "reading_distances": [25, 18, 20, 18], "reading_log_probs": [-128.73880004882812, -127.22338104248047, -127.25580596923828, -128.72296142578125], "reading_logprob_std": 0.8611211152908864, "tokenization_path_logprob_std": 57.71413590506524, "trajectory_balance_rmse": 1435.3170291929332} {"canonical_length": 64, "canonical_nll": 1.8651926517486572, "generation_rouge_l_f1": 0.16413373860182373, "generation_token_f1": 0.2613981762917933, "prompt_id": "d02d88b60737f21f", "reading_canonical_log_prob": -119.37232971191406, "reading_distances": [21, 23, 16, 19], "reading_log_probs": [-125.92314910888672, -133.63673400878906, -133.38494873046875, -130.1039581298828], "reading_logprob_std": 3.605156148129826, "tokenization_path_logprob_std": 29.31657781252133, "trajectory_balance_rmse": 1452.411739944483} {"canonical_length": 236, "canonical_nll": 1.5218880863512976, "generation_rouge_l_f1": 0.21157684630738524, "generation_token_f1": 0.4311377245508982, "prompt_id": "559f28b026670bdd", "reading_canonical_log_prob": -359.16558837890625, "reading_distances": [19, 13, 16, 14], "reading_log_probs": [-364.61907958984375, -357.3251953125, -363.4163818359375, -361.30596923828125], "reading_logprob_std": 3.2019120014973965, "tokenization_path_logprob_std": 135.12413707953718, "trajectory_balance_rmse": 5317.631968498965} {"canonical_length": 50, "canonical_nll": 0.9023011016845703, "generation_rouge_l_f1": 0.44827586206896547, "generation_token_f1": 0.5689655172413793, "prompt_id": "44cafb7b8f27398a", "reading_canonical_log_prob": -45.115055084228516, "reading_distances": [30, 23, 28, 26], "reading_log_probs": [-42.48210906982422, -41.97079849243164, -43.03776931762695, -43.08708572387695], "reading_logprob_std": 0.5262405662117051, "tokenization_path_logprob_std": 62.313456259651446, "trajectory_balance_rmse": 963.0484626370076} {"canonical_length": 274, "canonical_nll": 1.0878891770857093, "generation_rouge_l_f1": 0.2448979591836735, "generation_token_f1": 0.48237476808905383, "prompt_id": "55f3c7490c4e7162", "reading_canonical_log_prob": -298.0816345214844, "reading_distances": [14, 17, 22, 21], "reading_log_probs": [-302.7383117675781, -305.1336975097656, -305.3734130859375, -306.3791809082031], "reading_logprob_std": 1.5426660169059514, "tokenization_path_logprob_std": 301.3509789001351, "trajectory_balance_rmse": 4883.297482160699} {"canonical_length": 308, "canonical_nll": 1.1798935927353895, "generation_rouge_l_f1": 0.24179620034542315, "generation_token_f1": 0.40759930915371334, "prompt_id": "41cd8f9fa13781b8", "reading_canonical_log_prob": -363.4072265625, "reading_distances": [28, 27, 31, 17], "reading_log_probs": [-374.8011169433594, -371.0362854003906, -379.9033203125, -374.30963134765625], "reading_logprob_std": 3.663746396748457, "tokenization_path_logprob_std": 219.86850195154062, "trajectory_balance_rmse": 6423.44009363113} {"canonical_length": 373, "canonical_nll": 1.2715230742344588, "generation_rouge_l_f1": 0.20421393841166938, "generation_token_f1": 0.40842787682333875, "prompt_id": "28de6d5b6a032c7b", "reading_canonical_log_prob": -474.2781066894531, "reading_distances": [26, 21, 23, 25], "reading_log_probs": [-473.7197570800781, -472.92474365234375, -472.9901428222656, -470.7062683105469], "reading_logprob_std": 1.3034406570828225, "tokenization_path_logprob_std": 226.128323477807, "trajectory_balance_rmse": 6103.413239303773} {"canonical_length": 132, "canonical_nll": 1.7377508914831914, "generation_rouge_l_f1": 0.19774011299435026, "generation_token_f1": 0.35593220338983045, "prompt_id": "dc17c375b90b9957", "reading_canonical_log_prob": -229.38311767578125, "reading_distances": [23, 24, 22, 18], "reading_log_probs": [-226.06581115722656, -225.94590759277344, -228.12107849121094, -226.9673309326172], "reading_logprob_std": 1.0065223167703141, "tokenization_path_logprob_std": 253.456940937009, "trajectory_balance_rmse": 2925.0580938536827} {"canonical_length": 24, "canonical_nll": 0.5045696496963501, "generation_rouge_l_f1": 0.8444444444444443, "generation_token_f1": 0.8444444444444443, "prompt_id": "c1a810543bd499fe", "reading_canonical_log_prob": -12.109671592712402, "reading_distances": [28, 25, 24, 27], "reading_log_probs": [-12.936262130737305, -10.894858360290527, -11.37703800201416, -11.927830696105957], "reading_logprob_std": 0.8764675712415635, "tokenization_path_logprob_std": 41.0108104097279, "trajectory_balance_rmse": 388.95591850079364} {"canonical_length": 15, "canonical_nll": 0.8746392567952473, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.34782608695652173, "prompt_id": "f21bbbf6325c37df", "reading_canonical_log_prob": -13.119588851928711, "reading_distances": [42, 39, 38, 36], "reading_log_probs": [-13.606658935546875, -12.837121963500977, -12.893911361694336, -14.07944107055664], "reading_logprob_std": 0.5969213859923634, "tokenization_path_logprob_std": 56.562067975785865, "trajectory_balance_rmse": 321.0940913959869} {"canonical_length": 141, "canonical_nll": 1.700297254197141, "generation_rouge_l_f1": 0.22164948453608244, "generation_token_f1": 0.422680412371134, "prompt_id": "ff640b48674add94", "reading_canonical_log_prob": -239.74191284179688, "reading_distances": [48, 54, 50, 41], "reading_log_probs": [-249.4596405029297, -251.47970581054688, -254.4873809814453, -244.80442810058594], "reading_logprob_std": 4.066067506499793, "tokenization_path_logprob_std": 159.92025870106, "trajectory_balance_rmse": 3363.625122878226} {"canonical_length": 133, "canonical_nll": 2.0493416463522087, "generation_rouge_l_f1": 0.18639798488664988, "generation_token_f1": 0.3324937027707808, "prompt_id": "294ad70d174d121a", "reading_canonical_log_prob": -272.56243896484375, "reading_distances": [29, 22, 27, 28], "reading_log_probs": [-285.99700927734375, -285.01025390625, -285.8017272949219, -284.4919128417969], "reading_logprob_std": 0.7004675808002687, "tokenization_path_logprob_std": 360.450696684168, "trajectory_balance_rmse": 3537.954080497838} {"canonical_length": 49, "canonical_nll": 2.463639551279496, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.4857142857142857, "prompt_id": "fdd72207f77ed671", "reading_canonical_log_prob": -120.71833801269531, "reading_distances": [19, 23, 20, 23], "reading_log_probs": [-122.53704833984375, -121.38117218017578, -124.81944274902344, -123.13701629638672], "reading_logprob_std": 1.4329627617280145, "tokenization_path_logprob_std": 84.66951807170733, "trajectory_balance_rmse": 993.8854402705509} {"canonical_length": 9, "canonical_nll": 1.5689780977037218, "generation_rouge_l_f1": 0.4761904761904762, "generation_token_f1": 0.5714285714285715, "prompt_id": "bead008713b25f77", "reading_canonical_log_prob": -14.120802879333496, "reading_distances": [36, 42, 33, 40], "reading_log_probs": [-20.865982055664062, -20.8564453125, -17.368921279907227, -19.65953826904297], "reading_logprob_std": 1.646394246239094, "tokenization_path_logprob_std": 62.76286308909611, "trajectory_balance_rmse": 224.671932106486} {"canonical_length": 353, "canonical_nll": 1.0246397023835871, "generation_rouge_l_f1": 0.21664050235478807, "generation_token_f1": 0.4175824175824176, "prompt_id": "5e76d06be1ed709a", "reading_canonical_log_prob": -361.69781494140625, "reading_distances": [31, 31, 23, 25], "reading_log_probs": [-384.16607666015625, -384.03631591796875, -376.90264892578125, -380.009521484375], "reading_logprob_std": 3.4977148074755515, "tokenization_path_logprob_std": 408.754259665412, "trajectory_balance_rmse": 6282.110961228426} {"canonical_length": 238, "canonical_nll": 1.4829837574678308, "generation_rouge_l_f1": 0.21763602251407127, "generation_token_f1": 0.4577861163227017, "prompt_id": "f6f871799dc3c062", "reading_canonical_log_prob": -352.95013427734375, "reading_distances": [30, 29, 31, 29], "reading_log_probs": [-364.44976806640625, -360.6031188964844, -359.6120910644531, -351.7735290527344], "reading_logprob_std": 5.317218191307821, "tokenization_path_logprob_std": 186.74246540917872, "trajectory_balance_rmse": 3851.1794933419515} {"canonical_length": 108, "canonical_nll": 0.8857612609863281, "generation_rouge_l_f1": 0.33333333333333326, "generation_token_f1": 0.40123456790123463, "prompt_id": "e5243c35b3e882de", "reading_canonical_log_prob": -95.66221618652344, "reading_distances": [26, 27, 23, 32], "reading_log_probs": [-95.7811508178711, -96.77645111083984, -99.61434936523438, -94.71524810791016], "reading_logprob_std": 2.1040358008458284, "tokenization_path_logprob_std": 55.24429964996057, "trajectory_balance_rmse": 1478.3873388872873} {"canonical_length": 147, "canonical_nll": 1.6091709266714498, "generation_rouge_l_f1": 0.187192118226601, "generation_token_f1": 0.33004926108374383, "prompt_id": "26ddb945cf9f0d7d", "reading_canonical_log_prob": -236.54812622070312, "reading_distances": [33, 39, 29, 28], "reading_log_probs": [-247.50604248046875, -244.41098022460938, -250.27545166015625, -245.61370849609375], "reading_logprob_std": 2.556045045926417, "tokenization_path_logprob_std": 361.7806274239504, "trajectory_balance_rmse": 3842.609110549213} {"canonical_length": 48, "canonical_nll": 1.7629261016845703, "generation_rouge_l_f1": 0.12258064516129033, "generation_token_f1": 0.16774193548387095, "prompt_id": "e4581a40938db0d9", "reading_canonical_log_prob": -84.62045288085938, "reading_distances": [19, 20, 15, 22], "reading_log_probs": [-83.92082214355469, -82.8893051147461, -81.8038558959961, -76.68573760986328], "reading_logprob_std": 3.2113030360049466, "tokenization_path_logprob_std": 84.56147525520173, "trajectory_balance_rmse": 1052.2406811059157} {"canonical_length": 10, "canonical_nll": 0.08796284794807434, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "267c9641a2f8cfb9", "reading_canonical_log_prob": -0.8796284794807434, "reading_distances": [22, 25, 25, 23], "reading_log_probs": [-4.323722839355469, -2.546391487121582, -6.408597469329834, -5.4110517501831055], "reading_logprob_std": 1.653428358800354, "tokenization_path_logprob_std": 61.12025165440978, "trajectory_balance_rmse": 194.91685984567965} {"canonical_length": 63, "canonical_nll": 0.7804948110429067, "generation_rouge_l_f1": 0.36042402826855124, "generation_token_f1": 0.4381625441696113, "prompt_id": "df9c5775b476f764", "reading_canonical_log_prob": -49.171173095703125, "reading_distances": [29, 29, 30, 25], "reading_log_probs": [-51.3177490234375, -52.518310546875, -50.25736999511719, -50.262168884277344], "reading_logprob_std": 1.075563863557099, "tokenization_path_logprob_std": 86.81339790629498, "trajectory_balance_rmse": 905.6394139124088} {"canonical_length": 106, "canonical_nll": 1.8954994633512676, "generation_rouge_l_f1": 0.16666666666666669, "generation_token_f1": 0.3148148148148148, "prompt_id": "518ad8fe3c294839", "reading_canonical_log_prob": -200.92294311523438, "reading_distances": [18, 25, 21, 17], "reading_log_probs": [-204.64645385742188, -213.2547607421875, -207.02967834472656, -206.05191040039062], "reading_logprob_std": 3.8007318212838728, "tokenization_path_logprob_std": 53.01350913119842, "trajectory_balance_rmse": 2401.0346325131595} {"canonical_length": 203, "canonical_nll": 0.6925949322179034, "generation_rouge_l_f1": 0.38048780487804873, "generation_token_f1": 0.5756097560975609, "prompt_id": "6369662548b008b0", "reading_canonical_log_prob": -140.59677124023438, "reading_distances": [23, 23, 26, 23], "reading_log_probs": [-147.28628540039062, -148.78050231933594, -148.00576782226562, -144.4176483154297], "reading_logprob_std": 1.9036965811536004, "tokenization_path_logprob_std": 130.30674271257197, "trajectory_balance_rmse": 1950.6594980882878} {"canonical_length": 68, "canonical_nll": 0.34849901760325713, "generation_rouge_l_f1": 0.2459546925566343, "generation_token_f1": 0.3042071197411003, "prompt_id": "3f0899fb36d7151b", "reading_canonical_log_prob": -23.697933197021484, "reading_distances": [32, 35, 29, 37], "reading_log_probs": [-31.352413177490234, -39.033748626708984, -35.723670959472656, -38.69707489013672], "reading_logprob_std": 3.558627531039933, "tokenization_path_logprob_std": 83.04620064403751, "trajectory_balance_rmse": 1279.6742816040392} {"canonical_length": 59, "canonical_nll": 0.8161825406349311, "generation_rouge_l_f1": 0.33628318584070793, "generation_token_f1": 0.47787610619469023, "prompt_id": "ca44a0ae267d5905", "reading_canonical_log_prob": -48.15476989746094, "reading_distances": [16, 19, 18, 20], "reading_log_probs": [-61.16929626464844, -61.049190521240234, -59.11278533935547, -56.873985290527344], "reading_logprob_std": 2.018406722010568, "tokenization_path_logprob_std": 72.49328823409648, "trajectory_balance_rmse": 1091.1559308535602} {"canonical_length": 144, "canonical_nll": 2.5788182152642145, "generation_rouge_l_f1": 0.26771653543307083, "generation_token_f1": 0.39370078740157477, "prompt_id": "b3b2b51b9f629425", "reading_canonical_log_prob": -371.3498229980469, "reading_distances": [9, 12, 15, 11], "reading_log_probs": [-397.43182373046875, -408.5779113769531, -403.2015380859375, -394.29693603515625], "reading_logprob_std": 6.321233915602566, "tokenization_path_logprob_std": 160.94454294478751, "trajectory_balance_rmse": 2964.2983033704472} {"canonical_length": 120, "canonical_nll": 1.1903419494628906, "generation_rouge_l_f1": 0.15463917525773196, "generation_token_f1": 0.365979381443299, "prompt_id": "4412ea6c558d478d", "reading_canonical_log_prob": -142.84103393554688, "reading_distances": [24, 26, 31, 24], "reading_log_probs": [-145.104248046875, -144.35198974609375, -146.4036865234375, -146.0590362548828], "reading_logprob_std": 0.9313209563375833, "tokenization_path_logprob_std": 58.19290251480254, "trajectory_balance_rmse": 2414.3590223559313} {"canonical_length": 30, "canonical_nll": 0.9533123016357422, "generation_rouge_l_f1": 0.07905138339920949, "generation_token_f1": 0.13438735177865613, "prompt_id": "7938d8548e06b8bf", "reading_canonical_log_prob": -28.599369049072266, "reading_distances": [8, 9, 6, 8], "reading_log_probs": [-29.513214111328125, -29.269187927246094, -29.789987564086914, -31.502227783203125], "reading_logprob_std": 1.0116732648423183, "tokenization_path_logprob_std": 75.61635114291626, "trajectory_balance_rmse": 317.0088348667839} {"canonical_length": 332, "canonical_nll": 1.1430117136024567, "generation_rouge_l_f1": 0.4021543985637343, "generation_token_f1": 0.5888689407540395, "prompt_id": "8829bdef994324cc", "reading_canonical_log_prob": -379.4798889160156, "reading_distances": [34, 33, 33, 35], "reading_log_probs": [-386.4954833984375, -389.44580078125, -386.59515380859375, -385.35833740234375], "reading_logprob_std": 1.7409448518405521, "tokenization_path_logprob_std": 406.5793607738287, "trajectory_balance_rmse": 7180.028305415162} {"canonical_length": 14, "canonical_nll": 0.798762457711356, "generation_rouge_l_f1": 0.055999999999999994, "generation_token_f1": 0.06399999999999999, "prompt_id": "75996f4d9b7d5f4d", "reading_canonical_log_prob": -11.182674407958984, "reading_distances": [10, 18, 19, 23], "reading_log_probs": [-11.772798538208008, -12.119029998779297, -10.228736877441406, -10.448341369628906], "reading_logprob_std": 0.9429927598289528, "tokenization_path_logprob_std": 54.64570950121636, "trajectory_balance_rmse": 230.90349502686246} {"canonical_length": 164, "canonical_nll": 1.423182231623952, "generation_rouge_l_f1": 0.1788235294117647, "generation_token_f1": 0.32941176470588235, "prompt_id": "f5d8be75fb333542", "reading_canonical_log_prob": -233.40188598632812, "reading_distances": [18, 26, 19, 19], "reading_log_probs": [-222.75009155273438, -223.9676055908203, -229.87783813476562, -223.59469604492188], "reading_logprob_std": 3.2602179540243976, "tokenization_path_logprob_std": 186.88291058223413, "trajectory_balance_rmse": 3904.9556167615056} {"canonical_length": 45, "canonical_nll": 1.4235470241970487, "generation_rouge_l_f1": 0.23655913978494625, "generation_token_f1": 0.3655913978494624, "prompt_id": "61f6f6b3601f9e9d", "reading_canonical_log_prob": -64.05961608886719, "reading_distances": [19, 15, 27, 22], "reading_log_probs": [-67.61820983886719, -66.45625305175781, -81.04208374023438, -76.11770629882812], "reading_logprob_std": 6.976935830956799, "tokenization_path_logprob_std": 92.77302407549378, "trajectory_balance_rmse": 883.4008265285205} {"canonical_length": 26, "canonical_nll": 1.4990114065317006, "generation_rouge_l_f1": 0.5217391304347826, "generation_token_f1": 0.5217391304347826, "prompt_id": "4478590e9673bdfe", "reading_canonical_log_prob": -38.97429656982422, "reading_distances": [35, 29, 29, 31], "reading_log_probs": [-43.06544494628906, -42.73284149169922, -37.319793701171875, -42.926307678222656], "reading_logprob_std": 2.797528775404783, "tokenization_path_logprob_std": 124.49521390368533, "trajectory_balance_rmse": 886.833666008349} {"canonical_length": 90, "canonical_nll": 0.8542109171549479, "generation_rouge_l_f1": 0.17522658610271902, "generation_token_f1": 0.2537764350453172, "prompt_id": "b3f4ac8bd91cf595", "reading_canonical_log_prob": -76.87898254394531, "reading_distances": [30, 32, 23, 32], "reading_log_probs": [-85.47344207763672, -88.66556549072266, -92.8803939819336, -81.30076599121094], "reading_logprob_std": 4.903711599399541, "tokenization_path_logprob_std": 48.348554877969406, "trajectory_balance_rmse": 1662.7385662642944} {"canonical_length": 116, "canonical_nll": 1.0842450898269127, "generation_rouge_l_f1": 0.21024258760107817, "generation_token_f1": 0.42048517520215634, "prompt_id": "47d49f308f57d626", "reading_canonical_log_prob": -125.77243041992188, "reading_distances": [32, 35, 39, 36], "reading_log_probs": [-137.0469207763672, -136.58961486816406, -136.68971252441406, -134.80392456054688], "reading_logprob_std": 1.005098009868366, "tokenization_path_logprob_std": 155.5500402498612, "trajectory_balance_rmse": 2859.4683295136565} {"canonical_length": 264, "canonical_nll": 1.2018337827740293, "generation_rouge_l_f1": 0.23874755381604695, "generation_token_f1": 0.5401174168297455, "prompt_id": "d8021c0fb98991e7", "reading_canonical_log_prob": -317.28411865234375, "reading_distances": [15, 24, 20, 24], "reading_log_probs": [-325.6406555175781, -323.43804931640625, -325.937255859375, -326.40863037109375], "reading_logprob_std": 1.3172512508063705, "tokenization_path_logprob_std": 307.1617168443645, "trajectory_balance_rmse": 5217.195354346857} {"canonical_length": 165, "canonical_nll": 1.5906653201941288, "generation_rouge_l_f1": 0.1747126436781609, "generation_token_f1": 0.34022988505747126, "prompt_id": "fb0e2e105f1d060e", "reading_canonical_log_prob": -262.45977783203125, "reading_distances": [35, 27, 23, 27], "reading_log_probs": [-261.4080505371094, -264.178955078125, -257.9647521972656, -265.0614929199219], "reading_logprob_std": 3.196883587292662, "tokenization_path_logprob_std": 181.76391660128255, "trajectory_balance_rmse": 3124.423474154165} {"canonical_length": 12, "canonical_nll": 1.0195817947387695, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.6, "prompt_id": "d8467aeeb0933718", "reading_canonical_log_prob": -12.234981536865234, "reading_distances": [41, 42, 45, 36], "reading_log_probs": [-16.200132369995117, -16.857746124267578, -16.574880599975586, -14.592338562011719], "reading_logprob_std": 1.0124413676408013, "tokenization_path_logprob_std": 25.63186706978468, "trajectory_balance_rmse": 222.80624395772864} {"canonical_length": 62, "canonical_nll": 0.4315523947438886, "generation_rouge_l_f1": 0.7703703703703704, "generation_token_f1": 0.7851851851851852, "prompt_id": "b80e624242a44cf6", "reading_canonical_log_prob": -26.756248474121094, "reading_distances": [23, 29, 22, 21], "reading_log_probs": [-29.713289260864258, -27.936420440673828, -28.28411102294922, -28.471139907836914], "reading_logprob_std": 0.7737646987548187, "tokenization_path_logprob_std": 162.839674204213, "trajectory_balance_rmse": 1340.4201278020294} {"canonical_length": 225, "canonical_nll": 1.2395791286892361, "generation_rouge_l_f1": 0.19542619542619544, "generation_token_f1": 0.41164241164241167, "prompt_id": "aad9f0c226359f45", "reading_canonical_log_prob": -278.9053039550781, "reading_distances": [13, 20, 17, 19], "reading_log_probs": [-284.4671630859375, -288.9100036621094, -287.9930419921875, -285.00628662109375], "reading_logprob_std": 2.1882611384414408, "tokenization_path_logprob_std": 240.49735252604495, "trajectory_balance_rmse": 4404.513424054222} {"canonical_length": 221, "canonical_nll": 0.9436954826251414, "generation_rouge_l_f1": 0.26694045174537984, "generation_token_f1": 0.5503080082135523, "prompt_id": "2fd9ba0f00f7f678", "reading_canonical_log_prob": -208.55670166015625, "reading_distances": [36, 26, 21, 28], "reading_log_probs": [-217.77528381347656, -217.0169677734375, -213.36984252929688, -217.6605224609375], "reading_logprob_std": 2.08410144008968, "tokenization_path_logprob_std": 365.120526629954, "trajectory_balance_rmse": 4227.455449269165} {"canonical_length": 8, "canonical_nll": 1.6237720251083374, "generation_rouge_l_f1": 0.0975609756097561, "generation_token_f1": 0.14634146341463417, "prompt_id": "b1a346c0437945a9", "reading_canonical_log_prob": -12.9901762008667, "reading_distances": [28, 31, 24, 31], "reading_log_probs": [-14.21794605255127, -14.854523658752441, -15.379568099975586, -13.31866455078125], "reading_logprob_std": 0.8871847962837288, "tokenization_path_logprob_std": 49.81312661315018, "trajectory_balance_rmse": 164.9344868800165} {"canonical_length": 21, "canonical_nll": 2.3531851995558966, "generation_rouge_l_f1": 0.14906832298136646, "generation_token_f1": 0.16149068322981366, "prompt_id": "c2ce9d8e95420539", "reading_canonical_log_prob": -49.41688919067383, "reading_distances": [19, 16, 19, 22], "reading_log_probs": [-62.65310287475586, -57.15496826171875, -57.24235534667969, -57.92546081542969], "reading_logprob_std": 2.628754471166701, "tokenization_path_logprob_std": 56.888699915668454, "trajectory_balance_rmse": 329.25009071399217} {"canonical_length": 86, "canonical_nll": 1.2479394424793333, "generation_rouge_l_f1": 0.245014245014245, "generation_token_f1": 0.3247863247863248, "prompt_id": "ecdc1eebf63862d2", "reading_canonical_log_prob": -107.32279205322266, "reading_distances": [32, 35, 28, 36], "reading_log_probs": [-116.16070556640625, -117.29222869873047, -117.26922607421875, -116.60374450683594], "reading_logprob_std": 0.5494581526158394, "tokenization_path_logprob_std": 131.20954712886967, "trajectory_balance_rmse": 2136.3701327869553} {"canonical_length": 149, "canonical_nll": 0.9108795575647546, "generation_rouge_l_f1": 0.32020997375328086, "generation_token_f1": 0.4881889763779528, "prompt_id": "a473b04dd30c4a84", "reading_canonical_log_prob": -135.72105407714844, "reading_distances": [50, 50, 49, 47], "reading_log_probs": [-141.81570434570312, -144.667724609375, -142.46395874023438, -148.6134490966797], "reading_logprob_std": 3.06872862414242, "tokenization_path_logprob_std": 100.23850109818082, "trajectory_balance_rmse": 3612.9152279593272} {"canonical_length": 147, "canonical_nll": 1.4726192967421343, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.31578947368421056, "prompt_id": "30cb3266dd228382", "reading_canonical_log_prob": -216.47503662109375, "reading_distances": [17, 20, 16, 12], "reading_log_probs": [-223.25173950195312, -226.3710479736328, -222.34976196289062, -220.36788940429688], "reading_logprob_std": 2.499945216658209, "tokenization_path_logprob_std": 259.2514064373747, "trajectory_balance_rmse": 2653.3872533002104} {"canonical_length": 151, "canonical_nll": 1.1027471277097993, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.456140350877193, "prompt_id": "d49ee101a0892105", "reading_canonical_log_prob": -166.5148162841797, "reading_distances": [21, 17, 24, 23], "reading_log_probs": [-172.5528564453125, -176.8837890625, -173.1692352294922, -169.7387237548828], "reading_logprob_std": 2.939310014632186, "tokenization_path_logprob_std": 144.13073783623156, "trajectory_balance_rmse": 2544.1596323370754} {"canonical_length": 18, "canonical_nll": 0.7790531582302518, "generation_rouge_l_f1": 0.08888888888888889, "generation_token_f1": 0.12222222222222223, "prompt_id": "c17572e82dc23485", "reading_canonical_log_prob": -14.022956848144531, "reading_distances": [26, 29, 24, 21], "reading_log_probs": [-16.567026138305664, -17.181446075439453, -16.644481658935547, -15.223287582397461], "reading_logprob_std": 0.8332492921843482, "tokenization_path_logprob_std": 45.12111039427707, "trajectory_balance_rmse": 488.7114619462669} {"canonical_length": 267, "canonical_nll": 1.4420511195722143, "generation_rouge_l_f1": 0.24045801526717558, "generation_token_f1": 0.5458015267175572, "prompt_id": "50c4d271f8ba07ce", "reading_canonical_log_prob": -385.02764892578125, "reading_distances": [13, 16, 12, 13], "reading_log_probs": [-387.04302978515625, -390.40484619140625, -390.2977600097656, -386.9766845703125], "reading_logprob_std": 1.9298700668056081, "tokenization_path_logprob_std": 187.22806791462648, "trajectory_balance_rmse": 5372.605988613084} {"canonical_length": 27, "canonical_nll": 2.0359827677408853, "generation_rouge_l_f1": 0.2641509433962264, "generation_token_f1": 0.3773584905660378, "prompt_id": "71a578aa99112e47", "reading_canonical_log_prob": -54.971534729003906, "reading_distances": [10, 13, 11, 16], "reading_log_probs": [-55.637908935546875, -59.21019744873047, -58.055267333984375, -59.245418548583984], "reading_logprob_std": 1.6923986051732947, "tokenization_path_logprob_std": 58.53710788235749, "trajectory_balance_rmse": 676.8606012121944} {"canonical_length": 260, "canonical_nll": 1.089328355055589, "generation_rouge_l_f1": 0.22519083969465647, "generation_token_f1": 0.4732824427480916, "prompt_id": "c22b91c04ec53dee", "reading_canonical_log_prob": -283.2253723144531, "reading_distances": [38, 39, 38, 43], "reading_log_probs": [-281.03131103515625, -277.1014404296875, -280.88140869140625, -278.283447265625], "reading_logprob_std": 1.946188856327989, "tokenization_path_logprob_std": 175.53803920282056, "trajectory_balance_rmse": 5610.996607022098} {"canonical_length": 42, "canonical_nll": 1.8268623352050781, "generation_rouge_l_f1": 0.1896551724137931, "generation_token_f1": 0.24137931034482757, "prompt_id": "ce56aa7776fbfa4a", "reading_canonical_log_prob": -76.72821807861328, "reading_distances": [18, 19, 16, 16], "reading_log_probs": [-84.53604888916016, -79.50222778320312, -85.42313385009766, -76.31365203857422], "reading_logprob_std": 4.300589954005594, "tokenization_path_logprob_std": 36.48518928962086, "trajectory_balance_rmse": 913.4330294597593} {"canonical_length": 92, "canonical_nll": 1.2375999118970789, "generation_rouge_l_f1": 0.3492063492063492, "generation_token_f1": 0.36507936507936506, "prompt_id": "61cafca95080c624", "reading_canonical_log_prob": -113.85919189453125, "reading_distances": [28, 27, 23, 27], "reading_log_probs": [-140.20712280273438, -112.12403106689453, -109.78811645507812, -109.14283752441406], "reading_logprob_std": 14.982562657372178, "tokenization_path_logprob_std": 36.676335215646745, "trajectory_balance_rmse": 963.3130210351528} {"canonical_length": 9, "canonical_nll": 1.9871732923719618, "generation_rouge_l_f1": 0.6250000000000001, "generation_token_f1": 0.6250000000000001, "prompt_id": "33cbcf32b9a54c98", "reading_canonical_log_prob": -17.884559631347656, "reading_distances": [24, 22, 17, 21], "reading_log_probs": [-22.426525115966797, -17.967784881591797, -18.871700286865234, -17.564512252807617], "reading_logprob_std": 2.214436451149541, "tokenization_path_logprob_std": 27.072437029135745, "trajectory_balance_rmse": 170.08601012350962} {"canonical_length": 13, "canonical_nll": 0.5550167010380671, "generation_rouge_l_f1": 0.787878787878788, "generation_token_f1": 0.787878787878788, "prompt_id": "5e0db48c41c73148", "reading_canonical_log_prob": -7.215217113494873, "reading_distances": [39, 37, 34, 28], "reading_log_probs": [-7.487349987030029, -6.641359329223633, -8.035120010375977, -9.525455474853516], "reading_logprob_std": 1.2128240247905988, "tokenization_path_logprob_std": 33.808226372272564, "trajectory_balance_rmse": 247.45765392378672} {"canonical_length": 45, "canonical_nll": 1.4527511596679688, "generation_rouge_l_f1": 0.2247191011235955, "generation_token_f1": 0.31460674157303364, "prompt_id": "c70b50c5f56eb753", "reading_canonical_log_prob": -65.3738021850586, "reading_distances": [26, 33, 32, 25], "reading_log_probs": [-78.54763793945312, -103.07257843017578, -107.08184814453125, -90.2860107421875], "reading_logprob_std": 12.958697095917762, "tokenization_path_logprob_std": 64.98685376117874, "trajectory_balance_rmse": 794.7335065087146} {"canonical_length": 28, "canonical_nll": 1.0001674379621233, "generation_rouge_l_f1": 0.10810810810810811, "generation_token_f1": 0.10810810810810811, "prompt_id": "a944937cb2c4cc95", "reading_canonical_log_prob": -28.004688262939453, "reading_distances": [16, 13, 12, 14], "reading_log_probs": [-35.650760650634766, -27.305509567260742, -29.277021408081055, -31.31901741027832], "reading_logprob_std": 3.573007823650896, "tokenization_path_logprob_std": 71.0851062604962, "trajectory_balance_rmse": 611.6458515162212} {"canonical_length": 158, "canonical_nll": 1.30721833434286, "generation_rouge_l_f1": 0.2167487684729064, "generation_token_f1": 0.37438423645320196, "prompt_id": "b725cba739a4a837", "reading_canonical_log_prob": -206.54049682617188, "reading_distances": [35, 37, 30, 31], "reading_log_probs": [-202.40869140625, -205.04556274414062, -200.35488891601562, -203.54937744140625], "reading_logprob_std": 1.9773321275797782, "tokenization_path_logprob_std": 83.27675775921541, "trajectory_balance_rmse": 2999.186078874606} {"canonical_length": 501, "canonical_nll": 0.6830049008428456, "generation_rouge_l_f1": 0.21515151515151512, "generation_token_f1": 0.4090909090909091, "prompt_id": "c3741a95aac9caea", "reading_canonical_log_prob": -342.1854553222656, "reading_distances": [33, 35, 25, 30], "reading_log_probs": [-354.6100769042969, -354.6698913574219, -354.107666015625, -361.3112487792969], "reading_logprob_std": 3.433620894202807, "tokenization_path_logprob_std": 144.9569845770976, "trajectory_balance_rmse": 5156.221213965406} {"canonical_length": 14, "canonical_nll": 0.628596578325544, "generation_rouge_l_f1": 0.4651162790697674, "generation_token_f1": 0.4651162790697674, "prompt_id": "77d6fb3ebd2588d3", "reading_canonical_log_prob": -8.800352096557617, "reading_distances": [23, 18, 19, 26], "reading_log_probs": [-12.410881042480469, -12.803668975830078, -12.013792037963867, -12.607954025268555], "reading_logprob_std": 0.3373966736451933, "tokenization_path_logprob_std": 43.057134371879975, "trajectory_balance_rmse": 211.27139301128935} {"canonical_length": 242, "canonical_nll": 1.41348039611312, "generation_rouge_l_f1": 0.1936842105263158, "generation_token_f1": 0.4042105263157894, "prompt_id": "8d10c80bdd9627ab", "reading_canonical_log_prob": -342.062255859375, "reading_distances": [16, 20, 18, 18], "reading_log_probs": [-347.6668701171875, -348.120361328125, -346.68829345703125, -346.54974365234375], "reading_logprob_std": 0.7609255557066036, "tokenization_path_logprob_std": 341.8783770377282, "trajectory_balance_rmse": 5308.780470238561} {"canonical_length": 233, "canonical_nll": 1.3515797889283798, "generation_rouge_l_f1": 0.21473684210526314, "generation_token_f1": 0.47578947368421054, "prompt_id": "a741547771fa6f03", "reading_canonical_log_prob": -314.9180908203125, "reading_distances": [22, 28, 26, 20], "reading_log_probs": [-315.2017822265625, -315.9852600097656, -314.298828125, -313.593994140625], "reading_logprob_std": 1.0437569153457515, "tokenization_path_logprob_std": 135.71625572370203, "trajectory_balance_rmse": 4903.182094957655} {"canonical_length": 177, "canonical_nll": 1.0412483861890889, "generation_rouge_l_f1": 0.2529274004683841, "generation_token_f1": 0.4355971896955504, "prompt_id": "51a6d546f1967143", "reading_canonical_log_prob": -184.30096435546875, "reading_distances": [27, 33, 28, 28], "reading_log_probs": [-181.79087829589844, -184.2711181640625, -183.82037353515625, -183.67626953125], "reading_logprob_std": 1.0955562674021528, "tokenization_path_logprob_std": 220.47103505485794, "trajectory_balance_rmse": 3039.605144513512} {"canonical_length": 163, "canonical_nll": 1.2253228872100268, "generation_rouge_l_f1": 0.27433628318584075, "generation_token_f1": 0.5, "prompt_id": "715a0325621f71c4", "reading_canonical_log_prob": -199.72763061523438, "reading_distances": [46, 42, 32, 36], "reading_log_probs": [-192.2281036376953, -190.43280029296875, -189.763427734375, -190.51510620117188], "reading_logprob_std": 1.050870909757697, "tokenization_path_logprob_std": 116.9160270517153, "trajectory_balance_rmse": 2687.8006317425593} {"canonical_length": 264, "canonical_nll": 1.3142970692027698, "generation_rouge_l_f1": 0.23828125, "generation_token_f1": 0.44921875, "prompt_id": "0857b1f32ec14a13", "reading_canonical_log_prob": -346.97442626953125, "reading_distances": [17, 23, 21, 17], "reading_log_probs": [-356.3417663574219, -361.4893493652344, -354.5234375, -355.726806640625], "reading_logprob_std": 3.0735569807242613, "tokenization_path_logprob_std": 131.85018245523824, "trajectory_balance_rmse": 4519.900484437096} {"canonical_length": 2, "canonical_nll": 5.505666732788086, "generation_rouge_l_f1": 0.01785714285714286, "generation_token_f1": 0.01785714285714286, "prompt_id": "7c6d4e136a035811", "reading_canonical_log_prob": -11.011333465576172, "reading_distances": [65, 71, 83, 67], "reading_log_probs": [-12.485105514526367, -12.077385902404785, -10.538238525390625, -11.349777221679688], "reading_logprob_std": 0.8564695355568618, "tokenization_path_logprob_std": 0.0, "trajectory_balance_rmse": 10.891670227050781} {"canonical_length": 13, "canonical_nll": 0.9798189309927133, "generation_rouge_l_f1": 0.5599999999999999, "generation_token_f1": 0.7199999999999999, "prompt_id": "d3b27b0b1b980505", "reading_canonical_log_prob": -12.737646102905273, "reading_distances": [31, 27, 34, 37], "reading_log_probs": [-13.24582290649414, -13.234413146972656, -13.382360458374023, -13.284899711608887], "reading_logprob_std": 0.06722820570226033, "tokenization_path_logprob_std": 23.79733848693559, "trajectory_balance_rmse": 242.0164682162875} {"canonical_length": 131, "canonical_nll": 0.47141935625149095, "generation_rouge_l_f1": 0.459214501510574, "generation_token_f1": 0.5438066465256798, "prompt_id": "37ec341c729bc100", "reading_canonical_log_prob": -61.75593566894531, "reading_distances": [9, 8, 14, 7], "reading_log_probs": [-53.9974365234375, -56.35185623168945, -55.86846923828125, -54.9313850402832], "reading_logprob_std": 1.0426665973953528, "tokenization_path_logprob_std": 96.5966515649517, "trajectory_balance_rmse": 1222.7594368693658} {"canonical_length": 12, "canonical_nll": 3.73262087504069, "generation_rouge_l_f1": 0.23076923076923078, "generation_token_f1": 0.38461538461538464, "prompt_id": "b0f40cc1f3136930", "reading_canonical_log_prob": -44.79145050048828, "reading_distances": [10, 12, 9, 13], "reading_log_probs": [-63.84538269042969, -50.933265686035156, -45.28761672973633, -58.184940338134766], "reading_logprob_std": 8.134059791334785, "tokenization_path_logprob_std": 52.0630271735715, "trajectory_balance_rmse": 298.1475519629203} {"canonical_length": 264, "canonical_nll": 1.2554679639411694, "generation_rouge_l_f1": 0.29629629629629634, "generation_token_f1": 0.5614035087719299, "prompt_id": "dc37089a4982f612", "reading_canonical_log_prob": -331.44354248046875, "reading_distances": [16, 16, 14, 14], "reading_log_probs": [-355.3581848144531, -354.35577392578125, -335.81005859375, -334.024658203125], "reading_logprob_std": 11.542453892284406, "tokenization_path_logprob_std": 239.07799119228622, "trajectory_balance_rmse": 5477.509014471944} {"canonical_length": 3, "canonical_nll": 0.8658781051635742, "generation_rouge_l_f1": 0.05882352941176471, "generation_token_f1": 0.05882352941176471, "prompt_id": "9ba6dc4bdecfc908", "reading_canonical_log_prob": -2.5976343154907227, "reading_distances": [24, 25, 31, 15], "reading_log_probs": [-2.2138845920562744, -2.5517778396606445, -1.956922173500061, -2.232792854309082], "reading_logprob_std": 0.24363024762287977, "tokenization_path_logprob_std": 13.53078043460846, "trajectory_balance_rmse": 14.59539341221242} {"canonical_length": 261, "canonical_nll": 0.9591437445746528, "generation_rouge_l_f1": 0.6344294003868473, "generation_token_f1": 0.793036750483559, "prompt_id": "5cf08716fd03e235", "reading_canonical_log_prob": -250.33651733398438, "reading_distances": [30, 30, 30, 31], "reading_log_probs": [-269.9743347167969, -270.131591796875, -265.5967102050781, -271.3695983886719], "reading_logprob_std": 2.5258519391503773, "tokenization_path_logprob_std": 120.11992486877074, "trajectory_balance_rmse": 4265.2643112659225} {"canonical_length": 102, "canonical_nll": 0.9521930170994178, "generation_rouge_l_f1": 0.2142857142857143, "generation_token_f1": 0.34615384615384615, "prompt_id": "8cfc1971edf74b1e", "reading_canonical_log_prob": -97.12368774414062, "reading_distances": [38, 31, 32, 39], "reading_log_probs": [-96.15725708007812, -95.50528717041016, -95.34181213378906, -94.076904296875], "reading_logprob_std": 0.8700994427370189, "tokenization_path_logprob_std": 119.18882489648969, "trajectory_balance_rmse": 1879.1354502084305} {"canonical_length": 42, "canonical_nll": 2.7036485217866444, "generation_rouge_l_f1": 0.07843137254901962, "generation_token_f1": 0.1176470588235294, "prompt_id": "73cb687de2d9ad9d", "reading_canonical_log_prob": -113.55323791503906, "reading_distances": [24, 27, 23, 23], "reading_log_probs": [-114.4521713256836, -115.80624389648438, -117.13917541503906, -112.0700912475586], "reading_logprob_std": 2.1633108566123216, "tokenization_path_logprob_std": 58.58203351855345, "trajectory_balance_rmse": 793.445661492345} {"canonical_length": 50, "canonical_nll": 1.3209599304199218, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "5b7c88eabb38a156", "reading_canonical_log_prob": -66.0479965209961, "reading_distances": [29, 26, 30, 26], "reading_log_probs": [-58.88529968261719, -60.92375946044922, -59.38827133178711, -58.55931091308594], "reading_logprob_std": 1.0468239077375925, "tokenization_path_logprob_std": 89.75669697099656, "trajectory_balance_rmse": 987.8994442156314} {"canonical_length": 297, "canonical_nll": 1.8541937934027777, "generation_rouge_l_f1": 0.22263450834879409, "generation_token_f1": 0.43042671614100186, "prompt_id": "7564700da05daae9", "reading_canonical_log_prob": -550.695556640625, "reading_distances": [25, 22, 21, 30], "reading_log_probs": [-568.5263671875, -567.09130859375, -569.1441650390625, -572.203857421875], "reading_logprob_std": 2.1540484047346498, "tokenization_path_logprob_std": 326.2365351977119, "trajectory_balance_rmse": 6065.322303294232} {"canonical_length": 240, "canonical_nll": 1.5221951802571614, "generation_rouge_l_f1": 0.2300194931773879, "generation_token_f1": 0.40545808966861596, "prompt_id": "ca2d7d9e3b473a6f", "reading_canonical_log_prob": -365.32684326171875, "reading_distances": [17, 14, 21, 13], "reading_log_probs": [-374.93780517578125, -374.54730224609375, -370.80731201171875, -377.3034973144531], "reading_logprob_std": 2.686300543296277, "tokenization_path_logprob_std": 182.193239666402, "trajectory_balance_rmse": 5504.062880200658} {"canonical_length": 282, "canonical_nll": 1.237128399788065, "generation_rouge_l_f1": 0.26953125, "generation_token_f1": 0.39843749999999994, "prompt_id": "9e92f8d560ad3563", "reading_canonical_log_prob": -348.8702087402344, "reading_distances": [22, 25, 24, 15], "reading_log_probs": [-366.949462890625, -357.28179931640625, -366.15966796875, -362.6773681640625], "reading_logprob_std": 4.400695165571493, "tokenization_path_logprob_std": 192.889729691358, "trajectory_balance_rmse": 4314.8725583427595} {"canonical_length": 16, "canonical_nll": 2.014431953430176, "generation_rouge_l_f1": 0.45161290322580644, "generation_token_f1": 0.5161290322580646, "prompt_id": "8efbf41d41230763", "reading_canonical_log_prob": -32.23091125488281, "reading_distances": [37, 36, 42, 28], "reading_log_probs": [-35.93354797363281, -36.14146041870117, -33.13258361816406, -34.851776123046875], "reading_logprob_std": 1.3763207987921362, "tokenization_path_logprob_std": 36.4084196251416, "trajectory_balance_rmse": 363.68542458246304} {"canonical_length": 39, "canonical_nll": 1.9306034185947516, "generation_rouge_l_f1": 0.33962264150943394, "generation_token_f1": 0.3773584905660377, "prompt_id": "3f9b43f9811260d6", "reading_canonical_log_prob": -75.29353332519531, "reading_distances": [32, 31, 28, 35], "reading_log_probs": [-78.29763793945312, -81.93568420410156, -76.91874694824219, -78.64204406738281], "reading_logprob_std": 2.126084317679521, "tokenization_path_logprob_std": 64.41526139796262, "trajectory_balance_rmse": 676.4932832595478} {"canonical_length": 281, "canonical_nll": 1.0085476369620219, "generation_rouge_l_f1": 0.24518388791593695, "generation_token_f1": 0.4238178633975481, "prompt_id": "9e7206bd2d909c4b", "reading_canonical_log_prob": -283.4018859863281, "reading_distances": [28, 37, 33, 32], "reading_log_probs": [-291.921875, -299.17510986328125, -297.5544738769531, -289.0128479003906], "reading_logprob_std": 4.757935724335682, "tokenization_path_logprob_std": 63.42907552356942, "trajectory_balance_rmse": 4251.525783259128} {"canonical_length": 193, "canonical_nll": 1.3441003987208548, "generation_rouge_l_f1": 0.3830845771144279, "generation_token_f1": 0.5373134328358209, "prompt_id": "59cc1c6a894af7a8", "reading_canonical_log_prob": -259.411376953125, "reading_distances": [20, 23, 27, 23], "reading_log_probs": [-261.229736328125, -259.0027770996094, -261.2709655761719, -260.1835021972656], "reading_logprob_std": 1.0714873010154689, "tokenization_path_logprob_std": 190.0612040791962, "trajectory_balance_rmse": 3655.102388963147} {"canonical_length": 85, "canonical_nll": 1.1209692562327664, "generation_rouge_l_f1": 0.23906705539358597, "generation_token_f1": 0.36734693877551017, "prompt_id": "bb9e06240071dbbd", "reading_canonical_log_prob": -95.28238677978516, "reading_distances": [25, 25, 20, 24], "reading_log_probs": [-97.14262390136719, -97.91383361816406, -98.91827392578125, -99.14409637451172], "reading_logprob_std": 0.9276777483233662, "tokenization_path_logprob_std": 85.2374377918263, "trajectory_balance_rmse": 1636.4367926702641} {"canonical_length": 59, "canonical_nll": 0.9168294486352953, "generation_rouge_l_f1": 0.20689655172413793, "generation_token_f1": 0.3054187192118227, "prompt_id": "fbd2f42b1f0643a2", "reading_canonical_log_prob": -54.09293746948242, "reading_distances": [17, 17, 13, 17], "reading_log_probs": [-48.96332550048828, -48.72308349609375, -51.11525344848633, -50.02308654785156], "reading_logprob_std": 1.0960883915888806, "tokenization_path_logprob_std": 65.30650061908857, "trajectory_balance_rmse": 899.0855662225521} {"canonical_length": 73, "canonical_nll": 0.8639211262742134, "generation_rouge_l_f1": 0.2572347266881029, "generation_token_f1": 0.34083601286173637, "prompt_id": "1a0ff319dcddb777", "reading_canonical_log_prob": -63.06624221801758, "reading_distances": [48, 48, 55, 41], "reading_log_probs": [-64.57321166992188, -65.57461547851562, -66.39527130126953, -64.12638854980469], "reading_logprob_std": 1.0182103558483617, "tokenization_path_logprob_std": 104.48747894773147, "trajectory_balance_rmse": 643.2517247263177} {"canonical_length": 15, "canonical_nll": 0.8491415659586589, "generation_rouge_l_f1": 0.46153846153846156, "generation_token_f1": 0.5384615384615384, "prompt_id": "9f8d22c249bf7888", "reading_canonical_log_prob": -12.737123489379883, "reading_distances": [20, 21, 24, 21], "reading_log_probs": [-13.555301666259766, -16.250402450561523, -13.183633804321289, -14.228435516357422], "reading_logprob_std": 1.3674747178448947, "tokenization_path_logprob_std": 35.49901534957844, "trajectory_balance_rmse": 337.56010657012473} {"canonical_length": 168, "canonical_nll": 1.1611677805582683, "generation_rouge_l_f1": 0.17955112219451372, "generation_token_f1": 0.3441396508728179, "prompt_id": "9edad123b1e31f1d", "reading_canonical_log_prob": -195.07618713378906, "reading_distances": [28, 29, 28, 31], "reading_log_probs": [-218.79470825195312, -223.6187744140625, -221.59625244140625, -215.17620849609375], "reading_logprob_std": 3.660581503598963, "tokenization_path_logprob_std": 284.5733519455572, "trajectory_balance_rmse": 3323.030092123595} {"canonical_length": 82, "canonical_nll": 1.337618292831793, "generation_rouge_l_f1": 0.17791411042944782, "generation_token_f1": 0.26993865030674846, "prompt_id": "e027ef71856c6f38", "reading_canonical_log_prob": -109.68470001220703, "reading_distances": [29, 27, 25, 26], "reading_log_probs": [-119.43263244628906, -118.58676147460938, -122.30903625488281, -111.6940689086914], "reading_logprob_std": 4.499228029202058, "tokenization_path_logprob_std": 109.56155984742738, "trajectory_balance_rmse": 1954.2659163537635} {"canonical_length": 27, "canonical_nll": 1.9492051866319444, "generation_rouge_l_f1": 0.1891891891891892, "generation_token_f1": 0.1891891891891892, "prompt_id": "59a16e60dba35dc9", "reading_canonical_log_prob": -52.6285400390625, "reading_distances": [15, 14, 18, 15], "reading_log_probs": [-73.9598159790039, -65.74514770507812, -66.27659606933594, -63.47101593017578], "reading_logprob_std": 4.563012182265138, "tokenization_path_logprob_std": 64.82049597665394, "trajectory_balance_rmse": 634.9847296936279} {"canonical_length": 16, "canonical_nll": 0.5487178564071655, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.75, "prompt_id": "ca87800b07a107f6", "reading_canonical_log_prob": -8.779485702514648, "reading_distances": [27, 32, 30, 26], "reading_log_probs": [-9.109512329101562, -9.240385055541992, -8.984173774719238, -9.206440925598145], "reading_logprob_std": 0.11490274412798959, "tokenization_path_logprob_std": 14.07736083664683, "trajectory_balance_rmse": 235.5430348084208} {"canonical_length": 100, "canonical_nll": 1.5049606323242188, "generation_rouge_l_f1": 0.19662921348314608, "generation_token_f1": 0.34269662921348315, "prompt_id": "911e3e27940b6a5a", "reading_canonical_log_prob": -150.49606323242188, "reading_distances": [29, 27, 23, 30], "reading_log_probs": [-152.2185516357422, -149.85939025878906, -151.69180297851562, -151.75868225097656], "reading_logprob_std": 1.0417975525316099, "tokenization_path_logprob_std": 289.3734969377017, "trajectory_balance_rmse": 2061.1087840471214} {"canonical_length": 190, "canonical_nll": 2.196766421669408, "generation_rouge_l_f1": 0.2091503267973856, "generation_token_f1": 0.4139433551198257, "prompt_id": "1e23f4d5f326fadb", "reading_canonical_log_prob": -417.3856201171875, "reading_distances": [23, 25, 31, 29], "reading_log_probs": [-417.8870849609375, -428.8445739746094, -421.91424560546875, -430.2857666015625], "reading_logprob_std": 5.8466567800703455, "tokenization_path_logprob_std": 191.95310855891802, "trajectory_balance_rmse": 4235.409798633718} {"canonical_length": 151, "canonical_nll": 1.514494839093543, "generation_rouge_l_f1": 0.26344086021505375, "generation_token_f1": 0.521505376344086, "prompt_id": "75c88d19daf68f00", "reading_canonical_log_prob": -228.688720703125, "reading_distances": [18, 19, 24, 20], "reading_log_probs": [-235.15806579589844, -236.005126953125, -229.10397338867188, -234.42123413085938], "reading_logprob_std": 3.1134167615729607, "tokenization_path_logprob_std": 216.14201087366484, "trajectory_balance_rmse": 3761.822852757185} {"canonical_length": 165, "canonical_nll": 1.3388463800603694, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.48125, "prompt_id": "0efac862acbbc3cf", "reading_canonical_log_prob": -220.90965270996094, "reading_distances": [28, 33, 24, 31], "reading_log_probs": [-220.49880981445312, -223.6935577392578, -222.09115600585938, -220.0110626220703], "reading_logprob_std": 1.6692095961012579, "tokenization_path_logprob_std": 215.35484661179348, "trajectory_balance_rmse": 4066.0557506693103} {"canonical_length": 225, "canonical_nll": 0.9715812174479167, "generation_rouge_l_f1": 0.20833333333333334, "generation_token_f1": 0.425, "prompt_id": "867d14d7a7b70c9c", "reading_canonical_log_prob": -218.60577392578125, "reading_distances": [14, 16, 20, 20], "reading_log_probs": [-218.48703002929688, -214.6271209716797, -198.77996826171875, -204.30393981933594], "reading_logprob_std": 9.095061889263478, "tokenization_path_logprob_std": 62.74366136071494, "trajectory_balance_rmse": 3605.92655033307} {"canonical_length": 263, "canonical_nll": 1.5089322514407082, "generation_rouge_l_f1": 0.18687872763419483, "generation_token_f1": 0.34194831013916505, "prompt_id": "67b31fbc0903ce67", "reading_canonical_log_prob": -396.84918212890625, "reading_distances": [27, 20, 20, 14], "reading_log_probs": [-407.583251953125, -404.28472900390625, -404.48388671875, -405.86993408203125], "reading_logprob_std": 1.5246705343105647, "tokenization_path_logprob_std": 287.26711050518685, "trajectory_balance_rmse": 5148.953720465755} {"canonical_length": 38, "canonical_nll": 0.7189707003141704, "generation_rouge_l_f1": 0.36000000000000004, "generation_token_f1": 0.38, "prompt_id": "c50601d0ada030f4", "reading_canonical_log_prob": -27.320886611938477, "reading_distances": [28, 26, 24, 26], "reading_log_probs": [-28.655040740966797, -30.466136932373047, -28.91344451904297, -30.490493774414062], "reading_logprob_std": 0.9837962655395402, "tokenization_path_logprob_std": 32.69333495776394, "trajectory_balance_rmse": 229.7848585210045} {"canonical_length": 35, "canonical_nll": 0.5135156358991351, "generation_rouge_l_f1": 0.5542168674698795, "generation_token_f1": 0.5542168674698795, "prompt_id": "d0d4fbc6a94b1aa0", "reading_canonical_log_prob": -17.973047256469727, "reading_distances": [21, 19, 20, 21], "reading_log_probs": [-16.97222328186035, -16.88335418701172, -18.557748794555664, -15.858419418334961], "reading_logprob_std": 1.1144062536082389, "tokenization_path_logprob_std": 83.96355091485988, "trajectory_balance_rmse": 579.7565471717012} {"canonical_length": 74, "canonical_nll": 0.5694807928961676, "generation_rouge_l_f1": 0.425531914893617, "generation_token_f1": 0.5744680851063829, "prompt_id": "a83eef42db047836", "reading_canonical_log_prob": -42.141578674316406, "reading_distances": [20, 25, 26, 29], "reading_log_probs": [-41.206302642822266, -41.67047119140625, -40.727298736572266, -38.78705596923828], "reading_logprob_std": 1.2670783740073017, "tokenization_path_logprob_std": 89.9827206909935, "trajectory_balance_rmse": 1187.4937717912123} {"canonical_length": 107, "canonical_nll": 1.1646197310117918, "generation_rouge_l_f1": 0.26523297491039427, "generation_token_f1": 0.47311827956989244, "prompt_id": "8e875ab98664b402", "reading_canonical_log_prob": -124.61431121826172, "reading_distances": [18, 17, 18, 18], "reading_log_probs": [-135.71511840820312, -134.9852294921875, -134.4835205078125, -133.58274841308594], "reading_logprob_std": 0.8956668122560835, "tokenization_path_logprob_std": 73.29327450128892, "trajectory_balance_rmse": 2001.024640606028} {"canonical_length": 31, "canonical_nll": 0.7431251771988407, "generation_rouge_l_f1": 0.6222222222222222, "generation_token_f1": 0.6222222222222222, "prompt_id": "54f5b26b0d406b0a", "reading_canonical_log_prob": -23.036880493164062, "reading_distances": [29, 31, 27, 34], "reading_log_probs": [-27.33916664123535, -27.016515731811523, -29.405506134033203, -31.899791717529297], "reading_logprob_std": 2.2536689283255593, "tokenization_path_logprob_std": 70.73753278440807, "trajectory_balance_rmse": 612.513569677371} {"canonical_length": 262, "canonical_nll": 1.3443203991606036, "generation_rouge_l_f1": 0.22846441947565543, "generation_token_f1": 0.49063670411985016, "prompt_id": "652a0c6a273ec7b9", "reading_canonical_log_prob": -352.2119445800781, "reading_distances": [16, 17, 13, 13], "reading_log_probs": [-356.75, -352.06219482421875, -355.1026916503906, -353.09552001953125], "reading_logprob_std": 2.0893676467091, "tokenization_path_logprob_std": 238.73048494614562, "trajectory_balance_rmse": 4305.190520898793} {"canonical_length": 304, "canonical_nll": 1.6778772253739207, "generation_rouge_l_f1": 0.12457912457912458, "generation_token_f1": 0.2525252525252525, "prompt_id": "13a435b70dd47ea2", "reading_canonical_log_prob": -510.0746765136719, "reading_distances": [23, 14, 17, 11], "reading_log_probs": [-517.7884521484375, -521.5836181640625, -520.4683227539062, -515.899658203125], "reading_logprob_std": 2.5751458629088857, "tokenization_path_logprob_std": 370.55049314789403, "trajectory_balance_rmse": 6760.89394932125} {"canonical_length": 15, "canonical_nll": 0.793018086751302, "generation_rouge_l_f1": 0.3529411764705882, "generation_token_f1": 0.4705882352941177, "prompt_id": "1f8b04daa461e185", "reading_canonical_log_prob": -11.895271301269531, "reading_distances": [24, 18, 25, 23], "reading_log_probs": [-12.424539566040039, -10.279709815979004, -11.065324783325195, -12.421923637390137], "reading_logprob_std": 1.0604400701985985, "tokenization_path_logprob_std": 19.81389266669746, "trajectory_balance_rmse": 183.681845033232} {"canonical_length": 18, "canonical_nll": 0.3054545720418294, "generation_rouge_l_f1": 0.47619047619047616, "generation_token_f1": 0.47619047619047616, "prompt_id": "713c6185b310a4ca", "reading_canonical_log_prob": -5.49818229675293, "reading_distances": [14, 27, 24, 30], "reading_log_probs": [-5.4050750732421875, -10.939093589782715, -8.123337745666504, -6.678660869598389], "reading_logprob_std": 2.377028833290396, "tokenization_path_logprob_std": 58.779798715763036, "trajectory_balance_rmse": 391.2406802366449} {"canonical_length": 227, "canonical_nll": 0.8063396840368598, "generation_rouge_l_f1": 0.3123732251521298, "generation_token_f1": 0.4949290060851927, "prompt_id": "a7de6528410fa4b7", "reading_canonical_log_prob": -183.0391082763672, "reading_distances": [21, 28, 25, 25], "reading_log_probs": [-192.43504333496094, -198.64566040039062, -190.81935119628906, -189.8446807861328], "reading_logprob_std": 3.953379818711711, "tokenization_path_logprob_std": 86.73878609704282, "trajectory_balance_rmse": 3343.74942262196} {"canonical_length": 154, "canonical_nll": 1.729839176326603, "generation_rouge_l_f1": 0.2707838479809976, "generation_token_f1": 0.4370546318289786, "prompt_id": "b7245c913ba925d6", "reading_canonical_log_prob": -266.3952331542969, "reading_distances": [22, 28, 21, 28], "reading_log_probs": [-267.54693603515625, -268.79815673828125, -268.0393371582031, -266.97271728515625], "reading_logprob_std": 0.773706102280434, "tokenization_path_logprob_std": 123.61555373757567, "trajectory_balance_rmse": 3166.4388561824094} {"canonical_length": 9, "canonical_nll": 0.7106206681993272, "generation_rouge_l_f1": 0.8750000000000001, "generation_token_f1": 0.8750000000000001, "prompt_id": "402312ec99016f32", "reading_canonical_log_prob": -6.395586013793945, "reading_distances": [23, 21, 22, 20], "reading_log_probs": [-7.132452487945557, -7.360826015472412, -7.475787162780762, -6.691463947296143], "reading_logprob_std": 0.3465218379428095, "tokenization_path_logprob_std": 40.78940981776797, "trajectory_balance_rmse": 170.04873396746723} {"canonical_length": 290, "canonical_nll": 1.4355969659213361, "generation_rouge_l_f1": 0.2380106571936057, "generation_token_f1": 0.49378330373001783, "prompt_id": "083ca3a9f1344c38", "reading_canonical_log_prob": -416.3231201171875, "reading_distances": [30, 35, 32, 31], "reading_log_probs": [-424.122314453125, -425.85205078125, -426.0133361816406, -428.669921875], "reading_logprob_std": 1.8768888759855333, "tokenization_path_logprob_std": 223.76793754054282, "trajectory_balance_rmse": 6043.879281819117} {"canonical_length": 58, "canonical_nll": 1.6705818176269531, "generation_rouge_l_f1": 0.19591836734693877, "generation_token_f1": 0.2775510204081633, "prompt_id": "c06dd3bfbe8f1823", "reading_canonical_log_prob": -96.89374542236328, "reading_distances": [35, 29, 32, 34], "reading_log_probs": [-101.22756958007812, -101.20771026611328, -103.38756561279297, -101.84834289550781], "reading_logprob_std": 1.023992302808972, "tokenization_path_logprob_std": 121.88293775128167, "trajectory_balance_rmse": 1159.945515222223} {"canonical_length": 117, "canonical_nll": 0.9430011194995326, "generation_rouge_l_f1": 0.2011173184357542, "generation_token_f1": 0.32402234636871513, "prompt_id": "22135eb61f4e8a41", "reading_canonical_log_prob": -110.33113098144531, "reading_distances": [23, 16, 22, 21], "reading_log_probs": [-123.54287719726562, -125.29365539550781, -119.62701416015625, -117.16004943847656], "reading_logprob_std": 3.6911152505674676, "tokenization_path_logprob_std": 122.95212818946268, "trajectory_balance_rmse": 2225.321180638249} {"canonical_length": 21, "canonical_nll": 2.183457510811942, "generation_rouge_l_f1": 0.3529411764705882, "generation_token_f1": 0.4117647058823529, "prompt_id": "b36f1ed6b55c0058", "reading_canonical_log_prob": -45.85260772705078, "reading_distances": [27, 26, 24, 28], "reading_log_probs": [-59.48552322387695, -51.255043029785156, -56.30403518676758, -49.405670166015625], "reading_logprob_std": 4.618495875045387, "tokenization_path_logprob_std": 87.5854508063459, "trajectory_balance_rmse": 435.8118251054314} {"canonical_length": 25, "canonical_nll": 1.7651446533203126, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.2631578947368421, "prompt_id": "017529d62a606652", "reading_canonical_log_prob": -44.12861633300781, "reading_distances": [25, 19, 26, 27], "reading_log_probs": [-43.71147918701172, -42.53459548950195, -42.91149139404297, -41.66218185424805], "reading_logprob_std": 0.8509105742524149, "tokenization_path_logprob_std": 92.50105275424599, "trajectory_balance_rmse": 488.9851702159808} {"canonical_length": 231, "canonical_nll": 1.649572331151921, "generation_rouge_l_f1": 0.17751479289940827, "generation_token_f1": 0.37475345167652857, "prompt_id": "47cd1ce3b4c7133a", "reading_canonical_log_prob": -381.05120849609375, "reading_distances": [15, 12, 12, 9], "reading_log_probs": [-389.7129211425781, -385.671142578125, -385.835693359375, -386.2374267578125], "reading_logprob_std": 1.913919950447398, "tokenization_path_logprob_std": 158.21984493481514, "trajectory_balance_rmse": 5121.077724353289} {"canonical_length": 34, "canonical_nll": 0.6977959801169002, "generation_rouge_l_f1": 0.4571428571428571, "generation_token_f1": 0.4857142857142857, "prompt_id": "549cddbffb2c458b", "reading_canonical_log_prob": -23.72506332397461, "reading_distances": [11, 13, 17, 10], "reading_log_probs": [-29.168651580810547, -29.181406021118164, -30.692413330078125, -27.295413970947266], "reading_logprob_std": 1.3907655396177365, "tokenization_path_logprob_std": 100.96137296613398, "trajectory_balance_rmse": 444.5291828794505} {"canonical_length": 356, "canonical_nll": 1.4196941075700054, "generation_rouge_l_f1": 0.24957841483979762, "generation_token_f1": 0.47892074198988194, "prompt_id": "6b4ba85e93ef4f52", "reading_canonical_log_prob": -505.4111022949219, "reading_distances": [29, 24, 24, 28], "reading_log_probs": [-503.0926513671875, -496.92559814453125, -501.9715576171875, -499.1402893066406], "reading_logprob_std": 2.7882671843299605, "tokenization_path_logprob_std": 343.71519934960804, "trajectory_balance_rmse": 7921.432920885689} {"canonical_length": 420, "canonical_nll": 1.2938700358072917, "generation_rouge_l_f1": 0.23547880690737835, "generation_token_f1": 0.43014128728414447, "prompt_id": "71f47b6572a134fd", "reading_canonical_log_prob": -543.4254150390625, "reading_distances": [28, 37, 36, 36], "reading_log_probs": [-541.7032470703125, -542.8193359375, -540.565185546875, -539.6295166015625], "reading_logprob_std": 1.383618872206404, "tokenization_path_logprob_std": 297.03958027486357, "trajectory_balance_rmse": 7828.369661348962} {"canonical_length": 144, "canonical_nll": 1.3292505476209853, "generation_rouge_l_f1": 0.274932614555256, "generation_token_f1": 0.4582210242587601, "prompt_id": "c1009daeb59a17f6", "reading_canonical_log_prob": -191.41207885742188, "reading_distances": [18, 17, 20, 19], "reading_log_probs": [-195.83627319335938, -196.5247344970703, -200.96148681640625, -197.65213012695312], "reading_logprob_std": 2.272054206353305, "tokenization_path_logprob_std": 100.9035006929341, "trajectory_balance_rmse": 2677.094080917692} {"canonical_length": 19, "canonical_nll": 2.151770541542455, "generation_rouge_l_f1": 0.411764705882353, "generation_token_f1": 0.411764705882353, "prompt_id": "231e10798735d9e9", "reading_canonical_log_prob": -40.88364028930664, "reading_distances": [17, 20, 16, 21], "reading_log_probs": [-41.385643005371094, -42.02881622314453, -41.84560775756836, -44.6905403137207], "reading_logprob_std": 1.4933063791837131, "tokenization_path_logprob_std": 17.93105294129818, "trajectory_balance_rmse": 475.5733597041861} {"canonical_length": 133, "canonical_nll": 1.8580839687720276, "generation_rouge_l_f1": 0.14790996784565916, "generation_token_f1": 0.26366559485530544, "prompt_id": "948eaf298aee4503", "reading_canonical_log_prob": -247.1251678466797, "reading_distances": [24, 19, 19, 19], "reading_log_probs": [-229.3079376220703, -229.83560180664062, -231.535400390625, -231.25889587402344], "reading_logprob_std": 1.081580400536757, "tokenization_path_logprob_std": 210.27235177188712, "trajectory_balance_rmse": 2827.8509424403896} {"canonical_length": 43, "canonical_nll": 0.31613764651986054, "generation_rouge_l_f1": 0.3829787234042553, "generation_token_f1": 0.5957446808510638, "prompt_id": "cc672b4406958c5b", "reading_canonical_log_prob": -13.593918800354004, "reading_distances": [44, 54, 40, 46], "reading_log_probs": [-23.33747100830078, -17.400869369506836, -18.68670654296875, -20.35967254638672], "reading_logprob_std": 2.5649373211033692, "tokenization_path_logprob_std": 35.8848828623337, "trajectory_balance_rmse": 787.8818387217954} {"canonical_length": 189, "canonical_nll": 1.1571185399615576, "generation_rouge_l_f1": 0.18981481481481483, "generation_token_f1": 0.34259259259259256, "prompt_id": "f0c1c49e728adae3", "reading_canonical_log_prob": -218.69540405273438, "reading_distances": [17, 19, 17, 21], "reading_log_probs": [-214.26712036132812, -219.0343780517578, -218.64736938476562, -217.7457275390625], "reading_logprob_std": 2.1725060253347817, "tokenization_path_logprob_std": 84.05780690362613, "trajectory_balance_rmse": 3683.3206127591825} {"canonical_length": 202, "canonical_nll": 1.5799119401686261, "generation_rouge_l_f1": 0.24242424242424243, "generation_token_f1": 0.4761904761904762, "prompt_id": "91ba22c36f60a796", "reading_canonical_log_prob": -319.1422119140625, "reading_distances": [28, 31, 29, 23], "reading_log_probs": [-325.3409729003906, -325.93231201171875, -320.43646240234375, -319.4364929199219], "reading_logprob_std": 3.3249906839191516, "tokenization_path_logprob_std": 282.7266676695962, "trajectory_balance_rmse": 4722.331721033755} {"canonical_length": 91, "canonical_nll": 1.21202640743046, "generation_rouge_l_f1": 0.28455284552845533, "generation_token_f1": 0.42276422764227634, "prompt_id": "f1222deb29a5be25", "reading_canonical_log_prob": -110.29440307617188, "reading_distances": [13, 13, 9, 15], "reading_log_probs": [-116.31722259521484, -115.73531341552734, -115.8082504272461, -115.06210327148438], "reading_logprob_std": 0.5154498651457398, "tokenization_path_logprob_std": 136.42449866181423, "trajectory_balance_rmse": 2193.4586960915044} {"canonical_length": 9, "canonical_nll": 0.2488141589694553, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "6b02223ad9dd90ef", "reading_canonical_log_prob": -2.2393274307250977, "reading_distances": [25, 29, 27, 25], "reading_log_probs": [-2.6756515502929688, -1.9005117416381836, -1.842444658279419, -1.8165372610092163], "reading_logprob_std": 0.4127395771661793, "tokenization_path_logprob_std": 48.46614675180617, "trajectory_balance_rmse": 180.84951497506364} {"canonical_length": 170, "canonical_nll": 1.3309161017922795, "generation_rouge_l_f1": 0.2068965517241379, "generation_token_f1": 0.37356321839080464, "prompt_id": "a027618ad6b82f3e", "reading_canonical_log_prob": -226.2557373046875, "reading_distances": [23, 19, 18, 19], "reading_log_probs": [-237.26693725585938, -238.31997680664062, -236.73348999023438, -237.36827087402344], "reading_logprob_std": 0.660136233043112, "tokenization_path_logprob_std": 201.4095488525846, "trajectory_balance_rmse": 3769.332601290178} {"canonical_length": 38, "canonical_nll": 1.5119420101768093, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "252eb77faa4d024a", "reading_canonical_log_prob": -57.45379638671875, "reading_distances": [20, 24, 22, 13], "reading_log_probs": [-59.76047897338867, -57.557315826416016, -57.21916580200195, -57.888816833496094], "reading_logprob_std": 1.1360747669731384, "tokenization_path_logprob_std": 44.35787262735552, "trajectory_balance_rmse": 663.6701139169618} {"canonical_length": 273, "canonical_nll": 0.7597573528359661, "generation_rouge_l_f1": 0.28270042194092826, "generation_token_f1": 0.5274261603375529, "prompt_id": "75be9357ba7c860a", "reading_canonical_log_prob": -207.41375732421875, "reading_distances": [24, 18, 17, 16], "reading_log_probs": [-220.16049194335938, -221.4410400390625, -214.2175750732422, -207.9789276123047], "reading_logprob_std": 6.175733513431951, "tokenization_path_logprob_std": 183.87501654677808, "trajectory_balance_rmse": 4853.956831392736} {"canonical_length": 347, "canonical_nll": 1.0733235383926962, "generation_rouge_l_f1": 0.25407166123778496, "generation_token_f1": 0.5635179153094463, "prompt_id": "8d16eaea1c875841", "reading_canonical_log_prob": -372.4432678222656, "reading_distances": [41, 42, 47, 53], "reading_log_probs": [-383.57781982421875, -386.8433532714844, -384.6949462890625, -387.4228515625], "reading_logprob_std": 1.8048288772603014, "tokenization_path_logprob_std": 161.48437726088943, "trajectory_balance_rmse": 7611.635499679803} {"canonical_length": 355, "canonical_nll": 0.9550095249229753, "generation_rouge_l_f1": 0.2601626016260163, "generation_token_f1": 0.5008130081300812, "prompt_id": "f4746ccf1632be92", "reading_canonical_log_prob": -339.02838134765625, "reading_distances": [23, 26, 23, 20], "reading_log_probs": [-342.7545471191406, -340.18951416015625, -344.10443115234375, -340.0690612792969], "reading_logprob_std": 1.9840763834484436, "tokenization_path_logprob_std": 241.6895517261005, "trajectory_balance_rmse": 7460.901488391807} {"canonical_length": 450, "canonical_nll": 1.0521318901909722, "generation_rouge_l_f1": 0.19774011299435026, "generation_token_f1": 0.3418079096045198, "prompt_id": "06190df6c1e1573b", "reading_canonical_log_prob": -473.4593505859375, "reading_distances": [16, 15, 17, 12], "reading_log_probs": [-477.75689697265625, -477.7242431640625, -474.7752380371094, -476.562255859375], "reading_logprob_std": 1.4011543485292695, "tokenization_path_logprob_std": 312.53421934876314, "trajectory_balance_rmse": 7592.401814239468} {"canonical_length": 102, "canonical_nll": 0.9073232014973959, "generation_rouge_l_f1": 0.27042253521126763, "generation_token_f1": 0.37183098591549296, "prompt_id": "56fa8783ee42ce19", "reading_canonical_log_prob": -92.54696655273438, "reading_distances": [13, 19, 16, 19], "reading_log_probs": [-93.839111328125, -93.09306335449219, -92.89100646972656, -91.70962524414062], "reading_logprob_std": 0.8822599283178342, "tokenization_path_logprob_std": 112.95397918313178, "trajectory_balance_rmse": 1950.5346608196216} {"canonical_length": 202, "canonical_nll": 2.2453555871944615, "generation_rouge_l_f1": 0.28710462287104627, "generation_token_f1": 0.4914841849148418, "prompt_id": "11f8f65e51ee2cf5", "reading_canonical_log_prob": -453.56182861328125, "reading_distances": [22, 23, 23, 20], "reading_log_probs": [-458.53228759765625, -460.79425048828125, -459.1257019042969, -460.6204833984375], "reading_logprob_std": 1.1134705300001257, "tokenization_path_logprob_std": 161.9668982969156, "trajectory_balance_rmse": 3549.9619809746587} {"canonical_length": 11, "canonical_nll": 2.659000049937855, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "8980941d8fb125ec", "reading_canonical_log_prob": -29.249000549316406, "reading_distances": [15, 18, 16, 21], "reading_log_probs": [-32.01899337768555, -40.81427001953125, -39.345523834228516, -36.76975631713867], "reading_logprob_std": 3.8595614279335275, "tokenization_path_logprob_std": 42.330532442358894, "trajectory_balance_rmse": 239.31123729450175} {"canonical_length": 69, "canonical_nll": 0.5367729076440784, "generation_rouge_l_f1": 0.20588235294117646, "generation_token_f1": 0.3431372549019608, "prompt_id": "2d0e3ff402d06b28", "reading_canonical_log_prob": -37.037330627441406, "reading_distances": [16, 14, 13, 14], "reading_log_probs": [-50.25616455078125, -44.902076721191406, -48.460121154785156, -41.56532287597656], "reading_logprob_std": 3.8595617961999116, "tokenization_path_logprob_std": 131.61262538515697, "trajectory_balance_rmse": 947.9396063660722} {"canonical_length": 53, "canonical_nll": 0.5856976059247863, "generation_rouge_l_f1": 0.18930041152263377, "generation_token_f1": 0.2716049382716049, "prompt_id": "54b6e30181a1bec1", "reading_canonical_log_prob": -31.041973114013672, "reading_distances": [19, 24, 19, 16], "reading_log_probs": [-27.688323974609375, -29.05846405029297, -31.64278793334961, -28.595294952392578], "reading_logprob_std": 1.6960280657386917, "tokenization_path_logprob_std": 21.47594736342194, "trajectory_balance_rmse": 429.0170959423824} {"canonical_length": 9, "canonical_nll": 0.8935994042290581, "generation_rouge_l_f1": 0.41025641025641024, "generation_token_f1": 0.41025641025641024, "prompt_id": "2996ca79ae58108b", "reading_canonical_log_prob": -8.042394638061523, "reading_distances": [30, 28, 26, 36], "reading_log_probs": [-8.060506820678711, -7.914007186889648, -8.034048080444336, -7.543618202209473], "reading_logprob_std": 0.23830221237982654, "tokenization_path_logprob_std": 42.22922306575307, "trajectory_balance_rmse": 140.69519750988718} {"canonical_length": 11, "canonical_nll": 1.4016671614213423, "generation_rouge_l_f1": 0.16, "generation_token_f1": 0.16, "prompt_id": "e6cc80b8ba91ddd7", "reading_canonical_log_prob": -15.418338775634766, "reading_distances": [21, 22, 27, 25], "reading_log_probs": [-16.904268264770508, -19.186016082763672, -17.836898803710938, -18.968942642211914], "reading_logprob_std": 1.060181378957692, "tokenization_path_logprob_std": 44.263482912838526, "trajectory_balance_rmse": 226.5533809594063} {"canonical_length": 218, "canonical_nll": 1.8638242669061784, "generation_rouge_l_f1": 0.22641509433962265, "generation_token_f1": 0.4528301886792453, "prompt_id": "1826061f4f79f3ac", "reading_canonical_log_prob": -406.3136901855469, "reading_distances": [14, 16, 13, 20], "reading_log_probs": [-410.0091552734375, -409.3782958984375, -407.9631652832031, -408.52117919921875], "reading_logprob_std": 0.9058492614361238, "tokenization_path_logprob_std": 158.25805299686058, "trajectory_balance_rmse": 4676.83206725419} {"canonical_length": 182, "canonical_nll": 1.8046948778760301, "generation_rouge_l_f1": 0.21479713603818615, "generation_token_f1": 0.4534606205250597, "prompt_id": "d48fbaa3dac1814d", "reading_canonical_log_prob": -328.4544677734375, "reading_distances": [29, 25, 30, 24], "reading_log_probs": [-342.60308837890625, -341.4290771484375, -351.630126953125, -347.2237548828125], "reading_logprob_std": 4.6661199862682805, "tokenization_path_logprob_std": 32.66820395278296, "trajectory_balance_rmse": 3692.5845544855433} {"canonical_length": 105, "canonical_nll": 0.577705819266183, "generation_rouge_l_f1": 0.3496932515337423, "generation_token_f1": 0.43558282208588966, "prompt_id": "7fbf1f51feeeeb73", "reading_canonical_log_prob": -60.65911102294922, "reading_distances": [17, 20, 17, 20], "reading_log_probs": [-65.66716003417969, -62.974365234375, -61.19239807128906, -62.404029846191406], "reading_logprob_std": 1.8905774705415723, "tokenization_path_logprob_std": 72.98940356337387, "trajectory_balance_rmse": 857.4781274108417} {"canonical_length": 18, "canonical_nll": 0.3793629010518392, "generation_rouge_l_f1": 0.1037037037037037, "generation_token_f1": 0.11851851851851852, "prompt_id": "8cc0717e876412c6", "reading_canonical_log_prob": -6.8285322189331055, "reading_distances": [20, 21, 29, 26], "reading_log_probs": [-8.787773132324219, -11.084543228149414, -7.9134955406188965, -10.639596939086914], "reading_logprob_std": 1.5042693426599847, "tokenization_path_logprob_std": 85.6119751077066, "trajectory_balance_rmse": 364.14480051889876} {"canonical_length": 152, "canonical_nll": 1.9959291156969572, "generation_rouge_l_f1": 0.23469387755102042, "generation_token_f1": 0.3673469387755102, "prompt_id": "3f19d2fd29921f16", "reading_canonical_log_prob": -303.3812255859375, "reading_distances": [19, 17, 16, 12], "reading_log_probs": [-312.19073486328125, -314.116943359375, -320.2994689941406, -313.4577331542969], "reading_logprob_std": 3.611724699572672, "tokenization_path_logprob_std": 303.410375344361, "trajectory_balance_rmse": 3031.61475780782} {"canonical_length": 257, "canonical_nll": 1.9743070491093142, "generation_rouge_l_f1": 0.19915254237288135, "generation_token_f1": 0.38135593220338987, "prompt_id": "774504b35575be36", "reading_canonical_log_prob": -507.39691162109375, "reading_distances": [31, 24, 28, 29], "reading_log_probs": [-514.38427734375, -510.4630126953125, -515.7324829101562, -514.1043090820312], "reading_logprob_std": 2.253694423526105, "tokenization_path_logprob_std": 314.72284965798167, "trajectory_balance_rmse": 5495.415951844061} {"canonical_length": 9, "canonical_nll": 0.431236637963189, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "18b528d0286abbc1", "reading_canonical_log_prob": -3.881129741668701, "reading_distances": [34, 30, 33, 31], "reading_log_probs": [-1.8169488906860352, -2.7718164920806885, -2.1719167232513428, -2.456108331680298], "reading_logprob_std": 0.40687993057409116, "tokenization_path_logprob_std": 17.301323130973074, "trajectory_balance_rmse": 191.7662528949343} {"canonical_length": 4, "canonical_nll": 1.7426186800003052, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.5714285714285715, "prompt_id": "ad4fd52fcf4632b3", "reading_canonical_log_prob": -6.970474720001221, "reading_distances": [22, 29, 20, 25], "reading_log_probs": [-5.677864074707031, -5.1765336990356445, -5.813916206359863, -5.216721057891846], "reading_logprob_std": 0.32236221085708155, "tokenization_path_logprob_std": 30.30021798919276, "trajectory_balance_rmse": 92.86898957198696} {"canonical_length": 349, "canonical_nll": 1.5130268960422637, "generation_rouge_l_f1": 0.18493150684931509, "generation_token_f1": 0.410958904109589, "prompt_id": "ba9050b431bd2ba0", "reading_canonical_log_prob": -528.04638671875, "reading_distances": [18, 15, 21, 19], "reading_log_probs": [-522.6118774414062, -527.091064453125, -523.2296752929688, -521.246826171875], "reading_logprob_std": 2.5050830560252297, "tokenization_path_logprob_std": 274.0551410833943, "trajectory_balance_rmse": 6534.811135174135} {"canonical_length": 35, "canonical_nll": 0.6896263122558594, "generation_rouge_l_f1": 0.12781954887218044, "generation_token_f1": 0.19548872180451127, "prompt_id": "da7f7e74060fc4e6", "reading_canonical_log_prob": -24.136920928955078, "reading_distances": [27, 33, 26, 26], "reading_log_probs": [-24.015052795410156, -22.75440216064453, -23.454326629638672, -23.487041473388672], "reading_logprob_std": 0.5172180953568783, "tokenization_path_logprob_std": 76.21050499491386, "trajectory_balance_rmse": 682.642086344806} {"canonical_length": 366, "canonical_nll": 1.114656792312372, "generation_rouge_l_f1": 0.21664050235478807, "generation_token_f1": 0.43956043956043955, "prompt_id": "378c4d815b07ba59", "reading_canonical_log_prob": -407.9643859863281, "reading_distances": [29, 25, 29, 34], "reading_log_probs": [-415.68292236328125, -412.7955322265625, -418.077392578125, -420.3497009277344], "reading_logprob_std": 3.240064690145231, "tokenization_path_logprob_std": 365.54804097040125, "trajectory_balance_rmse": 7057.026579122546} {"canonical_length": 158, "canonical_nll": 0.5975696225709553, "generation_rouge_l_f1": 0.46913580246913583, "generation_token_f1": 0.6255144032921811, "prompt_id": "e2790ed28464c480", "reading_canonical_log_prob": -94.41600036621094, "reading_distances": [14, 24, 22, 19], "reading_log_probs": [-95.3733901977539, -95.28754425048828, -94.98316192626953, -95.69978332519531], "reading_logprob_std": 0.29471963903273946, "tokenization_path_logprob_std": 173.0602780608511, "trajectory_balance_rmse": 2080.2666355868305} {"canonical_length": 253, "canonical_nll": 1.2553328563102149, "generation_rouge_l_f1": 0.2671905697445972, "generation_token_f1": 0.5304518664047152, "prompt_id": "2481a184df55ff40", "reading_canonical_log_prob": -317.5992126464844, "reading_distances": [47, 39, 37, 41], "reading_log_probs": [-343.9942626953125, -332.8840637207031, -336.337890625, -343.7132873535156], "reading_logprob_std": 5.52066559509569, "tokenization_path_logprob_std": 240.5584767624945, "trajectory_balance_rmse": 4653.16967534523} {"canonical_length": 342, "canonical_nll": 1.5492291366844846, "generation_rouge_l_f1": 0.19865319865319866, "generation_token_f1": 0.393939393939394, "prompt_id": "130ff8d17ea30994", "reading_canonical_log_prob": -529.8363647460938, "reading_distances": [18, 20, 17, 21], "reading_log_probs": [-545.3275756835938, -550.9266357421875, -542.5791625976562, -541.7706909179688], "reading_logprob_std": 4.140450229607613, "tokenization_path_logprob_std": 139.5766657248354, "trajectory_balance_rmse": 7635.520147842602} {"canonical_length": 16, "canonical_nll": 0.754535436630249, "generation_rouge_l_f1": 0.7777777777777778, "generation_token_f1": 0.8888888888888888, "prompt_id": "16bf564574acdadf", "reading_canonical_log_prob": -12.072566986083984, "reading_distances": [27, 28, 22, 31], "reading_log_probs": [-16.984865188598633, -16.169652938842773, -14.849013328552246, -18.878997802734375], "reading_logprob_std": 1.6867025516784084, "tokenization_path_logprob_std": 27.07947410316818, "trajectory_balance_rmse": 220.2790189195804} {"canonical_length": 50, "canonical_nll": 0.6298949813842774, "generation_rouge_l_f1": 0.1890547263681592, "generation_token_f1": 0.31840796019900497, "prompt_id": "effedc7b02b1c679", "reading_canonical_log_prob": -31.494749069213867, "reading_distances": [29, 29, 27, 26], "reading_log_probs": [-37.19982147216797, -35.650917053222656, -36.36476135253906, -31.25717544555664], "reading_logprob_std": 2.6506830078275723, "tokenization_path_logprob_std": 62.4391268807624, "trajectory_balance_rmse": 781.8526077849417} {"canonical_length": 176, "canonical_nll": 1.4085471413352273, "generation_rouge_l_f1": 0.17674418604651163, "generation_token_f1": 0.3581395348837209, "prompt_id": "33cc463f90c8e3a1", "reading_canonical_log_prob": -247.904296875, "reading_distances": [25, 17, 25, 28], "reading_log_probs": [-246.01425170898438, -246.36546325683594, -246.04249572753906, -247.07485961914062], "reading_logprob_std": 0.49348782858414975, "tokenization_path_logprob_std": 100.68100019506808, "trajectory_balance_rmse": 3624.385102810858} {"canonical_length": 283, "canonical_nll": 1.2341067041188163, "generation_rouge_l_f1": 0.24686940966010731, "generation_token_f1": 0.5330948121645797, "prompt_id": "e32f1e4a8a806d4a", "reading_canonical_log_prob": -349.252197265625, "reading_distances": [15, 14, 15, 15], "reading_log_probs": [-358.95623779296875, -352.8197937011719, -354.6418151855469, -356.13177490234375], "reading_logprob_std": 2.5941720077868275, "tokenization_path_logprob_std": 358.71525461705687, "trajectory_balance_rmse": 6123.158634048985} {"canonical_length": 193, "canonical_nll": 1.2627100178614798, "generation_rouge_l_f1": 0.19438444924406048, "generation_token_f1": 0.38876889848812096, "prompt_id": "f16b4d6e1a4f11c6", "reading_canonical_log_prob": -243.70303344726562, "reading_distances": [11, 8, 9, 13], "reading_log_probs": [-250.18663024902344, -256.26788330078125, -253.5135498046875, -255.08648681640625], "reading_logprob_std": 2.6380991940873275, "tokenization_path_logprob_std": 189.83516905346087, "trajectory_balance_rmse": 4229.920723039991} {"canonical_length": 11, "canonical_nll": 0.900890437039462, "generation_rouge_l_f1": 0.7777777777777778, "generation_token_f1": 0.8888888888888888, "prompt_id": "7cf86e055cecbc26", "reading_canonical_log_prob": -9.909794807434082, "reading_distances": [9, 16, 13, 16], "reading_log_probs": [-10.2212495803833, -11.681831359863281, -10.98697280883789, -12.288783073425293], "reading_logprob_std": 0.8916396271059821, "tokenization_path_logprob_std": 35.457802930054974, "trajectory_balance_rmse": 197.45015838339444} {"canonical_length": 251, "canonical_nll": 1.5619380388601842, "generation_rouge_l_f1": 0.21696252465483237, "generation_token_f1": 0.42998027613412226, "prompt_id": "279ec9f263f853e8", "reading_canonical_log_prob": -392.04644775390625, "reading_distances": [29, 20, 27, 24], "reading_log_probs": [-390.5419921875, -393.2213439941406, -387.767822265625, -386.6112060546875], "reading_logprob_std": 2.9594285535455493, "tokenization_path_logprob_std": 201.9871168865536, "trajectory_balance_rmse": 5417.87170216087} {"canonical_length": 102, "canonical_nll": 1.2455711364746094, "generation_rouge_l_f1": 0.32098765432098764, "generation_token_f1": 0.4567901234567901, "prompt_id": "cf98e6698c25bbf0", "reading_canonical_log_prob": -127.04825592041016, "reading_distances": [110, 109, 100, 109], "reading_log_probs": [-119.18705749511719, -119.64510345458984, -125.99453735351562, -116.6880874633789], "reading_logprob_std": 3.9630134084825155, "tokenization_path_logprob_std": 159.3516753156645, "trajectory_balance_rmse": 1830.6312760839987} {"canonical_length": 125, "canonical_nll": 1.50201171875, "generation_rouge_l_f1": 0.1964735516372796, "generation_token_f1": 0.397984886649874, "prompt_id": "588a316f7d854f31", "reading_canonical_log_prob": -187.75146484375, "reading_distances": [34, 25, 33, 35], "reading_log_probs": [-194.733154296875, -191.7010498046875, -194.99783325195312, -190.99172973632812], "reading_logprob_std": 2.0551308104861437, "tokenization_path_logprob_std": 133.79307105630062, "trajectory_balance_rmse": 3213.279395413211} {"canonical_length": 116, "canonical_nll": 1.843872201853785, "generation_rouge_l_f1": 0.23268698060941828, "generation_token_f1": 0.371191135734072, "prompt_id": "6bdc8fa7262e4a79", "reading_canonical_log_prob": -213.88917541503906, "reading_distances": [18, 15, 18, 11], "reading_log_probs": [-213.30035400390625, -227.48992919921875, -216.11029052734375, -221.99244689941406], "reading_logprob_std": 6.318692625310894, "tokenization_path_logprob_std": 174.97913173619344, "trajectory_balance_rmse": 2485.320216215748} {"canonical_length": 7, "canonical_nll": 2.7416093008858815, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "73284d596fde1232", "reading_canonical_log_prob": -19.191265106201172, "reading_distances": [26, 25, 31, 27], "reading_log_probs": [-26.080543518066406, -22.304473876953125, -20.091121673583984, -21.73085594177246], "reading_logprob_std": 2.5325934000390795, "tokenization_path_logprob_std": 29.290445542640065, "trajectory_balance_rmse": 176.77938479364502} {"canonical_length": 47, "canonical_nll": 1.0833739422737283, "generation_rouge_l_f1": 0.15602836879432622, "generation_token_f1": 0.22695035460992907, "prompt_id": "08b3c472b8ed1840", "reading_canonical_log_prob": -50.918575286865234, "reading_distances": [37, 51, 48, 51], "reading_log_probs": [-50.27709197998047, -52.31874084472656, -54.178592681884766, -50.220787048339844], "reading_logprob_std": 1.8911612940219187, "tokenization_path_logprob_std": 63.6894411076199, "trajectory_balance_rmse": 638.2438920272166} {"canonical_length": 149, "canonical_nll": 0.7746859556876573, "generation_rouge_l_f1": 0.29649595687331537, "generation_token_f1": 0.45822102425876005, "prompt_id": "51fa8b33eba8ebe6", "reading_canonical_log_prob": -115.42820739746094, "reading_distances": [40, 36, 42, 45], "reading_log_probs": [-119.04322814941406, -122.62535858154297, -121.24957275390625, -121.67088317871094], "reading_logprob_std": 1.516176614374996, "tokenization_path_logprob_std": 71.72535434876143, "trajectory_balance_rmse": 2350.191411777714} {"canonical_length": 25, "canonical_nll": 0.8124961853027344, "generation_rouge_l_f1": 0.2686567164179105, "generation_token_f1": 0.3134328358208956, "prompt_id": "6e11957de8d37c71", "reading_canonical_log_prob": -20.31240463256836, "reading_distances": [26, 22, 27, 19], "reading_log_probs": [-21.340269088745117, -22.62666130065918, -21.595279693603516, -23.857208251953125], "reading_logprob_std": 1.1456123800738887, "tokenization_path_logprob_std": 30.037048314165784, "trajectory_balance_rmse": 477.7331151610306} {"canonical_length": 124, "canonical_nll": 1.1737483855216735, "generation_rouge_l_f1": 0.2210796915167095, "generation_token_f1": 0.4061696658097686, "prompt_id": "613e08997d787f9d", "reading_canonical_log_prob": -145.5447998046875, "reading_distances": [21, 18, 24, 20], "reading_log_probs": [-143.04635620117188, -145.76669311523438, -143.3068084716797, -147.36878967285156], "reading_logprob_std": 2.066981223639106, "tokenization_path_logprob_std": 200.7140928755919, "trajectory_balance_rmse": 2785.6575848175653} {"canonical_length": 7, "canonical_nll": 0.16052302292415074, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "82a7bb73a8e6bdca", "reading_canonical_log_prob": -1.1236611604690552, "reading_distances": [33, 34, 31, 40], "reading_log_probs": [-1.8113667964935303, -1.445715308189392, -2.253610610961914, -1.5671390295028687], "reading_logprob_std": 0.3567923739135296, "tokenization_path_logprob_std": 16.383844449223908, "trajectory_balance_rmse": 104.97244631531049} {"canonical_length": 500, "canonical_nll": 0.81297900390625, "generation_rouge_l_f1": 0.15172413793103448, "generation_token_f1": 0.2620689655172414, "prompt_id": "04c1fbf027769e8a", "reading_canonical_log_prob": -406.489501953125, "reading_distances": [24, 24, 28, 22], "reading_log_probs": [-417.69384765625, -417.3563537597656, -428.31036376953125, -417.96038818359375], "reading_logprob_std": 5.325821907423919, "tokenization_path_logprob_std": 200.21467146696898, "trajectory_balance_rmse": 8043.086034663614} {"canonical_length": 47, "canonical_nll": 1.896366038220994, "generation_rouge_l_f1": 0.17322834645669294, "generation_token_f1": 0.2677165354330709, "prompt_id": "5950f3bf48416566", "reading_canonical_log_prob": -89.12920379638672, "reading_distances": [32, 35, 36, 48], "reading_log_probs": [-96.4745101928711, -89.72310638427734, -92.9197006225586, -91.32283020019531], "reading_logprob_std": 2.8879816679367125, "tokenization_path_logprob_std": 73.11880625675266, "trajectory_balance_rmse": 1057.195269695691} {"canonical_length": 310, "canonical_nll": 1.1810854019657258, "generation_rouge_l_f1": 0.2756183745583039, "generation_token_f1": 0.4770318021201413, "prompt_id": "f93b23d1ff15b3d1", "reading_canonical_log_prob": -366.136474609375, "reading_distances": [22, 25, 23, 24], "reading_log_probs": [-361.2095947265625, -361.9642333984375, -361.508544921875, -359.565673828125], "reading_logprob_std": 1.0447015024886719, "tokenization_path_logprob_std": 249.32821451384623, "trajectory_balance_rmse": 5735.917032198753} {"canonical_length": 228, "canonical_nll": 1.3983487580951892, "generation_rouge_l_f1": 0.2414486921529175, "generation_token_f1": 0.45070422535211263, "prompt_id": "cb117bc3f0e8e883", "reading_canonical_log_prob": -318.8235168457031, "reading_distances": [25, 19, 19, 23], "reading_log_probs": [-328.0265808105469, -324.98382568359375, -320.8035583496094, -326.6986083984375], "reading_logprob_std": 3.1406030840963086, "tokenization_path_logprob_std": 300.3520167844739, "trajectory_balance_rmse": 4702.608865122326} {"canonical_length": 318, "canonical_nll": 1.8733478282232705, "generation_rouge_l_f1": 0.21832358674463936, "generation_token_f1": 0.49902534113060426, "prompt_id": "0ff4ea021b8e685b", "reading_canonical_log_prob": -595.724609375, "reading_distances": [16, 20, 14, 14], "reading_log_probs": [-599.964111328125, -606.0427856445312, -603.8404541015625, -600.5645141601562], "reading_logprob_std": 2.8567216566693108, "tokenization_path_logprob_std": 220.25505881231632, "trajectory_balance_rmse": 6493.275473559726} {"canonical_length": 70, "canonical_nll": 1.517994907924107, "generation_rouge_l_f1": 0.22352941176470587, "generation_token_f1": 0.3529411764705882, "prompt_id": "d8931b93e350f362", "reading_canonical_log_prob": -106.2596435546875, "reading_distances": [14, 16, 14, 11], "reading_log_probs": [-110.76921081542969, -112.06926727294922, -106.95133972167969, -112.42951965332031], "reading_logprob_std": 2.505931820691401, "tokenization_path_logprob_std": 115.00975243022906, "trajectory_balance_rmse": 1643.7055263595494} {"canonical_length": 43, "canonical_nll": 1.9972502242687136, "generation_rouge_l_f1": 0.3488372093023256, "generation_token_f1": 0.46511627906976744, "prompt_id": "cfd710328af17995", "reading_canonical_log_prob": -85.88175964355469, "reading_distances": [32, 31, 27, 30], "reading_log_probs": [-87.07557678222656, -88.95108032226562, -85.93759155273438, -83.38697814941406], "reading_logprob_std": 2.3267334053068076, "tokenization_path_logprob_std": 80.18251608949525, "trajectory_balance_rmse": 840.5582333979407} {"canonical_length": 237, "canonical_nll": 0.7680912581174183, "generation_rouge_l_f1": 0.2993492407809111, "generation_token_f1": 0.3774403470715835, "prompt_id": "3f921f925896c713", "reading_canonical_log_prob": -182.03762817382812, "reading_distances": [37, 35, 42, 39], "reading_log_probs": [-205.52755737304688, -196.45669555664062, -205.33743286132812, -201.58236694335938], "reading_logprob_std": 4.2536428626587925, "tokenization_path_logprob_std": 109.53918205693012, "trajectory_balance_rmse": 2244.669832983569} {"canonical_length": 242, "canonical_nll": 1.0406487835340263, "generation_rouge_l_f1": 0.3220675944333996, "generation_token_f1": 0.5049701789264414, "prompt_id": "74c9b8835b365ad8", "reading_canonical_log_prob": -251.83700561523438, "reading_distances": [21, 15, 17, 19], "reading_log_probs": [-262.7458190917969, -254.2945556640625, -252.0796661376953, -256.7156677246094], "reading_logprob_std": 4.599034570918568, "tokenization_path_logprob_std": 137.88174392557318, "trajectory_balance_rmse": 4514.873475395476} {"canonical_length": 243, "canonical_nll": 1.2568626874758873, "generation_rouge_l_f1": 0.2553191489361702, "generation_token_f1": 0.5609284332688588, "prompt_id": "d4cbdb2804902b19", "reading_canonical_log_prob": -305.4176330566406, "reading_distances": [26, 16, 23, 24], "reading_log_probs": [-334.05010986328125, -314.25537109375, -325.3602294921875, -317.67816162109375], "reading_logprob_std": 8.800728828162866, "tokenization_path_logprob_std": 187.66568468523275, "trajectory_balance_rmse": 4355.446967332672} {"canonical_length": 140, "canonical_nll": 1.2200587681361608, "generation_rouge_l_f1": 0.22895622895622894, "generation_token_f1": 0.3569023569023569, "prompt_id": "ad65facd3056e4a9", "reading_canonical_log_prob": -170.8082275390625, "reading_distances": [18, 13, 19, 17], "reading_log_probs": [-175.64523315429688, -168.63168334960938, -174.9516143798828, -175.52947998046875], "reading_logprob_std": 3.3855014634067575, "tokenization_path_logprob_std": 75.54832224164802, "trajectory_balance_rmse": 1822.6117958165637} {"canonical_length": 245, "canonical_nll": 1.1764820332429846, "generation_rouge_l_f1": 0.2434782608695652, "generation_token_f1": 0.4434782608695652, "prompt_id": "4eba5e8b52e3bb5e", "reading_canonical_log_prob": -288.23809814453125, "reading_distances": [19, 24, 22, 25], "reading_log_probs": [-296.53643798828125, -296.58551025390625, -297.6062927246094, -298.02459716796875], "reading_logprob_std": 0.7443991757473701, "tokenization_path_logprob_std": 265.61956187568114, "trajectory_balance_rmse": 4738.1188144174475} {"canonical_length": 118, "canonical_nll": 1.8130945755263506, "generation_rouge_l_f1": 0.17708333333333334, "generation_token_f1": 0.32812499999999994, "prompt_id": "4873d3597b3d71b4", "reading_canonical_log_prob": -213.94515991210938, "reading_distances": [21, 22, 20, 20], "reading_log_probs": [-221.48016357421875, -222.8440399169922, -219.7539520263672, -220.9866485595703], "reading_logprob_std": 1.2780717450316943, "tokenization_path_logprob_std": 149.8376958884325, "trajectory_balance_rmse": 2998.9292977953} {"canonical_length": 38, "canonical_nll": 1.5758445137425472, "generation_rouge_l_f1": 0.21739130434782608, "generation_token_f1": 0.26086956521739124, "prompt_id": "2089d8bccf6a73c9", "reading_canonical_log_prob": -59.8820915222168, "reading_distances": [12, 10, 9, 11], "reading_log_probs": [-69.55670166015625, -71.18232727050781, -68.76402282714844, -70.73959350585938], "reading_logprob_std": 1.103680228058016, "tokenization_path_logprob_std": 107.26446113894306, "trajectory_balance_rmse": 