{"canonical_length": 14, "canonical_nll": 2.697730473109654, "generation_rouge_l_f1": 0.23999999999999996, "generation_token_f1": 0.23999999999999996, "prompt_id": "54ef1334aadf1dfd", "reading_canonical_log_prob": -37.768226623535156, "reading_distances": [37, 36, 33, 38], "reading_log_probs": [-41.180145263671875, -41.319618225097656, -38.89855194091797, -40.7351188659668], "reading_logprob_std": 1.1180097202714658, "tokenization_path_logprob_std": 55.204573871416635, "trajectory_balance_rmse": 349.9694134935713} {"canonical_length": 23, "canonical_nll": 0.5685220801311991, "generation_rouge_l_f1": 0.08695652173913045, "generation_token_f1": 0.10276679841897232, "prompt_id": "d109cb9b45bb2b84", "reading_canonical_log_prob": -13.076007843017578, "reading_distances": [12, 12, 9, 10], "reading_log_probs": [-14.315933227539062, -19.70895004272461, -15.525222778320312, -17.747304916381836], "reading_logprob_std": 2.391139855564196, "tokenization_path_logprob_std": 55.75414425151582, "trajectory_balance_rmse": 288.6060332704559} {"canonical_length": 79, "canonical_nll": 2.1194620253164556, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.48484848484848475, "prompt_id": "f461f81672c6f974", "reading_canonical_log_prob": -167.4375, "reading_distances": [24, 27, 34, 27], "reading_log_probs": [-158.49493408203125, -168.1779022216797, -159.17575073242188, -155.59426879882812], "reading_logprob_std": 5.437913331317045, "tokenization_path_logprob_std": 166.2043884468614, "trajectory_balance_rmse": 2683.724493916132} {"canonical_length": 22, "canonical_nll": 0.8916719610040839, "generation_rouge_l_f1": 0.08727272727272727, "generation_token_f1": 0.1090909090909091, "prompt_id": "db2b92b024f2b9fd", "reading_canonical_log_prob": -19.616783142089844, "reading_distances": [20, 13, 17, 14], "reading_log_probs": [-19.203853607177734, -23.691020965576172, -19.938488006591797, -19.85326385498047], "reading_logprob_std": 2.039469675292245, "tokenization_path_logprob_std": 73.4863969457754, "trajectory_balance_rmse": 249.41571685355177} {"canonical_length": 22, "canonical_nll": 2.447098818692294, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "a3ca3e528a4c44fb", "reading_canonical_log_prob": -53.83617401123047, "reading_distances": [27, 30, 28, 27], "reading_log_probs": [-64.613037109375, -61.88005065917969, -62.30294418334961, -60.78023910522461], "reading_logprob_std": 1.612532685621921, "tokenization_path_logprob_std": 155.26571184772405, "trajectory_balance_rmse": 954.2089396115423} {"canonical_length": 46, "canonical_nll": 1.1224575872006624, "generation_rouge_l_f1": 0.37777777777777777, "generation_token_f1": 0.5555555555555556, "prompt_id": "bea2699cff758312", "reading_canonical_log_prob": -51.63304901123047, "reading_distances": [21, 15, 20, 25], "reading_log_probs": [-52.070098876953125, -52.1037712097168, -50.596954345703125, -50.04356002807617], "reading_logprob_std": 1.0448030431365167, "tokenization_path_logprob_std": 145.48534363503495, "trajectory_balance_rmse": 1821.9292568520095} {"canonical_length": 41, "canonical_nll": 1.8968457477848704, "generation_rouge_l_f1": 0.20689655172413793, "generation_token_f1": 0.21551724137931033, "prompt_id": "cf9306b8e22601b5", "reading_canonical_log_prob": -77.77067565917969, "reading_distances": [16, 21, 21, 18], "reading_log_probs": [-80.33616638183594, -79.51863861083984, -79.12164306640625, -79.35260009765625], "reading_logprob_std": 0.528309929231204, "tokenization_path_logprob_std": 79.01335868876073, "trajectory_balance_rmse": 534.0360608794279} {"canonical_length": 231, "canonical_nll": 1.173817424031047, "generation_rouge_l_f1": 0.25609756097560976, "generation_token_f1": 0.47560975609756095, "prompt_id": "ff6c469935e35f5e", "reading_canonical_log_prob": -271.1518249511719, "reading_distances": [17, 21, 23, 17], "reading_log_probs": [-267.14697265625, -269.9151611328125, -267.96917724609375, -265.13372802734375], "reading_logprob_std": 1.9807567970076685, "tokenization_path_logprob_std": 342.69191923193205, "trajectory_balance_rmse": 7179.4773074557825} {"canonical_length": 203, "canonical_nll": 1.5005865989647476, "generation_rouge_l_f1": 0.26966292134831454, "generation_token_f1": 0.5662921348314607, "prompt_id": "41c1bf9aacb45501", "reading_canonical_log_prob": -304.61907958984375, "reading_distances": [23, 