689.4835414053827} {"canonical_length": 277, "canonical_nll": 1.601714537031814, "generation_rouge_l_f1": 0.22310756972111553, "generation_token_f1": 0.42231075697211157, "prompt_id": "9b806d8ec328a266", "reading_canonical_log_prob": -443.6749267578125, "reading_distances": [31, 23, 30, 30], "reading_log_probs": [-452.97442626953125, -450.75103759765625, -454.1072692871094, -450.98162841796875], "reading_logprob_std": 1.6146498043547408, "tokenization_path_logprob_std": 234.9261651507821, "trajectory_balance_rmse": 6348.836685535419} {"canonical_length": 400, "canonical_nll": 0.9056741333007813, "generation_rouge_l_f1": 0.22666666666666666, "generation_token_f1": 0.39333333333333337, "prompt_id": "ce03043d2a4264da", "reading_canonical_log_prob": -362.2696533203125, "reading_distances": [20, 23, 24, 18], "reading_log_probs": [-365.8238525390625, -372.817626953125, -371.13653564453125, -366.89642333984375], "reading_logprob_std": 3.3435654999658624, "tokenization_path_logprob_std": 118.60631443223015, "trajectory_balance_rmse": 6669.494447092999} {"canonical_length": 12, "canonical_nll": 1.224385102589925, "generation_rouge_l_f1": 0.21428571428571427, "generation_token_f1": 0.21428571428571427, "prompt_id": "59741a0906fe97c9", "reading_canonical_log_prob": -14.692621231079102, "reading_distances": [27, 27, 30, 23], "reading_log_probs": [-11.568324089050293, -14.547901153564453, -12.97624397277832, -14.562300682067871], "reading_logprob_std": 1.4378775807688695, "tokenization_path_logprob_std": 51.714895933770954, "trajectory_balance_rmse": 205.3453929457434} {"canonical_length": 103, "canonical_nll": 0.48171693375967084, "generation_rouge_l_f1": 0.2827586206896552, "generation_token_f1": 0.45517241379310347, "prompt_id": "995c665a4ec8066d", "reading_canonical_log_prob": -49.616844177246094, "reading_distances": [31, 27, 34, 26], "reading_log_probs": [-67.41150665283203, -64.91985321044922, -64.62183380126953, -64.10440063476562], "reading_logprob_std": 1.4705179717718824, "tokenization_path_logprob_std": 162.61181192319495, "trajectory_balance_rmse": 1342.979148570965} {"canonical_length": 63, "canonical_nll": 0.8556002662295387, "generation_rouge_l_f1": 0.42168674698795183, "generation_token_f1": 0.4698795180722891, "prompt_id": "e7b51e88b3b34da7", "reading_canonical_log_prob": -53.90281677246094, "reading_distances": [29, 30, 26, 28], "reading_log_probs": [-56.89440155029297, -55.26811599731445, -56.09957504272461, -54.601585388183594], "reading_logprob_std": 0.9963734253282647, "tokenization_path_logprob_std": 95.79614289660866, "trajectory_balance_rmse": 1050.167617000383} {"canonical_length": 90, "canonical_nll": 1.1765779283311633, "generation_rouge_l_f1": 0.3699421965317919, "generation_token_f1": 0.6242774566473988, "prompt_id": "6b580b925c517c44", "reading_canonical_log_prob": -105.89201354980469, "reading_distances": [26, 23, 27, 20], "reading_log_probs": [-107.61618041992188, -108.80406188964844, -107.85337829589844, -107.2198257446289], "reading_logprob_std": 0.6732640975715509, "tokenization_path_logprob_std": 140.40753942138184, "trajectory_balance_rmse": 1635.1596774206303} {"canonical_length": 169, "canonical_nll": 0.7255596634904309, "generation_rouge_l_f1": 0.21957040572792363, "generation_token_f1": 0.39618138424821003, "prompt_id": "58b4edff73083c67", "reading_canonical_log_prob": -122.61958312988281, "reading_distances": [54, 51, 52, 62], "reading_log_probs": [-140.18661499023438, -139.57785034179688, -136.07691955566406, -144.39991760253906], "reading_logprob_std": 3.4131272975292037, "tokenization_path_logprob_std": 36.21620847574499, "trajectory_balance_rmse": 2451.950180581133} {"canonical_length": 6, "canonical_nll": 1.8029948870340984, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "dc845859825dc504", "reading_canonical_log_prob": -10.81796932220459, "reading_distances": [33, 29, 30, 35], "reading_log_probs": [-7.969677925109863, -18.538053512573242, -12.391934394836426, -13.603168487548828], "reading_logprob_std": 4.345285656181385, "tokenization_path_logprob_std": 16.252401357377522, "trajectory_balance_rmse": 111.70499672301683} {"canonical_length": 55, "canonical_nll": 0.4249051874334162, "generation_rouge_l_f1": 0.9902912621359222, "generation_token_f1": 0.9902912621359222, "prompt_id": "f074eb196b10e4a0", "reading_canonical_log_prob": -23.36978530883789, "reading_distances": [30, 31, 27, 29], "reading_log_probs": [-19.30567741394043, -20.80806541442871, -24.165637969970703, -21.925228118896484], "reading_logprob_std": 2.0469329593481635, "tokenization_path_logprob_std": 36.75659743401428, "trajectory_balance_rmse": 382.395393616463} {"canonical_length": 39, "canonical_nll": 2.3667479294996996, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "b1076398ceba44c5", "reading_canonical_log_prob": -92.30316925048828, "reading_distances": [9, 12, 15, 13], "reading_log_probs": [-94.03843688964844, -91.14710998535156, -92.61165618896484, -93.514404296875], "reading_logprob_std": 1.2660310909777466, "tokenization_path_logprob_std": 57.59941279187505, "trajectory_balance_rmse": 821.0170921714741} {"canonical_length": 278, "canonical_nll": 1.8083013081722121, "generation_rouge_l_f1": 0.2331511839708561, "generation_token_f1": 0.44080145719489977, "prompt_id": "43d587270196a951", "reading_canonical_log_prob": -502.707763671875, "reading_distances": [23, 24, 29, 22], "reading_log_probs": [-520.5206909179688, -513.0958251953125, -514.38818359375, -513.4569702148438], "reading_logprob_std": 3.479705466044351, "tokenization_path_logprob_std": 255.14816484487633, "trajectory_balance_rmse": 6187.289710890969} {"canonical_length": 77, "canonical_nll": 1.5067148580179586, "generation_rouge_l_f1": 0.16883116883116883, "generation_token_f1": 0.26623376623376627, "prompt_id": "9ec5781f60b3ed7d", "reading_canonical_log_prob": -116.01704406738281, "reading_distances": [14, 16, 9, 14], "reading_log_probs": [-114.97421264648438, -116.84053802490234, -114.42149353027344, -113.46202087402344], "reading_logprob_std": 1.421915813849965, "tokenization_path_logprob_std": 231.74195782865326, "trajectory_balance_rmse": 1698.556313088805} {"canonical_length": 50, "canonical_nll": 1.2700799560546876, "generation_rouge_l_f1": 0.044444444444444446, "generation_token_f1": 0.044444444444444446, "prompt_id": "9c285d8b80679a4e", "reading_canonical_log_prob": -63.503997802734375, "reading_distances": [21, 24, 26, 20], "reading_log_probs": [-66.5753402709961, -70.70352172851562, -69.35383605957031, -66.81613159179688], "reading_logprob_std": 2.0040230142407207, "tokenization_path_logprob_std": 38.1320604229205, "trajectory_balance_rmse": 1067.1899770885348} {"canonical_length": 260, "canonical_nll": 1.4365875244140625, "generation_rouge_l_f1": 0.25600000000000006, "generation_token_f1": 0.488, "prompt_id": "46fcd75d760af9c5", "reading_canonical_log_prob": -373.51275634765625, "reading_distances": [24, 20, 34, 25], "reading_log_probs": [-372.7444763183594, -370.1941223144531, -370.8607177734375, -372.339111328125], "reading_logprob_std": 1.2058243187880406, "tokenization_path_logprob_std": 128.67635284365122, "trajectory_balance_rmse": 5207.718481487184} {"canonical_length": 328, "canonical_nll": 1.0758116187118902, "generation_rouge_l_f1": 0.19377162629757785, "generation_token_f1": 0.3737024221453287, "prompt_id": "5feba3faef17f1e8", "reading_canonical_log_prob": -352.8662109375, "reading_distances": [27, 24, 28, 26], "reading_log_probs": [-368.6459045410156, -368.57647705078125, -370.11370849609375, -366.7442321777344], "reading_logprob_std": 1.379891265506685, "tokenization_path_logprob_std": 357.0160867335868, "trajectory_balance_rmse": 7035.446492771807} {"canonical_length": 232, "canonical_nll": 1.1531643045359645, "generation_rouge_l_f1": 0.2597938144329897, "generation_token_f1": 0.49896907216494846, "prompt_id": "49ec1dbe6be4038c", "reading_canonical_log_prob": -267.53411865234375, "reading_distances": [17, 24, 23, 16], "reading_log_probs": [-271.54888916015625, -273.10736083984375, -273.051513671875, -269.8177185058594], "reading_logprob_std": 1.5536506203498848, "tokenization_path_logprob_std": 222.4753686716794, "trajectory_balance_rmse": 5204.190478926365} {"canonical_length": 368, "canonical_nll": 1.2499258621879246, "generation_rouge_l_f1": 0.19374999999999998, "generation_token_f1": 0.365625, "prompt_id": "af5752cd92c29a4d", "reading_canonical_log_prob": -459.97271728515625, "reading_distances": [17, 20, 19, 15], "reading_log_probs": [-462.298828125, -460.7337646484375, -459.4855651855469, -458.6839904785156], "reading_logprob_std": 1.5767324154825866, "tokenization_path_logprob_std": 440.7308336767898, "trajectory_balance_rmse": 6615.488687342776} {"canonical_length": 152, "canonical_nll": 1.6690595526444285, "generation_rouge_l_f1": 0.15903614457831325, "generation_token_f1": 0.29397590361445786, "prompt_id": "a82619f7741e0303", "reading_canonical_log_prob": -253.69705200195312, "reading_distances": [26, 16, 20, 19], "reading_log_probs": [-260.63507080078125, -258.23809814453125, -259.18072509765625, -261.2874755859375], "reading_logprob_std": 1.3817825840646156, "tokenization_path_logprob_std": 116.05267086145122, "trajectory_balance_rmse": 3862.106710043352} {"canonical_length": 419, "canonical_nll": 1.4756415829169154, "generation_rouge_l_f1": 0.25941422594142255, "generation_token_f1": 0.502092050209205, "prompt_id": "9294062be51005dc", "reading_canonical_log_prob": -618.2938232421875, "reading_distances": [19, 22, 20, 26], "reading_log_probs": [-616.85498046875, -619.6953735351562, -620.9141845703125, -619.091796875], "reading_logprob_std": 1.7009610122229462, "tokenization_path_logprob_std": 150.0350866920364, "trajectory_balance_rmse": 6516.996952363051} {"canonical_length": 9, "canonical_nll": 0.10177124208874172, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "3e7c9dafc0fa7462", "reading_canonical_log_prob": -0.9159411787986755, "reading_distances": [23, 32, 26, 29], "reading_log_probs": [-1.3242636919021606, -1.3076157569885254, -0.6997504234313965, -1.0150034427642822], "reading_logprob_std": 0.2944546618050103, "tokenization_path_logprob_std": 20.052229306649036, "trajectory_balance_rmse": 192.54449683496665} {"canonical_length": 12, "canonical_nll": 0.7136015892028809, "generation_rouge_l_f1": 0.75, "generation_token_f1": 0.75, "prompt_id": "c34fdb5fae4b8ff6", "reading_canonical_log_prob": -8.56321907043457, "reading_distances": [54, 44, 42, 59], "reading_log_probs": [-8.982990264892578, -9.261734962463379, -9.082284927368164, -9.89128303527832], "reading_logprob_std": 0.4077953758563517, "tokenization_path_logprob_std": 91.07845320086382, "trajectory_balance_rmse": 294.30289580484373} {"canonical_length": 122, "canonical_nll": 1.6669535402391777, "generation_rouge_l_f1": 0.18658892128279883, "generation_token_f1": 0.32069970845481055, "prompt_id": "d51b2c6c349d91e4", "reading_canonical_log_prob": -203.3683319091797, "reading_distances": [20, 25, 30, 31], "reading_log_probs": [-205.41184997558594, -200.8694305419922, -204.72247314453125, -202.15994262695312], "reading_logprob_std": 2.1362279357531833, "tokenization_path_logprob_std": 50.672654815272075, "trajectory_balance_rmse": 2427.085251410316} {"canonical_length": 311, "canonical_nll": 1.3430158118343047, "generation_rouge_l_f1": 0.20738137082601055, "generation_token_f1": 0.36906854130052724, "prompt_id": "fc10a987783bea33", "reading_canonical_log_prob": -417.67791748046875, "reading_distances": [27, 25, 26, 25], "reading_log_probs": [-430.02996826171875, -432.321044921875, -427.3624267578125, -429.18438720703125], "reading_logprob_std": 2.058030467327892, "tokenization_path_logprob_std": 457.88389294623175, "trajectory_balance_rmse": 7155.480811617138} {"canonical_length": 31, "canonical_nll": 1.286988781344506, "generation_rouge_l_f1": 0.11678832116788321, "generation_token_f1": 0.13138686131386862, "prompt_id": "7386d29fb448f53c", "reading_canonical_log_prob": -39.89665222167969, "reading_distances": [15, 17, 12, 17], "reading_log_probs": [-35.09310531616211, -35.22724914550781, -37.87216567993164, -38.231353759765625], "reading_logprob_std": 1.6767781221332057, "tokenization_path_logprob_std": 44.923798799850715, "trajectory_balance_rmse": 371.77752121715366} {"canonical_length": 253, "canonical_nll": 0.9028679768558547, "generation_rouge_l_f1": 0.2067669172932331, "generation_token_f1": 0.3947368421052631, "prompt_id": "d5d2ea5bf0257ac3", "reading_canonical_log_prob": -228.42559814453125, "reading_distances": [34, 33, 30, 34], "reading_log_probs": [-260.81597900390625, -254.19732666015625, -252.66966247558594, -255.9256134033203], "reading_logprob_std": 3.535612357447625, "tokenization_path_logprob_std": 159.59248603896432, "trajectory_balance_rmse": 5667.497503089309} {"canonical_length": 379, "canonical_nll": 1.13995361328125, "generation_rouge_l_f1": 0.157973174366617, "generation_token_f1": 0.29210134128166915, "prompt_id": "0a54bd903d89ace0", "reading_canonical_log_prob": -432.04241943359375, "reading_distances": [21, 18, 22, 20], "reading_log_probs": [-431.86602783203125, -433.56390380859375, -435.83172607421875, -429.0708923339844], "reading_logprob_std": 2.849873836898258, "tokenization_path_logprob_std": 145.53084160893047, "trajectory_balance_rmse": 6996.6611848013845} {"canonical_length": 128, "canonical_nll": 1.2298705577850342, "generation_rouge_l_f1": 0.22395833333333331, "generation_token_f1": 0.38020833333333337, "prompt_id": "49a85c182167a715", "reading_canonical_log_prob": -157.42343139648438, "reading_distances": [15, 10, 18, 18], "reading_log_probs": [-164.73074340820312, -163.36647033691406, -166.1029815673828, -171.0482177734375], "reading_logprob_std": 3.3492270185402107, "tokenization_path_logprob_std": 131.9235507058643, "trajectory_balance_rmse": 2874.5046903454} {"canonical_length": 127, "canonical_nll": 1.2365133931317667, "generation_rouge_l_f1": 0.1208053691275168, "generation_token_f1": 0.18791946308724833, "prompt_id": "50a15e61cf0ab437", "reading_canonical_log_prob": -157.03720092773438, "reading_distances": [35, 28, 41, 28], "reading_log_probs": [-169.33316040039062, -178.7964630126953, -183.6971435546875, -182.1249237060547], "reading_logprob_std": 6.43606059941072, "tokenization_path_logprob_std": 205.67540621031873, "trajectory_balance_rmse": 3184.3173293633304} {"canonical_length": 14, "canonical_nll": 0.5653907571520124, "generation_rouge_l_f1": 0.6923076923076923, "generation_token_f1": 0.6923076923076923, "prompt_id": "394139a0ae5c1d75", "reading_canonical_log_prob": -7.915470600128174, "reading_distances": [20, 29, 26, 29], "reading_log_probs": [-5.899117469787598, -6.496757984161377, -6.894381523132324, -6.343422889709473], "reading_logprob_std": 0.4113295082362613, "tokenization_path_logprob_std": 43.350331212773966, "trajectory_balance_rmse": 225.48252854786776} {"canonical_length": 58, "canonical_nll": 0.533273302275559, "generation_rouge_l_f1": 0.03278688524590165, "generation_token_f1": 0.03278688524590165, "prompt_id": "ee3925e13ca1cdfe", "reading_canonical_log_prob": -30.929851531982422, "reading_distances": [21, 20, 14, 22], "reading_log_probs": [-29.850460052490234, -29.904926300048828, -30.892807006835938, -29.159687042236328], "reading_logprob_std": 0.7130694070339504, "tokenization_path_logprob_std": 13.688333731546107, "trajectory_balance_rmse": 861.8350839801363} {"canonical_length": 293, "canonical_nll": 1.1717400143984642, "generation_rouge_l_f1": 0.29824561403508765, "generation_token_f1": 0.5894736842105264, "prompt_id": "e5cf7a10fe204155", "reading_canonical_log_prob": -343.31982421875, "reading_distances": [28, 24, 27, 30], "reading_log_probs": [-375.53466796875, -361.07293701171875, -376.4525146484375, -368.8734436035156], "reading_logprob_std": 7.1249423630502715, "tokenization_path_logprob_std": 210.0727512096483, "trajectory_balance_rmse": 5648.077046696184} {"canonical_length": 302, "canonical_nll": 1.3239526811814466, "generation_rouge_l_f1": 0.24125874125874125, "generation_token_f1": 0.472027972027972, "prompt_id": "a6e94611dc8387e2", "reading_canonical_log_prob": -399.8337097167969, "reading_distances": [32, 34, 29, 28], "reading_log_probs": [-400.45550537109375, -401.34326171875, -399.42840576171875, -399.7647705078125], "reading_logprob_std": 0.8461485622776358, "tokenization_path_logprob_std": 284.32767173903, "trajectory_balance_rmse": 6488.9211625788275} {"canonical_length": 195, "canonical_nll": 1.8380167643229166, "generation_rouge_l_f1": 0.251207729468599, "generation_token_f1": 0.4975845410628019, "prompt_id": "747395c5d51f280d", "reading_canonical_log_prob": -358.41326904296875, "reading_distances": [21, 17, 16, 15], "reading_log_probs": [-381.1961669921875, -382.2222595214844, -379.86810302734375, -377.99871826171875], "reading_logprob_std": 1.823806713332709, "tokenization_path_logprob_std": 158.78771307678028, "trajectory_balance_rmse": 4037.4557622252382} {"canonical_length": 35, "canonical_nll": 0.8661948067801339, "generation_rouge_l_f1": 0.37209302325581395, "generation_token_f1": 0.37209302325581395, "prompt_id": "c31ebbb64961b272", "reading_canonical_log_prob": -30.316818237304688, "reading_distances": [16, 17, 15, 20], "reading_log_probs": [-34.01789474487305, -30.660837173461914, -33.79148483276367, -31.443504333496094], "reading_logprob_std": 1.6801566903934921, "tokenization_path_logprob_std": 36.842985949159264, "trajectory_balance_rmse": 360.0408623634905} {"canonical_length": 111, "canonical_nll": 1.3858903764604449, "generation_rouge_l_f1": 0.26912928759894456, "generation_token_f1": 0.40105540897097625, "prompt_id": "0a9b1baeb01418e9", "reading_canonical_log_prob": -153.83383178710938, "reading_distances": [25, 28, 28, 22], "reading_log_probs": [-156.8324432373047, -156.31260681152344, -155.3677978515625, -156.16915893554688], "reading_logprob_std": 0.6062715928369427, "tokenization_path_logprob_std": 123.66807467824212, "trajectory_balance_rmse": 2285.2726746152675} {"canonical_length": 86, "canonical_nll": 1.0531028924986374, "generation_rouge_l_f1": 0.2962962962962963, "generation_token_f1": 0.4722222222222222, "prompt_id": "ffff4b859d3e455e", "reading_canonical_log_prob": -90.56684875488281, "reading_distances": [18, 19, 17, 19], "reading_log_probs": [-92.73799896240234, -94.16004943847656, -95.02366638183594, -93.92874145507812], "reading_logprob_std": 0.9426279310127137, "tokenization_path_logprob_std": 72.8998269338888, "trajectory_balance_rmse": 2155.905826174077} {"canonical_length": 12, "canonical_nll": 0.7095138231913248, "generation_rouge_l_f1": 0.27397260273972607, "generation_token_f1": 0.27397260273972607, "prompt_id": "a9db5a2fccb43cd0", "reading_canonical_log_prob": -8.514165878295898, "reading_distances": [21, 28, 23, 23], "reading_log_probs": [-9.120794296264648, -10.764845848083496, -9.445552825927734, -11.468722343444824], "reading_logprob_std": 1.1049257447902538, "tokenization_path_logprob_std": 81.62866714435455, "trajectory_balance_rmse": 317.5652360531834} {"canonical_length": 115, "canonical_nll": 1.8979163128396739, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "6d2c7ff3221f2284", "reading_canonical_log_prob": -218.2603759765625, "reading_distances": [11, 13, 15, 13], "reading_log_probs": [-217.90655517578125, -212.68429565429688, -214.44485473632812, -215.3583526611328], "reading_logprob_std": 2.176260709597588, "tokenization_path_logprob_std": 171.4949643077366, "trajectory_balance_rmse": 1613.5395713354112} {"canonical_length": 16, "canonical_nll": 0.9616560935974121, "generation_rouge_l_f1": 0.23076923076923075, "generation_token_f1": 0.23076923076923075, "prompt_id": "686af343746a5e07", "reading_canonical_log_prob": -15.386497497558594, "reading_distances": [22, 25, 24, 26], "reading_log_probs": [-24.17514419555664, -18.717844009399414, -18.458589553833008, -19.70053482055664], "reading_logprob_std": 2.66237008291927, "tokenization_path_logprob_std": 69.5715694623994, "trajectory_balance_rmse": 385.5796434313235} {"canonical_length": 8, "canonical_nll": 3.2749288082122803, "generation_rouge_l_f1": 0.031746031746031744, "generation_token_f1": 0.06349206349206349, "prompt_id": "90063df346a3b4e5", "reading_canonical_log_prob": -26.199430465698242, "reading_distances": [19, 18, 22, 26], "reading_log_probs": [-27.854938507080078, -26.04607391357422, -29.172645568847656, -26.21005630493164], "reading_logprob_std": 1.4802380485392932, "tokenization_path_logprob_std": 27.95656264790823, "trajectory_balance_rmse": 243.36032610152907} {"canonical_length": 276, "canonical_nll": 0.8121829931286798, "generation_rouge_l_f1": 0.2664165103189493, "generation_token_f1": 0.5328330206378986, "prompt_id": "73f47d65b9b8ef25", "reading_canonical_log_prob": -224.16250610351562, "reading_distances": [25, 20, 25, 29], "reading_log_probs": [-240.641845703125, -236.28192138671875, -246.2879180908203, -236.24581909179688], "reading_logprob_std": 4.753745445643961, "tokenization_path_logprob_std": 297.19980905408187, "trajectory_balance_rmse": 4877.858275522187} {"canonical_length": 73, "canonical_nll": 0.8555358991231004, "generation_rouge_l_f1": 0.4787234042553191, "generation_token_f1": 0.5957446808510638, "prompt_id": "96bcd59d4201d57e", "reading_canonical_log_prob": -62.45412063598633, "reading_distances": [19, 20, 18, 20], "reading_log_probs": [-66.34358978271484, -66.43382263183594, -63.51869583129883, -65.1117935180664], "reading_logprob_std": 1.3628761047217184, "tokenization_path_logprob_std": 95.51139106455446, "trajectory_balance_rmse": 1115.547634742293} {"canonical_length": 159, "canonical_nll": 1.2828271254053656, "generation_rouge_l_f1": 0.22543352601156066, "generation_token_f1": 0.44508670520231214, "prompt_id": "3d7e3fa66ce5f5da", "reading_canonical_log_prob": -203.96951293945312, "reading_distances": [24, 22, 25, 26], "reading_log_probs": [-210.34188842773438, -216.30380249023438, -212.04168701171875, -212.59463500976562], "reading_logprob_std": 2.512268312974642, "tokenization_path_logprob_std": 137.714271485503, "trajectory_balance_rmse": 3784.995113461206} {"canonical_length": 40, "canonical_nll": 0.8881778717041016, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.37037037037037035, "prompt_id": "d99d42ddbcbf4292", "reading_canonical_log_prob": -35.52711486816406, "reading_distances": [18, 15, 21, 17], "reading_log_probs": [-34.535606384277344, -36.644264221191406, -36.96298599243164, -37.33318328857422], "reading_logprob_std": 1.254268885900437, "tokenization_path_logprob_std": 95.59807119475074, "trajectory_balance_rmse": 646.9769160379223} {"canonical_length": 15, "canonical_nll": 0.8902035395304362, "generation_rouge_l_f1": 0.10227272727272727, "generation_token_f1": 0.11363636363636363, "prompt_id": "4bb936b473e4256b", "reading_canonical_log_prob": -13.353053092956543, "reading_distances": [17, 17, 23, 18], "reading_log_probs": [-12.797606468200684, -13.454937934875488, -10.692862510681152, -11.381438255310059], "reading_logprob_std": 1.267220130745022, "tokenization_path_logprob_std": 35.3267600430657, "trajectory_balance_rmse": 209.75852224562234} {"canonical_length": 378, "canonical_nll": 1.360090144727596, "generation_rouge_l_f1": 0.2786885245901639, "generation_token_f1": 0.5377049180327869, "prompt_id": "d413361106deda8e", "reading_canonical_log_prob": -514.1140747070312, "reading_distances": [24, 24, 35, 20], "reading_log_probs": [-518.520263671875, -518.2807006835938, -517.896484375, -518.9713134765625], "reading_logprob_std": 0.4499780329860466, "tokenization_path_logprob_std": 75.22969165721995, "trajectory_balance_rmse": 8088.272659317319} {"canonical_length": 370, "canonical_nll": 0.593709337389147, "generation_rouge_l_f1": 0.32085561497326204, "generation_token_f1": 0.47058823529411764, "prompt_id": "2807bad41faf5ef6", "reading_canonical_log_prob": -219.67245483398438, "reading_distances": [20, 12, 12, 15], "reading_log_probs": [-229.22801208496094, -230.34725952148438, -230.90609741210938, -229.7265625], "reading_logprob_std": 0.7306449716434008, "tokenization_path_logprob_std": 225.0499358445882, "trajectory_balance_rmse": 5645.422406865508} {"canonical_length": 76, "canonical_nll": 1.497072420622173, "generation_rouge_l_f1": 0.24669603524229072, "generation_token_f1": 0.4229074889867841, "prompt_id": "f47b46ea69a09ffc", "reading_canonical_log_prob": -113.77750396728516, "reading_distances": [26, 30, 23, 20], "reading_log_probs": [-113.05683898925781, -111.29364776611328, -110.73210144042969, -116.90263366699219], "reading_logprob_std": 2.7861997675266816, "tokenization_path_logprob_std": 71.07138068082668, "trajectory_balance_rmse": 1635.2898531769915} {"canonical_length": 148, "canonical_nll": 1.3768650776631124, "generation_rouge_l_f1": 0.24299065420560748, "generation_token_f1": 0.35046728971962615, "prompt_id": "abdba20a44568a50", "reading_canonical_log_prob": -203.77603149414062, "reading_distances": [31, 31, 31, 29], "reading_log_probs": [-222.52093505859375, -212.57057189941406, -215.22506713867188, -219.36019897460938], "reading_logprob_std": 4.401461541287147, "tokenization_path_logprob_std": 196.97274709950037, "trajectory_balance_rmse": 3276.2576536087945} {"canonical_length": 19, "canonical_nll": 0.52247112675717, "generation_rouge_l_f1": 0.054901960784313725, "generation_token_f1": 0.07058823529411765, "prompt_id": "f10790becafaebfa", "reading_canonical_log_prob": -9.92695140838623, "reading_distances": [31, 29, 23, 29], "reading_log_probs": [-14.346477508544922, -13.345208168029785, -9.87272834777832, -13.165154457092285], "reading_logprob_std": 1.9438600156170616, "tokenization_path_logprob_std": 78.75711006988584, "trajectory_balance_rmse": 280.878702472041} {"canonical_length": 7, "canonical_nll": 0.4494248798915318, "generation_rouge_l_f1": 0.8, "generation_token_f1": 0.8, "prompt_id": "3eb075dbec50f70d", "reading_canonical_log_prob": -3.1459741592407227, "reading_distances": [65, 69, 68, 66], "reading_log_probs": [-3.126750946044922, -2.4081778526306152, -2.143880844116211, -1.885917067527771], "reading_logprob_std": 0.534727671212019, "tokenization_path_logprob_std": 51.69822223069899, "trajectory_balance_rmse": 141.53594243334874} {"canonical_length": 5, "canonical_nll": 1.0905746459960937, "generation_rouge_l_f1": 0.07619047619047618, "generation_token_f1": 0.07619047619047618, "prompt_id": "56ce6de59318b6ea", "reading_canonical_log_prob": -5.452873229980469, "reading_distances": [33, 24, 33, 24], "reading_log_probs": [-15.800033569335938, -11.985738754272461, -13.604837417602539, -14.779488563537598], "reading_logprob_std": 1.6384842330305867, "tokenization_path_logprob_std": 42.96899039852065, "trajectory_balance_rmse": 91.1176540536691} {"canonical_length": 128, "canonical_nll": 1.2636088132858276, "generation_rouge_l_f1": 0.23136246786632392, "generation_token_f1": 0.4370179948586118, "prompt_id": "7b4da759d302cb91", "reading_canonical_log_prob": -161.74192810058594, "reading_distances": [51, 42, 40, 46], "reading_log_probs": [-166.25057983398438, -161.8417510986328, -161.49795532226562, -162.21090698242188], "reading_logprob_std": 2.21936437947907, "tokenization_path_logprob_std": 77.15526110009071, "trajectory_balance_rmse": 2577.538882427533} {"canonical_length": 179, "canonical_nll": 1.9583157161094622, "generation_rouge_l_f1": 0.2559241706161137, "generation_token_f1": 0.4786729857819905, "prompt_id": "e1ce0d3f6081406f", "reading_canonical_log_prob": -350.53851318359375, "reading_distances": [15, 18, 20, 21], "reading_log_probs": [-345.51373291015625, -344.82342529296875, -344.2745361328125, -345.4786682128906], "reading_logprob_std": 0.5911795731726105, "tokenization_path_logprob_std": 168.00120263027554, "trajectory_balance_rmse": 4273.144471856584} {"canonical_length": 193, "canonical_nll": 1.043547160884877, "generation_rouge_l_f1": 0.1413612565445026, "generation_token_f1": 0.23036649214659685, "prompt_id": "50afdbd8ae3129b4", "reading_canonical_log_prob": -201.40460205078125, "reading_distances": [20, 20, 21, 18], "reading_log_probs": [-207.77027893066406, -206.54806518554688, -205.5362091064453, -204.04818725585938], "reading_logprob_std": 1.5765545838384196, "tokenization_path_logprob_std": 58.205421671219334, "trajectory_balance_rmse": 2654.4882781216675} {"canonical_length": 308, "canonical_nll": 0.9199671559519582, "generation_rouge_l_f1": 0.2323943661971831, "generation_token_f1": 0.4647887323943662, "prompt_id": "517e8812395e8d56", "reading_canonical_log_prob": -283.3498840332031, "reading_distances": [13, 15, 16, 16], "reading_log_probs": [-289.31390380859375, -291.67340087890625, -290.48052978515625, -288.0340576171875], "reading_logprob_std": 1.560428246754409, "tokenization_path_logprob_std": 277.9167925243984, "trajectory_balance_rmse": 4868.355878251643} {"canonical_length": 9, "canonical_nll": 1.3689329359266493, "generation_rouge_l_f1": 0.13333333333333333, "generation_token_f1": 0.13333333333333333, "prompt_id": "7febc4a2c9567c65", "reading_canonical_log_prob": -12.320396423339844, "reading_distances": [33, 34, 34, 39], "reading_log_probs": [-21.05249786376953, -16.37981605529785, -15.12768840789795, -19.33171844482422], "reading_logprob_std": 2.705762740496757, "tokenization_path_logprob_std": 50.887642763496174, "trajectory_balance_rmse": 115.52099349579937} {"canonical_length": 13, "canonical_nll": 0.7833364926851712, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.6, "prompt_id": "a3b4630e36098b06", "reading_canonical_log_prob": -10.183374404907227, "reading_distances": [39, 44, 46, 44], "reading_log_probs": [-14.478515625, -14.521512985229492, -14.629406929016113, -13.251228332519531], "reading_logprob_std": 0.6490693018183814, "tokenization_path_logprob_std": 43.805875773058986, "trajectory_balance_rmse": 362.15020875843686} {"canonical_length": 67, "canonical_nll": 1.0006482707920359, "generation_rouge_l_f1": 0.21764705882352942, "generation_token_f1": 0.2882352941176471, "prompt_id": "0d478580ddc19eb6", "reading_canonical_log_prob": -67.0434341430664, "reading_distances": [26, 22, 27, 24], "reading_log_probs": [-64.33454132080078, -66.58795166015625, -67.13223266601562, -65.74710845947266], "reading_logprob_std": 1.2186778734516022, "tokenization_path_logprob_std": 90.14129083273116, "trajectory_balance_rmse": 1485.1396407212258} {"canonical_length": 5, "canonical_nll": 4.103171157836914, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.5714285714285715, "prompt_id": "7c79a39b53212c05", "reading_canonical_log_prob": -20.51585578918457, "reading_distances": [33, 32, 33, 33], "reading_log_probs": [-20.016897201538086, -19.80022430419922, -20.26087760925293, -20.691909790039062], "reading_logprob_std": 0.3824492114878681, "tokenization_path_logprob_std": 25.7614712344112, "trajectory_balance_rmse": 118.16307318842824} {"canonical_length": 89, "canonical_nll": 0.9667280390021507, "generation_rouge_l_f1": 0.391304347826087, "generation_token_f1": 0.5434782608695652, "prompt_id": "1791e9720c6a20af", "reading_canonical_log_prob": -86.0387954711914, "reading_distances": [25, 25, 25, 20], "reading_log_probs": [-102.07977294921875, -115.73123931884766, -98.38719177246094, -97.13748931884766], "reading_logprob_std": 8.52706054202091, "tokenization_path_logprob_std": 122.92674719727867, "trajectory_balance_rmse": 2159.550434697693} {"canonical_length": 14, "canonical_nll": 1.3383862631661552, "generation_rouge_l_f1": 0.15789473684210523, "generation_token_f1": 0.15789473684210523, "prompt_id": "a58d338f47d878b7", "reading_canonical_log_prob": -18.737407684326172, "reading_distances": [9, 9, 9, 8], "reading_log_probs": [-21.233129501342773, -16.19882583618164, -18.35839080810547, -17.90896224975586], "reading_logprob_std": 2.090627842579562, "tokenization_path_logprob_std": 40.297614855150826, "trajectory_balance_rmse": 256.92644688106145} {"canonical_length": 5, "canonical_nll": 1.4150254249572753, "generation_rouge_l_f1": 0.39999999999999997, "generation_token_f1": 0.39999999999999997, "prompt_id": "c979ade41611b06d", "reading_canonical_log_prob": -7.075127124786377, "reading_distances": [30, 29, 36, 38], "reading_log_probs": [-9.753704071044922, -6.880867004394531, -8.295276641845703, -11.10317611694336], "reading_logprob_std": 1.8237787353711457, "tokenization_path_logprob_std": 17.496596084111424, "trajectory_balance_rmse": 95.05553387766965} {"canonical_length": 197, "canonical_nll": 1.0844559258010786, "generation_rouge_l_f1": 0.23965141612200436, "generation_token_f1": 0.4923747276688452, "prompt_id": "58bae05f46b8fa96", "reading_canonical_log_prob": -213.6378173828125, "reading_distances": [21, 21, 21, 18], "reading_log_probs": [-224.21939086914062, -222.85595703125, -223.84869384765625, -224.61866760253906], "reading_logprob_std": 0.7550565403700374, "tokenization_path_logprob_std": 52.925305431878435, "trajectory_balance_rmse": 4133.847567953597} {"canonical_length": 91, "canonical_nll": 1.158413562145862, "generation_rouge_l_f1": 0.021276595744680854, "generation_token_f1": 0.021276595744680854, "prompt_id": "005445d3abeec9e6", "reading_canonical_log_prob": -105.41563415527344, "reading_distances": [18, 23, 20, 21], "reading_log_probs": [-106.40045928955078, -107.20836639404297, -108.78958129882812, -110.24197387695312], "reading_logprob_std": 1.7061240509412745, "tokenization_path_logprob_std": 187.24555755921247, "trajectory_balance_rmse": 2090.151504945627} {"canonical_length": 48, "canonical_nll": 1.3072136243184407, "generation_rouge_l_f1": 0.16176470588235295, "generation_token_f1": 0.2352941176470588, "prompt_id": "3ba7853ae081fbb0", "reading_canonical_log_prob": -62.746253967285156, "reading_distances": [23, 12, 18, 19], "reading_log_probs": [-66.75735473632812, -64.93315124511719, -66.88782501220703, -69.98821258544922], "reading_logprob_std": 2.0970213221385534, "tokenization_path_logprob_std": 46.85519396310243, "trajectory_balance_rmse": 834.7463372583535} {"canonical_length": 60, "canonical_nll": 1.1194854736328126, "generation_rouge_l_f1": 0.161993769470405, "generation_token_f1": 0.24922118380062305, "prompt_id": "19c4e3cd0eb041a5", "reading_canonical_log_prob": -67.16912841796875, "reading_distances": [36, 31, 31, 26], "reading_log_probs": [-80.0254135131836, -67.61659240722656, -66.05419921875, -66.3183822631836], "reading_logprob_std": 6.7159796413948865, "tokenization_path_logprob_std": 157.97959515284364, "trajectory_balance_rmse": 1598.740866972732} {"canonical_length": 139, "canonical_nll": 1.2575565036252248, "generation_rouge_l_f1": 0.33613445378151263, "generation_token_f1": 0.4957983193277311, "prompt_id": "9a7d94fcb2412368", "reading_canonical_log_prob": -174.80035400390625, "reading_distances": [22, 32, 29, 23], "reading_log_probs": [-199.95062255859375, -202.54727172851562, -199.63018798828125, -196.87576293945312], "reading_logprob_std": 2.3195235456247927, "tokenization_path_logprob_std": 170.9708812899398, "trajectory_balance_rmse": 3333.892896169621} {"canonical_length": 27, "canonical_nll": 1.3626027990270544, "generation_rouge_l_f1": 0.19047619047619047, "generation_token_f1": 0.20634920634920637, "prompt_id": "2a25293cb727ba1d", "reading_canonical_log_prob": -36.79027557373047, "reading_distances": [31, 27, 32, 29], "reading_log_probs": [-41.1252326965332, -38.9827995300293, -41.99668884277344, -48.30134582519531], "reading_logprob_std": 4.005346310804879, "tokenization_path_logprob_std": 59.65599726888875, "trajectory_balance_rmse": 707.0508187014813} {"canonical_length": 33, "canonical_nll": 1.691409949100379, "generation_rouge_l_f1": 0.21621621621621623, "generation_token_f1": 0.21621621621621623, "prompt_id": "cc6712ec80ac0d9e", "reading_canonical_log_prob": -55.8165283203125, "reading_distances": [29, 23, 29, 25], "reading_log_probs": [-56.73794174194336, -56.721492767333984, -55.300621032714844, -53.98542022705078], "reading_logprob_std": 1.3190040014799103, "tokenization_path_logprob_std": 72.9543007532632, "trajectory_balance_rmse": 696.148247775823} {"canonical_length": 48, "canonical_nll": 0.5508571068445841, "generation_rouge_l_f1": 0.17167381974248924, "generation_token_f1": 0.27467811158798283, "prompt_id": "aa6e8f58031f9e55", "reading_canonical_log_prob": -26.44114112854004, "reading_distances": [30, 38, 36, 39], "reading_log_probs": [-29.70479393005371, -33.281585693359375, -32.19731521606445, -33.59529495239258], "reading_logprob_std": 1.7647120443670177, "tokenization_path_logprob_std": 122.37887056418212, "trajectory_balance_rmse": 1011.6642751113596} {"canonical_length": 259, "canonical_nll": 1.5134710952582047, "generation_rouge_l_f1": 0.3838771593090211, "generation_token_f1": 0.6295585412667947, "prompt_id": "ef58deb0251c5bd3", "reading_canonical_log_prob": -391.989013671875, "reading_distances": [24, 25, 18, 23], "reading_log_probs": [-393.02923583984375, -393.30413818359375, -393.7992248535156, -394.1148681640625], "reading_logprob_std": 0.4872607562196906, "tokenization_path_logprob_std": 268.9026485861376, "trajectory_balance_rmse": 5471.487014139393} {"canonical_length": 136, "canonical_nll": 1.5478145374971277, "generation_rouge_l_f1": 0.16748768472906403, "generation_token_f1": 0.32019704433497537, "prompt_id": "209efb2fe475a408", "reading_canonical_log_prob": -210.50277709960938, "reading_distances": [25, 28, 23, 24], "reading_log_probs": [-223.52932739257812, -221.99493408203125, -221.54354858398438, -221.8185577392578], "reading_logprob_std": 0.8913893945434324, "tokenization_path_logprob_std": 96.43391333942068, "trajectory_balance_rmse": 3547.3241791975747} {"canonical_length": 16, "canonical_nll": 0.7568477392196655, "generation_rouge_l_f1": 0.17910447761194032, "generation_token_f1": 0.208955223880597, "prompt_id": "aa1f86a1c5954cf7", "reading_canonical_log_prob": -12.109563827514648, "reading_distances": [28, 29, 27, 28], "reading_log_probs": [-12.243049621582031, -16.039432525634766, -12.607356071472168, -12.340551376342773], "reading_logprob_std": 1.8277218110930908, "tokenization_path_logprob_std": 19.13061150622438, "trajectory_balance_rmse": 164.81837645140934} {"canonical_length": 8, "canonical_nll": 0.18501946330070496, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "878c5da5498115bc", "reading_canonical_log_prob": -1.4801557064056396, "reading_distances": [32, 33, 32, 30], "reading_log_probs": [-5.565235614776611, -4.179627895355225, -6.807958126068115, -8.174968719482422], "reading_logprob_std": 1.708180341160144, "tokenization_path_logprob_std": 44.341558497148, "trajectory_balance_rmse": 123.16423261348199} {"canonical_length": 30, "canonical_nll": 0.27333898544311525, "generation_rouge_l_f1": 0.38461538461538464, "generation_token_f1": 0.4999999999999999, "prompt_id": "3e49b0b70cea783a", "reading_canonical_log_prob": -8.200169563293457, "reading_distances": [41, 34, 38, 39], "reading_log_probs": [-9.331878662109375, -10.037626266479492, -9.768878936767578, -9.044351577758789], "reading_logprob_std": 0.4430463081320618, "tokenization_path_logprob_std": 131.7767962191786, "trajectory_balance_rmse": 601.0318048501559} {"canonical_length": 406, "canonical_nll": 1.2534680296047567, "generation_rouge_l_f1": 0.2006172839506173, "generation_token_f1": 0.404320987654321, "prompt_id": "50289ab6cb9cf49e", "reading_canonical_log_prob": -508.90802001953125, "reading_distances": [25, 21, 20, 23], "reading_log_probs": [-501.05755615234375, -503.18878173828125, -500.6357421875, -500.030517578125], "reading_logprob_std": 1.3733717035395965, "tokenization_path_logprob_std": 204.80380653697785, "trajectory_balance_rmse": 7109.133249254779} {"canonical_length": 168, "canonical_nll": 0.3259641783578055, "generation_rouge_l_f1": 0.3146666666666667, "generation_token_f1": 0.56, "prompt_id": "97290896fed51229", "reading_canonical_log_prob": -54.76198196411133, "reading_distances": [30, 30, 34, 39], "reading_log_probs": [-57.10009765625, -55.58306884765625, -55.68038558959961, -60.2017707824707], "reading_logprob_std": 2.1548808380327946, "tokenization_path_logprob_std": 84.28958741636076, "trajectory_balance_rmse": 1465.4674562971486} {"canonical_length": 36, "canonical_nll": 1.1749977535671658, "generation_rouge_l_f1": 0.12949640287769784, "generation_token_f1": 0.15827338129496402, "prompt_id": "11c3ac95688ce9b6", "reading_canonical_log_prob": -42.29991912841797, "reading_distances": [23, 21, 17, 18], "reading_log_probs": [-42.76161575317383, -43.796363830566406, -43.872406005859375, -46.906288146972656], "reading_logprob_std": 1.7880324459540073, "tokenization_path_logprob_std": 53.103843100202624, "trajectory_balance_rmse": 639.6611426012367} {"canonical_length": 12, "canonical_nll": 0.7455742359161377, "generation_rouge_l_f1": 0.1923076923076923, "generation_token_f1": 0.1923076923076923, "prompt_id": "b6d0ecb07aebae2f", "reading_canonical_log_prob": -8.946890830993652, "reading_distances": [26, 27, 26, 27], "reading_log_probs": [-7.956788063049316, -8.356451034545898, -9.93035888671875, -10.820728302001953], "reading_logprob_std": 1.3416243902082716, "tokenization_path_logprob_std": 54.414265103115866, "trajectory_balance_rmse": 259.0950380068309} {"canonical_length": 27, "canonical_nll": 1.202668931749132, "generation_rouge_l_f1": 0.1518987341772152, "generation_token_f1": 0.25316455696202533, "prompt_id": "a7179a571daab3f1", "reading_canonical_log_prob": -32.47206115722656, "reading_distances": [28, 19, 18, 19], "reading_log_probs": [-39.639225006103516, -36.431514739990234, -41.45774459838867, -39.880279541015625], "reading_logprob_std": 2.107523965801572, "tokenization_path_logprob_std": 70.80655928681878, "trajectory_balance_rmse": 406.0885191978273} {"canonical_length": 226, "canonical_nll": 1.0145446642310219, "generation_rouge_l_f1": 0.23214285714285712, "generation_token_f1": 0.2857142857142857, "prompt_id": "a32be7c1cc7860a7", "reading_canonical_log_prob": -229.28709411621094, "reading_distances": [38, 29, 34, 38], "reading_log_probs": [-240.09170532226562, -242.02882385253906, -240.6178741455078, -239.81735229492188], "reading_logprob_std": 0.9843260034866652, "tokenization_path_logprob_std": 204.40341387998393, "trajectory_balance_rmse": 3491.5835811419133} {"canonical_length": 317, "canonical_nll": 1.1507170764429713, "generation_rouge_l_f1": 0.23570190641247832, "generation_token_f1": 0.4679376083188908, "prompt_id": "7ebfd1446f72fed6", "reading_canonical_log_prob": -364.7773132324219, "reading_distances": [25, 29, 19, 24], "reading_log_probs": [-371.8355712890625, -374.3475341796875, -374.27484130859375, -375.52197265625], "reading_logprob_std": 1.5489103002624176, "tokenization_path_logprob_std": 218.45816383692352, "trajectory_balance_rmse": 7413.28487244812} {"canonical_length": 7, "canonical_nll": 0.7846924236842564, "generation_rouge_l_f1": 0.7142857142857143, "generation_token_f1": 0.7142857142857143, "prompt_id": "05c61d7a9c151e3a", "reading_canonical_log_prob": -5.492846965789795, "reading_distances": [48, 57, 44, 50], "reading_log_probs": [-12.114837646484375, -6.958666801452637, -8.282870292663574, -7.0649590492248535], "reading_logprob_std": 2.4155652574488933, "tokenization_path_logprob_std": 60.10883088983452, "trajectory_balance_rmse": 193.49854050250343} {"canonical_length": 20, "canonical_nll": 1.298878002166748, "generation_rouge_l_f1": 0.4313725490196078, "generation_token_f1": 0.5490196078431372, "prompt_id": "3b1c1940eb0db061", "reading_canonical_log_prob": -25.97756004333496, "reading_distances": [19, 25, 18, 21], "reading_log_probs": [-26.45697021484375, -25.457998275756836, -26.107864379882812, -24.757530212402344], "reading_logprob_std": 0.7496832645591919, "tokenization_path_logprob_std": 55.18643279122828, "trajectory_balance_rmse": 536.3716305492595} {"canonical_length": 191, "canonical_nll": 1.4650896481818554, "generation_rouge_l_f1": 0.2271604938271605, "generation_token_f1": 0.4987654320987654, "prompt_id": "13ae7fb963a43898", "reading_canonical_log_prob": -279.8321228027344, "reading_distances": [42, 40, 46, 36], "reading_log_probs": [-306.8354797363281, -308.4251403808594, -310.4430236816406, -315.60211181640625], "reading_logprob_std": 3.81436096323371, "tokenization_path_logprob_std": 34.92730828929617, "trajectory_balance_rmse": 3026.049806925633} {"canonical_length": 346, "canonical_nll": 1.8984099812590318, "generation_rouge_l_f1": 0.17962003454231434, "generation_token_f1": 0.3765112262521589, "prompt_id": "6aeff1f7dd4bd317", "reading_canonical_log_prob": -656.849853515625, "reading_distances": [22, 24, 20, 26], "reading_log_probs": [-680.044189453125, -685.9493408203125, -679.8070068359375, -692.8668212890625], "reading_logprob_std": 6.160935479728225, "tokenization_path_logprob_std": 527.5597705965541, "trajectory_balance_rmse": 6749.471223298045} {"canonical_length": 35, "canonical_nll": 2.8178658621651786, "generation_rouge_l_f1": 0.2631578947368421, "generation_token_f1": 0.3157894736842105, "prompt_id": "a5fa29779141ee07", "reading_canonical_log_prob": -98.62530517578125, "reading_distances": [15, 16, 16, 13], "reading_log_probs": [-98.30657958984375, -97.01901245117188, -98.47029113769531, -99.38179016113281], "reading_logprob_std": 0.9729879745091212, "tokenization_path_logprob_std": 105.03008929493416, "trajectory_balance_rmse": 756.0209999331116} {"canonical_length": 44, "canonical_nll": 1.688167225230824, "generation_rouge_l_f1": 0.09302325581395349, "generation_token_f1": 0.09302325581395349, "prompt_id": "635690e839a377b9", "reading_canonical_log_prob": -74.27935791015625, "reading_distances": [32, 30, 23, 35], "reading_log_probs": [-78.25892639160156, -83.97386169433594, -84.06140899658203, -82.05133819580078], "reading_logprob_std": 2.715017233598987, "tokenization_path_logprob_std": 99.72301497267195, "trajectory_balance_rmse": 1353.6485143102398} {"canonical_length": 124, "canonical_nll": 1.6289793445217995, "generation_rouge_l_f1": 0.14758269720101783, "generation_token_f1": 0.2697201017811705, "prompt_id": "908852670070fd4b", "reading_canonical_log_prob": -201.99343872070312, "reading_distances": [21, 20, 23, 20], "reading_log_probs": [-210.3544921875, -199.8421630859375, -206.43466186523438, -198.17886352539062], "reading_logprob_std": 5.689946382464156, "tokenization_path_logprob_std": 129.89743386088506, "trajectory_balance_rmse": 2893.8941790466174} {"canonical_length": 14, "canonical_nll": 1.592501095363072, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "410cbaa296c6e2c6", "reading_canonical_log_prob": -22.295015335083008, "reading_distances": [25, 24, 23, 21], "reading_log_probs": [-28.350982666015625, -29.54331398010254, -22.938446044921875, -25.66535186767578], "reading_logprob_std": 2.9443265412058284, "tokenization_path_logprob_std": 23.655644740743266, "trajectory_balance_rmse": 350.2095921746504} {"canonical_length": 24, "canonical_nll": 0.9998162587483724, "generation_rouge_l_f1": 0.5098039215686274, "generation_token_f1": 0.5490196078431373, "prompt_id": "b9eb5c3d26d069d5", "reading_canonical_log_prob": -23.995590209960938, "reading_distances": [15, 12, 16, 17], "reading_log_probs": [-28.287372589111328, -25.91518211364746, -33.28711700439453, -24.14548110961914], "reading_logprob_std": 3.966807799390313, "tokenization_path_logprob_std": 82.37341535021659, "trajectory_balance_rmse": 469.6767269786989} {"canonical_length": 120, "canonical_nll": 0.7565411885579427, "generation_rouge_l_f1": 0.31901840490797545, "generation_token_f1": 0.4233128834355828, "prompt_id": "fd96f68afab8f1c0", "reading_canonical_log_prob": -90.78494262695312, "reading_distances": [32, 25, 24, 30], "reading_log_probs": [-96.06948852539062, -93.11294555664062, -93.95042419433594, -96.94761657714844], "reading_logprob_std": 1.788666247147228, "tokenization_path_logprob_std": 47.73611383137815, "trajectory_balance_rmse": 1401.8005477957347} {"canonical_length": 92, "canonical_nll": 2.5318382926609204, "generation_rouge_l_f1": 0.2175732217573222, "generation_token_f1": 0.4435146443514644, "prompt_id": "930a5444ff77055e", "reading_canonical_log_prob": -232.9291229248047, "reading_distances": [46, 50, 45, 43], "reading_log_probs": [-258.9430236816406, -257.96624755859375, -255.26840209960938, -256.1670837402344], "reading_logprob_std": 1.6704731660279024, "tokenization_path_logprob_std": 225.03893002841278, "trajectory_balance_rmse": 2237.1227488690183} {"canonical_length": 88, "canonical_nll": 1.0141091780229048, "generation_rouge_l_f1": 0.225609756097561, "generation_token_f1": 0.36585365853658536, "prompt_id": "6881f74e519464d9", "reading_canonical_log_prob": -89.24160766601562, "reading_distances": [14, 19, 18, 17], "reading_log_probs": [-95.11968994140625, -93.91822814941406, -92.77951049804688, -93.55767822265625], "reading_logprob_std": 0.9743395207821212, "tokenization_path_logprob_std": 219.79677646249894, "trajectory_balance_rmse": 1706.1911981035166} {"canonical_length": 242, "canonical_nll": 1.390915673626356, "generation_rouge_l_f1": 0.19646365422396858, "generation_token_f1": 0.4518664047151277, "prompt_id": "1afba92bdf32d325", "reading_canonical_log_prob": -336.6015930175781, "reading_distances": [32, 28, 26, 33], "reading_log_probs": [-342.6899719238281, -343.3204345703125, -344.50250244140625, -346.37933349609375], "reading_logprob_std": 1.62200443727639, "tokenization_path_logprob_std": 355.3895876849406, "trajectory_balance_rmse": 5384.804436577701} {"canonical_length": 114, "canonical_nll": 1.404857033177426, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.32978723404255317, "prompt_id": "19d150b4a3d84266", "reading_canonical_log_prob": -160.15370178222656, "reading_distances": [35, 36, 33, 35], "reading_log_probs": [-165.65835571289062, -165.70944213867188, -167.0151824951172, -165.3439483642578], "reading_logprob_std": 0.7401580810022593, "tokenization_path_logprob_std": 66.24831227467263, "trajectory_balance_rmse": 2441.066823606261} {"canonical_length": 6, "canonical_nll": 1.4569366772969563, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "dee72943ee9496d3", "reading_canonical_log_prob": -8.741620063781738, "reading_distances": [38, 45, 40, 41], "reading_log_probs": [-6.453496932983398, -4.36970853805542, -7.957554340362549, -5.476649761199951], "reading_logprob_std": 1.522373066960742, "tokenization_path_logprob_std": 20.77078975838271, "trajectory_balance_rmse": 122.1978132444942} {"canonical_length": 40, "canonical_nll": 0.6707370758056641, "generation_rouge_l_f1": 0.19277108433734938, "generation_token_f1": 0.24899598393574301, "prompt_id": "f1a8da244ce8319b", "reading_canonical_log_prob": -26.829483032226562, "reading_distances": [21, 23, 19, 25], "reading_log_probs": [-22.583385467529297, -22.646682739257812, -22.264934539794922, -21.504104614257812], "reading_logprob_std": 0.5244317029998646, "tokenization_path_logprob_std": 36.74788832637045, "trajectory_balance_rmse": 583.8442712159383} {"canonical_length": 286, "canonical_nll": 1.5379977993198208, "generation_rouge_l_f1": 0.2883895131086142, "generation_token_f1": 0.5131086142322099, "prompt_id": "201696a6ff512973", "reading_canonical_log_prob": -439.86737060546875, "reading_distances": [23, 21, 20, 17], "reading_log_probs": [-440.5660400390625, -440.52630615234375, -442.9278564453125, -441.3013000488281], "reading_logprob_std": 1.1230207723196304, "tokenization_path_logprob_std": 209.48996271989594, "trajectory_balance_rmse": 4458.615163617695} {"canonical_length": 202, "canonical_nll": 1.959287737855817, "generation_rouge_l_f1": 0.2616822429906542, "generation_token_f1": 0.39719626168224303, "prompt_id": "8d88867a9f02da1b", "reading_canonical_log_prob": -395.776123046875, "reading_distances": [25, 29, 29, 32], "reading_log_probs": [-399.99407958984375, -396.3551025390625, -399.92926025390625, -397.3677978515625], "reading_logprob_std": 1.8372297101980006, "tokenization_path_logprob_std": 142.0128269557168, "trajectory_balance_rmse": 3456.7835319639225} {"canonical_length": 71, "canonical_nll": 0.7262168669364821, "generation_rouge_l_f1": 0.27586206896551724, "generation_token_f1": 0.3754789272030651, "prompt_id": "de7efec5b0e0f3a3", "reading_canonical_log_prob": -51.561397552490234, "reading_distances": [21, 25, 25, 22], "reading_log_probs": [-53.506004333496094, -52.67863082885742, -51.744056701660156, -51.19600296020508], "reading_logprob_std": 1.020501911551809, "tokenization_path_logprob_std": 20.292203315425812, "trajectory_balance_rmse": 823.4381749166137} {"canonical_length": 8, "canonical_nll": 1.964295744895935, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.18181818181818182, "prompt_id": "5d9fef90efd251af", "reading_canonical_log_prob": -15.71436595916748, "reading_distances": [11, 13, 11, 15], "reading_log_probs": [-11.63593864440918, -13.93332290649414, -15.983759880065918, -11.427576065063477], "reading_logprob_std": 2.149938155726715, "tokenization_path_logprob_std": 21.15644758817164, "trajectory_balance_rmse": 185.07615789144103} {"canonical_length": 35, "canonical_nll": 1.9994035993303572, "generation_rouge_l_f1": 0.11267605633802816, "generation_token_f1": 0.18309859154929578, "prompt_id": "207d314006c9124a", "reading_canonical_log_prob": -69.9791259765625, "reading_distances": [27, 21, 21, 30], "reading_log_probs": [-83.63481140136719, -81.13050079345703, -78.86503601074219, -81.30377197265625], "reading_logprob_std": 1.9486290135909963, "tokenization_path_logprob_std": 82.71515584359108, "trajectory_balance_rmse": 895.672844515078} {"canonical_length": 243, "canonical_nll": 0.6703236897786459, "generation_rouge_l_f1": 0.32432432432432434, "generation_token_f1": 0.5057915057915058, "prompt_id": "182ccb8ddd147837", "reading_canonical_log_prob": -162.88865661621094, "reading_distances": [16, 6, 14, 12], "reading_log_probs": [-172.5432586669922, -164.405517578125, -168.04132080078125, -172.90350341796875], "reading_logprob_std": 4.038313375988071, "tokenization_path_logprob_std": 116.79461062709828, "trajectory_balance_rmse": 1756.9247297564575} {"canonical_length": 8, "canonical_nll": 1.0391247272491455, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "e6557b0f23801243", "reading_canonical_log_prob": -8.312997817993164, "reading_distances": [8, 6, 8, 9], "reading_log_probs": [-7.5451788902282715, -7.484490871429443, -7.08360481262207, -7.457517623901367], "reading_logprob_std": 0.2092976630353855, "tokenization_path_logprob_std": 29.070463283441903, "trajectory_balance_rmse": 218.07030030462087} {"canonical_length": 56, "canonical_nll": 0.6549959182739258, "generation_rouge_l_f1": 0.18930041152263377, "generation_token_f1": 0.23045267489711935, "prompt_id": "f461f97cdf966620", "reading_canonical_log_prob": -36.679771423339844, "reading_distances": [24, 22, 20, 22], "reading_log_probs": [-44.419273376464844, -44.03855514526367, -42.437416076660156, -42.46227264404297], "reading_logprob_std": 1.0388891216367855, "tokenization_path_logprob_std": 28.921368002048904, "trajectory_balance_rmse": 963.7438555449346} {"canonical_length": 319, "canonical_nll": 1.0572381572663596, "generation_rouge_l_f1": 0.28321678321678323, "generation_token_f1": 0.493006993006993, "prompt_id": "e632ea14cc0c360e", "reading_canonical_log_prob": -337.25897216796875, "reading_distances": [12, 15, 14, 15], "reading_log_probs": [-343.9011535644531, -343.17529296875, -340.5194091796875, -343.7041320800781], "reading_logprob_std": 1.567315529085253, "tokenization_path_logprob_std": 231.49381729640882, "trajectory_balance_rmse": 5870.499583610416} {"canonical_length": 123, "canonical_nll": 1.3240654177782012, "generation_rouge_l_f1": 0.24415584415584418, "generation_token_f1": 0.45194805194805193, "prompt_id": "fbb4ad27a6d06f7c", "reading_canonical_log_prob": -162.86004638671875, "reading_distances": [43, 40, 44, 45], "reading_log_probs": [-166.98526000976562, -163.89254760742188, -165.22528076171875, -164.6353759765625], "reading_logprob_std": 1.3184685316931768, "tokenization_path_logprob_std": 150.4286728217588, "trajectory_balance_rmse": 2442.781419707642} {"canonical_length": 285, "canonical_nll": 1.0170503315172696, "generation_rouge_l_f1": 0.27165354330708663, "generation_token_f1": 0.4881889763779527, "prompt_id": "db0aefd45e9cb0fd", "reading_canonical_log_prob": -289.8593444824219, "reading_distances": [23, 26, 25, 26], "reading_log_probs": [-297.361083984375, -296.9803466796875, -292.30316162109375, -292.8934326171875], "reading_logprob_std": 2.6554157669173013, "tokenization_path_logprob_std": 170.21001601591678, "trajectory_balance_rmse": 4790.556028167004} {"canonical_length": 32, "canonical_nll": 0.16632339358329773, "generation_rouge_l_f1": 0.2738095238095238, "generation_token_f1": 0.2738095238095238, "prompt_id": "5f8ab685b6a4b0df", "reading_canonical_log_prob": -5.322348594665527, "reading_distances": [27, 24, 32, 28], "reading_log_probs": [-6.46469259262085, -6.312196254730225, -5.91124963760376, -5.312600612640381], "reading_logprob_std": 0.5143928832795014, "tokenization_path_logprob_std": 95.05030709208366, "trajectory_balance_rmse": 389.6274441146211} {"canonical_length": 12, "canonical_nll": 1.154083251953125, "generation_rouge_l_f1": 0.380952380952381, "generation_token_f1": 0.380952380952381, "prompt_id": "6c7141219130f618", "reading_canonical_log_prob": -13.8489990234375, "reading_distances": [22, 20, 26, 20], "reading_log_probs": [-16.988754272460938, -12.81863784790039, -14.889547348022461, -18.16643714904785], "reading_logprob_std": 2.35953929039335, "tokenization_path_logprob_std": 27.17015261174827, "trajectory_balance_rmse": 310.17058530725893} {"canonical_length": 8, "canonical_nll": 0.10597799718379974, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1d180306e638dc8c", "reading_canonical_log_prob": -0.847823977470398, "reading_distances": [24, 26, 26, 24], "reading_log_probs": [-1.028914213180542, -0.8137321472167969, -0.9422630667686462, -0.9915755987167358], "reading_logprob_std": 0.09389095232697015, "tokenization_path_logprob_std": 17.609687058209243, "trajectory_balance_rmse": 123.86873987682145} {"canonical_length": 162, "canonical_nll": 1.2630660445601851, "generation_rouge_l_f1": 0.19387755102040816, "generation_token_f1": 0.336734693877551, "prompt_id": "042d56a64e3fa50f", "reading_canonical_log_prob": -204.61669921875, "reading_distances": [29, 27, 31, 21], "reading_log_probs": [-221.88540649414062, -217.93223571777344, -209.24545288085938, -209.56634521484375], "reading_logprob_std": 6.276321568594277, "tokenization_path_logprob_std": 100.79063888335001, "trajectory_balance_rmse": 3598.0675462320414} {"canonical_length": 278, "canonical_nll": 1.4809704238562276, "generation_rouge_l_f1": 0.2781954887218045, "generation_token_f1": 0.5488721804511277, "prompt_id": "519601f060a2cff2", "reading_canonical_log_prob": -411.70977783203125, "reading_distances": [22, 21, 30, 27], "reading_log_probs": [-410.06341552734375, -406.80389404296875, -404.9210510253906, -411.1739196777344], "reading_logprob_std": 2.8873603964362093, "tokenization_path_logprob_std": 149.79709029587585, "trajectory_balance_rmse": 6305.4853748738415} {"canonical_length": 41, "canonical_nll": 1.5049972534179688, "generation_rouge_l_f1": 0.19999999999999998, "generation_token_f1": 0.3090909090909091, "prompt_id": "acf7f177082b10c2", "reading_canonical_log_prob": -61.70488739013672, "reading_distances": [10, 11, 15, 17], "reading_log_probs": [-64.32560729980469, -77.73712921142578, -68.26397705078125, -60.506752014160156], "reading_logprob_std": 7.397997378015844, "tokenization_path_logprob_std": 145.08751289224531, "trajectory_balance_rmse": 922.4219796137077} {"canonical_length": 10, "canonical_nll": 1.2116401672363282, "generation_rouge_l_f1": 0.4761904761904762, "generation_token_f1": 0.6666666666666666, "prompt_id": "78f8a9aafe333ebf", "reading_canonical_log_prob": -12.116401672363281, "reading_distances": [23, 24, 29, 25], "reading_log_probs": [-12.458230972290039, -11.730351448059082, -12.064391136169434, -11.679327011108398], "reading_logprob_std": 0.35986921071673805, "tokenization_path_logprob_std": 47.125085262703024, "trajectory_balance_rmse": 296.5608080342059} {"canonical_length": 303, "canonical_nll": 1.127280055886448, "generation_rouge_l_f1": 0.16896551724137931, "generation_token_f1": 0.3586206896551724, "prompt_id": "c2a3bc6c2ee0dd9e", "reading_canonical_log_prob": -341.56585693359375, "reading_distances": [30, 28, 31, 32], "reading_log_probs": [-347.1539306640625, -347.3682556152344, -349.2416687011719, -348.72412109375], "reading_logprob_std": 1.0200484708299702, "tokenization_path_logprob_std": 86.80697873569284, "trajectory_balance_rmse": 6204.859372287625} {"canonical_length": 335, "canonical_nll": 1.30876610599347, "generation_rouge_l_f1": 0.21585160202360873, "generation_token_f1": 0.46880269814502534, "prompt_id": "994646dcabc2bac8", "reading_canonical_log_prob": -438.4366455078125, "reading_distances": [20, 18, 23, 20], "reading_log_probs": [-455.7921142578125, -454.152587890625, -449.3140869140625, -452.4200439453125], "reading_logprob_std": 2.7701310509030397, "tokenization_path_logprob_std": 352.5446451149835, "trajectory_balance_rmse": 7470.809080679328} {"canonical_length": 10, "canonical_nll": 1.1148740768432617, "generation_rouge_l_f1": 0.7000000000000001, "generation_token_f1": 0.7000000000000001, "prompt_id": "bb2e66b420114e36", "reading_canonical_log_prob": -11.148740768432617, "reading_distances": [25, 24, 29, 26], "reading_log_probs": [-10.212684631347656, -10.628886222839355, -9.832693099975586, -10.074382781982422], "reading_logprob_std": 0.3337359469474413, "tokenization_path_logprob_std": 18.66414389925607, "trajectory_balance_rmse": 202.51914342252866} {"canonical_length": 282, "canonical_nll": 0.9801728809978945, "generation_rouge_l_f1": 0.31739961759082214, "generation_token_f1": 0.5544933078393881, "prompt_id": "57dc4fcf5ed48b59", "reading_canonical_log_prob": -276.40875244140625, "reading_distances": [20, 17, 18, 18], "reading_log_probs": [-286.64068603515625, -287.8413391113281, -285.46063232421875, -289.89288330078125], "reading_logprob_std": 1.891479330913237, "tokenization_path_logprob_std": 226.27995633166995, "trajectory_balance_rmse": 5976.18843813417} {"canonical_length": 15, "canonical_nll": 0.7184852600097656, "generation_rouge_l_f1": 0.11627906976744186, "generation_token_f1": 0.11627906976744186, "prompt_id": "3b7a47e13ee4a0c9", "reading_canonical_log_prob": -10.777278900146484, "reading_distances": [16, 17, 24, 21], "reading_log_probs": [-12.53393840789795, -12.219380378723145, -14.09075927734375, -12.484039306640625], "reading_logprob_std": 0.8504302973667583, "tokenization_path_logprob_std": 19.808770481563513, "trajectory_balance_rmse": 227.77497015333947} {"canonical_length": 38, "canonical_nll": 1.4991009360865544, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.39285714285714285, "prompt_id": "4ae552c8baec5eb5", "reading_canonical_log_prob": -56.96583557128906, "reading_distances": [40, 48, 42, 38], "reading_log_probs": [-70.55859375, -72.91340637207031, -59.99890899658203, -65.02178192138672], "reading_logprob_std": 5.787915580133018, "tokenization_path_logprob_std": 86.18044936459599, "trajectory_balance_rmse": 792.5325729269668} {"canonical_length": 27, "canonical_nll": 0.856299082438151, "generation_rouge_l_f1": 0.7017543859649122, "generation_token_f1": 0.7719298245614034, "prompt_id": "17ccb345d2ac2714", "reading_canonical_log_prob": -23.120075225830078, "reading_distances": [33, 36, 31, 36], "reading_log_probs": [-22.828540802001953, -24.294933319091797, -23.608219146728516, -28.862375259399414], "reading_logprob_std": 2.7096220874815136, "tokenization_path_logprob_std": 80.40606182790854, "trajectory_balance_rmse": 591.0128246912046} {"canonical_length": 3, "canonical_nll": 2.505311965942383, "generation_rouge_l_f1": 0.05194805194805195, "generation_token_f1": 0.05194805194805195, "prompt_id": "5e428837a52a1f42", "reading_canonical_log_prob": -7.515935897827148, "reading_distances": [38, 45, 47, 50], "reading_log_probs": [-11.847381591796875, -16.2613525390625, -8.073710441589355, -7.550748348236084], "reading_logprob_std": 4.034955478270763, "tokenization_path_logprob_std": 27.516147232665293, "trajectory_balance_rmse": 53.66097141593659} {"canonical_length": 22, "canonical_nll": 1.1371042945168235, "generation_rouge_l_f1": 0.47619047619047616, "generation_token_f1": 0.5238095238095238, "prompt_id": "6a0a62922b0dc5d9", "reading_canonical_log_prob": -25.016294479370117, "reading_distances": [25, 21, 22, 14], "reading_log_probs": [-26.819303512573242, -27.377662658691406, -26.24802017211914, -25.71426773071289], "reading_logprob_std": 0.7180471301225565, "tokenization_path_logprob_std": 77.79090212087267, "trajectory_balance_rmse": 652.6820242443356} {"canonical_length": 138, "canonical_nll": 1.745796535326087, "generation_rouge_l_f1": 0.3268698060941828, "generation_token_f1": 0.41551246537396125, "prompt_id": "b8e84fe41fdcb18b", "reading_canonical_log_prob": -240.919921875, "reading_distances": [64, 67, 61, 50], "reading_log_probs": [-231.51397705078125, -232.172119140625, -232.64996337890625, -230.76515197753906], "reading_logprob_std": 0.8187775782231402, "tokenization_path_logprob_std": 159.86842689025346, "trajectory_balance_rmse": 2213.744561071291} {"canonical_length": 16, "canonical_nll": 0.4296753406524658, "generation_rouge_l_f1": 0.29850746268656714, "generation_token_f1": 0.35820895522388063, "prompt_id": "b63558947632bbd1", "reading_canonical_log_prob": -6.874805450439453, "reading_distances": [27, 27, 22, 28], "reading_log_probs": [-6.820241928100586, -7.3437724113464355, -6.929332733154297, -6.847342491149902], "reading_logprob_std": 0.2435235655028702, "tokenization_path_logprob_std": 53.87412555643853, "trajectory_balance_rmse": 276.29336706378524} {"canonical_length": 145, "canonical_nll": 1.3372999520137392, "generation_rouge_l_f1": 0.27469879518072293, "generation_token_f1": 0.5397590361445783, "prompt_id": "f6960da157e7277f", "reading_canonical_log_prob": -193.9084930419922, "reading_distances": [18, 13, 15, 21], "reading_log_probs": [-201.40724182128906, -198.9468536376953, -202.8739471435547, -203.3026580810547], "reading_logprob_std": 1.9658803400045795, "tokenization_path_logprob_std": 208.42777306028887, "trajectory_balance_rmse": 3192.7073661657496} {"canonical_length": 147, "canonical_nll": 1.3446472583173894, "generation_rouge_l_f1": 0.26262626262626265, "generation_token_f1": 0.42929292929292934, "prompt_id": "92be2f9693f01c35", "reading_canonical_log_prob": -197.66314697265625, "reading_distances": [70, 65, 62, 72], "reading_log_probs": [-184.170166015625, -185.53228759765625, -177.73919677734375, -196.01370239257812], "reading_logprob_std": 7.572053720635317, "tokenization_path_logprob_std": 131.83710920266236, "trajectory_balance_rmse": 2524.3866904516035} {"canonical_length": 21, "canonical_nll": 0.1376661913735526, "generation_rouge_l_f1": 0.9743589743589743, "generation_token_f1": 0.9743589743589743, "prompt_id": "45f3d6045c73150a", "reading_canonical_log_prob": -2.8909900188446045, "reading_distances": [33, 34, 22, 17], "reading_log_probs": [-2.683378219604492, -2.8109190464019775, -2.572695016860962, -3.179007053375244], "reading_logprob_std": 0.2636315499399863, "tokenization_path_logprob_std": 80.73257354323938, "trajectory_balance_rmse": 319.36593785970035} {"canonical_length": 98, "canonical_nll": 1.271170557761679, "generation_rouge_l_f1": 0.34024896265560167, "generation_token_f1": 0.4730290456431535, "prompt_id": "949e9bf0a5c54d25", "reading_canonical_log_prob": -124.57471466064453, "reading_distances": [17, 15, 17, 14], "reading_log_probs": [-123.30146026611328, -120.57134246826172, -121.33478546142578, -128.2280731201172], "reading_logprob_std": 3.4438187550924315, "tokenization_path_logprob_std": 100.74214806528197, "trajectory_balance_rmse": 2064.0738971579035} {"canonical_length": 14, "canonical_nll": 0.9594338962009975, "generation_rouge_l_f1": 0.16058394160583941, "generation_token_f1": 0.16058394160583941, "prompt_id": "5602e1c51a5b2686", "reading_canonical_log_prob": -13.432074546813965, "reading_distances": [19, 17, 19, 20], "reading_log_probs": [-9.49597454071045, -9.002147674560547, -10.897697448730469, -13.079639434814453], "reading_logprob_std": 1.8264552414847013, "tokenization_path_logprob_std": 61.18878184077786, "trajectory_balance_rmse": 336.4399435158537} {"canonical_length": 179, "canonical_nll": 1.409801206109244, "generation_rouge_l_f1": 0.26760563380281693, "generation_token_f1": 0.5023474178403755, "prompt_id": "94a9111b40eefe1b", "reading_canonical_log_prob": -252.3544158935547, "reading_distances": [37, 36, 37, 38], "reading_log_probs": [-267.12164306640625, -256.92724609375, -263.60198974609375, -260.101318359375], "reading_logprob_std": 4.401517596741942, "tokenization_path_logprob_std": 85.41588166380131, "trajectory_balance_rmse": 3908.2376259555685} {"canonical_length": 137, "canonical_nll": 1.7513145947978443, "generation_rouge_l_f1": 0.22110552763819097, "generation_token_f1": 0.4321608040201005, "prompt_id": "726f150c14796002", "reading_canonical_log_prob": -239.9300994873047, "reading_distances": [17, 16, 13, 22], "reading_log_probs": [-246.96458435058594, -246.96420288085938, -240.56597900390625, -243.98883056640625], "reading_logprob_std": 3.0455333840488525, "tokenization_path_logprob_std": 232.49044879453666, "trajectory_balance_rmse": 3483.0602442673467} {"canonical_length": 57, "canonical_nll": 0.3553804765667832, "generation_rouge_l_f1": 0.38043478260869573, "generation_token_f1": 0.47826086956521735, "prompt_id": "39559806a2fa846b", "reading_canonical_log_prob": -20.25668716430664, "reading_distances": [27, 21, 21, 22], "reading_log_probs": [-31.779691696166992, -27.88542366027832, -33.77168655395508, -29.206787109375], "reading_logprob_std": 2.6297275558941102, "tokenization_path_logprob_std": 23.156249421026537, "trajectory_balance_rmse": 640.4581286368126} {"canonical_length": 13, "canonical_nll": 1.0278352590707631, "generation_rouge_l_f1": 0.588235294117647, "generation_token_f1": 0.8235294117647058, "prompt_id": "3dc6a3c96d3d9987", "reading_canonical_log_prob": -13.361858367919922, "reading_distances": [24, 34, 30, 30], "reading_log_probs": [-16.345808029174805, -17.894086837768555, -17.386714935302734, -21.061677932739258], "reading_logprob_std": 2.0313479930944753, "tokenization_path_logprob_std": 27.247719293153494, "trajectory_balance_rmse": 219.49304209341636} {"canonical_length": 14, "canonical_nll": 0.9231204305376325, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.33333333333333337, "prompt_id": "8e74067a85468916", "reading_canonical_log_prob": -12.923686027526855, "reading_distances": [17, 17, 17, 23], "reading_log_probs": [-12.113627433776855, -13.030538558959961, -12.67210578918457, -12.003637313842773], "reading_logprob_std": 0.4825774749219413, "tokenization_path_logprob_std": 68.52636245285306, "trajectory_balance_rmse": 343.66191871955573} {"canonical_length": 27, "canonical_nll": 1.4952614395706743, "generation_rouge_l_f1": 0.11347517730496455, "generation_token_f1": 0.1702127659574468, "prompt_id": "3090724d0b15fbad", "reading_canonical_log_prob": -40.3720588684082, "reading_distances": [25, 17, 19, 26], "reading_log_probs": [-47.606475830078125, -40.42924499511719, -43.51251983642578, -41.807403564453125], "reading_logprob_std": 3.1120154719040687, "tokenization_path_logprob_std": 44.05475325585706, "trajectory_balance_rmse": 492.7749594175128} {"canonical_length": 273, "canonical_nll": 1.6054989322201236, "generation_rouge_l_f1": 0.23048327137546468, "generation_token_f1": 0.47583643122676583, "prompt_id": "d410f93c21deda57", "reading_canonical_log_prob": -438.30120849609375, "reading_distances": [22, 17, 18, 13], "reading_log_probs": [-435.5776062011719, -438.5889892578125, -435.64361572265625, -433.17584228515625], "reading_logprob_std": 2.2156365945071417, "tokenization_path_logprob_std": 430.91354323600683, "trajectory_balance_rmse": 6696.103587871833} {"canonical_length": 354, "canonical_nll": 1.2453755524198888, "generation_rouge_l_f1": 0.21658206429780033, "generation_token_f1": 0.4365482233502538, "prompt_id": "710012fa888db2a9", "reading_canonical_log_prob": -440.8629455566406, "reading_distances": [33, 26, 36, 26], "reading_log_probs": [-451.6876220703125, -443.8354797363281, -445.83184814453125, -443.7415771484375], "reading_logprob_std": 3.735520877979618, "tokenization_path_logprob_std": 283.3437207878894, "trajectory_balance_rmse": 7206.305866548486} {"canonical_length": 191, "canonical_nll": 1.4548544359456805, "generation_rouge_l_f1": 0.26106194690265483, "generation_token_f1": 0.4734513274336283, "prompt_id": "b61ed23f42bef95d", "reading_canonical_log_prob": -277.877197265625, "reading_distances": [20, 19, 23, 25], "reading_log_probs": [-279.60882568359375, -280.8762512207031, -279.8554992675781, -279.6552734375], "reading_logprob_std": 0.5945715840635601, "tokenization_path_logprob_std": 94.44786331754727, "trajectory_balance_rmse": 4490.060071560019} {"canonical_length": 255, "canonical_nll": 1.1079786113664216, "generation_rouge_l_f1": 0.25390625, "generation_token_f1": 0.5859375000000001, "prompt_id": "0cd98f075a94d815", "reading_canonical_log_prob": -282.5345458984375, "reading_distances": [17, 29, 27, 20], "reading_log_probs": [-285.99951171875, -298.16485595703125, -286.83697509765625, -286.243896484375], "reading_logprob_std": 5.912829067763025, "tokenization_path_logprob_std": 315.8546788928873, "trajectory_balance_rmse": 5370.686527328258} {"canonical_length": 315, "canonical_nll": 1.3602494497147817, "generation_rouge_l_f1": 0.1745200698080279, "generation_token_f1": 0.3560209424083769, "prompt_id": "3f6acc98eba7e06b", "reading_canonical_log_prob": -428.47857666015625, "reading_distances": [22, 23, 25, 22], "reading_log_probs": [-432.89910888671875, -434.66448974609375, -431.17291259765625, -435.8438720703125], "reading_logprob_std": 2.044665202012253, "tokenization_path_logprob_std": 121.50601268929779, "trajectory_balance_rmse": 7685.954587945636} {"canonical_length": 44, "canonical_nll": 1.3168940977616743, "generation_rouge_l_f1": 0.3440860215053763, "generation_token_f1": 0.4301075268817204, "prompt_id": "43a708261a941980", "reading_canonical_log_prob": -57.94334030151367, "reading_distances": [15, 12, 13, 12], "reading_log_probs": [-57.033607482910156, -58.91949462890625, -58.943748474121094, -57.88151550292969], "reading_logprob_std": 0.9188074129094344, "tokenization_path_logprob_std": 104.38805274204742, "trajectory_balance_rmse": 933.2594363748539} {"canonical_length": 186, "canonical_nll": 1.5509515577746975, "generation_rouge_l_f1": 0.2100456621004566, "generation_token_f1": 0.5114155251141553, "prompt_id": "6a69846ab5b6aa8a", "reading_canonical_log_prob": -288.47698974609375, "reading_distances": [19, 25, 20, 24], "reading_log_probs": [-287.3973693847656, -289.4256591796875, -284.9673767089844, -305.08258056640625], "reading_logprob_std": 9.094056733932842, "tokenization_path_logprob_std": 210.6326394460599, "trajectory_balance_rmse": 4073.6291040705914} {"canonical_length": 99, "canonical_nll": 1.2388084488685684, "generation_rouge_l_f1": 0.26484018264840187, "generation_token_f1": 0.410958904109589, "prompt_id": "692adc91823ed2ff", "reading_canonical_log_prob": -122.64203643798828, "reading_distances": [29, 20, 22, 31], "reading_log_probs": [-119.27561950683594, -118.3108901977539, -115.70274353027344, -118.06409454345703], "reading_logprob_std": 1.5166644833900607, "tokenization_path_logprob_std": 113.67205490192016, "trajectory_balance_rmse": 2381.156418396268} {"canonical_length": 11, "canonical_nll": 1.6587441184303977, "generation_rouge_l_f1": 0.0909090909090909, "generation_token_f1": 0.0909090909090909, "prompt_id": "9477d8c348a14e9c", "reading_canonical_log_prob": -18.246185302734375, "reading_distances": [24, 27, 24, 25], "reading_log_probs": [-19.496801376342773, -21.596302032470703, -19.608230590820312, -20.492523193359375], "reading_logprob_std": 0.9731611121102821, "tokenization_path_logprob_std": 47.769378907304265, "trajectory_balance_rmse": 221.22147494508522} {"canonical_length": 53, "canonical_nll": 0.640685387377469, "generation_rouge_l_f1": 0.5957446808510639, "generation_token_f1": 0.723404255319149, "prompt_id": "7e54ccf509187c23", "reading_canonical_log_prob": -33.95632553100586, "reading_distances": [14, 14, 14, 14], "reading_log_probs": [-35.41782760620117, -34.769676208496094, -31.876625061035156, -34.44838333129883], "reading_logprob_std": 1.5542134280135238, "tokenization_path_logprob_std": 77.59231806817228, "trajectory_balance_rmse": 826.046965719755} {"canonical_length": 65, "canonical_nll": 1.2143444354717547, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "ba0c8c0d4edf2258", "reading_canonical_log_prob": -78.93238830566406, "reading_distances": [25, 25, 22, 20], "reading_log_probs": [-77.94743347167969, -80.77973937988281, -79.10740661621094, -81.33875274658203], "reading_logprob_std": 1.5534005506232949, "tokenization_path_logprob_std": 111.27756109189218, "trajectory_balance_rmse": 1587.2835698736212} {"canonical_length": 181, "canonical_nll": 1.5919122010963398, "generation_rouge_l_f1": 0.27951807228915665, "generation_token_f1": 0.4578313253012048, "prompt_id": "1f441ed2a6f2f793", "reading_canonical_log_prob": -288.1361083984375, "reading_distances": [21, 19, 24, 17], "reading_log_probs": [-295.64727783203125, -297.5779724121094, -300.8790588378906, -292.525390625], "reading_logprob_std": 3.500652566608179, "tokenization_path_logprob_std": 187.35117220439938, "trajectory_balance_rmse": 3462.1656347551902} {"canonical_length": 234, "canonical_nll": 1.466380844768296, "generation_rouge_l_f1": 0.22997946611909648, "generation_token_f1": 0.5420944558521561, "prompt_id": "a1958dd73d8cbad2", "reading_canonical_log_prob": -343.13311767578125, "reading_distances": [16, 12, 15, 15], "reading_log_probs": [-344.1114501953125, -342.9207458496094, -344.26361083984375, -342.5207824707031], "reading_logprob_std": 0.864670521136957, "tokenization_path_logprob_std": 126.56713497261758, "trajectory_balance_rmse": 5314.733756176333} {"canonical_length": 248, "canonical_nll": 0.890288322202621, "generation_rouge_l_f1": 0.3524416135881104, "generation_token_f1": 0.5987261146496814, "prompt_id": "3e4f398cc2a68a1a", "reading_canonical_log_prob": -220.79150390625, "reading_distances": [14, 16, 12, 12], "reading_log_probs": [-228.60548400878906, -227.04983520507812, -224.90396118164062, -229.1787109375], "reading_logprob_std": 1.9118114498202574, "tokenization_path_logprob_std": 277.19225737579467, "trajectory_balance_rmse": 4630.486650982881} {"canonical_length": 7, "canonical_nll": 0.6269586426871163, "generation_rouge_l_f1": 0.07608695652173914, "generation_token_f1": 0.07608695652173914, "prompt_id": "96fa3fcdd459d2ed", "reading_canonical_log_prob": -4.3887104988098145, "reading_distances": [14, 15, 19, 19], "reading_log_probs": [-3.2793874740600586, -3.769725799560547, -3.3263444900512695, -3.2814202308654785], "reading_logprob_std": 0.23799304436786584, "tokenization_path_logprob_std": 8.765106230398208, "trajectory_balance_rmse": 101.48220071541262} {"canonical_length": 17, "canonical_nll": 0.31364718605490294, "generation_rouge_l_f1": 0.45454545454545453, "generation_token_f1": 0.45454545454545453, "prompt_id": "27d776f87907a327", "reading_canonical_log_prob": -5.33200216293335, "reading_distances": [23, 20, 24, 22], "reading_log_probs": [-7.417811393737793, -6.979030609130859, -9.153746604919434, -6.368009090423584], "reading_logprob_std": 1.1962146883933464, "tokenization_path_logprob_std": 64.70745389165408, "trajectory_balance_rmse": 299.83605525596045} {"canonical_length": 37, "canonical_nll": 0.8774164560678843, "generation_rouge_l_f1": 0.149812734082397, "generation_token_f1": 0.20224719101123595, "prompt_id": "38bff471c2090786", "reading_canonical_log_prob": -32.46440887451172, "reading_distances": [40, 34, 37, 46], "reading_log_probs": [-30.49312973022461, -32.4571533203125, -28.3737850189209, -27.050655364990234], "reading_logprob_std": 2.3779267813558516, "tokenization_path_logprob_std": 31.38256055154502, "trajectory_balance_rmse": 490.9326284011052} {"canonical_length": 60, "canonical_nll": 1.2994707743326823, "generation_rouge_l_f1": 0.3246753246753247, "generation_token_f1": 0.4025974025974026, "prompt_id": "68df2af9448f2f1d", "reading_canonical_log_prob": -77.96824645996094, "reading_distances": [31, 30, 28, 24], "reading_log_probs": [-76.50618743896484, -74.06370544433594, -75.89927673339844, -78.49948120117188], "reading_logprob_std": 1.8283364642360898, "tokenization_path_logprob_std": 117.51170374306733, "trajectory_balance_rmse": 1595.682097021364} {"canonical_length": 14, "canonical_nll": 0.6282217161996024, "generation_rouge_l_f1": 0.14285714285714288, "generation_token_f1": 0.14285714285714288, "prompt_id": "6da3ca8bcb458aa8", "reading_canonical_log_prob": -8.795104026794434, "reading_distances": [15, 21, 15, 19], "reading_log_probs": [-9.578786849975586, -10.047918319702148, -9.144937515258789, -11.768998146057129], "reading_logprob_std": 1.1499463537701173, "tokenization_path_logprob_std": 39.641993699459235, "trajectory_balance_rmse": 267.2225801990337} {"canonical_length": 29, "canonical_nll": 0.5809977301235857, "generation_rouge_l_f1": 0.1923076923076923, "generation_token_f1": 0.1923076923076923, "prompt_id": "eba9c966fdb6009a", "reading_canonical_log_prob": -16.848934173583984, "reading_distances": [35, 30, 32, 34], "reading_log_probs": [-20.037235260009766, -16.796438217163086, -19.120834350585938, -19.039457321166992], "reading_logprob_std": 1.3777605991206823, "tokenization_path_logprob_std": 122.49702965997108, "trajectory_balance_rmse": 550.7287742729818} {"canonical_length": 237, "canonical_nll": 1.2968286441851267, "generation_rouge_l_f1": 0.2738589211618257, "generation_token_f1": 0.5560165975103734, "prompt_id": "16c0d54d78758bbb", "reading_canonical_log_prob": -307.348388671875, "reading_distances": [20, 34, 21, 33], "reading_log_probs": [-315.42950439453125, -316.9176330566406, -315.67279052734375, -316.8008117675781], "reading_logprob_std": 0.7632120185727873, "tokenization_path_logprob_std": 179.5133710587393, "trajectory_balance_rmse": 5067.598818423994} {"canonical_length": 40, "canonical_nll": 1.2161100387573243, "generation_rouge_l_f1": 0.425531914893617, "generation_token_f1": 0.4680851063829787, "prompt_id": "51ab3680c9c02305", "reading_canonical_log_prob": -48.64440155029297, "reading_distances": [23, 18, 18, 19], "reading_log_probs": [-56.0472412109375, -50.67694091796875, -64.78913879394531, -49.52558517456055], "reading_logprob_std": 6.959697754905862, "tokenization_path_logprob_std": 80.41002836415689, "trajectory_balance_rmse": 691.0265406837113} {"canonical_length": 133, "canonical_nll": 1.225605441215343, "generation_rouge_l_f1": 0.2606516290726817, "generation_token_f1": 0.38095238095238093, "prompt_id": "a142f5697eaf1704", "reading_canonical_log_prob": -163.00552368164062, "reading_distances": [23, 19, 18, 26], "reading_log_probs": [-174.96405029296875, -169.81764221191406, -169.33978271484375, -180.23294067382812], "reading_logprob_std": 5.10919750581961, "tokenization_path_logprob_std": 243.76942912816003, "trajectory_balance_rmse": 3106.641136025406} {"canonical_length": 248, "canonical_nll": 1.138247336110761, "generation_rouge_l_f1": 0.32380952380952377, "generation_token_f1": 0.5295238095238095, "prompt_id": "879233d0df49ade7", "reading_canonical_log_prob": -282.28533935546875, "reading_distances": [21, 27, 28, 22], "reading_log_probs": [-286.36273193359375, -287.52374267578125, -285.6839904785156, -286.9559326171875], "reading_logprob_std": 0.7898025160045573, "tokenization_path_logprob_std": 128.03940277881466, "trajectory_balance_rmse": 4559.763350584863} {"canonical_length": 189, "canonical_nll": 1.1710523171399636, "generation_rouge_l_f1": 0.3062200956937799, "generation_token_f1": 0.5119617224880382, "prompt_id": "f6b26f3fe5bcfb2a", "reading_canonical_log_prob": -221.32888793945312, "reading_distances": [8, 13, 18, 16], "reading_log_probs": [-222.13876342773438, -219.40664672851562, -218.7164306640625, -219.28477478027344], "reading_logprob_std": 1.5312397054107536, "tokenization_path_logprob_std": 81.4623396700573, "trajectory_balance_rmse": 2897.304135014547} {"canonical_length": 12, "canonical_nll": 1.1011126041412354, "generation_rouge_l_f1": 0.10191082802547771, "generation_token_f1": 0.10191082802547771, "prompt_id": "61b8f081f2257b08", "reading_canonical_log_prob": -13.213351249694824, "reading_distances": [43, 37, 38, 42], "reading_log_probs": [-19.68390464782715, -21.882312774658203, -20.908430099487305, -20.805997848510742], "reading_logprob_std": 0.8994881728047092, "tokenization_path_logprob_std": 21.400342124441956, "trajectory_balance_rmse": 236.37045479843493} {"canonical_length": 7, "canonical_nll": 0.19470012187957764, "generation_rouge_l_f1": 0.7272727272727272, "generation_token_f1": 0.7272727272727272, "prompt_id": "756f69ca5eb428f4", "reading_canonical_log_prob": -1.3629008531570435, "reading_distances": [39, 46, 37, 36], "reading_log_probs": [-1.9283130168914795, -1.7911834716796875, -1.5821013450622559, -1.8834872245788574], "reading_logprob_std": 0.15377070385787825, "tokenization_path_logprob_std": 15.133383740551139, "trajectory_balance_rmse": 173.98073003528575} {"canonical_length": 59, "canonical_nll": 1.1552601183875133, "generation_rouge_l_f1": 0.2764227642276423, "generation_token_f1": 0.45528455284552843, "prompt_id": "f3fd9530a70f7802", "reading_canonical_log_prob": -68.16034698486328, "reading_distances": [25, 30, 31, 37], "reading_log_probs": [-82.51309204101562, -83.83744812011719, -89.70943450927734, -89.42204284667969], "reading_logprob_std": 3.7307883613030213, "tokenization_path_logprob_std": 102.04202735891681, "trajectory_balance_rmse": 1503.353548140588} {"canonical_length": 84, "canonical_nll": 0.8607978820800781, "generation_rouge_l_f1": 0.2934362934362934, "generation_token_f1": 0.4092664092664092, "prompt_id": "d431ad8fbed7b76c", "reading_canonical_log_prob": -72.30702209472656, "reading_distances": [27, 28, 21, 22], "reading_log_probs": [-71.22535705566406, -75.06098937988281, -73.74563598632812, -74.21499633789062], "reading_logprob_std": 1.6499490438956739, "tokenization_path_logprob_std": 68.36183836993487, "trajectory_balance_rmse": 1306.1702731592634} {"canonical_length": 376, "canonical_nll": 1.4556076374459774, "generation_rouge_l_f1": 0.20408163265306123, "generation_token_f1": 0.3673469387755102, "prompt_id": "455b2b67c74aee94", "reading_canonical_log_prob": -547.3084716796875, "reading_distances": [27, 19, 18, 24], "reading_log_probs": [-559.505126953125, -556.0010986328125, -557.4556884765625, -558.07275390625], "reading_logprob_std": 1.4525397875846724, "tokenization_path_logprob_std": 177.74161333440142, "trajectory_balance_rmse": 6939.889895449669} {"canonical_length": 307, "canonical_nll": 1.7439274989820848, "generation_rouge_l_f1": 0.20921985815602837, "generation_token_f1": 0.4042553191489362, "prompt_id": "b3e34a08eb2e7c34", "reading_canonical_log_prob": -535.3857421875, "reading_distances": [25, 26, 28, 27], "reading_log_probs": [-561.8927001953125, -563.3511352539062, -566.1680297851562, -560.5552368164062], "reading_logprob_std": 2.4057166637493346, "tokenization_path_logprob_std": 201.79597234933897, "trajectory_balance_rmse": 6426.924189161003} {"canonical_length": 165, "canonical_nll": 1.4203015876538825, "generation_rouge_l_f1": 0.24413145539906103, "generation_token_f1": 0.4835680751173709, "prompt_id": "7461cdb891fcaab7", "reading_canonical_log_prob": -234.34976196289062, "reading_distances": [18, 18, 17, 23], "reading_log_probs": [-239.6746368408203, -236.65142822265625, -241.35379028320312, -239.11749267578125], "reading_logprob_std": 1.9464597366553023, "tokenization_path_logprob_std": 199.02567010557547, "trajectory_balance_rmse": 3532.146927237003} {"canonical_length": 119, "canonical_nll": 1.283313398601628, "generation_rouge_l_f1": 0.19289340101522842, "generation_token_f1": 0.2893401015228426, "prompt_id": "eb07969998e95372", "reading_canonical_log_prob": -152.71429443359375, "reading_distances": [18, 19, 18, 17], "reading_log_probs": [-160.9700164794922, -169.12814331054688, -165.8467254638672, -173.29830932617188], "reading_logprob_std": 5.21223095358962, "tokenization_path_logprob_std": 85.07596910554503, "trajectory_balance_rmse": 1886.6419737988472} {"canonical_length": 134, "canonical_nll": 1.4986296696449393, "generation_rouge_l_f1": 0.23661971830985915, "generation_token_f1": 0.4112676056338028, "prompt_id": "b32326c583d0044a", "reading_canonical_log_prob": -200.81637573242188, "reading_distances": [23, 33, 27, 21], "reading_log_probs": [-197.71519470214844, -198.18003845214844, -195.93975830078125, -199.56057739257812], "reading_logprob_std": 1.4946784832510664, "tokenization_path_logprob_std": 121.03277095069168, "trajectory_balance_rmse": 2612.4408462108736} {"canonical_length": 324, "canonical_nll": 1.1645012372805748, "generation_rouge_l_f1": 0.28477905073649756, "generation_token_f1": 0.5171849427168577, "prompt_id": "e86ca4a6dca6aa44", "reading_canonical_log_prob": -377.29840087890625, "reading_distances": [38, 31, 34, 32], "reading_log_probs": [-375.0494384765625, -380.67724609375, -371.57208251953125, -376.5843200683594], "reading_logprob_std": 3.7737984504706232, "tokenization_path_logprob_std": 330.89774116995545, "trajectory_balance_rmse": 6227.517607251415} {"canonical_length": 17, "canonical_nll": 0.11626364203060374, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "39176769401db5cc", "reading_canonical_log_prob": -1.9764819145202637, "reading_distances": [14, 22, 19, 18], "reading_log_probs": [-2.938516616821289, -3.7822349071502686, -2.7345898151397705, -3.6927077770233154], "reading_logprob_std": 0.5280322255580082, "tokenization_path_logprob_std": 88.06406560825938, "trajectory_balance_rmse": 341.9762424116767} {"canonical_length": 46, "canonical_nll": 1.5213901685631794, "generation_rouge_l_f1": 0.10397553516819572, "generation_token_f1": 0.19571865443425074, "prompt_id": "6614284ebae06e44", "reading_canonical_log_prob": -69.98394775390625, "reading_distances": [15, 22, 21, 18], "reading_log_probs": [-74.18711853027344, -73.72288513183594, -71.05695343017578, -73.67460632324219], "reading_logprob_std": 1.421200909861827, "tokenization_path_logprob_std": 165.30292600772464, "trajectory_balance_rmse": 937.9199062550857} {"canonical_length": 169, "canonical_nll": 2.0954524835891273, "generation_rouge_l_f1": 0.21639344262295082, "generation_token_f1": 0.4852459016393442, "prompt_id": "130451267d979653", "reading_canonical_log_prob": -354.1314697265625, "reading_distances": [29, 24, 16, 21], "reading_log_probs": [-347.56439208984375, -346.98968505859375, -348.599609375, -348.7873840332031], "reading_logprob_std": 0.8542301369706836, "tokenization_path_logprob_std": 109.64655888042887, "trajectory_balance_rmse": 3587.3801250068072} {"canonical_length": 89, "canonical_nll": 1.7636701605293188, "generation_rouge_l_f1": 0.239202657807309, "generation_token_f1": 0.37209302325581395, "prompt_id": "13cc5e7d950f14d1", "reading_canonical_log_prob": -156.96664428710938, "reading_distances": [24, 22, 20, 26], "reading_log_probs": [-154.96253967285156, -156.1913299560547, -155.98788452148438, -157.0155029296875], "reading_logprob_std": 0.8442238289706735, "tokenization_path_logprob_std": 91.8148584940149, "trajectory_balance_rmse": 1955.9643079192554} {"canonical_length": 50, "canonical_nll": 1.154826431274414, "generation_rouge_l_f1": 0.3098591549295775, "generation_token_f1": 0.3661971830985915, "prompt_id": "aa5825f0f273154e", "reading_canonical_log_prob": -57.7413215637207, "reading_distances": [35, 32, 29, 29], "reading_log_probs": [-72.32946014404297, -69.01251220703125, -70.42032623291016, -73.20891571044922], "reading_logprob_std": 1.8883050594375206, "tokenization_path_logprob_std": 36.88875843904142, "trajectory_balance_rmse": 1327.7583756497531} {"canonical_length": 217, "canonical_nll": 1.479395993843606, "generation_rouge_l_f1": 0.30734966592427615, "generation_token_f1": 0.5567928730512249, "prompt_id": "046e2cee68c73a19", "reading_canonical_log_prob": -321.0289306640625, "reading_distances": [23, 31, 24, 29], "reading_log_probs": [-326.2210693359375, -337.37835693359375, -331.7938537597656, -332.55670166015625], "reading_logprob_std": 4.5707238842483076, "tokenization_path_logprob_std": 156.38197951720622, "trajectory_balance_rmse": 4200.689110563975} {"canonical_length": 114, "canonical_nll": 1.9337270636307566, "generation_rouge_l_f1": 0.16909620991253643, "generation_token_f1": 0.3556851311953353, "prompt_id": "f13ae64aa84fa6f5", "reading_canonical_log_prob": -220.44488525390625, "reading_distances": [25, 29, 25, 26], "reading_log_probs": [-227.1532440185547, -227.4529266357422, -225.3116455078125, -219.97048950195312], "reading_logprob_std": 3.4661837203716486, "tokenization_path_logprob_std": 174.78422144430516, "trajectory_balance_rmse": 2254.9700114823913} {"canonical_length": 180, "canonical_nll": 1.3794513278537326, "generation_rouge_l_f1": 0.23214285714285715, "generation_token_f1": 0.41517857142857145, "prompt_id": "55942a6d7a51de21", "reading_canonical_log_prob": -248.30123901367188, "reading_distances": [17, 19, 19, 15], "reading_log_probs": [-254.63113403320312, -253.59356689453125, -255.60586547851562, -255.72027587890625], "reading_logprob_std": 0.9915561571270499, "tokenization_path_logprob_std": 262.6903288737938, "trajectory_balance_rmse": 4079.472556499115} {"canonical_length": 158, "canonical_nll": 1.2223292483559138, "generation_rouge_l_f1": 0.2405345211581292, "generation_token_f1": 0.32962138084632514, "prompt_id": "ab7cc228a99beb55", "reading_canonical_log_prob": -193.12802124023438, "reading_distances": [34, 32, 33, 32], "reading_log_probs": [-186.8361053466797, -194.42037963867188, -194.86764526367188, -187.9633026123047], "reading_logprob_std": 4.211702559152678, "tokenization_path_logprob_std": 66.64876492462022, "trajectory_balance_rmse": 3108.9495134394165} {"canonical_length": 11, "canonical_nll": 0.5070665532892401, "generation_rouge_l_f1": 0.8000000000000002, "generation_token_f1": 0.9, "prompt_id": "c2611a4dd26cfbb1", "reading_canonical_log_prob": -5.577732086181641, "reading_distances": [14, 11, 14, 16], "reading_log_probs": [-6.798861503601074, -5.7834930419921875, -6.498479843139648, -6.2131876945495605], "reading_logprob_std": 0.4321893195964683, "tokenization_path_logprob_std": 28.73158311592937, "trajectory_balance_rmse": 160.22270904249208} {"canonical_length": 138, "canonical_nll": 1.3653955874235735, "generation_rouge_l_f1": 0.33438485804416407, "generation_token_f1": 0.5615141955835963, "prompt_id": "6f0ecfb4c606f967", "reading_canonical_log_prob": -188.42459106445312, "reading_distances": [28, 22, 23, 24], "reading_log_probs": [-192.4261474609375, -192.43560791015625, -187.49356079101562, -186.93753051757812], "reading_logprob_std": 3.0196201625768238, "tokenization_path_logprob_std": 123.06186761220552, "trajectory_balance_rmse": 2978.9984013410285} {"canonical_length": 202, "canonical_nll": 0.6479805672522818, "generation_rouge_l_f1": 0.5744234800838575, "generation_token_f1": 0.7169811320754716, "prompt_id": "5d860ca9ad882bc0", "reading_canonical_log_prob": -130.89207458496094, "reading_distances": [13, 10, 13, 6], "reading_log_probs": [-144.8391571044922, -141.45733642578125, -142.15850830078125, -139.0645751953125], "reading_logprob_std": 2.376232236529431, "tokenization_path_logprob_std": 194.6367171227036, "trajectory_balance_rmse": 3612.330124720921} {"canonical_length": 253, "canonical_nll": 1.6184437868623394, "generation_rouge_l_f1": 0.26883910386965376, "generation_token_f1": 0.47250509164969445, "prompt_id": "bef65ba5cd28353b", "reading_canonical_log_prob": -409.4662780761719, "reading_distances": [38, 36, 31, 36], "reading_log_probs": [-415.39666748046875, -406.81121826171875, -414.973876953125, -409.8282165527344], "reading_logprob_std": 4.154369511017505, "tokenization_path_logprob_std": 224.8890314269449, "trajectory_balance_rmse": 5157.499657894413} {"canonical_length": 49, "canonical_nll": 0.8084937115104831, "generation_rouge_l_f1": 0.2807017543859649, "generation_token_f1": 0.3508771929824562, "prompt_id": "ba40c7a321bb21e7", "reading_canonical_log_prob": -39.61619186401367, "reading_distances": [12, 16, 15, 19], "reading_log_probs": [-38.63020324707031, -36.74210739135742, -39.597660064697266, -42.583702087402344], "reading_logprob_std": 2.4379959044140964, "tokenization_path_logprob_std": 99.26334727070046, "trajectory_balance_rmse": 988.0523387521625} {"canonical_length": 15, "canonical_nll": 0.6189289093017578, "generation_rouge_l_f1": 0.4827586206896553, "generation_token_f1": 0.5172413793103449, "prompt_id": "a165c584ac95ad42", "reading_canonical_log_prob": -9.283933639526367, "reading_distances": [26, 34, 24, 22], "reading_log_probs": [-12.585525512695312, -19.986915588378906, -16.75680160522461, -12.745680809020996], "reading_logprob_std": 3.5492249920130905, "tokenization_path_logprob_std": 33.772010256054095, "trajectory_balance_rmse": 345.27763094589676} {"canonical_length": 242, "canonical_nll": 1.6346228733535642, "generation_rouge_l_f1": 0.25609756097560976, "generation_token_f1": 0.5325203252032521, "prompt_id": "9c4dbe029e6ba3bb", "reading_canonical_log_prob": -395.5787353515625, "reading_distances": [16, 16, 19, 19], "reading_log_probs": [-399.7325744628906, -396.72467041015625, -397.0830078125, -396.3992004394531], "reading_logprob_std": 1.524275825997032, "tokenization_path_logprob_std": 211.65422672664107, "trajectory_balance_rmse": 4958.078712984057} {"canonical_length": 62, "canonical_nll": 0.6147270202636719, "generation_rouge_l_f1": 0.7037037037037038, "generation_token_f1": 0.7222222222222222, "prompt_id": "dcfe7ed6b29a3c6d", "reading_canonical_log_prob": -38.113075256347656, "reading_distances": [24, 17, 20, 19], "reading_log_probs": [-48.99702453613281, -44.813880920410156, -45.880760192871094, -42.563018798828125], "reading_logprob_std": 2.6742326535637746, "tokenization_path_logprob_std": 93.94367519881057, "trajectory_balance_rmse": 1148.8860279177197} {"canonical_length": 148, "canonical_nll": 1.625121142413165, "generation_rouge_l_f1": 0.2375296912114014, "generation_token_f1": 0.47980997624703087, "prompt_id": "c0cc49ea5faf89df", "reading_canonical_log_prob": -240.51792907714844, "reading_distances": [28, 23, 21, 26], "reading_log_probs": [-250.95738220214844, -254.31967163085938, -250.54727172851562, -250.47317504882812], "reading_logprob_std": 1.84254499418271, "tokenization_path_logprob_std": 149.39633409507252, "trajectory_balance_rmse": 3724.7835572283975} {"canonical_length": 14, "canonical_nll": 1.5705161775861467, "generation_rouge_l_f1": 0.3870967741935484, "generation_token_f1": 0.3870967741935484, "prompt_id": "adf9198c8ba81554", "reading_canonical_log_prob": -21.987226486206055, "reading_distances": [18, 15, 10, 13], "reading_log_probs": [-22.092485427856445, -23.97096824645996, -23.281940460205078, -22.21892547607422], "reading_logprob_std": 0.8960051188256121, "tokenization_path_logprob_std": 75.5489613806014, "trajectory_balance_rmse": 354.39108646272547} {"canonical_length": 280, "canonical_nll": 1.0724979945591517, "generation_rouge_l_f1": 0.21973929236499068, "generation_token_f1": 0.435754189944134, "prompt_id": "7eceb236d5391917", "reading_canonical_log_prob": -300.2994384765625, "reading_distances": [21, 27, 22, 29], "reading_log_probs": [-309.2178955078125, -305.1175842285156, -306.0792541503906, -307.9174499511719], "reading_logprob_std": 1.837067448158179, "tokenization_path_logprob_std": 144.54265626915617, "trajectory_balance_rmse": 5551.722399796231} {"canonical_length": 7, "canonical_nll": 2.4040540967668806, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.25, "prompt_id": "8279f5e0aa9b5caf", "reading_canonical_log_prob": -16.828378677368164, "reading_distances": [22, 28, 24, 21], "reading_log_probs": [-18.284931182861328, -16.60403060913086, -17.337846755981445, -17.98064422607422], "reading_logprob_std": 0.7450797484031919, "tokenization_path_logprob_std": 11.010308327041768, "trajectory_balance_rmse": 191.73895110215184} {"canonical_length": 323, "canonical_nll": 0.7764751800442627, "generation_rouge_l_f1": 0.19272727272727275, "generation_token_f1": 0.42181818181818176, "prompt_id": "f4edb7d5c1554a64", "reading_canonical_log_prob": -250.80148315429688, "reading_distances": [21, 21, 22, 13], "reading_log_probs": [-276.32452392578125, -268.63714599609375, -274.3364562988281, -269.9615478515625], "reading_logprob_std": 3.6160712908699355, "tokenization_path_logprob_std": 54.73794542766445, "trajectory_balance_rmse": 6584.067700219827} {"canonical_length": 18, "canonical_nll": 0.8168849945068359, "generation_rouge_l_f1": 0.09375000000000001, "generation_token_f1": 0.125, "prompt_id": "39c97bd4effbfbe9", "reading_canonical_log_prob": -14.703929901123047, "reading_distances": [25, 27, 28, 26], "reading_log_probs": [-19.498743057250977, -17.84552574157715, -19.257234573364258, -17.068523406982422], "reading_logprob_std": 1.1577468207953316, "tokenization_path_logprob_std": 64.92088686809208, "trajectory_balance_rmse": 341.90588045510145} {"canonical_length": 32, "canonical_nll": 0.5553747415542603, "generation_rouge_l_f1": 0.18579234972677597, "generation_token_f1": 0.26229508196721313, "prompt_id": "e245da9a65e5ed0d", "reading_canonical_log_prob": -17.771991729736328, "reading_distances": [22, 20, 24, 17], "reading_log_probs": [-23.90508460998535, -31.00856590270996, -26.750484466552734, -32.44333267211914], "reading_logprob_std": 3.916373329599102, "tokenization_path_logprob_std": 61.20803935986338, "trajectory_balance_rmse": 373.1460137598614} {"canonical_length": 244, "canonical_nll": 1.2323634663566214, "generation_rouge_l_f1": 0.22580645161290325, "generation_token_f1": 0.45967741935483875, "prompt_id": "1dc2c6d9af2ea2a5", "reading_canonical_log_prob": -300.6966857910156, "reading_distances": [15, 15, 18, 18], "reading_log_probs": [-303.1695861816406, -302.33636474609375, -298.965576171875, -307.2049865722656], "reading_logprob_std": 3.386320425377507, "tokenization_path_logprob_std": 161.3920238764912, "trajectory_balance_rmse": 4424.31112640251} {"canonical_length": 143, "canonical_nll": 1.3716554441652098, "generation_rouge_l_f1": 0.199501246882793, "generation_token_f1": 0.344139650872818, "prompt_id": "8214cf7581d9c6a8", "reading_canonical_log_prob": -196.146728515625, "reading_distances": [13, 10, 10, 7], "reading_log_probs": [-196.27174377441406, -193.5574951171875, -198.0056610107422, -193.53538513183594], "reading_logprob_std": 2.191485578286839, "tokenization_path_logprob_std": 87.11384942379333, "trajectory_balance_rmse": 3206.9046329407083} {"canonical_length": 79, "canonical_nll": 0.6610775718206092, "generation_rouge_l_f1": 0.43, "generation_token_f1": 0.47, "prompt_id": "1669f97fd38169a6", "reading_canonical_log_prob": -52.225128173828125, "reading_distances": [15, 27, 30, 22], "reading_log_probs": [-53.493167877197266, -54.10374069213867, -54.6709098815918, -51.178611755371094], "reading_logprob_std": 1.532733174782016, "tokenization_path_logprob_std": 39.344826601812066, "trajectory_balance_rmse": 1194.7828994832735} {"canonical_length": 354, "canonical_nll": 0.9296202578787076, "generation_rouge_l_f1": 0.3012704174228675, "generation_token_f1": 0.48275862068965525, "prompt_id": "85c649ae8170e371", "reading_canonical_log_prob": -329.0855712890625, "reading_distances": [22, 24, 24, 26], "reading_log_probs": [-337.1012268066406, -333.6749572753906, -336.26007080078125, -335.5186462402344], "reading_logprob_std": 1.4601124070129672, "tokenization_path_logprob_std": 263.7041829437129, "trajectory_balance_rmse": 5335.25681073273} {"canonical_length": 364, "canonical_nll": 1.0071121048141312, "generation_rouge_l_f1": 0.2781456953642384, "generation_token_f1": 0.49668874172185434, "prompt_id": "ff47696a7202ba2a", "reading_canonical_log_prob": -366.58880615234375, "reading_distances": [34, 28, 33, 29], "reading_log_probs": [-372.5338134765625, -372.74725341796875, -371.68402099609375, -376.1024475097656], "reading_logprob_std": 1.9453606384733606, "tokenization_path_logprob_std": 125.97578315876252, "trajectory_balance_rmse": 7027.593880003428} {"canonical_length": 41, "canonical_nll": 0.9448856260718369, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "3a5ef38cf12c45a8", "reading_canonical_log_prob": -38.74031066894531, "reading_distances": [24, 17, 17, 21], "reading_log_probs": [-42.26313018798828, -43.687129974365234, -41.95326614379883, -41.83937072753906], "reading_logprob_std": 0.8532698567289192, "tokenization_path_logprob_std": 62.54446046814217, "trajectory_balance_rmse": 784.0961194060825} {"canonical_length": 230, "canonical_nll": 0.9904659105383832, "generation_rouge_l_f1": 0.35000000000000003, "generation_token_f1": 0.5541666666666667, "prompt_id": "13b43b11a6bf37ad", "reading_canonical_log_prob": -227.80715942382812, "reading_distances": [20, 19, 22, 18], "reading_log_probs": [-234.66061401367188, -235.28366088867188, -236.17501831054688, -234.5572509765625], "reading_logprob_std": 0.743396050855493, "tokenization_path_logprob_std": 326.92772787462167, "trajectory_balance_rmse": 4474.0727433608545} {"canonical_length": 111, "canonical_nll": 1.2193312086500563, "generation_rouge_l_f1": 0.24365482233502536, "generation_token_f1": 0.40609137055837563, "prompt_id": "d6b3da36a5892656", "reading_canonical_log_prob": -135.34576416015625, "reading_distances": [32, 35, 30, 40], "reading_log_probs": [-134.4136199951172, -132.95443725585938, -132.89120483398438, -134.08035278320312], "reading_logprob_std": 0.7769484207973013, "tokenization_path_logprob_std": 137.17709469554993, "trajectory_balance_rmse": 2396.5553296769435} {"canonical_length": 184, "canonical_nll": 1.322824644005817, "generation_rouge_l_f1": 0.2483069977426636, "generation_token_f1": 0.43340857787810383, "prompt_id": "f1036fae25a980bc", "reading_canonical_log_prob": -243.3997344970703, "reading_distances": [17, 8, 10, 14], "reading_log_probs": [-246.17254638671875, -244.97592163085938, -250.244384765625, -245.83975219726562], "reading_logprob_std": 2.34567377802373, "tokenization_path_logprob_std": 345.7523059142867, "trajectory_balance_rmse": 4635.645130592324} {"canonical_length": 199, "canonical_nll": 1.1154382600257144, "generation_rouge_l_f1": 0.30701754385964913, "generation_token_f1": 0.5833333333333334, "prompt_id": "272e476421ef865b", "reading_canonical_log_prob": -221.9722137451172, "reading_distances": [25, 28, 25, 30], "reading_log_probs": [-233.1898651123047, -226.70260620117188, -227.7006378173828, -231.20986938476562], "reading_logprob_std": 3.024383163874256, "tokenization_path_logprob_std": 199.8623288518021, "trajectory_balance_rmse": 3418.0364993102157} {"canonical_length": 202, "canonical_nll": 0.9804542465965347, "generation_rouge_l_f1": 0.26582278481012656, "generation_token_f1": 0.4767932489451477, "prompt_id": "548a4dcfac376298", "reading_canonical_log_prob": -198.0517578125, "reading_distances": [23, 23, 22, 25], "reading_log_probs": [-212.71978759765625, -216.91281127929688, -212.5640869140625, -216.01068115234375], "reading_logprob_std": 2.2368118103684886, "tokenization_path_logprob_std": 205.68006222139775, "trajectory_balance_rmse": 4197.027826945358} {"canonical_length": 415, "canonical_nll": 1.0004906344126505, "generation_rouge_l_f1": 0.2100456621004566, "generation_token_f1": 0.4170471841704719, "prompt_id": "8ea8da1e0ef16ea4", "reading_canonical_log_prob": -415.20361328125, "reading_distances": [32, 28, 28, 26], "reading_log_probs": [-432.6480407714844, -431.3941345214844, -434.7420349121094, -423.738037109375], "reading_logprob_std": 4.798068099553205, "tokenization_path_logprob_std": 81.14758914232164, "trajectory_balance_rmse": 3977.4036906910756} {"canonical_length": 70, "canonical_nll": 1.106312997000558, "generation_rouge_l_f1": 0.1286549707602339, "generation_token_f1": 0.2046783625730994, "prompt_id": "68d1893ed02225a9", "reading_canonical_log_prob": -77.44190979003906, "reading_distances": [12, 15, 21, 12], "reading_log_probs": [-75.17063903808594, -73.08802795410156, -70.99478149414062, -75.69754028320312], "reading_logprob_std": 2.147864358739906, "tokenization_path_logprob_std": 122.82827172973526, "trajectory_balance_rmse": 1528.119885667336} {"canonical_length": 207, "canonical_nll": 1.2219836838579408, "generation_rouge_l_f1": 0.3247863247863248, "generation_token_f1": 0.5384615384615384, "prompt_id": "06716de5ff9ed39f", "reading_canonical_log_prob": -252.95062255859375, "reading_distances": [27, 29, 30, 30], "reading_log_probs": [-260.60369873046875, -252.90362548828125, -259.0007019042969, -271.388916015625], "reading_logprob_std": 7.694856475913797, "tokenization_path_logprob_std": 286.1748683634035, "trajectory_balance_rmse": 4997.708725417616} {"canonical_length": 24, "canonical_nll": 0.5953617493311564, "generation_rouge_l_f1": 0.36363636363636365, "generation_token_f1": 0.5454545454545454, "prompt_id": "8d76cf27c40ca6eb", "reading_canonical_log_prob": -14.288681983947754, "reading_distances": [33, 27, 25, 27], "reading_log_probs": [-11.566034317016602, -12.813544273376465, -10.583664894104004, -13.987336158752441], "reading_logprob_std": 1.4809670516579279, "tokenization_path_logprob_std": 96.01257672909846, "trajectory_balance_rmse": 393.31238828975665} {"canonical_length": 37, "canonical_nll": 1.4977799492913324, "generation_rouge_l_f1": 0.1241830065359477, "generation_token_f1": 0.1764705882352941, "prompt_id": "f7669b85c591c347", "reading_canonical_log_prob": -55.4178581237793, "reading_distances": [13, 25, 26, 18], "reading_log_probs": [-57.89085388183594, -60.96092987060547, -58.924320220947266, -59.62055969238281], "reading_logprob_std": 1.2882294672427153, "tokenization_path_logprob_std": 53.48338341054953, "trajectory_balance_rmse": 808.7196001812412} {"canonical_length": 292, "canonical_nll": 1.2142578543049016, "generation_rouge_l_f1": 0.2903225806451613, "generation_token_f1": 0.5161290322580646, "prompt_id": "9875c8d91c22a365", "reading_canonical_log_prob": -354.56329345703125, "reading_distances": [29, 21, 25, 22], "reading_log_probs": [-352.0805358886719, -358.0799560546875, -351.40863037109375, -351.97064208984375], "reading_logprob_std": 3.143813192467503, "tokenization_path_logprob_std": 208.699401577166, "trajectory_balance_rmse": 6566.351369653379} {"canonical_length": 293, "canonical_nll": 1.3836567849429393, "generation_rouge_l_f1": 0.216696269982238, "generation_token_f1": 0.47602131438721135, "prompt_id": "967c6a5c34b145df", "reading_canonical_log_prob": -405.41143798828125, "reading_distances": [37, 32, 28, 37], "reading_log_probs": [-406.37786865234375, -408.0135192871094, -407.73345947265625, -407.1114501953125], "reading_logprob_std": 0.7263022235071737, "tokenization_path_logprob_std": 273.4163165635346, "trajectory_balance_rmse": 6774.899407799939} {"canonical_length": 9, "canonical_nll": 0.04739754067526923, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "abcadc4b6374d8f6", "reading_canonical_log_prob": -0.4265778660774231, "reading_distances": [43, 51, 41, 50], "reading_log_probs": [-0.9321314096450806, -2.5861546993255615, -4.904276371002197, -0.7899395227432251], "reading_logprob_std": 1.916198317758456, "tokenization_path_logprob_std": 51.46543530743161, "trajectory_balance_rmse": 180.11949337669296} {"canonical_length": 261, "canonical_nll": 1.2744061115601053, "generation_rouge_l_f1": 0.11013215859030838, "generation_token_f1": 0.2290748898678414, "prompt_id": "ba38639bb894f700", "reading_canonical_log_prob": -332.6199951171875, "reading_distances": [21, 30, 29, 25], "reading_log_probs": [-336.9472961425781, -335.7657165527344, -335.8953857421875, -333.8782043457031], "reading_logprob_std": 1.2770574083917412, "tokenization_path_logprob_std": 259.8962295685285, "trajectory_balance_rmse": 4365.410291599053} {"canonical_length": 258, "canonical_nll": 1.0998554081879846, "generation_rouge_l_f1": 0.12673267326732673, "generation_token_f1": 0.21386138613861386, "prompt_id": "49e9fe2e1aacf453", "reading_canonical_log_prob": -283.7626953125, "reading_distances": [15, 21, 17, 17], "reading_log_probs": [-286.92755126953125, -290.102294921875, -286.96917724609375, -284.7029724121094], "reading_logprob_std": 2.2198775371109805, "tokenization_path_logprob_std": 136.81241568308195, "trajectory_balance_rmse": 3978.896333527721} {"canonical_length": 255, "canonical_nll": 1.1135311351102941, "generation_rouge_l_f1": 0.23715415019762848, "generation_token_f1": 0.47430830039525695, "prompt_id": "6193ec67d5074a6a", "reading_canonical_log_prob": -283.950439453125, "reading_distances": [21, 21, 17, 15], "reading_log_probs": [-288.24530029296875, -290.72760009765625, -289.60772705078125, -287.67144775390625], "reading_logprob_std": 1.3750965430402822, "tokenization_path_logprob_std": 118.36510238731164, "trajectory_balance_rmse": 4387.598929284802} {"canonical_length": 289, "canonical_nll": 0.922762966485997, "generation_rouge_l_f1": 0.3303730017761989, "generation_token_f1": 0.5115452930728241, "prompt_id": "821b32f3ecbd67a9", "reading_canonical_log_prob": -266.6784973144531, "reading_distances": [17, 18, 17, 17], "reading_log_probs": [-276.9149169921875, -272.54327392578125, -276.6912841796875, -275.02984619140625], "reading_logprob_std": 2.017923304341071, "tokenization_path_logprob_std": 265.38859739920576, "trajectory_balance_rmse": 5608.345503511372} {"canonical_length": 22, "canonical_nll": 0.5868290987881747, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "ebadf5b1dd406fa9", "reading_canonical_log_prob": -12.910240173339844, "reading_distances": [30, 36, 37, 43], "reading_log_probs": [-13.4397611618042, -12.023569107055664, -13.32705307006836, -13.124493598937988], "reading_logprob_std": 0.6499892437618656, "tokenization_path_logprob_std": 55.786665573622216, "trajectory_balance_rmse": 459.5757548148007} {"canonical_length": 8, "canonical_nll": 1.0232824087142944, "generation_rouge_l_f1": 0.06530612244897958, "generation_token_f1": 0.06530612244897958, "prompt_id": "6edf6d71173f0e03", "reading_canonical_log_prob": -8.186259269714355, "reading_distances": [30, 32, 33, 27], "reading_log_probs": [-25.350719451904297, -19.549354553222656, -25.16889190673828, -23.183513641357422], "reading_logprob_std": 2.6943415424920505, "tokenization_path_logprob_std": 18.36218213644387, "trajectory_balance_rmse": 147.69861504013127} {"canonical_length": 16, "canonical_nll": 1.3733820915222168, "generation_rouge_l_f1": 0.6, "generation_token_f1": 0.6, "prompt_id": "938d726d90109110", "reading_canonical_log_prob": -21.97411346435547, "reading_distances": [25, 24, 32, 23], "reading_log_probs": [-23.495662689208984, -26.672348022460938, -23.158037185668945, -20.71816635131836], "reading_logprob_std": 2.4439625664245392, "tokenization_path_logprob_std": 62.615485698042114, "trajectory_balance_rmse": 455.38425535085065} {"canonical_length": 6, "canonical_nll": 0.2602604627609253, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "b4288f8c25b7da9a", "reading_canonical_log_prob": -1.5615627765655518, "reading_distances": [42, 33, 33, 37], "reading_log_probs": [-3.104525089263916, -1.5834659337997437, -2.305109739303589, -1.8872469663619995], "reading_logprob_std": 0.6596788283858469, "tokenization_path_logprob_std": 7.375947316822912, "trajectory_balance_rmse": 145.2863708202881} {"canonical_length": 266, "canonical_nll": 1.3246230505462875, "generation_rouge_l_f1": 0.2504816955684008, "generation_token_f1": 0.5394990366088632, "prompt_id": "3c7a013fe65f77bc", "reading_canonical_log_prob": -352.3497314453125, "reading_distances": [27, 25, 25, 26], "reading_log_probs": [-363.2305908203125, -364.503662109375, -362.1199951171875, -362.7124938964844], "reading_logprob_std": 1.0150436955002404, "tokenization_path_logprob_std": 100.30942717219983, "trajectory_balance_rmse": 5363.840028224754} {"canonical_length": 28, "canonical_nll": 1.6345694405691964, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "66c2a007594b76e9", "reading_canonical_log_prob": -45.7679443359375, "reading_distances": [61, 62, 51, 74], "reading_log_probs": [-49.87070083618164, -50.248558044433594, -46.52005386352539, -48.55038070678711], "reading_logprob_std": 1.6837614943160744, "tokenization_path_logprob_std": 85.74685040675718, "trajectory_balance_rmse": 543.4780383113766} {"canonical_length": 321, "canonical_nll": 1.483644289391063, "generation_rouge_l_f1": 0.23776223776223776, "generation_token_f1": 0.4335664335664336, "prompt_id": "dd27e1dd582c59a2", "reading_canonical_log_prob": -476.24981689453125, "reading_distances": [22, 25, 24, 24], "reading_log_probs": [-507.1310119628906, -505.6759948730469, -504.84722900390625, -501.3029479980469], "reading_logprob_std": 2.477757531167552, "tokenization_path_logprob_std": 152.8327549206193, "trajectory_balance_rmse": 6423.661654881058} {"canonical_length": 21, "canonical_nll": 0.3527716682070777, "generation_rouge_l_f1": 0.9047619047619048, "generation_token_f1": 0.9047619047619048, "prompt_id": "3c6903de8714c50f", "reading_canonical_log_prob": -7.408205032348633, "reading_distances": [18, 15, 16, 12], "reading_log_probs": [-6.385104656219482, -5.771248817443848, -7.729464530944824, -5.802967071533203], "reading_logprob_std": 0.916061038743184, "tokenization_path_logprob_std": 80.94907770415286, "trajectory_balance_rmse": 583.0961445055731} {"canonical_length": 45, "canonical_nll": 1.4196385701497396, "generation_rouge_l_f1": 0.044444444444444446, "generation_token_f1": 0.044444444444444446, "prompt_id": "0692a96622cd2aed", "reading_canonical_log_prob": -63.88373565673828, "reading_distances": [21, 27, 23, 25], "reading_log_probs": [-69.47220611572266, -65.36384582519531, -65.11241912841797, -68.6801986694336], "reading_logprob_std": 2.241727478810703, "tokenization_path_logprob_std": 46.39081362326951, "trajectory_balance_rmse": 1218.7966520143304} {"canonical_length": 322, "canonical_nll": 1.0158236485830745, "generation_rouge_l_f1": 0.2610229276895944, "generation_token_f1": 0.46560846560846564, "prompt_id": "e5a0acf44a48d7ab", "reading_canonical_log_prob": -327.09521484375, "reading_distances": [25, 33, 23, 26], "reading_log_probs": [-369.3889465332031, -369.9361267089844, -348.4163513183594, -341.7210693359375], "reading_logprob_std": 14.461666516014112, "tokenization_path_logprob_std": 208.25467235392384, "trajectory_balance_rmse": 5355.968427410843} {"canonical_length": 191, "canonical_nll": 0.9087136952664839, "generation_rouge_l_f1": 0.42898550724637685, "generation_token_f1": 0.6434782608695652, "prompt_id": "4e165cac00833611", "reading_canonical_log_prob": -173.56431579589844, "reading_distances": [18, 19, 18, 15], "reading_log_probs": [-172.92660522460938, -171.82333374023438, -170.16075134277344, -170.77769470214844], "reading_logprob_std": 1.2152891436621998, "tokenization_path_logprob_std": 68.34963078697216, "trajectory_balance_rmse": 2619.092808423963} {"canonical_length": 296, "canonical_nll": 1.234161376953125, "generation_rouge_l_f1": 0.1909722222222222, "generation_token_f1": 0.40277777777777773, "prompt_id": "eadf46baa67f8e23", "reading_canonical_log_prob": -365.311767578125, "reading_distances": [25, 30, 24, 31], "reading_log_probs": [-370.118896484375, -374.7658996582031, -371.1357116699219, -372.90643310546875], "reading_logprob_std": 2.044713654084003, "tokenization_path_logprob_std": 349.1629887785577, "trajectory_balance_rmse": 6830.771442277849} {"canonical_length": 97, "canonical_nll": 1.1321814625533586, "generation_rouge_l_f1": 0.2318840579710145, "generation_token_f1": 0.3188405797101449, "prompt_id": "228080ce5449beea", "reading_canonical_log_prob": -109.82160186767578, "reading_distances": [17, 17, 23, 21], "reading_log_probs": [-98.8594970703125, -98.39656829833984, -97.60993957519531, -99.05169677734375], "reading_logprob_std": 0.6415657676686654, "tokenization_path_logprob_std": 89.6491870523282, "trajectory_balance_rmse": 1083.6690170519068} {"canonical_length": 345, "canonical_nll": 1.0870218912760417, "generation_rouge_l_f1": 0.19694397283531406, "generation_token_f1": 0.400679117147708, "prompt_id": "0584d61b9b2b3717", "reading_canonical_log_prob": -375.0225524902344, "reading_distances": [39, 24, 40, 33], "reading_log_probs": [-402.4698486328125, -390.25872802734375, -404.2660217285156, -404.896240234375], "reading_logprob_std": 6.886480089289509, "tokenization_path_logprob_std": 222.6767385207419, "trajectory_balance_rmse": 6430.681308948946} {"canonical_length": 16, "canonical_nll": 1.3764069080352783, "generation_rouge_l_f1": 0.4347826086956522, "generation_token_f1": 0.5217391304347826, "prompt_id": "f605a0e57df216ae", "reading_canonical_log_prob": -22.022510528564453, "reading_distances": [34, 36, 38, 30], "reading_log_probs": [-25.89016342163086, -23.746932983398438, -26.284025192260742, -25.093814849853516], "reading_logprob_std": 1.1199056192052492, "tokenization_path_logprob_std": 32.66940965920646, "trajectory_balance_rmse": 257.75702944174833} {"canonical_length": 337, "canonical_nll": 1.1801877347227374, "generation_rouge_l_f1": 0.2296173044925125, "generation_token_f1": 0.4492512479201331, "prompt_id": "ab315c4f854a9130", "reading_canonical_log_prob": -397.7232666015625, "reading_distances": [27, 25, 31, 21], "reading_log_probs": [-411.7716064453125, -414.0267639160156, -413.48675537109375, -415.27618408203125], "reading_logprob_std": 1.4538526564049419, "tokenization_path_logprob_std": 157.4489814079576, "trajectory_balance_rmse": 7733.888214693269} {"canonical_length": 283, "canonical_nll": 1.2258957502277497, "generation_rouge_l_f1": 0.2318840579710145, "generation_token_f1": 0.47101449275362317, "prompt_id": "93f170181468154b", "reading_canonical_log_prob": -346.9284973144531, "reading_distances": [27, 25, 27, 26], "reading_log_probs": [-343.80316162109375, -342.7449645996094, -342.9481506347656, -341.43389892578125], "reading_logprob_std": 0.9796838965773034, "tokenization_path_logprob_std": 131.72230650141722, "trajectory_balance_rmse": 5978.73000146681} {"canonical_length": 32, "canonical_nll": 0.8168178796768188, "generation_rouge_l_f1": 0.34234234234234234, "generation_token_f1": 0.37837837837837834, "prompt_id": "ce3af5b67c725855", "reading_canonical_log_prob": -26.138172149658203, "reading_distances": [15, 15, 15, 17], "reading_log_probs": [-32.994903564453125, -29.355754852294922, -32.551429748535156, -31.323904037475586], "reading_logprob_std": 1.6285238823002248, "tokenization_path_logprob_std": 80.08875081668043, "trajectory_balance_rmse": 787.1808181284591} {"canonical_length": 64, "canonical_nll": 1.6132663488388062, "generation_rouge_l_f1": 0.1889763779527559, "generation_token_f1": 0.36220472440944884, "prompt_id": "5d5d479145243991", "reading_canonical_log_prob": -103.2490463256836, "reading_distances": [23, 19, 19, 21], "reading_log_probs": [-111.39912414550781, -110.14967346191406, -112.28376770019531, -107.41043090820312], "reading_logprob_std": 2.1225105897853984, "tokenization_path_logprob_std": 105.09143352982178, "trajectory_balance_rmse": 1315.4706639569572} {"canonical_length": 259, "canonical_nll": 1.7445977995294402, "generation_rouge_l_f1": 0.19659735349716448, "generation_token_f1": 0.4158790170132325, "prompt_id": "a2496cc91dc1f05e", "reading_canonical_log_prob": -451.850830078125, "reading_distances": [14, 15, 15, 16], "reading_log_probs": [-452.5496826171875, -456.4117126464844, -452.99053955078125, -451.53863525390625], "reading_logprob_std": 2.115245893231436, "tokenization_path_logprob_std": 99.78881573968228, "trajectory_balance_rmse": 6232.045477744104} {"canonical_length": 201, "canonical_nll": 1.0364182505441542, "generation_rouge_l_f1": 0.3049327354260089, "generation_token_f1": 0.4932735426008969, "prompt_id": "419c50c5b36f610d", "reading_canonical_log_prob": -208.320068359375, "reading_distances": [36, 29, 46, 37], "reading_log_probs": [-214.78277587890625, -215.07168579101562, -215.5267791748047, -222.56967163085938], "reading_logprob_std": 3.7338763490889106, "tokenization_path_logprob_std": 228.76665524438187, "trajectory_balance_rmse": 5023.317332315824} {"canonical_length": 120, "canonical_nll": 0.6275367101033529, "generation_rouge_l_f1": 0.4896265560165975, "generation_token_f1": 0.6721991701244814, "prompt_id": "07d185247528da20", "reading_canonical_log_prob": -75.30440521240234, "reading_distances": [20, 20, 15, 20], "reading_log_probs": [-75.84051513671875, -70.4469223022461, -77.79108428955078, -79.02315521240234], "reading_logprob_std": 3.7862827694120225, "tokenization_path_logprob_std": 107.48518422380803, "trajectory_balance_rmse": 1704.1921126730401} {"canonical_length": 187, "canonical_nll": 1.3765072746072862, "generation_rouge_l_f1": 0.2529832935560859, "generation_token_f1": 0.548926014319809, "prompt_id": "f75ec5897fcc8e0d", "reading_canonical_log_prob": -257.4068603515625, "reading_distances": [29, 29, 22, 32], "reading_log_probs": [-258.29608154296875, -254.7371826171875, -256.73370361328125, -255.3677520751953], "reading_logprob_std": 1.579331012500403, "tokenization_path_logprob_std": 114.90147670168204, "trajectory_balance_rmse": 3958.0340545379627} {"canonical_length": 362, "canonical_nll": 1.4650524834901588, "generation_rouge_l_f1": 0.29392971246006394, "generation_token_f1": 0.5079872204472843, "prompt_id": "cd40c2a793f29f44", "reading_canonical_log_prob": -530.3489990234375, "reading_distances": [25, 21, 17, 17], "reading_log_probs": [-530.64013671875, -532.1062622070312, -534.447998046875, -533.532470703125], "reading_logprob_std": 1.667605955001586, "tokenization_path_logprob_std": 311.59388458763743, "trajectory_balance_rmse": 7756.33037310848} {"canonical_length": 131, "canonical_nll": 1.047522391981751, "generation_rouge_l_f1": 0.29508196721311475, "generation_token_f1": 0.3934426229508196, "prompt_id": "79a9080b044df25c", "reading_canonical_log_prob": -137.22543334960938, "reading_distances": [32, 28, 28, 29], "reading_log_probs": [-145.19406127929688, -139.69644165039062, -141.55068969726562, -147.2845916748047], "reading_logprob_std": 3.4371037214469236, "tokenization_path_logprob_std": 139.6785922920995, "trajectory_balance_rmse": 2484.3516793241374} {"canonical_length": 59, "canonical_nll": 1.894548060530323, "generation_rouge_l_f1": 0.15384615384615385, "generation_token_f1": 0.23999999999999996, "prompt_id": "56223c1e8305751f", "reading_canonical_log_prob": -111.77833557128906, "reading_distances": [20, 19, 18, 21], "reading_log_probs": [-113.86820983886719, -112.5506591796875, -112.79173278808594, -112.75114440917969], "reading_logprob_std": 0.5945962203447719, "tokenization_path_logprob_std": 122.1368439452319, "trajectory_balance_rmse": 1288.7869694845383} {"canonical_length": 24, "canonical_nll": 0.7720607121785482, "generation_rouge_l_f1": 0.4329896907216495, "generation_token_f1": 0.49484536082474223, "prompt_id": "5528bdba3c5f6117", "reading_canonical_log_prob": -18.529457092285156, "reading_distances": [47, 41, 48, 44], "reading_log_probs": [-20.8382568359375, -19.522239685058594, -19.778160095214844, -19.067113876342773], "reading_logprob_std": 0.7511588329533597, "tokenization_path_logprob_std": 83.00213581242227, "trajectory_balance_rmse": 612.4646497532186} {"canonical_length": 351, "canonical_nll": 1.330199847533832, "generation_rouge_l_f1": 0.18849840255591058, "generation_token_f1": 0.37699680511182115, "prompt_id": "daf735e510f876d3", "reading_canonical_log_prob": -466.900146484375, "reading_distances": [19, 21, 16, 18], "reading_log_probs": [-485.147216796875, -488.3537902832031, -486.8211975097656, -479.28472900390625], "reading_logprob_std": 3.9670324625772944, "tokenization_path_logprob_std": 605.900600140992, "trajectory_balance_rmse": 7841.321089233095} {"canonical_length": 73, "canonical_nll": 1.499880751518354, "generation_rouge_l_f1": 0.1565217391304348, "generation_token_f1": 0.25507246376811593, "prompt_id": "d276e797ca2de2bf", "reading_canonical_log_prob": -109.49129486083984, "reading_distances": [53, 53, 48, 42], "reading_log_probs": [-113.35899353027344, -111.56916809082031, -113.16641235351562, -112.48445129394531], "reading_logprob_std": 0.8092893075554476, "tokenization_path_logprob_std": 172.36582421438183, "trajectory_balance_rmse": 1300.5807178726432} {"canonical_length": 107, "canonical_nll": 1.6569506208473277, "generation_rouge_l_f1": 0.22988505747126442, "generation_token_f1": 0.37356321839080464, "prompt_id": "60ad70d4930701a2", "reading_canonical_log_prob": -177.29371643066406, "reading_distances": [12, 13, 10, 12], "reading_log_probs": [-179.69287109375, -191.07872009277344, -183.66531372070312, -188.00527954101562], "reading_logprob_std": 4.981247553167889, "tokenization_path_logprob_std": 145.44930353072985, "trajectory_balance_rmse": 2224.3902626182307} {"canonical_length": 11, "canonical_nll": 0.11699270118366588, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.33333333333333337, "prompt_id": "91d144e62a7c6f3d", "reading_canonical_log_prob": -1.2869197130203247, "reading_distances": [35, 35, 25, 26], "reading_log_probs": [-0.7163445949554443, -0.9510237574577332, -0.5626870393753052, -0.616635799407959], "reading_logprob_std": 0.1717939607538536, "tokenization_path_logprob_std": 47.985175395972554, "trajectory_balance_rmse": 269.1648045583853} {"canonical_length": 31, "canonical_nll": 1.0728959114320817, "generation_rouge_l_f1": 0.1724137931034483, "generation_token_f1": 0.2413793103448276, "prompt_id": "663423620e0283d5", "reading_canonical_log_prob": -33.25977325439453, "reading_distances": [15, 15, 18, 18], "reading_log_probs": [-40.405921936035156, -41.13079071044922, -41.256919860839844, -43.03167724609375], "reading_logprob_std": 1.115170581754556, "tokenization_path_logprob_std": 33.39761274156904, "trajectory_balance_rmse": 431.97156931904726} {"canonical_length": 51, "canonical_nll": 0.06478690166099399, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "eece35ce9b12a6f2", "reading_canonical_log_prob": -3.3041319847106934, "reading_distances": [24, 22, 23, 23], "reading_log_probs": [-3.225281000137329, -3.190898895263672, -3.2749035358428955, -3.017089605331421], "reading_logprob_std": 0.11207239242560892, "tokenization_path_logprob_std": 44.81207453535667, "trajectory_balance_rmse": 1025.5347964517282} {"canonical_length": 156, "canonical_nll": 1.5385279533190606, "generation_rouge_l_f1": 0.20759493670886076, "generation_token_f1": 0.3848101265822784, "prompt_id": "25055edbaab1f2e6", "reading_canonical_log_prob": -240.01036071777344, "reading_distances": [9, 5, 8, 8], "reading_log_probs": [-239.73739624023438, -237.97666931152344, -238.9518585205078, -231.88914489746094], "reading_logprob_std": 3.573083975739592, "tokenization_path_logprob_std": 126.71068102800872, "trajectory_balance_rmse": 3202.2026105361033} {"canonical_length": 297, "canonical_nll": 0.9222429577349011, "generation_rouge_l_f1": 0.4041095890410959, "generation_token_f1": 0.6232876712328765, "prompt_id": "51338c6674b54120", "reading_canonical_log_prob": -273.9061584472656, "reading_distances": [28, 22, 22, 29], "reading_log_probs": [-273.4908752441406, -275.02777099609375, -275.7892150878906, -274.8765563964844], "reading_logprob_std": 0.957433013007864, "tokenization_path_logprob_std": 205.4661969444233, "trajectory_balance_rmse": 5207.482656709611} {"canonical_length": 130, "canonical_nll": 2.133148897611178, "generation_rouge_l_f1": 0.15816326530612246, "generation_token_f1": 0.2755102040816326, "prompt_id": "156614ecc1fc9430", "reading_canonical_log_prob": -277.3093566894531, "reading_distances": [67, 59, 68, 64], "reading_log_probs": [-295.39349365234375, -296.8414611816406, -294.4439392089844, -297.88739013671875], "reading_logprob_std": 1.5252654567247923, "tokenization_path_logprob_std": 100.05862472017927, "trajectory_balance_rmse": 3261.3755388769123} {"canonical_length": 234, "canonical_nll": 1.5243682861328125, "generation_rouge_l_f1": 0.24307036247334754, "generation_token_f1": 0.5159914712153517, "prompt_id": "39459679cbe7ca40", "reading_canonical_log_prob": -356.7021789550781, "reading_distances": [25, 26, 27, 29], "reading_log_probs": [-370.7032775878906, -367.27166748046875, -365.9843444824219, -366.9220886230469], "reading_logprob_std": 2.0615616557042054, "tokenization_path_logprob_std": 82.1760142698818, "trajectory_balance_rmse": 4022.451696879831} {"canonical_length": 81, "canonical_nll": 1.0973133040063174, "generation_rouge_l_f1": 0.2682926829268293, "generation_token_f1": 0.3841463414634146, "prompt_id": "0fc229d90ecbfd68", "reading_canonical_log_prob": -88.88237762451172, "reading_distances": [24, 26, 32, 30], "reading_log_probs": [-90.55499267578125, -88.65754699707031, -92.11275482177734, -90.77619934082031], "reading_logprob_std": 1.4227140968083156, "tokenization_path_logprob_std": 102.93874772634182, "trajectory_balance_rmse": 1844.8554617719499} {"canonical_length": 7, "canonical_nll": 2.7641078404017856, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "fc7a567baaf9e2df", "reading_canonical_log_prob": -19.3487548828125, "reading_distances": [19, 24, 24, 21], "reading_log_probs": [-22.024673461914062, -20.386377334594727, -23.817630767822266, -21.418752670288086], "reading_logprob_std": 1.4393217497296178, "tokenization_path_logprob_std": 55.38844267079742, "trajectory_balance_rmse": 127.35455907700694} {"canonical_length": 3, "canonical_nll": 1.863329251607259, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.2222222222222222, "prompt_id": "14871a1dde668d87", "reading_canonical_log_prob": -5.589987754821777, "reading_distances": [30, 26, 20, 26], "reading_log_probs": [-5.616631031036377, -5.301844596862793, -4.480679512023926, -5.3927321434021], "reading_logprob_std": 0.4961535136803672, "tokenization_path_logprob_std": 27.182659310482453, "trajectory_balance_rmse": 45.529837646589485} {"canonical_length": 3, "canonical_nll": 1.3763655026753743, "generation_rouge_l_f1": 0.034482758620689655, "generation_token_f1": 0.034482758620689655, "prompt_id": "408cd1b200daeec5", "reading_canonical_log_prob": -4.129096508026123, "reading_distances": [31, 29, 23, 30], "reading_log_probs": [-4.480188369750977, -5.3310980796813965, -4.889196395874023, -4.268194198608398], "reading_logprob_std": 0.4696596210773233, "tokenization_path_logprob_std": 18.26951182131665, "trajectory_balance_rmse": 53.10820733939125} {"canonical_length": 129, "canonical_nll": 1.6843098485192587, "generation_rouge_l_f1": 0.23280423280423282, "generation_token_f1": 0.31746031746031744, "prompt_id": "6db695863766c033", "reading_canonical_log_prob": -217.27597045898438, "reading_distances": [9, 13, 11, 12], "reading_log_probs": [-221.61489868164062, -218.47308349609375, -221.76470947265625, -221.7476806640625], "reading_logprob_std": 1.6193917363850352, "tokenization_path_logprob_std": 88.84891391908604, "trajectory_balance_rmse": 2567.1041044733415} {"canonical_length": 6, "canonical_nll": 0.09685635566711426, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "66a509bfd96df65b", "reading_canonical_log_prob": -0.5811381340026855, "reading_distances": [16, 17, 13, 19], "reading_log_probs": [-0.9682716131210327, -1.1141512393951416, -1.8134422302246094, -1.4065649509429932], "reading_logprob_std": 0.3727999995873892, "tokenization_path_logprob_std": 27.435154969349906, "trajectory_balance_rmse": 91.88051013397059} {"canonical_length": 3, "canonical_nll": 0.945428212483724, "generation_rouge_l_f1": 0.052631578947368425, "generation_token_f1": 0.052631578947368425, "prompt_id": "2c889c2667571076", "reading_canonical_log_prob": -2.836284637451172, "reading_distances": [13, 18, 20, 19], "reading_log_probs": [-4.861045837402344, -5.915553569793701, -3.5687108039855957, -3.854325771331787], "reading_logprob_std": 1.0658919817789814, "tokenization_path_logprob_std": 49.86259801954718, "trajectory_balance_rmse": 90.84000442888875} {"canonical_length": 15, "canonical_nll": 0.44223756790161134, "generation_rouge_l_f1": 0.9333333333333333, "generation_token_f1": 0.9333333333333333, "prompt_id": "79afdb8eb4cf0c62", "reading_canonical_log_prob": -6.63356351852417, "reading_distances": [23, 24, 26, 24], "reading_log_probs": [-8.619072914123535, -6.939155101776123, -9.478473663330078, -7.830822467803955], "reading_logprob_std": 1.0855104859355353, "tokenization_path_logprob_std": 65.13065186250086, "trajectory_balance_rmse": 253.38446085755177} {"canonical_length": 145, "canonical_nll": 1.3733514193830818, "generation_rouge_l_f1": 0.30412371134020616, "generation_token_f1": 0.4587628865979382, "prompt_id": "51fd58caa03ae57f", "reading_canonical_log_prob": -199.13595581054688, "reading_distances": [29, 30, 29, 22], "reading_log_probs": [-212.25648498535156, -214.55221557617188, -209.85794067382812, -213.09628295898438], "reading_logprob_std": 1.965779569638198, "tokenization_path_logprob_std": 195.74180570088248, "trajectory_balance_rmse": 3324.1888967157} {"canonical_length": 19, "canonical_nll": 0.8316924446507504, "generation_rouge_l_f1": 0.4117647058823529, "generation_token_f1": 0.4411764705882353, "prompt_id": "8671c943f4179a67", "reading_canonical_log_prob": -15.802156448364258, "reading_distances": [16, 13, 15, 16], "reading_log_probs": [-16.517929077148438, -19.168376922607422, -18.899250030517578, -19.10280990600586], "reading_logprob_std": 1.2746024577774728, "tokenization_path_logprob_std": 46.12121934481701, "trajectory_balance_rmse": 412.4851723440782} {"canonical_length": 289, "canonical_nll": 1.2635295300335208, "generation_rouge_l_f1": 0.24731182795698925, "generation_token_f1": 0.45878136200716846, "prompt_id": "9cb312e17f963300", "reading_canonical_log_prob": -365.1600341796875, "reading_distances": [21, 13, 21, 18], "reading_log_probs": [-377.24713134765625, -376.0048828125, -375.68853759765625, -374.391357421875], "reading_logprob_std": 1.1731032742756473, "tokenization_path_logprob_std": 198.0188183826618, "trajectory_balance_rmse": 6780.04841400437} {"canonical_length": 28, "canonical_nll": 2.1532557351248607, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.35000000000000003, "prompt_id": "9227ddd5a86a7e6a", "reading_canonical_log_prob": -60.291160583496094, "reading_distances": [25, 25, 27, 24], "reading_log_probs": [-60.47611999511719, -60.78692626953125, -57.39075469970703, -59.73680877685547], "reading_logprob_std": 1.5357777192895612, "tokenization_path_logprob_std": 29.988505305375213, "trajectory_balance_rmse": 621.2328582246503} {"canonical_length": 223, "canonical_nll": 1.590182864612528, "generation_rouge_l_f1": 0.21757322175732216, "generation_token_f1": 0.47280334728033474, "prompt_id": "3216cdb4c067a3d8", "reading_canonical_log_prob": -354.61077880859375, "reading_distances": [46, 47, 49, 44], "reading_log_probs": [-365.3714599609375, -367.16326904296875, -368.27655029296875, -364.3145751953125], "reading_logprob_std": 1.7752660243196818, "tokenization_path_logprob_std": 256.6847212077311, "trajectory_balance_rmse": 5335.715105556704} {"canonical_length": 108, "canonical_nll": 0.9012833348026982, "generation_rouge_l_f1": 0.2891566265060241, "generation_token_f1": 0.3795180722891566, "prompt_id": "1ef389c9aee61602", "reading_canonical_log_prob": -97.3386001586914, "reading_distances": [14, 9, 14, 15], "reading_log_probs": [-92.12455749511719, -92.36344909667969, -89.61309814453125, -91.33598327636719], "reading_logprob_std": 1.244146163533295, "tokenization_path_logprob_std": 65.05602789794756, "trajectory_balance_rmse": 1485.194908324616} {"canonical_length": 6, "canonical_nll": 3.2097765604654946, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "a730d1195f266d12", "reading_canonical_log_prob": -19.25865936279297, "reading_distances": [32, 30, 25, 29], "reading_log_probs": [-24.25063705444336, -21.378494262695312, -21.805261611938477, -25.733867645263672], "reading_logprob_std": 2.061843659074436, "tokenization_path_logprob_std": 38.036769228234036, "trajectory_balance_rmse": 129.78900291621625} {"canonical_length": 26, "canonical_nll": 1.4801122225247896, "generation_rouge_l_f1": 0.273972602739726, "generation_token_f1": 0.4657534246575343, "prompt_id": "7f409abdb6fbebf8", "reading_canonical_log_prob": -38.48291778564453, "reading_distances": [15, 11, 11, 11], "reading_log_probs": [-41.30583190917969, -43.40192413330078, -43.140663146972656, -46.09565353393555], "reading_logprob_std": 1.9739765925251205, "tokenization_path_logprob_std": 57.57184497280803, "trajectory_balance_rmse": 449.8797926633938} {"canonical_length": 168, "canonical_nll": 1.4048800695510137, "generation_rouge_l_f1": 0.25679012345679014, "generation_token_f1": 0.4592592592592592, "prompt_id": "1c78ad010cb27d62", "reading_canonical_log_prob": -236.0198516845703, "reading_distances": [39, 33, 35, 41], "reading_log_probs": [-239.37655639648438, -235.2880096435547, -236.80206298828125, -234.72817993164062], "reading_logprob_std": 2.0788296464663127, "tokenization_path_logprob_std": 83.38756260839754, "trajectory_balance_rmse": 3720.659835841355} {"canonical_length": 15, "canonical_nll": 0.8020760218302408, "generation_rouge_l_f1": 0.042735042735042736, "generation_token_f1": 0.05128205128205128, "prompt_id": "779d0398df55911b", "reading_canonical_log_prob": -12.031140327453613, "reading_distances": [6, 5, 8, 7], "reading_log_probs": [-16.078041076660156, -12.220148086547852, -12.816638946533203, -13.329687118530273], "reading_logprob_std": 1.7059608312485413, "tokenization_path_logprob_std": 53.82221515986815, "trajectory_balance_rmse": 197.8376902026034} {"canonical_length": 11, "canonical_nll": 0.9265428889881481, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.18181818181818182, "prompt_id": "3e344435d6b1501d", "reading_canonical_log_prob": -10.191971778869629, "reading_distances": [41, 47, 43, 38], "reading_log_probs": [-9.267268180847168, -10.119976043701172, -10.062997817993164, -8.722406387329102], "reading_logprob_std": 0.6714911004267519, "tokenization_path_logprob_std": 38.587103367168844, "trajectory_balance_rmse": 165.24028243963426} {"canonical_length": 376, "canonical_nll": 0.46502551626651845, "generation_rouge_l_f1": 0.3085714285714286, "generation_token_f1": 0.4533333333333333, "prompt_id": "e2124bce16d0a6b8", "reading_canonical_log_prob": -174.84959411621094, "reading_distances": [13, 17, 9, 12], "reading_log_probs": [-179.23492431640625, -176.18161010742188, -180.48977661132812, -183.424560546875], "reading_logprob_std": 3.0011667193305978, "tokenization_path_logprob_std": 236.5844173032574, "trajectory_balance_rmse": 5103.874936056901} {"canonical_length": 240, "canonical_nll": 2.148565673828125, "generation_rouge_l_f1": 0.27572016460905346, "generation_token_f1": 0.46090534979423864, "prompt_id": "6a9d3b5c30c1cdf4", "reading_canonical_log_prob": -515.65576171875, "reading_distances": [11, 18, 14, 13], "reading_log_probs": [-519.4200439453125, -534.0361328125, -519.5238037109375, -523.1048583984375], "reading_logprob_std": 6.892889369723125, "tokenization_path_logprob_std": 148.49787805065517, "trajectory_balance_rmse": 4379.173890368197} {"canonical_length": 271, "canonical_nll": 1.1150485668675045, "generation_rouge_l_f1": 0.3299798792756539, "generation_token_f1": 0.5875251509054327, "prompt_id": "63c15bf241bee263", "reading_canonical_log_prob": -302.17816162109375, "reading_distances": [33, 24, 32, 30], "reading_log_probs": [-314.3404846191406, -313.1299133300781, -314.42510986328125, -311.3929138183594], "reading_logprob_std": 1.415678050783409, "tokenization_path_logprob_std": 71.90557863377134, "trajectory_balance_rmse": 5949.57904071498} {"canonical_length": 25, "canonical_nll": 1.2768466186523437, "generation_rouge_l_f1": 0.4444444444444445, "generation_token_f1": 0.4444444444444445, "prompt_id": "9932e0e2da85e27d", "reading_canonical_log_prob": -31.921165466308594, "reading_distances": [24, 18, 24, 22], "reading_log_probs": [-33.982032775878906, -31.468414306640625, -31.416545867919922, -34.026222229003906], "reading_logprob_std": 1.479229416616764, "tokenization_path_logprob_std": 53.063894593354895, "trajectory_balance_rmse": 515.2034383652125} {"canonical_length": 114, "canonical_nll": 0.9726456759268778, "generation_rouge_l_f1": 0.2545454545454545, "generation_token_f1": 0.44155844155844154, "prompt_id": "f5ce2237f02c3721", "reading_canonical_log_prob": -110.88160705566406, "reading_distances": [24, 22, 21, 23], "reading_log_probs": [-114.78033447265625, -110.67975616455078, -115.25025177001953, -114.01936340332031], "reading_logprob_std": 2.0650289436179636, "tokenization_path_logprob_std": 151.46974503156414, "trajectory_balance_rmse": 2399.967752750525} {"canonical_length": 297, "canonical_nll": 1.4485192090172558, "generation_rouge_l_f1": 0.2787193973634651, "generation_token_f1": 0.5386064030131827, "prompt_id": "3c5a7e9edbe46ab0", "reading_canonical_log_prob": -430.210205078125, "reading_distances": [13, 21, 18, 16], "reading_log_probs": [-436.9396057128906, -439.20697021484375, -438.2203369140625, -440.58172607421875], "reading_logprob_std": 1.54071987042781, "tokenization_path_logprob_std": 562.7900795529991, "trajectory_balance_rmse": 5992.380100862704} {"canonical_length": 129, "canonical_nll": 1.31214538101078, "generation_rouge_l_f1": 0.19354838709677416, "generation_token_f1": 0.3920595533498759, "prompt_id": "6097f6fb796f153f", "reading_canonical_log_prob": -169.26675415039062, "reading_distances": [26, 25, 32, 21], "reading_log_probs": [-168.3463592529297, -168.55709838867188, -164.07229614257812, -167.85159301757812], "reading_logprob_std": 2.1105112723153097, "tokenization_path_logprob_std": 254.29473831024262, "trajectory_balance_rmse": 2666.490043333633} {"canonical_length": 10, "canonical_nll": 0.5182074069976806, "generation_rouge_l_f1": 0.1037037037037037, "generation_token_f1": 0.1037037037037037, "prompt_id": "b2680f937aebb340", "reading_canonical_log_prob": -5.182074069976807, "reading_distances": [7, 10, 5, 6], "reading_log_probs": [-7.626395225524902, -8.98571491241455, -4.217597484588623, -5.151373863220215], "reading_logprob_std": 2.1966344750930342, "tokenization_path_logprob_std": 15.79635967611309, "trajectory_balance_rmse": 47.895357821515184} {"canonical_length": 8, "canonical_nll": 1.2477418184280396, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "a6d7ae5badbbf976", "reading_canonical_log_prob": -9.981934547424316, "reading_distances": [29, 35, 25, 32], "reading_log_probs": [-13.1099214553833, -12.619868278503418, -13.786825180053711, -12.08974838256836], "reading_logprob_std": 0.7223796922812715, "tokenization_path_logprob_std": 16.62561272807141, "trajectory_balance_rmse": 86.74096163943273} {"canonical_length": 31, "canonical_nll": 2.0048353133663053, "generation_rouge_l_f1": 0.125, "generation_token_f1": 0.21875, "prompt_id": "5eee9bc9190ed48e", "reading_canonical_log_prob": -62.14989471435547, "reading_distances": [26, 31, 23, 22], "reading_log_probs": [-77.25784301757812, -81.4750747680664, -74.64743041992188, -74.64811706542969], "reading_logprob_std": 3.222758218287414, "tokenization_path_logprob_std": 145.447310986004, "trajectory_balance_rmse": 775.619226700552} {"canonical_length": 14, "canonical_nll": 0.6311964988708496, "generation_rouge_l_f1": 0.5882352941176471, "generation_token_f1": 0.5882352941176471, "prompt_id": "89b79b80175b4a32", "reading_canonical_log_prob": -8.836750984191895, "reading_distances": [28, 41, 37, 37], "reading_log_probs": [-18.42803192138672, -19.02079200744629, -19.9798526763916, -19.061752319335938], "reading_logprob_std": 0.640670033217819, "tokenization_path_logprob_std": 56.236938034878335, "trajectory_balance_rmse": 261.43748569079287} {"canonical_length": 10, "canonical_nll": 0.1590540051460266, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "5a773e9fe3bdab08", "reading_canonical_log_prob": -1.5905400514602661, "reading_distances": [19, 16, 20, 21], "reading_log_probs": [-1.7310510873794556, -1.5069025754928589, -2.2982592582702637, -1.825951099395752], "reading_logprob_std": 0.3331773843999956, "tokenization_path_logprob_std": 34.09314092237867, "trajectory_balance_rmse": 214.84836937373635} {"canonical_length": 37, "canonical_nll": 1.3539420462943412, "generation_rouge_l_f1": 0.33962264150943394, "generation_token_f1": 0.33962264150943394, "prompt_id": "bc2e832a6508e078", "reading_canonical_log_prob": -50.095855712890625, "reading_distances": [22, 12, 14, 17], "reading_log_probs": [-53.428993225097656, -49.73957824707031, -50.06222915649414, -50.04054260253906], "reading_logprob_std": 1.7469887009386351, "tokenization_path_logprob_std": 48.4753313898906, "trajectory_balance_rmse": 865.2172937643362} {"canonical_length": 245, "canonical_nll": 0.7604148397640306, "generation_rouge_l_f1": 0.24827586206896554, "generation_token_f1": 0.42758620689655175, "prompt_id": "66a81182d7c90c5b", "reading_canonical_log_prob": -186.3016357421875, "reading_distances": [16, 10, 19, 16], "reading_log_probs": [-188.4032745361328, -189.93588256835938, -192.36473083496094, -190.40103149414062], "reading_logprob_std": 1.6331166211089057, "tokenization_path_logprob_std": 217.69050560258916, "trajectory_balance_rmse": 3850.520457371567} {"canonical_length": 92, "canonical_nll": 1.194306000419285, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.34615384615384615, "prompt_id": "aed52e93b5146883", "reading_canonical_log_prob": -109.87615203857422, "reading_distances": [24, 29, 23, 28], "reading_log_probs": [-115.61148071289062, -118.25424194335938, -115.73379516601562, -118.455078125], "reading_logprob_std": 1.551439172215171, "tokenization_path_logprob_std": 163.78175134202382, "trajectory_balance_rmse": 1513.244239061427} {"canonical_length": 131, "canonical_nll": 0.9378121645395993, "generation_rouge_l_f1": 0.4819277108433735, "generation_token_f1": 0.6586345381526104, "prompt_id": "b1543c03f1b0c7a8", "reading_canonical_log_prob": -122.8533935546875, "reading_distances": [17, 20, 19, 19], "reading_log_probs": [-124.87332916259766, -121.44792175292969, -126.34773254394531, -126.40596008300781], "reading_logprob_std": 2.3246849088983583, "tokenization_path_logprob_std": 141.7943809203762, "trajectory_balance_rmse": 2410.462683043475} {"canonical_length": 501, "canonical_nll": 0.33412392743809255, "generation_rouge_l_f1": 0.08030592734225622, "generation_token_f1": 0.1147227533460803, "prompt_id": "c52f77a538102872", "reading_canonical_log_prob": -167.39608764648438, "reading_distances": [32, 29, 23, 31], "reading_log_probs": [-171.2750244140625, -172.45907592773438, -171.39077758789062, -174.3115234375], "reading_logprob_std": 1.406511125118254, "tokenization_path_logprob_std": 50.63143514045756, "trajectory_balance_rmse": 2943.6202723746846} {"canonical_length": 67, "canonical_nll": 0.6171673447338503, "generation_rouge_l_f1": 0.20224719101123595, "generation_token_f1": 0.24157303370786515, "prompt_id": "fdb23a8f949c30eb", "reading_canonical_log_prob": -41.35021209716797, "reading_distances": [19, 20, 23, 21], "reading_log_probs": [-47.279327392578125, -43.13516616821289, -44.70442199707031, -46.7696418762207], "reading_logprob_std": 1.9148781521868863, "tokenization_path_logprob_std": 55.95371812484584, "trajectory_balance_rmse": 851.5924211898919} {"canonical_length": 268, "canonical_nll": 1.2615864312470848, "generation_rouge_l_f1": 0.269449715370019, "generation_token_f1": 0.5009487666034156, "prompt_id": "8aa2f18f3606aa8c", "reading_canonical_log_prob": -338.10516357421875, "reading_distances": [22, 27, 24, 22], "reading_log_probs": [-345.9656066894531, -357.2061767578125, -359.372314453125, -349.2481689453125], "reading_logprob_std": 6.373027842713474, "tokenization_path_logprob_std": 361.3844293872569, "trajectory_balance_rmse": 5777.9666301262505} {"canonical_length": 123, "canonical_nll": 0.9008470240647231, "generation_rouge_l_f1": 0.14792899408284024, "generation_token_f1": 0.2485207100591716, "prompt_id": "75e528b4a6af3ee2", "reading_canonical_log_prob": -110.80418395996094, "reading_distances": [21, 18, 15, 19], "reading_log_probs": [-116.28736877441406, -113.47994232177734, -114.63737487792969, -113.17391204833984], "reading_logprob_std": 1.4104598551072378, "tokenization_path_logprob_std": 121.32321539392238, "trajectory_balance_rmse": 1446.1708680807276} {"canonical_length": 323, "canonical_nll": 1.5794653169141835, "generation_rouge_l_f1": 0.17894736842105263, "generation_token_f1": 0.36140350877192984, "prompt_id": "af23751b53e953f7", "reading_canonical_log_prob": -510.16729736328125, "reading_distances": [33, 36, 31, 28], "reading_log_probs": [-495.4378662109375, -501.36553955078125, -508.120849609375, -502.89471435546875], "reading_logprob_std": 5.219235597149943, "tokenization_path_logprob_std": 233.19042258372514, "trajectory_balance_rmse": 6681.590551097682} {"canonical_length": 250, "canonical_nll": 1.724156982421875, "generation_rouge_l_f1": 0.25051334702258726, "generation_token_f1": 0.5010266940451745, "prompt_id": "b8fd0dbaa9eea228", "reading_canonical_log_prob": -431.03924560546875, "reading_distances": [15, 20, 19, 21], "reading_log_probs": [-437.31304931640625, -426.3522033691406, -433.2516784667969, -432.99658203125], "reading_logprob_std": 4.537642242964879, "tokenization_path_logprob_std": 156.08139706954012, "trajectory_balance_rmse": 5369.995343559864} {"canonical_length": 13, "canonical_nll": 1.8855609893798828, "generation_rouge_l_f1": 0.23076923076923075, "generation_token_f1": 0.23076923076923075, "prompt_id": "44c43dead4708e04", "reading_canonical_log_prob": -24.512292861938477, "reading_distances": [34, 24, 29, 22], "reading_log_probs": [-32.89270782470703, -29.082143783569336, -33.42533874511719, -28.55088996887207], "reading_logprob_std": 2.5258876371175356, "tokenization_path_logprob_std": 39.90334193507522, "trajectory_balance_rmse": 384.0334037298002} {"canonical_length": 138, "canonical_nll": 1.7399495442708333, "generation_rouge_l_f1": 0.16666666666666666, "generation_token_f1": 0.24444444444444446, "prompt_id": "36e5fedd319115db", "reading_canonical_log_prob": -240.113037109375, "reading_distances": [30, 27, 35, 26], "reading_log_probs": [-246.95220947265625, -239.75567626953125, -241.19546508789062, -240.16323852539062], "reading_logprob_std": 3.3457051210996895, "tokenization_path_logprob_std": 102.28721919210494, "trajectory_balance_rmse": 3257.392262819546} {"canonical_length": 13, "canonical_nll": 0.5393135364239032, "generation_rouge_l_f1": 0.8695652173913043, "generation_token_f1": 0.8695652173913043, "prompt_id": "15b19e1f65328c81", "reading_canonical_log_prob": -7.011075973510742, "reading_distances": [24, 21, 16, 24], "reading_log_probs": [-7.6452765464782715, -9.149943351745605, -8.956006050109863, -8.722283363342285], "reading_logprob_std": 0.6718822284629462, "tokenization_path_logprob_std": 41.740142011751814, "trajectory_balance_rmse": 212.01210150268014} {"canonical_length": 20, "canonical_nll": 1.2797932624816895, "generation_rouge_l_f1": 0.1, "generation_token_f1": 0.1, "prompt_id": "7e1ececb041b4ecb", "reading_canonical_log_prob": -25.59586524963379, "reading_distances": [21, 24, 26, 23], "reading_log_probs": [-28.994354248046875, -24.335134506225586, -25.23343276977539, -24.710193634033203], "reading_logprob_std": 2.149191310078197, "tokenization_path_logprob_std": 68.92654170203585, "trajectory_balance_rmse": 423.9708677761578} {"canonical_length": 359, "canonical_nll": 1.3559613666162518, "generation_rouge_l_f1": 0.22580645161290322, "generation_token_f1": 0.33548387096774196, "prompt_id": "4db416527bef041c", "reading_canonical_log_prob": -486.7901306152344, "reading_distances": [18, 21, 22, 15], "reading_log_probs": [-480.94268798828125, -476.7535400390625, -481.2838134765625, -485.2657775878906], "reading_logprob_std": 3.4784100117193417, "tokenization_path_logprob_std": 409.464448975485, "trajectory_balance_rmse": 7856.108100260891} {"canonical_length": 78, "canonical_nll": 0.7729614453438001, "generation_rouge_l_f1": 0.31868131868131866, "generation_token_f1": 0.6263736263736264, "prompt_id": "ba11bfd32f28c7ee", "reading_canonical_log_prob": -60.290992736816406, "reading_distances": [16, 14, 14, 18], "reading_log_probs": [-70.97798156738281, -66.94534301757812, -67.0463638305664, -68.2847671508789], "reading_logprob_std": 1.8777446709721362, "tokenization_path_logprob_std": 96.33106309418986, "trajectory_balance_rmse": 1580.6446544802893} {"canonical_length": 292, "canonical_nll": 1.526546739552119, "generation_rouge_l_f1": 0.18726591760299627, "generation_token_f1": 0.3632958801498127, "prompt_id": "2bb0a1d6597bcbfc", "reading_canonical_log_prob": -445.75164794921875, "reading_distances": [50, 64, 52, 54], "reading_log_probs": [-459.9969787597656, -475.12298583984375, -458.7764892578125, -468.54974365234375], "reading_logprob_std": 7.688559883255023, "tokenization_path_logprob_std": 231.36229145327573, "trajectory_balance_rmse": 6647.387875659986} {"canonical_length": 2, "canonical_nll": 1.9058377742767334, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "bad903d11c6658df", "reading_canonical_log_prob": -3.811675548553467, "reading_distances": [28, 34, 34, 30], "reading_log_probs": [-11.389792442321777, -8.439462661743164, -9.69044017791748, -10.295644760131836], "reading_logprob_std": 1.2303804424456204, "tokenization_path_logprob_std": 27.474193359596775, "trajectory_balance_rmse": 43.59375600018005} {"canonical_length": 99, "canonical_nll": 0.8747849897904829, "generation_rouge_l_f1": 0.1134020618556701, "generation_token_f1": 0.20618556701030927, "prompt_id": "7c8a24a51c291f25", "reading_canonical_log_prob": -86.60371398925781, "reading_distances": [10, 13, 11, 10], "reading_log_probs": [-87.48460388183594, -85.89543914794922, -89.80709075927734, -90.87523651123047], "reading_logprob_std": 2.2482607153456047, "tokenization_path_logprob_std": 98.45807882875526, "trajectory_balance_rmse": 1952.9675304870846} {"canonical_length": 108, "canonical_nll": 1.9774873521592882, "generation_rouge_l_f1": 0.2446808510638298, "generation_token_f1": 0.35106382978723405, "prompt_id": "cef16e435e5a8a53", "reading_canonical_log_prob": -213.56863403320312, "reading_distances": [20, 21, 18, 23], "reading_log_probs": [-225.38633728027344, -225.10086059570312, -221.07321166992188, -229.8297882080078], "reading_logprob_std": 3.578770162528008, "tokenization_path_logprob_std": 221.72082213507036, "trajectory_balance_rmse": 2689.4711072607633} {"canonical_length": 38, "canonical_nll": 3.0625666568153784, "generation_rouge_l_f1": 0.225, "generation_token_f1": 0.375, "prompt_id": "289239d155fcb98c", "reading_canonical_log_prob": -116.37753295898438, "reading_distances": [28, 17, 22, 19], "reading_log_probs": [-116.94070434570312, -117.32616424560547, -120.13970947265625, -120.11226654052734], "reading_logprob_std": 1.7349393781267095, "tokenization_path_logprob_std": 48.43800564723438, "trajectory_balance_rmse": 745.5539476596604} {"canonical_length": 8, "canonical_nll": 2.4560093879699707, "generation_rouge_l_f1": 0.33333333333333326, "generation_token_f1": 0.4444444444444444, "prompt_id": "c036d970930b8362", "reading_canonical_log_prob": -19.648075103759766, "reading_distances": [24, 36, 31, 26], "reading_log_probs": [-20.45248794555664, -25.165279388427734, -25.92707061767578, -21.718536376953125], "reading_logprob_std": 2.645065546887118, "tokenization_path_logprob_std": 32.398483421509276, "trajectory_balance_rmse": 112.01978202140373} {"canonical_length": 99, "canonical_nll": 1.132212089769768, "generation_rouge_l_f1": 0.1792717086834734, "generation_token_f1": 0.3473389355742297, "prompt_id": "1d5889fbec769861", "reading_canonical_log_prob": -112.08899688720703, "reading_distances": [22, 16, 23, 24], "reading_log_probs": [-114.56573486328125, -118.58491516113281, -122.29148864746094, -115.1820068359375], "reading_logprob_std": 3.560010260428839, "tokenization_path_logprob_std": 115.58186735036956, "trajectory_balance_rmse": 1836.1996693380008} {"canonical_length": 185, "canonical_nll": 0.914030662742821, "generation_rouge_l_f1": 0.15456674473067913, "generation_token_f1": 0.23419203747072598, "prompt_id": "c20bcd26ca03079c", "reading_canonical_log_prob": -169.09567260742188, "reading_distances": [38, 36, 38, 37], "reading_log_probs": [-165.60914611816406, -164.11508178710938, -163.76744079589844, -166.5478515625], "reading_logprob_std": 1.2998455581489772, "tokenization_path_logprob_std": 189.63054998377388, "trajectory_balance_rmse": 3279.442716619219} {"canonical_length": 16, "canonical_nll": 0.803476870059967, "generation_rouge_l_f1": 0.12017167381974247, "generation_token_f1": 0.12017167381974247, "prompt_id": "93426790937338d2", "reading_canonical_log_prob": -12.855629920959473, "reading_distances": [15, 17, 15, 16], "reading_log_probs": [-13.268143653869629, -10.118547439575195, -12.893117904663086, -11.319831848144531], "reading_logprob_std": 1.4569681428008605, "tokenization_path_logprob_std": 35.63492947102252, "trajectory_balance_rmse": 200.58029337370036} {"canonical_length": 203, "canonical_nll": 1.4552719041044488, "generation_rouge_l_f1": 0.1687763713080169, "generation_token_f1": 0.3333333333333333, "prompt_id": "f380c029d22e1217", "reading_canonical_log_prob": -295.4201965332031, "reading_distances": [19, 12, 12, 11], "reading_log_probs": [-301.2754821777344, -300.30938720703125, -296.1219177246094, -294.9608154296875], "reading_logprob_std": 3.0937822249729887, "tokenization_path_logprob_std": 78.66723797342857, "trajectory_balance_rmse": 4024.3407646781384} {"canonical_length": 191, "canonical_nll": 0.7236777100887598, "generation_rouge_l_f1": 0.24786324786324787, "generation_token_f1": 0.5170940170940171, "prompt_id": "b8396f86fe7c1b16", "reading_canonical_log_prob": -138.22244262695312, "reading_distances": [41, 36, 35, 34], "reading_log_probs": [-143.41468811035156, -141.85028076171875, -143.23915100097656, -139.2340087890625], "reading_logprob_std": 1.9315638511047768, "tokenization_path_logprob_std": 196.10100594038568, "trajectory_balance_rmse": 2771.0292494703112} {"canonical_length": 149, "canonical_nll": 1.732212962720218, "generation_rouge_l_f1": 0.1826923076923077, "generation_token_f1": 0.4134615384615385, "prompt_id": "6cb0fdf8b6463b4b", "reading_canonical_log_prob": -258.0997314453125, "reading_distances": [45, 51, 38, 49], "reading_log_probs": [-259.91241455078125, -262.8824462890625, -259.37249755859375, -263.5625], "reading_logprob_std": 2.097102291868444, "tokenization_path_logprob_std": 116.30230154076655, "trajectory_balance_rmse": 3430.70936401547} {"canonical_length": 118, "canonical_nll": 1.4796941724874206, "generation_rouge_l_f1": 0.23129251700680273, "generation_token_f1": 0.272108843537415, "prompt_id": "7c85ca0add25adb8", "reading_canonical_log_prob": -174.60391235351562, "reading_distances": [33, 33, 38, 35], "reading_log_probs": [-175.53761291503906, -169.13296508789062, -177.1910400390625, -174.01675415039062], "reading_logprob_std": 3.4752266695255676, "tokenization_path_logprob_std": 173.45891356285728, "trajectory_balance_rmse": 2717.127161289327} {"canonical_length": 114, "canonical_nll": 1.5074853729783444, "generation_rouge_l_f1": 0.23868312757201646, "generation_token_f1": 0.4197530864197531, "prompt_id": "9a1872631342b4a7", "reading_canonical_log_prob": -171.85333251953125, "reading_distances": [30, 29, 23, 28], "reading_log_probs": [-167.36199951171875, -172.19631958007812, -167.39547729492188, -168.12933349609375], "reading_logprob_std": 2.3109806739988197, "tokenization_path_logprob_std": 209.30405219936128, "trajectory_balance_rmse": 2721.849052274824} {"canonical_length": 13, "canonical_nll": 2.2498077979454627, "generation_rouge_l_f1": 0.007462686567164179, "generation_token_f1": 0.007462686567164179, "prompt_id": "d49306af1b19f211", "reading_canonical_log_prob": -29.247501373291016, "reading_distances": [22, 22, 20, 22], "reading_log_probs": [-26.98549461364746, -25.73594856262207, -27.638620376586914, -28.926820755004883], "reading_logprob_std": 1.329723385323067, "tokenization_path_logprob_std": 86.23153438444764, "trajectory_balance_rmse": 245.64121570006574} {"canonical_length": 253, "canonical_nll": 1.606416121773098, "generation_rouge_l_f1": 0.15439856373429087, "generation_token_f1": 0.355475763016158, "prompt_id": "807f35218626237b", "reading_canonical_log_prob": -406.42327880859375, "reading_distances": [17, 15, 21, 21], "reading_log_probs": [-408.89410400390625, -412.2125549316406, -406.259521484375, -409.9690856933594], "reading_logprob_std": 2.4722013691009552, "tokenization_path_logprob_std": 185.84948826524112, "trajectory_balance_rmse": 5328.19529220624} {"canonical_length": 4, "canonical_nll": 1.8452303409576416, "generation_rouge_l_f1": 0.1, "generation_token_f1": 0.1, "prompt_id": "1a44d44c0ed30f8a", "reading_canonical_log_prob": -7.380921363830566, "reading_distances": [20, 20, 28, 22], "reading_log_probs": [-9.283254623413086, -8.795578002929688, -10.39392375946045, -9.582917213439941], "reading_logprob_std": 0.6704200108458642, "tokenization_path_logprob_std": 36.485526987467296, "trajectory_balance_rmse": 138.49940899833877} {"canonical_length": 119, "canonical_nll": 0.24156647369641215, "generation_rouge_l_f1": 0.5910290237467019, "generation_token_f1": 0.6490765171503957, "prompt_id": "c78f29d81d80f8fc", "reading_canonical_log_prob": -28.746410369873047, "reading_distances": [21, 27, 20, 22], "reading_log_probs": [-29.93809700012207, -28.073184967041016, -30.129619598388672, -30.78078842163086], "reading_logprob_std": 1.1622093660114647, "tokenization_path_logprob_std": 87.80581337830081, "trajectory_balance_rmse": 1669.2616408664696} {"canonical_length": 279, "canonical_nll": 0.9372779545390905, "generation_rouge_l_f1": 0.2376599634369287, "generation_token_f1": 0.41316270566727603, "prompt_id": "dbd0e34044b57777", "reading_canonical_log_prob": -261.50054931640625, "reading_distances": [25, 18, 19, 23], "reading_log_probs": [-284.45269775390625, -274.9421081542969, -285.08453369140625, -301.10015869140625], "reading_logprob_std": 10.845894517696948, "tokenization_path_logprob_std": 219.20625067301114, "trajectory_balance_rmse": 4243.141667548889} {"canonical_length": 11, "canonical_nll": 0.17282251878218216, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1c94de058e5cb066", "reading_canonical_log_prob": -1.901047706604004, "reading_distances": [20, 22, 23, 12], "reading_log_probs": [-1.1993008852005005, -1.3097734451293945, -1.1825698614120483, -1.2877814769744873], "reading_logprob_std": 0.0632763852726332, "tokenization_path_logprob_std": 36.04416855261435, "trajectory_balance_rmse": 210.8409452665693} {"canonical_length": 40, "canonical_nll": 0.746882963180542, "generation_rouge_l_f1": 0.12698412698412698, "generation_token_f1": 0.18253968253968253, "prompt_id": "91069c7b26de61af", "reading_canonical_log_prob": -29.87531852722168, "reading_distances": [45, 43, 46, 36], "reading_log_probs": [-33.91272735595703, -38.311973571777344, -33.176849365234375, -32.67640686035156], "reading_logprob_std": 2.5788066931680635, "tokenization_path_logprob_std": 74.44156524020488, "trajectory_balance_rmse": 675.9463700931085} {"canonical_length": 10, "canonical_nll": 0.3157517910003662, "generation_rouge_l_f1": 0.875, "generation_token_f1": 1.0, "prompt_id": "c9ec0273bafaec5d", "reading_canonical_log_prob": -3.157517910003662, "reading_distances": [41, 38, 42, 49], "reading_log_probs": [-6.078888416290283, -5.410486221313477, -4.819972991943359, -5.968066692352295], "reading_logprob_std": 0.5789093189689177, "tokenization_path_logprob_std": 58.191131330990146, "trajectory_balance_rmse": 176.11647937637352} {"canonical_length": 12, "canonical_nll": 0.15030216177304587, "generation_rouge_l_f1": 0.9565217391304348, "generation_token_f1": 0.9565217391304348, "prompt_id": "1cb768d05dd2b1b9", "reading_canonical_log_prob": -1.8036259412765503, "reading_distances": [41, 43, 37, 40], "reading_log_probs": [-2.529237747192383, -2.72155499458313, -2.7345430850982666, -4.244475364685059], "reading_logprob_std": 0.7968964237895738, "tokenization_path_logprob_std": 29.207784140813473, "trajectory_balance_rmse": 162.33109063888887} {"canonical_length": 117, "canonical_nll": 0.9146372477213541, "generation_rouge_l_f1": 0.33027522935779813, "generation_token_f1": 0.44648318042813456, "prompt_id": "31716783a47d0f33", "reading_canonical_log_prob": -107.01255798339844, "reading_distances": [29, 30, 29, 26], "reading_log_probs": [-108.6370620727539, -116.51718139648438, -109.81136322021484, -107.48780059814453], "reading_logprob_std": 4.048587448640545, "tokenization_path_logprob_std": 149.3460060966816, "trajectory_balance_rmse": 2249.12839842922} {"canonical_length": 38, "canonical_nll": 0.5173832742791427, "generation_rouge_l_f1": 0.6486486486486486, "generation_token_f1": 0.6486486486486486, "prompt_id": "64444cea846a2353", "reading_canonical_log_prob": -19.660564422607422, "reading_distances": [621, 595, 616, 591], "reading_log_probs": [-29.675092697143555, -30.696762084960938, -39.95773696899414, -35.78385925292969], "reading_logprob_std": 4.771085472058262, "tokenization_path_logprob_std": 51.52065349751102, "trajectory_balance_rmse": 873.7066597739558} {"canonical_length": 17, "canonical_nll": 1.3973330329446232, "generation_rouge_l_f1": 0.4117647058823529, "generation_token_f1": 0.4117647058823529, "prompt_id": "bcb6a50b95a69177", "reading_canonical_log_prob": -23.754661560058594, "reading_distances": [31, 30, 30, 36], "reading_log_probs": [-21.53719711303711, -22.15658950805664, -23.400293350219727, -23.758440017700195], "reading_logprob_std": 1.0420207894284095, "tokenization_path_logprob_std": 46.735898835731106, "trajectory_balance_rmse": 435.70494838417915} {"canonical_length": 47, "canonical_nll": 0.6643037187292221, "generation_rouge_l_f1": 0.5681818181818183, "generation_token_f1": 0.6363636363636364, "prompt_id": "5a3dfb14391da9f8", "reading_canonical_log_prob": -31.222274780273438, "reading_distances": [25, 28, 24, 23], "reading_log_probs": [-32.56884002685547, -31.10881996154785, -30.639768600463867, -31.679035186767578], "reading_logprob_std": 0.8301584905144977, "tokenization_path_logprob_std": 93.71887320841556, "trajectory_balance_rmse": 717.8352880898366} {"canonical_length": 255, "canonical_nll": 0.1665877099130668, "generation_rouge_l_f1": 0.940766550522648, "generation_token_f1": 0.9512195121951219, "prompt_id": "1b64247b2a4cbf46", "reading_canonical_log_prob": -42.47986602783203, "reading_distances": [40, 50, 41, 46], "reading_log_probs": [-46.41322326660156, -44.059024810791016, -43.40599060058594, -42.52495193481445], "reading_logprob_std": 1.6648427109284547, "tokenization_path_logprob_std": 121.43694150900231, "trajectory_balance_rmse": 3089.308444458537} {"canonical_length": 8, "canonical_nll": 0.25600558519363403, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "ffe7a66de245047f", "reading_canonical_log_prob": -2.0480446815490723, "reading_distances": [23, 21, 28, 26], "reading_log_probs": [-2.562122106552124, -2.8481855392456055, -1.9264132976531982, -1.7867214679718018], "reading_logprob_std": 0.5068767087765711, "tokenization_path_logprob_std": 24.522050594327357, "trajectory_balance_rmse": 156.00536710172696} {"canonical_length": 195, "canonical_nll": 1.3847238393930288, "generation_rouge_l_f1": 0.3034482758620689, "generation_token_f1": 0.5011494252873563, "prompt_id": "9b0fbce06c332eb0", "reading_canonical_log_prob": -270.0211486816406, "reading_distances": [22, 24, 21, 17], "reading_log_probs": [-273.1966552734375, -280.419189453125, -270.5654602050781, -272.06256103515625], "reading_logprob_std": 4.37364156730458, "tokenization_path_logprob_std": 278.26231485824513, "trajectory_balance_rmse": 4717.604468617204} {"canonical_length": 103, "canonical_nll": 1.5667220921192355, "generation_rouge_l_f1": 0.26262626262626265, "generation_token_f1": 0.48484848484848486, "prompt_id": "f63c128779f1525d", "reading_canonical_log_prob": -161.37237548828125, "reading_distances": [33, 28, 31, 28], "reading_log_probs": [-161.04400634765625, -156.09666442871094, -166.7769775390625, -164.17967224121094], "reading_logprob_std": 4.594613237103323, "tokenization_path_logprob_std": 149.07176202145246, "trajectory_balance_rmse": 2365.208610948626} {"canonical_length": 45, "canonical_nll": 0.7739800347222222, "generation_rouge_l_f1": 0.4347826086956522, "generation_token_f1": 0.5869565217391305, "prompt_id": "e0f5d30e662ee161", "reading_canonical_log_prob": -34.8291015625, "reading_distances": [36, 27, 29, 30], "reading_log_probs": [-46.15786361694336, -42.73017883300781, -41.015533447265625, -39.16532516479492], "reading_logprob_std": 2.9743295844038062, "tokenization_path_logprob_std": 13.559067966266557, "trajectory_balance_rmse": 1025.3847885061352} {"canonical_length": 10, "canonical_nll": 1.9035648345947265, "generation_rouge_l_f1": 0.5714285714285714, "generation_token_f1": 0.5714285714285714, "prompt_id": "10787d0a825f897d", "reading_canonical_log_prob": -19.035648345947266, "reading_distances": [35, 37, 43, 32], "reading_log_probs": [-22.802459716796875, -19.376150131225586, -19.347692489624023, -19.07863426208496], "reading_logprob_std": 1.7725595393584614, "tokenization_path_logprob_std": 52.6382482958698, "trajectory_balance_rmse": 220.21722161685565} {"canonical_length": 138, "canonical_nll": 0.9960625689962636, "generation_rouge_l_f1": 0.21395348837209302, "generation_token_f1": 0.4, "prompt_id": "55c070ce04e00179", "reading_canonical_log_prob": -137.45663452148438, "reading_distances": [20, 14, 18, 20], "reading_log_probs": [-150.344482421875, -149.0562286376953, -149.42445373535156, -149.8438262939453], "reading_logprob_std": 0.5544126237506374, "tokenization_path_logprob_std": 136.44502025645394, "trajectory_balance_rmse": 3102.2617017625635} {"canonical_length": 57, "canonical_nll": 1.745874237595943, "generation_rouge_l_f1": 0.32835820895522383, "generation_token_f1": 0.46268656716417905, "prompt_id": "9718a6939e1171c6", "reading_canonical_log_prob": -99.51483154296875, "reading_distances": [19, 17, 25, 27], "reading_log_probs": [-101.13616943359375, -101.6139907836914, -100.67315673828125, -100.48207092285156], "reading_logprob_std": 0.5060849666483715, "tokenization_path_logprob_std": 70.35102515984464, "trajectory_balance_rmse": 1304.1265719836474} {"canonical_length": 180, "canonical_nll": 1.4394795735677084, "generation_rouge_l_f1": 0.21171171171171171, "generation_token_f1": 0.45045045045045046, "prompt_id": "abf26ab3207cd57f", "reading_canonical_log_prob": -259.1063232421875, "reading_distances": [23, 22, 24, 26], "reading_log_probs": [-257.8349609375, -261.9275207519531, -259.4063720703125, -259.7969665527344], "reading_logprob_std": 1.6861163645554489, "tokenization_path_logprob_std": 350.309213132203, "trajectory_balance_rmse": 3178.7921649105733} {"canonical_length": 8, "canonical_nll": 0.2863231301307678, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "9706792c2f395e2c", "reading_canonical_log_prob": -2.2905850410461426, "reading_distances": [26, 27, 19, 24], "reading_log_probs": [-2.9655072689056396, -2.5725343227386475, -2.504331350326538, -4.465763092041016], "reading_logprob_std": 0.9153350841319052, "tokenization_path_logprob_std": 24.54457095278277, "trajectory_balance_rmse": 133.28397173239108} {"canonical_length": 198, "canonical_nll": 1.3547989045730744, "generation_rouge_l_f1": 0.22795698924731186, "generation_token_f1": 0.4344086021505376, "prompt_id": "4d13949d8c9ff709", "reading_canonical_log_prob": -268.25018310546875, "reading_distances": [31, 26, 23, 26], "reading_log_probs": [-269.553955078125, -268.4941711425781, -269.4770812988281, -271.8819580078125], "reading_logprob_std": 1.4368743106000859, "tokenization_path_logprob_std": 360.0341073856448, "trajectory_balance_rmse": 4701.755126809155} {"canonical_length": 259, "canonical_nll": 1.3902996755474784, "generation_rouge_l_f1": 0.35871404399323176, "generation_token_f1": 0.622673434856176, "prompt_id": "cf2b1f1029bc9513", "reading_canonical_log_prob": -360.0876159667969, "reading_distances": [29, 30, 35, 24], "reading_log_probs": [-364.4342041015625, -364.1883850097656, -364.2034912109375, -357.0611267089844], "reading_logprob_std": 3.60887026947571, "tokenization_path_logprob_std": 81.91307633727577, "trajectory_balance_rmse": 4231.837614599876} {"canonical_length": 159, "canonical_nll": 1.7340613791027908, "generation_rouge_l_f1": 0.20578778135048229, "generation_token_f1": 0.4180064308681672, "prompt_id": "0c1e711645beb310", "reading_canonical_log_prob": -275.71575927734375, "reading_distances": [20, 22, 27, 25], "reading_log_probs": [-275.6651611328125, -272.28302001953125, -272.0450439453125, -278.0213623046875], "reading_logprob_std": 2.8693407612469812, "tokenization_path_logprob_std": 71.83462388270762, "trajectory_balance_rmse": 2970.3384199669404} {"canonical_length": 36, "canonical_nll": 0.8665847778320312, "generation_rouge_l_f1": 0.05882352941176471, "generation_token_f1": 0.05882352941176471, "prompt_id": "602cd10603b86b5c", "reading_canonical_log_prob": -31.197052001953125, "reading_distances": [11, 13, 14, 14], "reading_log_probs": [-32.907203674316406, -34.12369918823242, -32.20379638671875, -30.841293334960938], "reading_logprob_std": 1.371108377578137, "tokenization_path_logprob_std": 75.79167642403009, "trajectory_balance_rmse": 639.8720019307945} {"canonical_length": 8, "canonical_nll": 3.525554895401001, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "b36657ba534be993", "reading_canonical_log_prob": -28.204439163208008, "reading_distances": [31, 35, 29, 34], "reading_log_probs": [-28.059131622314453, -30.406356811523438, -35.550262451171875, -31.672988891601562], "reading_logprob_std": 3.1329411615169698, "tokenization_path_logprob_std": 11.16603902329296, "trajectory_balance_rmse": 137.85021757555836} {"canonical_length": 33, "canonical_nll": 1.7941849448464133, "generation_rouge_l_f1": 0.20833333333333334, "generation_token_f1": 0.3333333333333333, "prompt_id": "f11461f699664b03", "reading_canonical_log_prob": -59.20810317993164, "reading_distances": [21, 20, 20, 18], "reading_log_probs": [-70.75711059570312, -76.27420043945312, -64.71699523925781, -61.83344268798828], "reading_logprob_std": 6.435399978389081, "tokenization_path_logprob_std": 68.40398922002792, "trajectory_balance_rmse": 711.5351981604604} {"canonical_length": 323, "canonical_nll": 1.3379897362676567, "generation_rouge_l_f1": 0.2743055555555556, "generation_token_f1": 0.4756944444444445, "prompt_id": "167bb9079aab9f14", "reading_canonical_log_prob": -432.1706848144531, "reading_distances": [19, 26, 28, 23], "reading_log_probs": [-436.1304016113281, -436.20941162109375, -440.8200378417969, -438.72918701171875], "reading_logprob_std": 2.249655815766135, "tokenization_path_logprob_std": 197.27045068479808, "trajectory_balance_rmse": 7926.878197130099} {"canonical_length": 330, "canonical_nll": 1.1570654666785039, "generation_rouge_l_f1": 0.2163120567375887, "generation_token_f1": 0.4148936170212766, "prompt_id": "e95dc10521ada79e", "reading_canonical_log_prob": -381.83160400390625, "reading_distances": [39, 43, 31, 31], "reading_log_probs": [-389.6556396484375, -390.0495910644531, -389.7505798339844, -386.9276123046875], "reading_logprob_std": 1.455210446095304, "tokenization_path_logprob_std": 150.2061495269255, "trajectory_balance_rmse": 6152.860429709102} {"canonical_length": 126, "canonical_nll": 1.3159620496961806, "generation_rouge_l_f1": 0.21621621621621623, "generation_token_f1": 0.3903903903903904, "prompt_id": "78584ee870db48f1", "reading_canonical_log_prob": -165.81121826171875, "reading_distances": [32, 35, 34, 38], "reading_log_probs": [-173.23443603515625, -178.58267211914062, -178.85255432128906, -177.5316162109375], "reading_logprob_std": 2.6069629866750463, "tokenization_path_logprob_std": 133.06455205917703, "trajectory_balance_rmse": 2797.18818450851} {"canonical_length": 318, "canonical_nll": 1.337570478331368, "generation_rouge_l_f1": 0.22105263157894736, "generation_token_f1": 0.4350877192982456, "prompt_id": "16eff2d1d4178803", "reading_canonical_log_prob": -425.347412109375, "reading_distances": [17, 25, 25, 22], "reading_log_probs": [-424.1650695800781, -427.9237060546875, -426.88848876953125, -427.34613037109375], "reading_logprob_std": 1.6652880711538622, "tokenization_path_logprob_std": 230.8235984529618, "trajectory_balance_rmse": 6995.1001925671535} {"canonical_length": 104, "canonical_nll": 1.1579959575946515, "generation_rouge_l_f1": 0.1940700808625337, "generation_token_f1": 0.3881401617250674, "prompt_id": "22619d053543f675", "reading_canonical_log_prob": -120.43157958984375, "reading_distances": [43, 39, 43, 38], "reading_log_probs": [-115.36868286132812, -124.34019470214844, -123.14923858642578, -120.12690734863281], "reading_logprob_std": 3.999691918101589, "tokenization_path_logprob_std": 123.91988874658614, "trajectory_balance_rmse": 1995.4148411208973} {"canonical_length": 180, "canonical_nll": 1.1385631985134548, "generation_rouge_l_f1": 0.2815884476534296, "generation_token_f1": 0.4115523465703971, "prompt_id": "3c84043f06521168", "reading_canonical_log_prob": -204.94137573242188, "reading_distances": [20, 28, 25, 24], "reading_log_probs": [-202.93856811523438, -201.3363800048828, -201.61428833007812, -202.04766845703125], "reading_logprob_std": 0.7003218620226422, "tokenization_path_logprob_std": 21.87869420556396, "trajectory_balance_rmse": 3659.85263648736} {"canonical_length": 238, "canonical_nll": 1.613931094898897, "generation_rouge_l_f1": 0.28686868686868683, "generation_token_f1": 0.5131313131313131, "prompt_id": "4c550a924c0f2e5b", "reading_canonical_log_prob": -384.1156005859375, "reading_distances": [31, 22, 30, 26], "reading_log_probs": [-394.559814453125, -388.09747314453125, -395.2391052246094, -389.74151611328125], "reading_logprob_std": 3.528084491146399, "tokenization_path_logprob_std": 172.75199150264984, "trajectory_balance_rmse": 4575.580291601237} {"canonical_length": 72, "canonical_nll": 1.4642170800103083, "generation_rouge_l_f1": 0.1714285714285714, "generation_token_f1": 0.2476190476190476, "prompt_id": "9e6e21f46355104c", "reading_canonical_log_prob": -105.42362976074219, "reading_distances": [19, 13, 20, 21], "reading_log_probs": [-107.2712173461914, -107.81300354003906, -107.62059020996094, -103.69599914550781], "reading_logprob_std": 1.9490797392113108, "tokenization_path_logprob_std": 77.23422292981948, "trajectory_balance_rmse": 1258.7510477636404} {"canonical_length": 27, "canonical_nll": 1.7264340718587239, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.4666666666666667, "prompt_id": "4a624dc7ffd3988d", "reading_canonical_log_prob": -46.61371994018555, "reading_distances": [30, 26, 25, 36], "reading_log_probs": [-47.19640350341797, -43.7358512878418, -43.52900695800781, -46.53666687011719], "reading_logprob_std": 1.8884261995319769, "tokenization_path_logprob_std": 108.19970270627486, "trajectory_balance_rmse": 731.2521958415468} {"canonical_length": 7, "canonical_nll": 0.3458918843950544, "generation_rouge_l_f1": 0.8, "generation_token_f1": 0.8, "prompt_id": "0cd953773dcd678b", "reading_canonical_log_prob": -2.421243190765381, "reading_distances": [27, 29, 36, 31], "reading_log_probs": [-2.5037002563476562, -2.8000566959381104, -2.165121555328369, -2.745856523513794], "reading_logprob_std": 0.28931466844708204, "tokenization_path_logprob_std": 11.703934419226524, "trajectory_balance_rmse": 167.00530653499416} {"canonical_length": 165, "canonical_nll": 0.601030106977983, "generation_rouge_l_f1": 0.30319148936170215, "generation_token_f1": 0.5638297872340425, "prompt_id": "a86300073620913b", "reading_canonical_log_prob": -99.16996765136719, "reading_distances": [135, 135, 154, 137], "reading_log_probs": [-108.26213073730469, -116.9215087890625, -110.41272735595703, -112.21411895751953], "reading_logprob_std": 3.6855220279761958, "tokenization_path_logprob_std": 118.95528207113243, "trajectory_balance_rmse": 2756.2838094939275} {"canonical_length": 294, "canonical_nll": 1.4503802863919004, "generation_rouge_l_f1": 0.18563922942206654, "generation_token_f1": 0.3257443082311734, "prompt_id": "2837da3fee02c3ea", "reading_canonical_log_prob": -426.41180419921875, "reading_distances": [32, 24, 25, 27], "reading_log_probs": [-436.41546630859375, -437.68731689453125, -433.27142333984375, -433.33746337890625], "reading_logprob_std": 2.2249047717064556, "tokenization_path_logprob_std": 216.16120256926402, "trajectory_balance_rmse": 5690.529878042734} {"canonical_length": 23, "canonical_nll": 1.2008044201394785, "generation_rouge_l_f1": 0.12903225806451613, "generation_token_f1": 0.13620071684587814, "prompt_id": "8041588a81157d9c", "reading_canonical_log_prob": -27.618501663208008, "reading_distances": [54, 53, 49, 37], "reading_log_probs": [-29.01919937133789, -30.336790084838867, -26.888057708740234, -30.99832534790039], "reading_logprob_std": 1.8124668658401268, "tokenization_path_logprob_std": 100.58956968565373, "trajectory_balance_rmse": 554.4140014035759} {"canonical_length": 308, "canonical_nll": 0.8722415280032467, "generation_rouge_l_f1": 0.25147347740667975, "generation_token_f1": 0.44793713163064836, "prompt_id": "bc95295408e3badf", "reading_canonical_log_prob": -268.650390625, "reading_distances": [30, 32, 28, 32], "reading_log_probs": [-289.4039001464844, -299.83056640625, -288.1817626953125, -291.0349426269531], "reading_logprob_std": 5.276264067345901, "tokenization_path_logprob_std": 155.21958186667365, "trajectory_balance_rmse": 5264.030470980987} {"canonical_length": 126, "canonical_nll": 1.393372793046255, "generation_rouge_l_f1": 0.3403141361256545, "generation_token_f1": 0.4764397905759162, "prompt_id": "2a5a99cace11afef", "reading_canonical_log_prob": -175.56497192382812, "reading_distances": [27, 28, 22, 22], "reading_log_probs": [-178.6173858642578, -178.66322326660156, -179.42445373535156, -178.04864501953125], "reading_logprob_std": 0.5647236518361736, "tokenization_path_logprob_std": 162.55163282340433, "trajectory_balance_rmse": 2758.8151251997638} {"canonical_length": 14, "canonical_nll": 0.5289716720581055, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4727272727272727, "prompt_id": "90db691d0627dd6d", "reading_canonical_log_prob": -7.405603408813477, "reading_distances": [16, 21, 16, 22], "reading_log_probs": [-7.451227188110352, -8.107370376586914, -7.620997428894043, -7.538363456726074], "reading_logprob_std": 0.2935549977149999, "tokenization_path_logprob_std": 