22, 21, 22], "reading_log_probs": [-310.15545654296875, -309.81121826171875, -308.8662109375, -309.02838134765625], "reading_logprob_std": 0.6180024568561853, "tokenization_path_logprob_std": 207.1432772668014, "trajectory_balance_rmse": 6565.6165120552905} {"canonical_length": 33, "canonical_nll": 1.343424941554214, "generation_rouge_l_f1": 0.1714285714285714, "generation_token_f1": 0.1714285714285714, "prompt_id": "aff22e4e1c81f916", "reading_canonical_log_prob": -44.33302307128906, "reading_distances": [25, 29, 34, 25], "reading_log_probs": [-45.57191467285156, -44.665306091308594, -44.40422821044922, -44.829139709472656], "reading_logprob_std": 0.5010570773495341, "tokenization_path_logprob_std": 217.64362371128038, "trajectory_balance_rmse": 1149.9129928307693} {"canonical_length": 10, "canonical_nll": 1.5348469734191894, "generation_rouge_l_f1": 0.17204301075268816, "generation_token_f1": 0.17204301075268816, "prompt_id": "f599bb889055c49d", "reading_canonical_log_prob": -15.348469734191895, "reading_distances": [63, 53, 60, 54], "reading_log_probs": [-16.76611328125, -16.618900299072266, -17.843299865722656, -16.228679656982422], "reading_logprob_std": 0.6909692689248166, "tokenization_path_logprob_std": 66.21813616914287, "trajectory_balance_rmse": 269.71666360578485} {"canonical_length": 90, "canonical_nll": 1.4982506646050346, "generation_rouge_l_f1": 0.17983651226158037, "generation_token_f1": 0.321525885558583, "prompt_id": "7fbc599a6c4c6b54", "reading_canonical_log_prob": -134.84255981445312, "reading_distances": [21, 34, 28, 33], "reading_log_probs": [-138.9425811767578, -143.42262268066406, -140.5062255859375, -144.35861206054688], "reading_logprob_std": 2.5177946026296936, "tokenization_path_logprob_std": 219.61576626286012, "trajectory_balance_rmse": 2845.535478042987} {"canonical_length": 294, "canonical_nll": 1.7462488291214924, "generation_rouge_l_f1": 0.2186379928315412, "generation_token_f1": 0.38351254480286745, "prompt_id": "9e624d36638d2bbd", "reading_canonical_log_prob": -513.3971557617188, "reading_distances": [20, 13, 22, 16], "reading_log_probs": [-502.44647216796875, -505.11712646484375, -505.78173828125, -509.296875], "reading_logprob_std": 2.8203517247569354, "tokenization_path_logprob_std": 304.16064529943026, "trajectory_balance_rmse": 8406.533565946742} {"canonical_length": 27, "canonical_nll": 1.1111765967475042, "generation_rouge_l_f1": 0.3225806451612903, "generation_token_f1": 0.3225806451612903, "prompt_id": "3982276a331d29f5", "reading_canonical_log_prob": -30.001768112182617, "reading_distances": [26, 23, 24, 25], "reading_log_probs": [-28.38424301147461, -30.569021224975586, -32.08875274658203, -29.588808059692383], "reading_logprob_std": 1.5670500513105456, "tokenization_path_logprob_std": 59.145110395236834, "trajectory_balance_rmse": 445.66230185831677} {"canonical_length": 10, "canonical_nll": 1.143953514099121, "generation_rouge_l_f1": 0.4615384615384615, "generation_token_f1": 0.6153846153846154, "prompt_id": "3b79dbc85baf9416", "reading_canonical_log_prob": -11.439535140991211, "reading_distances": [21, 25, 21, 19], "reading_log_probs": [-12.928622245788574, -15.06148624420166, -12.461203575134277, -14.132028579711914], "reading_logprob_std": 1.1773130976572066, "tokenization_path_logprob_std": 44.413780742476625, "trajectory_balance_rmse": 257.99883260970176} {"canonical_length": 174, "canonical_nll": 1.4945854099317528, "generation_rouge_l_f1": 0.25345622119815675, "generation_token_f1": 0.48387096774193544, "prompt_id": "bc017f1a688cb1fb", "reading_canonical_log_prob": -260.057861328125, "reading_distances": [23, 20, 25, 28], "reading_log_probs": [-263.003173828125, -262.032470703125, -261.7153625488281, -259.94927978515625], "reading_logprob_std": 1.2743063990532582, "tokenization_path_logprob_std": 335.0481893438071, "trajectory_balance_rmse": 5936.038031066898} {"canonical_length": 64, "canonical_nll": 1.1147500276565552, "generation_rouge_l_f1": 0.18085106382978727, "generation_token_f1": 0.26595744680851063, "prompt_id": "2060bf62bca6703a", "reading_canonical_log_prob": -71.34400177001953, "reading_distances": [17, 14, 23, 18], "reading_log_probs": [-68.81966400146484, -68.93302917480469, -68.30650329589844, -69.82566833496094], "reading_logprob_std": 