16.987213848766512, "trajectory_balance_rmse": 277.25816698410443} {"canonical_length": 14, "canonical_nll": 0.9777494158063617, "generation_rouge_l_f1": 0.5454545454545454, "generation_token_f1": 0.5454545454545454, "prompt_id": "07c0c6eb902aa919", "reading_canonical_log_prob": -13.688491821289062, "reading_distances": [26, 26, 23, 27], "reading_log_probs": [-13.458626747131348, -13.760231971740723, -14.28122329711914, -14.379498481750488], "reading_logprob_std": 0.4359105920022971, "tokenization_path_logprob_std": 40.09436082530594, "trajectory_balance_rmse": 293.23662378812827} {"canonical_length": 6, "canonical_nll": 0.8778841495513916, "generation_rouge_l_f1": 0.42857142857142855, "generation_token_f1": 0.42857142857142855, "prompt_id": "285daef948799d4d", "reading_canonical_log_prob": -5.26730489730835, "reading_distances": [43, 43, 52, 50], "reading_log_probs": [-6.455638885498047, -5.703510761260986, -5.690751075744629, -6.175311088562012], "reading_logprob_std": 0.37493227332303675, "tokenization_path_logprob_std": 16.884600138516756, "trajectory_balance_rmse": 114.21494967595869} {"canonical_length": 68, "canonical_nll": 0.6518881741692039, "generation_rouge_l_f1": 0.40625, "generation_token_f1": 0.5625, "prompt_id": "924060b32d38a1e4", "reading_canonical_log_prob": -44.32839584350586, "reading_distances": [34, 30, 33, 36], "reading_log_probs": [-45.69954299926758, -44.8997802734375, -45.18071746826172, -43.77649688720703], "reading_logprob_std": 0.8123747807014376, "tokenization_path_logprob_std": 85.86341448960054, "trajectory_balance_rmse": 1112.4671526199124} {"canonical_length": 147, "canonical_nll": 0.9794524316074086, "generation_rouge_l_f1": 0.419047619047619, "generation_token_f1": 0.5968253968253969, "prompt_id": "ac4b8108300a4ae5", "reading_canonical_log_prob": -143.97950744628906, "reading_distances": [32, 32, 30, 29], "reading_log_probs": [-147.22552490234375, -150.10861206054688, -149.55430603027344, -149.86270141601562], "reading_logprob_std": 1.3276843249965373, "tokenization_path_logprob_std": 69.89883229624662, "trajectory_balance_rmse": 2148.9676128928722} {"canonical_length": 53, "canonical_nll": 1.8604744965175413, "generation_rouge_l_f1": 0.1943573667711599, "generation_token_f1": 0.2445141065830721, "prompt_id": "10e2361fe555104f", "reading_canonical_log_prob": -98.60514831542969, "reading_distances": [25, 24, 18, 28], "reading_log_probs": [-114.04329681396484, -106.96611022949219, -103.89007568359375, -101.98916625976562], "reading_logprob_std": 5.2940371262887975, "tokenization_path_logprob_std": 78.93616990653666, "trajectory_balance_rmse": 984.4634723139186} {"canonical_length": 27, "canonical_nll": 1.1910028810854312, "generation_rouge_l_f1": 0.148471615720524, "generation_token_f1": 0.1572052401746725, "prompt_id": "e53f2608a52ca6d9", "reading_canonical_log_prob": -32.15707778930664, "reading_distances": [24, 33, 32, 36], "reading_log_probs": [-39.526283264160156, -43.66931915283203, -42.46046447753906, -44.18619918823242], "reading_logprob_std": 2.0855781987228426, "tokenization_path_logprob_std": 52.30715699037312, "trajectory_balance_rmse": 397.7199893302772} {"canonical_length": 18, "canonical_nll": 0.8452784750196669, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "3f4c8b5b47d9a555", "reading_canonical_log_prob": -15.215012550354004, "reading_distances": [17, 19, 15, 16], "reading_log_probs": [-15.580632209777832, -15.477836608886719, -16.513324737548828, -16.37799072265625], "reading_logprob_std": 0.5336268416619591, "tokenization_path_logprob_std": 65.51031436004445, "trajectory_balance_rmse": 386.99343818955055} {"canonical_length": 11, "canonical_nll": 0.5993902033025568, "generation_rouge_l_f1": 0.19512195121951217, "generation_token_f1": 0.29268292682926833, "prompt_id": "94a2d875be1c7a06", "reading_canonical_log_prob": -6.593292236328125, "reading_distances": [17, 16, 13, 17], "reading_log_probs": [-8.326345443725586, -6.435762882232666, -7.577872276306152, -6.43245267868042], "reading_logprob_std": 0.9281592531170676, "tokenization_path_logprob_std": 37.74590493823276, "trajectory_balance_rmse": 226.16813336752085} {"canonical_length": 18, "canonical_nll": 0.3636471430460612, "generation_rouge_l_f1": 0.2272727272727273, "generation_token_f1": 0.25000000000000006, "prompt_id": "a86c032862f0a2ee", "reading_canonical_log_prob": -6.545648574829102, "reading_distances": [17, 17, 15, 14], "reading_log_probs": [-6.191745758056641, -6.514629364013672, -6.449068069458008, -5.878154754638672], "reading_logprob_std": 0.28927324699528206, "tokenization_path_logprob_std": 47.567464206920775, "trajectory_balance_rmse": 275.80740566931485} {"canonical_length": 80, "canonical_nll": 1.3145111083984375, "generation_rouge_l_f1": 0.23030303030303031, "generation_token_f1": 0.3151515151515151, "prompt_id": "26b8508f97bd840b", "reading_canonical_log_prob": -105.160888671875, "reading_distances": [33, 29, 38, 28], "reading_log_probs": [-114.7528076171875, -114.50819396972656, -117.74787902832031, -118.25892639160156], "reading_logprob_std": 1.96103458752229, "tokenization_path_logprob_std": 68.43084033613115, "trajectory_balance_rmse": 1939.4471341300161} {"canonical_length": 282, "canonical_nll": 0.900173971838985, "generation_rouge_l_f1": 0.35095137420718814, "generation_token_f1": 0.4820295983086681, "prompt_id": "d2afb79833a3e8c4", "reading_canonical_log_prob": -253.84906005859375, "reading_distances": [21, 26, 22, 22], "reading_log_probs": [-257.5771789550781, -259.1357421875, -261.27301025390625, -260.5921325683594], "reading_logprob_std": 1.6414123943753407, "tokenization_path_logprob_std": 207.84718950141735, "trajectory_balance_rmse": 6075.774928999473} {"canonical_length": 206, "canonical_nll": 1.0510346495989458, "generation_rouge_l_f1": 0.2703862660944206, "generation_token_f1": 0.45064377682403434, "prompt_id": "8d6230df8e48bdfc", "reading_canonical_log_prob": -216.5131378173828, "reading_distances": [14, 14, 12, 17], "reading_log_probs": [-212.60421752929688, -210.42034912109375, -214.7750244140625, -214.35159301757812], "reading_logprob_std": 1.9818378356277129, "tokenization_path_logprob_std": 121.83040073758399, "trajectory_balance_rmse": 3924.955459449486} {"canonical_length": 14, "canonical_nll": 3.361528124128069, "generation_rouge_l_f1": 0.27586206896551724, "generation_token_f1": 0.27586206896551724, "prompt_id": "74ebf94d0f403056", "reading_canonical_log_prob": -47.06139373779297, "reading_distances": [22, 24, 30, 26], "reading_log_probs": [-57.03080368041992, -54.6933708190918, -61.09722137451172, -50.61948776245117], "reading_logprob_std": 4.382665130799198, "tokenization_path_logprob_std": 15.989453382571044, "trajectory_balance_rmse": 260.9301778688126} {"canonical_length": 129, "canonical_nll": 1.158163026321766, "generation_rouge_l_f1": 0.28418230563002683, "generation_token_f1": 0.47184986595174255, "prompt_id": "c307d4b735027f71", "reading_canonical_log_prob": -149.4030303955078, "reading_distances": [18, 17, 16, 15], "reading_log_probs": [-151.55075073242188, -153.28890991210938, -153.65670776367188, -151.8795623779297], "reading_logprob_std": 1.034575231008758, "tokenization_path_logprob_std": 218.50727397179207, "trajectory_balance_rmse": 2580.7473676611366} {"canonical_length": 73, "canonical_nll": 0.8659822124324433, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.4117647058823529, "prompt_id": "1044800b3405ca7b", "reading_canonical_log_prob": -63.21670150756836, "reading_distances": [12, 13, 13, 12], "reading_log_probs": [-72.13866424560547, -71.8779296875, -67.63488006591797, -71.80648803710938], "reading_logprob_std": 2.1578013993107725, "tokenization_path_logprob_std": 109.19188044264737, "trajectory_balance_rmse": 1317.271933944333} {"canonical_length": 205, "canonical_nll": 1.511232850609756, "generation_rouge_l_f1": 0.175, "generation_token_f1": 0.42083333333333334, "prompt_id": "c87b3b9596ed8821", "reading_canonical_log_prob": -309.802734375, "reading_distances": [30, 29, 32, 27], "reading_log_probs": [-322.97479248046875, -320.61248779296875, -324.12677001953125, -322.6510314941406], "reading_logprob_std": 1.4633295666675088, "tokenization_path_logprob_std": 77.0781841632279, "trajectory_balance_rmse": 4766.6800301807825} {"canonical_length": 11, "canonical_nll": 1.0908335772427646, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.18181818181818182, "prompt_id": "f2f04e5d3550740c", "reading_canonical_log_prob": -11.99916934967041, "reading_distances": [19, 11, 13, 17], "reading_log_probs": [-17.711563110351562, -13.778355598449707, -16.35578155517578, -17.91439437866211], "reading_logprob_std": 1.9045695219612127, "tokenization_path_logprob_std": 54.42895850649896, "trajectory_balance_rmse": 197.4518197401619} {"canonical_length": 149, "canonical_nll": 1.420266170629719, "generation_rouge_l_f1": 0.2121212121212121, "generation_token_f1": 0.34848484848484845, "prompt_id": "e64143ec52b6e7c2", "reading_canonical_log_prob": -211.61965942382812, "reading_distances": [20, 18, 16, 15], "reading_log_probs": [-215.80148315429688, -215.19046020507812, -216.06149291992188, -219.46627807617188], "reading_logprob_std": 1.9258235112847661, "tokenization_path_logprob_std": 107.81110648033226, "trajectory_balance_rmse": 2829.8912149069924} {"canonical_length": 48, "canonical_nll": 1.1979846954345703, "generation_rouge_l_f1": 0.30357142857142855, "generation_token_f1": 0.375, "prompt_id": "c95d33d4bb11096d", "reading_canonical_log_prob": -57.503265380859375, "reading_distances": [18, 26, 25, 24], "reading_log_probs": [-62.77712631225586, -62.59103775024414, -61.16206359863281, -63.272918701171875], "reading_logprob_std": 0.9060677235487009, "tokenization_path_logprob_std": 138.13818463997237, "trajectory_balance_rmse": 1194.1789335474161} {"canonical_length": 211, "canonical_nll": 1.3803867141217416, "generation_rouge_l_f1": 0.19555555555555554, "generation_token_f1": 0.46666666666666673, "prompt_id": "0accd9d43b5400d9", "reading_canonical_log_prob": -291.2615966796875, "reading_distances": [23, 20, 20, 30], "reading_log_probs": [-290.6258850097656, -293.39739990234375, -293.49029541015625, -291.794921875], "reading_logprob_std": 1.3754891784720564, "tokenization_path_logprob_std": 174.23758292156413, "trajectory_balance_rmse": 4462.900075555778} {"canonical_length": 28, "canonical_nll": 0.8952695301600865, "generation_rouge_l_f1": 0.1107011070110701, "generation_token_f1": 0.15498154981549814, "prompt_id": "c5287ae056861247", "reading_canonical_log_prob": -25.067546844482422, "reading_distances": [19, 16, 23, 19], "reading_log_probs": [-21.566503524780273, -25.329166412353516, -23.64122772216797, -24.493770599365234], "reading_logprob_std": 1.615158856231064, "tokenization_path_logprob_std": 42.102617184585945, "trajectory_balance_rmse": 630.5388514147099} {"canonical_length": 9, "canonical_nll": 1.1699852413601346, "generation_rouge_l_f1": 0.08045977011494253, "generation_token_f1": 0.08045977011494253, "prompt_id": "9ba45d22569ee03c", "reading_canonical_log_prob": -10.529867172241211, "reading_distances": [41, 44, 43, 46], "reading_log_probs": [-20.399234771728516, -13.407402038574219, -12.727315902709961, -15.131816864013672], "reading_logprob_std": 3.4726011388082285, "tokenization_path_logprob_std": 35.579853161872165, "trajectory_balance_rmse": 223.02708071599167} {"canonical_length": 9, "canonical_nll": 0.8203980657789443, "generation_rouge_l_f1": 0.3529411764705882, "generation_token_f1": 0.47058823529411764, "prompt_id": "61752b6441afd01d", "reading_canonical_log_prob": -7.383582592010498, "reading_distances": [20, 20, 16, 23], "reading_log_probs": [-8.464971542358398, -7.608294486999512, -7.964993000030518, -7.3055267333984375], "reading_logprob_std": 0.4984964295789649, "tokenization_path_logprob_std": 42.92336739605415, "trajectory_balance_rmse": 253.70111729114086} {"canonical_length": 147, "canonical_nll": 1.0339127105920494, "generation_rouge_l_f1": 0.3107769423558897, "generation_token_f1": 0.47619047619047616, "prompt_id": "1c1ecaad77fec061", "reading_canonical_log_prob": -151.98516845703125, "reading_distances": [14, 14, 17, 16], "reading_log_probs": [-153.93331909179688, -152.77552795410156, -157.30714416503906, -153.92623901367188], "reading_logprob_std": 1.958175748395682, "tokenization_path_logprob_std": 107.75683739583125, "trajectory_balance_rmse": 2767.558582320793} {"canonical_length": 58, "canonical_nll": 0.9742117256953798, "generation_rouge_l_f1": 0.16201117318435757, "generation_token_f1": 0.20670391061452512, "prompt_id": "697abdbcde745481", "reading_canonical_log_prob": -56.50428009033203, "reading_distances": [21, 25, 17, 22], "reading_log_probs": [-58.5679931640625, -63.848663330078125, -55.62532043457031, -59.28302001953125], "reading_logprob_std": 3.4022455460725944, "tokenization_path_logprob_std": 45.27500765493841, "trajectory_balance_rmse": 956.9592316010513} {"canonical_length": 338, "canonical_nll": 1.2606715817423262, "generation_rouge_l_f1": 0.12376237623762376, "generation_token_f1": 0.21287128712871287, "prompt_id": "4034832cbc1502c1", "reading_canonical_log_prob": -426.10699462890625, "reading_distances": [29, 20, 22, 17], "reading_log_probs": [-439.406494140625, -439.8609924316406, -436.83282470703125, -438.7072448730469], "reading_logprob_std": 1.3333440648813684, "tokenization_path_logprob_std": 82.26010039225535, "trajectory_balance_rmse": 7138.5383122907915} {"canonical_length": 13, "canonical_nll": 2.2884081326998196, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "228b1bd7f7bc8ed3", "reading_canonical_log_prob": -29.749305725097656, "reading_distances": [14, 16, 14, 15], "reading_log_probs": [-24.87112808227539, -24.167402267456055, -22.843124389648438, -23.83719825744629], "reading_logprob_std": 0.8430089621379374, "tokenization_path_logprob_std": 23.40406162242526, "trajectory_balance_rmse": 207.42715084371977} {"canonical_length": 115, "canonical_nll": 1.1556254511294157, "generation_rouge_l_f1": 0.33098591549295775, "generation_token_f1": 0.5, "prompt_id": "28e1d60f144a6d02", "reading_canonical_log_prob": -132.8969268798828, "reading_distances": [20, 24, 27, 23], "reading_log_probs": [-132.28399658203125, -132.07952880859375, -135.0620880126953, -132.7393341064453], "reading_logprob_std": 1.375170276568405, "tokenization_path_logprob_std": 90.54445560899504, "trajectory_balance_rmse": 2292.7497516567882} {"canonical_length": 131, "canonical_nll": 1.363728064617128, "generation_rouge_l_f1": 0.27485380116959063, "generation_token_f1": 0.4736842105263158, "prompt_id": "f38548affd15660d", "reading_canonical_log_prob": -178.64837646484375, "reading_distances": [27, 26, 30, 26], "reading_log_probs": [-180.80523681640625, -176.3970489501953, -178.00308227539062, -174.53970336914062], "reading_logprob_std": 2.6546369621806236, "tokenization_path_logprob_std": 192.55694624198637, "trajectory_balance_rmse": 2796.97704313355} {"canonical_length": 39, "canonical_nll": 1.2263014377691808, "generation_rouge_l_f1": 0.19753086419753088, "generation_token_f1": 0.4197530864197531, "prompt_id": "b15afd8e011cefe5", "reading_canonical_log_prob": -47.82575607299805, "reading_distances": [20, 19, 30, 25], "reading_log_probs": [-51.524600982666016, -51.330806732177734, -52.63684844970703, -52.183380126953125], "reading_logprob_std": 0.6018651898858578, "tokenization_path_logprob_std": 92.04653230048392, "trajectory_balance_rmse": 673.008083502629} {"canonical_length": 286, "canonical_nll": 0.9092558747404939, "generation_rouge_l_f1": 0.2648401826484018, "generation_token_f1": 0.3835616438356164, "prompt_id": "60d1476fd22fd625", "reading_canonical_log_prob": -260.04718017578125, "reading_distances": [17, 24, 20, 22], "reading_log_probs": [-266.15863037109375, -276.21734619140625, -263.0412902832031, -264.3773193359375], "reading_logprob_std": 5.983646470766595, "tokenization_path_logprob_std": 135.62147440627513, "trajectory_balance_rmse": 4770.862875094985} {"canonical_length": 58, "canonical_nll": 1.6545857396619073, "generation_rouge_l_f1": 0.10408921933085502, "generation_token_f1": 0.20817843866171004, "prompt_id": "f1aff688c2307612", "reading_canonical_log_prob": -95.96597290039062, "reading_distances": [35, 51, 43, 44], "reading_log_probs": [-95.47395324707031, -87.27033996582031, -92.68405151367188, -97.27096557617188], "reading_logprob_std": 4.365323983774336, "tokenization_path_logprob_std": 69.86287255075645, "trajectory_balance_rmse": 1301.998201079532} {"canonical_length": 123, "canonical_nll": 1.538190516029916, "generation_rouge_l_f1": 0.22399999999999998, "generation_token_f1": 0.39466666666666667, "prompt_id": "a3f93481aca6425b", "reading_canonical_log_prob": -189.1974334716797, "reading_distances": [24, 22, 23, 28], "reading_log_probs": [-187.34500122070312, -185.45596313476562, -188.9951171875, -190.49948120117188], "reading_logprob_std": 2.1692527231353558, "tokenization_path_logprob_std": 156.69741989542953, "trajectory_balance_rmse": 2426.3555591200043} {"canonical_length": 163, "canonical_nll": 1.2178735089448331, "generation_rouge_l_f1": 0.1953488372093023, "generation_token_f1": 0.3627906976744186, "prompt_id": "39b8bde022351616", "reading_canonical_log_prob": -198.5133819580078, "reading_distances": [17, 22, 26, 16], "reading_log_probs": [-200.46072387695312, -202.41177368164062, -205.14442443847656, -203.30978393554688], "reading_logprob_std": 1.947230894069131, "tokenization_path_logprob_std": 129.42857402012777, "trajectory_balance_rmse": 3491.643030022174} {"canonical_length": 56, "canonical_nll": 1.5338856833321708, "generation_rouge_l_f1": 0.14062500000000003, "generation_token_f1": 0.21093750000000003, "prompt_id": "969cc5c017329874", "reading_canonical_log_prob": -85.89759826660156, "reading_distances": [32, 31, 33, 33], "reading_log_probs": [-103.0071029663086, -104.88673400878906, -92.943359375, -97.02703857421875], "reading_logprob_std": 5.489904096235616, "tokenization_path_logprob_std": 66.1065448331786, "trajectory_balance_rmse": 819.7327172679245} {"canonical_length": 262, "canonical_nll": 1.565566084767116, "generation_rouge_l_f1": 0.27392120075046905, "generation_token_f1": 0.46904315196998125, "prompt_id": "a4cec9116b07e32b", "reading_canonical_log_prob": -410.1783142089844, "reading_distances": [12, 20, 13, 16], "reading_log_probs": [-421.56500244140625, -423.1644287109375, -416.5032958984375, -417.84027099609375], "reading_logprob_std": 3.1165888134912607, "tokenization_path_logprob_std": 229.48992211557805, "trajectory_balance_rmse": 5229.270641482039} {"canonical_length": 412, "canonical_nll": 0.9121857430171041, "generation_rouge_l_f1": 0.1721664275466284, "generation_token_f1": 0.4218077474892396, "prompt_id": "63a1199c25b071e9", "reading_canonical_log_prob": -375.8205261230469, "reading_distances": [209, 221, 217, 245], "reading_log_probs": [-390.619384765625, -390.1912841796875, -385.1368408203125, -388.77410888671875], "reading_logprob_std": 2.490517077574216, "tokenization_path_logprob_std": 129.14720507414427, "trajectory_balance_rmse": 6942.525941543411} {"canonical_length": 34, "canonical_nll": 1.5277252197265625, "generation_rouge_l_f1": 0.10596026490066225, "generation_token_f1": 0.1456953642384106, "prompt_id": "85280303ae668b99", "reading_canonical_log_prob": -51.942657470703125, "reading_distances": [10, 14, 10, 15], "reading_log_probs": [-57.40237808227539, -51.05070495605469, -47.643611907958984, -58.24309539794922], "reading_logprob_std": 5.098789790347507, "tokenization_path_logprob_std": 44.25174459847228, "trajectory_balance_rmse": 623.3987759841418} {"canonical_length": 18, "canonical_nll": 0.855766773223877, "generation_rouge_l_f1": 0.4117647058823529, "generation_token_f1": 0.4117647058823529, "prompt_id": "963484b861b43113", "reading_canonical_log_prob": -15.403801918029785, "reading_distances": [15, 15, 17, 12], "reading_log_probs": [-17.10584831237793, -17.312314987182617, -18.696260452270508, -19.269615173339844], "reading_logprob_std": 1.0539208934719482, "tokenization_path_logprob_std": 31.07106155941228, "trajectory_balance_rmse": 333.21897622442225} {"canonical_length": 242, "canonical_nll": 1.1544376089553203, "generation_rouge_l_f1": 0.18823529411764703, "generation_token_f1": 0.4745098039215686, "prompt_id": "edfd3e88db49ba9c", "reading_canonical_log_prob": -279.3739013671875, "reading_distances": [24, 28, 23, 31], "reading_log_probs": [-292.0556640625, -292.2870178222656, -292.05841064453125, -298.92816162109375], "reading_logprob_std": 3.398961673280459, "tokenization_path_logprob_std": 165.41325411797752, "trajectory_balance_rmse": 5531.374173386513} {"canonical_length": 3, "canonical_nll": 3.588791529337565, "generation_rouge_l_f1": 0.024242424242424246, "generation_token_f1": 0.024242424242424246, "prompt_id": "315bed08881fe880", "reading_canonical_log_prob": -10.766374588012695, "reading_distances": [31, 32, 31, 29], "reading_log_probs": [-13.726205825805664, -13.063669204711914, -13.060636520385742, -13.21189022064209], "reading_logprob_std": 0.3150812282704635, "tokenization_path_logprob_std": 21.18395759633128, "trajectory_balance_rmse": 85.43282714365078} {"canonical_length": 40, "canonical_nll": 1.3783669471740723, "generation_rouge_l_f1": 0.2028985507246377, "generation_token_f1": 0.2898550724637681, "prompt_id": "ec3686b4d6045b07", "reading_canonical_log_prob": -55.13467788696289, "reading_distances": [20, 30, 28, 25], "reading_log_probs": [-54.22146987915039, -53.30556869506836, -51.624908447265625, -51.38795471191406], "reading_logprob_std": 1.3591610090157011, "tokenization_path_logprob_std": 108.60655478260207, "trajectory_balance_rmse": 1065.5111021722958} {"canonical_length": 266, "canonical_nll": 1.5498584948087994, "generation_rouge_l_f1": 0.19083969465648856, "generation_token_f1": 0.4312977099236641, "prompt_id": "a1b8dba044d5aaa9", "reading_canonical_log_prob": -412.2623596191406, "reading_distances": [26, 27, 19, 23], "reading_log_probs": [-422.3701171875, -418.14501953125, -421.4830322265625, -418.58563232421875], "reading_logprob_std": 2.0954731008110543, "tokenization_path_logprob_std": 127.18945389191109, "trajectory_balance_rmse": 5996.334443253505} {"canonical_length": 154, "canonical_nll": 1.5067697500253652, "generation_rouge_l_f1": 0.3018867924528302, "generation_token_f1": 0.47641509433962265, "prompt_id": "e45ec8d7efaf0407", "reading_canonical_log_prob": -232.04254150390625, "reading_distances": [22, 15, 16, 15], "reading_log_probs": [-242.37808227539062, -245.14404296875, -245.9573974609375, -241.37063598632812], "reading_logprob_std": 2.187378765972982, "tokenization_path_logprob_std": 75.1354046421014, "trajectory_balance_rmse": 3850.7842070857937} {"canonical_length": 3, "canonical_nll": 2.2104438145955405, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "3521244aca15fc91", "reading_canonical_log_prob": -6.631331443786621, "reading_distances": [49, 49, 46, 48], "reading_log_probs": [-4.011022090911865, -4.618602752685547, -4.7353410720825195, -4.680583953857422], "reading_logprob_std": 0.33696849282329294, "tokenization_path_logprob_std": 16.758153449989692, "trajectory_balance_rmse": 63.946337758157206} {"canonical_length": 23, "canonical_nll": 0.8882930589758832, "generation_rouge_l_f1": 0.10752688172043011, "generation_token_f1": 0.13978494623655915, "prompt_id": "99de541f6c007900", "reading_canonical_log_prob": -20.430740356445312, "reading_distances": [26, 23, 26, 26], "reading_log_probs": [-24.762699127197266, -22.083892822265625, -23.745019912719727, -27.010770797729492], "reading_logprob_std": 2.060827080677889, "tokenization_path_logprob_std": 72.29174760928969, "trajectory_balance_rmse": 265.56848580653053} {"canonical_length": 176, "canonical_nll": 1.2882652282714844, "generation_rouge_l_f1": 0.2556053811659193, "generation_token_f1": 0.48430493273542596, "prompt_id": "124ab50eaad8157c", "reading_canonical_log_prob": -226.73468017578125, "reading_distances": [18, 19, 23, 25], "reading_log_probs": [-235.5067138671875, -229.06512451171875, -227.29220581054688, -230.04446411132812], "reading_logprob_std": 3.541252849852191, "tokenization_path_logprob_std": 209.4860749918217, "trajectory_balance_rmse": 4096.361179898095} {"canonical_length": 15, "canonical_nll": 0.3924101829528809, "generation_rouge_l_f1": 0.8888888888888888, "generation_token_f1": 0.8888888888888888, "prompt_id": "7efa869db692612b", "reading_canonical_log_prob": -5.886152744293213, "reading_distances": [48, 52, 54, 51], "reading_log_probs": [-16.46129035949707, -12.773311614990234, -20.264516830444336, -17.784196853637695], "reading_logprob_std": 3.1250988523535206, "tokenization_path_logprob_std": 34.63997265695272, "trajectory_balance_rmse": 216.4280234958604} {"canonical_length": 248, "canonical_nll": 1.384886710874496, "generation_rouge_l_f1": 0.25806451612903225, "generation_token_f1": 0.5120967741935484, "prompt_id": "b9153e23659ffc43", "reading_canonical_log_prob": -343.451904296875, "reading_distances": [30, 27, 29, 36], "reading_log_probs": [-338.96905517578125, -341.96026611328125, -339.9527282714844, -340.5910339355469], "reading_logprob_std": 1.2536018038876544, "tokenization_path_logprob_std": 230.90975485514977, "trajectory_balance_rmse": 6026.507636285862} {"canonical_length": 48, "canonical_nll": 0.8331316312154134, "generation_rouge_l_f1": 0.18237082066869298, "generation_token_f1": 0.2127659574468085, "prompt_id": "2f3de773aa866413", "reading_canonical_log_prob": -39.990318298339844, "reading_distances": [23, 33, 25, 29], "reading_log_probs": [-45.14973831176758, -50.5255126953125, -44.7657470703125, -44.77503967285156], "reading_logprob_std": 2.8200136665706386, "tokenization_path_logprob_std": 90.50631787990659, "trajectory_balance_rmse": 708.2126260638506} {"canonical_length": 301, "canonical_nll": 1.4493009751025228, "generation_rouge_l_f1": 0.16347826086956518, "generation_token_f1": 0.2747826086956522, "prompt_id": "54066b815b8806cf", "reading_canonical_log_prob": -436.2395935058594, "reading_distances": [32, 37, 45, 40], "reading_log_probs": [-453.16241455078125, -450.9010009765625, -456.84228515625, -455.0323486328125], "reading_logprob_std": 2.5461548440291324, "tokenization_path_logprob_std": 291.23591065143063, "trajectory_balance_rmse": 6122.770250789534} {"canonical_length": 394, "canonical_nll": 1.3647634438451777, "generation_rouge_l_f1": 0.2212121212121212, "generation_token_f1": 0.46060606060606063, "prompt_id": "38de2822dc803e85", "reading_canonical_log_prob": -537.716796875, "reading_distances": [26, 31, 38, 36], "reading_log_probs": [-537.41552734375, -540.78271484375, -537.6578369140625, -539.5496826171875], "reading_logprob_std": 1.6024875202526176, "tokenization_path_logprob_std": 254.0538600832025, "trajectory_balance_rmse": 7873.035831992829} {"canonical_length": 329, "canonical_nll": 1.0092309644519377, "generation_rouge_l_f1": 0.19410745233968807, "generation_token_f1": 0.3951473136915078, "prompt_id": "93d5783f4b9e5595", "reading_canonical_log_prob": -332.0369873046875, "reading_distances": [32, 31, 30, 32], "reading_log_probs": [-344.82244873046875, -345.8948974609375, -349.952392578125, -350.0054626464844], "reading_logprob_std": 2.7032840000017875, "tokenization_path_logprob_std": 237.1919966084242, "trajectory_balance_rmse": 7417.0995859355735} {"canonical_length": 25, "canonical_nll": 0.9644075012207032, "generation_rouge_l_f1": 0.08759124087591241, "generation_token_f1": 0.1240875912408759, "prompt_id": "fffceba44eca0753", "reading_canonical_log_prob": -24.110187530517578, "reading_distances": [27, 31, 30, 28], "reading_log_probs": [-21.670976638793945, -21.90738296508789, -23.360702514648438, -22.08664894104004], "reading_logprob_std": 0.7556085550287638, "tokenization_path_logprob_std": 70.42791564466124, "trajectory_balance_rmse": 382.13440870148577} {"canonical_length": 49, "canonical_nll": 2.1057222327407525, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.3333333333333333, "prompt_id": "4d1f490ef0029856", "reading_canonical_log_prob": -103.18038940429688, "reading_distances": [21, 21, 25, 25], "reading_log_probs": [-102.76264190673828, -104.96815490722656, -90.43221282958984, -101.48043823242188], "reading_logprob_std": 6.481183529548212, "tokenization_path_logprob_std": 127.08614852238408, "trajectory_balance_rmse": 1197.7184710647796} {"canonical_length": 58, "canonical_nll": 1.4556013962318157, "generation_rouge_l_f1": 0.25503355704697983, "generation_token_f1": 0.4295302013422819, "prompt_id": "04c2c0f3f6e1a037", "reading_canonical_log_prob": -84.42488098144531, "reading_distances": [11, 14, 10, 10], "reading_log_probs": [-92.43074035644531, -89.01815795898438, -91.40135192871094, -89.10472106933594], "reading_logprob_std": 1.7012099475198426, "tokenization_path_logprob_std": 97.4432157478516, "trajectory_balance_rmse": 1303.4552415894966} {"canonical_length": 68, "canonical_nll": 0.47962525311638327, "generation_rouge_l_f1": 0.3815789473684211, "generation_token_f1": 0.5, "prompt_id": "42b3ed92d904a820", "reading_canonical_log_prob": -32.61451721191406, "reading_distances": [10, 11, 11, 14], "reading_log_probs": [-33.55376434326172, -38.37251663208008, -34.909358978271484, -32.54035949707031], "reading_logprob_std": 2.544678698484569, "tokenization_path_logprob_std": 35.14384898164102, "trajectory_balance_rmse": 1284.1761163225417} {"canonical_length": 4, "canonical_nll": 0.4104052186012268, "generation_rouge_l_f1": 0.06666666666666667, "generation_token_f1": 0.06666666666666667, "prompt_id": "a0febc02441ecdc9", "reading_canonical_log_prob": -1.6416208744049072, "reading_distances": [48, 47, 48, 41], "reading_log_probs": [-4.994021892547607, -4.561961650848389, -4.950193405151367, -4.128662586212158], "reading_logprob_std": 0.40319852210756835, "tokenization_path_logprob_std": 29.56131696531379, "trajectory_balance_rmse": 114.790157293899} {"canonical_length": 84, "canonical_nll": 2.111334664481027, "generation_rouge_l_f1": 0.13569321533923304, "generation_token_f1": 0.20058997050147492, "prompt_id": "95d007f697f6007d", "reading_canonical_log_prob": -177.35211181640625, "reading_distances": [28, 23, 21, 24], "reading_log_probs": [-183.89244079589844, -182.87899780273438, -182.446533203125, -179.65292358398438], "reading_logprob_std": 1.8140667218596302, "tokenization_path_logprob_std": 110.39464258083274, "trajectory_balance_rmse": 1647.5000294899266} {"canonical_length": 178, "canonical_nll": 1.1330401817064606, "generation_rouge_l_f1": 0.2582417582417582, "generation_token_f1": 0.5054945054945055, "prompt_id": "42731119470564d3", "reading_canonical_log_prob": -201.68115234375, "reading_distances": [57, 55, 51, 60], "reading_log_probs": [-223.49964904785156, -225.76760864257812, -224.1773223876953, -225.04074096679688], "reading_logprob_std": 0.9908197285912812, "tokenization_path_logprob_std": 179.4518109929475, "trajectory_balance_rmse": 2463.869890401559} {"canonical_length": 302, "canonical_nll": 1.2495818485487376, "generation_rouge_l_f1": 0.24455611390284757, "generation_token_f1": 0.48576214405360135, "prompt_id": "f73070b2f6b0ff25", "reading_canonical_log_prob": -377.37371826171875, "reading_distances": [25, 27, 22, 27], "reading_log_probs": [-394.38818359375, -394.15069580078125, -393.62371826171875, -395.95123291015625], "reading_logprob_std": 1.000870225887155, "tokenization_path_logprob_std": 108.41210392981108, "trajectory_balance_rmse": 6416.506847403886} {"canonical_length": 143, "canonical_nll": 1.2215741564343858, "generation_rouge_l_f1": 0.3134796238244514, "generation_token_f1": 0.438871473354232, "prompt_id": "08c2f6d9de2bf434", "reading_canonical_log_prob": -174.6851043701172, "reading_distances": [21, 22, 20, 18], "reading_log_probs": [-184.8869171142578, -183.8318634033203, -182.4974365234375, -179.61312866210938], "reading_logprob_std": 2.2827843053724495, "tokenization_path_logprob_std": 205.1353034932194, "trajectory_balance_rmse": 2845.0260161904057} {"canonical_length": 118, "canonical_nll": 0.9481848377292439, "generation_rouge_l_f1": 0.3015873015873015, "generation_token_f1": 0.4523809523809524, "prompt_id": "5f825922a6155ea5", "reading_canonical_log_prob": -111.88581085205078, "reading_distances": [37, 29, 34, 36], "reading_log_probs": [-117.18097686767578, -116.78590393066406, -115.32303619384766, -120.61314392089844], "reading_logprob_std": 2.239062016714459, "tokenization_path_logprob_std": 110.77197470731473, "trajectory_balance_rmse": 2284.5030644564245} {"canonical_length": 55, "canonical_nll": 1.4601035378196023, "generation_rouge_l_f1": 0.20535714285714288, "generation_token_f1": 0.3482142857142857, "prompt_id": "3828b93ba3b0437a", "reading_canonical_log_prob": -80.30569458007812, "reading_distances": [22, 30, 26, 26], "reading_log_probs": [-76.91938781738281, -76.62975311279297, -76.21810913085938, -76.05403137207031], "reading_logprob_std": 0.39288989522626666, "tokenization_path_logprob_std": 125.22574517817007, "trajectory_balance_rmse": 1024.4278120624692} {"canonical_length": 109, "canonical_nll": 1.914849797520069, "generation_rouge_l_f1": 0.3008130081300814, "generation_token_f1": 0.4796747967479675, "prompt_id": "bf57ac1eadcdbbef", "reading_canonical_log_prob": -208.7186279296875, "reading_distances": [36, 45, 38, 37], "reading_log_probs": [-216.3612518310547, -211.8814239501953, -206.39967346191406, -211.09512329101562], "reading_logprob_std": 4.079921542208026, "tokenization_path_logprob_std": 129.9325891181113, "trajectory_balance_rmse": 2559.0744843752996} {"canonical_length": 15, "canonical_nll": 2.4907986958821615, "generation_rouge_l_f1": 0.14634146341463414, "generation_token_f1": 0.14634146341463414, "prompt_id": "fd15860ce78ef664", "reading_canonical_log_prob": -37.36198043823242, "reading_distances": [34, 33, 22, 31], "reading_log_probs": [-37.174373626708984, -37.118228912353516, -37.9869384765625, -38.03852081298828], "reading_logprob_std": 0.5012000973060564, "tokenization_path_logprob_std": 54.43345595027648, "trajectory_balance_rmse": 268.82024461048456} {"canonical_length": 25, "canonical_nll": 1.7652032470703125, "generation_rouge_l_f1": 0.09411764705882351, "generation_token_f1": 0.14117647058823532, "prompt_id": "b2df929cec29e4f9", "reading_canonical_log_prob": -44.13008117675781, "reading_distances": [31, 39, 30, 29], "reading_log_probs": [-57.515708923339844, -62.496273040771484, -55.69495391845703, -63.00640106201172], "reading_logprob_std": 3.631394355844773, "tokenization_path_logprob_std": 61.799158406025825, "trajectory_balance_rmse": 623.610452929844} {"canonical_length": 280, "canonical_nll": 0.7780582972935268, "generation_rouge_l_f1": 0.25287356321839083, "generation_token_f1": 0.47892720306513414, "prompt_id": "190309fcf6be3c92", "reading_canonical_log_prob": -217.8563232421875, "reading_distances": [24, 29, 28, 27], "reading_log_probs": [-219.3204803466797, -219.33351135253906, -224.56626892089844, -221.85943603515625], "reading_logprob_std": 2.500895757913179, "tokenization_path_logprob_std": 228.64121413733403, "trajectory_balance_rmse": 3571.501562833498} {"canonical_length": 20, "canonical_nll": 0.931951904296875, "generation_rouge_l_f1": 0.1, "generation_token_f1": 0.1, "prompt_id": "48be81c0aa95d468", "reading_canonical_log_prob": -18.6390380859375, "reading_distances": [14, 23, 18, 23], "reading_log_probs": [-20.209041595458984, -18.706296920776367, -19.12470054626465, -18.813316345214844], "reading_logprob_std": 0.6871172215687444, "tokenization_path_logprob_std": 78.02428045237704, "trajectory_balance_rmse": 500.7703410589411} {"canonical_length": 86, "canonical_nll": 2.0753031885901163, "generation_rouge_l_f1": 0.2119815668202765, "generation_token_f1": 0.3686635944700461, "prompt_id": "87fbf3fe9c551bf6", "reading_canonical_log_prob": -178.47607421875, "reading_distances": [32, 30, 29, 27], "reading_log_probs": [-173.1131591796875, -174.07212829589844, -174.70864868164062, -174.67758178710938], "reading_logprob_std": 0.7463982345639618, "tokenization_path_logprob_std": 53.30864105894305, "trajectory_balance_rmse": 1837.563120348406} {"canonical_length": 122, "canonical_nll": 1.3953674816694417, "generation_rouge_l_f1": 0.18877551020408162, "generation_token_f1": 0.32142857142857145, "prompt_id": "31fa13f44ebbb510", "reading_canonical_log_prob": -170.23483276367188, "reading_distances": [22, 22, 21, 22], "reading_log_probs": [-174.34475708007812, -165.42564392089844, -165.4058380126953, -172.62997436523438], "reading_logprob_std": 4.712450254943998, "tokenization_path_logprob_std": 124.13336009737623, "trajectory_balance_rmse": 2137.103856329746} {"canonical_length": 113, "canonical_nll": 1.2748009335678236, "generation_rouge_l_f1": 0.30303030303030304, "generation_token_f1": 0.5656565656565656, "prompt_id": "1e48963a206d1f06", "reading_canonical_log_prob": -144.05250549316406, "reading_distances": [13, 13, 15, 13], "reading_log_probs": [-145.45413208007812, -139.28976440429688, -133.18002319335938, -132.485595703125], "reading_logprob_std": 6.061792862425478, "tokenization_path_logprob_std": 110.54272434119036, "trajectory_balance_rmse": 2182.8655678290424} {"canonical_length": 7, "canonical_nll": 2.8257579803466797, "generation_rouge_l_f1": 0.7499999999999999, "generation_token_f1": 0.7499999999999999, "prompt_id": "b92fcbf5058fa547", "reading_canonical_log_prob": -19.780305862426758, "reading_distances": [21, 17, 20, 17], "reading_log_probs": [-18.54420280456543, -18.586034774780273, -20.90228271484375, -19.358257293701172], "reading_logprob_std": 1.1019044925334074, "tokenization_path_logprob_std": 28.043532172781333, "trajectory_balance_rmse": 115.16520708494122} {"canonical_length": 14, "canonical_nll": 1.157998766217913, "generation_rouge_l_f1": 0.46153846153846156, "generation_token_f1": 0.46153846153846156, "prompt_id": "5f82c19500742c50", "reading_canonical_log_prob": -16.21198272705078, "reading_distances": [28, 27, 25, 23], "reading_log_probs": [-20.605724334716797, -20.794879913330078, -22.260923385620117, -20.055187225341797], "reading_logprob_std": 0.9416411021853314, "tokenization_path_logprob_std": 59.9963274467012, "trajectory_balance_rmse": 421.9038912269116} {"canonical_length": 107, "canonical_nll": 1.4795823141793225, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.5238095238095238, "prompt_id": "cb0faecefbcafa87", "reading_canonical_log_prob": -158.3153076171875, "reading_distances": [26, 27, 26, 28], "reading_log_probs": [-166.10739135742188, -166.61968994140625, -166.5689697265625, -166.5221405029297], "reading_logprob_std": 0.23484084912931213, "tokenization_path_logprob_std": 183.6780550891232, "trajectory_balance_rmse": 2382.027916651834} {"canonical_length": 111, "canonical_nll": 1.668647559913429, "generation_rouge_l_f1": 0.15267175572519084, "generation_token_f1": 0.18320610687022898, "prompt_id": "935eaf327e8c6a5b", "reading_canonical_log_prob": -185.21987915039062, "reading_distances": [36, 33, 34, 33], "reading_log_probs": [-186.2438201904297, -181.5003662109375, -205.57077026367188, -196.93875122070312], "reading_logprob_std": 10.811462822167362, "tokenization_path_logprob_std": 191.95907726396663, "trajectory_balance_rmse": 2196.3904190394483} {"canonical_length": 72, "canonical_nll": 1.265170203314887, "generation_rouge_l_f1": 0.17575757575757575, "generation_token_f1": 0.2606060606060606, "prompt_id": "116cdf21f0043438", "reading_canonical_log_prob": -91.09225463867188, "reading_distances": [12, 18, 18, 16], "reading_log_probs": [-94.63438415527344, -89.90421295166016, -88.7596435546875, -92.14518737792969], "reading_logprob_std": 2.59610461073858, "tokenization_path_logprob_std": 217.9990490970798, "trajectory_balance_rmse": 1667.4061194168196} {"canonical_length": 31, "canonical_nll": 1.4788470114431074, "generation_rouge_l_f1": 0.21875, "generation_token_f1": 0.25, "prompt_id": "da79267701ae23c7", "reading_canonical_log_prob": -45.84425735473633, "reading_distances": [33, 35, 40, 31], "reading_log_probs": [-57.80133819580078, -52.95092010498047, -59.20721435546875, -56.23952102661133], "reading_logprob_std": 2.688023340775454, "tokenization_path_logprob_std": 45.40464306767191, "trajectory_balance_rmse": 865.5830171541023} {"canonical_length": 269, "canonical_nll": 0.798794487595115, "generation_rouge_l_f1": 0.3071428571428571, "generation_token_f1": 0.4928571428571428, "prompt_id": "3348c8386d3ef99b", "reading_canonical_log_prob": -214.87571716308594, "reading_distances": [15, 17, 11, 12], "reading_log_probs": [-218.5614776611328, -214.8040008544922, -218.517333984375, -216.1136474609375], "reading_logprob_std": 1.8572869858982082, "tokenization_path_logprob_std": 146.62979686242434, "trajectory_balance_rmse": 4431.970526884525} {"canonical_length": 10, "canonical_nll": 0.37860856056213377, "generation_rouge_l_f1": 0.8888888888888888, "generation_token_f1": 0.8888888888888888, "prompt_id": "d04f01367f709567", "reading_canonical_log_prob": -3.786085605621338, "reading_distances": [20, 27, 14, 25], "reading_log_probs": [-3.71677827835083, -5.389420509338379, -3.2447633743286133, -5.2291059494018555], "reading_logprob_std": 1.0751176496571964, "tokenization_path_logprob_std": 44.71298604644582, "trajectory_balance_rmse": 166.34919423409735} {"canonical_length": 281, "canonical_nll": 0.9163399149938835, "generation_rouge_l_f1": 0.21875, "generation_token_f1": 0.34375, "prompt_id": "f04ba139f8e40a51", "reading_canonical_log_prob": -257.49151611328125, "reading_distances": [14, 12, 15, 9], "reading_log_probs": [-280.6182556152344, -285.0244445800781, -273.4722900390625, -273.27008056640625], "reading_logprob_std": 5.745528974299703, "tokenization_path_logprob_std": 173.11843250380784, "trajectory_balance_rmse": 3940.612671021751} {"canonical_length": 501, "canonical_nll": 0.7717896726079092, "generation_rouge_l_f1": 0.4086330935251798, "generation_token_f1": 0.5352517985611511, "prompt_id": "16085c364de7d67f", "reading_canonical_log_prob": -386.6666259765625, "reading_distances": [14, 13, 17, 15], "reading_log_probs": [-385.78289794921875, -387.64093017578125, -385.91107177734375, -385.7672119140625], "reading_logprob_std": 0.912546081031269, "tokenization_path_logprob_std": 93.10895958137908, "trajectory_balance_rmse": 6916.181286374453} {"canonical_length": 405, "canonical_nll": 1.136451099537037, "generation_rouge_l_f1": 0.19753086419753085, "generation_token_f1": 0.41358024691358025, "prompt_id": "122921a791ee9365", "reading_canonical_log_prob": -460.2626953125, "reading_distances": [24, 22, 23, 21], "reading_log_probs": [-484.610595703125, -481.2486877441406, -471.99444580078125, -486.2576904296875], "reading_logprob_std": 6.372818801463443, "tokenization_path_logprob_std": 232.75028118710003, "trajectory_balance_rmse": 7271.813255191059} {"canonical_length": 67, "canonical_nll": 1.608788333722015, "generation_rouge_l_f1": 0.39999999999999997, "generation_token_f1": 0.5439999999999999, "prompt_id": "5090b423053cf048", "reading_canonical_log_prob": -107.788818359375, "reading_distances": [33, 38, 36, 33], "reading_log_probs": [-114.6448974609375, -107.13885498046875, -113.09622192382812, -104.21056365966797], "reading_logprob_std": 4.921334721215579, "tokenization_path_logprob_std": 131.22125755537957, "trajectory_balance_rmse": 1543.9286503215471} {"canonical_length": 349, "canonical_nll": 1.1001964667465078, "generation_rouge_l_f1": 0.22257551669316375, "generation_token_f1": 0.3910969793322735, "prompt_id": "4811a788bad2e4f1", "reading_canonical_log_prob": -383.96856689453125, "reading_distances": [31, 24, 29, 22], "reading_log_probs": [-391.29931640625, -391.9385986328125, -392.30316162109375, -390.8318176269531], "reading_logprob_std": 0.6555955057335787, "tokenization_path_logprob_std": 83.4977032541632, "trajectory_balance_rmse": 7384.801015369105} {"canonical_length": 14, "canonical_nll": 1.2775380270821708, "generation_rouge_l_f1": 0.3870967741935484, "generation_token_f1": 0.3870967741935484, "prompt_id": "b2fd6c1ae0104b61", "reading_canonical_log_prob": -17.88553237915039, "reading_distances": [24, 18, 27, 20], "reading_log_probs": [-21.002513885498047, -22.429183959960938, -20.734233856201172, -18.43927764892578], "reading_logprob_std": 1.6516803034674994, "tokenization_path_logprob_std": 38.61585123808332, "trajectory_balance_rmse": 314.0073775614396} {"canonical_length": 12, "canonical_nll": 0.6022164424260458, "generation_rouge_l_f1": 0.06278026905829596, "generation_token_f1": 0.07174887892376682, "prompt_id": "3b02f9d0f5b11977", "reading_canonical_log_prob": -7.226597309112549, "reading_distances": [26, 20, 18, 19], "reading_log_probs": [-7.32683801651001, -6.599616050720215, -6.810224533081055, -6.6846923828125], "reading_logprob_std": 0.3260169384071183, "tokenization_path_logprob_std": 32.53768112032121, "trajectory_balance_rmse": 197.88369085254638} {"canonical_length": 6, "canonical_nll": 0.7297150293986002, "generation_rouge_l_f1": 0.8000000000000002, "generation_token_f1": 0.8000000000000002, "prompt_id": "842d7e43c22f686e", "reading_canonical_log_prob": -4.378290176391602, "reading_distances": [23, 27, 22, 26], "reading_log_probs": [-5.581386566162109, -5.608363151550293, -6.275440216064453, -5.5148468017578125], "reading_logprob_std": 0.3557978364318378, "tokenization_path_logprob_std": 32.292131410973624, "trajectory_balance_rmse": 156.35531516024415} {"canonical_length": 74, "canonical_nll": 1.8020436054951436, "generation_rouge_l_f1": 0.2948717948717949, "generation_token_f1": 0.5641025641025641, "prompt_id": "f726d0133e785ebc", "reading_canonical_log_prob": -133.35122680664062, "reading_distances": [22, 22, 20, 22], "reading_log_probs": [-135.05532836914062, -136.72354125976562, -135.3929443359375, -132.4562225341797], "reading_logprob_std": 1.785522849456346, "tokenization_path_logprob_std": 76.4111911687003, "trajectory_balance_rmse": 1141.7991789770779} {"canonical_length": 48, "canonical_nll": 1.1818892161051433, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.4782608695652174, "prompt_id": "cb06f10a99b5f8fb", "reading_canonical_log_prob": -56.730682373046875, "reading_distances": [21, 19, 24, 17], "reading_log_probs": [-63.041160583496094, -60.974639892578125, -61.09449005126953, -59.394073486328125], "reading_logprob_std": 1.4934649014063441, "tokenization_path_logprob_std": 25.877351795271252, "trajectory_balance_rmse": 1028.0889035167252} {"canonical_length": 171, "canonical_nll": 0.8850882011547423, "generation_rouge_l_f1": 0.2826666666666667, "generation_token_f1": 0.45866666666666667, "prompt_id": "443ce871037b41c0", "reading_canonical_log_prob": -151.35008239746094, "reading_distances": [30, 36, 30, 28], "reading_log_probs": [-160.0960693359375, -161.70346069335938, -164.23651123046875, -160.08856201171875], "reading_logprob_std": 1.9569693387707188, "tokenization_path_logprob_std": 52.20665670694153, "trajectory_balance_rmse": 2200.146322972526} {"canonical_length": 307, "canonical_nll": 1.361438664240635, "generation_rouge_l_f1": 0.3168654173764906, "generation_token_f1": 0.555366269165247, "prompt_id": "d1c3cdcce82ccde0", "reading_canonical_log_prob": -417.961669921875, "reading_distances": [44, 41, 34, 36], "reading_log_probs": [-426.1721496582031, -426.0488586425781, -433.415771484375, -429.463623046875], "reading_logprob_std": 3.4745569613379117, "tokenization_path_logprob_std": 211.73462580069037, "trajectory_balance_rmse": 6430.745550841822} {"canonical_length": 61, "canonical_nll": 0.5720563794745773, "generation_rouge_l_f1": 0.35668789808917195, "generation_token_f1": 0.4585987261146497, "prompt_id": "66bf9508c6f9f94b", "reading_canonical_log_prob": -34.89543914794922, "reading_distances": [41, 46, 45, 47], "reading_log_probs": [-38.62407302856445, -45.59910583496094, -45.03715896606445, -42.568634033203125], "reading_logprob_std": 3.1745271876330046, "tokenization_path_logprob_std": 124.89303276375874, "trajectory_balance_rmse": 1177.6486512731376} {"canonical_length": 265, "canonical_nll": 0.8738923846550708, "generation_rouge_l_f1": 0.218809980806142, "generation_token_f1": 0.45681381957773515, "prompt_id": "c9e81615c39cdb45", "reading_canonical_log_prob": -231.58148193359375, "reading_distances": [19, 27, 23, 24], "reading_log_probs": [-236.095947265625, -238.2382354736328, -238.3468780517578, -239.316162109375], "reading_logprob_std": 1.358280544052314, "tokenization_path_logprob_std": 336.44828251900356, "trajectory_balance_rmse": 5687.092268641484} {"canonical_length": 41, "canonical_nll": 1.5723463849323551, "generation_rouge_l_f1": 0.08641975308641975, "generation_token_f1": 0.16049382716049382, "prompt_id": "768e651293911174", "reading_canonical_log_prob": -64.46620178222656, "reading_distances": [55, 54, 52, 61], "reading_log_probs": [-67.98294067382812, -69.85260772705078, -68.87808990478516, -69.46675109863281], "reading_logprob_std": 0.8136201374667977, "tokenization_path_logprob_std": 116.98123608578457, "trajectory_balance_rmse": 810.0380005637551} {"canonical_length": 367, "canonical_nll": 1.1433444737738419, "generation_rouge_l_f1": 0.16874999999999998, "generation_token_f1": 0.328125, "prompt_id": "bd6b470fc22dea74", "reading_canonical_log_prob": -419.607421875, "reading_distances": [48, 35, 46, 46], "reading_log_probs": [-428.1832275390625, -432.7828674316406, -428.5986328125, -428.44500732421875], "reading_logprob_std": 2.193669448449692, "tokenization_path_logprob_std": 290.2130385522451, "trajectory_balance_rmse": 7704.848975437611} {"canonical_length": 303, "canonical_nll": 1.2516209567734118, "generation_rouge_l_f1": 0.2385008517887564, "generation_token_f1": 0.4429301533219761, "prompt_id": "ca8e3b0a9ea7294b", "reading_canonical_log_prob": -379.24114990234375, "reading_distances": [24, 17, 25, 21], "reading_log_probs": [-384.5900573730469, -388.5174255371094, -386.9290771484375, -390.10430908203125], "reading_logprob_std": 2.352752890000191, "tokenization_path_logprob_std": 189.4611966798024, "trajectory_balance_rmse": 6684.205617742754} {"canonical_length": 246, "canonical_nll": 1.517755523929751, "generation_rouge_l_f1": 0.2922201138519923, "generation_token_f1": 0.5654648956356736, "prompt_id": "5cd1f02fb9a1877d", "reading_canonical_log_prob": -373.36785888671875, "reading_distances": [25, 17, 30, 29], "reading_log_probs": [-368.07147216796875, -371.20745849609375, -370.38818359375, -366.93634033203125], "reading_logprob_std": 1.9857611891908367, "tokenization_path_logprob_std": 194.33569167656103, "trajectory_balance_rmse": 5060.754134623818} {"canonical_length": 210, "canonical_nll": 1.0055236089797246, "generation_rouge_l_f1": 0.23076923076923078, "generation_token_f1": 0.39316239316239315, "prompt_id": "754987e5b5772585", "reading_canonical_log_prob": -211.1599578857422, "reading_distances": [27, 24, 27, 33], "reading_log_probs": [-222.33474731445312, -229.13565063476562, -227.5047607421875, -222.40264892578125], "reading_logprob_std": 3.5001261201803446, "tokenization_path_logprob_std": 278.01863190004593, "trajectory_balance_rmse": 4799.754239839985} {"canonical_length": 221, "canonical_nll": 1.171263267551612, "generation_rouge_l_f1": 0.36974789915966383, "generation_token_f1": 0.5798319327731092, "prompt_id": "5aac80ddf699bdd3", "reading_canonical_log_prob": -258.84918212890625, "reading_distances": [20, 20, 20, 20], "reading_log_probs": [-267.0321350097656, -265.406494140625, -265.5575866699219, -265.630859375], "reading_logprob_std": 0.7560374795966904, "tokenization_path_logprob_std": 245.91280842878083, "trajectory_balance_rmse": 4042.1215703207763} {"canonical_length": 25, "canonical_nll": 1.2374343872070312, "generation_rouge_l_f1": 0.2406015037593985, "generation_token_f1": 0.27067669172932335, "prompt_id": "f552b8f246a9b742", "reading_canonical_log_prob": -30.93585968017578, "reading_distances": [23, 22, 27, 26], "reading_log_probs": [-28.525964736938477, -27.460403442382812, -32.56800842285156, -32.474525451660156], "reading_logprob_std": 2.650510280673324, "tokenization_path_logprob_std": 73.56314706041077, "trajectory_balance_rmse": 462.03060332701233} {"canonical_length": 212, "canonical_nll": 1.263317755933078, "generation_rouge_l_f1": 0.2005012531328321, "generation_token_f1": 0.4160401002506265, "prompt_id": "1dcab7a046130541", "reading_canonical_log_prob": -267.8233642578125, "reading_distances": [20, 18, 19, 26], "reading_log_probs": [-272.4289855957031, -273.0601806640625, -269.5506591796875, -272.3839416503906], "reading_logprob_std": 1.5675545960762973, "tokenization_path_logprob_std": 97.37496113103059, "trajectory_balance_rmse": 4248.35745818682} {"canonical_length": 21, "canonical_nll": 1.3701246352422805, "generation_rouge_l_f1": 0.12121212121212123, "generation_token_f1": 0.12121212121212123, "prompt_id": "721929550e55e7e5", "reading_canonical_log_prob": -28.77261734008789, "reading_distances": [24, 25, 26, 27], "reading_log_probs": [-37.85713577270508, -39.93980026245117, -39.604618072509766, -39.603179931640625], "reading_logprob_std": 0.942758292889852, "tokenization_path_logprob_std": 23.460691519233563, "trajectory_balance_rmse": 321.2555213054675} {"canonical_length": 106, "canonical_nll": 2.3633350876142396, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.38888888888888884, "prompt_id": "a21c76c0694212ca", "reading_canonical_log_prob": -250.51351928710938, "reading_distances": [34, 36, 35, 33], "reading_log_probs": [-250.10531616210938, -248.18565368652344, -247.46083068847656, -254.27476501464844], "reading_logprob_std": 3.056346638834076, "tokenization_path_logprob_std": 57.94996794355719, "trajectory_balance_rmse": 1845.8335200218642} {"canonical_length": 10, "canonical_nll": 1.5135167121887207, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "7cb87ccdae4ac547", "reading_canonical_log_prob": -15.135167121887207, "reading_distances": [19, 12, 18, 15], "reading_log_probs": [-18.319454193115234, -17.562820434570312, -18.166357040405273, -16.3531436920166], "reading_logprob_std": 0.8933843551490716, "tokenization_path_logprob_std": 15.879523251942379, "trajectory_balance_rmse": 229.9165309616217} {"canonical_length": 263, "canonical_nll": 0.840719012706452, "generation_rouge_l_f1": 0.30071599045346065, "generation_token_f1": 0.5441527446300716, "prompt_id": "02274883d489f1d7", "reading_canonical_log_prob": -221.10910034179688, "reading_distances": [52, 51, 65, 54], "reading_log_probs": [-222.93812561035156, -229.67678833007812, -222.02407836914062, -229.65972900390625], "reading_logprob_std": 4.166257624982341, "tokenization_path_logprob_std": 237.62102662446262, "trajectory_balance_rmse": 4117.023978986888} {"canonical_length": 219, "canonical_nll": 1.2902525461972032, "generation_rouge_l_f1": 0.2658486707566462, "generation_token_f1": 0.4703476482617587, "prompt_id": "dd480c5be4f865e4", "reading_canonical_log_prob": -282.5653076171875, "reading_distances": [25, 18, 27, 22], "reading_log_probs": [-291.552490234375, -289.8797302246094, -292.3236083984375, -292.58209228515625], "reading_logprob_std": 1.2177487702957264, "tokenization_path_logprob_std": 223.39596725265852, "trajectory_balance_rmse": 4993.356105960603} {"canonical_length": 17, "canonical_nll": 0.656436190885656, "generation_rouge_l_f1": 0.07003891050583659, "generation_token_f1": 0.08560311284046693, "prompt_id": "48b6b6146fd695d3", "reading_canonical_log_prob": -11.159415245056152, "reading_distances": [17, 19, 18, 17], "reading_log_probs": [-14.315964698791504, -14.51150131225586, -14.23897647857666, -17.05747413635254], "reading_logprob_std": 1.3558580371358941, "tokenization_path_logprob_std": 20.213999004159366, "trajectory_balance_rmse": 238.6387414767538} {"canonical_length": 230, "canonical_nll": 2.0510175622027855, "generation_rouge_l_f1": 0.19999999999999998, "generation_token_f1": 0.39999999999999997, "prompt_id": "619b20aa15b49441", "reading_canonical_log_prob": -471.7340393066406, "reading_distances": [26, 22, 19, 24], "reading_log_probs": [-479.10723876953125, -478.41815185546875, -476.26220703125, -477.156494140625], "reading_logprob_std": 1.271943351813183, "tokenization_path_logprob_std": 153.0895875350774, "trajectory_balance_rmse": 6199.256093747116} {"canonical_length": 264, "canonical_nll": 1.2434322472774622, "generation_rouge_l_f1": 0.30018761726078796, "generation_token_f1": 0.547842401500938, "prompt_id": "49784b3e9fd7ae19", "reading_canonical_log_prob": -328.26611328125, "reading_distances": [38, 31, 38, 35], "reading_log_probs": [-338.8000183105469, -337.33624267578125, -342.8325500488281, -337.3695068359375], "reading_logprob_std": 2.5901360452918114, "tokenization_path_logprob_std": 304.659848308029, "trajectory_balance_rmse": 6181.716594384623} {"canonical_length": 128, "canonical_nll": 1.3421714305877686, "generation_rouge_l_f1": 0.144702842377261, "generation_token_f1": 0.35658914728682173, "prompt_id": "211ba51e9f80e5d0", "reading_canonical_log_prob": -171.79794311523438, "reading_distances": [11, 9, 14, 12], "reading_log_probs": [-174.4541015625, -173.19960021972656, -178.54476928710938, -176.8848419189453], "reading_logprob_std": 2.4000526442278285, "tokenization_path_logprob_std": 94.5184692843615, "trajectory_balance_rmse": 2577.7939503390603} {"canonical_length": 19, "canonical_nll": 0.6316652297973633, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "c295bb5aa04e8e4c", "reading_canonical_log_prob": -12.001639366149902, "reading_distances": [23, 23, 25, 18], "reading_log_probs": [-11.850578308105469, -13.485528945922852, -11.146251678466797, -12.644221305847168], "reading_logprob_std": 1.0092463558278912, "tokenization_path_logprob_std": 80.86072901314412, "trajectory_balance_rmse": 525.7314518936315} {"canonical_length": 163, "canonical_nll": 0.5758627675062308, "generation_rouge_l_f1": 0.4689265536723164, "generation_token_f1": 0.5197740112994351, "prompt_id": "55b91ee2bfd58fff", "reading_canonical_log_prob": -93.86563110351562, "reading_distances": [39, 24, 36, 32], "reading_log_probs": [-105.25642395019531, -103.98875427246094, -106.09088134765625, -105.30101776123047], "reading_logprob_std": 0.8693940589761006, "tokenization_path_logprob_std": 167.61513396016946, "trajectory_balance_rmse": 2242.756196389818} {"canonical_length": 79, "canonical_nll": 1.0360218482681467, "generation_rouge_l_f1": 0.35000000000000003, "generation_token_f1": 0.55, "prompt_id": "aacdb99cf2030827", "reading_canonical_log_prob": -81.8457260131836, "reading_distances": [13, 29, 28, 24], "reading_log_probs": [-88.23963928222656, -89.53333282470703, -87.87934112548828, -87.84588623046875], "reading_logprob_std": 0.7928210700227872, "tokenization_path_logprob_std": 106.72334854683963, "trajectory_balance_rmse": 1742.824244196302} {"canonical_length": 7, "canonical_nll": 0.5576424939291817, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.7142857142857143, "prompt_id": "732d6c9c54c4fcf0", "reading_canonical_log_prob": -3.9034974575042725, "reading_distances": [19, 17, 23, 18], "reading_log_probs": [-4.753417015075684, -4.004030227661133, -4.084160327911377, -5.178136348724365], "reading_logprob_std": 0.5606247358556574, "tokenization_path_logprob_std": 45.8000108746762, "trajectory_balance_rmse": 152.79908443160588} {"canonical_length": 234, "canonical_nll": 0.6266558394472823, "generation_rouge_l_f1": 0.4086444007858546, "generation_token_f1": 0.6365422396856583, "prompt_id": "1e288eaee9c72d47", "reading_canonical_log_prob": -146.63746643066406, "reading_distances": [34, 34, 31, 32], "reading_log_probs": [-165.8192138671875, -170.1934814453125, -168.3990020751953, -172.84226989746094], "reading_logprob_std": 2.959331363979089, "tokenization_path_logprob_std": 338.74825481854407, "trajectory_balance_rmse": 3892.515066115526} {"canonical_length": 85, "canonical_nll": 1.1757272159352021, "generation_rouge_l_f1": 0.2331288343558282, "generation_token_f1": 0.3558282208588957, "prompt_id": "34251c7523ed14e6", "reading_canonical_log_prob": -99.93681335449219, "reading_distances": [26, 25, 27, 25], "reading_log_probs": [-106.67927551269531, -103.80485534667969, -106.7031478881836, -109.54351043701172], "reading_logprob_std": 2.342837030068084, "tokenization_path_logprob_std": 121.61566466224633, "trajectory_balance_rmse": 1827.9931672512723} {"canonical_length": 62, "canonical_nll": 2.3637623940744708, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.3407407407407407, "prompt_id": "915708e38c0c1bff", "reading_canonical_log_prob": -146.5532684326172, "reading_distances": [13, 14, 11, 14], "reading_log_probs": [-161.2531280517578, -163.87623596191406, -145.03668212890625, -160.03067016601562], "reading_logprob_std": 8.494532844394039, "tokenization_path_logprob_std": 106.97419592065681, "trajectory_balance_rmse": 1004.7606388716459} {"canonical_length": 145, "canonical_nll": 1.6919209447400323, "generation_rouge_l_f1": 0.34810126582278483, "generation_token_f1": 0.518987341772152, "prompt_id": "fc76d6d236be83db", "reading_canonical_log_prob": -245.3285369873047, "reading_distances": [23, 27, 22, 25], "reading_log_probs": [-252.2273406982422, -245.5945587158203, -250.884521484375, -247.80593872070312], "reading_logprob_std": 2.9957933724630905, "tokenization_path_logprob_std": 158.2324502806448, "trajectory_balance_rmse": 2355.170956393513} {"canonical_length": 391, "canonical_nll": 0.7264428492397299, "generation_rouge_l_f1": 0.337228714524207, "generation_token_f1": 0.5275459098497496, "prompt_id": "cf72c2330315d4cd", "reading_canonical_log_prob": -284.0391540527344, "reading_distances": [23, 19, 28, 22], "reading_log_probs": [-283.86334228515625, -281.3121337890625, -280.40924072265625, -284.5303649902344], "reading_logprob_std": 1.9799051927234126, "tokenization_path_logprob_std": 112.65251865671226, "trajectory_balance_rmse": 4093.5006888328217} {"canonical_length": 247, "canonical_nll": 1.2369495963278088, "generation_rouge_l_f1": 0.32748538011695905, "generation_token_f1": 0.5497076023391813, "prompt_id": "12725fe24fa84c62", "reading_canonical_log_prob": -305.52655029296875, "reading_distances": [34, 37, 32, 31], "reading_log_probs": [-298.9849853515625, -307.7962646484375, -307.48089599609375, -308.1614990234375], "reading_logprob_std": 4.422702960425695, "tokenization_path_logprob_std": 358.58835623777077, "trajectory_balance_rmse": 6000.649687831664} {"canonical_length": 164, "canonical_nll": 1.6620397800352515, "generation_rouge_l_f1": 0.25118483412322273, "generation_token_f1": 0.47867298578199047, "prompt_id": "767d0dc7cec8c49a", "reading_canonical_log_prob": -272.57452392578125, "reading_distances": [13, 12, 17, 13], "reading_log_probs": [-276.7805480957031, -275.4636535644531, -276.423828125, -282.0924072265625], "reading_logprob_std": 2.9870889828888267, "tokenization_path_logprob_std": 207.76914071320195, "trajectory_balance_rmse": 3536.843243814597} {"canonical_length": 211, "canonical_nll": 1.371337890625, "generation_rouge_l_f1": 0.21868787276341947, "generation_token_f1": 0.3021868787276342, "prompt_id": "0305770999fedf01", "reading_canonical_log_prob": -289.352294921875, "reading_distances": [33, 26, 28, 37], "reading_log_probs": [-290.0286560058594, -299.0056457519531, -285.06591796875, -294.57073974609375], "reading_logprob_std": 5.987289852953969, "tokenization_path_logprob_std": 135.32308095676427, "trajectory_balance_rmse": 2772.6282548946224} {"canonical_length": 28, "canonical_nll": 0.6466276305062431, "generation_rouge_l_f1": 0.6086956521739131, "generation_token_f1": 0.6086956521739131, "prompt_id": "47deaa7fb6aebf76", "reading_canonical_log_prob": -18.105573654174805, "reading_distances": [17, 18, 14, 15], "reading_log_probs": [-18.301000595092773, -21.083972930908203, -18.89157485961914, -18.32685089111328], "reading_logprob_std": 1.317245409543438, "tokenization_path_logprob_std": 60.267106715536926, "trajectory_balance_rmse": 792.7031041445944} {"canonical_length": 55, "canonical_nll": 1.1639816284179687, "generation_rouge_l_f1": 0.28169014084507044, "generation_token_f1": 0.43661971830985913, "prompt_id": "83b035c8880eabe5", "reading_canonical_log_prob": -64.01898956298828, "reading_distances": [38, 38, 29, 28], "reading_log_probs": [-77.28264617919922, -72.64824676513672, -70.85000610351562, -69.15081787109375], "reading_logprob_std": 3.50399851241471, "tokenization_path_logprob_std": 80.72217018511972, "trajectory_balance_rmse": 1490.568642513338} {"canonical_length": 235, "canonical_nll": 1.4102439068733377, "generation_rouge_l_f1": 0.30078125, "generation_token_f1": 0.49609375, "prompt_id": "3f8f300bcd0a09f9", "reading_canonical_log_prob": -331.4073181152344, "reading_distances": [30, 24, 30, 33], "reading_log_probs": [-329.95538330078125, -338.0352783203125, -333.857177734375, -328.76812744140625], "reading_logprob_std": 4.194771771774149, "tokenization_path_logprob_std": 295.05247161292004, "trajectory_balance_rmse": 5144.089758926803} {"canonical_length": 32, "canonical_nll": 1.2798619270324707, "generation_rouge_l_f1": 0.16, "generation_token_f1": 0.1866666666666667, "prompt_id": "de8521c6b6a04f6f", "reading_canonical_log_prob": -40.95558166503906, "reading_distances": [20, 21, 24, 20], "reading_log_probs": [-48.11853790283203, -40.721614837646484, -36.99334716796875, -41.474063873291016], "reading_logprob_std": 4.629402016909306, "tokenization_path_logprob_std": 83.07289200361876, "trajectory_balance_rmse": 648.8600687713512} {"canonical_length": 12, "canonical_nll": 0.5417372385660807, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "89d6bb48c1695f7e", "reading_canonical_log_prob": -6.500846862792969, "reading_distances": [23, 26, 34, 26], "reading_log_probs": [-11.486183166503906, -9.218979835510254, -9.944332122802734, -10.177179336547852], "reading_logprob_std": 0.9455821249935902, "tokenization_path_logprob_std": 57.273426245328615, "trajectory_balance_rmse": 253.887290395112} {"canonical_length": 325, "canonical_nll": 1.6771035531850962, "generation_rouge_l_f1": 0.23322683706070285, "generation_token_f1": 0.4440894568690096, "prompt_id": "b1d79661f636c046", "reading_canonical_log_prob": -545.0586547851562, "reading_distances": [16, 19, 22, 14], "reading_log_probs": [-540.0594482421875, -541.91650390625, -540.1327514648438, -543.54345703125], "reading_logprob_std": 1.6596688436760447, "tokenization_path_logprob_std": 361.9612158512109, "trajectory_balance_rmse": 7014.167502194828} {"canonical_length": 15, "canonical_nll": 0.9565741221110026, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "db698be75c6d5259", "reading_canonical_log_prob": -14.348611831665039, "reading_distances": [22, 19, 26, 22], "reading_log_probs": [-16.575641632080078, -17.323959350585938, -17.596412658691406, -14.725578308105469], "reading_logprob_std": 1.2939665089315688, "tokenization_path_logprob_std": 42.276718878992675, "trajectory_balance_rmse": 352.4934638622565} {"canonical_length": 33, "canonical_nll": 1.0974806583288945, "generation_rouge_l_f1": 0.43181818181818177, "generation_token_f1": 0.5227272727272728, "prompt_id": "4c3924ee604d1034", "reading_canonical_log_prob": -36.216861724853516, "reading_distances": [31, 23, 26, 19], "reading_log_probs": [-36.833404541015625, -37.06498718261719, -36.59831237792969, -37.180694580078125], "reading_logprob_std": 0.258174976324872, "tokenization_path_logprob_std": 28.055452594301805, "trajectory_balance_rmse": 707.3297139903182} {"canonical_length": 63, "canonical_nll": 1.3386160229879713, "generation_rouge_l_f1": 0.3229813664596274, "generation_token_f1": 0.5590062111801242, "prompt_id": "1fd45a1cb9cbeccc", "reading_canonical_log_prob": -84.33280944824219, "reading_distances": [16, 16, 20, 19], "reading_log_probs": [-92.90748596191406, -95.20539093017578, -85.67257690429688, -96.56513214111328], "reading_logprob_std": 4.8508921077053415, "tokenization_path_logprob_std": 156.29499867564277, "trajectory_balance_rmse": 1222.8259456437036} {"canonical_length": 205, "canonical_nll": 1.1343900354897103, "generation_rouge_l_f1": 0.36320754716981124, "generation_token_f1": 0.5754716981132075, "prompt_id": "326b7dac8c34a0d5", "reading_canonical_log_prob": -232.54995727539062, "reading_distances": [21, 21, 21, 12], "reading_log_probs": [-236.9007568359375, -237.38235473632812, -236.79061889648438, -238.0803680419922], "reading_logprob_std": 0.5871098873716931, "tokenization_path_logprob_std": 59.07525014486692, "trajectory_balance_rmse": 4291.202410240395} {"canonical_length": 322, "canonical_nll": 1.4535105450553183, "generation_rouge_l_f1": 0.2021276595744681, "generation_token_f1": 0.4042553191489362, "prompt_id": "1e2cec78b7ee8d07", "reading_canonical_log_prob": -468.0303955078125, "reading_distances": [15, 19, 18, 17], "reading_log_probs": [-477.3238220214844, -482.10540771484375, -472.262939453125, -480.29449462890625], "reading_logprob_std": 4.300779542551256, "tokenization_path_logprob_std": 252.1784551973215, "trajectory_balance_rmse": 6066.590442253039} {"canonical_length": 360, "canonical_nll": 1.1168377346462675, "generation_rouge_l_f1": 0.2558922558922559, "generation_token_f1": 0.49158249158249157, "prompt_id": "e3633ae650b9ffe5", "reading_canonical_log_prob": -402.06158447265625, "reading_distances": [22, 15, 20, 21], "reading_log_probs": [-401.6678466796875, -400.591064453125, -394.0149841308594, -399.208984375], "reading_logprob_std": 3.3899899861145384, "tokenization_path_logprob_std": 253.6013088505718, "trajectory_balance_rmse": 5674.76708881111} {"canonical_length": 199, "canonical_nll": 1.0551836023378611, "generation_rouge_l_f1": 0.3010204081632653, "generation_token_f1": 0.5561224489795918, "prompt_id": "e7d8d0860929d3e3", "reading_canonical_log_prob": -209.98153686523438, "reading_distances": [24, 30, 28, 25], "reading_log_probs": [-217.4732208251953, -222.08053588867188, -217.9873046875, -220.10073852539062], "reading_logprob_std": 2.112191307416369, "tokenization_path_logprob_std": 297.28989539498696, "trajectory_balance_rmse": 3675.0883969026922} {"canonical_length": 58, "canonical_nll": 1.8898936304552802, "generation_rouge_l_f1": 0.19285714285714287, "generation_token_f1": 0.30714285714285716, "prompt_id": "8c4c85aa12a6cb37", "reading_canonical_log_prob": -109.61383056640625, "reading_distances": [30, 24, 30, 29], "reading_log_probs": [-109.44929504394531, -106.93726348876953, -106.3335952758789, -107.68402862548828], "reading_logprob_std": 1.3503023371165919, "tokenization_path_logprob_std": 70.74996769985651, "trajectory_balance_rmse": 1314.397405293234} {"canonical_length": 35, "canonical_nll": 1.9001896449497768, "generation_rouge_l_f1": 0.30952380952380953, "generation_token_f1": 0.45238095238095244, "prompt_id": "5f2bc0710dc7dd88", "reading_canonical_log_prob": -66.50663757324219, "reading_distances": [13, 16, 14, 25], "reading_log_probs": [-77.8819580078125, -82.91648864746094, -84.54911041259766, -82.3728256225586], "reading_logprob_std": 2.852813679559272, "tokenization_path_logprob_std": 49.72732474069264, "trajectory_balance_rmse": 733.0471741581837} {"canonical_length": 24, "canonical_nll": 1.1220448017120361, "generation_rouge_l_f1": 0.1119402985074627, "generation_token_f1": 0.11940298507462688, "prompt_id": "e06153205819c799", "reading_canonical_log_prob": -26.929075241088867, "reading_distances": [70, 71, 64, 77], "reading_log_probs": [-30.516162872314453, -28.359172821044922, -30.03626823425293, -28.41408920288086], "reading_logprob_std": 1.108630848556846, "tokenization_path_logprob_std": 89.72057035922256, "trajectory_balance_rmse": 459.2487815084205} {"canonical_length": 81, "canonical_nll": 0.8537124822169174, "generation_rouge_l_f1": 0.17667844522968199, "generation_token_f1": 0.2190812720848057, "prompt_id": "788acaefe94c61a6", "reading_canonical_log_prob": -69.15071105957031, "reading_distances": [19, 19, 26, 26], "reading_log_probs": [-75.93479919433594, -70.84272766113281, -75.303466796875, -75.6014175415039], "reading_logprob_std": 2.3991493993132154, "tokenization_path_logprob_std": 14.133337248496334, "trajectory_balance_rmse": 945.977186655655} {"canonical_length": 21, "canonical_nll": 1.993374960763114, "generation_rouge_l_f1": 0.20512820512820512, "generation_token_f1": 0.25641025641025644, "prompt_id": "24304bef6f934812", "reading_canonical_log_prob": -41.86087417602539, "reading_distances": [17, 17, 14, 14], "reading_log_probs": [-43.89740753173828, -46.13129425048828, -42.883277893066406, -41.61530303955078], "reading_logprob_std": 1.9100230542673986, "tokenization_path_logprob_std": 48.52698425394242, "trajectory_balance_rmse": 346.4130064603021} {"canonical_length": 315, "canonical_nll": 1.4385946606832838, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.5017301038062284, "prompt_id": "969a799cc7257588", "reading_canonical_log_prob": -453.1573181152344, "reading_distances": [20, 15, 16, 18], "reading_log_probs": [-451.2314453125, -454.52532958984375, -452.34417724609375, -449.55450439453125], "reading_logprob_std": 2.0846418581032498, "tokenization_path_logprob_std": 209.77463787301278, "trajectory_balance_rmse": 6583.474516193665} {"canonical_length": 152, "canonical_nll": 1.297210693359375, "generation_rouge_l_f1": 0.20246913580246914, "generation_token_f1": 0.36049382716049383, "prompt_id": "7f301b40ae5af8a5", "reading_canonical_log_prob": -197.176025390625, "reading_distances": [23, 32, 31, 24], "reading_log_probs": [-220.99496459960938, -207.46762084960938, -207.49658203125, -216.08033752441406], "reading_logprob_std": 6.6908488862127955, "tokenization_path_logprob_std": 240.64791584501194, "trajectory_balance_rmse": 2761.4466642573198} {"canonical_length": 354, "canonical_nll": 1.4285443839380296, "generation_rouge_l_f1": 0.22940226171243946, "generation_token_f1": 0.4168012924071083, "prompt_id": "d838a6ece3beb4b3", "reading_canonical_log_prob": -505.7047119140625, "reading_distances": [15, 19, 22, 17], "reading_log_probs": [-506.1630554199219, -504.7623291015625, -513.0413208007812, -507.43804931640625], "reading_logprob_std": 3.628544380510501, "tokenization_path_logprob_std": 176.03456830613214, "trajectory_balance_rmse": 7888.778390422644} {"canonical_length": 15, "canonical_nll": 2.327711232503255, "generation_rouge_l_f1": 0.21428571428571427, "generation_token_f1": 0.21428571428571427, "prompt_id": "3ad7d2823c1e6bd8", "reading_canonical_log_prob": -34.91566848754883, "reading_distances": [40, 44, 31, 41], "reading_log_probs": [-48.291229248046875, -38.581451416015625, -40.85565948486328, -47.77029037475586], "reading_logprob_std": 4.8926625731276765, "tokenization_path_logprob_std": 31.55766181847715, "trajectory_balance_rmse": 475.8523011447967} {"canonical_length": 8, "canonical_nll": 0.1356101930141449, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "7a4144ac9a54046a", "reading_canonical_log_prob": -1.0848815441131592, "reading_distances": [32, 28, 27, 30], "reading_log_probs": [-2.2996466159820557, -2.2144622802734375, -2.0469329357147217, -1.8452547788619995], "reading_logprob_std": 0.20055067028558718, "tokenization_path_logprob_std": 32.32454420206162, "trajectory_balance_rmse": 117.20154266943764} {"canonical_length": 34, "canonical_nll": 0.4353351032032686, "generation_rouge_l_f1": 0.5925925925925927, "generation_token_f1": 0.6666666666666666, "prompt_id": "c898f115e1329204", "reading_canonical_log_prob": -14.801393508911133, "reading_distances": [10, 7, 6, 7], "reading_log_probs": [-15.713576316833496, -13.906003952026367, -14.141672134399414, -13.799698829650879], "reading_logprob_std": 0.8937241213549236, "tokenization_path_logprob_std": 57.66675739581822, "trajectory_balance_rmse": 489.97514600768466} {"canonical_length": 104, "canonical_nll": 1.7992360041691706, "generation_rouge_l_f1": 0.20555555555555557, "generation_token_f1": 0.32222222222222224, "prompt_id": "5328de30202d5ac5", "reading_canonical_log_prob": -187.12054443359375, "reading_distances": [14, 12, 11, 14], "reading_log_probs": [-181.56805419921875, -185.96435546875, -189.05743408203125, -185.22100830078125], "reading_logprob_std": 3.0767977488072678, "tokenization_path_logprob_std": 37.27959719567724, "trajectory_balance_rmse": 2548.8788895331363} {"canonical_length": 35, "canonical_nll": 1.3104739597865513, "generation_rouge_l_f1": 0.30434782608695654, "generation_token_f1": 0.43478260869565216, "prompt_id": "f9e71d1877230558", "reading_canonical_log_prob": -45.8665885925293, "reading_distances": [29, 22, 28, 26], "reading_log_probs": [-46.347938537597656, -47.17424774169922, -44.097660064697266, -45.414649963378906], "reading_logprob_std": 1.3201517001885512, "tokenization_path_logprob_std": 147.9390785758136, "trajectory_balance_rmse": 772.9769333103768} {"canonical_length": 211, "canonical_nll": 1.5251435917135663, "generation_rouge_l_f1": 0.2454728370221328, "generation_token_f1": 0.45875251509054327, "prompt_id": "4df4b024265f29c3", "reading_canonical_log_prob": -321.8052978515625, "reading_distances": [27, 28, 31, 30], "reading_log_probs": [-337.2209167480469, -339.23260498046875, -341.8634033203125, -338.8587646484375], "reading_logprob_std": 1.92290637036427, "tokenization_path_logprob_std": 167.61437563067096, "trajectory_balance_rmse": 4726.988648454356} {"canonical_length": 21, "canonical_nll": 0.3314772787548247, "generation_rouge_l_f1": 0.30588235294117644, "generation_token_f1": 0.30588235294117644, "prompt_id": "6e5cc5a69c7269eb", "reading_canonical_log_prob": -6.961022853851318, "reading_distances": [19, 26, 29, 24], "reading_log_probs": [-5.414730072021484, -7.70972204208374, -8.032635688781738, -12.932923316955566], "reading_logprob_std": 3.1628385358852307, "tokenization_path_logprob_std": 87.40250772881842, "trajectory_balance_rmse": 467.7527661855451} {"canonical_length": 81, "canonical_nll": 1.4226288972077545, "generation_rouge_l_f1": 0.17765042979942694, "generation_token_f1": 0.2636103151862464, "prompt_id": "15d75a153e79b8a7", "reading_canonical_log_prob": -115.23294067382812, "reading_distances": [37, 41, 41, 42], "reading_log_probs": [-121.66873168945312, -125.74581909179688, -119.25517272949219, -123.92815399169922], "reading_logprob_std": 2.811020695005567, "tokenization_path_logprob_std": 92.08756277957545, "trajectory_balance_rmse": 2038.0613418699468} {"canonical_length": 4, "canonical_nll": 1.2988592386245728, "generation_rouge_l_f1": 0.0851063829787234, "generation_token_f1": 0.0851063829787234, "prompt_id": "9acb0ff71abeda7d", "reading_canonical_log_prob": -5.195436954498291, "reading_distances": [21, 16, 14, 15], "reading_log_probs": [-16.71489143371582, -14.639609336853027, -11.779657363891602, -8.308234214782715], "reading_logprob_std": 3.6475063583031044, "tokenization_path_logprob_std": 20.44536330143587, "trajectory_balance_rmse": 63.10367863250915} {"canonical_length": 114, "canonical_nll": 0.6419172454298588, "generation_rouge_l_f1": 0.20526315789473684, "generation_token_f1": 0.37894736842105264, "prompt_id": "1b111d554cbb9115", "reading_canonical_log_prob": -73.1785659790039, "reading_distances": [9, 12, 11, 10], "reading_log_probs": [-85.99668884277344, -82.77693176269531, -86.70562744140625, -84.26376342773438], "reading_logprob_std": 1.7673067570643455, "tokenization_path_logprob_std": 68.4314614345139, "trajectory_balance_rmse": 1459.677602513676} {"canonical_length": 135, "canonical_nll": 1.3706998471860532, "generation_rouge_l_f1": 0.1941747572815534, "generation_token_f1": 0.3640776699029126, "prompt_id": "4c468581b163e3ed", "reading_canonical_log_prob": -185.0444793701172, "reading_distances": [19, 23, 22, 28], "reading_log_probs": [-188.22897338867188, -185.83505249023438, -186.08367919921875, -188.4053497314453], "reading_logprob_std": 1.3669506895024113, "tokenization_path_logprob_std": 243.28442325196704, "trajectory_balance_rmse": 2743.2442121455597} {"canonical_length": 38, "canonical_nll": 0.891846405832391, "generation_rouge_l_f1": 0.23300970873786409, "generation_token_f1": 0.2524271844660194, "prompt_id": "5a9339c4c26875b0", "reading_canonical_log_prob": -33.89016342163086, "reading_distances": [20, 17, 15, 19], "reading_log_probs": [-37.92093276977539, -40.15513229370117, -35.57682800292969, -41.51567459106445], "reading_logprob_std": 2.605930539999014, "tokenization_path_logprob_std": 45.404865611165675, "trajectory_balance_rmse": 460.6650703459716} {"canonical_length": 245, "canonical_nll": 1.337932726801658, "generation_rouge_l_f1": 0.3058350100603622, "generation_token_f1": 0.482897384305835, "prompt_id": "dadb86ce514e9684", "reading_canonical_log_prob": -327.79351806640625, "reading_distances": [14, 16, 17, 17], "reading_log_probs": [-324.63909912109375, -322.70318603515625, -323.2264404296875, -324.15704345703125], "reading_logprob_std": 0.8769863300437399, "tokenization_path_logprob_std": 70.00489747421835, "trajectory_balance_rmse": 5248.349499668141} {"canonical_length": 153, "canonical_nll": 1.6267009061925553, "generation_rouge_l_f1": 0.18461538461538463, "generation_token_f1": 0.39487179487179486, "prompt_id": "8be6dabc44180a81", "reading_canonical_log_prob": -248.88523864746094, "reading_distances": [21, 22, 22, 24], "reading_log_probs": [-248.3231201171875, -251.05442810058594, -259.8227844238281, -257.68292236328125], "reading_logprob_std": 5.421471502212874, "tokenization_path_logprob_std": 168.48119598609023, "trajectory_balance_rmse": 3622.073472032519} {"canonical_length": 8, "canonical_nll": 1.1702277660369873, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "5786bd96f280e551", "reading_canonical_log_prob": -9.361822128295898, "reading_distances": [14, 23, 19, 22], "reading_log_probs": [-14.885632514953613, -17.146709442138672, -17.623462677001953, -13.721083641052246], "reading_logprob_std": 1.8519162342201252, "tokenization_path_logprob_std": 31.580995205462735, "trajectory_balance_rmse": 106.67183504036778} {"canonical_length": 103, "canonical_nll": 1.3492366457448421, "generation_rouge_l_f1": 0.16085790884718498, "generation_token_f1": 0.3324396782841823, "prompt_id": "edcbad1e6d6e267f", "reading_canonical_log_prob": -138.97137451171875, "reading_distances": [20, 28, 20, 24], "reading_log_probs": [-140.10614013671875, -138.8144073486328, -144.43626403808594, -148.6638946533203], "reading_logprob_std": 4.473479652204368, "tokenization_path_logprob_std": 203.52015842324684, "trajectory_balance_rmse": 2144.135060636296} {"canonical_length": 15, "canonical_nll": 1.9760457356770833, "generation_rouge_l_f1": 0.3384615384615385, "generation_token_f1": 0.3384615384615385, "prompt_id": "f125519b3e50ff0b", "reading_canonical_log_prob": -29.64068603515625, "reading_distances": [14, 14, 15, 14], "reading_log_probs": [-28.041397094726562, -29.034582138061523, -32.28110122680664, -30.279582977294922], "reading_logprob_std": 1.8272696455188775, "tokenization_path_logprob_std": 39.618168009167974, "trajectory_balance_rmse": 281.76816725291036} {"canonical_length": 335, "canonical_nll": 0.6902156089668843, "generation_rouge_l_f1": 0.2699822380106572, "generation_token_f1": 0.42984014209591476, "prompt_id": "805fd7652d88dc11", "reading_canonical_log_prob": -231.22222900390625, "reading_distances": [24, 26, 23, 25], "reading_log_probs": [-236.77835083007812, -238.609619140625, -234.1749725341797, -237.9347381591797], "reading_logprob_std": 1.9520447879463059, "tokenization_path_logprob_std": 167.09512224734374, "trajectory_balance_rmse": 4588.730280551544} {"canonical_length": 58, "canonical_nll": 1.45953395448882, "generation_rouge_l_f1": 0.29213483146067415, "generation_token_f1": 0.3820224719101123, "prompt_id": "7efa21552d311157", "reading_canonical_log_prob": -84.65296936035156, "reading_distances": [38, 36, 33, 36], "reading_log_probs": [-81.96827697753906, -83.7235107421875, -81.53913879394531, -84.88402557373047], "reading_logprob_std": 1.5565222473367861, "tokenization_path_logprob_std": 100.35665771125834, "trajectory_balance_rmse": 1489.938148938415} {"canonical_length": 337, "canonical_nll": 1.06898512118648, "generation_rouge_l_f1": 0.3181818181818182, "generation_token_f1": 0.5227272727272727, "prompt_id": "961872748d95ea69", "reading_canonical_log_prob": -360.24798583984375, "reading_distances": [27, 32, 27, 29], "reading_log_probs": [-375.64617919921875, -377.54840087890625, -374.43353271484375, -377.3243713378906], "reading_logprob_std": 1.4723798346067039, "tokenization_path_logprob_std": 355.9291770816018, "trajectory_balance_rmse": 7959.078768994791} {"canonical_length": 62, "canonical_nll": 0.851396253032069, "generation_rouge_l_f1": 0.2846153846153846, "generation_token_f1": 0.33076923076923076, "prompt_id": "e754a1aad6ea85da", "reading_canonical_log_prob": -52.78656768798828, "reading_distances": [6, 6, 9, 7], "reading_log_probs": [-49.78192138671875, -51.51549530029297, -50.559104919433594, -49.41069030761719], "reading_logprob_std": 0.9314347360329713, "tokenization_path_logprob_std": 82.35326407377664, "trajectory_balance_rmse": 885.1579666155159} {"canonical_length": 200, "canonical_nll": 2.00938720703125, "generation_rouge_l_f1": 0.1483050847457627, "generation_token_f1": 0.3516949152542373, "prompt_id": "4b3ba24fb950898f", "reading_canonical_log_prob": -401.87744140625, "reading_distances": [42, 39, 42, 44], "reading_log_probs": [-413.3886413574219, -411.0435791015625, -418.2554626464844, -411.99530029296875], "reading_logprob_std": 3.2045974259580112, "tokenization_path_logprob_std": 214.6228611516827, "trajectory_balance_rmse": 4933.756948056534} {"canonical_length": 33, "canonical_nll": 1.1921584389426492, "generation_rouge_l_f1": 0.10980392156862744, "generation_token_f1": 0.12549019607843137, "prompt_id": "7b02e9bcfb31548c", "reading_canonical_log_prob": -39.34122848510742, "reading_distances": [11, 16, 15, 12], "reading_log_probs": [-43.54468536376953, -45.13256072998047, -45.54804992675781, -49.54009246826172], "reading_logprob_std": 2.54975644342135, "tokenization_path_logprob_std": 86.28544113637629, "trajectory_balance_rmse": 715.5987888478486} {"canonical_length": 64, "canonical_nll": 0.7073568105697632, "generation_rouge_l_f1": 0.19512195121951217, "generation_token_f1": 0.25609756097560976, "prompt_id": "2e9002691e2c9828", "reading_canonical_log_prob": -45.270835876464844, "reading_distances": [33, 45, 41, 34], "reading_log_probs": [-60.5861930847168, -55.04359817504883, -56.71037292480469, -55.38829803466797], "reading_logprob_std": 2.539771828182532, "tokenization_path_logprob_std": 133.8585524395155, "trajectory_balance_rmse": 1168.580686625536} {"canonical_length": 110, "canonical_nll": 1.3341156005859376, "generation_rouge_l_f1": 0.2487046632124352, "generation_token_f1": 0.4041450777202072, "prompt_id": "660c03c6bae9ac45", "reading_canonical_log_prob": -146.75271606445312, "reading_distances": [31, 30, 29, 28], "reading_log_probs": [-147.52784729003906, -147.4603271484375, -144.44033813476562, -149.66452026367188], "reading_logprob_std": 2.1481294769636397, "tokenization_path_logprob_std": 166.63178044177695, "trajectory_balance_rmse": 2374.4171494901702} {"canonical_length": 129, "canonical_nll": 1.1394329218901404, "generation_rouge_l_f1": 0.23936170212765956, "generation_token_f1": 0.4255319148936171, "prompt_id": "dc5df7204109382e", "reading_canonical_log_prob": -146.98684692382812, "reading_distances": [34, 21, 31, 29], "reading_log_probs": [-147.19444274902344, -146.2755889892578, -145.65145874023438, -148.847412109375], "reading_logprob_std": 1.3897041128753487, "tokenization_path_logprob_std": 164.09403731686876, "trajectory_balance_rmse": 2608.3614285556027} {"canonical_length": 330, "canonical_nll": 1.2758629076408616, "generation_rouge_l_f1": 0.29372937293729373, "generation_token_f1": 0.5610561056105611, "prompt_id": "e7afe4dd1ceeee4d", "reading_canonical_log_prob": -421.0347595214844, "reading_distances": [24, 17, 18, 16], "reading_log_probs": [-421.4747009277344, -424.7118835449219, -425.11932373046875, -428.27880859375], "reading_logprob_std": 2.7828315048409507, "tokenization_path_logprob_std": 163.1909897712482, "trajectory_balance_rmse": 5822.825113419748} {"canonical_length": 142, "canonical_nll": 0.8303399959080656, "generation_rouge_l_f1": 0.18384401114206128, "generation_token_f1": 0.2618384401114206, "prompt_id": "64cb8edf4fa19a8d", "reading_canonical_log_prob": -117.90827941894531, "reading_distances": [14, 17, 19, 20], "reading_log_probs": [-126.27102661132812, -148.20225524902344, -153.96170043945312, -145.05062866210938], "reading_logprob_std": 11.982406613677615, "tokenization_path_logprob_std": 177.6613502965423, "trajectory_balance_rmse": 1893.017544858186} {"canonical_length": 67, "canonical_nll": 1.4882304632841652, "generation_rouge_l_f1": 0.1513157894736842, "generation_token_f1": 0.2631578947368421, "prompt_id": "d242cdc5b38c5c82", "reading_canonical_log_prob": -99.71144104003906, "reading_distances": [24, 28, 31, 27], "reading_log_probs": [-104.427734375, -108.62016296386719, -103.90792846679688, -105.21284484863281], "reading_logprob_std": 2.1209438065817414, "tokenization_path_logprob_std": 126.32511063194968, "trajectory_balance_rmse": 1432.4925806501283} {"canonical_length": 71, "canonical_nll": 3.14667177871919, "generation_rouge_l_f1": 0.2574257425742574, "generation_token_f1": 0.31683168316831684, "prompt_id": "406fdefc259c7cdd", "reading_canonical_log_prob": -223.4136962890625, "reading_distances": [32, 31, 32, 35], "reading_log_probs": [-232.9736785888672, -235.7882843017578, -235.66925048828125, -238.30430603027344], "reading_logprob_std": 2.177379047530882, "tokenization_path_logprob_std": 120.07480439006018, "trajectory_balance_rmse": 1261.1177283223262} {"canonical_length": 112, "canonical_nll": 1.230825696672712, "generation_rouge_l_f1": 0.23592493297587133, "generation_token_f1": 0.4075067024128686, "prompt_id": "8b97cb84ee36bc33", "reading_canonical_log_prob": -137.85247802734375, "reading_distances": [27, 21, 23, 24], "reading_log_probs": [-142.72882080078125, -144.68820190429688, -141.08384704589844, -144.72064208984375], "reading_logprob_std": 1.7495515976289917, "tokenization_path_logprob_std": 155.88390915062922, "trajectory_balance_rmse": 2351.7505593151313} {"canonical_length": 3, "canonical_nll": 1.624506155649821, "generation_rouge_l_f1": 0.03636363636363636, "generation_token_f1": 0.03636363636363636, "prompt_id": "3c3cb4aabb9bd15f", "reading_canonical_log_prob": -4.873518466949463, "reading_distances": [24, 22, 21, 22], "reading_log_probs": [-5.036066055297852, -6.3817949295043945, -6.439653396606445, -5.560227870941162], "reading_logprob_std": 0.6774645455456647, "tokenization_path_logprob_std": 14.103625868473337, "trajectory_balance_rmse": 37.678425831390165} {"canonical_length": 76, "canonical_nll": 2.343923468338816, "generation_rouge_l_f1": 0.15492957746478875, "generation_token_f1": 0.28169014084507044, "prompt_id": "4d20c1324c854a9e", "reading_canonical_log_prob": -178.13818359375, "reading_distances": [11, 17, 12, 15], "reading_log_probs": [-179.5197296142578, -175.31640625, -177.90419006347656, -175.5406494140625], "reading_logprob_std": 2.0092330428879484, "tokenization_path_logprob_std": 199.7193489193653, "trajectory_balance_rmse": 1322.3053141439864} {"canonical_length": 194, "canonical_nll": 0.8506219608267558, "generation_rouge_l_f1": 0.37939110070257615, "generation_token_f1": 0.5480093676814988, "prompt_id": "45005e42ada68a08", "reading_canonical_log_prob": -165.02066040039062, "reading_distances": [25, 26, 20, 23], "reading_log_probs": [-157.8638916015625, -158.0363006591797, -158.26629638671875, -156.2595672607422], "reading_logprob_std": 0.912969170420784, "tokenization_path_logprob_std": 152.66922931923264, "trajectory_balance_rmse": 2312.6006136058368} {"canonical_length": 3, "canonical_nll": 1.5779444376627605, "generation_rouge_l_f1": 0.14285714285714288, "generation_token_f1": 0.14285714285714288, "prompt_id": "1b11c060139de682", "reading_canonical_log_prob": -4.733833312988281, "reading_distances": [17, 18, 19, 19], "reading_log_probs": [-6.306671142578125, -5.736741065979004, -6.27664852142334, -6.513401031494141], "reading_logprob_std": 0.33156339394698053, "tokenization_path_logprob_std": 20.31175468093034, "trajectory_balance_rmse": 35.090725924433364} {"canonical_length": 180, "canonical_nll": 1.5486326429578994, "generation_rouge_l_f1": 0.24766355140186916, "generation_token_f1": 0.42990654205607476, "prompt_id": "fb27819d14f6e258", "reading_canonical_log_prob": -278.7538757324219, "reading_distances": [18, 20, 19, 19], "reading_log_probs": [-289.7762451171875, -299.8358154296875, -288.3120422363281, -297.2088623046875], "reading_logprob_std": 5.608284659516864, "tokenization_path_logprob_std": 218.56890030197306, "trajectory_balance_rmse": 4083.4358745791965} {"canonical_length": 92, "canonical_nll": 1.8235811979874321, "generation_rouge_l_f1": 0.17679558011049723, "generation_token_f1": 0.30939226519337015, "prompt_id": "b1a0ce36dec4c1a2", "reading_canonical_log_prob": -167.76947021484375, "reading_distances": [31, 34, 28, 28], "reading_log_probs": [-174.73150634765625, -175.874267578125, -170.4536590576172, -172.28036499023438], "reading_logprob_std": 2.4366979317806403, "tokenization_path_logprob_std": 179.49887653327323, "trajectory_balance_rmse": 1844.3860441191896} {"canonical_length": 205, "canonical_nll": 1.3261897389481707, "generation_rouge_l_f1": 0.18723404255319148, "generation_token_f1": 0.31489361702127655, "prompt_id": "fe99f791079b2601", "reading_canonical_log_prob": -271.868896484375, "reading_distances": [21, 31, 27, 25], "reading_log_probs": [-284.46771240234375, -290.6544189453125, -286.0123596191406, -285.6978454589844], "reading_logprob_std": 2.7140031147154153, "tokenization_path_logprob_std": 194.46778119116792, "trajectory_balance_rmse": 3634.7958229592377} {"canonical_length": 265, "canonical_nll": 1.7936078917305425, "generation_rouge_l_f1": 0.22519083969465647, "generation_token_f1": 0.44656488549618323, "prompt_id": "7fb35b5d3ec740ba", "reading_canonical_log_prob": -475.30609130859375, "reading_distances": [22, 26, 27, 27], "reading_log_probs": [-481.8858337402344, -484.5378112792969, -480.3522033691406, -478.763427734375], "reading_logprob_std": 2.458343838971538, "tokenization_path_logprob_std": 111.68283710901885, "trajectory_balance_rmse": 6448.831328911523} {"canonical_length": 11, "canonical_nll": 0.2943045876242898, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.6285714285714286, "prompt_id": "cb7972227cbb9dc7", "reading_canonical_log_prob": -3.2373504638671875, "reading_distances": [29, 19, 21, 21], "reading_log_probs": [-4.411165237426758, -3.7592203617095947, -3.2163398265838623, -3.0705018043518066], "reading_logprob_std": 0.6082995136409659, "tokenization_path_logprob_std": 68.6932047665722, "trajectory_balance_rmse": 243.6355661959429} {"canonical_length": 109, "canonical_nll": 1.931182301372563, "generation_rouge_l_f1": 0.14332247557003258, "generation_token_f1": 0.3127035830618892, "prompt_id": "ce54506448843c44", "reading_canonical_log_prob": -210.49887084960938, "reading_distances": [22, 21, 20, 20], "reading_log_probs": [-211.4232635498047, -213.6783447265625, -214.00225830078125, -211.7230224609375], "reading_logprob_std": 1.3212866716379115, "tokenization_path_logprob_std": 51.30106993397239, "trajectory_balance_rmse": 1989.8531330112075} {"canonical_length": 90, "canonical_nll": 0.6985735999213325, "generation_rouge_l_f1": 0.2526315789473684, "generation_token_f1": 0.3508771929824562, "prompt_id": "110898cad92a62bf", "reading_canonical_log_prob": -62.87162399291992, "reading_distances": [12, 16, 16, 17], "reading_log_probs": [-65.4766845703125, -68.7655258178711, -67.05813598632812, -66.32717895507812], "reading_logprob_std": 1.3974923873692748, "tokenization_path_logprob_std": 119.31902494890828, "trajectory_balance_rmse": 1096.3685014261725} {"canonical_length": 6, "canonical_nll": 1.0626872380574544, "generation_rouge_l_f1": 0.08695652173913045, "generation_token_f1": 0.08695652173913045, "prompt_id": "bb43b5a4398228cc", "reading_canonical_log_prob": -6.376123428344727, "reading_distances": [34, 27, 22, 30], "reading_log_probs": [-17.005077362060547, -18.358810424804688, -20.045961380004883, -17.07716941833496], "reading_logprob_std": 1.42558683259174, "tokenization_path_logprob_std": 23.66083531370116, "trajectory_balance_rmse": 111.13953304408221} {"canonical_length": 344, "canonical_nll": 1.3131666848825854, "generation_rouge_l_f1": 0.20289855072463767, "generation_token_f1": 0.3550724637681159, "prompt_id": "6dc0804e4c40522b", "reading_canonical_log_prob": -451.7293395996094, "reading_distances": [16, 21, 19, 19], "reading_log_probs": [-468.57232666015625, -463.4148254394531, -471.11883544921875, -471.8951721191406], "reading_logprob_std": 3.8296711832516923, "tokenization_path_logprob_std": 173.68370905981013, "trajectory_balance_rmse": 5678.253699022825} {"canonical_length": 325, "canonical_nll": 1.1572716346153846, "generation_rouge_l_f1": 0.26013513513513514, "generation_token_f1": 0.46621621621621623, "prompt_id": "8ed2c6a41c30f4da", "reading_canonical_log_prob": -376.11328125, "reading_distances": [30, 35, 29, 23], "reading_log_probs": [-375.8719482421875, -379.2647705078125, -376.62066650390625, -378.1043701171875], "reading_logprob_std": 1.5164300415677487, "tokenization_path_logprob_std": 33.71623860771175, "trajectory_balance_rmse": 7580.057820265621} {"canonical_length": 116, "canonical_nll": 0.7489460254537648, "generation_rouge_l_f1": 0.28493150684931506, "generation_token_f1": 0.4767123287671233, "prompt_id": "b2cdbde556d30889", "reading_canonical_log_prob": -86.87773895263672, "reading_distances": [15, 15, 13, 12], "reading_log_probs": [-87.9789047241211, -89.54727172851562, -85.77488708496094, -84.88255310058594], "reading_logprob_std": 2.1152542776051653, "tokenization_path_logprob_std": 131.22145201848414, "trajectory_balance_rmse": 1327.1997796409637} {"canonical_length": 15, "canonical_nll": 1.3308114369710287, "generation_rouge_l_f1": 0.5599999999999999, "generation_token_f1": 0.5599999999999999, "prompt_id": "eb54d1c2e52f2aa1", "reading_canonical_log_prob": -19.96217155456543, "reading_distances": [21, 23, 21, 19], "reading_log_probs": [-18.910213470458984, -18.4296875, -18.944950103759766, -18.738834381103516], "reading_logprob_std": 0.2354134077005747, "tokenization_path_logprob_std": 37.40072217423574, "trajectory_balance_rmse": 308.40599239558526} {"canonical_length": 61, "canonical_nll": 1.5088454699907146, "generation_rouge_l_f1": 0.3957219251336898, "generation_token_f1": 0.4598930481283423, "prompt_id": "2d9f537706729121", "reading_canonical_log_prob": -92.0395736694336, "reading_distances": [11, 15, 10, 10], "reading_log_probs": [-89.37062072753906, -89.73491668701172, -91.2440185546875, -89.46156311035156], "reading_logprob_std": 0.874636091839651, "tokenization_path_logprob_std": 56.85267260392961, "trajectory_balance_rmse": 1231.8750176334308} {"canonical_length": 2, "canonical_nll": 1.9942615032196045, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "e6a02b338700cb67", "reading_canonical_log_prob": -3.988523006439209, "reading_distances": [22, 19, 25, 21], "reading_log_probs": [-4.927846908569336, -3.953230857849121, -4.2201151847839355, -3.7802717685699463], "reading_logprob_std": 0.5051644668185521, "tokenization_path_logprob_std": 24.144932519181243, "trajectory_balance_rmse": 43.019062904463006} {"canonical_length": 169, "canonical_nll": 0.8616919884314904, "generation_rouge_l_f1": 0.2755344418052256, "generation_token_f1": 0.44655581947743467, "prompt_id": "56062aaca5480bb3", "reading_canonical_log_prob": -145.62594604492188, "reading_distances": [27, 24, 27, 23], "reading_log_probs": [-148.22586059570312, -148.32583618164062, -147.481201171875, -146.39236450195312], "reading_logprob_std": 0.8927255564990297, "tokenization_path_logprob_std": 134.34342100347612, "trajectory_balance_rmse": 1993.4301171992322} {"canonical_length": 3, "canonical_nll": 0.5414434274037679, "generation_rouge_l_f1": 0.10526315789473684, "generation_token_f1": 0.10526315789473684, "prompt_id": "a6378b47a0666677", "reading_canonical_log_prob": -1.6243302822113037, "reading_distances": [29, 31, 27, 30], "reading_log_probs": [-2.2709784507751465, -3.454357624053955, -3.0212461948394775, -2.679255962371826], "reading_logprob_std": 0.5029333866735299, "tokenization_path_logprob_std": 21.593470245068968, "trajectory_balance_rmse": 60.98764771299053} {"canonical_length": 199, "canonical_nll": 1.099143560208268, "generation_rouge_l_f1": 0.3615560640732265, "generation_token_f1": 0.5720823798627002, "prompt_id": "41d1448457af2a4f", "reading_canonical_log_prob": -218.7295684814453, "reading_distances": [13, 18, 21, 16], "reading_log_probs": [-230.09378051757812, -227.68917846679688, -231.75991821289062, -225.3192138671875], "reading_logprob_std": 2.8140264641904897, "tokenization_path_logprob_std": 140.2852812547755, "trajectory_balance_rmse": 4423.56405241665} {"canonical_length": 94, "canonical_nll": 1.8830592378656914, "generation_rouge_l_f1": 0.17751479289940827, "generation_token_f1": 0.2781065088757397, "prompt_id": "5749ffa6095e5a40", "reading_canonical_log_prob": -177.007568359375, "reading_distances": [31, 23, 22, 19], "reading_log_probs": [-174.657958984375, -176.40802001953125, -176.76040649414062, -178.0992431640625], "reading_logprob_std": 1.4172253274230544, "tokenization_path_logprob_std": 138.87790746289184, "trajectory_balance_rmse": 2182.9951372932214} {"canonical_length": 58, "canonical_nll": 1.3547018643083244, "generation_rouge_l_f1": 0.26153846153846155, "generation_token_f1": 0.41538461538461535, "prompt_id": "e503fe8212044208", "reading_canonical_log_prob": -78.57270812988281, "reading_distances": [31, 31, 34, 34], "reading_log_probs": [-75.353271484375, -77.65940856933594, -83.7754898071289, -74.66932678222656], "reading_logprob_std": 4.143157551145278, "tokenization_path_logprob_std": 61.629919986801994, "trajectory_balance_rmse": 1514.9709462822834} {"canonical_length": 161, "canonical_nll": 0.97540889763684, "generation_rouge_l_f1": 0.2580645161290323, "generation_token_f1": 0.5017921146953406, "prompt_id": "cf2e46f2d41b486e", "reading_canonical_log_prob": -157.04083251953125, "reading_distances": [25, 21, 22, 22], "reading_log_probs": [-157.90679931640625, -158.3642578125, -157.99020385742188, -155.42311096191406], "reading_logprob_std": 1.3467602956582618, "tokenization_path_logprob_std": 189.50192266843885, "trajectory_balance_rmse": 2954.644492886987} {"canonical_length": 62, "canonical_nll": 0.6893789845128213, "generation_rouge_l_f1": 0.3037974683544304, "generation_token_f1": 0.3544303797468354, "prompt_id": "45e020ec1f1a052d", "reading_canonical_log_prob": -42.74149703979492, "reading_distances": [12, 12, 12, 18], "reading_log_probs": [-44.079254150390625, -46.39651107788086, -46.55066680908203, -45.797691345214844], "reading_logprob_std": 1.1321037828226155, "tokenization_path_logprob_std": 27.633641062400837, "trajectory_balance_rmse": 1021.135992950033} {"canonical_length": 284, "canonical_nll": 1.4465686636911312, "generation_rouge_l_f1": 0.2246376811594203, "generation_token_f1": 0.5, "prompt_id": "238ad0e8d97d1bdf", "reading_canonical_log_prob": -410.82550048828125, "reading_distances": [13, 19, 17, 26], "reading_log_probs": [-429.3948974609375, -425.02386474609375, -435.0646057128906, -426.93865966796875], "reading_logprob_std": 4.356972770095468, "tokenization_path_logprob_std": 300.2984267357353, "trajectory_balance_rmse": 5187.522855853252} {"canonical_length": 30, "canonical_nll": 0.8230169932047526, "generation_rouge_l_f1": 0.07171314741035856, "generation_token_f1": 0.07171314741035856, "prompt_id": "aca807bc21ebbe28", "reading_canonical_log_prob": -24.690509796142578, "reading_distances": [9, 11, 9, 11], "reading_log_probs": [-35.26555633544922, -25.69086456298828, -20.018714904785156, -30.943771362304688], "reading_logprob_std": 6.595085870624246, "tokenization_path_logprob_std": 20.283010214150103, "trajectory_balance_rmse": 416.562789255012} {"canonical_length": 39, "canonical_nll": 1.0193864871294072, "generation_rouge_l_f1": 0.26262626262626265, "generation_token_f1": 0.30303030303030304, "prompt_id": "1c53f5181f02d61c", "reading_canonical_log_prob": -39.756072998046875, "reading_distances": [21, 21, 21, 22], "reading_log_probs": [-47.2385139465332, -45.723018646240234, -47.2462043762207, -47.13249588012695], "reading_logprob_std": 0.7431731203241704, "tokenization_path_logprob_std": 35.57377322796554, "trajectory_balance_rmse": 441.0126665663961} {"canonical_length": 10, "canonical_nll": 3.36035041809082, "generation_rouge_l_f1": 0.36363636363636365, "generation_token_f1": 0.36363636363636365, "prompt_id": "1a83cc65795284e0", "reading_canonical_log_prob": -33.6035041809082, "reading_distances": [21, 14, 22, 17], "reading_log_probs": [-27.31140899658203, -28.87749481201172, -27.03780746459961, -28.64046287536621], "reading_logprob_std": 0.9265979365930537, "tokenization_path_logprob_std": 61.141144311938795, "trajectory_balance_rmse": 144.56288056851872} {"canonical_length": 30, "canonical_nll": 1.665926996866862, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.463768115942029, "prompt_id": "a5ddf34551bf57f2", "reading_canonical_log_prob": -49.97780990600586, "reading_distances": [24, 20, 18, 17], "reading_log_probs": [-48.329200744628906, -51.089149475097656, -47.534446716308594, -49.46675109863281], "reading_logprob_std": 1.5423159044515262, "tokenization_path_logprob_std": 97.28026770268754, "trajectory_balance_rmse": 866.4601979146237} {"canonical_length": 33, "canonical_nll": 0.8438203985040839, "generation_rouge_l_f1": 0.06451612903225806, "generation_token_f1": 0.06451612903225806, "prompt_id": "626308b96b881737", "reading_canonical_log_prob": -27.846073150634766, "reading_distances": [22, 18, 14, 14], "reading_log_probs": [-30.860679626464844, -31.105388641357422, -33.23359680175781, -30.690427780151367], "reading_logprob_std": 1.186338037714152, "tokenization_path_logprob_std": 12.33383708385424, "trajectory_balance_rmse": 671.6730566087083} {"canonical_length": 79, "canonical_nll": 0.7899691907665397, "generation_rouge_l_f1": 0.14067278287461774, "generation_token_f1": 0.19571865443425077, "prompt_id": "b884d1bbe7793313", "reading_canonical_log_prob": -62.40756607055664, "reading_distances": [19, 23, 24, 21], "reading_log_probs": [-69.25619506835938, -74.74956512451172, -67.77152252197266, -66.13023376464844], "reading_logprob_std": 3.7397947989051423, "tokenization_path_logprob_std": 52.23988309050335, "trajectory_balance_rmse": 954.2302646069569} {"canonical_length": 50, "canonical_nll": 1.5189080810546876, "generation_rouge_l_f1": 0.2386363636363636, "generation_token_f1": 0.3068181818181818, "prompt_id": "d8395ff4ce8fbfae", "reading_canonical_log_prob": -75.94540405273438, "reading_distances": [38, 31, 29, 29], "reading_log_probs": [-81.65582275390625, -81.04574584960938, -81.85169219970703, -80.39958190917969], "reading_logprob_std": 0.6560234613383905, "tokenization_path_logprob_std": 130.59077452498047, "trajectory_balance_rmse": 1103.4707967544832} {"canonical_length": 228, "canonical_nll": 1.509675946152001, "generation_rouge_l_f1": 0.2326530612244898, "generation_token_f1": 0.4979591836734693, "prompt_id": "cabaeed82d3c2f7b", "reading_canonical_log_prob": -344.20611572265625, "reading_distances": [24, 21, 27, 23], "reading_log_probs": [-348.8831481933594, -348.5140380859375, -345.65167236328125, -347.06048583984375], "reading_logprob_std": 1.4773705087307694, "tokenization_path_logprob_std": 160.30751772023623, "trajectory_balance_rmse": 5441.38034361745} {"canonical_length": 60, "canonical_nll": 0.07974156538645426, "generation_rouge_l_f1": 0.6486486486486486, "generation_token_f1": 0.954954954954955, "prompt_id": "ef7598410b16ef36", "reading_canonical_log_prob": -4.784493923187256, "reading_distances": [32, 36, 31, 32], "reading_log_probs": [-8.354711532592773, -7.603092193603516, -5.552677631378174, -5.726247310638428], "reading_logprob_std": 1.3869037098770622, "tokenization_path_logprob_std": 156.4878591454637, "trajectory_balance_rmse": 1145.2516447545224} {"canonical_length": 239, "canonical_nll": 1.2851559946227773, "generation_rouge_l_f1": 0.3354297693920335, "generation_token_f1": 0.5366876310272537, "prompt_id": "39157986b3a30c4b", "reading_canonical_log_prob": -307.15228271484375, "reading_distances": [42, 44, 40, 40], "reading_log_probs": [-307.49737548828125, -299.2003173828125, -308.8614196777344, -308.61322021484375], "reading_logprob_std": 4.600254879367091, "tokenization_path_logprob_std": 110.09398876779323, "trajectory_balance_rmse": 4245.091034868026} {"canonical_length": 113, "canonical_nll": 1.7486589819984097, "generation_rouge_l_f1": 0.2541436464088398, "generation_token_f1": 0.3535911602209945, "prompt_id": "952e6be856268bb7", "reading_canonical_log_prob": -197.5984649658203, "reading_distances": [25, 27, 29, 33], "reading_log_probs": [-199.7388458251953, -211.51544189453125, -194.83029174804688, -203.97088623046875], "reading_logprob_std": 7.06845873801002, "tokenization_path_logprob_std": 105.80830871064435, "trajectory_balance_rmse": 2275.2703963169624} {"canonical_length": 324, "canonical_nll": 0.9138532096956983, "generation_rouge_l_f1": 0.23121387283236994, "generation_token_f1": 0.3930635838150289, "prompt_id": "b30cec3b9127ceb8", "reading_canonical_log_prob": -296.08843994140625, "reading_distances": [24, 19, 21, 20], "reading_log_probs": [-304.3084716796875, -304.796142578125, -305.91748046875, -302.259521484375], "reading_logprob_std": 1.5301807698469512, "tokenization_path_logprob_std": 275.04133622585425, "trajectory_balance_rmse": 6760.859887841715} {"canonical_length": 211, "canonical_nll": 1.2081162873037619, "generation_rouge_l_f1": 0.23419203747072598, "generation_token_f1": 0.4355971896955504, "prompt_id": "ce18c132bb5bd7f2", "reading_canonical_log_prob": -254.91253662109375, "reading_distances": [25, 28, 28, 24], "reading_log_probs": [-257.4071960449219, -257.0425109863281, -255.39505004882812, -258.77410888671875], "reading_logprob_std": 1.3898673461215398, "tokenization_path_logprob_std": 418.2924132127256, "trajectory_balance_rmse": 4387.845007541657} {"canonical_length": 53, "canonical_nll": 1.4499608165812943, "generation_rouge_l_f1": 0.08955223880597016, "generation_token_f1": 0.16716417910447762, "prompt_id": "f1a703b446eae2b2", "reading_canonical_log_prob": -76.8479232788086, "reading_distances": [54, 59, 45, 56], "reading_log_probs": [-82.18938446044922, -80.57261657714844, -79.24674987792969, -83.00477600097656], "reading_logprob_std": 1.6766529741221607, "tokenization_path_logprob_std": 73.60406370301807, "trajectory_balance_rmse": 970.5540051719388} {"canonical_length": 56, "canonical_nll": 1.7012721470424108, "generation_rouge_l_f1": 0.1804511278195489, "generation_token_f1": 0.28571428571428575, "prompt_id": "7188e613a859c6bb", "reading_canonical_log_prob": -95.271240234375, "reading_distances": [26, 25, 28, 24], "reading_log_probs": [-102.6282958984375, -91.47330474853516, -101.20074462890625, -102.02842712402344], "reading_logprob_std": 5.272177965734747, "tokenization_path_logprob_std": 96.58771038212805, "trajectory_balance_rmse": 1006.2848809391373} {"canonical_length": 198, "canonical_nll": 1.7005134351325757, "generation_rouge_l_f1": 0.30068337129840544, "generation_token_f1": 0.5056947608200456, "prompt_id": "85df6700a3ed9560", "reading_canonical_log_prob": -336.70166015625, "reading_distances": [40, 30, 32, 41], "reading_log_probs": [-347.25128173828125, -337.5235595703125, -337.5125732421875, -344.32489013671875], "reading_logprob_std": 4.921895546088149, "tokenization_path_logprob_std": 111.42998852413102, "trajectory_balance_rmse": 3931.6091365424177} {"canonical_length": 94, "canonical_nll": 1.3757676469518783, "generation_rouge_l_f1": 0.18328840970350405, "generation_token_f1": 0.3126684636118598, "prompt_id": "0baa5b0629968fb8", "reading_canonical_log_prob": -129.32215881347656, "reading_distances": [26, 24, 29, 21], "reading_log_probs": [-138.393798828125, -138.41128540039062, -134.85223388671875, -140.58714294433594], "reading_logprob_std": 2.3742390369368533, "tokenization_path_logprob_std": 73.79665589173852, "trajectory_balance_rmse": 2331.0878632286276} {"canonical_length": 331, "canonical_nll": 1.1820858497273885, "generation_rouge_l_f1": 0.21478873239436622, "generation_token_f1": 0.4929577464788733, "prompt_id": "3d704d3c66ac7a28", "reading_canonical_log_prob": -391.2704162597656, "reading_distances": [25, 32, 31, 32], "reading_log_probs": [-389.55633544921875, -390.52020263671875, -388.8435363769531, -393.28076171875], "reading_logprob_std": 1.9457076355326874, "tokenization_path_logprob_std": 134.7924787091062, "trajectory_balance_rmse": 6016.435311160839} {"canonical_length": 14, "canonical_nll": 0.6421580995832171, "generation_rouge_l_f1": 0.6363636363636364, "generation_token_f1": 0.6363636363636364, "prompt_id": "d990b55cc5500b31", "reading_canonical_log_prob": -8.990213394165039, "reading_distances": [59, 60, 60, 60], "reading_log_probs": [-8.587112426757812, -9.100207328796387, -7.877326965332031, -8.58211898803711], "reading_logprob_std": 0.502300575819712, "tokenization_path_logprob_std": 38.66831488602737, "trajectory_balance_rmse": 316.1771229280041} {"canonical_length": 125, "canonical_nll": 1.730137451171875, "generation_rouge_l_f1": 0.2841823056300268, "generation_token_f1": 0.41823056300268097, "prompt_id": "96948513433b0e29", "reading_canonical_log_prob": -216.26718139648438, "reading_distances": [14, 19, 14, 18], "reading_log_probs": [-221.71539306640625, -219.31768798828125, -219.25460815429688, -218.83273315429688], "reading_logprob_std": 1.3080302929181755, "tokenization_path_logprob_std": 191.0942442930997, "trajectory_balance_rmse": 2872.061472353953} {"canonical_length": 88, "canonical_nll": 1.7932454889470881, "generation_rouge_l_f1": 0.33480176211453744, "generation_token_f1": 0.4669603524229075, "prompt_id": "5b5a0a9c17c96287", "reading_canonical_log_prob": -157.80560302734375, "reading_distances": [43, 45, 35, 38], "reading_log_probs": [-169.07327270507812, -170.43609619140625, -171.34642028808594, -167.51345825195312], "reading_logprob_std": 1.671317079774744, "tokenization_path_logprob_std": 144.6474632701504, "trajectory_balance_rmse": 2457.315977281499} {"canonical_length": 310, "canonical_nll": 1.3558241321194557, "generation_rouge_l_f1": 0.27949183303085295, "generation_token_f1": 0.5408348457350272, "prompt_id": "9ae64366f18d52ad", "reading_canonical_log_prob": -420.30548095703125, "reading_distances": [20, 27, 15, 21], "reading_log_probs": [-435.37841796875, -431.39324951171875, -428.39202880859375, -433.13250732421875], "reading_logprob_std": 2.947312274988787, "tokenization_path_logprob_std": 155.1702374642809, "trajectory_balance_rmse": 6950.195814792922} {"canonical_length": 192, "canonical_nll": 1.3827284177144368, "generation_rouge_l_f1": 0.20454545454545453, "generation_token_f1": 0.4363636363636364, "prompt_id": "eaabe2e9ef0c1d39", "reading_canonical_log_prob": -265.4838562011719, "reading_distances": [28, 29, 31, 35], "reading_log_probs": [-284.96478271484375, -283.57537841796875, -273.70172119140625, -288.8389587402344], "reading_logprob_std": 6.4428680499116515, "tokenization_path_logprob_std": 304.7740277893854, "trajectory_balance_rmse": 4195.377467363224} {"canonical_length": 208, "canonical_nll": 1.517182863675631, "generation_rouge_l_f1": 0.28210526315789475, "generation_token_f1": 0.5221052631578946, "prompt_id": "74426ee6c2f5003e", "reading_canonical_log_prob": -315.57403564453125, "reading_distances": [20, 21, 28, 22], "reading_log_probs": [-325.53533935546875, -323.1500244140625, -324.9849548339844, -328.38128662109375], "reading_logprob_std": 2.1671834613675727, "tokenization_path_logprob_std": 157.7277633215922, "trajectory_balance_rmse": 4632.347808315344} {"canonical_length": 341, "canonical_nll": 1.3118696911943273, "generation_rouge_l_f1": 0.3420647149460709, "generation_token_f1": 0.576271186440678, "prompt_id": "6909a2a1a6b2d043", "reading_canonical_log_prob": -447.3475646972656, "reading_distances": [36, 44, 34, 33], "reading_log_probs": [-465.598876953125, -474.605712890625, -469.6709289550781, -471.7733154296875], "reading_logprob_std": 3.792789083339424, "tokenization_path_logprob_std": 356.4613978397814, "trajectory_balance_rmse": 5532.553232467566} {"canonical_length": 19, "canonical_nll": 1.6200170015033923, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4799999999999999, "prompt_id": "fcaeb274bce92488", "reading_canonical_log_prob": -30.780323028564453, "reading_distances": [24, 30, 32, 34], "reading_log_probs": [-36.21379470825195, -54.61771774291992, -33.345252990722656, -41.473602294921875], "reading_logprob_std": 9.424940673795735, "tokenization_path_logprob_std": 28.57710542014322, "trajectory_balance_rmse": 310.977418158221} {"canonical_length": 20, "canonical_nll": 0.6327784538269043, "generation_rouge_l_f1": 0.12578616352201258, "generation_token_f1": 0.13836477987421383, "prompt_id": "59d62ad713859ec4", "reading_canonical_log_prob": -12.655569076538086, "reading_distances": [13, 15, 14, 16], "reading_log_probs": [-14.94149398803711, -12.220111846923828, -13.127281188964844, -18.26007843017578], "reading_logprob_std": 2.6670841068979994, "tokenization_path_logprob_std": 45.262759552699364, "trajectory_balance_rmse": 190.68035071738123} {"canonical_length": 227, "canonical_nll": 1.3213205211488162, "generation_rouge_l_f1": 0.20392156862745095, "generation_token_f1": 0.38431372549019605, "prompt_id": "0205b5c76a68ce56", "reading_canonical_log_prob": -299.93975830078125, "reading_distances": [32, 25, 25, 26], "reading_log_probs": [-322.1390686035156, -312.3348083496094, -326.36083984375, -312.96209716796875], "reading_logprob_std": 6.921061047556848, "tokenization_path_logprob_std": 157.37285420067388, "trajectory_balance_rmse": 3842.972847523604} {"canonical_length": 251, "canonical_nll": 1.560176773375249, "generation_rouge_l_f1": 0.20158102766798422, "generation_token_f1": 0.4782608695652174, "prompt_id": "28e138f0fd885003", "reading_canonical_log_prob": -391.6043701171875, "reading_distances": [23, 25, 29, 28], "reading_log_probs": [-403.7083740234375, -401.0279846191406, -409.92279052734375, -404.68231201171875], "reading_logprob_std": 3.7269995858731764, "tokenization_path_logprob_std": 125.9570882800957, "trajectory_balance_rmse": 4538.068545123759} {"canonical_length": 25, "canonical_nll": 0.23832878112792968, "generation_rouge_l_f1": 0.8695652173913043, "generation_token_f1": 0.8695652173913043, "prompt_id": "3cdea1f3e20549fa", "reading_canonical_log_prob": -5.958219528198242, "reading_distances": [31, 36, 33, 37], "reading_log_probs": [-11.455368995666504, -11.553264617919922, -8.509830474853516, -11.276679992675781], "reading_logprob_std": 1.4637895221711483, "tokenization_path_logprob_std": 40.33219932339865, "trajectory_balance_rmse": 430.41567488634} {"canonical_length": 143, "canonical_nll": 2.1405071978802446, "generation_rouge_l_f1": 0.18446601941747573, "generation_token_f1": 0.37378640776699035, "prompt_id": "b7ff5b60544bcb0d", "reading_canonical_log_prob": -306.092529296875, "reading_distances": [17, 16, 16, 18], "reading_log_probs": [-326.83636474609375, -333.93377685546875, -331.6239318847656, -325.7936706542969], "reading_logprob_std": 3.8726598040753175, "tokenization_path_logprob_std": 185.087489179374, "trajectory_balance_rmse": 2798.0960441574143} {"canonical_length": 297, "canonical_nll": 1.3123903627748843, "generation_rouge_l_f1": 0.2610364683301344, "generation_token_f1": 0.5297504798464491, "prompt_id": "2c96188d7d0e9844", "reading_canonical_log_prob": -389.7799377441406, "reading_distances": [20, 19, 18, 15], "reading_log_probs": [-390.4190673828125, -392.2711181640625, -390.09918212890625, -393.79632568359375], "reading_logprob_std": 1.7236281231896546, "tokenization_path_logprob_std": 233.598424470628, "trajectory_balance_rmse": 5348.612310517799} {"canonical_length": 101, "canonical_nll": 1.0793691200785118, "generation_rouge_l_f1": 0.38620689655172413, "generation_token_f1": 0.5517241379310345, "prompt_id": "f320a0730139386c", "reading_canonical_log_prob": -109.01628112792969, "reading_distances": [35, 41, 38, 37], "reading_log_probs": [-114.09321594238281, -111.06389617919922, -108.49237823486328, -110.31037902832031], "reading_logprob_std": 2.333481588594284, "tokenization_path_logprob_std": 138.21121911610626, "trajectory_balance_rmse": 1526.6493211841437} {"canonical_length": 109, "canonical_nll": 1.895912100415711, "generation_rouge_l_f1": 0.2576271186440678, "generation_token_f1": 0.4677966101694915, "prompt_id": "2a4224ce9ad0c001", "reading_canonical_log_prob": -206.6544189453125, "reading_distances": [17, 13, 19, 14], "reading_log_probs": [-213.20965576171875, -229.11468505859375, -224.05218505859375, -219.82254028320312], "reading_logprob_std": 6.733766874772129, "tokenization_path_logprob_std": 127.63100962081496, "trajectory_balance_rmse": 2193.4556838695385} {"canonical_length": 20, "canonical_nll": 0.595949125289917, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5833333333333334, "prompt_id": "af538126b4c7deeb", "reading_canonical_log_prob": -11.91898250579834, "reading_distances": [27, 37, 35, 32], "reading_log_probs": [-18.444461822509766, -19.324867248535156, -18.284944534301758, -28.34178352355957], "reading_logprob_std": 4.850118255875472, "tokenization_path_logprob_std": 51.072311841490794, "trajectory_balance_rmse": 340.5876785533998} {"canonical_length": 48, "canonical_nll": 1.7559456825256348, "generation_rouge_l_f1": 0.1423487544483986, "generation_token_f1": 0.22775800711743774, "prompt_id": "6c3b3c80c0d299e0", "reading_canonical_log_prob": -84.28539276123047, "reading_distances": [29, 23, 19, 32], "reading_log_probs": [-80.04049682617188, -82.11489868164062, -76.93449401855469, -80.8387222290039], "reading_logprob_std": 2.204082897402736, "tokenization_path_logprob_std": 46.02250938810417, "trajectory_balance_rmse": 1166.8012256451416} {"canonical_length": 131, "canonical_nll": 1.452438703930105, "generation_rouge_l_f1": 0.24810126582278483, "generation_token_f1": 0.41012658227848103, "prompt_id": "783057aa1290e5ce", "reading_canonical_log_prob": -190.26947021484375, "reading_distances": [35, 27, 24, 27], "reading_log_probs": [-196.41690063476562, -190.69082641601562, -190.4134521484375, -194.03536987304688], "reading_logprob_std": 2.8705720086155617, "tokenization_path_logprob_std": 198.21572170007127, "trajectory_balance_rmse": 2908.6275075133194} {"canonical_length": 358, "canonical_nll": 1.586273704827165, "generation_rouge_l_f1": 0.21666666666666667, "generation_token_f1": 0.44, "prompt_id": "609ea42a956a4ffa", "reading_canonical_log_prob": -567.885986328125, "reading_distances": [13, 18, 13, 14], "reading_log_probs": [-568.7904052734375, -571.513427734375, -571.2559814453125, -570.4148559570312], "reading_logprob_std": 1.2285963479475515, "tokenization_path_logprob_std": 273.9468856355131, "trajectory_balance_rmse": 6692.612757932044} {"canonical_length": 185, "canonical_nll": 1.0777847702438768, "generation_rouge_l_f1": 0.3006535947712418, "generation_token_f1": 0.5751633986928104, "prompt_id": "0fd242df3fc4b686", "reading_canonical_log_prob": -199.3901824951172, "reading_distances": [21, 23, 25, 23], "reading_log_probs": [-198.5188446044922, -201.73519897460938, -199.8358154296875, -197.55172729492188], "reading_logprob_std": 1.810635598160273, "tokenization_path_logprob_std": 149.88033144304737, "trajectory_balance_rmse": 3647.41907300689} {"canonical_length": 78, "canonical_nll": 0.4864812508607522, "generation_rouge_l_f1": 0.26315789473684215, "generation_token_f1": 0.375, "prompt_id": "c7ad7826ab1191ed", "reading_canonical_log_prob": -37.94553756713867, "reading_distances": [33, 30, 29, 24], "reading_log_probs": [-40.06452560424805, -40.96092987060547, -39.69902801513672, -42.737892150878906], "reading_logprob_std": 1.3561189802424278, "tokenization_path_logprob_std": 91.13915976486065, "trajectory_balance_rmse": 1502.226308437836} {"canonical_length": 9, "canonical_nll": 2.168461693657769, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.2222222222222222, "prompt_id": "f4e51cd8f0bc2528", "reading_canonical_log_prob": -19.516155242919922, "reading_distances": [30, 27, 27, 24], "reading_log_probs": [-23.767353057861328, -19.041141510009766, -19.579872131347656, -20.234111785888672], "reading_logprob_std": 2.1310659359032016, "tokenization_path_logprob_std": 39.213344462190165, "trajectory_balance_rmse": 177.48977507134347} {"canonical_length": 110, "canonical_nll": 1.2734373612837357, "generation_rouge_l_f1": 0.2185430463576159, "generation_token_f1": 0.33112582781456956, "prompt_id": "67c67cc47e6c2592", "reading_canonical_log_prob": -140.07810974121094, "reading_distances": [20, 31, 25, 28], "reading_log_probs": [-143.19949340820312, -142.85494995117188, -144.09619140625, -145.5966339111328], "reading_logprob_std": 1.2239927519134823, "tokenization_path_logprob_std": 58.43569906398673, "trajectory_balance_rmse": 1757.9514785311217} {"canonical_length": 18, "canonical_nll": 0.7549547089470757, "generation_rouge_l_f1": 0.6666666666666667, "generation_token_f1": 0.6666666666666667, "prompt_id": "82aeadf42ab79a24", "reading_canonical_log_prob": -13.589184761047363, "reading_distances": [41, 46, 37, 40], "reading_log_probs": [-15.54796314239502, -16.527193069458008, -16.315471649169922, -15.782506942749023], "reading_logprob_std": 0.45519282706380537, "tokenization_path_logprob_std": 29.150941458953803, "trajectory_balance_rmse": 396.1588710627447} {"canonical_length": 12, "canonical_nll": 0.3647817373275757, "generation_rouge_l_f1": 0.5555555555555556, "generation_token_f1": 0.888888888888889, "prompt_id": "736b52db67801c89", "reading_canonical_log_prob": -4.377380847930908, "reading_distances": [24, 21, 27, 28], "reading_log_probs": [-4.091658115386963, -4.402646064758301, -3.690957546234131, -4.848612308502197], "reading_logprob_std": 0.4895410340145511, "tokenization_path_logprob_std": 29.0358855846956, "trajectory_balance_rmse": 187.25748559113927} {"canonical_length": 29, "canonical_nll": 0.9361759054249731, "generation_rouge_l_f1": 0.30769230769230765, "generation_token_f1": 0.5384615384615384, "prompt_id": "3c9993234c0b5914", "reading_canonical_log_prob": -27.14910125732422, "reading_distances": [21, 21, 19, 29], "reading_log_probs": [-27.570756912231445, -29.162282943725586, -28.806903839111328, -30.401325225830078], "reading_logprob_std": 1.1646469763756726, "tokenization_path_logprob_std": 36.016574627178805, "trajectory_balance_rmse": 736.1195835187738} {"canonical_length": 221, "canonical_nll": 1.1667434899515696, "generation_rouge_l_f1": 0.2268041237113402, "generation_token_f1": 0.49484536082474223, "prompt_id": "1560e8f9b6710dda", "reading_canonical_log_prob": -257.8503112792969, "reading_distances": [14, 14, 16, 10], "reading_log_probs": [-260.02630615234375, -264.7900390625, -265.60479736328125, -260.9146423339844], "reading_logprob_std": 2.773114099292599, "tokenization_path_logprob_std": 177.6928003707871, "trajectory_balance_rmse": 4348.987054016894} {"canonical_length": 184, "canonical_nll": 1.8725857942000679, "generation_rouge_l_f1": 0.18018018018018017, "generation_token_f1": 0.31981981981981983, "prompt_id": "1073dfb1d10b700d", "reading_canonical_log_prob": -344.5557861328125, "reading_distances": [15, 18, 21, 20], "reading_log_probs": [-346.35614013671875, -352.675048828125, -352.51043701171875, -347.138916015625], "reading_logprob_std": 3.390499183155992, "tokenization_path_logprob_std": 94.19295619516804, "trajectory_balance_rmse": 4196.171738633124} {"canonical_length": 67, "canonical_nll": 1.7042877425008744, "generation_rouge_l_f1": 0.22727272727272727, "generation_token_f1": 0.375, "prompt_id": "2045b4674c0f9cd0", "reading_canonical_log_prob": -114.1872787475586, "reading_distances": [29, 22, 24, 21], "reading_log_probs": [-110.11964416503906, -111.78443908691406, -109.36824798583984, -111.35723114013672], "reading_logprob_std": 1.112216253167673, "tokenization_path_logprob_std": 184.6603274550881, "trajectory_balance_rmse": 1550.4904046100403} {"canonical_length": 18, "canonical_nll": 0.5288469526502821, "generation_rouge_l_f1": 0.5833333333333334, "generation_token_f1": 0.5833333333333334, "prompt_id": "e0eb2633a5eb27ba", "reading_canonical_log_prob": -9.519245147705078, "reading_distances": [17, 23, 24, 27], "reading_log_probs": [-10.615922927856445, -10.168660163879395, -9.037089347839355, -10.73819637298584], "reading_logprob_std": 0.7749354470434598, "tokenization_path_logprob_std": 58.96364004452574, "trajectory_balance_rmse": 485.99096906015967} {"canonical_length": 10, "canonical_nll": 1.694986915588379, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.375, "prompt_id": "d46e9e081fc93a1e", "reading_canonical_log_prob": -16.94986915588379, "reading_distances": [15, 16, 10, 18], "reading_log_probs": [-16.665103912353516, -16.581899642944336, -16.247678756713867, -15.39695930480957], "reading_logprob_std": 0.5794299100719206, "tokenization_path_logprob_std": 32.95760946995046, "trajectory_balance_rmse": 294.75438399648016} {"canonical_length": 104, "canonical_nll": 1.6586221548227162, "generation_rouge_l_f1": 0.16931216931216933, "generation_token_f1": 0.30158730158730157, "prompt_id": "d0a49199c6983c98", "reading_canonical_log_prob": -172.4967041015625, "reading_distances": [26, 23, 34, 26], "reading_log_probs": [-174.6502685546875, -174.02674865722656, -178.82086181640625, -172.02027893066406], "reading_logprob_std": 2.857108677961219, "tokenization_path_logprob_std": 183.38995719723457, "trajectory_balance_rmse": 1629.9737678234355} {"canonical_length": 2, "canonical_nll": 1.9681872129440308, "generation_rouge_l_f1": 0.04878048780487806, "generation_token_f1": 0.04878048780487806, "prompt_id": "22a76db8343e67a9", "reading_canonical_log_prob": -3.9363744258880615, "reading_distances": [28, 20, 26, 22], "reading_log_probs": [-3.2143137454986572, -5.041039943695068, -2.925973415374756, -2.705958127975464], "reading_logprob_std": 1.0666542088478954, "tokenization_path_logprob_std": 16.553819223821463, "trajectory_balance_rmse": 35.67431142283516} {"canonical_length": 193, "canonical_nll": 1.2308126657120304, "generation_rouge_l_f1": 0.24728850325379612, "generation_token_f1": 0.5162689804772235, "prompt_id": "ab11d19cfada82a0", "reading_canonical_log_prob": -237.54684448242188, "reading_distances": [23, 24, 17, 23], "reading_log_probs": [-243.2758026123047, -248.70309448242188, -244.00112915039062, -244.70738220214844], "reading_logprob_std": 2.42562713086427, "tokenization_path_logprob_std": 119.7050257789445, "trajectory_balance_rmse": 4448.437495225161} {"canonical_length": 246, "canonical_nll": 1.3416913040285188, "generation_rouge_l_f1": 0.22813688212927757, "generation_token_f1": 0.4144486692015209, "prompt_id": "65ef693d02b228c6", "reading_canonical_log_prob": -330.0560607910156, "reading_distances": [76, 81, 86, 85], "reading_log_probs": [-342.4330139160156, -334.0799560546875, -336.5326232910156, -333.9574890136719], "reading_logprob_std": 3.969515071424298, "tokenization_path_logprob_std": 218.00596215408908, "trajectory_balance_rmse": 5410.25004684534} {"canonical_length": 24, "canonical_nll": 0.6206630865732828, "generation_rouge_l_f1": 0.5714285714285714, "generation_token_f1": 0.6428571428571429, "prompt_id": "4fb4dab7cec52a74", "reading_canonical_log_prob": -14.895914077758789, "reading_distances": [19, 27, 21, 27], "reading_log_probs": [-21.418684005737305, -22.28954315185547, -20.100988388061523, -22.128385543823242], "reading_logprob_std": 0.9968482257563799, "tokenization_path_logprob_std": 76.23822642927581, "trajectory_balance_rmse": 395.89160434992874} {"canonical_length": 36, "canonical_nll": 1.0783372455173068, "generation_rouge_l_f1": 0.16107382550335572, "generation_token_f1": 0.19463087248322147, "prompt_id": "d445e9b3b0bb59b8", "reading_canonical_log_prob": -38.82014083862305, "reading_distances": [49, 43, 38, 42], "reading_log_probs": [-49.46466064453125, -41.691368103027344, -42.574737548828125, -47.820152282714844], "reading_logprob_std": 3.8346690185510766, "tokenization_path_logprob_std": 109.13045842633849, "trajectory_balance_rmse": 787.1863276523893} {"canonical_length": 17, "canonical_nll": 0.7001398871926701, "generation_rouge_l_f1": 0.7333333333333333, "generation_token_f1": 0.7333333333333333, "prompt_id": "9384aa5dbac32997", "reading_canonical_log_prob": -11.90237808227539, "reading_distances": [27, 24, 26, 25], "reading_log_probs": [-8.671656608581543, -10.849802017211914, -7.305756568908691, -9.674481391906738], "reading_logprob_std": 1.504663635849132, "tokenization_path_logprob_std": 66.14553097811236, "trajectory_balance_rmse": 288.66533211323446} {"canonical_length": 227, "canonical_nll": 1.4861989294379818, "generation_rouge_l_f1": 0.1927710843373494, "generation_token_f1": 0.37751004016064255, "prompt_id": "7e391a560a4818b8", "reading_canonical_log_prob": -337.3671569824219, "reading_distances": [12, 17, 13, 11], "reading_log_probs": [-338.38079833984375, -335.64404296875, -337.0443420410156, -338.60076904296875], "reading_logprob_std": 1.3677799558738408, "tokenization_path_logprob_std": 109.12908284770566, "trajectory_balance_rmse": 4209.441270317173} {"canonical_length": 30, "canonical_nll": 4.334758504231771, "generation_rouge_l_f1": 0.125, "generation_token_f1": 0.125, "prompt_id": "62ca6c153160a733", "reading_canonical_log_prob": -130.04275512695312, "reading_distances": [47, 42, 48, 45], "reading_log_probs": [-152.5946044921875, -140.44375610351562, -159.19479370117188, -159.26219177246094], "reading_logprob_std": 8.857208969787985, "tokenization_path_logprob_std": 0.0, "trajectory_balance_rmse": 129.94097900390625} {"canonical_length": 85, "canonical_nll": 0.3728288089527803, "generation_rouge_l_f1": 0.2325581395348837, "generation_token_f1": 0.37790697674418605, "prompt_id": "5fd334d7d6294d39", "reading_canonical_log_prob": -31.690448760986328, "reading_distances": [25, 29, 20, 25], "reading_log_probs": [-35.78541564941406, -34.95534133911133, -36.2408447265625, -37.475608825683594], "reading_logprob_std": 1.0520654891507852, "tokenization_path_logprob_std": 141.72163130702714, "trajectory_balance_rmse": 1539.3394518354212} {"canonical_length": 161, "canonical_nll": 1.8646244025378493, "generation_rouge_l_f1": 0.2458628841607565, "generation_token_f1": 0.5011820330969267, "prompt_id": "ec23e8fead4fc5cf", "reading_canonical_log_prob": -300.20452880859375, "reading_distances": [25, 31, 28, 26], "reading_log_probs": [-309.5953674316406, -306.44891357421875, -307.1407165527344, -308.2904052734375], "reading_logprob_std": 1.379005566898028, "tokenization_path_logprob_std": 158.79352257064508, "trajectory_balance_rmse": 2867.6350694368443} {"canonical_length": 296, "canonical_nll": 1.1602028511665963, "generation_rouge_l_f1": 0.33756805807622503, "generation_token_f1": 0.5372050816696915, "prompt_id": "6d37aa3b89a989f6", "reading_canonical_log_prob": -343.4200439453125, "reading_distances": [16, 20, 24, 16], "reading_log_probs": [-360.09088134765625, -353.6018981933594, -360.22540283203125, -354.6764831542969], "reading_logprob_std": 3.5030553607944523, "tokenization_path_logprob_std": 130.13212504685012, "trajectory_balance_rmse": 6232.744149635676} {"canonical_length": 19, "canonical_nll": 3.339654621325041, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.25, "prompt_id": "b2ecf966e39675a0", "reading_canonical_log_prob": -63.45343780517578, "reading_distances": [23, 29, 29, 24], "reading_log_probs": [-62.59711456298828, -62.89313507080078, -70.85516357421875, -65.31388854980469], "reading_logprob_std": 3.82561037286772, "tokenization_path_logprob_std": 51.79778645247395, "trajectory_balance_rmse": 405.2401032837804} {"canonical_length": 17, "canonical_nll": 1.1076927185058594, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "efd422672564486e", "reading_canonical_log_prob": -18.83077621459961, "reading_distances": [14, 19, 18, 17], "reading_log_probs": [-20.76076889038086, -22.186176300048828, -18.335329055786133, -22.86850357055664], "reading_logprob_std": 2.0041905336253176, "tokenization_path_logprob_std": 34.66695649842751, "trajectory_balance_rmse": 232.08633606034581} {"canonical_length": 196, "canonical_nll": 1.6116902876873405, "generation_rouge_l_f1": 0.22272727272727275, "generation_token_f1": 0.4454545454545455, "prompt_id": "743badce8c66694d", "reading_canonical_log_prob": -315.89129638671875, "reading_distances": [18, 18, 22, 15], "reading_log_probs": [-320.56939697265625, -331.330322265625, -321.9815673828125, -322.1912841796875], "reading_logprob_std": 4.927706301029968, "tokenization_path_logprob_std": 339.40107684154157, "trajectory_balance_rmse": 4491.20193026216} {"canonical_length": 14, "canonical_nll": 0.3152073792048863, "generation_rouge_l_f1": 0.923076923076923, "generation_token_f1": 0.923076923076923, "prompt_id": "685c7dde1bc34a6f", "reading_canonical_log_prob": -4.412903308868408, "reading_distances": [33, 28, 26, 35], "reading_log_probs": [-5.9277520179748535, -4.359610080718994, -3.8617305755615234, -3.78761625289917], "reading_logprob_std": 0.9953328045764316, "tokenization_path_logprob_std": 22.380742292707183, "trajectory_balance_rmse": 228.93087817226638} {"canonical_length": 214, "canonical_nll": 1.450612576208382, "generation_rouge_l_f1": 0.24827586206896549, "generation_token_f1": 0.45977011494252873, "prompt_id": "d0bf7e97f3a575eb", "reading_canonical_log_prob": -310.43109130859375, "reading_distances": [28, 25, 35, 29], "reading_log_probs": [-323.1570739746094, -322.755859375, -321.1546936035156, -326.1384582519531], "reading_logprob_std": 2.0797189228320176, "tokenization_path_logprob_std": 352.19714245561573, "trajectory_balance_rmse": 4569.750837802165} {"canonical_length": 120, "canonical_nll": 0.724737803141276, "generation_rouge_l_f1": 0.18691588785046728, "generation_token_f1": 0.34890965732087226, "prompt_id": "181490950025f621", "reading_canonical_log_prob": -86.96853637695312, "reading_distances": [31, 28, 22, 22], "reading_log_probs": [-90.12725830078125, -83.37738037109375, -84.72586059570312, -84.87542724609375], "reading_logprob_std": 2.977731130125958, "tokenization_path_logprob_std": 178.2991711102583, "trajectory_balance_rmse": 2153.0726373447674} {"canonical_length": 38, "canonical_nll": 1.6903847142269737, "generation_rouge_l_f1": 0.33043478260869563, "generation_token_f1": 0.4869565217391304, "prompt_id": "70f652d8f02698c0", "reading_canonical_log_prob": -64.234619140625, "reading_distances": [28, 29, 27, 30], "reading_log_probs": [-70.73436737060547, -68.495849609375, -69.65128326416016, -69.1072006225586], "reading_logprob_std": 0.9502853732673797, "tokenization_path_logprob_std": 33.246619915218965, "trajectory_balance_rmse": 939.1639289446031} {"canonical_length": 278, "canonical_nll": 1.1910231336415242, "generation_rouge_l_f1": 0.21771217712177118, "generation_token_f1": 0.43542435424354237, "prompt_id": "fe3deab131dcdb79", "reading_canonical_log_prob": -331.10443115234375, "reading_distances": [25, 32, 23, 30], "reading_log_probs": [-349.16656494140625, -354.71661376953125, -347.7957763671875, -362.6981506347656], "reading_logprob_std": 6.766766796744333, "tokenization_path_logprob_std": 184.60284107132537, "trajectory_balance_rmse": 6486.14043534041} {"canonical_length": 16, "canonical_nll": 1.309575080871582, "generation_rouge_l_f1": 0.1818181818181818, "generation_token_f1": 0.1818181818181818, "prompt_id": "16c98792a8b466f0", "reading_canonical_log_prob": -20.953201293945312, "reading_distances": [33, 33, 34, 31], "reading_log_probs": [-33.621978759765625, -31.490821838378906, -38.72056579589844, -32.410011291503906], "reading_logprob_std": 3.2267564961800432, "tokenization_path_logprob_std": 32.14089084783827, "trajectory_balance_rmse": 291.1683198474904} {"canonical_length": 53, "canonical_nll": 1.2751958595131927, "generation_rouge_l_f1": 0.2391304347826087, "generation_token_f1": 0.34782608695652173, "prompt_id": "52f4380d6ba7596b", "reading_canonical_log_prob": -67.58538055419922, "reading_distances": [13, 13, 14, 16], "reading_log_probs": [-74.88432312011719, -75.89701843261719, -75.97760009765625, -75.38577270507812], "reading_logprob_std": 0.5074760497720683, "tokenization_path_logprob_std": 129.31946270766474, "trajectory_balance_rmse": 949.2716078198448} {"canonical_length": 78, "canonical_nll": 0.7753962003267728, "generation_rouge_l_f1": 0.13333333333333336, "generation_token_f1": 0.26, "prompt_id": "9b002f163aa12acd", "reading_canonical_log_prob": -60.48090362548828, "reading_distances": [18, 19, 21, 20], "reading_log_probs": [-63.373634338378906, -63.48828887939453, -66.15548706054688, -64.4481201171875], "reading_logprob_std": 1.2863609505867266, "tokenization_path_logprob_std": 90.51660755635284, "trajectory_balance_rmse": 1293.4247393151077} {"canonical_length": 22, "canonical_nll": 2.6570250771262427, "generation_rouge_l_f1": 0.10256410256410256, "generation_token_f1": 0.15384615384615385, "prompt_id": "6e455bb516dd4dc1", "reading_canonical_log_prob": -58.454551696777344, "reading_distances": [12, 14, 16, 15], "reading_log_probs": [-60.208595275878906, -64.59745788574219, -64.7347183227539, -61.5252571105957], "reading_logprob_std": 2.2590448301113915, "tokenization_path_logprob_std": 50.018891309706824, "trajectory_balance_rmse": 482.4444245977225} {"canonical_length": 130, "canonical_nll": 1.5664682241586538, "generation_rouge_l_f1": 0.193029490616622, "generation_token_f1": 0.386058981233244, "prompt_id": "7c910d294599c2d9", "reading_canonical_log_prob": -203.640869140625, "reading_distances": [14, 14, 13, 6], "reading_log_probs": [-203.25927734375, -205.0265350341797, -204.52468872070312, -203.58607482910156], "reading_logprob_std": 0.8184860849972009, "tokenization_path_logprob_std": 198.3807649908871, "trajectory_balance_rmse": 2608.648599043371} {"canonical_length": 22, "canonical_nll": 1.3620908910577947, "generation_rouge_l_f1": 0.09090909090909091, "generation_token_f1": 0.11363636363636362, "prompt_id": "e55841df337a493f", "reading_canonical_log_prob": -29.965999603271484, "reading_distances": [19, 21, 21, 24], "reading_log_probs": [-32.794830322265625, -31.41887092590332, -34.28435516357422, -32.91501235961914], "reading_logprob_std": 1.1708590527251315, "tokenization_path_logprob_std": 26.368343313009667, "trajectory_balance_rmse": 453.1568255203237} {"canonical_length": 440, "canonical_nll": 2.0820451216264204, "generation_rouge_l_f1": 0.2518301610541728, "generation_token_f1": 0.4626647144948755, "prompt_id": "58ceb01ab8ce3495", "reading_canonical_log_prob": -916.099853515625, "reading_distances": [80, 81, 74, 88], "reading_log_probs": [-929.8277587890625, -936.2935791015625, -939.180908203125, -948.95556640625], "reading_logprob_std": 7.954913551531989, "tokenization_path_logprob_std": 142.63438283139394, "trajectory_balance_rmse": 7654.523471356076} {"canonical_length": 32, "canonical_nll": 1.1785571575164795, "generation_rouge_l_f1": 0.1923076923076923, "generation_token_f1": 0.26923076923076916, "prompt_id": "a8059315f9c843b5", "reading_canonical_log_prob": -37.713829040527344, "reading_distances": [23, 31, 21, 20], "reading_log_probs": [-41.63325500488281, -41.677738189697266, -40.778564453125, -44.65098571777344], "reading_logprob_std": 1.6951778193642024, "tokenization_path_logprob_std": 82.31617503272585, "trajectory_balance_rmse": 766.4145114953083} {"canonical_length": 12, "canonical_nll": 1.1392656167348225, "generation_rouge_l_f1": 0.125, "generation_token_f1": 0.14583333333333331, "prompt_id": "3715d1d4c18fe319", "reading_canonical_log_prob": -13.671187400817871, "reading_distances": [13, 9, 11, 19], "reading_log_probs": [-18.32990264892578, -14.595026016235352, -15.376398086547852, -15.31956958770752], "reading_logprob_std": 1.6551267065891182, "tokenization_path_logprob_std": 62.140775853227844, "trajectory_balance_rmse": 231.5718731706457} {"canonical_length": 264, "canonical_nll": 1.9172686952533144, "generation_rouge_l_f1": 0.19193857965451055, "generation_token_f1": 0.41842610364683297, "prompt_id": "996c63537529bc50", "reading_canonical_log_prob": -506.158935546875, "reading_distances": [31, 19, 25, 22], "reading_log_probs": [-502.8643798828125, -522.3219604492188, -520.1690673828125, -507.4490966796875], "reading_logprob_std": 9.516231567822983, "tokenization_path_logprob_std": 191.94078516159496, "trajectory_balance_rmse": 6072.615506220098} {"canonical_length": 15, "canonical_nll": 0.5795244216918946, "generation_rouge_l_f1": 0.8148148148148148, "generation_token_f1": 0.888888888888889, "prompt_id": "a9d8f6ac4429b707", "reading_canonical_log_prob": -8.692866325378418, "reading_distances": [23, 23, 20, 23], "reading_log_probs": [-9.916847229003906, -9.74316692352295, -9.664498329162598, -9.337455749511719], "reading_logprob_std": 0.24277672452140897, "tokenization_path_logprob_std": 64.69785050975689, "trajectory_balance_rmse": 210.9230012354708} {"canonical_length": 17, "canonical_nll": 1.0418153650620405, "generation_rouge_l_f1": 0.5925925925925927, "generation_token_f1": 0.6666666666666667, "prompt_id": "b1822d4ae9e3de29", "reading_canonical_log_prob": -17.710861206054688, "reading_distances": [21, 22, 25, 16], "reading_log_probs": [-15.156855583190918, -14.75013542175293, -15.745104789733887, -12.829526901245117], "reading_logprob_std": 1.2618509146170431, "tokenization_path_logprob_std": 35.63123485718128, "trajectory_balance_rmse": 402.1184671053903} {"canonical_length": 14, "canonical_nll": 0.8168620382036481, "generation_rouge_l_f1": 0.3188405797101449, "generation_token_f1": 0.34782608695652173, "prompt_id": "92e5aa90e47873e8", "reading_canonical_log_prob": -11.436068534851074, "reading_distances": [27, 17, 21, 21], "reading_log_probs": [-13.361311912536621, -11.269205093383789, -12.912697792053223, -12.54876708984375], "reading_logprob_std": 0.8994992759911336, "tokenization_path_logprob_std": 49.59064469821852, "trajectory_balance_rmse": 299.3945365132964} {"canonical_length": 176, "canonical_nll": 1.6993390863591975, "generation_rouge_l_f1": 0.19157088122605365, "generation_token_f1": 0.2988505747126437, "prompt_id": "d3105cc3dd2ecc61", "reading_canonical_log_prob": -299.08367919921875, "reading_distances": [31, 31, 33, 35], "reading_log_probs": [-308.9253234863281, -308.3286437988281, -308.1663818359375, -308.0736999511719], "reading_logprob_std": 0.38266575554777266, "tokenization_path_logprob_std": 61.88099640519229, "trajectory_balance_rmse": 3816.180878931038} {"canonical_length": 46, "canonical_nll": 0.7492967688519022, "generation_rouge_l_f1": 0.18749999999999997, "generation_token_f1": 0.19642857142857142, "prompt_id": "f2effbcb126018c5", "reading_canonical_log_prob": -34.4676513671875, "reading_distances": [23, 26, 29, 23], "reading_log_probs": [-44.99707794189453, -39.63800811767578, -48.181243896484375, -40.147743225097656], "reading_logprob_std": 4.084099552859204, "tokenization_path_logprob_std": 31.200155928050187, "trajectory_balance_rmse": 444.20878627767723} {"canonical_length": 226, "canonical_nll": 1.4218474531595686, "generation_rouge_l_f1": 0.23293172690763053, "generation_token_f1": 0.5140562248995983, "prompt_id": "e73472036e75a1c6", "reading_canonical_log_prob": -321.3375244140625, "reading_distances": [41, 37, 54, 44], "reading_log_probs": [-322.9071044921875, -322.80560302734375, -321.9835510253906, -326.48468017578125], "reading_logprob_std": 2.0027864570768568, "tokenization_path_logprob_std": 207.50606971254842, "trajectory_balance_rmse": 4107.035949288971} {"canonical_length": 8, "canonical_nll": 1.3588497638702393, "generation_rouge_l_f1": 0.2745098039215686, "generation_token_f1": 0.2745098039215686, "prompt_id": "dea67577bfdb2da1", "reading_canonical_log_prob": -10.870798110961914, "reading_distances": [22, 26, 30, 27], "reading_log_probs": [-10.630261421203613, -10.499467849731445, -11.232511520385742, -10.402853965759277], "reading_logprob_std": 0.3726644172606094, "tokenization_path_logprob_std": 14.782317004784273, "trajectory_balance_rmse": 132.65699605974694} {"canonical_length": 107, "canonical_nll": 1.2507016190858644, "generation_rouge_l_f1": 0.26530612244897955, "generation_token_f1": 0.36734693877551017, "prompt_id": "90fd3570044a7e38", "reading_canonical_log_prob": -133.8250732421875, "reading_distances": [24, 25, 24, 18], "reading_log_probs": [-150.98410034179688, -151.07907104492188, -172.0290985107422, -150.24600219726562], "reading_logprob_std": 10.636206622358685, "tokenization_path_logprob_std": 81.37685667479577, "trajectory_balance_rmse": 2047.5062234012835} {"canonical_length": 222, "canonical_nll": 0.673827025267455, "generation_rouge_l_f1": 0.20454545454545453, "generation_token_f1": 0.40909090909090906, "prompt_id": "1c7402f9ee727b52", "reading_canonical_log_prob": -149.589599609375, "reading_distances": [78, 90, 86, 86], "reading_log_probs": [-143.25038146972656, -145.12008666992188, -145.46676635742188, -147.3059539794922], "reading_logprob_std": 1.6617421003763417, "tokenization_path_logprob_std": 75.56357188512342, "trajectory_balance_rmse": 2384.108214817852} {"canonical_length": 19, "canonical_nll": 1.4695774881463302, "generation_rouge_l_f1": 0.2580645161290323, "generation_token_f1": 0.38709677419354843, "prompt_id": "219202257fbe0a6c", "reading_canonical_log_prob": -27.921972274780273, "reading_distances": [21, 22, 20, 19], "reading_log_probs": [-28.36272621154785, -28.042469024658203, -28.994699478149414, -28.088350296020508], "reading_logprob_std": 0.4385158478738848, "tokenization_path_logprob_std": 48.20175147085446, "trajectory_balance_rmse": 409.45110921485633} {"canonical_length": 47, "canonical_nll": 1.732538750831117, "generation_rouge_l_f1": 0.3170731707317073, "generation_token_f1": 0.46341463414634143, "prompt_id": "1eb21676f35636bf", "reading_canonical_log_prob": -81.4293212890625, "reading_distances": [20, 29, 28, 23], "reading_log_probs": [-78.60418701171875, -97.04328155517578, -84.40904235839844, -91.04656982421875], "reading_logprob_std": 8.000785199780497, "tokenization_path_logprob_std": 49.86429246877467, "trajectory_balance_rmse": 970.207846918641} {"canonical_length": 15, "canonical_nll": 0.1363634745279948, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "5f8f43706c38c575", "reading_canonical_log_prob": -2.045452117919922, "reading_distances": [29, 31, 29, 33], "reading_log_probs": [-4.271304130554199, -6.440186023712158, -3.5088744163513184, -3.7655465602874756], "reading_logprob_std": 1.3339572408089613, "tokenization_path_logprob_std": 44.887860551576814, "trajectory_balance_rmse": 277.40373374900105} {"canonical_length": 18, "canonical_nll": 0.664484871758355, "generation_rouge_l_f1": 0.8387096774193549, "generation_token_f1": 0.8387096774193549, "prompt_id": "46d2c294deb9298b", "reading_canonical_log_prob": -11.96072769165039, "reading_distances": [73, 64, 59, 69], "reading_log_probs": [-13.466334342956543, -8.39310073852539, -11.4552640914917, -12.24624252319336], "reading_logprob_std": 2.1625586639929324, "tokenization_path_logprob_std": 41.14586052656816, "trajectory_balance_rmse": 427.13908658388544} {"canonical_length": 228, "canonical_nll": 1.243626644736842, "generation_rouge_l_f1": 0.21339950372208435, "generation_token_f1": 0.401985111662531, "prompt_id": "7cb804cc64427944", "reading_canonical_log_prob": -283.546875, "reading_distances": [24, 19, 19, 19], "reading_log_probs": [-280.81072998046875, -281.0289306640625, -278.18359375, -279.997314453125], "reading_logprob_std": 1.2929571429487472, "tokenization_path_logprob_std": 121.44418475982351, "trajectory_balance_rmse": 4679.860782454192} {"canonical_length": 171, "canonical_nll": 1.4459895083778782, "generation_rouge_l_f1": 0.3004926108374384, "generation_token_f1": 0.5024630541871921, "prompt_id": "d5a4beae5409050e", "reading_canonical_log_prob": -247.2642059326172, "reading_distances": [29, 18, 28, 22], "reading_log_probs": [-247.72958374023438, -252.6669464111328, -253.35354614257812, -257.77001953125], "reading_logprob_std": 4.111314325267481, "tokenization_path_logprob_std": 346.01361916690547, "trajectory_balance_rmse": 4155.044288753871} {"canonical_length": 168, "canonical_nll": 1.7465021042596727, "generation_rouge_l_f1": 0.18426966292134828, "generation_token_f1": 0.3280898876404495, "prompt_id": "ad2f9057c92483ba", "reading_canonical_log_prob": -293.412353515625, "reading_distances": [13, 14, 14, 11], "reading_log_probs": [-295.0787048339844, -303.84796142578125, -293.750244140625, -291.56512451171875], "reading_logprob_std": 5.389933831447726, "tokenization_path_logprob_std": 205.67665896635566, "trajectory_balance_rmse": 3420.93417371813} {"canonical_length": 159, "canonical_nll": 1.2722325355001964, "generation_rouge_l_f1": 0.34096692111959287, "generation_token_f1": 0.5343511450381679, "prompt_id": "240077b10fcaa781", "reading_canonical_log_prob": -202.28497314453125, "reading_distances": [27, 25, 26, 24], "reading_log_probs": [-205.7554168701172, -205.075439453125, -204.42691040039062, -206.14352416992188], "reading_logprob_std": 0.7575224821815676, "tokenization_path_logprob_std": 191.2293058367104, "trajectory_balance_rmse": 3499.1752660722236} {"canonical_length": 327, "canonical_nll": 0.8742391137172687, "generation_rouge_l_f1": 0.24817518248175185, "generation_token_f1": 0.4562043795620438, "prompt_id": "dcea5e8dbdd12a1a", "reading_canonical_log_prob": -285.8761901855469, "reading_distances": [68, 68, 70, 69], "reading_log_probs": [-301.268310546875, -302.1013488769531, -292.8637390136719, -302.20928955078125], "reading_logprob_std": 4.517564046943357, "tokenization_path_logprob_std": 81.27158959497363, "trajectory_balance_rmse": 5268.44508774434} {"canonical_length": 4, "canonical_nll": 2.387115478515625, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "09f4cd01fd363213", "reading_canonical_log_prob": -9.5484619140625, "reading_distances": [30, 33, 33, 28], "reading_log_probs": [-17.501720428466797, -19.966259002685547, -17.017881393432617, -14.636714935302734], "reading_logprob_std": 2.1848575709821723, "tokenization_path_logprob_std": 16.325661022023656, "trajectory_balance_rmse": 98.74459919139309} {"canonical_length": 139, "canonical_nll": 1.0730430548139613, "generation_rouge_l_f1": 0.18674698795180725, "generation_token_f1": 0.3855421686746988, "prompt_id": "0eb3ed1212b155f2", "reading_canonical_log_prob": -149.15298461914062, "reading_distances": [26, 19, 22, 22], "reading_log_probs": [-156.83636474609375, -156.5906524658203, -160.2344970703125, -160.9141845703125], "reading_logprob_std": 2.248495808206875, "tokenization_path_logprob_std": 181.50393274742794, "trajectory_balance_rmse": 2632.2487883320164} {"canonical_length": 3, "canonical_nll": 2.8462441762288413, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "423d33be2ee3658e", "reading_canonical_log_prob": -8.538732528686523, "reading_distances": [22, 21, 21, 23], "reading_log_probs": [-9.972702026367188, -9.74113655090332, -9.510371208190918, -10.573066711425781], "reading_logprob_std": 0.45666292071829356, "tokenization_path_logprob_std": 12.005987171527147, "trajectory_balance_rmse": 33.01678400039036} {"canonical_length": 33, "canonical_nll": 1.2715643680456914, "generation_rouge_l_f1": 0.3225806451612903, "generation_token_f1": 0.3870967741935484, "prompt_id": "82dd428e952ac035", "reading_canonical_log_prob": -41.96162414550781, "reading_distances": [36, 31, 30, 32], "reading_log_probs": [-41.248138427734375, -42.60540008544922, -43.080196380615234, -42.84624481201172], "reading_logprob_std": 0.8211125890295252, "tokenization_path_logprob_std": 73.02859519231362, "trajectory_balance_rmse": 731.1498134639085} {"canonical_length": 293, "canonical_nll": 1.23785400390625, "generation_rouge_l_f1": 0.20186915887850468, "generation_token_f1": 0.42242990654205603, "prompt_id": "a3746b2776781f66", "reading_canonical_log_prob": -362.69122314453125, "reading_distances": [21, 25, 23, 27], "reading_log_probs": [-373.1911926269531, -377.12823486328125, -375.88079833984375, -378.3154602050781], "reading_logprob_std": 2.1963146967117737, "tokenization_path_logprob_std": 231.5618795486543, "trajectory_balance_rmse": 6276.509825888971} {"canonical_length": 64, "canonical_nll": 1.9353671073913574, "generation_rouge_l_f1": 0.10638297872340427, "generation_token_f1": 0.1843971631205674, "prompt_id": "d4e9c302f1022491", "reading_canonical_log_prob": -123.86349487304688, "reading_distances": [30, 25, 32, 28], "reading_log_probs": [-126.26878356933594, -128.79039001464844, -130.60855102539062, -133.87765502929688], "reading_logprob_std": 3.2010417745697244, "tokenization_path_logprob_std": 124.53153203035261, "trajectory_balance_rmse": 1289.5486185006885} {"canonical_length": 99, "canonical_nll": 1.1993950737847223, "generation_rouge_l_f1": 0.19220779220779224, "generation_token_f1": 0.28571428571428575, "prompt_id": "ed00a510b9a4ea44", "reading_canonical_log_prob": -118.7401123046875, "reading_distances": [30, 33, 35, 28], "reading_log_probs": [-131.45791625976562, -139.9716339111328, -138.4097137451172, -140.21839904785156], "reading_logprob_std": 4.1163167735947, "tokenization_path_logprob_std": 248.4940144064881, "trajectory_balance_rmse": 2330.0547169286947} {"canonical_length": 275, "canonical_nll": 1.3525319602272727, "generation_rouge_l_f1": 0.21454545454545457, "generation_token_f1": 0.4436363636363637, "prompt_id": "00a5d180fc30eed4", "reading_canonical_log_prob": -371.9462890625, "reading_distances": [30, 25, 32, 26], "reading_log_probs": [-383.15753173828125, -380.884765625, -382.98724365234375, -381.8463439941406], "reading_logprob_std": 1.0630293184976451, "tokenization_path_logprob_std": 161.27290854280827, "trajectory_balance_rmse": 5656.521114507434} {"canonical_length": 115, "canonical_nll": 1.1670238992442254, "generation_rouge_l_f1": 0.2872340425531915, "generation_token_f1": 0.45212765957446804, "prompt_id": "300b5dc6095f893e", "reading_canonical_log_prob": -134.20774841308594, "reading_distances": [6, 10, 5, 6], "reading_log_probs": [-135.8631591796875, -133.16818237304688, -133.30133056640625, -133.8509063720703], "reading_logprob_std": 1.2470265053427898, "tokenization_path_logprob_std": 110.7508622763413, "trajectory_balance_rmse": 2224.908322422784} {"canonical_length": 6, "canonical_nll": 2.7684399286905923, "generation_rouge_l_f1": 0.13333333333333333, "generation_token_f1": 0.13333333333333333, "prompt_id": "2c288dbc9a356333", "reading_canonical_log_prob": -16.610639572143555, "reading_distances": [19, 21, 19, 22], "reading_log_probs": [-19.858686447143555, -21.205076217651367, -20.221590042114258, -23.327486038208008], "reading_logprob_std": 1.5571205658404321, "tokenization_path_logprob_std": 24.880769942226998, "trajectory_balance_rmse": 136.37646731151844} {"canonical_length": 233, "canonical_nll": 1.0097039349600994, "generation_rouge_l_f1": 0.3574660633484163, "generation_token_f1": 0.5701357466063348, "prompt_id": "810a6641955f3c0f", "reading_canonical_log_prob": -235.26101684570312, "reading_distances": [42, 39, 33, 35], "reading_log_probs": [-241.60861206054688, -238.23268127441406, -238.49424743652344, -233.88865661621094], "reading_logprob_std": 3.1733830743525147, "tokenization_path_logprob_std": 85.23517542275863, "trajectory_balance_rmse": 2769.3003302062916} {"canonical_length": 167, "canonical_nll": 2.0111913623924025, "generation_rouge_l_f1": 0.2882882882882883, "generation_token_f1": 0.5105105105105104, "prompt_id": "eb867c21f2710dc4", "reading_canonical_log_prob": -335.86895751953125, "reading_distances": [23, 31, 25, 26], "reading_log_probs": [-342.8240051269531, -343.2503662109375, -339.6632385253906, -340.7568054199219], "reading_logprob_std": 1.7011457637908205, "tokenization_path_logprob_std": 192.66809247661072, "trajectory_balance_rmse": 3931.8527802864855} {"canonical_length": 48, "canonical_nll": 2.0486672719319663, "generation_rouge_l_f1": 0.18556701030927836, "generation_token_f1": 0.2783505154639175, "prompt_id": "ee8008c4016a6c6e", "reading_canonical_log_prob": -98.33602905273438, "reading_distances": [24, 23, 32, 28], "reading_log_probs": [-100.70294189453125, -102.22210693359375, -102.49652862548828, -104.68665313720703], "reading_logprob_std": 1.6416635788622695, "tokenization_path_logprob_std": 48.9744440589389, "trajectory_balance_rmse": 1135.72063942121} {"canonical_length": 232, "canonical_nll": 1.2516818868702855, "generation_rouge_l_f1": 0.24267782426778242, "generation_token_f1": 0.5313807531380753, "prompt_id": "5adec8000ecf22fe", "reading_canonical_log_prob": -290.39019775390625, "reading_distances": [22, 18, 20, 23], "reading_log_probs": [-291.1209716796875, -293.10638427734375, -290.0760498046875, -291.28973388671875], "reading_logprob_std": 1.258914224613501, "tokenization_path_logprob_std": 139.23205408212402, "trajectory_balance_rmse": 4900.656180958802} {"canonical_length": 283, "canonical_nll": 1.4482439128754416, "generation_rouge_l_f1": 0.2803030303030304, "generation_token_f1": 0.4810606060606061, "prompt_id": "893df9640ff6dc30", "reading_canonical_log_prob": -409.85302734375, "reading_distances": [22, 23, 24, 19], "reading_log_probs": [-408.09112548828125, -409.75360107421875, -416.1960144042969, -421.1897277832031], "reading_logprob_std": 6.036370932560288, "tokenization_path_logprob_std": 90.57169009802719, "trajectory_balance_rmse": 5856.820651378994} {"canonical_length": 13, "canonical_nll": 0.8298521041870117, "generation_rouge_l_f1": 0.12345679012345681, "generation_token_f1": 0.12345679012345681, "prompt_id": "5f407731bdd7e46c", "reading_canonical_log_prob": -10.788077354431152, "reading_distances": [25, 31, 31, 29], "reading_log_probs": [-14.435845375061035, -19.34140968322754, -14.316780090332031, -12.833436965942383], "reading_logprob_std": 2.835009827916565, "tokenization_path_logprob_std": 43.02503803272751, "trajectory_balance_rmse": 134.91530089813435} {"canonical_length": 226, "canonical_nll": 1.2120939271639934, "generation_rouge_l_f1": 0.256198347107438, "generation_token_f1": 0.4752066115702479, "prompt_id": "c371f7349e2c9b37", "reading_canonical_log_prob": -273.9332275390625, "reading_distances": [24, 30, 21, 19], "reading_log_probs": [-276.6036376953125, -273.3338317871094, -271.4679870605469, -274.60406494140625], "reading_logprob_std": 2.160144335505237, "tokenization_path_logprob_std": 241.12529280601746, "trajectory_balance_rmse": 5014.807454581122} {"canonical_length": 9, "canonical_nll": 1.0244793362087674, "generation_rouge_l_f1": 0.7999999999999999, "generation_token_f1": 0.7999999999999999, "prompt_id": "59b66b10bf9d5bcd", "reading_canonical_log_prob": -9.220314025878906, "reading_distances": [64, 50, 67, 58], "reading_log_probs": [-11.846272468566895, -16.697689056396484, -17.893198013305664, -12.092485427856445], "reading_logprob_std": 3.1151187898559325, "tokenization_path_logprob_std": 14.811675986946215, "trajectory_balance_rmse": 181.02478845494107} {"canonical_length": 387, "canonical_nll": 1.3925502096959788, "generation_rouge_l_f1": 0.15384615384615385, "generation_token_f1": 0.2781065088757397, "prompt_id": "d9970f787452110a", "reading_canonical_log_prob": -538.9169311523438, "reading_distances": [32, 28, 31, 31], "reading_log_probs": [-549.3381958007812, -549.1727294921875, -550.459228515625, -550.4732055664062], "reading_logprob_std": 0.7023089141042389, "tokenization_path_logprob_std": 319.6252814885803, "trajectory_balance_rmse": 7160.697423238505} {"canonical_length": 199, "canonical_nll": 1.1438414607215766, "generation_rouge_l_f1": 0.3180827886710239, "generation_token_f1": 0.531590413943355, "prompt_id": "4f2d283e14485973", "reading_canonical_log_prob": -227.62445068359375, "reading_distances": [20, 21, 24, 20], "reading_log_probs": [-234.77944946289062, -233.8711395263672, -232.24282836914062, -231.898681640625], "reading_logprob_std": 1.3607204300899072, "tokenization_path_logprob_std": 198.68983628349503, "trajectory_balance_rmse": 4149.5613705351425} {"canonical_length": 366, "canonical_nll": 1.345063589961151, "generation_rouge_l_f1": 0.2156521739130435, "generation_token_f1": 0.4034782608695652, "prompt_id": "74c803f89db0a539", "reading_canonical_log_prob": -492.29327392578125, "reading_distances": [12, 15, 17, 12], "reading_log_probs": [-497.11492919921875, -489.6858215332031, -494.6996154785156, -491.93463134765625], "reading_logprob_std": 3.2365269382615103, "tokenization_path_logprob_std": 175.12506496873877, "trajectory_balance_rmse": 7204.213255144621} {"canonical_length": 132, "canonical_nll": 1.3240615382339016, "generation_rouge_l_f1": 0.3405797101449275, "generation_token_f1": 0.5652173913043478, "prompt_id": "456727f2988746cd", "reading_canonical_log_prob": -174.776123046875, "reading_distances": [26, 29, 30, 31], "reading_log_probs": [-180.36795043945312, -182.54110717773438, -184.0253448486328, -186.79127502441406], "reading_logprob_std": 2.6968423004685933, "tokenization_path_logprob_std": 174.1548838298587, "trajectory_balance_rmse": 3174.1963671391195} {"canonical_length": 12, "canonical_nll": 0.3255306084950765, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "4105efabcbebe823", "reading_canonical_log_prob": -3.906367301940918, "reading_distances": [29, 18, 24, 23], "reading_log_probs": [-3.4181275367736816, -3.348874807357788, -3.614330291748047, -5.422823905944824], "reading_logprob_std": 0.9876098936520268, "tokenization_path_logprob_std": 25.43405743525082, "trajectory_balance_rmse": 268.36357897191965} {"canonical_length": 57, "canonical_nll": 1.5481218036852384, "generation_rouge_l_f1": 0.13259668508287292, "generation_token_f1": 0.17679558011049723, "prompt_id": "61331d091d23a9af", "reading_canonical_log_prob": -88.2429428100586, "reading_distances": [16, 16, 15, 16], "reading_log_probs": [-103.94621276855469, -114.00994873046875, -104.85232543945312, -103.42672729492188], "reading_logprob_std": 5.0022383090344515, "tokenization_path_logprob_std": 110.66463459298241, "trajectory_balance_rmse": 1188.634239951792} {"canonical_length": 32, "canonical_nll": 2.770867109298706, "generation_rouge_l_f1": 0.11707317073170731, "generation_token_f1": 0.16585365853658537, "prompt_id": "1a1747f5e15bef6b", "reading_canonical_log_prob": -88.6677474975586, "reading_distances": [21, 38, 30, 33], "reading_log_probs": [-87.76734924316406, -89.10191345214844, -92.43193054199219, -91.21572875976562], "reading_logprob_std": 2.0909941288028455, "tokenization_path_logprob_std": 59.679939630373724, "trajectory_balance_rmse": 856.0914205972797} {"canonical_length": 9, "canonical_nll": 0.6680299970838759, "generation_rouge_l_f1": 0.42857142857142855, "generation_token_f1": 0.5714285714285715, "prompt_id": "8d4efdacc1d27322", "reading_canonical_log_prob": -6.012269973754883, "reading_distances": [33, 36, 36, 33], "reading_log_probs": [-7.5354390144348145, -8.473259925842285, -5.715976715087891, -8.630756378173828], "reading_logprob_std": 1.3389370027833827, "tokenization_path_logprob_std": 16.674266500226217, "trajectory_balance_rmse": 235.9258201998062} {"canonical_length": 263, "canonical_nll": 1.629330943292538, "generation_rouge_l_f1": 0.17938931297709923, "generation_token_f1": 0.40458015267175573, "prompt_id": "123b7e61f56166d9", "reading_canonical_log_prob": -428.5140380859375, "reading_distances": [34, 36, 32, 39], "reading_log_probs": [-461.0875549316406, -448.3791198730469, -450.65191650390625, -456.3132019042969], "reading_logprob_std": 5.7254310787568174, "tokenization_path_logprob_std": 191.9611967075028, "trajectory_balance_rmse": 5017.657677946173} {"canonical_length": 63, "canonical_nll": 1.1953047495039681, "generation_rouge_l_f1": 0.5468749999999999, "generation_token_f1": 0.5468749999999999, "prompt_id": "0407e3355bf3ebea", "reading_canonical_log_prob": -75.30419921875, "reading_distances": [18, 15, 20, 19], "reading_log_probs": [-81.36368560791016, -86.07547760009766, -84.2758560180664, -81.03868103027344], "reading_logprob_std": 2.4130652649830875, "tokenization_path_logprob_std": 47.78381386978652, "trajectory_balance_rmse": 1083.2806951169885} {"canonical_length": 29, "canonical_nll": 0.7511422387484846, "generation_rouge_l_f1": 0.5483870967741935, "generation_token_f1": 0.6129032258064517, "prompt_id": "f217efc1d8ea8fee", "reading_canonical_log_prob": -21.783124923706055, "reading_distances": [10, 17, 13, 15], "reading_log_probs": [-22.522075653076172, -23.132022857666016, -21.95505714416504, -21.6812686920166], "reading_logprob_std": 0.6432597961449186, "tokenization_path_logprob_std": 35.48980706189737, "trajectory_balance_rmse": 511.78931261111154} {"canonical_length": 125, "canonical_nll": 1.6074600830078125, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.421875, "prompt_id": "65c5ecab43401060", "reading_canonical_log_prob": -200.93251037597656, "reading_distances": [31, 17, 21, 32], "reading_log_probs": [-206.49827575683594, -205.03173828125, -206.8839874267578, -204.0282745361328], "reading_logprob_std": 1.322663795413371, "tokenization_path_logprob_std": 190.71595823448848, "trajectory_balance_rmse": 2726.6587880409006} {"canonical_length": 41, "canonical_nll": 0.8965763929413586, "generation_rouge_l_f1": 0.4814814814814815, "generation_token_f1": 0.4814814814814815, "prompt_id": "5cd588db0ae30114", "reading_canonical_log_prob": -36.7596321105957, "reading_distances": [26, 32, 31, 25], "reading_log_probs": [-36.00370788574219, -36.393096923828125, -37.09809112548828, -36.23659896850586], "reading_logprob_std": 0.47145288963888005, "tokenization_path_logprob_std": 26.958504677249312, "trajectory_balance_rmse": 761.1773005386617} {"canonical_length": 145, "canonical_nll": 1.6802721747036637, "generation_rouge_l_f1": 0.2376237623762376, "generation_token_f1": 0.44554455445544555, "prompt_id": "f88ab53f7308d0a9", "reading_canonical_log_prob": -243.63946533203125, "reading_distances": [26, 29, 27, 18], "reading_log_probs": [-262.5938720703125, -259.20184326171875, -264.55908203125, -260.1706848144531], "reading_logprob_std": 2.417582081100888, "tokenization_path_logprob_std": 205.41279887385983, "trajectory_balance_rmse": 3414.5554108280535} {"canonical_length": 55, "canonical_nll": 1.290421641956676, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.5034013605442177, "prompt_id": "8c737ded95e1b468", "reading_canonical_log_prob": -70.97319030761719, "reading_distances": [12, 20, 18, 16], "reading_log_probs": [-87.92411804199219, -92.6528091430664, -91.41007232666016, -89.13389587402344], "reading_logprob_std": 2.1425079509864875, "tokenization_path_logprob_std": 109.86118247859675, "trajectory_balance_rmse": 1013.8690260750152} {"canonical_length": 22, "canonical_nll": 0.552081975069913, "generation_rouge_l_f1": 0.13071895424836602, "generation_token_f1": 0.13071895424836602, "prompt_id": "cd208c8b818d2286", "reading_canonical_log_prob": -12.145803451538086, "reading_distances": [24, 27, 28, 30], "reading_log_probs": [-11.536735534667969, -15.650022506713867, -16.093873977661133, -16.920936584472656], "reading_logprob_std": 2.4009116449906407, "tokenization_path_logprob_std": 46.41330469487402, "trajectory_balance_rmse": 330.9429701382389} {"canonical_length": 165, "canonical_nll": 1.0195562189275569, "generation_rouge_l_f1": 0.302158273381295, "generation_token_f1": 0.4892086330935252, "prompt_id": "a88030476f29d6bf", "reading_canonical_log_prob": -168.22677612304688, "reading_distances": [14, 12, 16, 17], "reading_log_probs": [-170.55810546875, -175.73287963867188, -172.3150634765625, -174.15040588378906], "reading_logprob_std": 2.242096929168282, "tokenization_path_logprob_std": 102.00718091278519, "trajectory_balance_rmse": 3787.678148423015} {"canonical_length": 377, "canonical_nll": 1.290530015049942, "generation_rouge_l_f1": 0.20031298904538342, "generation_token_f1": 0.41627543035993736, "prompt_id": "842add8204336566", "reading_canonical_log_prob": -486.5298156738281, "reading_distances": [35, 31, 24, 24], "reading_log_probs": [-504.9465026855469, -501.7647705078125, -495.7621154785156, -502.10205078125], "reading_logprob_std": 3.861214183839699, "tokenization_path_logprob_std": 160.22851578471102, "trajectory_balance_rmse": 7602.5553516483205} {"canonical_length": 112, "canonical_nll": 1.2014612470354353, "generation_rouge_l_f1": 0.2987804878048781, "generation_token_f1": 0.524390243902439, "prompt_id": "95ed700077e4b7ce", "reading_canonical_log_prob": -134.56365966796875, "reading_distances": [12, 13, 12, 14], "reading_log_probs": [-131.74481201171875, -129.77496337890625, -130.16619873046875, -132.34417724609375], "reading_logprob_std": 1.2325136112262083, "tokenization_path_logprob_std": 225.06121374423415, "trajectory_balance_rmse": 1401.051051130146} {"canonical_length": 11, "canonical_nll": 1.2001592462713069, "generation_rouge_l_f1": 0.380952380952381, "generation_token_f1": 0.380952380952381, "prompt_id": "e5cef34a838082a8", "reading_canonical_log_prob": -13.201751708984375, "reading_distances": [35, 36, 30, 38], "reading_log_probs": [-12.605958938598633, -13.865150451660156, -13.353683471679688, -13.759684562683105], "reading_logprob_std": 0.5710587003430984, "tokenization_path_logprob_std": 32.38711021327868, "trajectory_balance_rmse": 343.5594495669773} {"canonical_length": 10, "canonical_nll": 0.3591897487640381, "generation_rouge_l_f1": 0.8888888888888888, "generation_token_f1": 0.8888888888888888, "prompt_id": "47116aef05367776", "reading_canonical_log_prob": -3.591897487640381, "reading_distances": [32, 37, 32, 38], "reading_log_probs": [-7.0063581466674805, -6.230525970458984, -9.66596508026123, -7.208677768707275], "reading_logprob_std": 1.4864286501943162, "tokenization_path_logprob_std": 28.589049284666423, "trajectory_balance_rmse": 139.32086221195036} {"canonical_length": 375, "canonical_nll": 1.4496888020833334, "generation_rouge_l_f1": 0.1890243902439024, "generation_token_f1": 0.3628048780487805, "prompt_id": "7dea199ced8c5f4e", "reading_canonical_log_prob": -543.63330078125, "reading_distances": [32, 26, 28, 27], "reading_log_probs": [-546.881103515625, -540.7261962890625, -544.4821166992188, -551.7045288085938], "reading_logprob_std": 4.597983362158216, "tokenization_path_logprob_std": 125.0536312492317, "trajectory_balance_rmse": 6965.366404744249} {"canonical_length": 377, "canonical_nll": 1.2978944651960378, "generation_rouge_l_f1": 0.2702702702702703, "generation_token_f1": 0.48966613672496023, "prompt_id": "c009e5e48dec12c7", "reading_canonical_log_prob": -489.30621337890625, "reading_distances": [27, 29, 32, 26], "reading_log_probs": [-487.48699951171875, -489.8751525878906, -485.3941345214844, -487.5289306640625], "reading_logprob_std": 1.830909427984542, "tokenization_path_logprob_std": 337.3766757361184, "trajectory_balance_rmse": 6923.076884703229} {"canonical_length": 206, "canonical_nll": 1.0951241798771238, "generation_rouge_l_f1": 0.3191011235955056, "generation_token_f1": 0.5123595505617977, "prompt_id": "12ce96bbad2517ca", "reading_canonical_log_prob": -225.5955810546875, "reading_distances": [21, 23, 23, 20], "reading_log_probs": [-231.7244110107422, -231.04693603515625, -233.56887817382812, -232.33740234375], "reading_logprob_std": 1.0715572001018427, "tokenization_path_logprob_std": 121.93052299437966, "trajectory_balance_rmse": 4472.880632384204} {"canonical_length": 193, "canonical_nll": 1.3699944846988341, "generation_rouge_l_f1": 0.3672566371681416, "generation_token_f1": 0.6238938053097346, "prompt_id": "cda8b27fc8b243fa", "reading_canonical_log_prob": -264.408935546875, "reading_distances": [22, 12, 18, 13], "reading_log_probs": [-265.5367431640625, -267.88726806640625, -266.2191162109375, -266.3933410644531], "reading_logprob_std": 0.9903392959925494, "tokenization_path_logprob_std": 212.8354130511294, "trajectory_balance_rmse": 4038.9520435919044} {"canonical_length": 247, "canonical_nll": 0.9340132122580339, "generation_rouge_l_f1": 0.19767441860465115, "generation_token_f1": 0.3798449612403101, "prompt_id": "1ba77d2f5f7df1e4", "reading_canonical_log_prob": -230.70126342773438, "reading_distances": [17, 17, 17, 19], "reading_log_probs": [-234.57846069335938, -239.02609252929688, -234.66644287109375, -238.86891174316406], "reading_logprob_std": 2.4981515764149447, "tokenization_path_logprob_std": 163.06358017977334, "trajectory_balance_rmse": 5705.280400078084} {"canonical_length": 10, "canonical_nll": 0.7781905174255371, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.5714285714285715, "prompt_id": "78c9a1c087a98da8", "reading_canonical_log_prob": -7.781905174255371, "reading_distances": [16, 21, 21, 21], "reading_log_probs": [-6.25946044921875, -6.1109514236450195, -7.034836769104004, -5.8169708251953125], "reading_logprob_std": 0.5197953617078058, "tokenization_path_logprob_std": 50.027796518435714, "trajectory_balance_rmse": 170.58978646618783} {"canonical_length": 128, "canonical_nll": 0.9665576219558716, "generation_rouge_l_f1": 0.36538461538461536, "generation_token_f1": 0.5448717948717948, "prompt_id": "6de4ae8779beb5f7", "reading_canonical_log_prob": -123.71937561035156, "reading_distances": [38, 40, 33, 35], "reading_log_probs": [-130.42352294921875, -131.84742736816406, -144.5048828125, -126.57121276855469], "reading_logprob_std": 7.771843588627852, "tokenization_path_logprob_std": 147.17130844803864, "trajectory_balance_rmse": 2686.9697633172145} {"canonical_length": 292, "canonical_nll": 0.9341179834653254, "generation_rouge_l_f1": 0.29380530973451324, "generation_token_f1": 0.5274336283185841, "prompt_id": "bcb2851b6b713edf", "reading_canonical_log_prob": -272.762451171875, "reading_distances": [46, 47, 43, 50], "reading_log_probs": [-288.560546875, -287.9048767089844, -291.4202880859375, -286.53631591796875], "reading_logprob_std": 2.0572998900855266, "tokenization_path_logprob_std": 227.8827631744119, "trajectory_balance_rmse": 5056.425234815382} {"canonical_length": 207, "canonical_nll": 1.6832921124886775, "generation_rouge_l_f1": 0.25806451612903225, "generation_token_f1": 0.4559139784946237, "prompt_id": "790529293961a686", "reading_canonical_log_prob": -348.44146728515625, "reading_distances": [30, 26, 26, 35], "reading_log_probs": [-344.6898498535156, -342.9771728515625, -343.07806396484375, -344.8731689453125], "reading_logprob_std": 1.0162063540654633, "tokenization_path_logprob_std": 167.7643246323325, "trajectory_balance_rmse": 4708.139508180195} {"canonical_length": 130, "canonical_nll": 1.779036653958834, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.368421052631579, "prompt_id": "ecb22237f8778fd5", "reading_canonical_log_prob": -231.27476501464844, "reading_distances": [19, 21, 20, 14], "reading_log_probs": [-238.4727325439453, -238.56138610839844, -238.62661743164062, -232.453125], "reading_logprob_std": 3.0508787353496407, "tokenization_path_logprob_std": 193.75607954833822, "trajectory_balance_rmse": 2573.3612556895005} {"canonical_length": 37, "canonical_nll": 1.2729719007337414, "generation_rouge_l_f1": 0.027874564459930317, "generation_token_f1": 0.04181184668989547, "prompt_id": "db3b67bebed71ebb", "reading_canonical_log_prob": -47.09996032714844, "reading_distances": [29, 33, 32, 29], "reading_log_probs": [-45.0899658203125, -48.43071746826172, -48.39970397949219, -46.045936584472656], "reading_logprob_std": 1.689606241297139, "tokenization_path_logprob_std": 54.594432007108985, "trajectory_balance_rmse": 735.744385266469} {"canonical_length": 10, "canonical_nll": 0.19459301233291626, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "a75ceffa04e21449", "reading_canonical_log_prob": -1.9459301233291626, "reading_distances": [38, 37, 35, 32], "reading_log_probs": [-1.165687084197998, -1.7524139881134033, -1.3256689310073853, -1.82191801071167], "reading_logprob_std": 0.32063569436729905, "tokenization_path_logprob_std": 32.825455397259795, "trajectory_balance_rmse": 137.01092281297184} {"canonical_length": 7, "canonical_nll": 0.044092229434422085, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "7af908712adcd78e", "reading_canonical_log_prob": -0.3086456060409546, "reading_distances": [31, 36, 38, 38], "reading_log_probs": [-2.050161838531494, -8.120285034179688, -5.072097301483154, -3.224264621734619], "reading_logprob_std": 2.646286027046577, "tokenization_path_logprob_std": 64.75034190763412, "trajectory_balance_rmse": 180.20712540786752} {"canonical_length": 45, "canonical_nll": 1.0759835137261284, "generation_rouge_l_f1": 0.12598425196850396, "generation_token_f1": 0.2047244094488189, "prompt_id": "3c24680236bfa43e", "reading_canonical_log_prob": -48.41925811767578, "reading_distances": [26, 20, 23, 28], "reading_log_probs": [-41.97296142578125, -43.20393371582031, -41.485809326171875, -40.825592041015625], "reading_logprob_std": 1.0047144642522388, "tokenization_path_logprob_std": 144.09499993089725, "trajectory_balance_rmse": 826.9999524402823} {"canonical_length": 22, "canonical_nll": 0.41737487099387427, "generation_rouge_l_f1": 0.3148148148148148, "generation_token_f1": 0.3148148148148148, "prompt_id": "8be7f076246178b6", "reading_canonical_log_prob": -9.182247161865234, "reading_distances": [28, 32, 38, 35], "reading_log_probs": [-12.661697387695312, -14.5625581741333, -13.46589469909668, -12.004827499389648], "reading_logprob_std": 1.101924734245693, "tokenization_path_logprob_std": 55.351879934792805, "trajectory_balance_rmse": 448.6043902408566} {"canonical_length": 20, "canonical_nll": 1.7637624740600586, "generation_rouge_l_f1": 0.3478260869565218, "generation_token_f1": 0.3478260869565218, "prompt_id": "ba65e8996b196d2c", "reading_canonical_log_prob": -35.27524948120117, "reading_distances": [39, 40, 26, 40], "reading_log_probs": [-34.51456069946289, -36.70458984375, -30.604049682617188, -34.840511322021484], "reading_logprob_std": 2.5630963385612486, "tokenization_path_logprob_std": 44.7082546693838, "trajectory_balance_rmse": 308.9218602539718} {"canonical_length": 8, "canonical_nll": 0.9824860095977783, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.4444444444444444, "prompt_id": "e026b81849f6617b", "reading_canonical_log_prob": -7.859888076782227, "reading_distances": [29, 25, 43, 28], "reading_log_probs": [-12.86572551727295, -9.677931785583496, -7.92563009262085, -13.630949020385742], "reading_logprob_std": 2.683277057994246, "tokenization_path_logprob_std": 57.06460636270658, "trajectory_balance_rmse": 208.13534833665435} {"canonical_length": 84, "canonical_nll": 2.7466031029110862, "generation_rouge_l_f1": 0.4230769230769231, "generation_token_f1": 0.4423076923076923, "prompt_id": "1c6873b5ff06ab16", "reading_canonical_log_prob": -230.71466064453125, "reading_distances": [22, 21, 23, 21], "reading_log_probs": [-238.09303283691406, -245.54637145996094, -232.58963012695312, -235.73239135742188], "reading_logprob_std": 5.518760623151345, "tokenization_path_logprob_std": 82.50462930495925, "trajectory_balance_rmse": 2042.3310582949766} {"canonical_length": 212, "canonical_nll": 0.9524669287339697, "generation_rouge_l_f1": 0.27293064876957496, "generation_token_f1": 0.4787472035794184, "prompt_id": "9cabb6227a1b5412", "reading_canonical_log_prob": -201.92298889160156, "reading_distances": [16, 14, 17, 20], "reading_log_probs": [-205.4437255859375, -203.7789306640625, -205.3084259033203, -207.7887725830078], "reading_logprob_std": 1.6547759442435537, "tokenization_path_logprob_std": 105.34735138390124, "trajectory_balance_rmse": 3499.3696445157966} {"canonical_length": 123, "canonical_nll": 1.4218480800225484, "generation_rouge_l_f1": 0.24489795918367346, "generation_token_f1": 0.44387755102040816, "prompt_id": "3aaeb505baa59e0c", "reading_canonical_log_prob": -174.88731384277344, "reading_distances": [16, 18, 18, 19], "reading_log_probs": [-182.09442138671875, -179.60939025878906, -184.3189697265625, -187.6394805908203], "reading_logprob_std": 3.410278073994133, "tokenization_path_logprob_std": 219.14041277964125, "trajectory_balance_rmse": 2726.842563300302} {"canonical_length": 305, "canonical_nll": 1.0151317158683402, "generation_rouge_l_f1": 0.2765957446808511, "generation_token_f1": 0.5035460992907802, "prompt_id": "5964386ba6fbd7d7", "reading_canonical_log_prob": -309.61517333984375, "reading_distances": [30, 32, 37, 40], "reading_log_probs": [-307.4267578125, -309.13226318359375, -306.66802978515625, -305.581787109375], "reading_logprob_std": 1.4929440554365268, "tokenization_path_logprob_std": 154.05964982857049, "trajectory_balance_rmse": 5960.779779400419} {"canonical_length": 125, "canonical_nll": 1.1134493408203125, "generation_rouge_l_f1": 0.22465753424657534, "generation_token_f1": 0.39452054794520547, "prompt_id": "a562f2a21ae863a6", "reading_canonical_log_prob": -139.18116760253906, "reading_distances": [10, 8, 8, 7], "reading_log_probs": [-142.1015167236328, -140.02301025390625, -139.90061950683594, -140.17718505859375], "reading_logprob_std": 1.0401293160972573, "tokenization_path_logprob_std": 85.21431709865325, "trajectory_balance_rmse": 2489.0580434315975} {"canonical_length": 166, "canonical_nll": 1.206667245152485, "generation_rouge_l_f1": 0.22966507177033493, "generation_token_f1": 0.35885167464114837, "prompt_id": "f84f62888219bcaf", "reading_canonical_log_prob": -200.3067626953125, "reading_distances": [87, 91, 90, 94], "reading_log_probs": [-200.49407958984375, -198.35992431640625, -202.20013427734375, -198.7978057861328], "reading_logprob_std": 1.7525547550351783, "tokenization_path_logprob_std": 57.646530390497105, "trajectory_balance_rmse": 2432.395827192909} {"canonical_length": 86, "canonical_nll": 0.670362250749455, "generation_rouge_l_f1": 0.06896551724137931, "generation_token_f1": 0.06896551724137931, "prompt_id": "6b3af0ffe0ba6c9d", "reading_canonical_log_prob": -57.651153564453125, "reading_distances": [32, 26, 25, 21], "reading_log_probs": [-70.80048370361328, -64.86492919921875, -66.02185821533203, -66.72183990478516], "reading_logprob_std": 2.581633535702253, "tokenization_path_logprob_std": 122.40219852351039, "trajectory_balance_rmse": 1899.242758127129} {"canonical_length": 61, "canonical_nll": 1.0483203325115267, "generation_rouge_l_f1": 0.4523809523809524, "generation_token_f1": 0.5714285714285715, "prompt_id": "53e2d36ebe126c1f", "reading_canonical_log_prob": -63.947540283203125, "reading_distances": [22, 21, 23, 24], "reading_log_probs": [-67.65516662597656, -64.06208801269531, -65.0958480834961, -68.19351959228516], "reading_logprob_std": 1.9892022018180744, "tokenization_path_logprob_std": 100.76322769799454, "trajectory_balance_rmse": 1211.8200779280419} {"canonical_length": 133, "canonical_nll": 0.8605572693330005, "generation_rouge_l_f1": 0.3739837398373984, "generation_token_f1": 0.6016260162601625, "prompt_id": "bce0ef5bdff3466e", "reading_canonical_log_prob": -114.45411682128906, "reading_distances": [23, 21, 18, 19], "reading_log_probs": [-118.2449951171875, -122.91222381591797, -117.01213073730469, -117.52991485595703], "reading_logprob_std": 2.7059004440257235, "tokenization_path_logprob_std": 87.79416687628391, "trajectory_balance_rmse": 2568.113410853719} {"canonical_length": 74, "canonical_nll": 2.3166087382548564, "generation_rouge_l_f1": 0.15017064846416384, "generation_token_f1": 0.21160409556313994, "prompt_id": "2c030e502077ba6b", "reading_canonical_log_prob": -171.42904663085938, "reading_distances": [32, 25, 29, 23], "reading_log_probs": [-182.505615234375, -176.78067016601562, -178.9287109375, -175.9214324951172], "reading_logprob_std": 2.934234637427425, "tokenization_path_logprob_std": 37.44769854742927, "trajectory_balance_rmse": 1241.1774687945751} {"canonical_length": 152, "canonical_nll": 1.4326962922748767, "generation_rouge_l_f1": 0.17961165048543692, "generation_token_f1": 0.32038834951456313, "prompt_id": "b73bca8ae43941f1", "reading_canonical_log_prob": -217.76983642578125, "reading_distances": [18, 23, 21, 12], "reading_log_probs": [-222.11550903320312, -226.2329864501953, -224.84616088867188, -222.55380249023438], "reading_logprob_std": 1.9432954777453766, "tokenization_path_logprob_std": 162.08125462756362, "trajectory_balance_rmse": 3273.071002911467} {"canonical_length": 139, "canonical_nll": 1.4328301518941098, "generation_rouge_l_f1": 0.2266009852216749, "generation_token_f1": 0.4088669950738916, "prompt_id": "98eb458a641b4b4a", "reading_canonical_log_prob": -199.16339111328125, "reading_distances": [37, 38, 30, 23], "reading_log_probs": [-202.7073974609375, -197.16424560546875, -207.15045166015625, -203.12210083007812], "reading_logprob_std": 4.103730037504825, "tokenization_path_logprob_std": 155.36039517188647, "trajectory_balance_rmse": 3380.142292337568} {"canonical_length": 79, "canonical_nll": 1.9025701208959651, "generation_rouge_l_f1": 0.23636363636363633, "generation_token_f1": 0.38181818181818183, "prompt_id": "3cfac8c1932346e3", "reading_canonical_log_prob": -150.30303955078125, "reading_distances": [22, 28, 26, 21], "reading_log_probs": [-152.00543212890625, -154.56163024902344, -160.18019104003906, -150.5010528564453], "reading_logprob_std": 4.256037076714316, "tokenization_path_logprob_std": 127.72833740697381, "trajectory_balance_rmse": 1577.842581516834} {"canonical_length": 22, "canonical_nll": 1.2422289414839311, "generation_rouge_l_f1": 0.16326530612244897, "generation_token_f1": 0.20408163265306123, "prompt_id": "7f70df4099ae17f5", "reading_canonical_log_prob": -27.329036712646484, "reading_distances": [28, 29, 27, 31], "reading_log_probs": [-30.740577697753906, -29.039113998413086, -30.001554489135742, -32.16090393066406], "reading_logprob_std": 1.3163432531340462, "tokenization_path_logprob_std": 59.281836686848, "trajectory_balance_rmse": 453.72407427203586} {"canonical_length": 87, "canonical_nll": 0.5936050853510012, "generation_rouge_l_f1": 0.49218749999999994, "generation_token_f1": 0.609375, "prompt_id": "0f424e2f0c95df2d", "reading_canonical_log_prob": -51.64364242553711, "reading_distances": [21, 22, 21, 18], "reading_log_probs": [-53.50151062011719, -51.2110481262207, -54.45730972290039, -54.893287658691406], "reading_logprob_std": 1.6427617187476835, "tokenization_path_logprob_std": 85.1359703381122, "trajectory_balance_rmse": 1509.6311833169186} {"canonical_length": 222, "canonical_nll": 1.070265142767279, "generation_rouge_l_f1": 0.2985685071574642, "generation_token_f1": 0.49079754601226994, "prompt_id": "7d4605b87d9b1016", "reading_canonical_log_prob": -237.59886169433594, "reading_distances": [26, 19, 24, 24], "reading_log_probs": [-242.6002960205078, -242.52256774902344, -243.8340606689453, -245.32688903808594], "reading_logprob_std": 1.315779062204634, "tokenization_path_logprob_std": 252.95414047780415, "trajectory_balance_rmse": 5243.251882280219} {"canonical_length": 68, "canonical_nll": 2.1900695351993336, "generation_rouge_l_f1": 0.17316017316017315, "generation_token_f1": 0.2510822510822511, "prompt_id": "7bba6cec3efa18f3", "reading_canonical_log_prob": -148.9247283935547, "reading_distances": [16, 18, 20, 17], "reading_log_probs": [-153.70697021484375, -150.70919799804688, -152.7168731689453, -152.96722412109375], "reading_logprob_std": 1.2814793492535703, "tokenization_path_logprob_std": 159.63654201092766, "trajectory_balance_rmse": 1526.666811450276} {"canonical_length": 10, "canonical_nll": 0.6443393707275391, "generation_rouge_l_f1": 0.631578947368421, "generation_token_f1": 0.631578947368421, "prompt_id": "23ac4432bd5b4331", "reading_canonical_log_prob": -6.443393707275391, "reading_distances": [23, 18, 24, 23], "reading_log_probs": [-10.979696273803711, -8.043773651123047, -8.438203811645508, -8.440049171447754], "reading_logprob_std": 1.349112233747662, "tokenization_path_logprob_std": 64.18709135639567, "trajectory_balance_rmse": 236.19882106687817} {"canonical_length": 35, "canonical_nll": 0.7511127471923829, "generation_rouge_l_f1": 0.15533980582524273, "generation_token_f1": 0.1941747572815534, "prompt_id": "f25ec758946024eb", "reading_canonical_log_prob": -26.2889461517334, "reading_distances": [37, 39, 42, 43], "reading_log_probs": [-32.7425537109375, -34.132652282714844, -30.646303176879883, -31.901954650878906], "reading_logprob_std": 1.4645974918373836, "tokenization_path_logprob_std": 33.413698602379526, "trajectory_balance_rmse": 375.7889386885178} {"canonical_length": 225, "canonical_nll": 1.5907000054253473, "generation_rouge_l_f1": 0.2597938144329897, "generation_token_f1": 0.5278350515463917, "prompt_id": "e70426e74522ced9", "reading_canonical_log_prob": -357.9075012207031, "reading_distances": [29, 30, 30, 31], "reading_log_probs": [-382.3168640136719, -387.95819091796875, -383.479248046875, -388.8794250488281], "reading_logprob_std": 3.2444109177777505, "tokenization_path_logprob_std": 279.7021718687069, "trajectory_balance_rmse": 4470.219331576924} {"canonical_length": 17, "canonical_nll": 0.43979372697718006, "generation_rouge_l_f1": 0.8333333333333333, "generation_token_f1": 0.8333333333333333, "prompt_id": "fa1698c1de8ba5af", "reading_canonical_log_prob": -7.4764933586120605, "reading_distances": [22, 23, 21, 22], "reading_log_probs": [-11.920595169067383, -9.595667839050293, -9.523252487182617, -8.130912780761719], "reading_logprob_std": 1.5706578465334537, "tokenization_path_logprob_std": 42.05024893242124, "trajectory_balance_rmse": 327.63313316864105} {"canonical_length": 41, "canonical_nll": 1.1855360821979801, "generation_rouge_l_f1": 0.5411764705882353, "generation_token_f1": 0.5411764705882353, "prompt_id": "2ce41f448f87f273", "reading_canonical_log_prob": -48.60697937011719, "reading_distances": [38, 37, 28, 41], "reading_log_probs": [-46.48384475708008, -46.48104476928711, -45.93769073486328, -45.47064971923828], "reading_logprob_std": 0.48811843805118327, "tokenization_path_logprob_std": 76.91022471623575, "trajectory_balance_rmse": 733.0926168660467} {"canonical_length": 17, "canonical_nll": 0.2547653422636144, "generation_rouge_l_f1": 0.9696969696969697, "generation_token_f1": 0.9696969696969697, "prompt_id": "f7f7943a60f6d582", "reading_canonical_log_prob": -4.331010818481445, "reading_distances": [33, 25, 33, 33], "reading_log_probs": [-3.4946129322052, -3.476958751678467, -4.276611804962158, -4.0853047370910645], "reading_logprob_std": 0.40894976802918936, "tokenization_path_logprob_std": 18.869879933728477, "trajectory_balance_rmse": 304.0788130015145} {"canonical_length": 45, "canonical_nll": 2.3315466986762154, "generation_rouge_l_f1": 0.27522935779816515, "generation_token_f1": 0.3669724770642202, "prompt_id": "d6cc1411da921bc9", "reading_canonical_log_prob": -104.91960144042969, "reading_distances": [17, 21, 19, 19], "reading_log_probs": [-107.37858581542969, -106.666748046875, -101.98953247070312, -109.93190002441406], "reading_logprob_std": 3.3126875764689823, "tokenization_path_logprob_std": 45.32736541154779, "trajectory_balance_rmse": 1097.1892839412353} {"canonical_length": 19, "canonical_nll": 0.21273643092105263, "generation_rouge_l_f1": 0.19298245614035087, "generation_token_f1": 0.2631578947368421, "prompt_id": "9f836d8d503579f9", "reading_canonical_log_prob": -4.0419921875, "reading_distances": [34, 42, 46, 38], "reading_log_probs": [-7.058588981628418, -7.261124610900879, -5.595542907714844, -4.394725799560547], "reading_logprob_std": 1.345054454634114, "tokenization_path_logprob_std": 61.08617281857506, "trajectory_balance_rmse": 305.4784474125672} {"canonical_length": 193, "canonical_nll": 1.4528008495587759, "generation_rouge_l_f1": 0.22522522522522523, "generation_token_f1": 0.42342342342342343, "prompt_id": "d3c6858742c763af", "reading_canonical_log_prob": -280.39056396484375, "reading_distances": [35, 34, 29, 42], "reading_log_probs": [-293.6461486816406, -287.970458984375, -296.06982421875, -297.84564208984375], "reading_logprob_std": 4.301122810887145, "tokenization_path_logprob_std": 123.36104144596882, "trajectory_balance_rmse": 4495.8373317546575} {"canonical_length": 19, "canonical_nll": 0.799637894881399, "generation_rouge_l_f1": 0.6341463414634146, "generation_token_f1": 0.6341463414634146, "prompt_id": "923c3d05ff40a1cb", "reading_canonical_log_prob": -15.193120002746582, "reading_distances": [31, 34, 27, 31], "reading_log_probs": [-18.581125259399414, -16.65011978149414, -22.29979705810547, -18.494518280029297], "reading_logprob_std": 2.369346052556864, "tokenization_path_logprob_std": 21.320865642645952, "trajectory_balance_rmse": 596.0058373561976} {"canonical_length": 200, "canonical_nll": 1.4670208740234374, "generation_rouge_l_f1": 0.2937365010799136, "generation_token_f1": 0.509719222462203, "prompt_id": "180d045694c4d4f7", "reading_canonical_log_prob": -293.4041748046875, "reading_distances": [12, 11, 15, 8], "reading_log_probs": [-291.99810791015625, -293.36090087890625, -292.7268981933594, -300.76708984375], "reading_logprob_std": 4.074122198579759, "tokenization_path_logprob_std": 252.49568968999108, "trajectory_balance_rmse": 4311.020219759218} {"canonical_length": 314, "canonical_nll": 1.0183821757128284, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.5504273504273505, "prompt_id": "1350704a8f43bb33", "reading_canonical_log_prob": -319.7720031738281, "reading_distances": [33, 45, 36, 44], "reading_log_probs": [-336.68450927734375, -335.3896789550781, -338.2474365234375, -337.6414794921875], "reading_logprob_std": 1.2463174767523943, "tokenization_path_logprob_std": 369.5684227145181, "trajectory_balance_rmse": 6655.283482774238} {"canonical_length": 79, "canonical_nll": 0.6736268635037579, "generation_rouge_l_f1": 0.26822157434402333, "generation_token_f1": 0.37900874635568516, "prompt_id": "3269989441292641", "reading_canonical_log_prob": -53.216522216796875, "reading_distances": [29, 31, 32, 34], "reading_log_probs": [-59.7449836730957, -58.91060256958008, -61.25739288330078, -61.45512771606445], "reading_logprob_std": 1.2223371791528217, "tokenization_path_logprob_std": 59.04666471023274, "trajectory_balance_rmse": 1898.6595448362116} {"canonical_length": 204, "canonical_nll": 1.2148681341433059, "generation_rouge_l_f1": 0.2838427947598254, "generation_token_f1": 0.5851528384279476, "prompt_id": "0f5981f2ed369cdc", "reading_canonical_log_prob": -247.83309936523438, "reading_distances": [28, 42, 38, 33], "reading_log_probs": [-257.72161865234375, -265.2259826660156, -256.5826416015625, -262.0830383300781], "reading_logprob_std": 3.994522683245533, "tokenization_path_logprob_std": 422.26694774447566, "trajectory_balance_rmse": 5384.001285509446} {"canonical_length": 53, "canonical_nll": 0.8358042015219634, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "801510c180dde0a4", "reading_canonical_log_prob": -44.29762268066406, "reading_distances": [34, 36, 39, 36], "reading_log_probs": [-42.188907623291016, -42.73442840576172, -43.656898498535156, -42.413021087646484], "reading_logprob_std": 0.6457748400391762, "tokenization_path_logprob_std": 53.66166729263354, "trajectory_balance_rmse": 1191.6245818244151} {"canonical_length": 18, "canonical_nll": 2.1550827026367188, "generation_rouge_l_f1": 0.21301775147928995, "generation_token_f1": 0.23668639053254437, "prompt_id": "a62ba8086bae83ba", "reading_canonical_log_prob": -38.79148864746094, "reading_distances": [21, 20, 17, 18], "reading_log_probs": [-36.82172393798828, -41.963050842285156, -41.81856918334961, -38.480751037597656], "reading_logprob_std": 2.5403804261334177, "tokenization_path_logprob_std": 35.2723641721119, "trajectory_balance_rmse": 348.40865546184887} {"canonical_length": 278, "canonical_nll": 1.1499037708309914, "generation_rouge_l_f1": 0.2531645569620253, "generation_token_f1": 0.4767932489451477, "prompt_id": "6197952fa284f196", "reading_canonical_log_prob": -319.6732482910156, "reading_distances": [20, 24, 23, 23], "reading_log_probs": [-314.84429931640625, -315.0085754394531, -315.7027587890625, -320.8558349609375], "reading_logprob_std": 2.8596186893808757, "tokenization_path_logprob_std": 141.10663669212897, "trajectory_balance_rmse": 5914.820775810243} {"canonical_length": 13, "canonical_nll": 0.9383912453284631, "generation_rouge_l_f1": 0.3636363636363636, "generation_token_f1": 0.45454545454545453, "prompt_id": "d2ae9ca225f37e77", "reading_canonical_log_prob": -12.19908618927002, "reading_distances": [24, 28, 33, 31], "reading_log_probs": [-11.060937881469727, -11.385209083557129, -12.788249969482422, -11.509937286376953], "reading_logprob_std": 0.7587557520622181, "tokenization_path_logprob_std": 12.540352093269005, "trajectory_balance_rmse": 239.89314862638136} {"canonical_length": 14, "canonical_nll": 2.7168472834995816, "generation_rouge_l_f1": 0.2608695652173913, "generation_token_f1": 0.34782608695652173, "prompt_id": "3d0956e900302c45", "reading_canonical_log_prob": -38.03586196899414, "reading_distances": [19, 18, 20, 17], "reading_log_probs": [-40.09354782104492, -39.37928009033203, -37.11897277832031, -39.00089645385742], "reading_logprob_std": 1.2697100104547188, "tokenization_path_logprob_std": 36.18729349439533, "trajectory_balance_rmse": 280.5756769586484} {"canonical_length": 13, "canonical_nll": 0.3200990970318134, "generation_rouge_l_f1": 0.2195121951219512, "generation_token_f1": 0.2195121951219512, "prompt_id": "360b4e609ec3ff2b", "reading_canonical_log_prob": -4.161288261413574, "reading_distances": [22, 16, 20, 23], "reading_log_probs": [-4.91032075881958, -3.426063060760498, -3.8051517009735107, -4.517561435699463], "reading_logprob_std": 0.672141104749779, "tokenization_path_logprob_std": 20.06404148010655, "trajectory_balance_rmse": 114.82528879438074} {"canonical_length": 86, "canonical_nll": 1.771823616914971, "generation_rouge_l_f1": 0.1945288753799392, "generation_token_f1": 0.3221884498480243, "prompt_id": "1ad837078d8b4fb2", "reading_canonical_log_prob": -152.3768310546875, "reading_distances": [31, 36, 32, 30], "reading_log_probs": [-148.65956115722656, -149.0208740234375, -150.61585998535156, -143.31814575195312], "reading_logprob_std": 3.1729336269628754, "tokenization_path_logprob_std": 231.57037104813713, "trajectory_balance_rmse": 1860.6971666517936} {"canonical_length": 20, "canonical_nll": 0.655671501159668, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2125, "prompt_id": "24f992c3a95d7b8a", "reading_canonical_log_prob": -13.11343002319336, "reading_distances": [38, 41, 32, 31], "reading_log_probs": [-15.915163040161133, -12.040850639343262, -13.521208763122559, -12.874611854553223], "reading_logprob_std": 1.6656072634523635, "tokenization_path_logprob_std": 51.734601408437086, "trajectory_balance_rmse": 508.4413658074277} {"canonical_length": 29, "canonical_nll": 0.80294760342302, "generation_rouge_l_f1": 0.11650485436893206, "generation_token_f1": 0.1553398058252427, "prompt_id": "429af0c846ac90d6", "reading_canonical_log_prob": -23.285480499267578, "reading_distances": [20, 20, 14, 22], "reading_log_probs": [-23.953411102294922, -25.540498733520508, -23.36055564880371, -24.498855590820312], "reading_logprob_std": 0.9264957939875819, "tokenization_path_logprob_std": 53.633905626028884, "trajectory_balance_rmse": 640.6840054870985} {"canonical_length": 132, "canonical_nll": 1.2007735281279592, "generation_rouge_l_f1": 0.22330097087378642, "generation_token_f1": 0.4029126213592233, "prompt_id": "a4d2b99ec03649f1", "reading_canonical_log_prob": -158.50210571289062, "reading_distances": [19, 21, 19, 21], "reading_log_probs": [-163.26513671875, -163.1295166015625, -161.92811584472656, -160.25991821289062], "reading_logprob_std": 1.393383111093799, "tokenization_path_logprob_std": 85.6639141732628, "trajectory_balance_rmse": 2717.0869997863024} {"canonical_length": 27, "canonical_nll": 0.35156546698676217, "generation_rouge_l_f1": 0.419047619047619, "generation_token_f1": 0.43809523809523804, "prompt_id": "363a6f01446a3b83", "reading_canonical_log_prob": -9.492267608642578, "reading_distances": [45, 40, 44, 33], "reading_log_probs": [-10.975530624389648, -10.896849632263184, -10.236871719360352, -10.6309175491333], "reading_logprob_std": 0.33318271794636833, "tokenization_path_logprob_std": 71.19395838240577, "trajectory_balance_rmse": 662.9145920874137} {"canonical_length": 56, "canonical_nll": 2.067905834742955, "generation_rouge_l_f1": 0.1111111111111111, "generation_token_f1": 0.22875816993464054, "prompt_id": "15d4cbd9fbd2e327", "reading_canonical_log_prob": -115.80272674560547, "reading_distances": [28, 29, 30, 28], "reading_log_probs": [-117.78805541992188, -118.4791259765625, -118.26309967041016, -118.13816833496094], "reading_logprob_std": 0.2893025988401759, "tokenization_path_logprob_std": 77.48431450653048, "trajectory_balance_rmse": 1169.3733475259396} {"canonical_length": 56, "canonical_nll": 1.1565919603620256, "generation_rouge_l_f1": 0.29629629629629634, "generation_token_f1": 0.35802469135802467, "prompt_id": "c741581c2c318b9d", "reading_canonical_log_prob": -64.76914978027344, "reading_distances": [19, 28, 30, 29], "reading_log_probs": [-60.033390045166016, -63.46987533569336, -57.694480895996094, -58.9355583190918], "reading_logprob_std": 2.4822813191716593, "tokenization_path_logprob_std": 98.67460784426727, "trajectory_balance_rmse": 1014.0524285770193} {"canonical_length": 10, "canonical_nll": 0.8280990600585938, "generation_rouge_l_f1": 0.15311004784688995, "generation_token_f1": 0.15311004784688995, "prompt_id": "1551065a718dd716", "reading_canonical_log_prob": -8.280990600585938, "reading_distances": [19, 24, 17, 29], "reading_log_probs": [-11.68053913116455, -10.66373062133789, -11.183500289916992, -10.579060554504395], "reading_logprob_std": 0.5112793701868531, "tokenization_path_logprob_std": 54.34629939400439, "trajectory_balance_rmse": 170.82495510139196} {"canonical_length": 22, "canonical_nll": 1.9351052370938389, "generation_rouge_l_f1": 0.08333333333333334, "generation_token_f1": 0.08333333333333334, "prompt_id": "0281e958164e52e0", "reading_canonical_log_prob": -42.57231521606445, "reading_distances": [20, 23, 25, 22], "reading_log_probs": [-41.51156234741211, -43.12125015258789, -39.21335220336914, -41.96870422363281], "reading_logprob_std": 1.639965361594017, "tokenization_path_logprob_std": 38.96573460744457, "trajectory_balance_rmse": 366.99556350189897} {"canonical_length": 338, "canonical_nll": 1.3099116940470137, "generation_rouge_l_f1": 0.2023217247097844, "generation_token_f1": 0.43781094527363185, "prompt_id": "059afc2f6690f050", "reading_canonical_log_prob": -442.7501525878906, "reading_distances": [28, 20, 24, 34], "reading_log_probs": [-447.0152893066406, -444.9050598144531, -449.0482177734375, -446.07135009765625], "reading_logprob_std": 1.7527269217795527, "tokenization_path_logprob_std": 348.6458082681006, "trajectory_balance_rmse": 6287.540128871158} {"canonical_length": 11, "canonical_nll": 1.1293325424194336, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "3463b16713cbef14", "reading_canonical_log_prob": -12.42265796661377, "reading_distances": [37, 40, 33, 39], "reading_log_probs": [-16.514720916748047, -10.663093566894531, -10.852685928344727, -12.119529724121094], "reading_logprob_std": 2.729162466424709, "tokenization_path_logprob_std": 43.89098274733872, "trajectory_balance_rmse": 283.9603191445168} {"canonical_length": 184, "canonical_nll": 1.944715748662534, "generation_rouge_l_f1": 0.22779043280182235, "generation_token_f1": 0.38268792710706145, "prompt_id": "75052e7415fb5476", "reading_canonical_log_prob": -357.82769775390625, "reading_distances": [25, 20, 30, 28], "reading_log_probs": [-377.6701354980469, -372.42218017578125, -371.6884765625, -372.96881103515625], "reading_logprob_std": 2.7064754608394392, "tokenization_path_logprob_std": 212.8437443741359, "trajectory_balance_rmse": 3691.3013678396796} {"canonical_length": 155, "canonical_nll": 2.30206298828125, "generation_rouge_l_f1": 0.2287390029325513, "generation_token_f1": 0.39882697947214074, "prompt_id": "3e8a52f189b6e093", "reading_canonical_log_prob": -356.81976318359375, "reading_distances": [27, 25, 28, 31], "reading_log_probs": [-360.8912353515625, -358.2234191894531, -358.7425537109375, -355.427734375], "reading_logprob_std": 2.2482826826183, "tokenization_path_logprob_std": 130.23500646906447, "trajectory_balance_rmse": 3504.9955334797573} {"canonical_length": 278, "canonical_nll": 1.453103703560589, "generation_rouge_l_f1": 0.27889908256880735, "generation_token_f1": 0.5211009174311927, "prompt_id": "39b34b632ecf238e", "reading_canonical_log_prob": -403.96282958984375, "reading_distances": [27, 21, 17, 23], "reading_log_probs": [-407.70623779296875, -408.6581115722656, -407.936767578125, -407.857177734375], "reading_logprob_std": 0.4232963159594323, "tokenization_path_logprob_std": 45.62692338098748, "trajectory_balance_rmse": 6094.75143607652} {"canonical_length": 144, "canonical_nll": 2.0807745191786022, "generation_rouge_l_f1": 0.163855421686747, "generation_token_f1": 0.35662650602409646, "prompt_id": "76cab7f49b66aca5", "reading_canonical_log_prob": -299.63153076171875, "reading_distances": [17, 14, 15, 11], "reading_log_probs": [-298.8077697753906, -299.66729736328125, -298.3683166503906, -298.50360107421875], "reading_logprob_std": 0.583400030222267, "tokenization_path_logprob_std": 99.50011741499067, "trajectory_balance_rmse": 2640.0852809086764} {"canonical_length": 40, "canonical_nll": 1.4617931365966796, "generation_rouge_l_f1": 0.21052631578947367, "generation_token_f1": 0.21052631578947367, "prompt_id": "b6ca714d2fc1e411", "reading_canonical_log_prob": -58.47172546386719, "reading_distances": [14, 16, 13, 18], "reading_log_probs": [-62.701332092285156, -65.2686538696289, -63.02703094482422, -64.75572204589844], "reading_logprob_std": 1.2647161818702808, "tokenization_path_logprob_std": 75.52785351803735, "trajectory_balance_rmse": 799.3299456992418} {"canonical_length": 39, "canonical_nll": 1.4300419734074519, "generation_rouge_l_f1": 0.3655913978494624, "generation_token_f1": 0.4731182795698924, "prompt_id": "56f7c951f93e4273", "reading_canonical_log_prob": -55.771636962890625, "reading_distances": [34, 27, 29, 26], "reading_log_probs": [-61.858909606933594, -60.480655670166016, -61.16297912597656, -62.96022415161133], "reading_logprob_std": 1.0583285449847442, "tokenization_path_logprob_std": 102.17148797635157, "trajectory_balance_rmse": 774.5533199228717} {"canonical_length": 17, "canonical_nll": 0.5828750273760628, "generation_rouge_l_f1": 0.20634920634920637, "generation_token_f1": 0.20634920634920637, "prompt_id": "3f9be20b25197a77", "reading_canonical_log_prob": -9.908875465393066, "reading_distances": [19, 14, 11, 10], "reading_log_probs": [-19.368385314941406, -13.57264518737793, -15.51441764831543, -17.280391693115234], "reading_logprob_std": 2.473861969576513, "tokenization_path_logprob_std": 60.76706600800636, "trajectory_balance_rmse": 250.9906367218362} {"canonical_length": 231, "canonical_nll": 1.3768030472132036, "generation_rouge_l_f1": 0.23236514522821577, "generation_token_f1": 0.44813278008298757, "prompt_id": "053c343603667759", "reading_canonical_log_prob": -318.04150390625, "reading_distances": [22, 20, 18, 20], "reading_log_probs": [-324.0564270019531, -320.5529479980469, -322.15966796875, -323.1474914550781], "reading_logprob_std": 1.4996450647859954, "tokenization_path_logprob_std": 136.8133596649161, "trajectory_balance_rmse": 4868.835954032723} {"canonical_length": 53, "canonical_nll": 1.3520404167895048, "generation_rouge_l_f1": 0.44642857142857145, "generation_token_f1": 0.5178571428571429, "prompt_id": "e23c57e239d810b9", "reading_canonical_log_prob": -71.65814208984375, "reading_distances": [29, 21, 23, 29], "reading_log_probs": [-77.05313873291016, -77.31550598144531, -78.3550796508789, -76.10090637207031], "reading_logprob_std": 0.9268178530229145, "tokenization_path_logprob_std": 92.9912285628782, "trajectory_balance_rmse": 972.7930513681085} {"canonical_length": 66, "canonical_nll": 1.1749191284179688, "generation_rouge_l_f1": 0.1512027491408935, "generation_token_f1": 0.24054982817869416, "prompt_id": "e9b590e25ca62203", "reading_canonical_log_prob": -77.54466247558594, "reading_distances": [21, 22, 30, 20], "reading_log_probs": [-80.10789489746094, -79.5748519897461, -78.5836181640625, -80.96514892578125], "reading_logprob_std": 0.9970622517784048, "tokenization_path_logprob_std": 152.3224681832855, "trajectory_balance_rmse": 1377.474661232484} {"canonical_length": 16, "canonical_nll": 1.5168668031692505, "generation_rouge_l_f1": 0.31111111111111117, "generation_token_f1": 0.4, "prompt_id": "d046e49475ca84ec", "reading_canonical_log_prob": -24.269868850708008, "reading_distances": [37, 44, 42, 37], "reading_log_probs": [-30.042701721191406, -26.605274200439453, -24.604759216308594, -29.217483520507812], "reading_logprob_std": 2.4861455268096537, "tokenization_path_logprob_std": 99.59867824745251, "trajectory_balance_rmse": 468.59114552932436} {"canonical_length": 56, "canonical_nll": 0.98408385685512, "generation_rouge_l_f1": 0.505050505050505, "generation_token_f1": 0.6060606060606061, "prompt_id": "750881d0bf75959b", "reading_canonical_log_prob": -55.10869598388672, "reading_distances": [48, 44, 52, 48], "reading_log_probs": [-59.052391052246094, -58.01084899902344, -58.06509780883789, -55.5865478515625], "reading_logprob_std": 1.4746472958669359, "tokenization_path_logprob_std": 71.2242049595935, "trajectory_balance_rmse": 1304.8994248860788} {"canonical_length": 2, "canonical_nll": 0.19716720283031464, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "73f8da9f02677521", "reading_canonical_log_prob": -0.3943344056606293, "reading_distances": [18, 14, 15, 17], "reading_log_probs": [-0.8789637088775635, -0.8756914734840393, -1.008843183517456, -0.9755443334579468], "reading_logprob_std": 0.0677101589168984, "tokenization_path_logprob_std": 9.309850920411277, "trajectory_balance_rmse": 48.70583821800508} {"canonical_length": 156, "canonical_nll": 2.29576658591246, "generation_rouge_l_f1": 0.27586206896551724, "generation_token_f1": 0.3931034482758621, "prompt_id": "9c25897ff633e0cd", "reading_canonical_log_prob": -358.13958740234375, "reading_distances": [29, 34, 28, 35], "reading_log_probs": [-368.16448974609375, -364.379150390625, -363.348876953125, -369.03363037109375], "reading_logprob_std": 2.788613674414823, "tokenization_path_logprob_std": 177.638824306618, "trajectory_balance_rmse": 2665.064734951152} {"canonical_length": 263, "canonical_nll": 1.1723449474958412, "generation_rouge_l_f1": 0.27599243856332706, "generation_token_f1": 0.47258979206049156, "prompt_id": "5d0b3ece32e43007", "reading_canonical_log_prob": -308.32672119140625, "reading_distances": [13, 21, 23, 25], "reading_log_probs": [-319.28765869140625, -317.24932861328125, -316.9119873046875, -317.6079406738281], "reading_logprob_std": 1.0546251120702974, "tokenization_path_logprob_std": 140.08733652765167, "trajectory_balance_rmse": 6321.626014734788} {"canonical_length": 21, "canonical_nll": 0.6431457882835752, "generation_rouge_l_f1": 0.5625000000000001, "generation_token_f1": 0.5625000000000001, "prompt_id": "16cea9158da9fd5a", "reading_canonical_log_prob": -13.506061553955078, "reading_distances": [31, 29, 34, 38], "reading_log_probs": [-18.018306732177734, -18.750370025634766, -20.21039390563965, -24.370159149169922], "reading_logprob_std": 2.838786982342457, "tokenization_path_logprob_std": 28.166029213278232, "trajectory_balance_rmse": 449.30478084081085} {"canonical_length": 18, "canonical_nll": 1.106355455186632, "generation_rouge_l_f1": 0.3478260869565218, "generation_token_f1": 0.3478260869565218, "prompt_id": "d643de98c824ec5b", "reading_canonical_log_prob": -19.914398193359375, "reading_distances": [27, 23, 18, 22], "reading_log_probs": [-24.317880630493164, -22.292861938476562, -20.977890014648438, -23.743274688720703], "reading_logprob_std": 1.501849157632807, "tokenization_path_logprob_std": 61.078153683954085, "trajectory_balance_rmse": 317.9779229550838} {"canonical_length": 97, "canonical_nll": 1.8102777225455058, "generation_rouge_l_f1": 0.25125628140703515, "generation_token_f1": 0.4321608040201005, "prompt_id": "cdeaf78291feca5e", "reading_canonical_log_prob": -175.59693908691406, "reading_distances": [37, 29, 32, 39], "reading_log_probs": [-198.122314453125, -199.85211181640625, -202.46615600585938, -192.10971069335938], "reading_logprob_std": 4.397409502675019, "tokenization_path_logprob_std": 38.91870179035253, "trajectory_balance_rmse": 1879.7238789070625} {"canonical_length": 308, "canonical_nll": 1.760803817154525, "generation_rouge_l_f1": 0.22437137330754353, "generation_token_f1": 0.3868471953578337, "prompt_id": "26a8450ba4613a47", "reading_canonical_log_prob": -542.3275756835938, "reading_distances": [22, 17, 20, 17], "reading_log_probs": [-557.72314453125, -555.6915283203125, -553.1800537109375, -552.9551391601562], "reading_logprob_std": 2.261027250931157, "tokenization_path_logprob_std": 150.66300863441873, "trajectory_balance_rmse": 5294.079528978215} {"canonical_length": 104, "canonical_nll": 0.6460158274723933, "generation_rouge_l_f1": 0.5416666666666666, "generation_token_f1": 0.6354166666666667, "prompt_id": "c15330337653f483", "reading_canonical_log_prob": -67.1856460571289, "reading_distances": [281, 304, 302, 287], "reading_log_probs": [-81.28590393066406, -87.8902587890625, -89.2505111694336, -83.171875], "reading_logprob_std": 3.7823340262927436, "tokenization_path_logprob_std": 177.49348347550333, "trajectory_balance_rmse": 1814.9690718690385} {"canonical_length": 11, "canonical_nll": 1.687682411887429, "generation_rouge_l_f1": 0.5714285714285714, "generation_token_f1": 0.5714285714285714, "prompt_id": "8609ed9bb93a15ed", "reading_canonical_log_prob": -18.56450653076172, "reading_distances": [17, 14, 15, 19], "reading_log_probs": [-15.931892395019531, -16.545330047607422, -16.141399383544922, -19.079099655151367], "reading_logprob_std": 1.4588335454899122, "tokenization_path_logprob_std": 34.923737412722076, "trajectory_balance_rmse": 147.61448921620087} {"canonical_length": 367, "canonical_nll": 1.3853635034379257, "generation_rouge_l_f1": 0.23137876386687797, "generation_token_f1": 0.46275752773375595, "prompt_id": "4cb8d591dfd52931", "reading_canonical_log_prob": -508.42840576171875, "reading_distances": [22, 25, 30, 31], "reading_log_probs": [-512.48486328125, -511.70599365234375, -512.3101806640625, -512.250244140625], "reading_logprob_std": 0.3362837322229343, "tokenization_path_logprob_std": 115.2445928376765, "trajectory_balance_rmse": 7873.6346540204195} {"canonical_length": 277, "canonical_nll": 0.41103464945989393, "generation_rouge_l_f1": 0.41262135922330095, "generation_token_f1": 0.6359223300970873, "prompt_id": "7e37b68aba5b5c9c", "reading_canonical_log_prob": -113.85659790039062, "reading_distances": [35, 28, 36, 36], "reading_log_probs": [-119.9264144897461, -120.32003021240234, -121.989990234375, -122.55003356933594], "reading_logprob_std": 1.2705636954283648, "tokenization_path_logprob_std": 94.72490079165539, "trajectory_balance_rmse": 4261.584319354416} {"canonical_length": 87, "canonical_nll": 1.3329918521574173, "generation_rouge_l_f1": 0.20168067226890754, "generation_token_f1": 0.3473389355742297, "prompt_id": "fef695db887a3a23", "reading_canonical_log_prob": -115.97029113769531, "reading_distances": [17, 19, 19, 15], "reading_log_probs": [-134.9976806640625, -126.7298355102539, -131.94908142089844, -125.72817993164062], "reading_logprob_std": 4.382905445916586, "tokenization_path_logprob_std": 60.165368776281156, "trajectory_balance_rmse": 1818.9834033424363} {"canonical_length": 174, "canonical_nll": 0.6388787017471489, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.6918604651162791, "prompt_id": "471b9b9cabf2decb", "reading_canonical_log_prob": -111.1648941040039, "reading_distances": [22, 24, 22, 22], "reading_log_probs": [-109.65475463867188, -108.59583282470703, -108.053955078125, -109.63975524902344], "reading_logprob_std": 0.7948934585360306, "tokenization_path_logprob_std": 71.08442703474692, "trajectory_balance_rmse": 2554.474291151385} {"canonical_length": 10, "canonical_nll": 1.0494939804077148, "generation_rouge_l_f1": 0.05755395683453238, "generation_token_f1": 0.05755395683453238, "prompt_id": "71f95c366e84f3b3", "reading_canonical_log_prob": -10.494939804077148, "reading_distances": [29, 41, 33, 31], "reading_log_probs": [-7.523638725280762, -7.567723274230957, -7.800560474395752, -8.583145141601562], "reading_logprob_std": 0.4915041912905907, "tokenization_path_logprob_std": 32.01062547932589, "trajectory_balance_rmse": 267.45429509065224} {"canonical_length": 7, "canonical_nll": 0.6036975043160575, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "78998e3018151a20", "reading_canonical_log_prob": -4.225882530212402, "reading_distances": [15, 19, 16, 24], "reading_log_probs": [-4.332919597625732, -5.4798054695129395, -4.658535957336426, -4.7054643630981445], "reading_logprob_std": 0.48618008110063937, "tokenization_path_logprob_std": 21.821338377643745, "trajectory_balance_rmse": 113.95937952935124} {"canonical_length": 179, "canonical_nll": 1.425860697996683, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "3025be82a097d5b7", "reading_canonical_log_prob": -255.22906494140625, "reading_distances": [24, 21, 30, 26], "reading_log_probs": [-260.79779052734375, -252.75115966796875, -261.7080993652344, -260.82647705078125], "reading_logprob_std": 4.201116228287459, "tokenization_path_logprob_std": 109.4813267901535, "trajectory_balance_rmse": 3581.3352636588916} {"canonical_length": 259, "canonical_nll": 1.8424319705447636, "generation_rouge_l_f1": 0.11650485436893204, "generation_token_f1": 0.23300970873786409, "prompt_id": "317ccf6fb834b830", "reading_canonical_log_prob": -477.18988037109375, "reading_distances": [20, 24, 29, 20], "reading_log_probs": [-479.28509521484375, -480.49566650390625, -482.4497375488281, -479.6882629394531], "reading_logprob_std": 1.4065034762616038, "tokenization_path_logprob_std": 198.20986958772698, "trajectory_balance_rmse": 5095.95975966169} {"canonical_length": 35, "canonical_nll": 1.4838160923549106, "generation_rouge_l_f1": 0.4166666666666667, "generation_token_f1": 0.4166666666666667, "prompt_id": "43805108665c2c05", "reading_canonical_log_prob": -51.933563232421875, "reading_distances": [31, 17, 23, 25], "reading_log_probs": [-49.569427490234375, -50.424110412597656, -51.43107986450195, -48.3175048828125], "reading_logprob_std": 1.3200273411676573, "tokenization_path_logprob_std": 50.987603006858734, "trajectory_balance_rmse": 754.9544509821366} {"canonical_length": 118, "canonical_nll": 1.5151933573060117, "generation_rouge_l_f1": 0.2912621359223301, "generation_token_f1": 0.5048543689320388, "prompt_id": "5056b26f2553e0fc", "reading_canonical_log_prob": -178.79281616210938, "reading_distances": [26, 26, 33, 14], "reading_log_probs": [-180.54196166992188, -181.77914428710938, -181.48934936523438, -179.10765075683594], "reading_logprob_std": 1.2034242501151413, "tokenization_path_logprob_std": 130.67721321431586, "trajectory_balance_rmse": 2561.7201423143606} {"canonical_length": 372, "canonical_nll": 0.9416867327946489, "generation_rouge_l_f1": 0.3423137876386688, "generation_token_f1": 0.5895404120443741, "prompt_id": "1ea4408820560cde", "reading_canonical_log_prob": -350.3074645996094, "reading_distances": [12, 13, 13, 14], "reading_log_probs": [-354.4993896484375, -353.010009765625, -352.67242431640625, -353.5459899902344], "reading_logprob_std": 0.7973592556693937, "tokenization_path_logprob_std": 377.2071252159126, "trajectory_balance_rmse": 6426.405870432818} {"canonical_length": 9, "canonical_nll": 1.0395607418484158, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.2857142857142857, "prompt_id": "2b79f5bffeeb5daf", "reading_canonical_log_prob": -9.356046676635742, "reading_distances": [35, 37, 32, 32], "reading_log_probs": [-11.447834968566895, -11.622906684875488, -10.516850471496582, -9.874281883239746], "reading_logprob_std": 0.8199286039367083, "tokenization_path_logprob_std": 50.72491626454617, "trajectory_balance_rmse": 177.68605281740082} {"canonical_length": 36, "canonical_nll": 0.8412617047627767, "generation_rouge_l_f1": 0.1714285714285714, "generation_token_f1": 0.2285714285714286, "prompt_id": "91748c9b4543a8c6", "reading_canonical_log_prob": -30.28542137145996, "reading_distances": [10, 13, 13, 11], "reading_log_probs": [-34.144962310791016, -38.81221389770508, -33.162513732910156, -36.51054382324219], "reading_logprob_std": 2.529336557368902, "tokenization_path_logprob_std": 93.50595516075926, "trajectory_balance_rmse": 479.7378155156978} {"canonical_length": 187, "canonical_nll": 1.185685346470797, "generation_rouge_l_f1": 0.2407407407407407, "generation_token_f1": 0.5046296296296298, "prompt_id": "9a3c9e43bea33f6c", "reading_canonical_log_prob": -221.72315979003906, "reading_distances": [21, 17, 25, 16], "reading_log_probs": [-223.7670135498047, -228.18475341796875, -229.55506896972656, -226.7507781982422], "reading_logprob_std": 2.4785573582450837, "tokenization_path_logprob_std": 107.90805795277934, "trajectory_balance_rmse": 4275.311012146912} {"canonical_length": 376, "canonical_nll": 1.2833286041909076, "generation_rouge_l_f1": 0.19488817891373805, "generation_token_f1": 0.43450479233226835, "prompt_id": "7473e8ab94e776ba", "reading_canonical_log_prob": -482.53155517578125, "reading_distances": [22, 19, 22, 14], "reading_log_probs": [-496.0376281738281, -489.21893310546875, -495.9058837890625, -488.71728515625], "reading_logprob_std": 4.049097546847472, "tokenization_path_logprob_std": 198.1671402473448, "trajectory_balance_rmse": 7669.301932233483} {"canonical_length": 87, "canonical_nll": 0.4000641395305765, "generation_rouge_l_f1": 0.24605678233438485, "generation_token_f1": 0.4037854889589905, "prompt_id": "81563152acabd240", "reading_canonical_log_prob": -34.805580139160156, "reading_distances": [40, 35, 40, 34], "reading_log_probs": [-41.27854919433594, -42.13875961303711, -43.89518737792969, -42.47654724121094], "reading_logprob_std": 1.0890988431244961, "tokenization_path_logprob_std": 84.96817731163763, "trajectory_balance_rmse": 1016.911477963771} {"canonical_length": 19, "canonical_nll": 2.3522551687140214, "generation_rouge_l_f1": 0.13793103448275862, "generation_token_f1": 0.13793103448275862, "prompt_id": "49b74500a1071228", "reading_canonical_log_prob": -44.692848205566406, "reading_distances": [22, 20, 18, 28], "reading_log_probs": [-42.10021209716797, -42.62196350097656, -45.94844436645508, -44.96874237060547], "reading_logprob_std": 1.8448632304182109, "tokenization_path_logprob_std": 71.36913578910372, "trajectory_balance_rmse": 482.66585090600626} {"canonical_length": 83, "canonical_nll": 1.1872129325407097, "generation_rouge_l_f1": 0.19753086419753088, "generation_token_f1": 0.25925925925925924, "prompt_id": "c1a613d92b944071", "reading_canonical_log_prob": -98.5386734008789, "reading_distances": [36, 36, 31, 34], "reading_log_probs": [-107.92977905273438, -104.84716033935547, -108.32515716552734, -110.00688171386719], "reading_logprob_std": 2.1509804255071328, "tokenization_path_logprob_std": 77.86828647321659, "trajectory_balance_rmse": 1676.5140604094975} {"canonical_length": 190, "canonical_nll": 1.273883699115954, "generation_rouge_l_f1": 0.2735042735042735, "generation_token_f1": 0.4957264957264957, "prompt_id": "00050e95a0ce36d0", "reading_canonical_log_prob": -242.03790283203125, "reading_distances": [34, 29, 22, 38], "reading_log_probs": [-254.1286163330078, -249.73419189453125, -247.96551513671875, -255.1200408935547], "reading_logprob_std": 3.4351192468332648, "tokenization_path_logprob_std": 149.62173673284136, "trajectory_balance_rmse": 3842.365939213206} {"canonical_length": 309, "canonical_nll": 1.3852987443744944, "generation_rouge_l_f1": 0.23859649122807014, "generation_token_f1": 0.4210526315789474, "prompt_id": "3e9a72eac16389eb", "reading_canonical_log_prob": -428.05731201171875, "reading_distances": [18, 19, 20, 22], "reading_log_probs": [-429.3834533691406, -430.3614501953125, -429.2032470703125, -429.7364196777344], "reading_logprob_std": 0.510710940186217, "tokenization_path_logprob_std": 382.64465827677185, "trajectory_balance_rmse": 6811.285052684681} {"canonical_length": 335, "canonical_nll": 1.1352140967525652, "generation_rouge_l_f1": 0.25089605734767023, "generation_token_f1": 0.4802867383512545, "prompt_id": "bef02b456531b6b6", "reading_canonical_log_prob": -380.2967224121094, "reading_distances": [18, 20, 20, 21], "reading_log_probs": [-382.2574462890625, -380.2530212402344, -380.6251220703125, -382.58282470703125], "reading_logprob_std": 1.1614328823277733, "tokenization_path_logprob_std": 206.22665676198557, "trajectory_balance_rmse": 5968.373044793859} {"canonical_length": 110, "canonical_nll": 1.514206210049716, "generation_rouge_l_f1": 0.18282548476454294, "generation_token_f1": 0.3545706371191135, "prompt_id": "9d2a25bad82737d8", "reading_canonical_log_prob": -166.56268310546875, "reading_distances": [16, 19, 17, 18], "reading_log_probs": [-175.151611328125, -175.9156951904297, -177.4982147216797, -176.85528564453125], "reading_logprob_std": 1.0325306249747337, "tokenization_path_logprob_std": 147.62815088269568, "trajectory_balance_rmse": 2827.131130369535} {"canonical_length": 10, "canonical_nll": 1.8601825714111329, "generation_rouge_l_f1": 0.19999999999999998, "generation_token_f1": 0.2285714285714286, "prompt_id": "77bd9a6bbbf6a3af", "reading_canonical_log_prob": -18.601825714111328, "reading_distances": [21, 22, 17, 19], "reading_log_probs": [-21.318693161010742, -19.334226608276367, -18.91596031188965, -20.968143463134766], "reading_logprob_std": 1.1863882305617655, "tokenization_path_logprob_std": 43.9839801056995, "trajectory_balance_rmse": 160.9366902152012} {"canonical_length": 18, "canonical_nll": 0.7763183381822374, "generation_rouge_l_f1": 0.3684210526315789, "generation_token_f1": 0.3684210526315789, "prompt_id": "45f03bee685e0c93", "reading_canonical_log_prob": -13.973730087280273, "reading_distances": [23, 23, 27, 23], "reading_log_probs": [-12.04904556274414, -15.63500690460205, -10.694262504577637, -11.582584381103516], "reading_logprob_std": 2.170528760571512, "tokenization_path_logprob_std": 59.15249499819606, "trajectory_balance_rmse": 391.900407258044} {"canonical_length": 300, "canonical_nll": 1.0329307047526042, "generation_rouge_l_f1": 0.20593368237347298, "generation_token_f1": 0.41535776614310643, "prompt_id": "1c2335f122ec1aa9", "reading_canonical_log_prob": -309.87921142578125, "reading_distances": [29, 22, 25, 22], "reading_log_probs": [-321.8633117675781, -319.31097412109375, -322.437255859375, -318.0854187011719], "reading_logprob_std": 2.068220361190157, "tokenization_path_logprob_std": 138.22531729301727, "trajectory_balance_rmse": 5069.299767022016} {"canonical_length": 24, "canonical_nll": 0.7165329456329346, "generation_rouge_l_f1": 0.4999999999999999, "generation_token_f1": 0.4999999999999999, "prompt_id": "0b6723f81e5854b3", "reading_canonical_log_prob": -17.19679069519043, "reading_distances": [20, 28, 25, 33], "reading_log_probs": [-24.94295883178711, -23.243839263916016, -22.832561492919922, -22.293954849243164], "reading_logprob_std": 1.1445353225989607, "tokenization_path_logprob_std": 84.73916297290859, "trajectory_balance_rmse": 568.0980905359105} {"canonical_length": 43, "canonical_nll": 0.1720623859139376, "generation_rouge_l_f1": 0.6666666666666667, "generation_token_f1": 0.7246376811594204, "prompt_id": "6d26cacb241c8ba0", "reading_canonical_log_prob": -7.398682594299316, "reading_distances": [32, 37, 42, 38], "reading_log_probs": [-9.798201560974121, -8.368529319763184, -10.801013946533203, -6.976495742797852], "reading_logprob_std": 1.670660985753445, "tokenization_path_logprob_std": 62.78407328478267, "trajectory_balance_rmse": 901.2280946353231} {"canonical_length": 10, "canonical_nll": 1.4487996101379395, "generation_rouge_l_f1": 0.10404624277456648, "generation_token_f1": 0.11560693641618495, "prompt_id": "d7683fe01587262b", "reading_canonical_log_prob": -14.487996101379395, "reading_distances": [40, 33, 45, 37], "reading_log_probs": [-19.05337142944336, -18.469425201416016, -20.479217529296875, -21.1487979888916], "reading_logprob_std": 1.2393373394372387, "tokenization_path_logprob_std": 23.634231817992568, "trajectory_balance_rmse": 215.27903307606562} {"canonical_length": 71, "canonical_nll": 0.7622511957732725, "generation_rouge_l_f1": 0.5671641791044777, "generation_token_f1": 0.6417910447761195, "prompt_id": "e455ade71af9293b", "reading_canonical_log_prob": -54.119834899902344, "reading_distances": [52, 34, 39, 53], "reading_log_probs": [-51.27135467529297, -52.0094108581543, -53.460479736328125, -50.98249435424805], "reading_logprob_std": 1.107587682608335, "tokenization_path_logprob_std": 124.6828759373478, "trajectory_balance_rmse": 1608.9130611015357} {"canonical_length": 334, "canonical_nll": 1.8465130286302396, "generation_rouge_l_f1": 0.19064124783362216, "generation_token_f1": 0.3466204506065858, "prompt_id": "83deb2f6c2a5fe75", "reading_canonical_log_prob": -616.7353515625, "reading_distances": [43, 35, 31, 37], "reading_log_probs": [-629.404296875, -618.5322265625, -619.6880493164062, -618.5322875976562], "reading_logprob_std": 5.271619585250597, "tokenization_path_logprob_std": 65.24173718053684, "trajectory_balance_rmse": 6113.043393813702} {"canonical_length": 251, "canonical_nll": 1.307596275056026, "generation_rouge_l_f1": 0.1509433962264151, "generation_token_f1": 0.2688679245283018, "prompt_id": "cc738e3b9180c6b5", "reading_canonical_log_prob": -328.2066650390625, "reading_distances": [11, 11, 12, 15], "reading_log_probs": [-346.5573425292969, -348.4682922363281, -324.6369934082031, -347.0729064941406], "reading_logprob_std": 11.393226576128166, "tokenization_path_logprob_std": 192.61432786431908, "trajectory_balance_rmse": 4511.085110052579} {"canonical_length": 15, "canonical_nll": 0.16384458541870117, "generation_rouge_l_f1": 0.6153846153846153, "generation_token_f1": 0.923076923076923, "prompt_id": "d829b51f56a6d0c5", "reading_canonical_log_prob": -2.4576687812805176, "reading_distances": [18, 16, 15, 19], "reading_log_probs": [-2.364928722381592, -2.7900102138519287, -3.0604965686798096, -1.991255760192871], "reading_logprob_std": 0.4706899048017651, "tokenization_path_logprob_std": 48.38973175111533, "trajectory_balance_rmse": 382.9666729223068} {"canonical_length": 158, "canonical_nll": 1.4765708054168314, "generation_rouge_l_f1": 0.21052631578947367, "generation_token_f1": 0.47058823529411764, "prompt_id": "f53fbba7ee710f29", "reading_canonical_log_prob": -233.29818725585938, "reading_distances": [22, 23, 22, 29], "reading_log_probs": [-254.536865234375, -245.90869140625, -245.07139587402344, -248.4102020263672], "reading_logprob_std": 4.278655762407804, "tokenization_path_logprob_std": 162.80279691829705, "trajectory_balance_rmse": 3898.416230687123} {"canonical_length": 268, "canonical_nll": 1.302620389568272, "generation_rouge_l_f1": 0.22399999999999998, "generation_token_f1": 0.47600000000000003, "prompt_id": "2ce5516156798d8a", "reading_canonical_log_prob": -349.1022644042969, "reading_distances": [31, 29, 26, 30], "reading_log_probs": [-361.0677490234375, -358.2259216308594, -357.37042236328125, -358.4622497558594], "reading_logprob_std": 1.5946433987242092, "tokenization_path_logprob_std": 437.12720647699894, "trajectory_balance_rmse": 5262.320733744826} {"canonical_length": 28, "canonical_nll": 0.26665360586983816, "generation_rouge_l_f1": 0.4740740740740741, "generation_token_f1": 0.4740740740740741, "prompt_id": "83deea31656f880e", "reading_canonical_log_prob": -7.466300964355469, "reading_distances": [33, 26, 28, 23], "reading_log_probs": [-12.738486289978027, -9.184443473815918, -12.530357360839844, -8.001565933227539], "reading_logprob_std": 2.38427562840858, "tokenization_path_logprob_std": 60.42512853474002, "trajectory_balance_rmse": 522.5778131036889} {"canonical_length": 16, "canonical_nll": 0.1774182915687561, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "2a011ac439c0c89f", "reading_canonical_log_prob": -2.8386926651000977, "reading_distances": [50, 50, 40, 47], "reading_log_probs": [-8.261852264404297, -3.847137451171875, -5.705896377563477, -4.886801242828369], "reading_logprob_std": 1.8845952076807606, "tokenization_path_logprob_std": 48.83337261084163, "trajectory_balance_rmse": 372.439352475725} {"canonical_length": 27, "canonical_nll": 1.1335885789659288, "generation_rouge_l_f1": 0.07999999999999999, "generation_token_f1": 0.07999999999999999, "prompt_id": "a62ae1af2ae45a72", "reading_canonical_log_prob": -30.606891632080078, "reading_distances": [27, 23, 29, 22], "reading_log_probs": [-31.209400177001953, -33.85502243041992, -36.90823745727539, -35.66803741455078], "reading_logprob_std": 2.4749200483676397, "tokenization_path_logprob_std": 39.596820073701856, "trajectory_balance_rmse": 407.27727696557116} {"canonical_length": 99, "canonical_nll": 1.6808559533321497, "generation_rouge_l_f1": 0.25308641975308643, "generation_token_f1": 0.3518518518518518, "prompt_id": "53aea8a2764798f8", "reading_canonical_log_prob": -166.4047393798828, "reading_distances": [34, 31, 28, 33], "reading_log_probs": [-178.39654541015625, -178.918212890625, -180.18527221679688, -177.23080444335938], "reading_logprob_std": 1.2251632196999014, "tokenization_path_logprob_std": 170.9194579641266, "trajectory_balance_rmse": 2091.220303871217} {"canonical_length": 159, "canonical_nll": 0.5048736956134532, "generation_rouge_l_f1": 0.3452768729641694, "generation_token_f1": 0.5732899022801303, "prompt_id": "7dff62aaf925a1c6", "reading_canonical_log_prob": -80.27491760253906, "reading_distances": [30, 32, 34, 32], "reading_log_probs": [-91.64088439941406, -92.42383575439453, -96.37913513183594, -91.86808776855469], "reading_logprob_std": 2.225205074842002, "tokenization_path_logprob_std": 84.95873853209903, "trajectory_balance_rmse": 2139.434119353328} {"canonical_length": 217, "canonical_nll": 1.3631974321356566, "generation_rouge_l_f1": 0.1888412017167382, "generation_token_f1": 0.351931330472103, "prompt_id": "5470eb105173aff8", "reading_canonical_log_prob": -295.8138427734375, "reading_distances": [25, 38, 33, 35], "reading_log_probs": [-316.7926330566406, -318.51263427734375, -306.7774963378906, -318.9991455078125], "reading_logprob_std": 5.740564206879463, "tokenization_path_logprob_std": 212.35698111081615, "trajectory_balance_rmse": 4545.016214905585} {"canonical_length": 15, "canonical_nll": 0.6839942932128906, "generation_rouge_l_f1": 0.5000000000000001, "generation_token_f1": 0.5000000000000001, "prompt_id": "861fa387c936a923", "reading_canonical_log_prob": -10.25991439819336, "reading_distances": [9, 9, 11, 10], "reading_log_probs": [-9.700505256652832, -13.213242530822754, -10.288997650146484, -10.727677345275879], "reading_logprob_std": 1.5454883597877112, "tokenization_path_logprob_std": 49.853285538652905, "trajectory_balance_rmse": 165.70578808094223} {"canonical_length": 11, "canonical_nll": 1.3218628276478162, "generation_rouge_l_f1": 0.09210526315789473, "generation_token_f1": 0.10526315789473685, "prompt_id": "6c2751afcc2de989", "reading_canonical_log_prob": -14.540491104125977, "reading_distances": [11, 9, 9, 14], "reading_log_probs": [-22.056886672973633, -20.174705505371094, -27.79510498046875, -22.26156234741211], "reading_logprob_std": 3.2857926635612356, "tokenization_path_logprob_std": 39.464483313944925, "trajectory_balance_rmse": 184.39601701421236} {"canonical_length": 69, "canonical_nll": 0.7856331424436708, "generation_rouge_l_f1": 0.3221476510067114, "generation_token_f1": 0.46979865771812085, "prompt_id": "2dc95060fe386398", "reading_canonical_log_prob": -54.20868682861328, "reading_distances": [21, 18, 24, 20], "reading_log_probs": [-53.574913024902344, -54.467594146728516, -53.546417236328125, -53.317970275878906], "reading_logprob_std": 0.5071243379691313, "tokenization_path_logprob_std": 109.32296349523274, "trajectory_balance_rmse": 1616.863684236489} {"canonical_length": 367, "canonical_nll": 1.3406854364462704, "generation_rouge_l_f1": 0.24749163879598662, "generation_token_f1": 0.42474916387959866, "prompt_id": "346df7037bb677e7", "reading_canonical_log_prob": -492.03155517578125, "reading_distances": [27, 25, 27, 28], "reading_log_probs": [-509.3713073730469, -511.2138366699219, -500.6146240234375, -508.8030700683594], "reading_logprob_std": 4.704634807761451, "tokenization_path_logprob_std": 249.70057445554124, "trajectory_balance_rmse": 6945.931407208274} {"canonical_length": 168, "canonical_nll": 1.7212495349702381, "generation_rouge_l_f1": 0.2293577981651376, "generation_token_f1": 0.48165137614678893, "prompt_id": "cccaf997f15db0c8", "reading_canonical_log_prob": -289.169921875, "reading_distances": [19, 12, 16, 12], "reading_log_probs": [-296.414306640625, -292.9389953613281, -292.2615966796875, -294.17462158203125], "reading_logprob_std": 1.8253831350689678, "tokenization_path_logprob_std": 120.77418867873244, "trajectory_balance_rmse": 3944.675931738369} {"canonical_length": 15, "canonical_nll": 0.40064481099446614, "generation_rouge_l_f1": 0.20183486238532108, "generation_token_f1": 0.23853211009174308, "prompt_id": "c7d4acf2c03f4f98", "reading_canonical_log_prob": -6.009672164916992, "reading_distances": [15, 14, 19, 20], "reading_log_probs": [-5.107651710510254, -5.655983924865723, -6.123502731323242, -6.02344274520874], "reading_logprob_std": 0.45961093627123606, "tokenization_path_logprob_std": 31.236764767257657, "trajectory_balance_rmse": 209.59984976496546} {"canonical_length": 31, "canonical_nll": 2.0639921619046118, "generation_rouge_l_f1": 0.22727272727272727, "generation_token_f1": 0.2727272727272727, "prompt_id": "f4a99dd0e66e3bbe", "reading_canonical_log_prob": -63.98375701904297, "reading_distances": [22, 18, 23, 19], "reading_log_probs": [-65.27302551269531, -65.51756286621094, -67.11817932128906, -67.08475494384766], "reading_logprob_std": 0.9901993006384096, "tokenization_path_logprob_std": 56.70387050659657, "trajectory_balance_rmse": 730.0524534288852} {"canonical_length": 51, "canonical_nll": 1.2621827967026655, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "b701439d534ac688", "reading_canonical_log_prob": -64.37132263183594, "reading_distances": [21, 24, 23, 21], "reading_log_probs": [-68.86381530761719, -67.11294555664062, -63.780208587646484, -65.86580657958984], "reading_logprob_std": 2.1390977229142245, "tokenization_path_logprob_std": 44.44122829319478, "trajectory_balance_rmse": 1130.7679786820022} {"canonical_length": 448, "canonical_nll": 1.3626689910888672, "generation_rouge_l_f1": 0.19694868238557559, "generation_token_f1": 0.37447988904299584, "prompt_id": "7fdb8e03f79c4b21", "reading_canonical_log_prob": -610.4757080078125, "reading_distances": [15, 14, 13, 13], "reading_log_probs": [-618.82470703125, -619.4595947265625, -613.8372192382812, -617.765380859375], "reading_logprob_std": 2.5217753218966097, "tokenization_path_logprob_std": 336.3627771163364, "trajectory_balance_rmse": 7427.191099837249} {"canonical_length": 392, "canonical_nll": 1.3291928038305165, "generation_rouge_l_f1": 0.2101910828025478, "generation_token_f1": 0.43630573248407645, "prompt_id": "79079ac51debbea1", "reading_canonical_log_prob": -521.0435791015625, "reading_distances": [26, 23, 23, 25], "reading_log_probs": [-541.124267578125, -537.5511474609375, -539.0130615234375, -541.33349609375], "reading_logprob_std": 1.804991353864911, "tokenization_path_logprob_std": 310.09027920664, "trajectory_balance_rmse": 7290.041861100535} {"canonical_length": 5, "canonical_nll": 2.573864555358887, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "e0c4de6979cf5d26", "reading_canonical_log_prob": -12.869322776794434, "reading_distances": [19, 19, 26, 24], "reading_log_probs": [-11.36561107635498, -11.751190185546875, -11.315932273864746, -12.055913925170898], "reading_logprob_std": 0.34851304463720917, "tokenization_path_logprob_std": 13.679033551213472, "trajectory_balance_rmse": 117.38871030297085} {"canonical_length": 199, "canonical_nll": 1.5741925263524654, "generation_rouge_l_f1": 0.19764705882352943, "generation_token_f1": 0.34823529411764703, "prompt_id": "a5453cc0769ffb94", "reading_canonical_log_prob": -313.2643127441406, "reading_distances": [22, 13, 22, 20], "reading_log_probs": [-360.82171630859375, -318.97186279296875, -347.84063720703125, -352.0797424316406], "reading_logprob_std": 18.12881822915692, "tokenization_path_logprob_std": 144.1766316635996, "trajectory_balance_rmse": 3214.831895638875} {"canonical_length": 16, "canonical_nll": 1.412653923034668, "generation_rouge_l_f1": 0.16666666666666669, "generation_token_f1": 0.25, "prompt_id": "c1820a57ac4bcbc9", "reading_canonical_log_prob": -22.602462768554688, "reading_distances": [17, 13, 13, 14], "reading_log_probs": [-23.620315551757812, -29.650894165039062, -27.580650329589844, -23.786170959472656], "reading_logprob_std": 2.960273065756334, "tokenization_path_logprob_std": 50.56574838092979, "trajectory_balance_rmse": 391.55078259512356} {"canonical_length": 87, "canonical_nll": 0.4301348280632633, "generation_rouge_l_f1": 0.44859813084112143, "generation_token_f1": 0.5046728971962616, "prompt_id": "326b224fa8e4dc3a", "reading_canonical_log_prob": -37.421730041503906, "reading_distances": [25, 28, 21, 30], "reading_log_probs": [-42.22368621826172, -45.970462799072266, -39.46440887451172, -41.736412048339844], "reading_logprob_std": 2.697333624369753, "tokenization_path_logprob_std": 136.96217357354865, "trajectory_balance_rmse": 1065.8648273847925} {"canonical_length": 210, "canonical_nll": 0.8883130754743304, "generation_rouge_l_f1": 0.23370786516853934, "generation_token_f1": 0.45842696629213486, "prompt_id": "a8f7e780841cbd23", "reading_canonical_log_prob": -186.54574584960938, "reading_distances": [15, 21, 12, 16], "reading_log_probs": [-194.53607177734375, -192.35069274902344, -194.77085876464844, -196.3507843017578], "reading_logprob_std": 1.6451516811869897, "tokenization_path_logprob_std": 213.91607094695567, "trajectory_balance_rmse": 2842.548308817153} {"canonical_length": 221, "canonical_nll": 1.5136501950915582, "generation_rouge_l_f1": 0.2595744680851064, "generation_token_f1": 0.5021276595744681, "prompt_id": "675093cc227e317b", "reading_canonical_log_prob": -334.5166931152344, "reading_distances": [15, 17, 14, 19], "reading_log_probs": [-336.02142333984375, -338.62548828125, -336.1220703125, -338.5320739746094], "reading_logprob_std": 1.4485221219212003, "tokenization_path_logprob_std": 85.28060926485799, "trajectory_balance_rmse": 4755.406441838851} {"canonical_length": 327, "canonical_nll": 1.137734789367116, "generation_rouge_l_f1": 0.2303839732888147, "generation_token_f1": 0.4540901502504174, "prompt_id": "df931a7b15ed70cf", "reading_canonical_log_prob": -372.0392761230469, "reading_distances": [34, 35, 42, 34], "reading_log_probs": [-370.97344970703125, -373.193115234375, -373.8367919921875, -373.83154296875], "reading_logprob_std": 1.3575799080452686, "tokenization_path_logprob_std": 253.17697069994418, "trajectory_balance_rmse": 7350.117652386556} {"canonical_length": 267, "canonical_nll": 1.3798293210147472, "generation_rouge_l_f1": 0.23619047619047623, "generation_token_f1": 0.4799999999999999, "prompt_id": "64f98636b7dde269", "reading_canonical_log_prob": -368.4144287109375, "reading_distances": [27, 26, 32, 24], "reading_log_probs": [-368.8928527832031, -371.7281188964844, -371.50006103515625, -374.26275634765625], "reading_logprob_std": 2.1943301157074093, "tokenization_path_logprob_std": 24.893913569045885, "trajectory_balance_rmse": 4642.303976499739} {"canonical_length": 110, "canonical_nll": 1.2080194646661933, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.4183673469387755, "prompt_id": "3485bbb40ac08228", "reading_canonical_log_prob": -132.88214111328125, "reading_distances": [16, 15, 10, 14], "reading_log_probs": [-135.4797821044922, -146.25364685058594, -138.8005828857422, -145.4296875], "reading_logprob_std": 5.214375931320965, "tokenization_path_logprob_std": 63.125512552029605, "trajectory_balance_rmse": 2469.054749884204} {"canonical_length": 20, "canonical_nll": 1.274479579925537, "generation_rouge_l_f1": 0.0784313725490196, "generation_token_f1": 0.08627450980392158, "prompt_id": "5de8b78dfeffe43c", "reading_canonical_log_prob": -25.489591598510742, "reading_distances": [13, 19, 12, 11], "reading_log_probs": [-27.615354537963867, -26.40329933166504, -27.582477569580078, -31.37088394165039], "reading_logprob_std": 2.1601223120115005, "tokenization_path_logprob_std": 39.807666896590476, "trajectory_balance_rmse": 290.3761881005666} {"canonical_length": 14, "canonical_nll": 3.231402260916574, "generation_rouge_l_f1": 0.06486486486486487, "generation_token_f1": 0.07567567567567567, "prompt_id": "75d59497e8843334", "reading_canonical_log_prob": -45.23963165283203, "reading_distances": [7, 7, 11, 10], "reading_log_probs": [-46.68138885498047, -45.57091522216797, -49.908626556396484, -43.16383361816406], "reading_logprob_std": 2.800646100481782, "tokenization_path_logprob_std": 38.428865128015964, "trajectory_balance_rmse": 240.2111470836308} {"canonical_length": 53, "canonical_nll": 2.2796792084316038, "generation_rouge_l_f1": 0.2262773722627737, "generation_token_f1": 0.2554744525547445, "prompt_id": "9fe6000f50208baa", "reading_canonical_log_prob": -120.822998046875, "reading_distances": [24, 22, 20, 23], "reading_log_probs": [-123.28175354003906, -123.96531677246094, -122.89289093017578, -123.171875], "reading_logprob_std": 0.4553410132648301, "tokenization_path_logprob_std": 63.19270784088064, "trajectory_balance_rmse": 1305.6396748190966} {"canonical_length": 13, "canonical_nll": 0.29115786919227016, "generation_rouge_l_f1": 0.7857142857142857, "generation_token_f1": 0.7857142857142857, "prompt_id": "2321fcc6e5fd7b96", "reading_canonical_log_prob": -3.7850522994995117, "reading_distances": [31, 35, 36, 45], "reading_log_probs": [-3.70505428314209, -4.194101333618164, -3.7755603790283203, -3.210510730743408], "reading_logprob_std": 0.4031769216811006, "tokenization_path_logprob_std": 53.14716234852484, "trajectory_balance_rmse": 181.26215716507332} {"canonical_length": 6, "canonical_nll": 1.0380016962687175, "generation_rouge_l_f1": 0.47058823529411764, "generation_token_f1": 0.47058823529411764, "prompt_id": "9ecc7ea82de4a5d6", "reading_canonical_log_prob": -6.228010177612305, "reading_distances": [47, 46, 47, 40], "reading_log_probs": [-7.506657600402832, -6.852087020874023, -6.301504135131836, -6.6411333084106445], "reading_logprob_std": 0.5076892731356677, "tokenization_path_logprob_std": 13.578750641830164, "trajectory_balance_rmse": 61.604181550646054} {"canonical_length": 268, "canonical_nll": 1.1492120543522621, "generation_rouge_l_f1": 0.18725099601593623, "generation_token_f1": 0.37051792828685254, "prompt_id": "aceb2f4e30eb8855", "reading_canonical_log_prob": -307.98883056640625, "reading_distances": [25, 27, 32, 27], "reading_log_probs": [-322.71246337890625, -321.92718505859375, -321.9488525390625, -318.9952087402344], "reading_logprob_std": 1.6416128224323878, "tokenization_path_logprob_std": 159.15842736975176, "trajectory_balance_rmse": 4573.1999950091995} {"canonical_length": 118, "canonical_nll": 1.3128925905389301, "generation_rouge_l_f1": 0.2546816479400749, "generation_token_f1": 0.4719101123595506, "prompt_id": "c575c5fadea559fc", "reading_canonical_log_prob": -154.92132568359375, "reading_distances": [53, 46, 38, 46], "reading_log_probs": [-153.29049682617188, -152.14556884765625, -151.0029296875, -153.77899169921875], "reading_logprob_std": 1.2403852051597215, "tokenization_path_logprob_std": 79.04450081283947, "trajectory_balance_rmse": 2614.9638139421018} {"canonical_length": 61, "canonical_nll": 2.369197407706839, "generation_rouge_l_f1": 0.2278481012658228, "generation_token_f1": 0.3291139240506329, "prompt_id": "3adbad55ce76ffec", "reading_canonical_log_prob": -144.5210418701172, "reading_distances": [25, 21, 24, 19], "reading_log_probs": [-143.54525756835938, -142.30221557617188, -142.69041442871094, -143.00527954101562], "reading_logprob_std": 0.5253271825317228, "tokenization_path_logprob_std": 94.734510014396, "trajectory_balance_rmse": 1136.7916750784582} {"canonical_length": 12, "canonical_nll": 1.5636126200358074, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.4210526315789474, "prompt_id": "70c54675265e6b63", "reading_canonical_log_prob": -18.763351440429688, "reading_distances": [27, 34, 22, 28], "reading_log_probs": [-18.54994010925293, -19.450794219970703, -23.163671493530273, -17.74013328552246], "reading_logprob_std": 2.3958371480846097, "tokenization_path_logprob_std": 40.54027503897718, "trajectory_balance_rmse": 335.86232464381277} {"canonical_length": 11, "canonical_nll": 1.494557640769265, "generation_rouge_l_f1": 0.42857142857142855, "generation_token_f1": 0.5714285714285714, "prompt_id": "5e3e21467d43fd88", "reading_canonical_log_prob": -16.440134048461914, "reading_distances": [16, 20, 12, 17], "reading_log_probs": [-19.41611099243164, -19.258771896362305, -18.81546401977539, -18.716777801513672], "reading_logprob_std": 0.3384545428367657, "tokenization_path_logprob_std": 21.22100767480661, "trajectory_balance_rmse": 236.40772165270707} {"canonical_length": 17, "canonical_nll": 0.5426231833065257, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4210526315789474, "prompt_id": "46d8b242e85d5c4e", "reading_canonical_log_prob": -9.224594116210938, "reading_distances": [12, 8, 11, 10], "reading_log_probs": [-9.98400592803955, -9.607699394226074, -9.405468940734863, -9.017105102539062], "reading_logprob_std": 0.40329208851800175, "tokenization_path_logprob_std": 18.39156716577332, "trajectory_balance_rmse": 113.15484778952057} {"canonical_length": 48, "canonical_nll": 1.3241264820098877, "generation_rouge_l_f1": 0.11445783132530121, "generation_token_f1": 0.14457831325301204, "prompt_id": "6509666af0922805", "reading_canonical_log_prob": -63.55807113647461, "reading_distances": [30, 28, 32, 23], "reading_log_probs": [-64.02175903320312, -60.954933166503906, -60.71479415893555, -66.41841125488281], "reading_logprob_std": 2.7160615940228516, "tokenization_path_logprob_std": 69.62764046637781, "trajectory_balance_rmse": 874.5454363209528} {"canonical_length": 175, "canonical_nll": 1.130288783482143, "generation_rouge_l_f1": 0.3180722891566265, "generation_token_f1": 0.5590361445783132, "prompt_id": "42f6dd40e9d218d5", "reading_canonical_log_prob": -197.800537109375, "reading_distances": [8, 15, 12, 13], "reading_log_probs": [-204.0872802734375, -206.32901000976562, -203.1543426513672, -206.18759155273438], "reading_logprob_std": 1.570725501200759, "tokenization_path_logprob_std": 158.75640740221843, "trajectory_balance_rmse": 3077.7227268543015} {"canonical_length": 85, "canonical_nll": 1.1904465619255515, "generation_rouge_l_f1": 0.1662049861495845, "generation_token_f1": 0.3157894736842105, "prompt_id": "68880c3a241cb55b", "reading_canonical_log_prob": -101.18795776367188, "reading_distances": [35, 41, 42, 37], "reading_log_probs": [-107.61569213867188, -101.49320983886719, -106.39942169189453, -104.03436279296875], "reading_logprob_std": 2.7066569150865973, "tokenization_path_logprob_std": 185.41799547099794, "trajectory_balance_rmse": 1706.2266468180192} {"canonical_length": 14, "canonical_nll": 1.091750144958496, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5714285714285714, "prompt_id": "a0d2e6c5b9eebb8b", "reading_canonical_log_prob": -15.284502029418945, "reading_distances": [26, 28, 25, 32], "reading_log_probs": [-19.345632553100586, -20.305580139160156, -17.78435516357422, -21.420618057250977], "reading_logprob_std": 1.540750468270014, "tokenization_path_logprob_std": 46.73778669549237, "trajectory_balance_rmse": 332.5748532582817} {"canonical_length": 256, "canonical_nll": 0.5915359854698181, "generation_rouge_l_f1": 0.20430107526881722, "generation_token_f1": 0.3405017921146953, "prompt_id": "c51ea1a2a74eb3ff", "reading_canonical_log_prob": -151.43321228027344, "reading_distances": [28, 24, 27, 20], "reading_log_probs": [-158.91278076171875, -154.70980834960938, -159.1629180908203, -157.0435791015625], "reading_logprob_std": 2.061345245788564, "tokenization_path_logprob_std": 243.38693425199412, "trajectory_balance_rmse": 3037.626504783758} {"canonical_length": 61, "canonical_nll": 0.725127923684042, "generation_rouge_l_f1": 0.3717948717948718, "generation_token_f1": 0.5384615384615384, "prompt_id": "9ce480763a8f4a02", "reading_canonical_log_prob": -44.23280334472656, "reading_distances": [30, 32, 30, 29], "reading_log_probs": [-52.07429122924805, -54.58271408081055, -51.65983200073242, -58.44916915893555], "reading_logprob_std": 3.118370886720041, "tokenization_path_logprob_std": 71.17298449675619, "trajectory_balance_rmse": 860.0816835912193} {"canonical_length": 17, "canonical_nll": 0.6446560691384708, "generation_rouge_l_f1": 0.8125000000000001, "generation_token_f1": 0.8125000000000001, "prompt_id": "690ac1ef381d5161", "reading_canonical_log_prob": -10.959153175354004, "reading_distances": [32, 37, 39, 41], "reading_log_probs": [-8.55289363861084, -13.097533226013184, -8.214041709899902, -13.374156951904297], "reading_logprob_std": 2.8072072261936687, "tokenization_path_logprob_std": 28.6226691295238, "trajectory_balance_rmse": 288.4234051992164} {"canonical_length": 49, "canonical_nll": 1.2312184547891423, "generation_rouge_l_f1": 0.2898550724637681, "generation_token_f1": 0.3864734299516908, "prompt_id": "f544199aa0f474b9", "reading_canonical_log_prob": -60.32970428466797, "reading_distances": [20, 18, 19, 19], "reading_log_probs": [-58.855037689208984, -59.68475341796875, -55.91701126098633, -56.195770263671875], "reading_logprob_std": 1.8894161734527295, "tokenization_path_logprob_std": 65.36470036910637, "trajectory_balance_rmse": 993.754322835788} {"canonical_length": 236, "canonical_nll": 0.5079078027757548, "generation_rouge_l_f1": 0.1408450704225352, "generation_token_f1": 0.25821596244131456, "prompt_id": "672d37d3ef4c74d7", "reading_canonical_log_prob": -119.86624145507812, "reading_distances": [20, 24, 25, 26], "reading_log_probs": [-135.51504516601562, -131.42469787597656, -138.17080688476562, -132.3135986328125], "reading_logprob_std": 3.090850624364087, "tokenization_path_logprob_std": 158.65080518623404, "trajectory_balance_rmse": 4088.9552866248473} {"canonical_length": 76, "canonical_nll": 0.9948431316174959, "generation_rouge_l_f1": 0.2723404255319149, "generation_token_f1": 0.451063829787234, "prompt_id": "e905bcd075fa835d", "reading_canonical_log_prob": -75.60807800292969, "reading_distances": [19, 17, 18, 19], "reading_log_probs": [-81.56578063964844, -81.63644409179688, -76.9011459350586, -80.33465576171875], "reading_logprob_std": 2.2208513745689835, "tokenization_path_logprob_std": 74.3636267510197, "trajectory_balance_rmse": 1556.34728614702} {"canonical_length": 10, "canonical_nll": 1.3692880630493165, "generation_rouge_l_f1": 0.30000000000000004, "generation_token_f1": 0.39999999999999997, "prompt_id": "1d805df4e7b3cce0", "reading_canonical_log_prob": -13.692880630493164, "reading_distances": [34, 41, 36, 34], "reading_log_probs": [-13.080294609069824, -14.382575988769531, -13.533218383789062, -13.168816566467285], "reading_logprob_std": 0.594160600861489, "tokenization_path_logprob_std": 29.973525995203737, "trajectory_balance_rmse": 221.46291301777006} {"canonical_length": 140, "canonical_nll": 1.5647683279854911, "generation_rouge_l_f1": 0.17777777777777778, "generation_token_f1": 0.3160493827160494, "prompt_id": "6a438344f95ac9cf", "reading_canonical_log_prob": -219.06756591796875, "reading_distances": [25, 26, 18, 25], "reading_log_probs": [-222.97579956054688, -221.30917358398438, -227.66229248046875, -222.62066650390625], "reading_logprob_std": 2.7743933781635266, "tokenization_path_logprob_std": 170.84367772181014, "trajectory_balance_rmse": 3270.0945209963525} {"canonical_length": 103, "canonical_nll": 1.894799093598301, "generation_rouge_l_f1": 0.22662889518413598, "generation_token_f1": 0.38526912181303113, "prompt_id": "208ec2e52d9d1ce6", "reading_canonical_log_prob": -195.164306640625, "reading_distances": [32, 29, 36, 30], "reading_log_probs": [-206.39776611328125, -205.4331512451172, -204.4995574951172, -204.0592041015625], "reading_logprob_std": 1.0390616455543682, "tokenization_path_logprob_std": 149.71857749187478, "trajectory_balance_rmse": 2079.8137775214086} {"canonical_length": 14, "canonical_nll": 0.2981821468898228, "generation_rouge_l_f1": 0.7199999999999999, "generation_token_f1": 0.7199999999999999, "prompt_id": "6a81e01870d01f92", "reading_canonical_log_prob": -4.1745500564575195, "reading_distances": [25, 28, 31, 24], "reading_log_probs": [-7.497313022613525, -6.575770854949951, -9.665862083435059, -6.630683898925781], "reading_logprob_std": 1.4453049530715003, "tokenization_path_logprob_std": 41.266696364491, "trajectory_balance_rmse": 350.5503639900057} {"canonical_length": 274, "canonical_nll": 1.2786660298813868, "generation_rouge_l_f1": 0.19521178637200734, "generation_token_f1": 0.40147329650092084, "prompt_id": "7857d9054d4c3033", "reading_canonical_log_prob": -350.3544921875, "reading_distances": [29, 31, 23, 30], "reading_log_probs": [-356.1824035644531, -359.84130859375, -356.279052734375, -353.06365966796875], "reading_logprob_std": 2.770205090295779, "tokenization_path_logprob_std": 213.01861610510053, "trajectory_balance_rmse": 4347.59007232781} {"canonical_length": 158, "canonical_nll": 1.349999439867237, "generation_rouge_l_f1": 0.13775510204081634, "generation_token_f1": 0.2704081632653061, "prompt_id": "77515334ad84a428", "reading_canonical_log_prob": -213.29991149902344, "reading_distances": [19, 18, 20, 16], "reading_log_probs": [-216.87786865234375, -218.43328857421875, -217.01710510253906, -214.99099731445312], "reading_logprob_std": 1.4130068081825196, "tokenization_path_logprob_std": 109.01498235682408, "trajectory_balance_rmse": 2560.701020214237} {"canonical_length": 43, "canonical_nll": 0.8047634390897529, "generation_rouge_l_f1": 0.2529644268774704, "generation_token_f1": 0.27667984189723316, "prompt_id": "78695f86d2d683db", "reading_canonical_log_prob": -34.604827880859375, "reading_distances": [40, 37, 35, 31], "reading_log_probs": [-34.224266052246094, -36.1228141784668, -35.051979064941406, -36.08882522583008], "reading_logprob_std": 0.9123718074962339, "tokenization_path_logprob_std": 125.78101012431644, "trajectory_balance_rmse": 1119.082098546621} {"canonical_length": 8, "canonical_nll": 0.3904587924480438, "generation_rouge_l_f1": 0.13559322033898305, "generation_token_f1": 0.13559322033898305, "prompt_id": "73a081a89d8a40bd", "reading_canonical_log_prob": -3.1236703395843506, "reading_distances": [28, 28, 32, 25], "reading_log_probs": [-2.25958514213562, -1.8453845977783203, -2.6182167530059814, -1.6422438621520996], "reading_logprob_std": 0.4351578029423466, "tokenization_path_logprob_std": 26.52698254479372, "trajectory_balance_rmse": 97.76526013058101} {"canonical_length": 84, "canonical_nll": 1.1723253159295945, "generation_rouge_l_f1": 0.1838235294117647, "generation_token_f1": 0.27205882352941174, "prompt_id": "7879ca5705298aac", "reading_canonical_log_prob": -98.47532653808594, "reading_distances": [46, 46, 41, 35], "reading_log_probs": [-103.00103759765625, -106.93067932128906, -102.64180755615234, -101.25856018066406], "reading_logprob_std": 2.433928369813383, "tokenization_path_logprob_std": 120.32276162719025, "trajectory_balance_rmse": 1264.7092971010245} {"canonical_length": 355, "canonical_nll": 1.2680547150088028, "generation_rouge_l_f1": 0.20253164556962025, "generation_token_f1": 0.4335443037974684, "prompt_id": "2e88e8dc9b895643", "reading_canonical_log_prob": -450.159423828125, "reading_distances": [23, 20, 20, 21], "reading_log_probs": [-455.3570861816406, -455.9150390625, -449.8663330078125, -451.66802978515625], "reading_logprob_std": 2.914602073253356, "tokenization_path_logprob_std": 269.3595193335543, "trajectory_balance_rmse": 7868.322195510426} {"canonical_length": 343, "canonical_nll": 1.405822664933719, "generation_rouge_l_f1": 0.18512396694214878, "generation_token_f1": 0.3900826446280992, "prompt_id": "a81b52bde10fe226", "reading_canonical_log_prob": -482.1971740722656, "reading_distances": [19, 20, 20, 19], "reading_log_probs": [-496.8898010253906, -484.4146423339844, -488.67645263671875, -483.7689208984375], "reading_logprob_std": 6.04090240517643, "tokenization_path_logprob_std": 125.52223810702513, "trajectory_balance_rmse": 6834.962865539237} {"canonical_length": 269, "canonical_nll": 1.2115388891510803, "generation_rouge_l_f1": 0.3692870201096892, "generation_token_f1": 0.5886654478976234, "prompt_id": "ff000a10ec04c16c", "reading_canonical_log_prob": -325.9039611816406, "reading_distances": [33, 25, 32, 26], "reading_log_probs": [-326.9682312011719, -330.1791076660156, -331.322021484375, -332.968994140625], "reading_logprob_std": 2.534373867874073, "tokenization_path_logprob_std": 388.58735965668507, "trajectory_balance_rmse": 5857.819185250494} {"canonical_length": 405, "canonical_nll": 1.2578441478587963, "generation_rouge_l_f1": 0.22082018927444796, "generation_token_f1": 0.41009463722397477, "prompt_id": "e7f10884569482cf", "reading_canonical_log_prob": -509.4268798828125, "reading_distances": [31, 38, 41, 31], "reading_log_probs": [-503.25616455078125, -508.20196533203125, -504.5242614746094, -506.2477111816406], "reading_logprob_std": 2.147348369899537, "tokenization_path_logprob_std": 231.30631672543345, "trajectory_balance_rmse": 6974.611307191246} {"canonical_length": 10, "canonical_nll": 0.9229780197143554, "generation_rouge_l_f1": 0.7368421052631577, "generation_token_f1": 0.7368421052631577, "prompt_id": "7d1f1626ed9a7e46", "reading_canonical_log_prob": -9.229780197143555, "reading_distances": [34, 38, 35, 40], "reading_log_probs": [-10.740545272827148, -11.873703956604004, -11.221661567687988, -11.086811065673828], "reading_logprob_std": 0.47416293329477677, "tokenization_path_logprob_std": 31.652477420378887, "trajectory_balance_rmse": 265.94310100967033} {"canonical_length": 41, "canonical_nll": 1.6783170002262766, "generation_rouge_l_f1": 0.1625, "generation_token_f1": 0.275, "prompt_id": "8de4f5d47e6be004", "reading_canonical_log_prob": -68.81099700927734, "reading_distances": [13, 16, 15, 15], "reading_log_probs": [-68.67618560791016, -64.67617797851562, -68.5402603149414, -68.70152282714844], "reading_logprob_std": 1.9828371454832423, "tokenization_path_logprob_std": 88.15705640228481, "trajectory_balance_rmse": 883.5759797085109} {"canonical_length": 187, "canonical_nll": 1.2196674856910095, "generation_rouge_l_f1": 0.32456140350877194, "generation_token_f1": 0.5394736842105263, "prompt_id": "37dc889754565fdd", "reading_canonical_log_prob": -228.07781982421875, "reading_distances": [32, 40, 40, 36], "reading_log_probs": [-225.0496368408203, -225.44187927246094, -226.07867431640625, -225.20968627929688], "reading_logprob_std": 0.45211998746675786, "tokenization_path_logprob_std": 140.76750815130467, "trajectory_balance_rmse": 3966.5214624188866} {"canonical_length": 81, "canonical_nll": 1.3838169427565585, "generation_rouge_l_f1": 0.2945736434108527, "generation_token_f1": 0.46511627906976744, "prompt_id": "fccb0a4a57193f97", "reading_canonical_log_prob": -112.08917236328125, "reading_distances": [24, 26, 24, 21], "reading_log_probs": [-113.28289031982422, -115.66033172607422, -113.77716064453125, -114.92509460449219], "reading_logprob_std": 1.0800472609398224, "tokenization_path_logprob_std": 113.38939259696802, "trajectory_balance_rmse": 1784.4685161479629} {"canonical_length": 8, "canonical_nll": 1.0169200897216797, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "a29eafa220e503a1", "reading_canonical_log_prob": -8.135360717773438, "reading_distances": [35, 27, 28, 37], "reading_log_probs": [-10.455377578735352, -11.25088882446289, -8.935020446777344, -9.088533401489258], "reading_logprob_std": 1.1133725230479488, "tokenization_path_logprob_std": 47.03055160026336, "trajectory_balance_rmse": 193.7236504036322} {"canonical_length": 20, "canonical_nll": 1.3625965118408203, "generation_rouge_l_f1": 0.35294117647058826, "generation_token_f1": 0.5294117647058824, "prompt_id": "276e3276fb9b91f2", "reading_canonical_log_prob": -27.251930236816406, "reading_distances": [40, 49, 46, 48], "reading_log_probs": [-28.199478149414062, -41.347957611083984, -33.493995666503906, -31.545156478881836], "reading_logprob_std": 5.580364338617929, "tokenization_path_logprob_std": 36.63599926776767, "trajectory_balance_rmse": 422.3530946969384} {"canonical_length": 229, "canonical_nll": 1.2636891993893287, "generation_rouge_l_f1": 0.2385685884691849, "generation_token_f1": 0.40159045725646125, "prompt_id": "14d153fd0e1b36b1", "reading_canonical_log_prob": -289.38482666015625, "reading_distances": [15, 21, 16, 22], "reading_log_probs": [-290.581787109375, -290.3868408203125, -285.7801208496094, -286.3772277832031], "reading_logprob_std": 2.5564905035853993, "tokenization_path_logprob_std": 83.3526774566742, "trajectory_balance_rmse": 4227.66734749475} {"canonical_length": 315, "canonical_nll": 0.5075072273375496, "generation_rouge_l_f1": 0.4526112185686654, "generation_token_f1": 0.5531914893617021, "prompt_id": "0ee97cbd046603bc", "reading_canonical_log_prob": -159.86477661132812, "reading_distances": [25, 28, 24, 29], "reading_log_probs": [-168.86965942382812, -166.63128662109375, -165.94046020507812, -166.2635498046875], "reading_logprob_std": 1.3259960597825189, "tokenization_path_logprob_std": 202.43246597832618, "trajectory_balance_rmse": 4898.426336146015} {"canonical_length": 282, "canonical_nll": 0.8676481855676529, "generation_rouge_l_f1": 0.25225225225225223, "generation_token_f1": 0.5081081081081081, "prompt_id": "740c83c01085ccfc", "reading_canonical_log_prob": -244.67678833007812, "reading_distances": [16, 17, 28, 24], "reading_log_probs": [-265.677978515625, -264.30548095703125, -265.6216735839844, -265.38555908203125], "reading_logprob_std": 0.6407749711740988, "tokenization_path_logprob_std": 304.3252793374053, "trajectory_balance_rmse": 6220.028368302407} {"canonical_length": 128, "canonical_nll": 1.3720507621765137, "generation_rouge_l_f1": 0.24742268041237112, "generation_token_f1": 0.4278350515463917, "prompt_id": "055891d31fa97386", "reading_canonical_log_prob": -175.62249755859375, "reading_distances": [26, 24, 21, 22], "reading_log_probs": [-178.44363403320312, -183.04437255859375, -185.30491638183594, -183.80551147460938], "reading_logprob_std": 2.957072398581378, "tokenization_path_logprob_std": 298.5298515003667, "trajectory_balance_rmse": 2438.0911178786773} {"canonical_length": 197, "canonical_nll": 1.108434376982868, "generation_rouge_l_f1": 0.2953020134228188, "generation_token_f1": 0.49217002237136465, "prompt_id": "54415ac83a3ddec9", "reading_canonical_log_prob": -218.361572265625, "reading_distances": [20, 19, 19, 19], "reading_log_probs": [-224.1509246826172, -222.53689575195312, -224.51153564453125, -223.87841796875], "reading_logprob_std": 0.8616414625045161, "tokenization_path_logprob_std": 183.0025517690412, "trajectory_balance_rmse": 3197.391368580276} {"canonical_length": 71, "canonical_nll": 1.4515772000165053, "generation_rouge_l_f1": 0.18085106382978725, "generation_token_f1": 0.44680851063829785, "prompt_id": "6bb4c6466fd9ddad", "reading_canonical_log_prob": -103.06198120117188, "reading_distances": [11, 17, 13, 18], "reading_log_probs": [-114.41809844970703, -114.00636291503906, -112.04137420654297, -111.36524200439453], "reading_logprob_std": 1.4841420585134142, "tokenization_path_logprob_std": 133.9896365592681, "trajectory_balance_rmse": 1348.0059836201365} {"canonical_length": 96, "canonical_nll": 0.9653031826019287, "generation_rouge_l_f1": 0.30508474576271183, "generation_token_f1": 0.46610169491525427, "prompt_id": "5ca3d0261fdb6645", "reading_canonical_log_prob": -92.66910552978516, "reading_distances": [20, 27, 25, 23], "reading_log_probs": [-97.86467742919922, -97.80188751220703, -101.42068481445312, -108.50765991210938], "reading_logprob_std": 5.032036365936118, "tokenization_path_logprob_std": 208.27625077984226, "trajectory_balance_rmse": 1914.812616672545} {"canonical_length": 120, "canonical_nll": 1.425806427001953, "generation_rouge_l_f1": 0.2638522427440633, "generation_token_f1": 0.41688654353562005, "prompt_id": "2749992f70103af2", "reading_canonical_log_prob": -171.09677124023438, "reading_distances": [38, 44, 37, 44], "reading_log_probs": [-178.6058807373047, -186.45626831054688, -180.51901245117188, -182.4122772216797], "reading_logprob_std": 3.3536874575948503, "tokenization_path_logprob_std": 196.8135681418511, "trajectory_balance_rmse": 2914.65022047519} {"canonical_length": 18, "canonical_nll": 0.030521962377760146, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "67a8cb2f3ba290b7", "reading_canonical_log_prob": -0.5493953227996826, "reading_distances": [39, 50, 47, 39], "reading_log_probs": [-1.4767341613769531, -1.476159691810608, -2.23921537399292, -0.9180870652198792], "reading_logprob_std": 0.5425667045438155, "tokenization_path_logprob_std": 36.93238983074377, "trajectory_balance_rmse": 330.851400225097} {"canonical_length": 144, "canonical_nll": 0.9279309378729926, "generation_rouge_l_f1": 0.3012345679012346, "generation_token_f1": 0.46419753086419746, "prompt_id": "14aa2b70fa0eb52a", "reading_canonical_log_prob": -133.62205505371094, "reading_distances": [32, 26, 28, 30], "reading_log_probs": [-148.80567932128906, -145.09500122070312, -144.20068359375, -147.28598022460938], "reading_logprob_std": 2.0897338601502073, "tokenization_path_logprob_std": 156.655675958984, "trajectory_balance_rmse": 2342.510440595476} {"canonical_length": 46, "canonical_nll": 1.4888172978940217, "generation_rouge_l_f1": 0.08469055374592832, "generation_token_f1": 0.11074918566775244, "prompt_id": "632380e80642e576", "reading_canonical_log_prob": -68.485595703125, "reading_distances": [26, 32, 28, 25], "reading_log_probs": [-77.54107666015625, -79.4707260131836, -77.96553039550781, -75.72920227050781], "reading_logprob_std": 1.5398152933090063, "tokenization_path_logprob_std": 71.56654559104642, "trajectory_balance_rmse": 856.7050197243757} {"canonical_length": 186, "canonical_nll": 1.3593382476478495, "generation_rouge_l_f1": 0.26388888888888884, "generation_token_f1": 0.462962962962963, "prompt_id": "e0e28697fbbba739", "reading_canonical_log_prob": -252.8369140625, "reading_distances": [36, 28, 34, 28], "reading_log_probs": [-264.6396484375, -260.8481140136719, -262.8082275390625, -258.2536315917969], "reading_logprob_std": 2.7360064458470226, "tokenization_path_logprob_std": 332.93159389834, "trajectory_balance_rmse": 3996.307395255762} {"canonical_length": 41, "canonical_nll": 1.1292495727539062, "generation_rouge_l_f1": 0.2631578947368421, "generation_token_f1": 0.40350877192982454, "prompt_id": "dd2bdffacb538a3a", "reading_canonical_log_prob": -46.299232482910156, "reading_distances": [145, 138, 159, 126], "reading_log_probs": [-53.15928649902344, -46.28835678100586, -51.789085388183594, -47.875003814697266], "reading_logprob_std": 3.2288577488177315, "tokenization_path_logprob_std": 52.623742660192114, "trajectory_balance_rmse": 955.929700698024} {"canonical_length": 11, "canonical_nll": 0.4339276227084073, "generation_rouge_l_f1": 0.06504065040650406, "generation_token_f1": 0.06504065040650406, "prompt_id": "fbdd1075d5b5f58e", "reading_canonical_log_prob": -4.7732038497924805, "reading_distances": [25, 19, 24, 15], "reading_log_probs": [-6.140774250030518, -5.877415657043457, -5.514123916625977, -5.570040702819824], "reading_logprob_std": 0.29117178619125034, "tokenization_path_logprob_std": 41.69500551642099, "trajectory_balance_rmse": 170.35625419425767} {"canonical_length": 25, "canonical_nll": 2.204437561035156, "generation_rouge_l_f1": 0.1527777777777778, "generation_token_f1": 0.1527777777777778, "prompt_id": "71ff4b9aa3b9c919", "reading_canonical_log_prob": -55.110939025878906, "reading_distances": [30, 22, 22, 23], "reading_log_probs": [-55.415313720703125, -53.32189178466797, -55.063682556152344, -58.06766891479492], "reading_logprob_std": 1.9604680065645275, "tokenization_path_logprob_std": 90.8628694962536, "trajectory_balance_rmse": 496.5834307563694} {"canonical_length": 227, "canonical_nll": 0.6302915060572687, "generation_rouge_l_f1": 0.33671399594320484, "generation_token_f1": 0.4949290060851927, "prompt_id": "84c9e255e3abf202", "reading_canonical_log_prob": -143.076171875, "reading_distances": [29, 23, 23, 24], "reading_log_probs": [-164.78683471679688, -163.88150024414062, -160.0535430908203, -157.81442260742188], "reading_logprob_std": 3.2699969202799224, "tokenization_path_logprob_std": 105.51229179522065, "trajectory_balance_rmse": 4056.0943104456146} {"canonical_length": 20, "canonical_nll": 0.3434009075164795, "generation_rouge_l_f1": 0.6274509803921569, "generation_token_f1": 0.6274509803921569, "prompt_id": "7aac721bb785059a", "reading_canonical_log_prob": -6.86801815032959, "reading_distances": [19, 22, 28, 23], "reading_log_probs": [-6.250827789306641, -7.458775520324707, -8.724678039550781, -6.6084489822387695], "reading_logprob_std": 1.099657353137282, "tokenization_path_logprob_std": 51.054238797207645, "trajectory_balance_rmse": 364.2099121898094} {"canonical_length": 16, "canonical_nll": 0.6306356191635132, "generation_rouge_l_f1": 0.8387096774193549, "generation_token_f1": 0.8387096774193549, "prompt_id": "e8e5a8f77a9d33c4", "reading_canonical_log_prob": -10.090169906616211, "reading_distances": [43, 41, 42, 43], "reading_log_probs": [-14.404842376708984, -13.52766227722168, -9.568952560424805, -13.52842903137207], "reading_logprob_std": 2.1654909514130423, "tokenization_path_logprob_std": 48.32832155060148, "trajectory_balance_rmse": 281.818418811745} {"canonical_length": 139, "canonical_nll": 1.553718182680418, "generation_rouge_l_f1": 0.07671232876712329, "generation_token_f1": 0.11506849315068493, "prompt_id": "b0bb5fcadef34afc", "reading_canonical_log_prob": -215.96682739257812, "reading_distances": [16, 10, 16, 17], "reading_log_probs": [-232.81820678710938, -217.7360076904297, -215.75180053710938, -217.0843505859375], "reading_logprob_std": 8.023022121177457, "tokenization_path_logprob_std": 191.5361553937166, "trajectory_balance_rmse": 2403.2538586540186} {"canonical_length": 214, "canonical_nll": 1.6051886727876752, "generation_rouge_l_f1": 0.2287390029325513, "generation_token_f1": 0.404692082111437, "prompt_id": "6db457998b4e4d8a", "reading_canonical_log_prob": -343.5103759765625, "reading_distances": [39, 38, 39, 35], "reading_log_probs": [-350.8918151855469, -343.53533935546875, -345.13482666015625, -344.76239013671875], "reading_logprob_std": 3.2791414929675513, "tokenization_path_logprob_std": 252.00401505053762, "trajectory_balance_rmse": 3876.6189202505007} {"canonical_length": 124, "canonical_nll": 1.6418073100428427, "generation_rouge_l_f1": 0.18421052631578946, "generation_token_f1": 0.2631578947368421, "prompt_id": "b7fe2d1d82e7517d", "reading_canonical_log_prob": -203.5841064453125, "reading_distances": [21, 28, 32, 26], "reading_log_probs": [-199.79266357421875, -202.25698852539062, -204.4772491455078, -200.3535614013672], "reading_logprob_std": 2.1191633266670102, "tokenization_path_logprob_std": 112.78855589029112, "trajectory_balance_rmse": 2291.0318539970867} {"canonical_length": 3, "canonical_nll": 0.8737468719482422, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "fe8dff1c5ccd0457", "reading_canonical_log_prob": -2.6212406158447266, "reading_distances": [57, 77, 54, 55], "reading_log_probs": [-1.4705345630645752, -2.544837236404419, -1.395871877670288, -2.9925172328948975], "reading_logprob_std": 0.7929873455793783, "tokenization_path_logprob_std": 11.75703613521195, "trajectory_balance_rmse": 63.37061096443167} {"canonical_length": 36, "canonical_nll": 1.598713133070204, "generation_rouge_l_f1": 0.1857142857142857, "generation_token_f1": 0.2857142857142857, "prompt_id": "007cda46c19b41d2", "reading_canonical_log_prob": -57.553672790527344, "reading_distances": [22, 16, 22, 16], "reading_log_probs": [-59.81294250488281, -64.94406127929688, -62.518348693847656, -61.66290283203125], "reading_logprob_std": 2.130176551020772, "tokenization_path_logprob_std": 116.86523339097364, "trajectory_balance_rmse": 785.7234805558297} {"canonical_length": 121, "canonical_nll": 1.6858853426846592, "generation_rouge_l_f1": 0.12293144208037825, "generation_token_f1": 0.2458628841607565, "prompt_id": "0f24b7ddfdbc9dca", "reading_canonical_log_prob": -203.99212646484375, "reading_distances": [20, 21, 20, 22], "reading_log_probs": [-210.17156982421875, -210.41656494140625, -207.6470489501953, -210.69081115722656], "reading_logprob_std": 1.4057251390340064, "tokenization_path_logprob_std": 99.06904964658284, "trajectory_balance_rmse": 1863.5150110034933} {"canonical_length": 4, "canonical_nll": 0.4188348352909088, "generation_rouge_l_f1": 0.031746031746031744, "generation_token_f1": 0.031746031746031744, "prompt_id": "0131c4e6a39cb194", "reading_canonical_log_prob": -1.6753393411636353, "reading_distances": [17, 27, 22, 21], "reading_log_probs": [-1.3881325721740723, -1.5175082683563232, -1.5005831718444824, -1.5132218599319458], "reading_logprob_std": 0.061573199436616995, "tokenization_path_logprob_std": 5.336241694288752, "trajectory_balance_rmse": 57.28036352475126} {"canonical_length": 182, "canonical_nll": 1.135721898340917, "generation_rouge_l_f1": 0.10050251256281408, "generation_token_f1": 0.17085427135678394, "prompt_id": "21b0719467934ebb", "reading_canonical_log_prob": -206.70138549804688, "reading_distances": [46, 42, 48, 46], "reading_log_probs": [-213.48968505859375, -219.322998046875, -216.30316162109375, -216.7347869873047], "reading_logprob_std": 2.3888356002249482, "tokenization_path_logprob_std": 204.2213934762067, "trajectory_balance_rmse": 2981.08715784625} {"canonical_length": 246, "canonical_nll": 1.4133690314564278, "generation_rouge_l_f1": 0.21956087824351297, "generation_token_f1": 0.47504990019960086, "prompt_id": "5d16c846337cb10b", "reading_canonical_log_prob": -347.68878173828125, "reading_distances": [17, 19, 21, 15], "reading_log_probs": [-353.26800537109375, -350.6430358886719, -351.9069519042969, -351.15960693359375], "reading_logprob_std": 1.1405820567667306, "tokenization_path_logprob_std": 147.37044108772892, "trajectory_balance_rmse": 5203.303773378137} {"canonical_length": 116, "canonical_nll": 0.5031298933358028, "generation_rouge_l_f1": 0.39252336448598135, "generation_token_f1": 0.49221183800623053, "prompt_id": "d66394bc1b63c0cb", "reading_canonical_log_prob": -58.363067626953125, "reading_distances": [22, 14, 17, 19], "reading_log_probs": [-58.956214904785156, -59.057518005371094, -57.93809509277344, -57.25898742675781], "reading_logprob_std": 0.8600590339945177, "tokenization_path_logprob_std": 42.11018550984232, "trajectory_balance_rmse": 1057.1871312156868} {"canonical_length": 75, "canonical_nll": 1.0627448527018228, "generation_rouge_l_f1": 0.3826086956521739, "generation_token_f1": 0.5043478260869566, "prompt_id": "4b9d4e7ec5d5b8b5", "reading_canonical_log_prob": -79.70586395263672, "reading_distances": [39, 32, 36, 40], "reading_log_probs": [-76.35138702392578, -73.75105285644531, -80.06172180175781, -79.04305267333984], "reading_logprob_std": 2.837852434895782, "tokenization_path_logprob_std": 142.603038512713, "trajectory_balance_rmse": 1484.4927709240965} {"canonical_length": 363, "canonical_nll": 1.2505197236032197, "generation_rouge_l_f1": 0.3142857142857143, "generation_token_f1": 0.5714285714285714, "prompt_id": "cddcde885956c90e", "reading_canonical_log_prob": -453.93865966796875, "reading_distances": [21, 21, 25, 24], "reading_log_probs": [-453.1285400390625, -452.5092468261719, -457.39013671875, -457.9449157714844], "reading_logprob_std": 2.8198635954536453, "tokenization_path_logprob_std": 151.62032813673696, "trajectory_balance_rmse": 7525.257218660112} {"canonical_length": 204, "canonical_nll": 1.606908312030867, "generation_rouge_l_f1": 0.19426048565121412, "generation_token_f1": 0.40618101545253865, "prompt_id": "e708ffc93eb420d5", "reading_canonical_log_prob": -327.8092956542969, "reading_distances": [40, 42, 33, 39], "reading_log_probs": [-327.2403564453125, -330.802978515625, -329.1978759765625, -328.76434326171875], "reading_logprob_std": 1.4653507265582382, "tokenization_path_logprob_std": 51.65451219866045, "trajectory_balance_rmse": 4240.521330441405} {"canonical_length": 73, "canonical_nll": 1.9505544166042381, "generation_rouge_l_f1": 0.15384615384615385, "generation_token_f1": 0.24260355029585795, "prompt_id": "276028c9e0375416", "reading_canonical_log_prob": -142.39047241210938, "reading_distances": [30, 25, 29, 25], "reading_log_probs": [-142.12721252441406, -147.43707275390625, -144.08139038085938, -142.97671508789062], "reading_logprob_std": 2.3293558400906806, "tokenization_path_logprob_std": 104.48019610551225, "trajectory_balance_rmse": 1579.31900724266} {"canonical_length": 100, "canonical_nll": 1.0888401794433593, "generation_rouge_l_f1": 0.2754491017964072, "generation_token_f1": 0.40718562874251496, "prompt_id": "328ee04612e4bcf9", "reading_canonical_log_prob": -108.88401794433594, "reading_distances": [28, 27, 25, 29], "reading_log_probs": [-118.40330505371094, -112.89990234375, -110.24313354492188, -116.87950897216797], "reading_logprob_std": 3.7208305873567453, "tokenization_path_logprob_std": 102.91014870649441, "trajectory_balance_rmse": 1665.778709826765} {"canonical_length": 90, "canonical_nll": 1.105615743001302, "generation_rouge_l_f1": 0.19571865443425074, "generation_token_f1": 0.2996941896024465, "prompt_id": "bdf95ab3a5ffa3ff", "reading_canonical_log_prob": -99.50541687011719, "reading_distances": [25, 34, 30, 30], "reading_log_probs": [-99.1829833984375, -101.04961395263672, -91.08280181884766, -98.66581726074219], "reading_logprob_std": 4.395894543327885, "tokenization_path_logprob_std": 92.55074955906389, "trajectory_balance_rmse": 1102.7348059657252} {"canonical_length": 132, "canonical_nll": 1.6935803962476326, "generation_rouge_l_f1": 0.2721518987341772, "generation_token_f1": 0.4113924050632911, "prompt_id": "80019145cee29a1a", "reading_canonical_log_prob": -223.5526123046875, "reading_distances": [18, 18, 19, 24], "reading_log_probs": [-229.69400024414062, -226.69541931152344, -229.5462188720703, -225.03643798828125], "reading_logprob_std": 2.271629626055234, "tokenization_path_logprob_std": 71.1227942320173, "trajectory_balance_rmse": 2488.488287468821} {"canonical_length": 250, "canonical_nll": 1.41506591796875, "generation_rouge_l_f1": 0.27822580645161293, "generation_token_f1": 0.40322580645161293, "prompt_id": "77663077eb493ddd", "reading_canonical_log_prob": -353.7664794921875, "reading_distances": [31, 30, 29, 30], "reading_log_probs": [-359.04071044921875, -359.3779296875, -357.19219970703125, -359.45684814453125], "reading_logprob_std": 1.0652134205548074, "tokenization_path_logprob_std": 158.79830623351947, "trajectory_balance_rmse": 4823.891356148339} {"canonical_length": 369, "canonical_nll": 1.253948511469978, "generation_rouge_l_f1": 0.2974683544303797, "generation_token_f1": 0.5664556962025317, "prompt_id": "4825c22b71594c0c", "reading_canonical_log_prob": -462.7070007324219, "reading_distances": [32, 32, 36, 32], "reading_log_probs": [-466.509765625, -468.2431945800781, -466.8395080566406, -464.80145263671875], "reading_logprob_std": 1.4142557711534502, "tokenization_path_logprob_std": 231.3081743086184, "trajectory_balance_rmse": 7388.681719940177} {"canonical_length": 198, "canonical_nll": 1.1415505457406092, "generation_rouge_l_f1": 0.358974358974359, "generation_token_f1": 0.564102564102564, "prompt_id": "c520430df28e5bdb", "reading_canonical_log_prob": -226.02700805664062, "reading_distances": [17, 21, 22, 21], "reading_log_probs": [-228.5535430908203, -236.67405700683594, -231.97293090820312, -241.22994995117188], "reading_logprob_std": 5.5292811382705995, "tokenization_path_logprob_std": 192.5138089350511, "trajectory_balance_rmse": 4167.619824841814} {"canonical_length": 290, "canonical_nll": 1.3436386634563577, "generation_rouge_l_f1": 0.1996370235934664, "generation_token_f1": 0.47549909255898365, "prompt_id": "c3e62d8cbe21b9a4", "reading_canonical_log_prob": -389.65521240234375, "reading_distances": [25, 23, 28, 26], "reading_log_probs": [-394.7208557128906, -395.9253845214844, -395.0234069824219, -392.8139343261719], "reading_logprob_std": 1.3087917874306976, "tokenization_path_logprob_std": 175.74983053546285, "trajectory_balance_rmse": 6196.043315402711} {"canonical_length": 38, "canonical_nll": 0.6761549899452611, "generation_rouge_l_f1": 0.2158273381294964, "generation_token_f1": 0.2589928057553957, "prompt_id": "ef7fdf70b61d5a5a", "reading_canonical_log_prob": -25.693889617919922, "reading_distances": [34, 27, 27, 31], "reading_log_probs": [-29.427885055541992, -31.270713806152344, -33.94676208496094, -29.76439094543457], "reading_logprob_std": 2.058558859270995, "tokenization_path_logprob_std": 85.38634862633523, "trajectory_balance_rmse": 614.2799220055351} {"canonical_length": 262, "canonical_nll": 1.0305512406443822, "generation_rouge_l_f1": 0.3121387283236994, "generation_token_f1": 0.605009633911368, "prompt_id": "f0bfbb195bb03bbd", "reading_canonical_log_prob": -270.0044250488281, "reading_distances": [38, 26, 27, 32], "reading_log_probs": [-278.6496276855469, -279.5231628417969, -276.51715087890625, -279.42852783203125], "reading_logprob_std": 1.3975688983481456, "tokenization_path_logprob_std": 94.457501852574, "trajectory_balance_rmse": 5625.054291176957} {"canonical_length": 12, "canonical_nll": 0.7141803105672201, "generation_rouge_l_f1": 0.7000000000000001, "generation_token_f1": 0.7000000000000001, "prompt_id": "064f3342c7039feb", "reading_canonical_log_prob": -8.57016372680664, "reading_distances": [32, 31, 33, 30], "reading_log_probs": [-8.967866897583008, -10.203532218933105, -9.591312408447266, -9.847519874572754], "reading_logprob_std": 0.5209400639483133, "tokenization_path_logprob_std": 51.65386445056644, "trajectory_balance_rmse": 290.7220786278993} {"canonical_length": 174, "canonical_nll": 1.6028326626481681, "generation_rouge_l_f1": 0.2102803738317757, "generation_token_f1": 0.37850467289719625, "prompt_id": "5319439232a42bef", "reading_canonical_log_prob": -278.89288330078125, "reading_distances": [27, 27, 25, 31], "reading_log_probs": [-279.8824157714844, -278.2611083984375, -276.64227294921875, -272.2665100097656], "reading_logprob_std": 3.2765876944539247, "tokenization_path_logprob_std": 121.06946815671712, "trajectory_balance_rmse": 3930.1264302725613} {"canonical_length": 274, "canonical_nll": 1.1687800275148266, "generation_rouge_l_f1": 0.18909090909090912, "generation_token_f1": 0.45090909090909087, "prompt_id": "7f189f92ecaccbe7", "reading_canonical_log_prob": -320.2457275390625, "reading_distances": [21, 13, 20, 19], "reading_log_probs": [-326.2923278808594, -326.8443603515625, -326.5905456542969, -326.64898681640625], "reading_logprob_std": 0.22856205824426037, "tokenization_path_logprob_std": 121.82166834476116, "trajectory_balance_rmse": 5438.306909140352} {"canonical_length": 225, "canonical_nll": 1.0905249701605904, "generation_rouge_l_f1": 0.304932735426009, "generation_token_f1": 0.6188340807174887, "prompt_id": "89ab6e8e7235950e", "reading_canonical_log_prob": -245.3681182861328, "reading_distances": [30, 18, 26, 22], "reading_log_probs": [-249.03228759765625, -245.75863647460938, -246.73638916015625, -246.21217346191406], "reading_logprob_std": 1.4542318493677053, "tokenization_path_logprob_std": 222.91481451674812, "trajectory_balance_rmse": 4858.831171801854} {"canonical_length": 188, "canonical_nll": 1.4227705610559342, "generation_rouge_l_f1": 0.25352112676056343, "generation_token_f1": 0.4553990610328638, "prompt_id": "3d38b4ece7a08884", "reading_canonical_log_prob": -267.4808654785156, "reading_distances": [26, 21, 21, 29], "reading_log_probs": [-270.6415100097656, -269.6282043457031, -269.30511474609375, -271.2972412109375], "reading_logprob_std": 0.9174858253489109, "tokenization_path_logprob_std": 62.63473566538263, "trajectory_balance_rmse": 3999.9895126613264} {"canonical_length": 169, "canonical_nll": 1.8017904970067493, "generation_rouge_l_f1": 0.2684931506849315, "generation_token_f1": 0.526027397260274, "prompt_id": "49135037e7b65f9f", "reading_canonical_log_prob": -304.5025939941406, "reading_distances": [10, 15, 17, 16], "reading_log_probs": [-307.51116943359375, -314.60186767578125, -318.4444580078125, -312.7666320800781], "reading_logprob_std": 4.544281989991234, "tokenization_path_logprob_std": 217.73849290960322, "trajectory_balance_rmse": 4224.694531695396} {"canonical_length": 250, "canonical_nll": 1.35721923828125, "generation_rouge_l_f1": 0.26719056974459726, "generation_token_f1": 0.43222003929273084, "prompt_id": "daaf9a26e9fd85ac", "reading_canonical_log_prob": -339.3048095703125, "reading_distances": [37, 30, 33, 34], "reading_log_probs": [-335.9688415527344, -342.4467468261719, -337.09344482421875, -340.7320251464844], "reading_logprob_std": 3.0379994338670224, "tokenization_path_logprob_std": 163.80506815317037, "trajectory_balance_rmse": 6346.543430418661} {"canonical_length": 32, "canonical_nll": 1.5981435775756836, "generation_rouge_l_f1": 0.24427480916030533, "generation_token_f1": 0.32061068702290074, "prompt_id": "06a9af54aedbd3a8", "reading_canonical_log_prob": -51.140594482421875, "reading_distances": [43, 44, 45, 39], "reading_log_probs": [-54.764469146728516, -56.283653259277344, -55.21186065673828, -53.91869354248047], "reading_logprob_std": 0.9847787794778254, "tokenization_path_logprob_std": 54.314385825389394, "trajectory_balance_rmse": 699.9881643623504} {"canonical_length": 48, "canonical_nll": 0.5147873560587565, "generation_rouge_l_f1": 0.1842105263157895, "generation_token_f1": 0.2697368421052631, "prompt_id": "4148986bb9e9364b", "reading_canonical_log_prob": -24.709793090820312, "reading_distances": [11, 9, 6, 10], "reading_log_probs": [-25.111188888549805, -24.548873901367188, -22.238182067871094, -24.706069946289062], "reading_logprob_std": 1.2970789496731365, "tokenization_path_logprob_std": 132.84148731409684, "trajectory_balance_rmse": 991.8437901871389} {"canonical_length": 19, "canonical_nll": 0.6590744821648848, "generation_rouge_l_f1": 0.6363636363636365, "generation_token_f1": 0.6363636363636365, "prompt_id": "1c5e2235e187365f", "reading_canonical_log_prob": -12.522415161132812, "reading_distances": [23, 24, 19, 15], "reading_log_probs": [-16.252342224121094, -17.442787170410156, -14.581331253051758, -12.34151554107666], "reading_logprob_std": 2.2123079729705792, "tokenization_path_logprob_std": 50.26113615358611, "trajectory_balance_rmse": 288.4815913171505} {"canonical_length": 369, "canonical_nll": 1.0914046951748815, "generation_rouge_l_f1": 0.2649434571890145, "generation_token_f1": 0.5460420032310178, "prompt_id": "6125f7b585f4ca3f", "reading_canonical_log_prob": -402.72833251953125, "reading_distances": [32, 19, 29, 26], "reading_log_probs": [-406.07025146484375, -407.28607177734375, -414.111572265625, -408.56341552734375], "reading_logprob_std": 3.5515024222108402, "tokenization_path_logprob_std": 420.30443755243886, "trajectory_balance_rmse": 7537.849272308401} {"canonical_length": 21, "canonical_nll": 0.623191651843843, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.18181818181818182, "prompt_id": "bac6bf1b8815878e", "reading_canonical_log_prob": -13.087024688720703, "reading_distances": [19, 20, 17, 16], "reading_log_probs": [-12.249857902526855, -12.305549621582031, -13.333539009094238, -12.39435863494873], "reading_logprob_std": 0.5119452906029015, "tokenization_path_logprob_std": 70.40303445126773, "trajectory_balance_rmse": 382.20416228712156} {"canonical_length": 81, "canonical_nll": 2.8134479287229937, "generation_rouge_l_f1": 0.07894736842105264, "generation_token_f1": 0.10526315789473685, "prompt_id": "3e2b40881e16b7da", "reading_canonical_log_prob": -227.8892822265625, "reading_distances": [22, 14, 19, 18], "reading_log_probs": [-225.22787475585938, -226.7690887451172, -225.85499572753906, -225.71531677246094], "reading_logprob_std": 0.6436699865475642, "tokenization_path_logprob_std": 91.39987729875752, "trajectory_balance_rmse": 1533.2278947925572} {"canonical_length": 32, "canonical_nll": 1.2973930835723877, "generation_rouge_l_f1": 0.18348623853211007, "generation_token_f1": 0.18960244648318042, "prompt_id": "e01d08fb030556a3", "reading_canonical_log_prob": -41.516578674316406, "reading_distances": [32, 27, 24, 30], "reading_log_probs": [-43.3019905090332, -44.24224853515625, -41.08909606933594, -44.52953338623047], "reading_logprob_std": 1.5585613147530086, "tokenization_path_logprob_std": 106.98160814880156, "trajectory_balance_rmse": 698.9180493086932} {"canonical_length": 64, "canonical_nll": 1.4368412494659424, "generation_rouge_l_f1": 0.1370716510903427, "generation_token_f1": 0.23052959501557635, "prompt_id": "68a9acf034dc69e5", "reading_canonical_log_prob": -91.95783996582031, "reading_distances": [28, 30, 23, 21], "reading_log_probs": [-97.62751770019531, -102.76956176757812, -98.23662567138672, -96.91020202636719], "reading_logprob_std": 2.645202849854502, "tokenization_path_logprob_std": 89.75265143584524, "trajectory_balance_rmse": 1420.7366839237131} {"canonical_length": 196, "canonical_nll": 1.0268665235869738, "generation_rouge_l_f1": 0.20754716981132076, "generation_token_f1": 0.3333333333333333, "prompt_id": "12fd41da356aa8b2", "reading_canonical_log_prob": -201.26583862304688, "reading_distances": [32, 31, 33, 31], "reading_log_probs": [-202.3257293701172, -212.11285400390625, -206.49330139160156, -208.2574462890625], "reading_logprob_std": 4.060967435922307, "tokenization_path_logprob_std": 165.36307832788964, "trajectory_balance_rmse": 3544.0983759228493} {"canonical_length": 404, "canonical_nll": 0.9039802173576733, "generation_rouge_l_f1": 0.16534181240063595, "generation_token_f1": 0.3243243243243243, "prompt_id": "7a350261a78f28df", "reading_canonical_log_prob": -365.2080078125, "reading_distances": [20, 20, 20, 25], "reading_log_probs": [-360.14349365234375, -360.6820373535156, -361.41094970703125, -364.0481872558594], "reading_logprob_std": 1.7310980758629584, "tokenization_path_logprob_std": 226.32736965668335, "trajectory_balance_rmse": 6623.390287708811} {"canonical_length": 316, "canonical_nll": 1.387033872966525, "generation_rouge_l_f1": 0.1993127147766323, "generation_token_f1": 0.436426116838488, "prompt_id": "732a9ef0a365d23a", "reading_canonical_log_prob": -438.3027038574219, "reading_distances": [20, 24, 26, 24], "reading_log_probs": [-437.026123046875, -442.943115234375, -438.19097900390625, -445.441650390625], "reading_logprob_std": 3.964283233438971, "tokenization_path_logprob_std": 108.33172011176862, "trajectory_balance_rmse": 7891.3371532759} {"canonical_length": 244, "canonical_nll": 1.6047623431096312, "generation_rouge_l_f1": 0.3166666666666667, "generation_token_f1": 0.5791666666666667, "prompt_id": "26e0cf383cca3d0a", "reading_canonical_log_prob": -391.56201171875, "reading_distances": [27, 33, 27, 30], "reading_log_probs": [-387.95703125, -387.5565490722656, -388.1624755859375, -385.49261474609375], "reading_logprob_std": 1.2258014657571814, "tokenization_path_logprob_std": 38.54272387738561, "trajectory_balance_rmse": 5526.984147141184} {"canonical_length": 123, "canonical_nll": 1.6950609160632621, "generation_rouge_l_f1": 0.25545171339563866, "generation_token_f1": 0.4299065420560747, "prompt_id": "92f09a8894ee3f5f", "reading_canonical_log_prob": -208.49249267578125, "reading_distances": [14, 20, 14, 18], "reading_log_probs": [-228.0850830078125, -234.5413055419922, -222.78955078125, -225.26747131347656], "reading_logprob_std": 5.065506210736452, "tokenization_path_logprob_std": 118.44748461488142, "trajectory_balance_rmse": 2816.381046148396} {"canonical_length": 114, "canonical_nll": 1.314373350979989, "generation_rouge_l_f1": 0.22888283378746593, "generation_token_f1": 0.3487738419618529, "prompt_id": "eb8e055ee1d3ca60", "reading_canonical_log_prob": -149.83856201171875, "reading_distances": [21, 21, 17, 21], "reading_log_probs": [-154.914794921875, -155.56668090820312, -152.21829223632812, -155.117919921875], "reading_logprob_std": 1.5154283133143813, "tokenization_path_logprob_std": 150.32346455380437, "trajectory_balance_rmse": 2552.7133182233356} {"canonical_length": 221, "canonical_nll": 1.6768544745121607, "generation_rouge_l_f1": 0.20722891566265061, "generation_token_f1": 0.453012048192771, "prompt_id": "5a05098dc8854389", "reading_canonical_log_prob": -370.5848388671875, "reading_distances": [22, 20, 19, 25], "reading_log_probs": [-377.41949462890625, -377.97845458984375, -375.92913818359375, -381.72119140625], "reading_logprob_std": 2.4629564023769657, "tokenization_path_logprob_std": 44.51366204570803, "trajectory_balance_rmse": 4642.95267742826} {"canonical_length": 6, "canonical_nll": 1.2702617645263672, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "0744be99f0e84455", "reading_canonical_log_prob": -7.621570587158203, "reading_distances": [20, 24, 17, 15], "reading_log_probs": [-6.016844749450684, -5.955646514892578, -6.7825751304626465, -7.668698310852051], "reading_logprob_std": 0.802199066646603, "tokenization_path_logprob_std": 45.23871420610429, "trajectory_balance_rmse": 153.59481511030725} {"canonical_length": 3, "canonical_nll": 2.6159515380859375, "generation_rouge_l_f1": 0.04938271604938272, "generation_token_f1": 0.04938271604938272, "prompt_id": "1e2019bd6514f4f1", "reading_canonical_log_prob": -7.8478546142578125, "reading_distances": [45, 55, 50, 54], "reading_log_probs": [-7.736717224121094, -8.472345352172852, -7.403501987457275, -8.129349708557129], "reading_logprob_std": 0.464871708881601, "tokenization_path_logprob_std": 14.406730897199347, "trajectory_balance_rmse": 34.6283754685387} {"canonical_length": 17, "canonical_nll": 0.8860734490787282, "generation_rouge_l_f1": 0.6923076923076923, "generation_token_f1": 0.6923076923076923, "prompt_id": "982c9409d78c0796", "reading_canonical_log_prob": -15.063248634338379, "reading_distances": [16, 16, 19, 17], "reading_log_probs": [-22.868070602416992, -21.503765106201172, -17.03420639038086, -21.199281692504883], "reading_logprob_std": 2.5182281319258206, "tokenization_path_logprob_std": 56.10671123748049, "trajectory_balance_rmse": 419.7704492201815} {"canonical_length": 185, "canonical_nll": 1.512831569362331, "generation_rouge_l_f1": 0.2533936651583711, "generation_token_f1": 0.4751131221719456, "prompt_id": "f2e5fb29aaf06273", "reading_canonical_log_prob": -279.87384033203125, "reading_distances": [35, 39, 36, 37], "reading_log_probs": [-284.1559143066406, -296.70574951171875, -291.6110534667969, -287.165771484375], "reading_logprob_std": 5.468579525021872, "tokenization_path_logprob_std": 265.6172591236739, "trajectory_balance_rmse": 3755.485229692831} {"canonical_length": 309, "canonical_nll": 1.0432956535067759, "generation_rouge_l_f1": 0.20205479452054792, "generation_token_f1": 0.4109589041095891, "prompt_id": "3aa4d1f8b5499bdf", "reading_canonical_log_prob": -322.37835693359375, "reading_distances": [65, 56, 54, 65], "reading_log_probs": [-362.96990966796875, -356.049560546875, -356.0225830078125, -360.25537109375], "reading_logprob_std": 3.405037499499035, "tokenization_path_logprob_std": 279.77232335563775, "trajectory_balance_rmse": 5224.792162130389} {"canonical_length": 253, "canonical_nll": 1.6282035013432559, "generation_rouge_l_f1": 0.1937984496124031, "generation_token_f1": 0.4922480620155039, "prompt_id": "aa727754a09ad816", "reading_canonical_log_prob": -411.93548583984375, "reading_distances": [24, 22, 21, 24], "reading_log_probs": [-429.93426513671875, -423.21905517578125, -431.2757873535156, -432.71624755859375], "reading_logprob_std": 4.201345766691522, "tokenization_path_logprob_std": 126.5346472222013, "trajectory_balance_rmse": 5518.693870385308} {"canonical_length": 218, "canonical_nll": 1.314598713446101, "generation_rouge_l_f1": 0.2401656314699793, "generation_token_f1": 0.40579710144927533, "prompt_id": "4d9af47029652710", "reading_canonical_log_prob": -286.58251953125, "reading_distances": [22, 24, 22, 19], "reading_log_probs": [-312.7008056640625, -306.85791015625, -314.4905700683594, -311.0486145019531], "reading_logprob_std": 3.26265457542356, "tokenization_path_logprob_std": 204.57254386744998, "trajectory_balance_rmse": 4960.571538340245} {"canonical_length": 286, "canonical_nll": 1.2409540990015844, "generation_rouge_l_f1": 0.18355640535372844, "generation_token_f1": 0.3441682600382409, "prompt_id": "4d68d14019e3f5eb", "reading_canonical_log_prob": -354.9128723144531, "reading_distances": [25, 24, 24, 22], "reading_log_probs": [-349.87762451171875, -354.5746765136719, -350.7348327636719, -355.7185363769531], "reading_logprob_std": 2.8548711592349276, "tokenization_path_logprob_std": 306.65706060751984, "trajectory_balance_rmse": 5017.312171333582} {"canonical_length": 104, "canonical_nll": 1.4029714144193208, "generation_rouge_l_f1": 0.12631578947368421, "generation_token_f1": 0.30526315789473685, "prompt_id": "9752f3aa2c81e0d1", "reading_canonical_log_prob": -145.90902709960938, "reading_distances": [23, 32, 17, 28], "reading_log_probs": [-147.552734375, -143.25674438476562, -148.1270751953125, -138.74208068847656], "reading_logprob_std": 4.364568021255688, "tokenization_path_logprob_std": 162.69839830718246, "trajectory_balance_rmse": 2196.901676953811} {"canonical_length": 129, "canonical_nll": 0.3522415456845779, "generation_rouge_l_f1": 0.43526170798898073, "generation_token_f1": 0.5509641873278237, "prompt_id": "46cccd71028358c4", "reading_canonical_log_prob": -45.43915939331055, "reading_distances": [20, 25, 24, 25], "reading_log_probs": [-49.352088928222656, -49.78082275390625, -49.72364807128906, -51.02932357788086], "reading_logprob_std": 0.7303992481517105, "tokenization_path_logprob_std": 73.19909423531722, "trajectory_balance_rmse": 1230.5071965003729} {"canonical_length": 3, "canonical_nll": 1.4465166727701824, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "f385579e0d438649", "reading_canonical_log_prob": -4.339550018310547, "reading_distances": [15, 11, 14, 12], "reading_log_probs": [-6.023060321807861, -5.548548221588135, -6.2537736892700195, -5.436739921569824], "reading_logprob_std": 0.3872499447706493, "tokenization_path_logprob_std": 13.498130997637052, "trajectory_balance_rmse": 33.943749264974684} {"canonical_length": 3, "canonical_nll": 2.82985782623291, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "f015c3c34e66dddf", "reading_canonical_log_prob": -8.48957347869873, "reading_distances": [26, 19, 21, 22], "reading_log_probs": [-6.978287696838379, -7.511871337890625, -7.628276824951172, -6.437513828277588], "reading_logprob_std": 0.5466065431146055, "tokenization_path_logprob_std": 19.64057281028665, "trajectory_balance_rmse": 35.46608178247034} {"canonical_length": 47, "canonical_nll": 0.7251724892474235, "generation_rouge_l_f1": 0.22754491017964074, "generation_token_f1": 0.31137724550898205, "prompt_id": "f91ba1a01be59af5", "reading_canonical_log_prob": -34.083106994628906, "reading_distances": [25, 28, 29, 28], "reading_log_probs": [-35.65321731567383, -35.150455474853516, -36.569400787353516, -35.16603088378906], "reading_logprob_std": 0.6653700550092668, "tokenization_path_logprob_std": 98.5388022335577, "trajectory_balance_rmse": 739.5796519987091} {"canonical_length": 179, "canonical_nll": 1.419406826935667, "generation_rouge_l_f1": 0.24537037037037038, "generation_token_f1": 0.4351851851851851, "prompt_id": "0764d55b71d1b589", "reading_canonical_log_prob": -254.07382202148438, "reading_distances": [14, 15, 18, 17], "reading_log_probs": [-258.32330322265625, -255.24473571777344, -255.2236785888672, -257.8697509765625], "reading_logprob_std": 1.6629243314345818, "tokenization_path_logprob_std": 181.41005070549386, "trajectory_balance_rmse": 3794.866942796406} {"canonical_length": 75, "canonical_nll": 1.5864464314778646, "generation_rouge_l_f1": 0.2530120481927711, "generation_token_f1": 0.4819277108433735, "prompt_id": "a40186552cfde2fc", "reading_canonical_log_prob": -118.98348236083984, "reading_distances": [12, 12, 6, 5], "reading_log_probs": [-126.83995056152344, -125.89086151123047, -122.71443939208984, -121.0546646118164], "reading_logprob_std": 2.7022129234586707, "tokenization_path_logprob_std": 60.09863904001264, "trajectory_balance_rmse": 1405.59122327741} {"canonical_length": 263, "canonical_nll": 1.085537870573907, "generation_rouge_l_f1": 0.3414634146341463, "generation_token_f1": 0.6003752345215759, "prompt_id": "a3ec61d54b8539e4", "reading_canonical_log_prob": -285.4964599609375, "reading_distances": [22, 25, 26, 23], "reading_log_probs": [-286.0145568847656, -289.12982177734375, -288.2120361328125, -286.18365478515625], "reading_logprob_std": 1.5329423185257587, "tokenization_path_logprob_std": 111.83789621206363, "trajectory_balance_rmse": 5604.897094304917} {"canonical_length": 210, "canonical_nll": 1.6185782296316964, "generation_rouge_l_f1": 0.20703933747412012, "generation_token_f1": 0.38923395445134573, "prompt_id": "ca23abf600fc132e", "reading_canonical_log_prob": -339.90142822265625, "reading_distances": [43, 45, 50, 45], "reading_log_probs": [-353.79913330078125, -352.76611328125, -355.3511962890625, -360.2409362792969], "reading_logprob_std": 3.309557443675643, "tokenization_path_logprob_std": 110.000496453823, "trajectory_balance_rmse": 5095.695571652786} {"canonical_length": 132, "canonical_nll": 1.496706991484671, "generation_rouge_l_f1": 0.2118863049095607, "generation_token_f1": 0.3462532299741602, "prompt_id": "a83cfaa0169d4436", "reading_canonical_log_prob": -197.56532287597656, "reading_distances": [40, 56, 39, 47], "reading_log_probs": [-218.57232666015625, -228.35923767089844, -214.625, -225.36557006835938], "reading_logprob_std": 6.261418545384485, "tokenization_path_logprob_std": 88.67097024068808, "trajectory_balance_rmse": 996.2002959527597} {"canonical_length": 115, "canonical_nll": 1.5574807871942935, "generation_rouge_l_f1": 0.2215568862275449, "generation_token_f1": 0.36526946107784436, "prompt_id": "bd9008593d2eaa20", "reading_canonical_log_prob": -179.11029052734375, "reading_distances": [41, 42, 41, 41], "reading_log_probs": [-173.8137664794922, -175.93788146972656, -178.29075622558594, -173.23062133789062], "reading_logprob_std": 2.297926638921856, "tokenization_path_logprob_std": 100.64580885951433, "trajectory_balance_rmse": 2839.744810632563} {"canonical_length": 7, "canonical_nll": 0.8916527203151158, "generation_rouge_l_f1": 0.13953488372093026, "generation_token_f1": 0.13953488372093026, "prompt_id": "48bde41f90ff68c1", "reading_canonical_log_prob": -6.2415690422058105, "reading_distances": [31, 34, 27, 35], "reading_log_probs": [-5.797557830810547, -7.52967643737793, -7.143650531768799, -7.690412998199463], "reading_logprob_std": 0.8596948367485001, "tokenization_path_logprob_std": 39.508216675725166, "trajectory_balance_rmse": 154.9764155089969} {"canonical_length": 130, "canonical_nll": 0.7050266559307392, "generation_rouge_l_f1": 0.36170212765957444, "generation_token_f1": 0.5425531914893617, "prompt_id": "8dbb7f1da28c2195", "reading_canonical_log_prob": -91.6534652709961, "reading_distances": [11, 9, 18, 15], "reading_log_probs": [-84.99537658691406, -85.7732925415039, -87.84649658203125, -88.19454193115234], "reading_logprob_std": 1.5612616832846642, "tokenization_path_logprob_std": 121.13761130935897, "trajectory_balance_rmse": 1594.1495440169583} {"canonical_length": 260, "canonical_nll": 2.0958726149338944, "generation_rouge_l_f1": 0.2692307692307692, "generation_token_f1": 0.48846153846153845, "prompt_id": "d859a6a8a580026f", "reading_canonical_log_prob": -544.9268798828125, "reading_distances": [38, 35, 33, 31], "reading_log_probs": [-560.8668823242188, -564.7914428710938, -564.6492309570312, -568.0185546875], "reading_logprob_std": 2.924630625719215, "tokenization_path_logprob_std": 221.50770350275408, "trajectory_balance_rmse": 5783.323584620784} {"canonical_length": 242, "canonical_nll": 1.3735919038126292, "generation_rouge_l_f1": 0.2204724409448819, "generation_token_f1": 0.45275590551181105, "prompt_id": "b4d85be43e32fd7f", "reading_canonical_log_prob": -332.40924072265625, "reading_distances": [37, 43, 34, 38], "reading_log_probs": [-343.82342529296875, -347.76422119140625, -356.06591796875, -348.18133544921875], "reading_logprob_std": 5.128829263372995, "tokenization_path_logprob_std": 162.0249687148957, "trajectory_balance_rmse": 5601.399814283968} {"canonical_length": 19, "canonical_nll": 1.3555649205258018, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3703703703703704, "prompt_id": "64c4e6c526576026", "reading_canonical_log_prob": -25.755733489990234, "reading_distances": [27, 28, 30, 29], "reading_log_probs": [-39.86664581298828, -33.573238372802734, -31.886165618896484, -38.4946174621582], "reading_logprob_std": 3.8287870742005556, "tokenization_path_logprob_std": 15.373242376716608, "trajectory_balance_rmse": 439.7267962851603} {"canonical_length": 56, "canonical_nll": 1.3955659866333008, "generation_rouge_l_f1": 0.07761194029850747, "generation_token_f1": 0.1492537313432836, "prompt_id": "505a139a155b8b27", "reading_canonical_log_prob": -78.15169525146484, "reading_distances": [26, 29, 18, 32], "reading_log_probs": [-84.1380615234375, -83.57157897949219, -86.54705810546875, -85.80043029785156], "reading_logprob_std": 1.3924302573367224, "tokenization_path_logprob_std": 86.64026753400155, "trajectory_balance_rmse": 1175.2597820279852} {"canonical_length": 171, "canonical_nll": 1.202891834995203, "generation_rouge_l_f1": 0.3877551020408163, "generation_token_f1": 0.5969387755102041, "prompt_id": "ca5c2935eb3033fb", "reading_canonical_log_prob": -205.6945037841797, "reading_distances": [18, 24, 21, 20], "reading_log_probs": [-204.35150146484375, -206.50540161132812, -205.76773071289062, -203.42599487304688], "reading_logprob_std": 1.3848039732904005, "tokenization_path_logprob_std": 213.25199939925088, "trajectory_balance_rmse": 3676.843248788093} {"canonical_length": 78, "canonical_nll": 0.7798337691869491, "generation_rouge_l_f1": 0.30927835051546393, "generation_token_f1": 0.4329896907216495, "prompt_id": "f8aa35c9d5781c5a", "reading_canonical_log_prob": -60.82703399658203, "reading_distances": [22, 28, 23, 24], "reading_log_probs": [-77.68382263183594, -76.70021057128906, -71.42616271972656, -71.01527404785156], "reading_logprob_std": 3.474889134862121, "tokenization_path_logprob_std": 78.31122801602172, "trajectory_balance_rmse": 1365.55659053146} {"canonical_length": 339, "canonical_nll": 1.1748068480365044, "generation_rouge_l_f1": 0.26465661641541044, "generation_token_f1": 0.4857621440536013, "prompt_id": "de613d5bcfdf1d4f", "reading_canonical_log_prob": -398.259521484375, "reading_distances": [30, 37, 37, 36], "reading_log_probs": [-400.97540283203125, -402.85205078125, -404.82550048828125, -399.0584411621094], "reading_logprob_std": 2.47596365906493, "tokenization_path_logprob_std": 323.94665315014, "trajectory_balance_rmse": 7514.04416342832} {"canonical_length": 19, "canonical_nll": 1.9584420856676603, "generation_rouge_l_f1": 0.12121212121212122, "generation_token_f1": 0.24242424242424243, "prompt_id": "cb6d19eef2166daa", "reading_canonical_log_prob": -37.21039962768555, "reading_distances": [23, 30, 27, 26], "reading_log_probs": [-39.59358215332031, -46.21973419189453, -37.2214469909668, -41.9531364440918], "reading_logprob_std": 3.8369076794842543, "tokenization_path_logprob_std": 89.57718276512963, "trajectory_balance_rmse": 478.98189846692804} {"canonical_length": 389, "canonical_nll": 1.1692366023909464, "generation_rouge_l_f1": 0.1806451612903226, "generation_token_f1": 0.37741935483870964, "prompt_id": "980fa61309c5ee0c", "reading_canonical_log_prob": -454.8330383300781, "reading_distances": [20, 23, 21, 26], "reading_log_probs": [-470.001220703125, -470.5516357421875, -466.465576171875, -463.1095275878906], "reading_logprob_std": 3.4598060115610894, "tokenization_path_logprob_std": 151.40607521800268, "trajectory_balance_rmse": 7573.439564881811} {"canonical_length": 14, "canonical_nll": 2.905287061418806, "generation_rouge_l_f1": 0.30769230769230765, "generation_token_f1": 0.30769230769230765, "prompt_id": "0f4edb0d2ff1676b", "reading_canonical_log_prob": -40.67401885986328, "reading_distances": [14, 13, 14, 17], "reading_log_probs": [-36.72511672973633, -44.58092498779297, -35.33143615722656, -42.509666442871094], "reading_logprob_std": 4.458019271770242, "tokenization_path_logprob_std": 108.83263296654349, "trajectory_balance_rmse": 332.85436893435315} {"canonical_length": 80, "canonical_nll": 0.9632139205932617, "generation_rouge_l_f1": 0.1750841750841751, "generation_token_f1": 0.2962962962962963, "prompt_id": "7bc8145a69124b53", "reading_canonical_log_prob": -77.05711364746094, "reading_distances": [25, 28, 26, 23], "reading_log_probs": [-84.14311218261719, -80.46277618408203, -81.50479888916016, -84.32122802734375], "reading_logprob_std": 1.924470281786787, "tokenization_path_logprob_std": 70.30110066409091, "trajectory_balance_rmse": 1203.5060127853928} {"canonical_length": 12, "canonical_nll": 0.47668131192525226, "generation_rouge_l_f1": 0.16666666666666669, "generation_token_f1": 0.16666666666666669, "prompt_id": "5167b2d9f21714f8", "reading_canonical_log_prob": -5.720175743103027, "reading_distances": [27, 20, 22, 22], "reading_log_probs": [-5.169466495513916, -4.3477325439453125, -5.066494464874268, -4.661626815795898], "reading_logprob_std": 0.37890381284910113, "tokenization_path_logprob_std": 10.53474180377651, "trajectory_balance_rmse": 235.18249234419991} {"canonical_length": 130, "canonical_nll": 1.1914283165564903, "generation_rouge_l_f1": 0.27906976744186046, "generation_token_f1": 0.4909560723514212, "prompt_id": "786583b4f8ed0931", "reading_canonical_log_prob": -154.88568115234375, "reading_distances": [23, 23, 21, 20], "reading_log_probs": [-160.83505249023438, -160.4123077392578, -160.75828552246094, -159.11172485351562], "reading_logprob_std": 0.7998355529404808, "tokenization_path_logprob_std": 135.3975562025562, "trajectory_balance_rmse": 3247.4662626313257} {"canonical_length": 351, "canonical_nll": 1.191967216991631, "generation_rouge_l_f1": 0.2233502538071066, "generation_token_f1": 0.38917089678510997, "prompt_id": "99a05025a306c903", "reading_canonical_log_prob": -418.3804931640625, "reading_distances": [18, 15, 13, 19], "reading_log_probs": [-430.235107421875, -429.1170654296875, -430.00531005859375, -424.0945129394531], "reading_logprob_std": 2.886206563749914, "tokenization_path_logprob_std": 427.8124487701777, "trajectory_balance_rmse": 6476.676150826765} {"canonical_length": 32, "canonical_nll": 1.1330312490463257, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "4d4e1a944b4c608a", "reading_canonical_log_prob": -36.25699996948242, "reading_distances": [12, 11, 14, 14], "reading_log_probs": [-40.8856201171875, -38.61418914794922, -43.100196838378906, -38.38993835449219], "reading_logprob_std": 2.2108306237832656, "tokenization_path_logprob_std": 34.814902841245754, "trajectory_balance_rmse": 576.2047360912934} {"canonical_length": 375, "canonical_nll": 1.3036437174479167, "generation_rouge_l_f1": 0.283987915407855, "generation_token_f1": 0.5438066465256798, "prompt_id": "8de18ae965c9405f", "reading_canonical_log_prob": -488.86639404296875, "reading_distances": [29, 25, 28, 29], "reading_log_probs": [-498.1494140625, -498.3665466308594, -495.1833801269531, -502.5911865234375], "reading_logprob_std": 3.0472606998629232, "tokenization_path_logprob_std": 295.9076515547213, "trajectory_balance_rmse": 7161.524112983343} {"canonical_length": 154, "canonical_nll": 1.059421985180347, "generation_rouge_l_f1": 0.3744075829383886, "generation_token_f1": 0.5260663507109005, "prompt_id": "577845af12167b4c", "reading_canonical_log_prob": -163.15098571777344, "reading_distances": [9, 11, 10, 8], "reading_log_probs": [-177.0057373046875, -175.59213256835938, -173.98681640625, -177.029052734375], "reading_logprob_std": 1.4436552788797645, "tokenization_path_logprob_std": 69.25297357742097, "trajectory_balance_rmse": 2872.414346109661} {"canonical_length": 94, "canonical_nll": 1.1147704428814826, "generation_rouge_l_f1": 0.19373219373219372, "generation_token_f1": 0.3646723646723647, "prompt_id": "aea07a40db5e5411", "reading_canonical_log_prob": -104.78842163085938, "reading_distances": [18, 24, 30, 29], "reading_log_probs": [-115.05433654785156, -111.40818786621094, -114.6631088256836, -117.95970916748047], "reading_logprob_std": 2.681311324474305, "tokenization_path_logprob_std": 92.5415608900638, "trajectory_balance_rmse": 2059.4155211749407} {"canonical_length": 111, "canonical_nll": 1.5192890339069538, "generation_rouge_l_f1": 0.14775725593667546, "generation_token_f1": 0.32189973614775724, "prompt_id": "fdb4e9b8596975db", "reading_canonical_log_prob": -168.64108276367188, "reading_distances": [24, 23, 23, 20], "reading_log_probs": [-168.70547485351562, -168.39443969726562, -168.67660522460938, -166.5957794189453], "reading_logprob_std": 1.0080105557354646, "tokenization_path_logprob_std": 141.17221268305795, "trajectory_balance_rmse": 2667.895099573544} {"canonical_length": 364, "canonical_nll": 1.6453238686362466, "generation_rouge_l_f1": 0.2673434856175973, "generation_token_f1": 0.5042301184433163, "prompt_id": "219a3ab27e90cb77", "reading_canonical_log_prob": -598.8978881835938, "reading_distances": [13, 15, 18, 13], "reading_log_probs": [-606.8853759765625, -606.630126953125, -606.325439453125, -604.9329833984375], "reading_logprob_std": 0.8709469145453177, "tokenization_path_logprob_std": 413.1124740974518, "trajectory_balance_rmse": 8166.33024129458} {"canonical_length": 42, "canonical_nll": 1.3088213602701824, "generation_rouge_l_f1": 0.1388888888888889, "generation_token_f1": 0.18750000000000003, "prompt_id": "c1bdd5a5ad1d792e", "reading_canonical_log_prob": -54.970497131347656, "reading_distances": [14, 21, 19, 18], "reading_log_probs": [-55.172279357910156, -55.14170837402344, -52.954551696777344, -55.55127716064453], "reading_logprob_std": 1.181710411331093, "tokenization_path_logprob_std": 29.10699988085212, "trajectory_balance_rmse": 534.9717694065935} {"canonical_length": 262, "canonical_nll": 1.7858928651300094, "generation_rouge_l_f1": 0.26591760299625467, "generation_token_f1": 0.5056179775280899, "prompt_id": "52d1f167e2d1bb05", "reading_canonical_log_prob": -467.9039306640625, "reading_distances": [30, 24, 24, 28], "reading_log_probs": [-479.5964660644531, -478.8572692871094, -479.214111328125, -484.058349609375], "reading_logprob_std": 2.436634103890299, "tokenization_path_logprob_std": 297.49943209398253, "trajectory_balance_rmse": 5779.154277871616} {"canonical_length": 17, "canonical_nll": 1.9816169738769531, "generation_rouge_l_f1": 0.5185185185185185, "generation_token_f1": 0.5185185185185185, "prompt_id": "c86b63feb26be380", "reading_canonical_log_prob": -33.6874885559082, "reading_distances": [37, 37, 36, 36], "reading_log_probs": [-30.32654571533203, -32.673095703125, -37.15692138671875, -30.410118103027344], "reading_logprob_std": 3.2004222108144793, "tokenization_path_logprob_std": 35.59641073064653, "trajectory_balance_rmse": 394.7013348390615} {"canonical_length": 10, "canonical_nll": 2.0308353424072267, "generation_rouge_l_f1": 0.11764705882352941, "generation_token_f1": 0.11764705882352941, "prompt_id": "197710f492b62fe4", "reading_canonical_log_prob": -20.308353424072266, "reading_distances": [14, 7, 9, 16], "reading_log_probs": [-26.159889221191406, -26.451244354248047, -26.29648208618164, -25.969839096069336], "reading_logprob_std": 0.2045447179290885, "tokenization_path_logprob_std": 49.42635742055952, "trajectory_balance_rmse": 209.82883386020345} {"canonical_length": 25, "canonical_nll": 1.179532012939453, "generation_rouge_l_f1": 0.18918918918918917, "generation_token_f1": 0.2162162162162162, "prompt_id": "23c24264db31cb91", "reading_canonical_log_prob": -29.488300323486328, "reading_distances": [31, 23, 27, 30], "reading_log_probs": [-29.865306854248047, -29.814311981201172, -29.201061248779297, -30.677940368652344], "reading_logprob_std": 0.6060320616751749, "tokenization_path_logprob_std": 73.85345318375903, "trajectory_balance_rmse": 421.08506968910746} {"canonical_length": 169, "canonical_nll": 0.8781460192076553, "generation_rouge_l_f1": 0.4271186440677966, "generation_token_f1": 0.5559322033898305, "prompt_id": "742a95dae13136e2", "reading_canonical_log_prob": -148.40667724609375, "reading_distances": [19, 18, 18, 16], "reading_log_probs": [-152.3227996826172, -153.23062133789062, -151.15933227539062, -154.43777465820312], "reading_logprob_std": 1.3888411743071862, "tokenization_path_logprob_std": 169.78037436631504, "trajectory_balance_rmse": 2638.473524520655} {"canonical_length": 19, "canonical_nll": 2.0900142067357113, "generation_rouge_l_f1": 0.45714285714285713, "generation_token_f1": 0.5142857142857142, "prompt_id": "87a7422c0b830ad1", "reading_canonical_log_prob": -39.710269927978516, "reading_distances": [48, 51, 43, 43], "reading_log_probs": [-46.2946662902832, -45.870601654052734, -44.759517669677734, -46.3359375], "reading_logprob_std": 0.7345270568587543, "tokenization_path_logprob_std": 55.4559263111949, "trajectory_balance_rmse": 406.073234435507} {"canonical_length": 44, "canonical_nll": 0.8019238385287198, "generation_rouge_l_f1": 0.42500000000000004, "generation_token_f1": 0.5500000000000002, "prompt_id": "73b66ff67ed60622", "reading_canonical_log_prob": -35.28464889526367, "reading_distances": [29, 26, 25, 22], "reading_log_probs": [-38.95512008666992, -40.94438171386719, -37.991188049316406, -39.107059478759766], "reading_logprob_std": 1.233277720681613, "tokenization_path_logprob_std": 92.0283641330875, "trajectory_balance_rmse": 1167.7932330820695} {"canonical_length": 126, "canonical_nll": 1.075755406939794, "generation_rouge_l_f1": 0.29714285714285715, "generation_token_f1": 0.4457142857142857, "prompt_id": "f610e308eb24edb1", "reading_canonical_log_prob": -135.54518127441406, "reading_distances": [23, 21, 24, 25], "reading_log_probs": [-140.96337890625, -138.85401916503906, -140.29168701171875, -140.44447326660156], "reading_logprob_std": 0.9032168578712659, "tokenization_path_logprob_std": 115.95449463906218, "trajectory_balance_rmse": 2089.7829678077246} {"canonical_length": 251, "canonical_nll": 1.212821808469248, "generation_rouge_l_f1": 0.21115537848605578, "generation_token_f1": 0.4701195219123506, "prompt_id": "da58850d3c7fdcfd", "reading_canonical_log_prob": -304.41827392578125, "reading_distances": [18, 23, 25, 22], "reading_log_probs": [-297.1063232421875, -300.933349609375, -300.3934631347656, -297.98663330078125], "reading_logprob_std": 1.8482832021983713, "tokenization_path_logprob_std": 25.96947481411621, "trajectory_balance_rmse": 5441.9734661212115} {"canonical_length": 296, "canonical_nll": 1.3015995541134395, "generation_rouge_l_f1": 0.21550094517958412, "generation_token_f1": 0.43856332703213613, "prompt_id": "0a2b0e3667ca975a", "reading_canonical_log_prob": -385.2734680175781, "reading_distances": [12, 16, 15, 19], "reading_log_probs": [-402.02008056640625, -402.5557861328125, -400.29071044921875, -399.5611267089844], "reading_logprob_std": 1.41288714105121, "tokenization_path_logprob_std": 231.47094984203616, "trajectory_balance_rmse": 6028.489148704944} {"canonical_length": 111, "canonical_nll": 1.0601274645006336, "generation_rouge_l_f1": 0.2507462686567164, "generation_token_f1": 0.4119402985074626, "prompt_id": "b3bdd02a596bc2d5", "reading_canonical_log_prob": -117.67414855957031, "reading_distances": [39, 36, 44, 46], "reading_log_probs": [-119.60747528076172, -115.25917053222656, -116.78125, -118.35335540771484], "reading_logprob_std": 1.8892313591909389, "tokenization_path_logprob_std": 82.23864174292581, "trajectory_balance_rmse": 1465.7045738452314} {"canonical_length": 14, "canonical_nll": 0.2903468608856201, "generation_rouge_l_f1": 0.5217391304347826, "generation_token_f1": 0.5217391304347826, "prompt_id": "3a6cc5e39c6939e7", "reading_canonical_log_prob": -4.064856052398682, "reading_distances": [22, 27, 23, 27], "reading_log_probs": [-5.5707316398620605, -5.084306240081787, -6.729076385498047, -4.24662971496582], "reading_logprob_std": 1.0368674935764994, "tokenization_path_logprob_std": 57.545990661149446, "trajectory_balance_rmse": 298.24199956179865} {"canonical_length": 79, "canonical_nll": 0.8459442718119561, "generation_rouge_l_f1": 0.31671554252199413, "generation_token_f1": 0.404692082111437, "prompt_id": "9431e7948140e38d", "reading_canonical_log_prob": -66.82959747314453, "reading_distances": [36, 34, 37, 37], "reading_log_probs": [-60.5301513671875, -62.442867279052734, -58.95789337158203, -58.08087921142578], "reading_logprob_std": 1.9163887117054208, "tokenization_path_logprob_std": 123.0719503072995, "trajectory_balance_rmse": 1374.0174376091459} {"canonical_length": 56, "canonical_nll": 1.0269077164786202, "generation_rouge_l_f1": 0.21472392638036808, "generation_token_f1": 0.3006134969325153, "prompt_id": "8230c7c1426f816d", "reading_canonical_log_prob": -57.506832122802734, "reading_distances": [24, 23, 25, 21], "reading_log_probs": [-55.31754684448242, -55.25116729736328, -56.49013137817383, -55.60479736328125], "reading_logprob_std": 0.5705097230009839, "tokenization_path_logprob_std": 92.16186102843778, "trajectory_balance_rmse": 1197.2087543130392} {"canonical_length": 101, "canonical_nll": 1.1630946244343672, "generation_rouge_l_f1": 0.2829268292682927, "generation_token_f1": 0.5170731707317074, "prompt_id": "211cd27653ee878e", "reading_canonical_log_prob": -117.4725570678711, "reading_distances": [27, 24, 31, 27], "reading_log_probs": [-132.08726501464844, -131.38345336914062, -131.1021728515625, -130.71646118164062], "reading_logprob_std": 0.5786213032418549, "tokenization_path_logprob_std": 85.18767778614082, "trajectory_balance_rmse": 2141.813030446215} {"canonical_length": 19, "canonical_nll": 2.135331806383635, "generation_rouge_l_f1": 0.125, "generation_token_f1": 0.18749999999999997, "prompt_id": "bd8cff3238489dd7", "reading_canonical_log_prob": -40.57130432128906, "reading_distances": [31, 29, 37, 36], "reading_log_probs": [-53.91114807128906, -49.4501838684082, -52.3900146484375, -56.110748291015625], "reading_logprob_std": 2.7973480286103984, "tokenization_path_logprob_std": 31.87348327886413, "trajectory_balance_rmse": 470.3499321402293} {"canonical_length": 21, "canonical_nll": 1.3588241395496188, "generation_rouge_l_f1": 0.3870967741935484, "generation_token_f1": 0.3870967741935484, "prompt_id": "9444b92ac8847c28", "reading_canonical_log_prob": -28.535306930541992, "reading_distances": [28, 21, 24, 23], "reading_log_probs": [-38.95271301269531, -32.200050354003906, -36.893619537353516, -48.02130126953125], "reading_logprob_std": 6.634785131147234, "tokenization_path_logprob_std": 46.442397302185476, "trajectory_balance_rmse": 501.1094238539552} {"canonical_length": 13, "canonical_nll": 2.259365668663612, "generation_rouge_l_f1": 0.32, "generation_token_f1": 0.32, "prompt_id": "d790284919634e0e", "reading_canonical_log_prob": -29.371753692626953, "reading_distances": [32, 30, 32, 24], "reading_log_probs": [-26.546918869018555, -27.36354637145996, -29.799436569213867, -28.318246841430664], "reading_logprob_std": 1.3970889886427962, "tokenization_path_logprob_std": 23.78534100559624, "trajectory_balance_rmse": 388.0015837290661} {"canonical_length": 86, "canonical_nll": 0.9776165097258812, "generation_rouge_l_f1": 0.10869565217391304, "generation_token_f1": 0.10869565217391304, "prompt_id": "58457960c61ffd91", "reading_canonical_log_prob": -84.07501983642578, "reading_distances": [32, 30, 29, 29], "reading_log_probs": [-86.519287109375, -84.06855773925781, -87.1138916015625, -85.08993530273438], "reading_logprob_std": 1.3788988691002175, "tokenization_path_logprob_std": 119.80995833585973, "trajectory_balance_rmse": 1850.1824968306535} {"canonical_length": 6, "canonical_nll": 1.4948693911234539, "generation_rouge_l_f1": 0.0425531914893617, "generation_token_f1": 0.0425531914893617, "prompt_id": "02cbb58a2f84ec5b", "reading_canonical_log_prob": -8.969216346740723, "reading_distances": [19, 16, 17, 20], "reading_log_probs": [-8.184678077697754, -6.845335006713867, -6.750216484069824, -7.567527770996094], "reading_logprob_std": 0.6727432700218775, "tokenization_path_logprob_std": 18.37277265455202, "trajectory_balance_rmse": 163.70128459275338} {"canonical_length": 50, "canonical_nll": 2.1157522583007813, "generation_rouge_l_f1": 0.21383647798742136, "generation_token_f1": 0.3018867924528302, "prompt_id": "d0f9c73543784d9e", "reading_canonical_log_prob": -105.78761291503906, "reading_distances": [40, 32, 29, 40], "reading_log_probs": [-110.34780883789062, -111.24732971191406, -113.23310089111328, -112.09288024902344], "reading_logprob_std": 1.229419519690043, "tokenization_path_logprob_std": 46.55896887303865, "trajectory_balance_rmse": 497.48430618096637} {"canonical_length": 76, "canonical_nll": 1.5733467905144942, "generation_rouge_l_f1": 0.15902140672782875, "generation_token_f1": 0.28134556574923547, "prompt_id": "e67d8b619feb7d26", "reading_canonical_log_prob": -119.57435607910156, "reading_distances": [6, 10, 9, 8], "reading_log_probs": [-120.87890625, -122.74034118652344, -122.01025390625, -121.240478515625], "reading_logprob_std": 0.8291959793410878, "tokenization_path_logprob_std": 109.04403056459563, "trajectory_balance_rmse": 1963.2596137001428} {"canonical_length": 260, "canonical_nll": 1.2875279353215143, "generation_rouge_l_f1": 0.2819548872180451, "generation_token_f1": 0.518796992481203, "prompt_id": "aa422d282fda03cd", "reading_canonical_log_prob": -334.75726318359375, "reading_distances": [28, 30, 37, 23], "reading_log_probs": [-344.86712646484375, -344.88580322265625, -347.313232421875, -346.0699462890625], "reading_logprob_std": 1.1644357156587841, "tokenization_path_logprob_std": 145.3614083528922, "trajectory_balance_rmse": 3715.318663246823} {"canonical_length": 281, "canonical_nll": 1.5061387031528024, "generation_rouge_l_f1": 0.19734345351043642, "generation_token_f1": 0.364326375711575, "prompt_id": "050b55490a11894a", "reading_canonical_log_prob": -423.2249755859375, "reading_distances": [19, 17, 26, 18], "reading_log_probs": [-429.0363464355469, -434.06402587890625, -429.26953125, -430.13519287109375], "reading_logprob_std": 2.3400797141676355, "tokenization_path_logprob_std": 278.8678312532969, "trajectory_balance_rmse": 6276.163635956876} {"canonical_length": 251, "canonical_nll": 0.8527957262745891, "generation_rouge_l_f1": 0.26953125, "generation_token_f1": 0.44921875, "prompt_id": "2d149cddaf627a2b", "reading_canonical_log_prob": -214.05172729492188, "reading_distances": [26, 29, 33, 35], "reading_log_probs": [-226.15625, -223.35548400878906, -230.3042755126953, -230.23751831054688], "reading_logprob_std": 3.383288745603635, "tokenization_path_logprob_std": 155.25372749158677, "trajectory_balance_rmse": 5498.836345431012} {"canonical_length": 157, "canonical_nll": 1.5555354804749701, "generation_rouge_l_f1": 0.2657004830917874, "generation_token_f1": 0.48792270531400955, "prompt_id": "f7af4006bdd897fe", "reading_canonical_log_prob": -244.2190704345703, "reading_distances": [29, 22, 22, 32], "reading_log_probs": [-246.1629638671875, -243.0739288330078, -245.9693603515625, -251.33673095703125], "reading_logprob_std": 3.4377024243363747, "tokenization_path_logprob_std": 112.93812117355634, "trajectory_balance_rmse": 3664.0864796162887} {"canonical_length": 21, "canonical_nll": 0.6307901654924665, "generation_rouge_l_f1": 0.7096774193548387, "generation_token_f1": 0.7096774193548387, "prompt_id": "2d0d3b6d72c67323", "reading_canonical_log_prob": -13.246593475341797, "reading_distances": [16, 31, 24, 23], "reading_log_probs": [-16.2669734954834, -14.050228118896484, -15.666415214538574, -13.638908386230469], "reading_logprob_std": 1.2607334215097419, "tokenization_path_logprob_std": 26.787344609765395, "trajectory_balance_rmse": 349.5274417642441} {"canonical_length": 7, "canonical_nll": 0.4562579904283796, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.33333333333333337, "prompt_id": "974b052b2f839275", "reading_canonical_log_prob": -3.1938059329986572, "reading_distances": [15, 25, 19, 23], "reading_log_probs": [-3.300539016723633, -3.8656294345855713, -3.6060285568237305, -3.818925380706787], "reading_logprob_std": 0.2575969726524603, "tokenization_path_logprob_std": 32.84239005497364, "trajectory_balance_rmse": 103.62676651256285} {"canonical_length": 71, "canonical_nll": 1.435912871024978, "generation_rouge_l_f1": 0.20353982300884957, "generation_token_f1": 0.3539823008849558, "prompt_id": "7f8aa3c30a8d7e14", "reading_canonical_log_prob": -101.94981384277344, "reading_distances": [32, 30, 32, 37], "reading_log_probs": [-101.53438568115234, -103.47474670410156, -100.7933349609375, -104.0520248413086], "reading_logprob_std": 1.5490573136204502, "tokenization_path_logprob_std": 113.44659428246396, "trajectory_balance_rmse": 1503.9402085226257} {"canonical_length": 12, "canonical_nll": 0.6232759555180868, "generation_rouge_l_f1": 0.75, "generation_token_f1": 0.75, "prompt_id": "537e5de8a8d8abfa", "reading_canonical_log_prob": -7.479311466217041, "reading_distances": [57, 49, 64, 54], "reading_log_probs": [-8.70363712310791, -8.925910949707031, -6.437108993530273, -10.677125930786133], "reading_logprob_std": 1.7397280967014794, "tokenization_path_logprob_std": 28.66172187873586, "trajectory_balance_rmse": 279.2039800062298} {"canonical_length": 235, "canonical_nll": 1.4811115670711437, "generation_rouge_l_f1": 0.2933884297520661, "generation_token_f1": 0.5413223140495869, "prompt_id": "219d66b292bdf703", "reading_canonical_log_prob": -348.06121826171875, "reading_distances": [38, 33, 45, 36], "reading_log_probs": [-357.2311706542969, -355.19622802734375, -355.0118408203125, -355.527587890625], "reading_logprob_std": 1.0156444457145746, "tokenization_path_logprob_std": 217.62037470249143, "trajectory_balance_rmse": 5241.712833268373} {"canonical_length": 233, "canonical_nll": 1.6930090122468482, "generation_rouge_l_f1": 0.1833688699360341, "generation_token_f1": 0.32409381663113, "prompt_id": "b3b45f7b904e5cc3", "reading_canonical_log_prob": -394.4710998535156, "reading_distances": [26, 14, 21, 19], "reading_log_probs": [-404.8982238769531, -405.1630859375, -394.86962890625, -401.02264404296875], "reading_logprob_std": 4.80122467064381, "tokenization_path_logprob_std": 190.97581065890748, "trajectory_balance_rmse": 5337.440190300485} {"canonical_length": 121, "canonical_nll": 1.5712415206530863, "generation_rouge_l_f1": 0.35454545454545455, "generation_token_f1": 0.5363636363636363, "prompt_id": "dd6de628b051978a", "reading_canonical_log_prob": -190.12022399902344, "reading_distances": [38, 39, 33, 35], "reading_log_probs": [-187.84426879882812, -184.47372436523438, -188.04563903808594, -186.41537475585938], "reading_logprob_std": 1.648969381097638, "tokenization_path_logprob_std": 72.40300950487843, "trajectory_balance_rmse": 2618.039004197086} {"canonical_length": 105, "canonical_nll": 0.9777313959030878, "generation_rouge_l_f1": 0.1907514450867052, "generation_token_f1": 0.24855491329479767, "prompt_id": "d2e124ed83b7b0d6", "reading_canonical_log_prob": -102.66179656982422, "reading_distances": [29, 23, 31, 31], "reading_log_probs": [-104.89208221435547, -107.91545104980469, -104.35890197753906, -106.01670837402344], "reading_logprob_std": 1.5730122742580452, "tokenization_path_logprob_std": 126.05353564744965, "trajectory_balance_rmse": 1732.9324714486013} {"canonical_length": 102, "canonical_nll": 0.8767568550857843, "generation_rouge_l_f1": 0.21052631578947367, "generation_token_f1": 0.33436532507739936, "prompt_id": "40e867680ebfa198", "reading_canonical_log_prob": -89.42919921875, "reading_distances": [36, 38, 40, 29], "reading_log_probs": [-101.06377410888672, -112.00315856933594, -111.26568603515625, -106.63459777832031], "reading_logprob_std": 5.046413782982453, "tokenization_path_logprob_std": 88.57690910521036, "trajectory_balance_rmse": 1455.631835412163} {"canonical_length": 217, "canonical_nll": 1.4723867952548964, "generation_rouge_l_f1": 0.19631901840490795, "generation_token_f1": 0.4008179959100205, "prompt_id": "7763d96632870ad4", "reading_canonical_log_prob": -319.5079345703125, "reading_distances": [13, 12, 12, 12], "reading_log_probs": [-311.419921875, -319.0244140625, -312.89105224609375, -315.47540283203125], "reading_logprob_std": 3.333315761961917, "tokenization_path_logprob_std": 88.51388577814511, "trajectory_balance_rmse": 4987.81533570568} {"canonical_length": 11, "canonical_nll": 0.9907182346690785, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "151a13661f1584a8", "reading_canonical_log_prob": -10.897900581359863, "reading_distances": [15, 12, 13, 15], "reading_log_probs": [-9.799845695495605, -12.918067932128906, -11.05373764038086, -10.04052734375], "reading_logprob_std": 1.418216161297504, "tokenization_path_logprob_std": 37.471570896953544, "trajectory_balance_rmse": 210.83714323236657} {"canonical_length": 166, "canonical_nll": 0.8966575530638178, "generation_rouge_l_f1": 0.3645083932853717, "generation_token_f1": 0.47002398081534774, "prompt_id": "f8789af8051fda52", "reading_canonical_log_prob": -148.84515380859375, "reading_distances": [19, 21, 25, 28], "reading_log_probs": [-149.57669067382812, -147.63931274414062, -149.74972534179688, -151.06179809570312], "reading_logprob_std": 1.4106056772068016, "tokenization_path_logprob_std": 181.30971217753077, "trajectory_balance_rmse": 3693.6610584341934} {"canonical_length": 199, "canonical_nll": 0.7158546639447236, "generation_rouge_l_f1": 0.22907488986784139, "generation_token_f1": 0.4405286343612335, "prompt_id": "b3a3308a3d93f328", "reading_canonical_log_prob": -142.455078125, "reading_distances": [7, 10, 10, 6], "reading_log_probs": [-148.84832763671875, -146.37911987304688, -145.68023681640625, -149.18385314941406], "reading_logprob_std": 1.753013060488881, "tokenization_path_logprob_std": 125.34604962554543, "trajectory_balance_rmse": 3515.5909924259854} {"canonical_length": 33, "canonical_nll": 0.546922914909594, "generation_rouge_l_f1": 0.16161616161616163, "generation_token_f1": 0.18181818181818182, "prompt_id": "06bad79ff1fc809f", "reading_canonical_log_prob": -18.0484561920166, "reading_distances": [21, 25, 19, 26], "reading_log_probs": [-20.284873962402344, -21.335895538330078, -21.19195556640625, -16.21790313720703], "reading_logprob_std": 2.405262591074536, "tokenization_path_logprob_std": 46.38104753501209, "trajectory_balance_rmse": 537.4038906531594} {"canonical_length": 153, "canonical_nll": 1.7223618451286764, "generation_rouge_l_f1": 0.20722891566265061, "generation_token_f1": 0.4000000000000001, "prompt_id": "df98426df5120860", "reading_canonical_log_prob": -263.5213623046875, "reading_distances": [21, 18, 21, 20], "reading_log_probs": [-259.76214599609375, -259.9242248535156, -264.28387451171875, -262.68182373046875], "reading_logprob_std": 2.2017848784526324, "tokenization_path_logprob_std": 261.2122501210815, "trajectory_balance_rmse": 3488.6325659325835} {"canonical_length": 25, "canonical_nll": 1.1165077209472656, "generation_rouge_l_f1": 0.09243697478991596, "generation_token_f1": 0.09243697478991596, "prompt_id": "e01eb25b0bbbab91", "reading_canonical_log_prob": -27.91269302368164, "reading_distances": [28, 30, 28, 22], "reading_log_probs": [-33.0933837890625, -39.26009750366211, -27.10674285888672, -25.080814361572266], "reading_logprob_std": 6.396157585590649, "tokenization_path_logprob_std": 32.37007448134235, "trajectory_balance_rmse": 219.78247421622174} {"canonical_length": 42, "canonical_nll": 1.7454156421479725, "generation_rouge_l_f1": 0.3529411764705882, "generation_token_f1": 0.4264705882352941, "prompt_id": "050fda8d8b928503", "reading_canonical_log_prob": -73.30745697021484, "reading_distances": [29, 30, 28, 33], "reading_log_probs": [-73.44132995605469, -73.38262939453125, -73.68287658691406, -71.382568359375], "reading_logprob_std": 1.0677897852313805, "tokenization_path_logprob_std": 98.67324684356157, "trajectory_balance_rmse": 976.2286443803163} {"canonical_length": 191, "canonical_nll": 1.1998495531331805, "generation_rouge_l_f1": 0.3161512027491409, "generation_token_f1": 0.48797250859106533, "prompt_id": "c86e9e8e522d0430", "reading_canonical_log_prob": -229.1712646484375, "reading_distances": [30, 21, 19, 24], "reading_log_probs": [-237.54788208007812, -237.69097900390625, -237.90309143066406, -242.42398071289062], "reading_logprob_std": 2.359514869809996, "tokenization_path_logprob_std": 231.02101005071907, "trajectory_balance_rmse": 3385.964415570276} {"canonical_length": 71, "canonical_nll": 1.06070107473454, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.3058823529411765, "prompt_id": "0d01c13272391fbb", "reading_canonical_log_prob": -75.30977630615234, "reading_distances": [27, 26, 27, 32], "reading_log_probs": [-75.43063354492188, -80.82584381103516, -78.36912536621094, -83.63896179199219], "reading_logprob_std": 3.4980941818060423, "tokenization_path_logprob_std": 52.120661441807265, "trajectory_balance_rmse": 1425.8578556162095} {"canonical_length": 96, "canonical_nll": 1.423648198445638, "generation_rouge_l_f1": 0.22292993630573246, "generation_token_f1": 0.34394904458598724, "prompt_id": "f4314cbcdd2435c8", "reading_canonical_log_prob": -136.67022705078125, "reading_distances": [23, 30, 30, 30], "reading_log_probs": [-139.37094116210938, -132.16055297851562, -138.45639038085938, -147.94264221191406], "reading_logprob_std": 6.487173516169593, "tokenization_path_logprob_std": 123.91442068825567, "trajectory_balance_rmse": 2126.679825898329} {"canonical_length": 28, "canonical_nll": 1.1002257210867745, "generation_rouge_l_f1": 0.136986301369863, "generation_token_f1": 0.17808219178082194, "prompt_id": "6039b41734030698", "reading_canonical_log_prob": -30.806320190429688, "reading_distances": [13, 18, 17, 15], "reading_log_probs": [-35.92156982421875, -32.876651763916016, -33.289466857910156, -35.10633087158203], "reading_logprob_std": 1.4522105746720546, "tokenization_path_logprob_std": 50.24597568994327, "trajectory_balance_rmse": 311.41450785399604} {"canonical_length": 180, "canonical_nll": 1.0747244093153212, "generation_rouge_l_f1": 0.3286384976525822, "generation_token_f1": 0.5352112676056339, "prompt_id": "bc991d771f2ffcd2", "reading_canonical_log_prob": -193.4503936767578, "reading_distances": [28, 34, 27, 26], "reading_log_probs": [-195.15695190429688, -198.6260986328125, -197.2825164794922, -201.15895080566406], "reading_logprob_std": 2.513700638777288, "tokenization_path_logprob_std": 61.706631532569666, "trajectory_balance_rmse": 3371.7629812182186} {"canonical_length": 322, "canonical_nll": 0.997580865895526, "generation_rouge_l_f1": 0.3355263157894737, "generation_token_f1": 0.6151315789473684, "prompt_id": "ee15b1eaf2aa94fb", "reading_canonical_log_prob": -321.2210388183594, "reading_distances": [31, 29, 29, 32], "reading_log_probs": [-339.78961181640625, -339.3483581542969, -340.1778564453125, -340.43902587890625], "reading_logprob_std": 0.4754831125741441, "tokenization_path_logprob_std": 150.02888068448686, "trajectory_balance_rmse": 6034.528620912911} {"canonical_length": 42, "canonical_nll": 1.5668344043550038, "generation_rouge_l_f1": 0.15434083601286172, "generation_token_f1": 0.21864951768488747, "prompt_id": "e36603aac9ffa742", "reading_canonical_log_prob": -65.80704498291016, "reading_distances": [24, 27, 27, 29], "reading_log_probs": [-60.83356475830078, -62.56185531616211, -63.993621826171875, -69.16314697265625], "reading_logprob_std": 3.590565019414547, "tokenization_path_logprob_std": 96.99280331869096, "trajectory_balance_rmse": 959.4626301135573} {"canonical_length": 248, "canonical_nll": 1.4531900959630166, "generation_rouge_l_f1": 0.21285140562248994, "generation_token_f1": 0.4337349397590362, "prompt_id": "46c0bc7a74b4f8f8", "reading_canonical_log_prob": -360.3911437988281, "reading_distances": [19, 11, 10, 18], "reading_log_probs": [-355.97216796875, -356.18218994140625, -352.4412841796875, -354.38665771484375], "reading_logprob_std": 1.7327214243321591, "tokenization_path_logprob_std": 215.299902462808, "trajectory_balance_rmse": 4867.1564108757675} {"canonical_length": 193, "canonical_nll": 1.5970667705634716, "generation_rouge_l_f1": 0.24347826086956523, "generation_token_f1": 0.44782608695652176, "prompt_id": "17af0bbd6fc62e57", "reading_canonical_log_prob": -308.23388671875, "reading_distances": [22, 30, 30, 29], "reading_log_probs": [-305.87493896484375, -305.86090087890625, -302.015625, -303.959716796875], "reading_logprob_std": 1.8426140334028538, "tokenization_path_logprob_std": 36.30976223015333, "trajectory_balance_rmse": 3429.639956101329} {"canonical_length": 27, "canonical_nll": 1.7293126141583477, "generation_rouge_l_f1": 0.09523809523809525, "generation_token_f1": 0.1111111111111111, "prompt_id": "917c56964f6f4a40", "reading_canonical_log_prob": -46.69144058227539, "reading_distances": [36, 35, 28, 33], "reading_log_probs": [-45.077781677246094, -44.46943664550781, -45.74365234375, -44.5295295715332], "reading_logprob_std": 0.5926901034682279, "tokenization_path_logprob_std": 66.0542339716962, "trajectory_balance_rmse": 505.1819939301882} {"canonical_length": 350, "canonical_nll": 1.4673427036830358, "generation_rouge_l_f1": 0.1996615905245347, "generation_token_f1": 0.4094754653130288, "prompt_id": "74f8c2e7c18cb610", "reading_canonical_log_prob": -513.5699462890625, "reading_distances": [43, 41, 32, 39], "reading_log_probs": [-525.9314575195312, -525.0418701171875, -526.3507690429688, -532.869384765625], "reading_logprob_std": 3.589075731545918, "tokenization_path_logprob_std": 196.78730205928858, "trajectory_balance_rmse": 6016.889903186999} {"canonical_length": 26, "canonical_nll": 1.0901861924391527, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "c533f2b5441efb8c", "reading_canonical_log_prob": -28.34484100341797, "reading_distances": [25, 25, 23, 27], "reading_log_probs": [-31.606098175048828, -29.89946174621582, -28.480958938598633, -29.341474533081055], "reading_logprob_std": 1.3188255640520645, "tokenization_path_logprob_std": 41.869360021934185, "trajectory_balance_rmse": 249.76722934927363} {"canonical_length": 91, "canonical_nll": 1.1515662224738152, "generation_rouge_l_f1": 0.254335260115607, "generation_token_f1": 0.4624277456647399, "prompt_id": "ce28dbb2751f3b46", "reading_canonical_log_prob": -104.79252624511719, "reading_distances": [21, 16, 15, 17], "reading_log_probs": [-103.98932647705078, -103.58506774902344, -102.07203674316406, -103.64530944824219], "reading_logprob_std": 0.85273343092104, "tokenization_path_logprob_std": 26.501833949342476, "trajectory_balance_rmse": 2005.5316363975533} {"canonical_length": 13, "canonical_nll": 0.8035841721754807, "generation_rouge_l_f1": 0.22580645161290322, "generation_token_f1": 0.2903225806451613, "prompt_id": "a24d7fb2090fb702", "reading_canonical_log_prob": -10.44659423828125, "reading_distances": [25, 25, 28, 31], "reading_log_probs": [-10.642887115478516, -10.824567794799805, -11.531105995178223, -19.843219757080078], "reading_logprob_std": 4.438417285472255, "tokenization_path_logprob_std": 36.08338201464311, "trajectory_balance_rmse": 212.89841552363626} {"canonical_length": 133, "canonical_nll": 1.155289269927749, "generation_rouge_l_f1": 0.5206611570247934, "generation_token_f1": 0.628099173553719, "prompt_id": "4973bf9719e2751b", "reading_canonical_log_prob": -153.65347290039062, "reading_distances": [18, 18, 23, 24], "reading_log_probs": [-164.40200805664062, -165.08787536621094, -165.80323791503906, -165.43450927734375], "reading_logprob_std": 0.5963623972348043, "tokenization_path_logprob_std": 115.41401476654526, "trajectory_balance_rmse": 2663.485861646829} {"canonical_length": 15, "canonical_nll": 2.6382423400878907, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.4166666666666667, "prompt_id": "c4461f4a56e11ca5", "reading_canonical_log_prob": -39.57363510131836, "reading_distances": [18, 23, 21, 22], "reading_log_probs": [-55.2528190612793, -56.707271575927734, -49.204246520996094, -45.55284881591797], "reading_logprob_std": 5.21887058294174, "tokenization_path_logprob_std": 17.77167284505997, "trajectory_balance_rmse": 291.36159798391964} {"canonical_length": 51, "canonical_nll": 1.0415215585746018, "generation_rouge_l_f1": 0.2736842105263158, "generation_token_f1": 0.37894736842105264, "prompt_id": "cdeadd1b86cf8871", "reading_canonical_log_prob": -53.11759948730469, "reading_distances": [13, 10, 11, 10], "reading_log_probs": [-62.508934020996094, -59.314544677734375, -60.9810791015625, -61.51054382324219], "reading_logprob_std": 1.3358932418214626, "tokenization_path_logprob_std": 77.62244556641889, "trajectory_balance_rmse": 671.4332351043452} {"canonical_length": 11, "canonical_nll": 1.076110059564764, "generation_rouge_l_f1": 0.11764705882352941, "generation_token_f1": 0.11764705882352941, "prompt_id": "55ee0fbefdef0ed3", "reading_canonical_log_prob": -11.837210655212402, "reading_distances": [15, 14, 16, 12], "reading_log_probs": [-25.215679168701172, -23.494855880737305, -23.91701316833496, -21.17041015625], "reading_logprob_std": 1.686640160189208, "tokenization_path_logprob_std": 53.4351541776222, "trajectory_balance_rmse": 224.1626112517164} {"canonical_length": 282, "canonical_nll": 1.480954866882757, "generation_rouge_l_f1": 0.20577617328519854, "generation_token_f1": 0.38989169675090257, "prompt_id": "0b1cafe0e71072e8", "reading_canonical_log_prob": -417.6292724609375, "reading_distances": [30, 22, 21, 21], "reading_log_probs": [-435.556640625, -427.4442138671875, -433.18402099609375, -429.4368896484375], "reading_logprob_std": 3.649763460398183, "tokenization_path_logprob_std": 257.18943703093095, "trajectory_balance_rmse": 5755.817473311273} {"canonical_length": 98, "canonical_nll": 1.292642710160236, "generation_rouge_l_f1": 0.3978494623655915, "generation_token_f1": 0.6559139784946236, "prompt_id": "3a37b4e0a87fe5c3", "reading_canonical_log_prob": -126.67898559570312, "reading_distances": [15, 15, 20, 16], "reading_log_probs": [-133.41278076171875, -130.82452392578125, -135.3768310546875, -137.36721801757812], "reading_logprob_std": 2.7941326538746902, "tokenization_path_logprob_std": 60.0064173026637, "trajectory_balance_rmse": 2091.4936150056633} {"canonical_length": 261, "canonical_nll": 0.9976088717522749, "generation_rouge_l_f1": 0.3204868154158215, "generation_token_f1": 0.5233265720081136, "prompt_id": "7236a1de81b123d6", "reading_canonical_log_prob": -260.37591552734375, "reading_distances": [24, 22, 23, 24], "reading_log_probs": [-261.97125244140625, -261.81805419921875, -264.78607177734375, -263.96807861328125], "reading_logprob_std": 1.4729459850256645, "tokenization_path_logprob_std": 259.36943224476903, "trajectory_balance_rmse": 5379.528430700034} {"canonical_length": 181, "canonical_nll": 1.4844066978159531, "generation_rouge_l_f1": 0.2896551724137931, "generation_token_f1": 0.5103448275862069, "prompt_id": "564e5eeb9b5a33fe", "reading_canonical_log_prob": -268.6776123046875, "reading_distances": [42, 46, 36, 36], "reading_log_probs": [-271.61651611328125, -274.517578125, -277.56048583984375, -275.4190673828125], "reading_logprob_std": 2.464142452501882, "tokenization_path_logprob_std": 237.43110660869536, "trajectory_balance_rmse": 3618.0179631398573} {"canonical_length": 141, "canonical_nll": 1.185851293252715, "generation_rouge_l_f1": 0.27717391304347827, "generation_token_f1": 0.4239130434782609, "prompt_id": "8686f6b687c7ab8c", "reading_canonical_log_prob": -167.2050323486328, "reading_distances": [22, 20, 25, 24], "reading_log_probs": [-171.22097778320312, -167.7799835205078, -170.43978881835938, -179.57696533203125], "reading_logprob_std": 5.099056811123319, "tokenization_path_logprob_std": 133.40847622312634, "trajectory_balance_rmse": 3023.126516976616} {"canonical_length": 101, "canonical_nll": 1.3410501007986542, "generation_rouge_l_f1": 0.2317880794701987, "generation_token_f1": 0.3973509933774834, "prompt_id": "79b5d532fcbe3323", "reading_canonical_log_prob": -135.44606018066406, "reading_distances": [28, 26, 24, 28], "reading_log_probs": [-138.09344482421875, -140.997802734375, -132.9862518310547, -143.57080078125], "reading_logprob_std": 4.540173333002945, "tokenization_path_logprob_std": 167.44609080623368, "trajectory_balance_rmse": 2696.544795558863} {"canonical_length": 333, "canonical_nll": 1.036856986380912, "generation_rouge_l_f1": 0.2608695652173913, "generation_token_f1": 0.48953301127214177, "prompt_id": "f5411f1c63da4002", "reading_canonical_log_prob": -345.27337646484375, "reading_distances": [45, 44, 44, 54], "reading_log_probs": [-367.4693603515625, -364.31231689453125, -363.03607177734375, -359.0690612792969], "reading_logprob_std": 3.4766326957372957, "tokenization_path_logprob_std": 180.12092709354192, "trajectory_balance_rmse": 7557.575580974138} {"canonical_length": 105, "canonical_nll": 1.2366811116536458, "generation_rouge_l_f1": 0.1717171717171717, "generation_token_f1": 0.30303030303030304, "prompt_id": "0709a316fd9c3cd1", "reading_canonical_log_prob": -129.8515167236328, "reading_distances": [30, 26, 31, 25], "reading_log_probs": [-161.10389709472656, -158.72195434570312, -149.4134521484375, -135.45040893554688], "reading_logprob_std": 11.631928844540814, "tokenization_path_logprob_std": 42.61241339508699, "trajectory_balance_rmse": 1552.0965589159257} {"canonical_length": 68, "canonical_nll": 0.6346193201401654, "generation_rouge_l_f1": 0.6086956521739131, "generation_token_f1": 0.7536231884057971, "prompt_id": "28c4e87a8705b13a", "reading_canonical_log_prob": -43.15411376953125, "reading_distances": [29, 29, 25, 27], "reading_log_probs": [-44.16466522216797, -44.03709411621094, -45.2847785949707, -42.16059875488281], "reading_logprob_std": 1.2950438162838036, "tokenization_path_logprob_std": 139.86922423074301, "trajectory_balance_rmse": 1234.4379800936597} {"canonical_length": 10, "canonical_nll": 0.6178737640380859, "generation_rouge_l_f1": 0.3902439024390244, "generation_token_f1": 0.3902439024390244, "prompt_id": "d86571d08d8aa4d7", "reading_canonical_log_prob": -6.178737640380859, "reading_distances": [36, 42, 39, 47], "reading_log_probs": [-7.866137504577637, -10.280590057373047, -5.711801052093506, -6.043260097503662], "reading_logprob_std": 2.096272326133881, "tokenization_path_logprob_std": 32.21331925644244, "trajectory_balance_rmse": 247.53566514729093} {"canonical_length": 30, "canonical_nll": 0.41890891393025714, "generation_rouge_l_f1": 0.43243243243243246, "generation_token_f1": 0.5405405405405406, "prompt_id": "f8bb815af5d18371", "reading_canonical_log_prob": -12.567267417907715, "reading_distances": [26, 20, 27, 18], "reading_log_probs": [-11.60881233215332, -11.724847793579102, -10.586267471313477, -11.467935562133789], "reading_logprob_std": 0.5178976855484139, "tokenization_path_logprob_std": 117.67395290187274, "trajectory_balance_rmse": 524.9712161683544} {"canonical_length": 39, "canonical_nll": 1.7974710709009416, "generation_rouge_l_f1": 0.2777777777777778, "generation_token_f1": 0.3611111111111111, "prompt_id": "e8ce7622bc908c26", "reading_canonical_log_prob": -70.10137176513672, "reading_distances": [19, 20, 22, 15], "reading_log_probs": [-92.33235168457031, -89.1807861328125, -79.36502838134766, -75.18232727050781], "reading_logprob_std": 8.072629248567694, "tokenization_path_logprob_std": 83.81290755299794, "trajectory_balance_rmse": 795.1616462096138} {"canonical_length": 35, "canonical_nll": 0.8125597272600447, "generation_rouge_l_f1": 0.1237785016286645, "generation_token_f1": 0.15635179153094464, "prompt_id": "300a5e4b8236fdd5", "reading_canonical_log_prob": -28.439590454101562, "reading_distances": [33, 33, 27, 25], "reading_log_probs": [-30.89866828918457, -33.637325286865234, -32.171142578125, -29.59853744506836], "reading_logprob_std": 1.7293922178004844, "tokenization_path_logprob_std": 115.25136509802786, "trajectory_balance_rmse": 742.1992980478318} {"canonical_length": 64, "canonical_nll": 0.8907232284545898, "generation_rouge_l_f1": 0.20238095238095238, "generation_token_f1": 0.27380952380952384, "prompt_id": "20d20afcda52df5b", "reading_canonical_log_prob": -57.00628662109375, "reading_distances": [28, 29, 26, 27], "reading_log_probs": [-60.323123931884766, -61.260616302490234, -60.54691696166992, -62.02543640136719], "reading_logprob_std": 0.7695875126312494, "tokenization_path_logprob_std": 166.09687345052401, "trajectory_balance_rmse": 1371.8296071874006} {"canonical_length": 92, "canonical_nll": 1.9172857533330503, "generation_rouge_l_f1": 0.2608695652173913, "generation_token_f1": 0.44347826086956527, "prompt_id": "26bff204aae37f53", "reading_canonical_log_prob": -176.39028930664062, "reading_distances": [15, 14, 19, 14], "reading_log_probs": [-178.6876220703125, -179.03399658203125, -181.4735107421875, -178.33387756347656], "reading_logprob_std": 1.4231707160765428, "tokenization_path_logprob_std": 123.01799573608135, "trajectory_balance_rmse": 2085.9113725406537} {"canonical_length": 259, "canonical_nll": 1.5410330636160714, "generation_rouge_l_f1": 0.2209737827715356, "generation_token_f1": 0.449438202247191, "prompt_id": "d7bb721e18ff9d29", "reading_canonical_log_prob": -399.1275634765625, "reading_distances": [21, 22, 26, 25], "reading_log_probs": [-427.459228515625, -421.75177001953125, -408.8062744140625, -425.40765380859375], "reading_logprob_std": 8.372941928864607, "tokenization_path_logprob_std": 200.47358515778882, "trajectory_balance_rmse": 5995.473470586819} {"canonical_length": 110, "canonical_nll": 0.8924879594282671, "generation_rouge_l_f1": 0.3561643835616438, "generation_token_f1": 0.41643835616438357, "prompt_id": "6df1e56bb497e6a2", "reading_canonical_log_prob": -98.17367553710938, "reading_distances": [41, 43, 35, 42], "reading_log_probs": [-105.55255889892578, -106.9046859741211, -103.98382568359375, -106.332275390625], "reading_logprob_std": 1.2672616171336697, "tokenization_path_logprob_std": 87.36941662778487, "trajectory_balance_rmse": 1815.717251719462} {"canonical_length": 9, "canonical_nll": 0.31414053175184464, "generation_rouge_l_f1": 0.09580838323353293, "generation_token_f1": 0.10778443113772454, "prompt_id": "196c78145c93b0b5", "reading_canonical_log_prob": -2.8272647857666016, "reading_distances": [36, 41, 38, 35], "reading_log_probs": [-2.287820339202881, -2.680889368057251, -3.0181353092193604, -2.929934024810791], "reading_logprob_std": 0.32707169634230426, "tokenization_path_logprob_std": 46.70381209359108, "trajectory_balance_rmse": 181.7284081752476} {"canonical_length": 61, "canonical_nll": 0.49415294459608733, "generation_rouge_l_f1": 0.75, "generation_token_f1": 0.7857142857142858, "prompt_id": "31d1718162ea0ff4", "reading_canonical_log_prob": -30.143329620361328, "reading_distances": [25, 22, 22, 19], "reading_log_probs": [-32.45354080200195, -32.29869842529297, -32.506134033203125, -32.30196762084961], "reading_logprob_std": 0.10584610756488937, "tokenization_path_logprob_std": 99.25619267900062, "trajectory_balance_rmse": 898.236095322615} {"canonical_length": 117, "canonical_nll": 0.9254915938418136, "generation_rouge_l_f1": 0.2177650429799427, "generation_token_f1": 0.3667621776504298, "prompt_id": "823548660cad1f9e", "reading_canonical_log_prob": -108.28251647949219, "reading_distances": [23, 23, 25, 25], "reading_log_probs": [-102.3230209350586, -100.55082702636719, -104.45199584960938, -98.09579467773438], "reading_logprob_std": 2.69552472666913, "tokenization_path_logprob_std": 114.69241390534522, "trajectory_balance_rmse": 1587.2614594736672} {"canonical_length": 14, "canonical_nll": 0.6203445707048688, "generation_rouge_l_f1": 0.14285714285714288, "generation_token_f1": 0.14285714285714288, "prompt_id": "912aaeeaa0cce8d9", "reading_canonical_log_prob": -8.684823989868164, "reading_distances": [32, 32, 28, 26], "reading_log_probs": [-9.843095779418945, -8.079562187194824, -7.8372087478637695, -8.412090301513672], "reading_logprob_std": 0.8982023310087216, "tokenization_path_logprob_std": 49.67851264726858, "trajectory_balance_rmse": 340.01560093534954} {"canonical_length": 368, "canonical_nll": 1.504454239555027, "generation_rouge_l_f1": 0.20440251572327042, "generation_token_f1": 0.4528301886792453, "prompt_id": "4547f9ca1d45d608", "reading_canonical_log_prob": -553.63916015625, "reading_distances": [25, 26, 27, 25], "reading_log_probs": [-557.8208618164062, -559.4083251953125, -559.0858154296875, -560.695068359375], "reading_logprob_std": 1.1807704541830226, "tokenization_path_logprob_std": 177.82283956616055, "trajectory_balance_rmse": 7588.735842493768} {"canonical_length": 228, "canonical_nll": 1.5021006266276042, "generation_rouge_l_f1": 0.3142857142857143, "generation_token_f1": 0.5306122448979592, "prompt_id": "99e3f7e86895d3aa", "reading_canonical_log_prob": -342.47894287109375, "reading_distances": [20, 23, 20, 19], "reading_log_probs": [-344.0309753417969, -346.1311340332031, -344.00299072265625, -344.35650634765625], "reading_logprob_std": 1.0132740158115954, "tokenization_path_logprob_std": 65.79621033569468, "trajectory_balance_rmse": 5069.809368275123} {"canonical_length": 318, "canonical_nll": 0.7976547817014298, "generation_rouge_l_f1": 0.3073929961089494, "generation_token_f1": 0.5136186770428016, "prompt_id": "d3e72bee48c9454b", "reading_canonical_log_prob": -253.6542205810547, "reading_distances": [17, 25, 23, 29], "reading_log_probs": [-263.09686279296875, -267.10430908203125, -269.02197265625, -267.690185546875], "reading_logprob_std": 2.550482116088824, "tokenization_path_logprob_std": 105.18839214956763, "trajectory_balance_rmse": 5253.215016544275} {"canonical_length": 281, "canonical_nll": 1.4176722624972198, "generation_rouge_l_f1": 0.22097378277153557, "generation_token_f1": 0.45692883895131087, "prompt_id": "3126c4e8c381cc6a", "reading_canonical_log_prob": -398.36590576171875, "reading_distances": [25, 25, 24, 26], "reading_log_probs": [-405.6690368652344, -405.678955078125, -400.8115234375, -401.15411376953125], "reading_logprob_std": 2.71206430312464, "tokenization_path_logprob_std": 326.35215558514955, "trajectory_balance_rmse": 6488.088234196615} {"canonical_length": 23, "canonical_nll": 0.37201661648957624, "generation_rouge_l_f1": 0.20512820512820512, "generation_token_f1": 0.20512820512820512, "prompt_id": "963018756c1975a3", "reading_canonical_log_prob": -8.556382179260254, "reading_distances": [44, 40, 38, 46], "reading_log_probs": [-8.954931259155273, -8.060184478759766, -8.419588088989258, -7.029996871948242], "reading_logprob_std": 0.8120938839750002, "tokenization_path_logprob_std": 43.3871483219989, "trajectory_balance_rmse": 326.8138045247213} {"canonical_length": 50, "canonical_nll": 2.1546142578125, "generation_rouge_l_f1": 0.3035714285714286, "generation_token_f1": 0.4285714285714286, "prompt_id": "b9207d2d35a02140", "reading_canonical_log_prob": -107.730712890625, "reading_distances": [35, 37, 39, 39], "reading_log_probs": [-119.77658081054688, -118.51406860351562, -112.77034759521484, -125.90503692626953], "reading_logprob_std": 5.388073425750405, "tokenization_path_logprob_std": 39.94497090531887, "trajectory_balance_rmse": 1169.7899901792327} {"canonical_length": 59, "canonical_nll": 0.6400767180879238, "generation_rouge_l_f1": 0.16842105263157897, "generation_token_f1": 0.28070175438596495, "prompt_id": "cc824ceca446b0f2", "reading_canonical_log_prob": -37.7645263671875, "reading_distances": [12, 11, 16, 14], "reading_log_probs": [-41.39405822753906, -39.535552978515625, -40.650882720947266, -48.97467041015625], "reading_logprob_std": 4.292419407855078, "tokenization_path_logprob_std": 116.23131829514175, "trajectory_balance_rmse": 1165.6195133673546} {"canonical_length": 27, "canonical_nll": 1.0325613375063296, "generation_rouge_l_f1": 0.7391304347826089, "generation_token_f1": 0.7391304347826089, "prompt_id": "9cebf3474b92652d", "reading_canonical_log_prob": -27.8791561126709, "reading_distances": [21, 26, 27, 25], "reading_log_probs": [-27.808229446411133, -28.02596092224121, -27.0772705078125, -30.724536895751953], "reading_logprob_std": 1.5961264307795404, "tokenization_path_logprob_std": 46.22925574529828, "trajectory_balance_rmse": 607.862958753528} {"canonical_length": 5, "canonical_nll": 2.6742454528808595, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "c004648702d6143d", "reading_canonical_log_prob": -13.371227264404297, "reading_distances": [22, 19, 24, 21], "reading_log_probs": [-17.924762725830078, -15.620748519897461, -16.436389923095703, -16.29757308959961], "reading_logprob_std": 0.9710009304552747, "tokenization_path_logprob_std": 2.153383515095175, "trajectory_balance_rmse": 58.370657197769354} {"canonical_length": 120, "canonical_nll": 1.2370304107666015, "generation_rouge_l_f1": 0.26126126126126126, "generation_token_f1": 0.4954954954954955, "prompt_id": "ebe4917bfacc9af5", "reading_canonical_log_prob": -148.4436492919922, "reading_distances": [28, 27, 18, 32], "reading_log_probs": [-155.21185302734375, -154.15884399414062, -152.75015258789062, -158.50228881835938], "reading_logprob_std": 2.4483457068898957, "tokenization_path_logprob_std": 136.12737400737825, "trajectory_balance_rmse": 2796.1803881663864} {"canonical_length": 127, "canonical_nll": 1.4735085795244833, "generation_rouge_l_f1": 0.22842639593908629, "generation_token_f1": 0.4619289340101523, "prompt_id": "bd51aa5053398c97", "reading_canonical_log_prob": -187.13558959960938, "reading_distances": [31, 21, 23, 18], "reading_log_probs": [-189.3699188232422, -188.84762573242188, -189.94515991210938, -189.7922821044922], "reading_logprob_std": 0.49180040712037554, "tokenization_path_logprob_std": 148.5107935065221, "trajectory_balance_rmse": 3031.2001431073572} {"canonical_length": 168, "canonical_nll": 1.4973231724330358, "generation_rouge_l_f1": 0.17218543046357618, "generation_token_f1": 0.2958057395143488, "prompt_id": "0d7273ba3d245c40", "reading_canonical_log_prob": -251.55029296875, "reading_distances": [28, 22, 20, 15], "reading_log_probs": [-258.8712463378906, -257.8172912597656, -259.3872985839844, -258.1954345703125], "reading_logprob_std": 0.6989461090991953, "tokenization_path_logprob_std": 143.63585746529873, "trajectory_balance_rmse": 3338.0866824523405} {"canonical_length": 211, "canonical_nll": 1.7615194456272216, "generation_rouge_l_f1": 0.21729490022172948, "generation_token_f1": 0.4523281596452328, "prompt_id": "16d4d8a6fd8e0475", "reading_canonical_log_prob": -371.68060302734375, "reading_distances": [32, 31, 29, 33], "reading_log_probs": [-377.2010498046875, -365.9719543457031, -371.71356201171875, -372.178955078125], "reading_logprob_std": 4.592893513684352, "tokenization_path_logprob_std": 228.09442097333098, "trajectory_balance_rmse": 4564.470393122564} {"canonical_length": 131, "canonical_nll": 1.8884278508543058, "generation_rouge_l_f1": 0.26857142857142857, "generation_token_f1": 0.4, "prompt_id": "5669cff2cb390e91", "reading_canonical_log_prob": -247.38404846191406, "reading_distances": [26, 24, 21, 20], "reading_log_probs": [-259.30963134765625, -260.87786865234375, -277.1001892089844, -248.91937255859375], "reading_logprob_std": 11.644914967180382, "tokenization_path_logprob_std": 200.02843550334097, "trajectory_balance_rmse": 2866.4759931430776} {"canonical_length": 314, "canonical_nll": 1.0583035414385948, "generation_rouge_l_f1": 0.25806451612903225, "generation_token_f1": 0.465195246179966, "prompt_id": "ea8af27e7f1b048e", "reading_canonical_log_prob": -332.30731201171875, "reading_distances": [27, 27, 28, 30], "reading_log_probs": [-338.93670654296875, -342.39117431640625, -342.27569580078125, -342.3727722167969], "reading_logprob_std": 1.7056729349378223, "tokenization_path_logprob_std": 324.8061356080011, "trajectory_balance_rmse": 6360.420882417653} {"canonical_length": 8, "canonical_nll": 1.1221692562103271, "generation_rouge_l_f1": 0.03861003861003861, "generation_token_f1": 0.04633204633204632, "prompt_id": "b75f2d55eed057da", "reading_canonical_log_prob": -8.977354049682617, "reading_distances": [14, 17, 18, 17], "reading_log_probs": [-8.682971954345703, -9.259958267211914, -9.084246635437012, -8.782257080078125], "reading_logprob_std": 0.2667806221703936, "tokenization_path_logprob_std": 42.849988722247886, "trajectory_balance_rmse": 161.68777665790606} {"canonical_length": 22, "canonical_nll": 0.7716738960959695, "generation_rouge_l_f1": 0.2298850574712644, "generation_token_f1": 0.2298850574712644, "prompt_id": "55725f89c2c9fced", "reading_canonical_log_prob": -16.976825714111328, "reading_distances": [35, 29, 30, 32], "reading_log_probs": [-19.988800048828125, -20.305206298828125, -18.486913681030273, -18.457672119140625], "reading_logprob_std": 0.9755577025424436, "tokenization_path_logprob_std": 28.264008340595073, "trajectory_balance_rmse": 418.2278664505941} {"canonical_length": 6, "canonical_nll": 1.0835871696472168, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "e8e15c8f276bd3e5", "reading_canonical_log_prob": -6.501523017883301, "reading_distances": [21, 24, 21, 25], "reading_log_probs": [-8.144867897033691, -7.3240156173706055, -9.719908714294434, -9.045257568359375], "reading_logprob_std": 1.045760702221825, "tokenization_path_logprob_std": 19.536457840703495, "trajectory_balance_rmse": 93.08988499224563} {"canonical_length": 37, "canonical_nll": 1.4958452791781038, "generation_rouge_l_f1": 0.1644736842105263, "generation_token_f1": 0.1907894736842105, "prompt_id": "6e9b2ca73b16e2a8", "reading_canonical_log_prob": -55.346275329589844, "reading_distances": [36, 33, 30, 32], "reading_log_probs": [-61.7927360534668, -59.143409729003906, -58.54730987548828, -60.55193328857422], "reading_logprob_std": 1.4562824025889702, "tokenization_path_logprob_std": 83.54672207538023, "trajectory_balance_rmse": 817.7135137999857} {"canonical_length": 266, "canonical_nll": 1.4203810154047227, "generation_rouge_l_f1": 0.22, "generation_token_f1": 0.48, "prompt_id": "fb1406f3dcf9e9ba", "reading_canonical_log_prob": -377.82135009765625, "reading_distances": [24, 25, 22, 33], "reading_log_probs": [-381.64837646484375, -377.1102600097656, -376.284423828125, -376.708251953125], "reading_logprob_std": 2.4965737387492526, "tokenization_path_logprob_std": 125.38594263066877, "trajectory_balance_rmse": 5447.277624816893} {"canonical_length": 3, "canonical_nll": 0.8520001570383707, "generation_rouge_l_f1": 0.0625, "generation_token_f1": 0.0625, "prompt_id": "f6b05fa9b7a6a33d", "reading_canonical_log_prob": -2.5560004711151123, "reading_distances": [34, 24, 24, 32], "reading_log_probs": [-3.58085298538208, -3.4154179096221924, -2.81269907951355, -12.830286026000977], "reading_logprob_std": 4.791698556076731, "tokenization_path_logprob_std": 30.63512717692898, "trajectory_balance_rmse": 111.69219558824639} {"canonical_length": 274, "canonical_nll": 1.3229422464858007, "generation_rouge_l_f1": 0.23703703703703705, "generation_token_f1": 0.43333333333333335, "prompt_id": "b4b2d4b733fbf2fd", "reading_canonical_log_prob": -362.4861755371094, "reading_distances": [27, 40, 31, 33], "reading_log_probs": [-384.6554260253906, -397.70989990234375, -386.8680114746094, -398.7254943847656], "reading_logprob_std": 7.259818202547301, "tokenization_path_logprob_std": 276.6080932623046, "trajectory_balance_rmse": 5152.100001589057} {"canonical_length": 309, "canonical_nll": 0.7533473042608465, "generation_rouge_l_f1": 0.2424242424242424, "generation_token_f1": 0.4015151515151516, "prompt_id": "8c9f3f4649945cbf", "reading_canonical_log_prob": -232.78431701660156, "reading_distances": [52, 46, 43, 42], "reading_log_probs": [-248.76119995117188, -245.58990478515625, -244.19369506835938, -242.541748046875], "reading_logprob_std": 2.63897685209266, "tokenization_path_logprob_std": 142.70046807997113, "trajectory_balance_rmse": 2659.49180863106} {"canonical_length": 8, "canonical_nll": 0.11915481090545654, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "eccb7866ba7ed3da", "reading_canonical_log_prob": -0.9532384872436523, "reading_distances": [20, 26, 26, 26], "reading_log_probs": [-1.1362289190292358, -1.6837446689605713, -1.2712520360946655, -1.157543420791626], "reading_logprob_std": 0.2546937084499167, "tokenization_path_logprob_std": 43.4639167589184, "trajectory_balance_rmse": 174.81696812643023} {"canonical_length": 71, "canonical_nll": 2.022575915699274, "generation_rouge_l_f1": 0.18023255813953487, "generation_token_f1": 0.2674418604651163, "prompt_id": "f78297865f8b46fc", "reading_canonical_log_prob": -143.60289001464844, "reading_distances": [26, 17, 21, 20], "reading_log_probs": [-145.29168701171875, -142.2943572998047, -146.35667419433594, -141.79576110839844], "reading_logprob_std": 2.234066283590724, "tokenization_path_logprob_std": 95.22698329477014, "trajectory_balance_rmse": 1866.8573099739244} {"canonical_length": 147, "canonical_nll": 1.0958812480070153, "generation_rouge_l_f1": 0.19402985074626866, "generation_token_f1": 0.39800995024875624, "prompt_id": "65091372e7851e26", "reading_canonical_log_prob": -161.09454345703125, "reading_distances": [27, 26, 19, 24], "reading_log_probs": [-168.04530334472656, -170.2093963623047, -168.94705200195312, -170.66644287109375], "reading_logprob_std": 1.1946244668798827, "tokenization_path_logprob_std": 208.87897072865613, "trajectory_balance_rmse": 3106.927722244476} {"canonical_length": 16, "canonical_nll": 1.1276034116744995, "generation_rouge_l_f1": 0.32, "generation_token_f1": 0.4, "prompt_id": "aa225c057d9ef9cd", "reading_canonical_log_prob": -18.041654586791992, "reading_distances": [28, 29, 24, 25], "reading_log_probs": [-24.35530662536621, -22.64667510986328, -19.457860946655273, -17.91785430908203], "reading_logprob_std": 2.9332446558557224, "tokenization_path_logprob_std": 22.638996591604137, "trajectory_balance_rmse": 219.45596481660468} {"canonical_length": 352, "canonical_nll": 1.017350283536044, "generation_rouge_l_f1": 0.27426160337552746, "generation_token_f1": 0.4514767932489452, "prompt_id": "1b2704eec1f82afa", "reading_canonical_log_prob": -358.1072998046875, "reading_distances": [31, 31, 32, 34], "reading_log_probs": [-373.65350341796875, -377.7998046875, -387.588134765625, -391.6393127441406], "reading_logprob_std": 8.3596776957008, "tokenization_path_logprob_std": 319.2070391431778, "trajectory_balance_rmse": 6327.936317424934} {"canonical_length": 284, "canonical_nll": 1.35638642646897, "generation_rouge_l_f1": 0.26838235294117646, "generation_token_f1": 0.5, "prompt_id": "aa3d50bf234f7b48", "reading_canonical_log_prob": -385.2137451171875, "reading_distances": [36, 33, 34, 30], "reading_log_probs": [-392.7276611328125, -392.3687744140625, -390.0644836425781, -397.3455810546875], "reading_logprob_std": 3.0501262029706684, "tokenization_path_logprob_std": 343.02606129539697, "trajectory_balance_rmse": 7011.915359207577} {"canonical_length": 11, "canonical_nll": 0.9035430388017134, "generation_rouge_l_f1": 0.053231939163498096, "generation_token_f1": 0.053231939163498096, "prompt_id": "38ebe3c79fd5a7f3", "reading_canonical_log_prob": -9.938973426818848, "reading_distances": [23, 13, 19, 23], "reading_log_probs": [-14.975369453430176, -12.13317584991455, -10.839996337890625, -15.516755104064941], "reading_logprob_std": 2.2447247062785767, "tokenization_path_logprob_std": 34.99036976039302, "trajectory_balance_rmse": 116.81074217328698} {"canonical_length": 266, "canonical_nll": 1.4350323210981555, "generation_rouge_l_f1": 0.20702402957486135, "generation_token_f1": 0.37707948243992606, "prompt_id": "df8d0cc99eb20244", "reading_canonical_log_prob": -381.7185974121094, "reading_distances": [24, 18, 23, 28], "reading_log_probs": [-387.947021484375, -393.2436828613281, -393.1351623535156, -388.76397705078125], "reading_logprob_std": 2.8110740180303075, "tokenization_path_logprob_std": 271.38142977163506, "trajectory_balance_rmse": 5606.751349862909} {"canonical_length": 131, "canonical_nll": 1.6657139435979247, "generation_rouge_l_f1": 0.2473684210526316, "generation_token_f1": 0.4368421052631578, "prompt_id": "0f4d389ccaad372e", "reading_canonical_log_prob": -218.20852661132812, "reading_distances": [32, 29, 29, 26], "reading_log_probs": [-224.56015014648438, -217.05711364746094, -212.84820556640625, -215.10092163085938], "reading_logprob_std": 5.079029851350721, "tokenization_path_logprob_std": 122.07178520423078, "trajectory_balance_rmse": 2265.1825648708827} {"canonical_length": 30, "canonical_nll": 1.2697739919026694, "generation_rouge_l_f1": 0.3829787234042554, "generation_token_f1": 0.4255319148936171, "prompt_id": "936a9a79e3069ca4", "reading_canonical_log_prob": -38.09321975708008, "reading_distances": [27, 27, 24, 26], "reading_log_probs": [-41.010223388671875, -40.37162399291992, -39.85005187988281, -44.59843826293945], "reading_logprob_std": 2.146979713595982, "tokenization_path_logprob_std": 104.32211880674043, "trajectory_balance_rmse": 690.2323052408083} {"canonical_length": 427, "canonical_nll": 1.569092082753952, "generation_rouge_l_f1": 0.21180030257186083, "generation_token_f1": 0.35703479576399394, "prompt_id": "82391a0f7d399b89", "reading_canonical_log_prob": -670.0023193359375, "reading_distances": [24, 29, 30, 30], "reading_log_probs": [-680.4127807617188, -679.1854248046875, -682.9671020507812, -684.5543212890625], "reading_logprob_std": 2.4294840844058556, "tokenization_path_logprob_std": 309.5990917848397, "trajectory_balance_rmse": 7571.780478733019} {"canonical_length": 189, "canonical_nll": 0.9985999859199322, "generation_rouge_l_f1": 0.35526315789473684, "generation_token_f1": 0.493421052631579, "prompt_id": "0bf403e155ca208a", "reading_canonical_log_prob": -188.7353973388672, "reading_distances": [23, 21, 22, 24], "reading_log_probs": [-197.712646484375, -199.15945434570312, -195.49920654296875, -198.83062744140625], "reading_logprob_std": 1.654447468065378, "tokenization_path_logprob_std": 64.72610102220216, "trajectory_balance_rmse": 3251.0372387930233} {"canonical_length": 175, "canonical_nll": 0.735687517438616, "generation_rouge_l_f1": 0.06857142857142857, "generation_token_f1": 0.13714285714285715, "prompt_id": "3b3aada8838b1247", "reading_canonical_log_prob": -128.7453155517578, "reading_distances": [23, 18, 21, 16], "reading_log_probs": [-134.83236694335938, -125.71662902832031, -135.28509521484375, -125.14053344726562], "reading_logprob_std": 5.568010131372844, "tokenization_path_logprob_std": 89.1702475113243, "trajectory_balance_rmse": 2800.6514215543984} {"canonical_length": 62, "canonical_nll": 1.7948362288936492, "generation_rouge_l_f1": 0.2698412698412699, "generation_token_f1": 0.31746031746031744, "prompt_id": "0221b725811d035d", "reading_canonical_log_prob": -111.27984619140625, "reading_distances": [14, 20, 17, 12], "reading_log_probs": [-118.92937469482422, -122.69081115722656, -115.96241760253906, -121.67340850830078], "reading_logprob_std": 3.0194199956303462, "tokenization_path_logprob_std": 92.25549865230971, "trajectory_balance_rmse": 1234.7257917663537} {"canonical_length": 109, "canonical_nll": 0.9187447084199398, "generation_rouge_l_f1": 0.2666666666666667, "generation_token_f1": 0.39999999999999997, "prompt_id": "7ac2b23c73ad8de9", "reading_canonical_log_prob": -100.14317321777344, "reading_distances": [39, 38, 40, 33], "reading_log_probs": [-105.60698699951172, -107.48947143554688, -105.0766372680664, -106.67577362060547], "reading_logprob_std": 1.0804487207792122, "tokenization_path_logprob_std": 149.10187925139033, "trajectory_balance_rmse": 2094.0993718605314} {"canonical_length": 98, "canonical_nll": 1.4429427166374362, "generation_rouge_l_f1": 0.19318181818181818, "generation_token_f1": 0.35795454545454547, "prompt_id": "a2a34e9410c7790e", "reading_canonical_log_prob": -141.40838623046875, "reading_distances": [21, 22, 25, 22], "reading_log_probs": [-144.21238708496094, -146.69342041015625, -141.70144653320312, -144.3192138671875], "reading_logprob_std": 2.0388135129760747, "tokenization_path_logprob_std": 81.55967135082669, "trajectory_balance_rmse": 2188.167536711957} {"canonical_length": 139, "canonical_nll": 1.1587770311094874, "generation_rouge_l_f1": 0.23741007194244607, "generation_token_f1": 0.38848920863309355, "prompt_id": "1c2356a54b0c56f0", "reading_canonical_log_prob": -161.07000732421875, "reading_distances": [31, 27, 25, 29], "reading_log_probs": [-165.9300079345703, -173.41815185546875, -164.44146728515625, -183.08358764648438], "reading_logprob_std": 8.534554841200043, "tokenization_path_logprob_std": 175.8292303029964, "trajectory_balance_rmse": 3420.83378912669} {"canonical_length": 150, "canonical_nll": 1.515656941731771, "generation_rouge_l_f1": 0.23076923076923078, "generation_token_f1": 0.4567307692307692, "prompt_id": "55d3c9fe027b6cf4", "reading_canonical_log_prob": -227.34854125976562, "reading_distances": [20, 18, 19, 24], "reading_log_probs": [-230.43992614746094, -241.7742919921875, -249.31277465820312, -242.28741455078125], "reading_logprob_std": 7.807384014418911, "tokenization_path_logprob_std": 77.68381387615412, "trajectory_balance_rmse": 3226.972473236403} {"canonical_length": 22, "canonical_nll": 2.545736486261541, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "99817bb6bf526c93", "reading_canonical_log_prob": -56.006202697753906, "reading_distances": [17, 20, 20, 23], "reading_log_probs": [-58.39988708496094, -54.673316955566406, -56.29883575439453, -55.60900115966797], "reading_logprob_std": 1.583363348861088, "tokenization_path_logprob_std": 44.09378903512777, "trajectory_balance_rmse": 431.85309119160365} {"canonical_length": 177, "canonical_nll": 1.5960014041534252, "generation_rouge_l_f1": 0.2705570291777188, "generation_token_f1": 0.5145888594164456, "prompt_id": "a99bf80957c76958", "reading_canonical_log_prob": -282.49224853515625, "reading_distances": [22, 28, 30, 26], "reading_log_probs": [-298.33563232421875, -298.7603759765625, -299.1532287597656, -295.33734130859375], "reading_logprob_std": 1.7385606307949166, "tokenization_path_logprob_std": 357.14928320749425, "trajectory_balance_rmse": 3910.9347710417874} {"canonical_length": 333, "canonical_nll": 1.696397019578172, "generation_rouge_l_f1": 0.31045751633986934, "generation_token_f1": 0.5392156862745099, "prompt_id": "6b033b3db0a5c2fe", "reading_canonical_log_prob": -564.9002075195312, "reading_distances": [23, 27, 26, 29], "reading_log_probs": [-566.0426635742188, -570.3458862304688, -577.066650390625, -579.2728271484375], "reading_logprob_std": 6.088304359550122, "tokenization_path_logprob_std": 210.61600392389207, "trajectory_balance_rmse": 7576.22399235756} {"canonical_length": 190, "canonical_nll": 1.1783792596114309, "generation_rouge_l_f1": 0.2924731182795699, "generation_token_f1": 0.4903225806451613, "prompt_id": "f5608ead26273538", "reading_canonical_log_prob": -223.89205932617188, "reading_distances": [34, 36, 38, 46], "reading_log_probs": [-233.48031616210938, -230.24208068847656, -232.18441772460938, -230.01864624023438], "reading_logprob_std": 1.649793992901066, "tokenization_path_logprob_std": 55.41033441355598, "trajectory_balance_rmse": 3482.1095419034204} {"canonical_length": 305, "canonical_nll": 1.1722990442494876, "generation_rouge_l_f1": 0.225, "generation_token_f1": 0.4857142857142857, "prompt_id": "eb21c6c6320f2679", "reading_canonical_log_prob": -357.55120849609375, "reading_distances": [32, 25, 34, 31], "reading_log_probs": [-371.1248779296875, -360.7789001464844, -372.76483154296875, -366.83441162109375], "reading_logprob_std": 5.3513118049031725, "tokenization_path_logprob_std": 281.4773792629559, "trajectory_balance_rmse": 6428.302079176631} {"canonical_length": 222, "canonical_nll": 0.8548471949121974, "generation_rouge_l_f1": 0.3463203463203463, "generation_token_f1": 0.593073593073593, "prompt_id": "bd9aa33662e77c04", "reading_canonical_log_prob": -189.7760772705078, "reading_distances": [18, 17, 19, 19], "reading_log_probs": [-184.9100799560547, -184.5764617919922, -180.5577850341797, -179.66845703125], "reading_logprob_std": 2.7011966632749806, "tokenization_path_logprob_std": 112.11760608216704, "trajectory_balance_rmse": 2050.867201226591} {"canonical_length": 85, "canonical_nll": 0.8100515926585478, "generation_rouge_l_f1": 0.18960244648318042, "generation_token_f1": 0.27522935779816515, "prompt_id": "4fcfacdaa4123cbd", "reading_canonical_log_prob": -68.85438537597656, "reading_distances": [28, 26, 23, 26], "reading_log_probs": [-66.88584899902344, -66.60953521728516, -65.55970764160156, -66.3616943359375], "reading_logprob_std": 0.5712912286874602, "tokenization_path_logprob_std": 159.303311872414, "trajectory_balance_rmse": 1138.0244376890057} {"canonical_length": 214, "canonical_nll": 1.7198041398948598, "generation_rouge_l_f1": 0.24686192468619245, "generation_token_f1": 0.4476987447698745, "prompt_id": "42e23f667d7b63a7", "reading_canonical_log_prob": -368.0380859375, "reading_distances": [17, 18, 22, 22], "reading_log_probs": [-369.36102294921875, -368.1517639160156, -376.07415771484375, -373.9488525390625], "reading_logprob_std": 3.7468225230814425, "tokenization_path_logprob_std": 170.9452556068227, "trajectory_balance_rmse": 4510.35572710611} {"canonical_length": 13, "canonical_nll": 1.3869056701660156, "generation_rouge_l_f1": 0.06716417910447761, "generation_token_f1": 0.08208955223880596, "prompt_id": "f4ba682f79ee92c5", "reading_canonical_log_prob": -18.029773712158203, "reading_distances": [12, 12, 19, 18], "reading_log_probs": [-16.655616760253906, -32.2617073059082, -18.60418128967285, -17.49492645263672], "reading_logprob_std": 7.381664709210115, "tokenization_path_logprob_std": 55.20570279680431, "trajectory_balance_rmse": 293.5817475624154} {"canonical_length": 112, "canonical_nll": 1.2805886949811662, "generation_rouge_l_f1": 0.27884615384615385, "generation_token_f1": 0.4903846153846154, "prompt_id": "8c5355410cf1d176", "reading_canonical_log_prob": -143.42593383789062, "reading_distances": [22, 20, 30, 24], "reading_log_probs": [-149.59423828125, -152.33961486816406, -150.70138549804688, -150.73773193359375], "reading_logprob_std": 1.1299554108144614, "tokenization_path_logprob_std": 101.64815945475254, "trajectory_balance_rmse": 2248.1789259936163} {"canonical_length": 255, "canonical_nll": 1.2821111940870098, "generation_rouge_l_f1": 0.2256214149139579, "generation_token_f1": 0.4397705544933079, "prompt_id": "131fb5156c48d9ba", "reading_canonical_log_prob": -326.9383544921875, "reading_distances": [22, 25, 26, 26], "reading_log_probs": [-328.83111572265625, -331.2213134765625, -327.53741455078125, -328.017822265625], "reading_logprob_std": 1.6358589252170281, "tokenization_path_logprob_std": 280.2611433844345, "trajectory_balance_rmse": 5824.046435119505} {"canonical_length": 17, "canonical_nll": 1.3566809261546415, "generation_rouge_l_f1": 0.19999999999999998, "generation_token_f1": 0.26666666666666666, "prompt_id": "20a85de31d07820a", "reading_canonical_log_prob": -23.063575744628906, "reading_distances": [17, 18, 17, 23], "reading_log_probs": [-22.897470474243164, -23.422075271606445, -24.34431266784668, -22.07070541381836], "reading_logprob_std": 0.95298282797049, "tokenization_path_logprob_std": 81.76914432000758, "trajectory_balance_rmse": 269.7466104102021} {"canonical_length": 236, "canonical_nll": 1.1453878111758475, "generation_rouge_l_f1": 0.22903885480572594, "generation_token_f1": 0.49079754601226994, "prompt_id": "8bdd0bdfad8b9b1b", "reading_canonical_log_prob": -270.3115234375, "reading_distances": [20, 18, 17, 16], "reading_log_probs": [-278.3125, -279.6015319824219, -283.6059875488281, -277.90740966796875], "reading_logprob_std": 2.6017016342239008, "tokenization_path_logprob_std": 166.03715250164177, "trajectory_balance_rmse": 5380.344927676016} {"canonical_length": 314, "canonical_nll": 1.2700247794959196, "generation_rouge_l_f1": 0.23793103448275862, "generation_token_f1": 0.5103448275862069, "prompt_id": "f543bc422c7c86fd", "reading_canonical_log_prob": -398.78778076171875, "reading_distances": [23, 23, 27, 26], "reading_log_probs": [-400.23614501953125, -401.47052001953125, -405.6004333496094, -403.10614013671875], "reading_logprob_std": 2.3182083386222923, "tokenization_path_logprob_std": 239.22292958002578, "trajectory_balance_rmse": 7883.84626996252} {"canonical_length": 17, "canonical_nll": 0.8973639432121726, "generation_rouge_l_f1": 0.3448275862068965, "generation_token_f1": 0.5517241379310344, "prompt_id": "6b9813758e34b6e5", "reading_canonical_log_prob": -15.255187034606934, "reading_distances": [28, 34, 32, 32], "reading_log_probs": [-12.542182922363281, -13.586296081542969, -14.055156707763672, -13.551521301269531], "reading_logprob_std": 0.6372279891877871, "tokenization_path_logprob_std": 62.60878944523789, "trajectory_balance_rmse": 510.2942592440994} {"canonical_length": 133, "canonical_nll": 1.3821170204564144, "generation_rouge_l_f1": 0.196969696969697, "generation_token_f1": 0.30303030303030304, "prompt_id": "89680ea889169bf7", "reading_canonical_log_prob": -183.82156372070312, "reading_distances": [24, 22, 25, 25], "reading_log_probs": [-173.98858642578125, -183.6540069580078, -178.11331176757812, -182.1690216064453], "reading_logprob_std": 4.346516804099044, "tokenization_path_logprob_std": 73.10143519641157, "trajectory_balance_rmse": 3062.383368286575} {"canonical_length": 15, "canonical_nll": 1.1090807596842447, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.33333333333333337, "prompt_id": "5d244886eb8450ed", "reading_canonical_log_prob": -16.636211395263672, "reading_distances": [33, 28, 32, 31], "reading_log_probs": [-12.407360076904297, -11.857039451599121, -10.202251434326172, -14.831377983093262], "reading_logprob_std": 1.9160516131764935, "tokenization_path_logprob_std": 67.75254195066387, "trajectory_balance_rmse": 468.59702365593085} {"canonical_length": 245, "canonical_nll": 0.9347950838050064, "generation_rouge_l_f1": 0.23908918406072105, "generation_token_f1": 0.48956356736242884, "prompt_id": "8dfd747741dad44f", "reading_canonical_log_prob": -229.02479553222656, "reading_distances": [21, 20, 17, 20], "reading_log_probs": [-242.17208862304688, -236.08309936523438, -233.77525329589844, -238.16497802734375], "reading_logprob_std": 3.565688625410228, "tokenization_path_logprob_std": 161.77059510265263, "trajectory_balance_rmse": 4643.617243836411} {"canonical_length": 18, "canonical_nll": 0.7259901894463433, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "73b880653e3d865c", "reading_canonical_log_prob": -13.06782341003418, "reading_distances": [18, 22, 27, 17], "reading_log_probs": [-16.67378044128418, -15.827432632446289, -14.263797760009766, -15.036516189575195], "reading_logprob_std": 1.0357184291103207, "tokenization_path_logprob_std": 39.546754594642735, "trajectory_balance_rmse": 321.9994176554681} {"canonical_length": 18, "canonical_nll": 1.559250725640191, "generation_rouge_l_f1": 0.3448275862068966, "generation_token_f1": 0.41379310344827586, "prompt_id": "50d05075f2282d29", "reading_canonical_log_prob": -28.066513061523438, "reading_distances": [22, 24, 29, 30], "reading_log_probs": [-31.27535057067871, -30.758337020874023, -28.05164337158203, -28.194704055786133], "reading_logprob_std": 1.6849541268749988, "tokenization_path_logprob_std": 29.905769710057886, "trajectory_balance_rmse": 369.20235290709456} {"canonical_length": 12, "canonical_nll": 0.8653172651926676, "generation_rouge_l_f1": 0.32352941176470584, "generation_token_f1": 0.32352941176470584, "prompt_id": "4907b79891256514", "reading_canonical_log_prob": -10.383807182312012, "reading_distances": [36, 32, 29, 40], "reading_log_probs": [-12.319332122802734, -10.178464889526367, -10.913588523864746, -13.021928787231445], "reading_logprob_std": 1.2949859076062717, "tokenization_path_logprob_std": 34.09672825811791, "trajectory_balance_rmse": 226.08113083212066} {"canonical_length": 58, "canonical_nll": 2.4448268495757004, "generation_rouge_l_f1": 0.2117647058823529, "generation_token_f1": 0.2823529411764706, "prompt_id": "d9bc5b8ac130964c", "reading_canonical_log_prob": -141.79995727539062, "reading_distances": [13, 12, 10, 14], "reading_log_probs": [-141.10629272460938, -139.79226684570312, -145.509765625, -142.27639770507812], "reading_logprob_std": 2.4461151135281174, "tokenization_path_logprob_std": 47.79943009698913, "trajectory_balance_rmse": 1315.8510953655461} {"canonical_length": 14, "canonical_nll": 0.7322731018066406, "generation_rouge_l_f1": 0.6923076923076924, "generation_token_f1": 0.8461538461538461, "prompt_id": "0946ba5ceddc203c", "reading_canonical_log_prob": -10.251823425292969, "reading_distances": [17, 14, 12, 15], "reading_log_probs": [-10.685576438903809, -11.066291809082031, -11.036063194274902, -10.961051940917969], "reading_logprob_std": 0.1735146002802494, "tokenization_path_logprob_std": 4.672314063668438, "trajectory_balance_rmse": 189.69239457584007} {"canonical_length": 17, "canonical_nll": 1.3605977226706112, "generation_rouge_l_f1": 0.10596026490066225, "generation_token_f1": 0.11258278145695365, "prompt_id": "d08a5a0e92cb2564", "reading_canonical_log_prob": -23.13016128540039, "reading_distances": [27, 16, 21, 20], "reading_log_probs": [-29.574234008789062, -29.47311019897461, -26.821578979492188, -28.53359603881836], "reading_logprob_std": 1.2752310312324258, "tokenization_path_logprob_std": 61.73379374310965, "trajectory_balance_rmse": 291.0083704302567} {"canonical_length": 65, "canonical_nll": 0.6336263803335337, "generation_rouge_l_f1": 0.3067484662576687, "generation_token_f1": 0.5889570552147239, "prompt_id": "02aff4ae4d6fb42e", "reading_canonical_log_prob": -41.18571472167969, "reading_distances": [18, 26, 25, 20], "reading_log_probs": [-43.30072784423828, -44.650604248046875, -44.299163818359375, -42.775535583496094], "reading_logprob_std": 0.8687013046877137, "tokenization_path_logprob_std": 67.26241280391689, "trajectory_balance_rmse": 1048.8298120080442} {"canonical_length": 281, "canonical_nll": 1.1809837911476868, "generation_rouge_l_f1": 0.19366852886405958, "generation_token_f1": 0.4245810055865922, "prompt_id": "a79f040945146e56", "reading_canonical_log_prob": -331.8564453125, "reading_distances": [35, 35, 41, 35], "reading_log_probs": [-347.1697998046875, -346.943115234375, -347.0458068847656, -344.63824462890625], "reading_logprob_std": 1.21088334992427, "tokenization_path_logprob_std": 354.4911689708359, "trajectory_balance_rmse": 5801.8734880486} {"canonical_length": 341, "canonical_nll": 1.4031592226447946, "generation_rouge_l_f1": 0.21201413427561838, "generation_token_f1": 0.3745583038869258, "prompt_id": "79e68a828bcf94f4", "reading_canonical_log_prob": -478.477294921875, "reading_distances": [30, 25, 29, 23], "reading_log_probs": [-474.7095947265625, -475.2646484375, -477.16357421875, -473.48760986328125], "reading_logprob_std": 1.5302463346311523, "tokenization_path_logprob_std": 296.3151343324492, "trajectory_balance_rmse": 7343.120469039681} {"canonical_length": 50, "canonical_nll": 0.6913883209228515, "generation_rouge_l_f1": 0.4888888888888889, "generation_token_f1": 0.5777777777777778, "prompt_id": "5debb498a4c42600", "reading_canonical_log_prob": -34.56941604614258, "reading_distances": [33, 32, 30, 30], "reading_log_probs": [-32.92588424682617, -34.63972854614258, -33.81696701049805, -34.4069938659668], "reading_logprob_std": 0.7639922634856078, "tokenization_path_logprob_std": 82.56775780567311, "trajectory_balance_rmse": 981.0077974420385} {"canonical_length": 34, "canonical_nll": 1.363683027379653, "generation_rouge_l_f1": 0.3928571428571428, "generation_token_f1": 0.46428571428571425, "prompt_id": "fa77d10032456ab0", "reading_canonical_log_prob": -46.3652229309082, "reading_distances": [27, 28, 32, 27], "reading_log_probs": [-53.90747833251953, -61.8238525390625, -69.48308563232422, -49.64945983886719], "reading_logprob_std": 8.77331012284038, "tokenization_path_logprob_std": 122.04912564406723, "trajectory_balance_rmse": 919.6694216776003} {"canonical_length": 8, "canonical_nll": 0.3674553632736206, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.8571428571428571, "prompt_id": "253388f751e77ae9", "reading_canonical_log_prob": -2.939642906188965, "reading_distances": [48, 51, 45, 55], "reading_log_probs": [-5.19413423538208, -4.698676109313965, -5.111936569213867, -8.495677947998047], "reading_logprob_std": 1.7604483090609735, "tokenization_path_logprob_std": 32.23773522409397, "trajectory_balance_rmse": 136.88551541885036} {"canonical_length": 51, "canonical_nll": 1.2161248898973651, "generation_rouge_l_f1": 0.25263157894736843, "generation_token_f1": 0.4210526315789474, "prompt_id": "0182330e9a378205", "reading_canonical_log_prob": -62.022369384765625, "reading_distances": [25, 22, 24, 26], "reading_log_probs": [-67.08851623535156, -69.28956604003906, -65.40241241455078, -68.74669647216797], "reading_logprob_std": 1.7565599564893428, "tokenization_path_logprob_std": 96.25818698309445, "trajectory_balance_rmse": 795.536413261808} {"canonical_length": 111, "canonical_nll": 1.7568059697881475, "generation_rouge_l_f1": 0.2608695652173913, "generation_token_f1": 0.4505928853754941, "prompt_id": "8709e23cb34a4482", "reading_canonical_log_prob": -195.00546264648438, "reading_distances": [22, 24, 26, 28], "reading_log_probs": [-191.25308227539062, -191.08982849121094, -190.04443359375, -195.2314453125], "reading_logprob_std": 2.281552593542214, "tokenization_path_logprob_std": 137.7208015307521, "trajectory_balance_rmse": 2962.566598271905} {"canonical_length": 6, "canonical_nll": 0.36980171998341876, "generation_rouge_l_f1": 0.888888888888889, "generation_token_f1": 0.888888888888889, "prompt_id": "0852cc501290063e", "reading_canonical_log_prob": -2.2188103199005127, "reading_distances": [25, 34, 28, 28], "reading_log_probs": [-3.157588005065918, -2.6148877143859863, -2.187561273574829, -2.8632590770721436], "reading_logprob_std": 0.41058583406673704, "tokenization_path_logprob_std": 22.118937743541583, "trajectory_balance_rmse": 81.14182964503028} {"canonical_length": 12, "canonical_nll": 1.498209794362386, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.3157894736842105, "prompt_id": "0108b707d4dec974", "reading_canonical_log_prob": -17.978517532348633, "reading_distances": [35, 40, 31, 43], "reading_log_probs": [-19.141029357910156, -18.900278091430664, -17.45411491394043, -20.223487854003906], "reading_logprob_std": 1.1397023150149168, "tokenization_path_logprob_std": 69.08389368177615, "trajectory_balance_rmse": 321.64492525146585} {"canonical_length": 104, "canonical_nll": 1.715912598830003, "generation_rouge_l_f1": 0.25409836065573776, "generation_token_f1": 0.4344262295081967, "prompt_id": "5ffba76c8fed8b0d", "reading_canonical_log_prob": -178.4549102783203, "reading_distances": [18, 19, 19, 19], "reading_log_probs": [-185.468505859375, -184.4597625732422, -176.12255859375, -179.789794921875], "reading_logprob_std": 4.333765121766135, "tokenization_path_logprob_std": 92.07342633209605, "trajectory_balance_rmse": 2381.202584712232} {"canonical_length": 245, "canonical_nll": 1.1049873196348852, "generation_rouge_l_f1": 0.36217303822937624, "generation_token_f1": 0.6519114688128772, "prompt_id": "66d201ed6635605c", "reading_canonical_log_prob": -270.7218933105469, "reading_distances": [43, 40, 50, 36], "reading_log_probs": [-277.3688659667969, -278.78582763671875, -281.055908203125, -281.0821533203125], "reading_logprob_std": 1.8215750973987268, "tokenization_path_logprob_std": 363.77966495654465, "trajectory_balance_rmse": 5088.744270485147} {"canonical_length": 62, "canonical_nll": 1.01577881843813, "generation_rouge_l_f1": 0.26086956521739135, "generation_token_f1": 0.28985507246376807, "prompt_id": "d9f443de104a8944", "reading_canonical_log_prob": -62.97828674316406, "reading_distances": [22, 17, 18, 17], "reading_log_probs": [-66.46078491210938, -65.38833618164062, -66.2870101928711, -65.78300476074219], "reading_logprob_std": 0.48794890507137606, "tokenization_path_logprob_std": 34.06616030511122, "trajectory_balance_rmse": 594.9655806881282} {"canonical_length": 19, "canonical_nll": 1.8187428524619655, "generation_rouge_l_f1": 0.08888888888888889, "generation_token_f1": 0.0962962962962963, "prompt_id": "d416f7f8cc539e7b", "reading_canonical_log_prob": -34.556114196777344, "reading_distances": [23, 30, 33, 28], "reading_log_probs": [-34.26998519897461, -43.48752975463867, -48.799468994140625, -38.44269943237305], "reading_logprob_std": 6.287621823096926, "tokenization_path_logprob_std": 103.41525373318127, "trajectory_balance_rmse": 564.2519687493531} {"canonical_length": 12, "canonical_nll": 1.2161842981974285, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.35714285714285715, "prompt_id": "111f88e5ddbba7a9", "reading_canonical_log_prob": -14.59421157836914, "reading_distances": [24, 22, 18, 19], "reading_log_probs": [-41.281158447265625, -41.122642517089844, -31.454587936401367, -42.860816955566406], "reading_logprob_std": 5.209578729764217, "tokenization_path_logprob_std": 25.357789300471882, "trajectory_balance_rmse": 269.0835373050113} {"canonical_length": 11, "canonical_nll": 1.4436532800847834, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.4444444444444444, "prompt_id": "47566df03369ba07", "reading_canonical_log_prob": -15.880186080932617, "reading_distances": [50, 40, 54, 48], "reading_log_probs": [-19.309391021728516, -21.513334274291992, -17.261211395263672, -23.931808471679688], "reading_logprob_std": 2.870040367051697, "tokenization_path_logprob_std": 31.707789200878665, "trajectory_balance_rmse": 203.42860149787612} {"canonical_length": 26, "canonical_nll": 0.9156254254854642, "generation_rouge_l_f1": 0.48387096774193544, "generation_token_f1": 0.5161290322580646, "prompt_id": "0ad3db1386b28ee8", "reading_canonical_log_prob": -23.80626106262207, "reading_distances": [30, 31, 33, 37], "reading_log_probs": [-25.055233001708984, -25.598793029785156, -26.338407516479492, -25.736003875732422], "reading_logprob_std": 0.5271139878271387, "tokenization_path_logprob_std": 59.024661943457915, "trajectory_balance_rmse": 596.1141227022534} {"canonical_length": 104, "canonical_nll": 1.790935076200045, "generation_rouge_l_f1": 0.15425531914893617, "generation_token_f1": 0.3191489361702128, "prompt_id": "136b8765ce15fa6a", "reading_canonical_log_prob": -186.2572479248047, "reading_distances": [18, 21, 18, 21], "reading_log_probs": [-197.2946014404297, -190.26409912109375, -197.63418579101562, -194.467529296875], "reading_logprob_std": 3.410160472946071, "tokenization_path_logprob_std": 80.25458859246196, "trajectory_balance_rmse": 2079.677797822536} {"canonical_length": 35, "canonical_nll": 0.714876229422433, "generation_rouge_l_f1": 0.3114754098360656, "generation_token_f1": 0.40983606557377056, "prompt_id": "92e8091e1c9b04af", "reading_canonical_log_prob": -25.020668029785156, "reading_distances": [28, 31, 24, 38], "reading_log_probs": [-22.74212074279785, -31.867721557617188, -31.720932006835938, -25.685321807861328], "reading_logprob_std": 4.539000245722303, "tokenization_path_logprob_std": 126.39074999770406, "trajectory_balance_rmse": 711.1734877969216} {"canonical_length": 20, "canonical_nll": 2.298870849609375, "generation_rouge_l_f1": 0.20512820512820512, "generation_token_f1": 0.358974358974359, "prompt_id": "3bfad973aaa202e7", "reading_canonical_log_prob": -45.9774169921875, "reading_distances": [21, 15, 21, 17], "reading_log_probs": [-44.786346435546875, -40.74661636352539, -43.43976974487305, -45.23225402832031], "reading_logprob_std": 2.019045602127616, "tokenization_path_logprob_std": 117.16076948258501, "trajectory_balance_rmse": 536.2226764704986} {"canonical_length": 292, "canonical_nll": 0.9874105584131528, "generation_rouge_l_f1": 0.2961538461538462, "generation_token_f1": 0.55, "prompt_id": "be5dbfe122219a1b", "reading_canonical_log_prob": -288.3238830566406, "reading_distances": [18, 28, 24, 21], "reading_log_probs": [-297.01263427734375, -299.8280029296875, -296.1871337890625, -297.148193359375], "reading_logprob_std": 1.5807980816520268, "tokenization_path_logprob_std": 167.57170688168105, "trajectory_balance_rmse": 6144.9718608720505} {"canonical_length": 39, "canonical_nll": 1.3118687165089142, "generation_rouge_l_f1": 0.1496062992125984, "generation_token_f1": 0.2047244094488189, "prompt_id": "6acd370678d1b813", "reading_canonical_log_prob": -51.162879943847656, "reading_distances": [16, 16, 14, 17], "reading_log_probs": [-53.441261291503906, -54.356021881103516, -52.220375061035156, -56.29255676269531], "reading_logprob_std": 1.7163686105431795, "tokenization_path_logprob_std": 30.54395816997974, "trajectory_balance_rmse": 358.8520653078272} {"canonical_length": 197, "canonical_nll": 1.2082701552337802, "generation_rouge_l_f1": 0.31147540983606553, "generation_token_f1": 0.5191256830601093, "prompt_id": "2bb939d56ab15ee3", "reading_canonical_log_prob": -238.0292205810547, "reading_distances": [22, 20, 21, 20], "reading_log_probs": [-242.06504821777344, -242.83242797851562, -241.3851318359375, -245.361328125], "reading_logprob_std": 1.7372573742980575, "tokenization_path_logprob_std": 445.8857877635834, "trajectory_balance_rmse": 4524.946767829334} {"canonical_length": 14, "canonical_nll": 1.3861558096749442, "generation_rouge_l_f1": 0.4666666666666667, "generation_token_f1": 0.4666666666666667, "prompt_id": "9d34cf3c82e91821", "reading_canonical_log_prob": -19.40618133544922, "reading_distances": [48, 30, 40, 42], "reading_log_probs": [-20.214488983154297, -21.08754539489746, -16.43211555480957, -20.136554718017578], "reading_logprob_std": 2.069116755538841, "tokenization_path_logprob_std": 54.27824812541373, "trajectory_balance_rmse": 211.67414804741912} {"canonical_length": 13, "canonical_nll": 0.12742819235875055, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "b8a06c7365e05ce4", "reading_canonical_log_prob": -1.6565665006637573, "reading_distances": [30, 24, 29, 22], "reading_log_probs": [-2.491306781768799, -3.8326704502105713, -3.12450909614563, -3.802642822265625], "reading_logprob_std": 0.6378390200111962, "tokenization_path_logprob_std": 30.446833047092447, "trajectory_balance_rmse": 277.4349132534592} {"canonical_length": 44, "canonical_nll": 1.5280099348588423, "generation_rouge_l_f1": 0.19047619047619047, "generation_token_f1": 0.22222222222222224, "prompt_id": "3c94fb62a787deaa", "reading_canonical_log_prob": -67.23243713378906, "reading_distances": [26, 19, 21, 24], "reading_log_probs": [-73.24627685546875, -70.61956024169922, -70.21127319335938, -74.35646057128906], "reading_logprob_std": 2.0136429687306605, "tokenization_path_logprob_std": 43.39021960920602, "trajectory_balance_rmse": 786.1304649280098} {"canonical_length": 230, "canonical_nll": 1.0475613801375678, "generation_rouge_l_f1": 0.2839506172839506, "generation_token_f1": 0.51440329218107, "prompt_id": "84f970de38e4a1fd", "reading_canonical_log_prob": -240.93911743164062, "reading_distances": [25, 25, 21, 29], "reading_log_probs": [-250.48866271972656, -247.0537109375, -248.12954711914062, -256.8578796386719], "reading_logprob_std": 4.391224016015738, "tokenization_path_logprob_std": 181.17856708360185, "trajectory_balance_rmse": 4028.401477784685} {"canonical_length": 43, "canonical_nll": 1.697984207508176, "generation_rouge_l_f1": 0.09375, "generation_token_f1": 0.15625, "prompt_id": "b9d9b86bb406a853", "reading_canonical_log_prob": -73.01332092285156, "reading_distances": [16, 18, 17, 12], "reading_log_probs": [-79.18098449707031, -77.78288269042969, -85.11553955078125, -79.67964935302734], "reading_logprob_std": 3.218912622194335, "tokenization_path_logprob_std": 77.75388346797452, "trajectory_balance_rmse": 956.94051281518} {"canonical_length": 217, "canonical_nll": 0.9794540053688436, "generation_rouge_l_f1": 0.33649289099526064, "generation_token_f1": 0.5687203791469193, "prompt_id": "2dfd04cd2fcb5555", "reading_canonical_log_prob": -212.54151916503906, "reading_distances": [42, 36, 40, 40], "reading_log_probs": [-223.21029663085938, -206.34573364257812, -210.4010009765625, -208.34780883789062], "reading_logprob_std": 7.605119873792863, "tokenization_path_logprob_std": 168.4705681161714, "trajectory_balance_rmse": 5017.890540054538} {"canonical_length": 193, "canonical_nll": 0.9812688481622409, "generation_rouge_l_f1": 0.3758389261744967, "generation_token_f1": 0.5548098434004474, "prompt_id": "daff264becf42549", "reading_canonical_log_prob": -189.3848876953125, "reading_distances": [16, 11, 14, 14], "reading_log_probs": [-195.08966064453125, -196.85183715820312, -195.08445739746094, -195.17417907714844], "reading_logprob_std": 0.8688428290924304, "tokenization_path_logprob_std": 167.5380741378698, "trajectory_balance_rmse": 4060.7530286747924} {"canonical_length": 254, "canonical_nll": 1.2778942678857037, "generation_rouge_l_f1": 0.1893491124260355, "generation_token_f1": 0.4260355029585799, "prompt_id": "f80fc3edc248f324", "reading_canonical_log_prob": -324.58514404296875, "reading_distances": [30, 28, 30, 27], "reading_log_probs": [-330.1671142578125, -332.6480712890625, -334.4849853515625, -327.97821044921875], "reading_logprob_std": 2.8447382952671556, "tokenization_path_logprob_std": 141.25040122758656, "trajectory_balance_rmse": 5945.652762217913} {"canonical_length": 72, "canonical_nll": 1.8363026512993708, "generation_rouge_l_f1": 0.26855123674911663, "generation_token_f1": 0.3745583038869258, "prompt_id": "7da5b774aaa5c557", "reading_canonical_log_prob": -132.2137908935547, "reading_distances": [25, 18, 24, 29], "reading_log_probs": [-136.07994079589844, -136.5620880126953, -132.4335479736328, -135.5257568359375], "reading_logprob_std": 1.8600256992980702, "tokenization_path_logprob_std": 128.49254957222865, "trajectory_balance_rmse": 1740.4843872192018} {"canonical_length": 138, "canonical_nll": 1.6266436369522759, "generation_rouge_l_f1": 0.2121212121212121, "generation_token_f1": 0.40404040404040403, "prompt_id": "0e6878e78767cf2b", "reading_canonical_log_prob": -224.47682189941406, "reading_distances": [16, 20, 24, 22], "reading_log_probs": [-228.5765380859375, -225.4202880859375, -228.85430908203125, -223.92062377929688], "reading_logprob_std": 2.4169428472106653, "tokenization_path_logprob_std": 81.44450481727851, "trajectory_balance_rmse": 2857.9063282628954} {"canonical_length": 152, "canonical_nll": 1.2619285583496094, "generation_rouge_l_f1": 0.20047732696897377, "generation_token_f1": 0.3436754176610978, "prompt_id": "0ac01e4320c0e1b4", "reading_canonical_log_prob": -191.81314086914062, "reading_distances": [29, 28, 26, 29], "reading_log_probs": [-199.49154663085938, -194.600341796875, -203.52078247070312, -200.9375762939453], "reading_logprob_std": 3.748970803271466, "tokenization_path_logprob_std": 73.490922701711, "trajectory_balance_rmse": 3114.331722354071} {"canonical_length": 259, "canonical_nll": 1.3122697631364624, "generation_rouge_l_f1": 0.16761904761904758, "generation_token_f1": 0.3961904761904762, "prompt_id": "cc034e8024f676e6", "reading_canonical_log_prob": -339.87786865234375, "reading_distances": [23, 15, 30, 24], "reading_log_probs": [-345.7435302734375, -345.6122131347656, -346.46484375, -346.5417175292969], "reading_logprob_std": 0.4805818679152741, "tokenization_path_logprob_std": 238.06169912767388, "trajectory_balance_rmse": 5373.977913077907} {"canonical_length": 253, "canonical_nll": 1.3334219107043601, "generation_rouge_l_f1": 0.22868217054263568, "generation_token_f1": 0.5038759689922481, "prompt_id": "e0ae35419a54967e", "reading_canonical_log_prob": -337.3557434082031, "reading_distances": [23, 23, 27, 30], "reading_log_probs": [-337.5723876953125, -336.4244384765625, -337.44586181640625, -332.44635009765625], "reading_logprob_std": 2.406132233047682, "tokenization_path_logprob_std": 153.66748705211913, "trajectory_balance_rmse": 4948.34486978512} {"canonical_length": 6, "canonical_nll": 0.988019863764445, "generation_rouge_l_f1": 0.8333333333333333, "generation_token_f1": 0.8333333333333333, "prompt_id": "9ad445292250739f", "reading_canonical_log_prob": -5.92811918258667, "reading_distances": [27, 19, 18, 27], "reading_log_probs": [-8.194297790527344, -6.735508918762207, -5.166739463806152, -8.853368759155273], "reading_logprob_std": 1.6397701160897522, "tokenization_path_logprob_std": 24.91740723658508, "trajectory_balance_rmse": 135.39510066275923} {"canonical_length": 21, "canonical_nll": 0.9873062315441313, "generation_rouge_l_f1": 0.4000000000000001, "generation_token_f1": 0.45, "prompt_id": "1a95335c2e86bbe5", "reading_canonical_log_prob": -20.733430862426758, "reading_distances": [17, 23, 23, 26], "reading_log_probs": [-29.867755889892578, -25.458763122558594, -25.051105499267578, -23.426166534423828], "reading_logprob_std": 2.754866480427432, "tokenization_path_logprob_std": 15.978862152294104, "trajectory_balance_rmse": 436.03793894986234} {"canonical_length": 25, "canonical_nll": 0.21876804351806642, "generation_rouge_l_f1": 0.19095477386934673, "generation_token_f1": 0.19095477386934673, "prompt_id": "8fe26f83cb600778", "reading_canonical_log_prob": -5.46920108795166, "reading_distances": [22, 23, 25, 23], "reading_log_probs": [-6.634970664978027, -5.633066177368164, -6.803696632385254, -5.728008270263672], "reading_logprob_std": 0.6049350485394414, "tokenization_path_logprob_std": 33.17988565621091, "trajectory_balance_rmse": 178.0223050985528} {"canonical_length": 17, "canonical_nll": 1.0483253703397863, "generation_rouge_l_f1": 0.16, "generation_token_f1": 0.16, "prompt_id": "8bd44913279a5599", "reading_canonical_log_prob": -17.821531295776367, "reading_distances": [14, 12, 11, 13], "reading_log_probs": [-30.552410125732422, -29.877592086791992, -22.68134117126465, -29.12914276123047], "reading_logprob_std": 3.6326659727883226, "tokenization_path_logprob_std": 13.067434949181461, "trajectory_balance_rmse": 252.71654133359772} {"canonical_length": 355, "canonical_nll": 1.2625761649977993, "generation_rouge_l_f1": 0.21453287197231832, "generation_token_f1": 0.4186851211072664, "prompt_id": "e131aed6216cf197", "reading_canonical_log_prob": -448.21453857421875, "reading_distances": [45, 39, 43, 45], "reading_log_probs": [-452.9766845703125, -454.15374755859375, -453.4085998535156, -454.3211364746094], "reading_logprob_std": 0.6321633655972426, "tokenization_path_logprob_std": 64.53184327968142, "trajectory_balance_rmse": 7762.968248952753} {"canonical_length": 46, "canonical_nll": 1.3621428945790166, "generation_rouge_l_f1": 0.14925373134328357, "generation_token_f1": 0.17910447761194032, "prompt_id": "87d0cec84789988b", "reading_canonical_log_prob": -62.658573150634766, "reading_distances": [15, 21, 25, 22], "reading_log_probs": [-72.33589935302734, -70.92578125, -72.84111022949219, -71.53834533691406], "reading_logprob_std": 0.8475788915294636, "tokenization_path_logprob_std": 104.81580161800497, "trajectory_balance_rmse": 982.8742187517121} {"canonical_length": 138, "canonical_nll": 0.9922979603643003, "generation_rouge_l_f1": 0.2943722943722944, "generation_token_f1": 0.49350649350649345, "prompt_id": "d60af44d5acef8d0", "reading_canonical_log_prob": -136.93711853027344, "reading_distances": [28, 34, 31, 35], "reading_log_probs": [-144.61868286132812, -143.4129180908203, -140.14564514160156, -143.2681884765625], "reading_logprob_std": 1.909015153419572, "tokenization_path_logprob_std": 161.79547197945305, "trajectory_balance_rmse": 3030.8365899026494} {"canonical_length": 35, "canonical_nll": 1.860331508091518, "generation_rouge_l_f1": 0.20588235294117646, "generation_token_f1": 0.20588235294117646, "prompt_id": "a9f740ec35ab9678", "reading_canonical_log_prob": -65.11160278320312, "reading_distances": [14, 15, 17, 11], "reading_log_probs": [-66.97135925292969, -68.10646057128906, -71.08074951171875, -66.57296752929688], "reading_logprob_std": 2.038225398058297, "tokenization_path_logprob_std": 123.13725790001487, "trajectory_balance_rmse": 727.5511563206212} {"canonical_length": 257, "canonical_nll": 1.359730286357004, "generation_rouge_l_f1": 0.0982456140350877, "generation_token_f1": 0.21052631578947367, "prompt_id": "a25427203b062343", "reading_canonical_log_prob": -349.45068359375, "reading_distances": [32, 33, 30, 31], "reading_log_probs": [-351.61151123046875, -353.3495788574219, -351.56182861328125, -349.970947265625], "reading_logprob_std": 1.38012388912162, "tokenization_path_logprob_std": 229.99539565314282, "trajectory_balance_rmse": 5270.161880355273} {"canonical_length": 28, "canonical_nll": 1.317697252546038, "generation_rouge_l_f1": 0.20168067226890754, "generation_token_f1": 0.25210084033613445, "prompt_id": "b0bf00b191d5ab96", "reading_canonical_log_prob": -36.89552307128906, "reading_distances": [26, 29, 27, 31], "reading_log_probs": [-38.598609924316406, -39.0526123046875, -37.49486541748047, -35.28867721557617], "reading_logprob_std": 1.6793122038503838, "tokenization_path_logprob_std": 49.54753202907772, "trajectory_balance_rmse": 798.0065986171602} {"canonical_length": 31, "canonical_nll": 0.9030148290818737, "generation_rouge_l_f1": 0.07692307692307693, "generation_token_f1": 0.07692307692307693, "prompt_id": "77df38b1cb264cc8", "reading_canonical_log_prob": -27.993459701538086, "reading_distances": [8, 10, 12, 10], "reading_log_probs": [-32.15764617919922, -34.61170959472656, -38.558170318603516, -30.51410675048828], "reading_logprob_std": 3.497168120569677, "tokenization_path_logprob_std": 77.68398536014895, "trajectory_balance_rmse": 684.8250900706412} {"canonical_length": 54, "canonical_nll": 1.7237749452944156, "generation_rouge_l_f1": 0.34108527131782945, "generation_token_f1": 0.49612403100775193, "prompt_id": "93f9306a97ca7347", "reading_canonical_log_prob": -93.08384704589844, "reading_distances": [13, 14, 13, 13], "reading_log_probs": [-94.07341003417969, -106.52723693847656, -108.16521453857422, -95.42245483398438], "reading_logprob_std": 7.325035618341917, "tokenization_path_logprob_std": 90.00201264193436, "trajectory_balance_rmse": 1128.0124420249367} {"canonical_length": 31, "canonical_nll": 1.287997830298639, "generation_rouge_l_f1": 0.27027027027027023, "generation_token_f1": 0.27027027027027023, "prompt_id": "1cc77327b9384870", "reading_canonical_log_prob": -39.92793273925781, "reading_distances": [26, 23, 26, 19], "reading_log_probs": [-33.69731140136719, -36.3967170715332, -33.89116668701172, -35.834716796875], "reading_logprob_std": 1.3621027474791778, "tokenization_path_logprob_std": 51.97660019670038, "trajectory_balance_rmse": 591.0746790501207} {"canonical_length": 182, "canonical_nll": 1.4068070296402817, "generation_rouge_l_f1": 0.29245283018867924, "generation_token_f1": 0.5424528301886793, "prompt_id": "0d40240cbf3216d7", "reading_canonical_log_prob": -256.03887939453125, "reading_distances": [27, 22, 27, 25], "reading_log_probs": [-263.459716796875, -262.57806396484375, -259.3365478515625, -267.5051574707031], "reading_logprob_std": 3.362207662029784, "tokenization_path_logprob_std": 212.61896449191048, "trajectory_balance_rmse": 3727.577590920414} {"canonical_length": 136, "canonical_nll": 1.7871273265165442, "generation_rouge_l_f1": 0.18673218673218672, "generation_token_f1": 0.32432432432432434, "prompt_id": "c887cead4589c177", "reading_canonical_log_prob": -243.04931640625, "reading_distances": [12, 12, 19, 17], "reading_log_probs": [-249.96627807617188, -253.6979522705078, -255.31138610839844, -255.841552734375], "reading_logprob_std": 2.653514851842913, "tokenization_path_logprob_std": 225.41454507513686, "trajectory_balance_rmse": 3123.749755567801} {"canonical_length": 169, "canonical_nll": 1.9518255651349852, "generation_rouge_l_f1": 0.30158730158730157, "generation_token_f1": 0.49206349206349204, "prompt_id": "5686c4dfdf197ade", "reading_canonical_log_prob": -329.8585205078125, "reading_distances": [13, 11, 11, 10], "reading_log_probs": [-330.36639404296875, -330.4832763671875, -333.52935791015625, -329.5208740234375], "reading_logprob_std": 1.7560770701108868, "tokenization_path_logprob_std": 170.93482268391097, "trajectory_balance_rmse": 3219.1475605392743} {"canonical_length": 348, "canonical_nll": 1.1830709172391343, "generation_rouge_l_f1": 0.18589743589743588, "generation_token_f1": 0.4198717948717948, "prompt_id": "497e17449fccf12b", "reading_canonical_log_prob": -411.70867919921875, "reading_distances": [35, 24, 33, 27], "reading_log_probs": [-415.3135986328125, -411.8149108886719, -417.37969970703125, -412.0138854980469], "reading_logprob_std": 2.695615637244495, "tokenization_path_logprob_std": 357.8904676795519, "trajectory_balance_rmse": 7651.279964721104} {"canonical_length": 38, "canonical_nll": 1.3171429885061163, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "faaabffbce1c5a62", "reading_canonical_log_prob": -50.05143356323242, "reading_distances": [35, 38, 39, 31], "reading_log_probs": [-54.91401290893555, -56.8781852722168, -55.811119079589844, -54.062294006347656], "reading_logprob_std": 1.2081135660116087, "tokenization_path_logprob_std": 73.49464690355977, "trajectory_balance_rmse": 780.4032106520709} {"canonical_length": 76, "canonical_nll": 0.24808020340768913, "generation_rouge_l_f1": 0.10810810810810813, "generation_token_f1": 0.11891891891891891, "prompt_id": "bf9932d84bc4ffec", "reading_canonical_log_prob": -18.854095458984375, "reading_distances": [21, 18, 17, 16], "reading_log_probs": [-30.265438079833984, -24.292678833007812, -21.57828140258789, -25.334823608398438], "reading_logprob_std": 3.6287868922074447, "tokenization_path_logprob_std": 25.73106522977166, "trajectory_balance_rmse": 350.99670556017566} {"canonical_length": 165, "canonical_nll": 1.867849823922822, "generation_rouge_l_f1": 0.27296587926509186, "generation_token_f1": 0.4934383202099737, "prompt_id": "d69cbb797502e917", "reading_canonical_log_prob": -308.1952209472656, "reading_distances": [18, 20, 13, 19], "reading_log_probs": [-311.6611328125, -305.28564453125, -309.6507263183594, -308.1605224609375], "reading_logprob_std": 2.6845612240013375, "tokenization_path_logprob_std": 184.29085209879796, "trajectory_balance_rmse": 4120.896379556914} {"canonical_length": 119, "canonical_nll": 1.4101903578814339, "generation_rouge_l_f1": 0.29201101928374656, "generation_token_f1": 0.42424242424242425, "prompt_id": "98ec80f0ace1a8dc", "reading_canonical_log_prob": -167.81265258789062, "reading_distances": [34, 25, 35, 36], "reading_log_probs": [-168.66848754882812, -165.8802490234375, -166.8165740966797, -171.53573608398438], "reading_logprob_std": 2.492600717929855, "tokenization_path_logprob_std": 222.9401743377049, "trajectory_balance_rmse": 2829.0939622977435} {"canonical_length": 14, "canonical_nll": 1.3715694972446986, "generation_rouge_l_f1": 0.0923076923076923, "generation_token_f1": 0.12307692307692307, "prompt_id": "e2aae622b59df148", "reading_canonical_log_prob": -19.20197296142578, "reading_distances": [33, 23, 26, 32], "reading_log_probs": [-29.24536895751953, -26.060197830200195, -32.471622467041016, -29.855079650878906], "reading_logprob_std": 2.6343810810641073, "tokenization_path_logprob_std": 52.316788783194, "trajectory_balance_rmse": 259.7392294786268} {"canonical_length": 10, "canonical_nll": 1.1643675804138183, "generation_rouge_l_f1": 0.6363636363636364, "generation_token_f1": 0.6363636363636364, "prompt_id": "aefa9cb1b8d05c05", "reading_canonical_log_prob": -11.643675804138184, "reading_distances": [33, 28, 33, 26], "reading_log_probs": [-17.008588790893555, -16.4467716217041, -14.900426864624023, -17.86250877380371], "reading_logprob_std": 1.246949197189402, "tokenization_path_logprob_std": 16.74208191145268, "trajectory_balance_rmse": 293.16644917316063} {"canonical_length": 121, "canonical_nll": 1.114362480226627, "generation_rouge_l_f1": 0.2701298701298701, "generation_token_f1": 0.43636363636363634, "prompt_id": "7467b21b2ac04b77", "reading_canonical_log_prob": -134.83786010742188, "reading_distances": [11, 5, 12, 9], "reading_log_probs": [-138.49862670898438, -135.14732360839844, -132.99386596679688, -133.4669952392578], "reading_logprob_std": 2.492249203340874, "tokenization_path_logprob_std": 117.67461578954881, "trajectory_balance_rmse": 2808.457606759669} {"canonical_length": 92, "canonical_nll": 1.474998142408288, "generation_rouge_l_f1": 0.22299651567944248, "generation_token_f1": 0.2787456445993031, "prompt_id": "80fb331e54c070fd", "reading_canonical_log_prob": -135.6998291015625, "reading_distances": [18, 21, 19, 23], "reading_log_probs": [-138.8749542236328, -135.6410369873047, -135.90060424804688, -136.95481872558594], "reading_logprob_std": 1.469032124191849, "tokenization_path_logprob_std": 27.87340980364162, "trajectory_balance_rmse": 1667.6597348223447} {"canonical_length": 288, "canonical_nll": 1.3016331990559895, "generation_rouge_l_f1": 0.29151291512915123, "generation_token_f1": 0.5498154981549814, "prompt_id": "1baa92d9506c1060", "reading_canonical_log_prob": -374.870361328125, "reading_distances": [17, 27, 17, 16], "reading_log_probs": [-373.921142578125, -377.1115417480469, -377.8565673828125, -382.03887939453125], "reading_logprob_std": 3.340275930711817, "tokenization_path_logprob_std": 49.18047928782511, "trajectory_balance_rmse": 6650.690616992576} {"canonical_length": 232, "canonical_nll": 1.576153722302667, "generation_rouge_l_f1": 0.20967741935483872, "generation_token_f1": 0.4596774193548387, "prompt_id": "23fd1d0b807b2a3d", "reading_canonical_log_prob": -365.66766357421875, "reading_distances": [27, 26, 23, 21], "reading_log_probs": [-358.09552001953125, -360.2049560546875, -358.168701171875, -364.8804931640625], "reading_logprob_std": 3.1825837170816245, "tokenization_path_logprob_std": 41.55308571989327, "trajectory_balance_rmse": 4975.6059844672145} {"canonical_length": 23, "canonical_nll": 1.2302491561226223, "generation_rouge_l_f1": 0.4285714285714286, "generation_token_f1": 0.4857142857142858, "prompt_id": "e7e3fd3a1ae85dcc", "reading_canonical_log_prob": -28.295730590820312, "reading_distances": [20, 26, 14, 28], "reading_log_probs": [-30.43441390991211, -30.187908172607422, -28.042165756225586, -28.66429328918457], "reading_logprob_std": 1.1629559613680913, "tokenization_path_logprob_std": 52.78023551766947, "trajectory_balance_rmse": 506.8623338725698} {"canonical_length": 11, "canonical_nll": 0.32504853335293854, "generation_rouge_l_f1": 0.7368421052631577, "generation_token_f1": 0.7368421052631577, "prompt_id": "7a61f927f0ede558", "reading_canonical_log_prob": -3.575533866882324, "reading_distances": [24, 18, 17, 20], "reading_log_probs": [-4.495759010314941, -7.139394283294678, -6.391578674316406, -1.1962664127349854], "reading_logprob_std": 2.651113879463929, "tokenization_path_logprob_std": 47.47567444904602, "trajectory_balance_rmse": 241.09480556564947} {"canonical_length": 244, "canonical_nll": 1.6285455422323258, "generation_rouge_l_f1": 0.24615384615384617, "generation_token_f1": 0.4692307692307693, "prompt_id": "d3428656e283f9ef", "reading_canonical_log_prob": -397.3651123046875, "reading_distances": [27, 34, 40, 40], "reading_log_probs": [-411.84442138671875, -411.938232421875, -413.08843994140625, -426.3785400390625], "reading_logprob_std": 7.06676027668645, "tokenization_path_logprob_std": 120.41322655171993, "trajectory_balance_rmse": 6008.574796490236} {"canonical_length": 18, "canonical_nll": 0.5807623863220215, "generation_rouge_l_f1": 0.05882352941176471, "generation_token_f1": 0.05882352941176471, "prompt_id": "f6a9e161c915b813", "reading_canonical_log_prob": -10.453722953796387, "reading_distances": [12, 15, 12, 19], "reading_log_probs": [-7.7885823249816895, -9.227909088134766, -8.642093658447266, -8.451546669006348], "reading_logprob_std": 0.5931477702064653, "tokenization_path_logprob_std": 23.052660197316857, "trajectory_balance_rmse": 346.29363777621336} {"canonical_length": 4, "canonical_nll": 1.1837563514709473, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "60d17bcb4782ac31", "reading_canonical_log_prob": -4.735025405883789, "reading_distances": [14, 12, 12, 19], "reading_log_probs": [-6.704819202423096, -5.466732025146484, -6.762348175048828, -10.844365119934082], "reading_logprob_std": 2.3440070942153555, "tokenization_path_logprob_std": 27.326652076055318, "trajectory_balance_rmse": 60.56946395095816} {"canonical_length": 12, "canonical_nll": 1.31163223584493, "generation_rouge_l_f1": 0.20253164556962025, "generation_token_f1": 0.2784810126582279, "prompt_id": "469967c1d14dd489", "reading_canonical_log_prob": -15.73958683013916, "reading_distances": [26, 23, 30, 26], "reading_log_probs": [-19.88237190246582, -17.277624130249023, -16.7691650390625, -16.299781799316406], "reading_logprob_std": 1.600696376469799, "tokenization_path_logprob_std": 35.97322518753308, "trajectory_balance_rmse": 201.56742205311983} {"canonical_length": 269, "canonical_nll": 1.347776391692321, "generation_rouge_l_f1": 0.2359154929577465, "generation_token_f1": 0.47535211267605637, "prompt_id": "812c384ebd7a8b2e", "reading_canonical_log_prob": -362.5518493652344, "reading_distances": [30, 30, 26, 37], "reading_log_probs": [-356.38226318359375, -356.6771240234375, -357.2724914550781, -357.21514892578125], "reading_logprob_std": 0.4301520360706938, "tokenization_path_logprob_std": 131.4854171457014, "trajectory_balance_rmse": 4319.784761286093} {"canonical_length": 99, "canonical_nll": 1.1420172489050664, "generation_rouge_l_f1": 0.2162162162162162, "generation_token_f1": 0.34594594594594597, "prompt_id": "40633ecaff73b957", "reading_canonical_log_prob": -113.05970764160156, "reading_distances": [19, 26, 23, 15], "reading_log_probs": [-117.4289321899414, -116.05750274658203, -118.8385238647461, -122.27763366699219], "reading_logprob_std": 2.671286468788155, "tokenization_path_logprob_std": 223.87780050261415, "trajectory_balance_rmse": 2531.208838083714} {"canonical_length": 210, "canonical_nll": 1.247254871186756, "generation_rouge_l_f1": 0.2743362831858407, "generation_token_f1": 0.4690265486725664, "prompt_id": "cd8cb6a38db69f38", "reading_canonical_log_prob": -261.92352294921875, "reading_distances": [16, 23, 20, 20], "reading_log_probs": [-263.4103698730469, -266.979736328125, -264.636474609375, -265.11895751953125], "reading_logprob_std": 1.4818103926297352, "tokenization_path_logprob_std": 125.80186892215308, "trajectory_balance_rmse": 4477.893313544925} {"canonical_length": 105, "canonical_nll": 1.2221865699404761, "generation_rouge_l_f1": 0.08695652173913043, "generation_token_f1": 0.13043478260869565, "prompt_id": "2de9ac9d5457eaa3", "reading_canonical_log_prob": -128.32958984375, "reading_distances": [30, 28, 31, 30], "reading_log_probs": [-133.68325805664062, -133.25067138671875, -135.5526885986328, -136.00015258789062], "reading_logprob_std": 1.3573582242875344, "tokenization_path_logprob_std": 67.88625824024307, "trajectory_balance_rmse": 1127.7385747098306} {"canonical_length": 99, "canonical_nll": 1.3057815089370266, "generation_rouge_l_f1": 0.180327868852459, "generation_token_f1": 0.31693989071038253, "prompt_id": "0b44770e70f57065", "reading_canonical_log_prob": -129.27236938476562, "reading_distances": [28, 23, 29, 32], "reading_log_probs": [-129.97579956054688, -131.27212524414062, -130.52035522460938, -130.40914916992188], "reading_logprob_std": 0.5390613513855975, "tokenization_path_logprob_std": 81.19178131909457, "trajectory_balance_rmse": 2120.336173852943} {"canonical_length": 111, "canonical_nll": 1.7720802925728463, "generation_rouge_l_f1": 0.2082191780821918, "generation_token_f1": 0.3616438356164384, "prompt_id": "f5df13177d4b7c6c", "reading_canonical_log_prob": -196.70091247558594, "reading_distances": [16, 19, 15, 16], "reading_log_probs": [-205.6563720703125, -202.98025512695312, -205.01609802246094, -201.5233917236328], "reading_logprob_std": 1.8955910216852605, "tokenization_path_logprob_std": 151.75115001248665, "trajectory_balance_rmse": 2357.1680320669475} {"canonical_length": 129, "canonical_nll": 1.9206151444782582, "generation_rouge_l_f1": 0.24352331606217612, "generation_token_f1": 0.383419689119171, "prompt_id": "b85464f813712359", "reading_canonical_log_prob": -247.7593536376953, "reading_distances": [41, 46, 44, 43], "reading_log_probs": [-242.435791015625, -242.5469512939453, -242.03341674804688, -245.1116485595703], "reading_logprob_std": 1.4039045591309691, "tokenization_path_logprob_std": 148.8454158845777, "trajectory_balance_rmse": 3237.2967923735337} {"canonical_length": 52, "canonical_nll": 1.3429298400878906, "generation_rouge_l_f1": 0.13745704467353953, "generation_token_f1": 0.20618556701030927, "prompt_id": "fa52ad4bc1072fe5", "reading_canonical_log_prob": -69.83235168457031, "reading_distances": [47, 48, 45, 59], "reading_log_probs": [-73.33645629882812, -77.7055435180664, -76.37361145019531, -67.52415466308594], "reading_logprob_std": 4.526253899237011, "tokenization_path_logprob_std": 27.905124351301232, "trajectory_balance_rmse": 805.9589594624061} {"canonical_length": 192, "canonical_nll": 1.2574830849965413, "generation_rouge_l_f1": 0.4372294372294372, "generation_token_f1": 0.6103896103896104, "prompt_id": "63e9efbaa63103ce", "reading_canonical_log_prob": -241.43675231933594, "reading_distances": [16, 24, 25, 23], "reading_log_probs": [-243.67640686035156, -243.56903076171875, -243.52203369140625, -244.94313049316406], "reading_logprob_std": 0.6800627655050084, "tokenization_path_logprob_std": 160.5875230514035, "trajectory_balance_rmse": 4328.583977893782} {"canonical_length": 265, "canonical_nll": 1.324787758881191, "generation_rouge_l_f1": 0.24196597353497165, "generation_token_f1": 0.4045368620037807, "prompt_id": "867f08ace4ea14ea", "reading_canonical_log_prob": -351.0687561035156, "reading_distances": [22, 25, 33, 21], "reading_log_probs": [-353.90106201171875, -346.7161865234375, -355.481201171875, -351.64080810546875], "reading_logprob_std": 3.8194024585694417, "tokenization_path_logprob_std": 72.47212917095297, "trajectory_balance_rmse": 5697.8336959156695} {"canonical_length": 6, "canonical_nll": 0.12491849064826965, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "126ead873a35c47e", "reading_canonical_log_prob": -0.7495109438896179, "reading_distances": [31, 37, 32, 27], "reading_log_probs": [-2.0938777923583984, -2.4753291606903076, -2.4735167026519775, -2.566157102584839], "reading_logprob_std": 0.2100622604554425, "tokenization_path_logprob_std": 32.55508787546521, "trajectory_balance_rmse": 102.75220044676023} {"canonical_length": 51, "canonical_nll": 2.7424884870940565, "generation_rouge_l_f1": 0.16176470588235295, "generation_token_f1": 0.27941176470588236, "prompt_id": "9800df60020cf919", "reading_canonical_log_prob": -139.86691284179688, "reading_distances": [51, 49, 47, 41], "reading_log_probs": [-142.82847595214844, -145.4696807861328, -142.90780639648438, -141.16445922851562], "reading_logprob_std": 1.7769027441863465, "tokenization_path_logprob_std": 154.5914165501321, "trajectory_balance_rmse": 1074.933906931178} {"canonical_length": 11, "canonical_nll": 0.8493843945589933, "generation_rouge_l_f1": 0.07092198581560284, "generation_token_f1": 0.07801418439716312, "prompt_id": "0aba661f60975419", "reading_canonical_log_prob": -9.343228340148926, "reading_distances": [12, 13, 20, 19], "reading_log_probs": [-5.793405532836914, -6.596480369567871, -7.233154296875, -7.685612678527832], "reading_logprob_std": 0.8213066338871483, "tokenization_path_logprob_std": 26.893170741802873, "trajectory_balance_rmse": 199.10045095566414} {"canonical_length": 245, "canonical_nll": 1.1670943279655612, "generation_rouge_l_f1": 0.31712473572938693, "generation_token_f1": 0.6596194503171248, "prompt_id": "a78682b2f45a9311", "reading_canonical_log_prob": -285.9381103515625, "reading_distances": [15, 17, 17, 15], "reading_log_probs": [-288.20404052734375, -288.44488525390625, -287.50701904296875, -289.1790771484375], "reading_logprob_std": 0.6897433472039077, "tokenization_path_logprob_std": 162.90130346679075, "trajectory_balance_rmse": 4928.875852801075} {"canonical_length": 41, "canonical_nll": 1.0329703354254, "generation_rouge_l_f1": 0.10676156583629892, "generation_token_f1": 0.13523131672597866, "prompt_id": "e970a5e256be139f", "reading_canonical_log_prob": -42.351783752441406, "reading_distances": [14, 16, 12, 15], "reading_log_probs": [-45.61189651489258, -38.96422576904297, -39.44526672363281, -35.68555450439453], "reading_logprob_std": 4.141936309165168, "tokenization_path_logprob_std": 38.29090634132755, "trajectory_balance_rmse": 320.2661047420579} {"canonical_length": 112, "canonical_nll": 1.5152403967721122, "generation_rouge_l_f1": 0.25837320574162675, "generation_token_f1": 0.5741626794258374, "prompt_id": "0b4d03b967102211", "reading_canonical_log_prob": -169.70692443847656, "reading_distances": [23, 28, 26, 24], "reading_log_probs": [-177.87408447265625, -178.115966796875, -176.06500244140625, -175.8497314453125], "reading_logprob_std": 1.1838464530120676, "tokenization_path_logprob_std": 208.68413712832955, "trajectory_balance_rmse": 2246.3650324827345} {"canonical_length": 354, "canonical_nll": 1.1864830922272245, "generation_rouge_l_f1": 0.23140495867768596, "generation_token_f1": 0.4727272727272727, "prompt_id": "e03c0c9ce1e8af04", "reading_canonical_log_prob": -420.0150146484375, "reading_distances": [11, 15, 15, 18], "reading_log_probs": [-416.63909912109375, -415.6044921875, -414.90118408203125, -417.2986145019531], "reading_logprob_std": 1.066070903424611, "tokenization_path_logprob_std": 238.1127150596475, "trajectory_balance_rmse": 7375.538161644955} {"canonical_length": 114, "canonical_nll": 2.050294976485403, "generation_rouge_l_f1": 0.12043010752688171, "generation_token_f1": 0.18064516129032257, "prompt_id": "6aab22322acba4ab", "reading_canonical_log_prob": -233.73362731933594, "reading_distances": [49, 44, 49, 43], "reading_log_probs": [-252.04971313476562, -242.0851593017578, -243.37803649902344, -247.4014434814453], "reading_logprob_std": 4.49276206997036, "tokenization_path_logprob_std": 76.11920671503619, "trajectory_balance_rmse": 2319.1995287834143} {"canonical_length": 72, "canonical_nll": 0.674685902065701, "generation_rouge_l_f1": 0.3770491803278689, "generation_token_f1": 0.5081967213114754, "prompt_id": "1498d74fffe04fb0", "reading_canonical_log_prob": -48.57738494873047, "reading_distances": [27, 29, 20, 31], "reading_log_probs": [-77.69330596923828, -83.66244506835938, -59.035980224609375, -84.21566772460938], "reading_logprob_std": 11.786479248837328, "tokenization_path_logprob_std": 91.58887984973131, "trajectory_balance_rmse": 1359.379136493017} {"canonical_length": 203, "canonical_nll": 1.6713295922490763, "generation_rouge_l_f1": 0.20219780219780217, "generation_token_f1": 0.43516483516483523, "prompt_id": "363570662fa8a4e6", "reading_canonical_log_prob": -339.2799072265625, "reading_distances": [31, 30, 29, 32], "reading_log_probs": [-357.77264404296875, -363.76318359375, -357.87103271484375, -361.52569580078125], "reading_logprob_std": 2.930614238975641, "tokenization_path_logprob_std": 159.12673521400546, "trajectory_balance_rmse": 4758.770268325458} {"canonical_length": 267, "canonical_nll": 1.6634076864978347, "generation_rouge_l_f1": 0.22879684418145957, "generation_token_f1": 0.5009861932938855, "prompt_id": "2b4383e7d7d3624e", "reading_canonical_log_prob": -444.1298522949219, "reading_distances": [25, 34, 34, 33], "reading_log_probs": [-455.50360107421875, -451.17669677734375, -458.16326904296875, -456.6822204589844], "reading_logprob_std": 3.006958126474933, "tokenization_path_logprob_std": 302.17487361687864, "trajectory_balance_rmse": 5891.867985047548} {"canonical_length": 72, "canonical_nll": 1.3436703152126737, "generation_rouge_l_f1": 0.25925925925925924, "generation_token_f1": 0.35802469135802467, "prompt_id": "fadbe59576c37aff", "reading_canonical_log_prob": -96.7442626953125, "reading_distances": [19, 22, 16, 20], "reading_log_probs": [-107.729248046875, -99.01010131835938, -97.67723083496094, -101.40804290771484], "reading_logprob_std": 4.4578566451736625, "tokenization_path_logprob_std": 71.34799348734246, "trajectory_balance_rmse": 1165.9536417659372} {"canonical_length": 119, "canonical_nll": 1.5834396746980042, "generation_rouge_l_f1": 0.21359223300970873, "generation_token_f1": 0.42718446601941745, "prompt_id": "6caa7e0c0b6b5ae9", "reading_canonical_log_prob": -188.4293212890625, "reading_distances": [26, 30, 24, 30], "reading_log_probs": [-197.8441162109375, -197.22705078125, -194.04360961914062, -198.08457946777344], "reading_logprob_std": 1.872643698871497, "tokenization_path_logprob_std": 98.49968588531561, "trajectory_balance_rmse": 2920.2048825548136} {"canonical_length": 246, "canonical_nll": 1.1773814379684324, "generation_rouge_l_f1": 0.217741935483871, "generation_token_f1": 0.41532258064516125, "prompt_id": "09ec255dc57dfab9", "reading_canonical_log_prob": -289.6358337402344, "reading_distances": [26, 27, 21, 29], "reading_log_probs": [-314.31866455078125, -307.95684814453125, -309.26690673828125, -321.58251953125], "reading_logprob_std": 6.176609418365568, "tokenization_path_logprob_std": 185.60002297362405, "trajectory_balance_rmse": 5408.772418798216} {"canonical_length": 83, "canonical_nll": 1.573542767260448, "generation_rouge_l_f1": 0.2831858407079646, "generation_token_f1": 0.37168141592920356, "prompt_id": "a8f654f305639b3f", "reading_canonical_log_prob": -130.6040496826172, "reading_distances": [17, 20, 13, 15], "reading_log_probs": [-153.70635986328125, -146.6162109375, -149.4910888671875, -150.2054443359375], "reading_logprob_std": 2.914803511454636, "tokenization_path_logprob_std": 43.02322447728502, "trajectory_balance_rmse": 738.7749375176188} {"canonical_length": 286, "canonical_nll": 1.3585162396197552, "generation_rouge_l_f1": 0.21333333333333335, "generation_token_f1": 0.44571428571428573, "prompt_id": "0e8185255effe518", "reading_canonical_log_prob": -388.53564453125, "reading_distances": [30, 29, 26, 25], "reading_log_probs": [-405.40692138671875, -398.456298828125, -394.038818359375, -387.9271240234375], "reading_logprob_std": 7.364434237542696, "tokenization_path_logprob_std": 222.1829144646255, "trajectory_balance_rmse": 4993.513017115068} {"canonical_length": 51, "canonical_nll": 1.1264509313246782, "generation_rouge_l_f1": 0.43478260869565216, "generation_token_f1": 0.5217391304347826, "prompt_id": "d03a9ea989bf62b9", "reading_canonical_log_prob": -57.448997497558594, "reading_distances": [30, 36, 34, 33], "reading_log_probs": [-61.92875671386719, -58.604820251464844, -61.556034088134766, -59.53139114379883], "reading_logprob_std": 1.5969296900273913, "tokenization_path_logprob_std": 79.92640347436118, "trajectory_balance_rmse": 828.6417837050199} {"canonical_length": 69, "canonical_nll": 0.967655126599298, "generation_rouge_l_f1": 0.030303030303030304, "generation_token_f1": 0.030303030303030304, "prompt_id": "fd11f4c3417e5ef0", "reading_canonical_log_prob": -66.76820373535156, "reading_distances": [44, 42, 39, 41], "reading_log_probs": [-58.24106216430664, -57.877445220947266, -57.15531921386719, -64.80110931396484], "reading_logprob_std": 3.550375218896581, "tokenization_path_logprob_std": 104.8284556435513, "trajectory_balance_rmse": 1387.8803465799392} {"canonical_length": 40, "canonical_nll": 1.5662714958190918, "generation_rouge_l_f1": 0.24719101123595502, "generation_token_f1": 0.29213483146067415, "prompt_id": "43fe74e6e2773223", "reading_canonical_log_prob": -62.65085983276367, "reading_distances": [16, 14, 13, 20], "reading_log_probs": [-59.99805450439453, -62.40883255004883, -64.3152084350586, -64.43606567382812], "reading_logprob_std": 2.0797468117484073, "tokenization_path_logprob_std": 94.02269726120146, "trajectory_balance_rmse": 632.9955221094597} {"canonical_length": 82, "canonical_nll": 0.9477648851348133, "generation_rouge_l_f1": 0.17567567567567566, "generation_token_f1": 0.21621621621621626, "prompt_id": "7eeeafb10f4366ae", "reading_canonical_log_prob": -77.71672058105469, "reading_distances": [30, 28, 34, 28], "reading_log_probs": [-80.33099365234375, -80.30680847167969, -80.7094497680664, -81.94387817382812], "reading_logprob_std": 0.7698019846219808, "tokenization_path_logprob_std": 29.501943127233634, "trajectory_balance_rmse": 1079.847964681983} {"canonical_length": 3, "canonical_nll": 2.784764607747396, "generation_rouge_l_f1": 0.05714285714285715, "generation_token_f1": 0.05714285714285715, "prompt_id": "6b50f3d234f6e0b4", "reading_canonical_log_prob": -8.354293823242188, "reading_distances": [20, 25, 26, 21], "reading_log_probs": [-11.225442886352539, -10.020376205444336, -11.172416687011719, -11.18372631072998], "reading_logprob_std": 0.5871858310450004, "tokenization_path_logprob_std": 18.836305810208255, "trajectory_balance_rmse": 28.494927078739206} {"canonical_length": 348, "canonical_nll": 1.1961505933739673, "generation_rouge_l_f1": 0.25212947189097107, "generation_token_f1": 0.4735945485519592, "prompt_id": "b62b5836a5079a74", "reading_canonical_log_prob": -416.2604064941406, "reading_distances": [31, 37, 29, 30], "reading_log_probs": [-431.46209716796875, -428.09503173828125, -423.988037109375, -431.78802490234375], "reading_logprob_std": 3.636050343635666, "tokenization_path_logprob_std": 351.6130038316864, "trajectory_balance_rmse": 7568.543784734926} {"canonical_length": 39, "canonical_nll": 1.2524918776292067, "generation_rouge_l_f1": 0.36923076923076925, "generation_token_f1": 0.36923076923076925, "prompt_id": "495dc4d7314e6352", "reading_canonical_log_prob": -48.84718322753906, "reading_distances": [27, 21, 27, 27], "reading_log_probs": [-46.43663787841797, -49.082523345947266, -48.11585998535156, -48.515899658203125], "reading_logprob_std": 1.1386892514589853, "tokenization_path_logprob_std": 98.22430569148193, "trajectory_balance_rmse": 923.797993823417} {"canonical_length": 54, "canonical_nll": 1.190069693106192, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.46956521739130436, "prompt_id": "0c8a53a7cd7794b9", "reading_canonical_log_prob": -64.26376342773438, "reading_distances": [24, 25, 26, 30], "reading_log_probs": [-69.4559326171875, -64.95152282714844, -63.096458435058594, -68.49269104003906], "reading_logprob_std": 2.9827408578849117, "tokenization_path_logprob_std": 94.9600169995948, "trajectory_balance_rmse": 1085.474914910377} {"canonical_length": 185, "canonical_nll": 1.7841095795502533, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.49074074074074076, "prompt_id": "bfd0b674c803f580", "reading_canonical_log_prob": -330.0602722167969, "reading_distances": [31, 33, 30, 37], "reading_log_probs": [-335.6417236328125, -341.9890441894531, -341.6355895996094, -339.19921875], "reading_logprob_std": 2.9256714048804127, "tokenization_path_logprob_std": 117.18766861526227, "trajectory_balance_rmse": 4591.802141379868} {"canonical_length": 215, "canonical_nll": 1.5270632721656976, "generation_rouge_l_f1": 0.20316027088036118, "generation_token_f1": 0.4108352144469526, "prompt_id": "f65fb83f66989dab", "reading_canonical_log_prob": -328.318603515625, "reading_distances": [23, 31, 22, 26], "reading_log_probs": [-332.7908935546875, -331.11767578125, -334.8688049316406, -332.02215576171875], "reading_logprob_std": 1.5994998342386282, "tokenization_path_logprob_std": 457.4079963472611, "trajectory_balance_rmse": 3929.818978047104} {"canonical_length": 20, "canonical_nll": 0.5984992980957031, "generation_rouge_l_f1": 0.7999999999999999, "generation_token_f1": 0.7999999999999999, "prompt_id": "ed802f2ffb8ef795", "reading_canonical_log_prob": -11.969985961914062, "reading_distances": [22, 18, 19, 22], "reading_log_probs": [-14.083541870117188, -14.897165298461914, -14.057515144348145, -15.397360801696777], "reading_logprob_std": 0.6544192697128668, "tokenization_path_logprob_std": 64.06671317039807, "trajectory_balance_rmse": 499.3202293109331} {"canonical_length": 17, "canonical_nll": 0.7780826792997473, "generation_rouge_l_f1": 0.32835820895522383, "generation_token_f1": 0.3880597014925373, "prompt_id": "578c6b312be28f5a", "reading_canonical_log_prob": -13.227405548095703, "reading_distances": [29, 21, 29, 32], "reading_log_probs": [-14.011760711669922, -13.357919692993164, -14.417922019958496, -14.211668014526367], "reading_logprob_std": 0.45893625739672556, "tokenization_path_logprob_std": 30.959550212883745, "trajectory_balance_rmse": 371.8065639230724} {"canonical_length": 20, "canonical_nll": 1.369499397277832, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.22222222222222224, "prompt_id": "016449fd389e919a", "reading_canonical_log_prob": -27.38998794555664, "reading_distances": [13, 19, 14, 16], "reading_log_probs": [-30.418800354003906, -27.834505081176758, -30.199432373046875, -31.56871795654297], "reading_logprob_std": 1.5668754674087388, "tokenization_path_logprob_std": 47.202519083269614, "trajectory_balance_rmse": 466.3473634168835} {"canonical_length": 172, "canonical_nll": 1.3650463902673056, "generation_rouge_l_f1": 0.24824355971896955, "generation_token_f1": 0.468384074941452, "prompt_id": "977004fae436eabd", "reading_canonical_log_prob": -234.78797912597656, "reading_distances": [26, 27, 25, 25], "reading_log_probs": [-240.20675659179688, -240.96693420410156, -240.64048767089844, -239.43887329101562], "reading_logprob_std": 0.6608739180010746, "tokenization_path_logprob_std": 171.96950249192759, "trajectory_balance_rmse": 3609.1449008287896} {"canonical_length": 223, "canonical_nll": 1.571718498195768, "generation_rouge_l_f1": 0.18989898989898993, "generation_token_f1": 0.4202020202020202, "prompt_id": "3d21a3aabdc36c10", "reading_canonical_log_prob": -350.49322509765625, "reading_distances": [32, 23, 34, 31], "reading_log_probs": [-364.05706787109375, -363.6971435546875, -362.00732421875, -357.43621826171875], "reading_logprob_std": 3.042974541005283, "tokenization_path_logprob_std": 319.4911913367641, "trajectory_balance_rmse": 4242.359200508377} {"canonical_length": 248, "canonical_nll": 0.409264410695722, "generation_rouge_l_f1": 0.5084745762711864, "generation_token_f1": 0.673728813559322, "prompt_id": "99f1a0976d255260", "reading_canonical_log_prob": -101.49757385253906, "reading_distances": [34, 32, 34, 32], "reading_log_probs": [-112.80690002441406, -109.37739562988281, -112.7172622680664, -109.10336303710938], "reading_logprob_std": 2.0366596878645806, "tokenization_path_logprob_std": 160.01208436580524, "trajectory_balance_rmse": 3677.1613869450484} {"canonical_length": 183, "canonical_nll": 1.0624152011558659, "generation_rouge_l_f1": 0.37264150943396224, "generation_token_f1": 0.6037735849056602, "prompt_id": "24640c4b7c2fb283", "reading_canonical_log_prob": -194.42198181152344, "reading_distances": [72, 61, 66, 65], "reading_log_probs": [-182.50875854492188, -186.1229248046875, -184.7954559326172, -186.12368774414062], "reading_logprob_std": 1.7050236983939595, "tokenization_path_logprob_std": 162.18854781510913, "trajectory_balance_rmse": 2975.3565522706403} {"canonical_length": 6, "canonical_nll": 0.35425861676534015, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "94e6b38d533edcd5", "reading_canonical_log_prob": -2.125551700592041, "reading_distances": [35, 35, 38, 30], "reading_log_probs": [-5.0027031898498535, -4.097500324249268, -3.8876187801361084, -4.5985870361328125], "reading_logprob_std": 0.5022225063649411, "tokenization_path_logprob_std": 34.54689729690296, "trajectory_balance_rmse": 105.94263537775616} {"canonical_length": 252, "canonical_nll": 1.483216300843254, "generation_rouge_l_f1": 0.22050290135396522, "generation_token_f1": 0.41779497098646035, "prompt_id": "22a8bdc8dd7cf586", "reading_canonical_log_prob": -373.7705078125, "reading_distances": [27, 22, 23, 22], "reading_log_probs": [-382.16595458984375, -379.4609375, -380.2267761230469, -378.02581787109375], "reading_logprob_std": 1.725025531333772, "tokenization_path_logprob_std": 95.23197312323863, "trajectory_balance_rmse": 5656.369313179925} {"canonical_length": 27, "canonical_nll": 1.3596750895182292, "generation_rouge_l_f1": 0.11666666666666667, "generation_token_f1": 0.15000000000000002, "prompt_id": "6585d8646962bd31", "reading_canonical_log_prob": -36.71122741699219, "reading_distances": [16, 14, 18, 12], "reading_log_probs": [-39.3947639465332, -35.05282211303711, -37.2265625, -35.01927185058594], "reading_logprob_std": 2.087600144921675, "tokenization_path_logprob_std": 63.16277289262182, "trajectory_balance_rmse": 280.9567172045488} {"canonical_length": 81, "canonical_nll": 1.4151329699857735, "generation_rouge_l_f1": 0.2959183673469388, "generation_token_f1": 0.510204081632653, "prompt_id": "04a5aa53ab7903f1", "reading_canonical_log_prob": -114.62577056884766, "reading_distances": [22, 17, 27, 27], "reading_log_probs": [-117.29414367675781, -118.31484985351562, -121.85684204101562, -119.62787628173828], "reading_logprob_std": 1.96944156102634, "tokenization_path_logprob_std": 131.02266015780236, "trajectory_balance_rmse": 1968.5188840109724} {"canonical_length": 198, "canonical_nll": 1.0849241006253945, "generation_rouge_l_f1": 0.24516129032258063, "generation_token_f1": 0.46021505376344085, "prompt_id": "f2143d23c11c6f8b", "reading_canonical_log_prob": -214.81497192382812, "reading_distances": [20, 24, 23, 20], "reading_log_probs": [-225.4429931640625, -228.00521850585938, -233.2384490966797, -231.20541381835938], "reading_logprob_std": 3.443602868536688, "tokenization_path_logprob_std": 251.6011375088442, "trajectory_balance_rmse": 3889.8550745052116} {"canonical_length": 12, "canonical_nll": 0.8008391062418619, "generation_rouge_l_f1": 0.7199999999999999, "generation_token_f1": 0.7199999999999999, "prompt_id": "5633164a0de34e8b", "reading_canonical_log_prob": -9.610069274902344, "reading_distances": [36, 42, 40, 33], "reading_log_probs": [-10.955211639404297, -9.714254379272461, -10.086682319641113, -8.803153038024902], "reading_logprob_std": 0.8917179097155581, "tokenization_path_logprob_std": 41.48357705809787, "trajectory_balance_rmse": 237.99937027010523} {"canonical_length": 134, "canonical_nll": 1.1812177060255364, "generation_rouge_l_f1": 0.18565400843881855, "generation_token_f1": 0.2869198312236287, "prompt_id": "fe842e52ca168c74", "reading_canonical_log_prob": -158.28317260742188, "reading_distances": [26, 25, 19, 27], "reading_log_probs": [-161.0008544921875, -165.14222717285156, -160.83114624023438, -160.27743530273438], "reading_logprob_std": 2.240932068806072, "tokenization_path_logprob_std": 227.0618887318721, "trajectory_balance_rmse": 2426.804846969693} {"canonical_length": 46, "canonical_nll": 0.2839079525159753, "generation_rouge_l_f1": 0.4457831325301204, "generation_token_f1": 0.4457831325301204, "prompt_id": "6327942806f18cd4", "reading_canonical_log_prob": -13.059765815734863, "reading_distances": [7, 13, 10, 12], "reading_log_probs": [-13.363529205322266, -15.203275680541992, -12.618634223937988, -13.854217529296875], "reading_logprob_std": 1.0880910699009547, "tokenization_path_logprob_std": 32.622554284707974, "trajectory_balance_rmse": 501.55804967340936} {"canonical_length": 54, "canonical_nll": 0.9071335969147859, "generation_rouge_l_f1": 0.3580246913580247, "generation_token_f1": 0.4197530864197531, "prompt_id": "95494e563c3fc530", "reading_canonical_log_prob": -48.98521423339844, "reading_distances": [29, 29, 30, 30], "reading_log_probs": [-53.49695587158203, -54.30955123901367, -50.97801208496094, -53.60148239135742], "reading_logprob_std": 1.457721564097446, "tokenization_path_logprob_std": 74.25160823350288, "trajectory_balance_rmse": 1045.8853763145778} {"canonical_length": 44, "canonical_nll": 2.0492640408602627, "generation_rouge_l_f1": 0.1386138613861386, "generation_token_f1": 0.198019801980198, "prompt_id": "15a442e093e950e5", "reading_canonical_log_prob": -90.16761779785156, "reading_distances": [23, 29, 30, 26], "reading_log_probs": [-90.18582153320312, -86.63855743408203, -89.03138732910156, -90.2447738647461], "reading_logprob_std": 1.686268440326189, "tokenization_path_logprob_std": 57.88166390209246, "trajectory_balance_rmse": 788.3224483565851} {"canonical_length": 303, "canonical_nll": 0.814246955091017, "generation_rouge_l_f1": 0.2924901185770751, "generation_token_f1": 0.4308300395256917, "prompt_id": "452a884e801c54f0", "reading_canonical_log_prob": -246.71682739257812, "reading_distances": [15, 24, 29, 28], "reading_log_probs": [-249.18527221679688, -247.6695098876953, -247.45904541015625, -248.0507354736328], "reading_logprob_std": 0.7694312522232185, "tokenization_path_logprob_std": 199.95145747197972, "trajectory_balance_rmse": 5176.374286625991} {"canonical_length": 340, "canonical_nll": 1.2187986486098346, "generation_rouge_l_f1": 0.2140468227424749, "generation_token_f1": 0.42140468227424754, "prompt_id": "6f1aa7f491e54018", "reading_canonical_log_prob": -414.39154052734375, "reading_distances": [25, 23, 16, 18], "reading_log_probs": [-411.2794189453125, -411.6778564453125, -415.00128173828125, -418.05352783203125], "reading_logprob_std": 3.174246811271325, "tokenization_path_logprob_std": 349.0571380620911, "trajectory_balance_rmse": 6829.665356708397} {"canonical_length": 197, "canonical_nll": 0.860818156131028, "generation_rouge_l_f1": 0.26126126126126126, "generation_token_f1": 0.5405405405405406, "prompt_id": "b6dbd4688fd838f3", "reading_canonical_log_prob": -169.5811767578125, "reading_distances": [20, 19, 18, 22], "reading_log_probs": [-178.47116088867188, -179.92236328125, -179.02464294433594, -180.33563232421875], "reading_logprob_std": 0.8457723671043597, "tokenization_path_logprob_std": 369.09809734830566, "trajectory_balance_rmse": 4547.969791327801} {"canonical_length": 138, "canonical_nll": 1.7843521781589673, "generation_rouge_l_f1": 0.20833333333333334, "generation_token_f1": 0.40476190476190477, "prompt_id": "4e91817c9d8f7b21", "reading_canonical_log_prob": -246.2406005859375, "reading_distances": [17, 20, 19, 20], "reading_log_probs": [-250.8447265625, -251.89370727539062, -249.91302490234375, -249.57040405273438], "reading_logprob_std": 1.0420554988555217, "tokenization_path_logprob_std": 60.24076677124703, "trajectory_balance_rmse": 1907.8997709926523} {"canonical_length": 10, "canonical_nll": 1.511479949951172, "generation_rouge_l_f1": 0.4615384615384615, "generation_token_f1": 0.4615384615384615, "prompt_id": "8f6ce2573cd1b9d1", "reading_canonical_log_prob": -15.114799499511719, "reading_distances": [37, 33, 38, 35], "reading_log_probs": [-18.272062301635742, -18.30222511291504, -19.777767181396484, -18.524192810058594], "reading_logprob_std": 0.7147009591608956, "tokenization_path_logprob_std": 22.293526880964382, "trajectory_balance_rmse": 199.31221992789304} {"canonical_length": 9, "canonical_nll": 0.3820211622450087, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.33333333333333337, "prompt_id": "c246d0eff1fb5aac", "reading_canonical_log_prob": -3.438190460205078, "reading_distances": [12, 18, 17, 17], "reading_log_probs": [-6.174264430999756, -5.134289264678955, -7.029783248901367, -6.652885913848877], "reading_logprob_std": 0.820752913816311, "tokenization_path_logprob_std": 21.894165909900376, "trajectory_balance_rmse": 157.30839169780901} {"canonical_length": 222, "canonical_nll": 0.6078346871040963, "generation_rouge_l_f1": 0.3983050847457627, "generation_token_f1": 0.5932203389830507, "prompt_id": "877651642570fc39", "reading_canonical_log_prob": -134.93930053710938, "reading_distances": [31, 33, 32, 30], "reading_log_probs": [-150.58396911621094, -142.56666564941406, -146.6351318359375, -139.35992431640625], "reading_logprob_std": 4.878643116967074, "tokenization_path_logprob_std": 152.6531341347162, "trajectory_balance_rmse": 3079.458866136251} {"canonical_length": 53, "canonical_nll": 0.8214186902316112, "generation_rouge_l_f1": 0.35658914728682173, "generation_token_f1": 0.4031007751937985, "prompt_id": "77a973a8d7f91484", "reading_canonical_log_prob": -43.53519058227539, "reading_distances": [31, 29, 31, 33], "reading_log_probs": [-48.45074462890625, -48.70758819580078, -48.67692565917969, -49.506591796875], "reading_logprob_std": 0.4618470674132073, "tokenization_path_logprob_std": 99.88976235304756, "trajectory_balance_rmse": 1083.5641858526915} {"canonical_length": 21, "canonical_nll": 0.9256453741164434, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.5263157894736842, "prompt_id": "f9bfdd6eb133d424", "reading_canonical_log_prob": -19.438552856445312, "reading_distances": [24, 24, 22, 26], "reading_log_probs": [-21.23746681213379, -24.638822555541992, -22.43543243408203, -25.495330810546875], "reading_logprob_std": 1.9597036520457867, "tokenization_path_logprob_std": 48.598394831502524, "trajectory_balance_rmse": 387.3794804419376} {"canonical_length": 247, "canonical_nll": 1.568133025999494, "generation_rouge_l_f1": 0.24951267056530216, "generation_token_f1": 0.42495126705653025, "prompt_id": "3a822df2306dda39", "reading_canonical_log_prob": -387.328857421875, "reading_distances": [39, 36, 48, 42], "reading_log_probs": [-390.57513427734375, -398.9679870605469, -399.66595458984375, -398.09490966796875], "reading_logprob_std": 4.216511380593217, "tokenization_path_logprob_std": 154.0682256991962, "trajectory_balance_rmse": 5630.614838289024} {"canonical_length": 88, "canonical_nll": 1.2990074157714844, "generation_rouge_l_f1": 0.21978021978021978, "generation_token_f1": 0.4505494505494506, "prompt_id": "c28f9f6ed8615ffa", "reading_canonical_log_prob": -114.31265258789062, "reading_distances": [26, 27, 29, 29], "reading_log_probs": [-115.78028869628906, -109.56770324707031, -112.03556060791016, -110.71595001220703], "reading_logprob_std": 2.6990259884851437, "tokenization_path_logprob_std": 108.85695120686697, "trajectory_balance_rmse": 1760.182657925609}