0.6314950550327754, "tokenization_path_logprob_std": 58.869269543722154, "trajectory_balance_rmse": 1592.9538106501157} {"canonical_length": 12, "canonical_nll": 0.9247584342956543, "generation_rouge_l_f1": 0.047244094488188976, "generation_token_f1": 0.047244094488188976, "prompt_id": "8300e5ea4c06c29e", "reading_canonical_log_prob": -11.097101211547852, "reading_distances": [27, 29, 30, 27], "reading_log_probs": [-10.25023078918457, -11.573229789733887, -12.420486450195312, -11.795438766479492], "reading_logprob_std": 0.9131405955848482, "tokenization_path_logprob_std": 19.36617219902223, "trajectory_balance_rmse": 174.4927222623082} {"canonical_length": 142, "canonical_nll": 1.5908942424075705, "generation_rouge_l_f1": 0.1979695431472081, "generation_token_f1": 0.3959390862944162, "prompt_id": "5c25240178b5a183", "reading_canonical_log_prob": -225.906982421875, "reading_distances": [20, 19, 20, 21], "reading_log_probs": [-239.26541137695312, -245.10223388671875, -239.74864196777344, -248.1778106689453], "reading_logprob_std": 4.309909839786842, "tokenization_path_logprob_std": 173.80713122270822, "trajectory_balance_rmse": 3934.8636853089474} {"canonical_length": 14, "canonical_nll": 1.6125375202723913, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "571e075c79645e81", "reading_canonical_log_prob": -22.575525283813477, "reading_distances": [28, 15, 24, 19], "reading_log_probs": [-22.44387435913086, -21.638944625854492, -21.938232421875, -22.922199249267578], "reading_logprob_std": 0.5654560604440909, "tokenization_path_logprob_std": 15.006182534617407, "trajectory_balance_rmse": 443.9152496371094} {"canonical_length": 153, "canonical_nll": 1.6021624795751634, "generation_rouge_l_f1": 0.19298245614035087, "generation_token_f1": 0.4678362573099415, "prompt_id": "fc69046f95700385", "reading_canonical_log_prob": -245.130859375, "reading_distances": [39, 41, 39, 38], "reading_log_probs": [-252.72555541992188, -246.6273956298828, -256.92596435546875, -250.55532836914062], "reading_logprob_std": 4.297431885951026, "tokenization_path_logprob_std": 295.73186912230085, "trajectory_balance_rmse": 4901.149747444821} {"canonical_length": 56, "canonical_nll": 2.03547055380685, "generation_rouge_l_f1": 0.15, "generation_token_f1": 0.2, "prompt_id": "fcbb32fd544b170a", "reading_canonical_log_prob": -113.9863510131836, "reading_distances": [24, 22, 24, 25], "reading_log_probs": [-117.33390808105469, -116.94377136230469, -117.33523559570312, -116.16514587402344], "reading_logprob_std": 0.5512731757795318, "tokenization_path_logprob_std": 92.95812990701107, "trajectory_balance_rmse": 1529.8146674157406} {"canonical_length": 83, "canonical_nll": 0.9626684533544334, "generation_rouge_l_f1": 0.3274336283185841, "generation_token_f1": 0.4424778761061947, "prompt_id": "f6d789446e38536d", "reading_canonical_log_prob": -79.90148162841797, "reading_distances": [8, 7, 8, 9], "reading_log_probs": [-79.42628479003906, -77.96641540527344, -79.88370513916016, -77.09201049804688], "reading_logprob_std": 1.2917504693661352, "tokenization_path_logprob_std": 164.17099120030255, "trajectory_balance_rmse": 2275.3852019366227} {"canonical_length": 270, "canonical_nll": 1.5119194878472222, "generation_rouge_l_f1": 0.15500000000000003, "generation_token_f1": 0.30500000000000005, "prompt_id": "86891b80d6ccb140", "reading_canonical_log_prob": -408.21826171875, "reading_distances": [20, 24, 27, 23], "reading_log_probs": [-408.2852478027344, -411.21337890625, -405.8559265136719, -413.60406494140625], "reading_logprob_std": 3.3815269528259946, "tokenization_path_logprob_std": 171.60521205949388, "trajectory_balance_rmse": 4956.612596565607} {"canonical_length": 27, "canonical_nll": 1.794401522035952, "generation_rouge_l_f1": 0.23684210526315785, "generation_token_f1": 0.42105263157894735, "prompt_id": "01a2b8a844748c41", "reading_canonical_log_prob": -48.4488410949707, "reading_distances": [33, 37, 33, 36], "reading_log_probs": [-48.973480224609375, -49.213077545166016, -47.125572204589844, -47.267417907714844], "reading_logprob_std": 1.1009923452214088, "tokenization_path_logprob_std": 61.37493984574047, "trajectory_balance_rmse": 953.608721765436} {"canonical_length": 66, "canonical_nll": 1.250297430789832, "generation_rouge_l_f1": 0.17337461300309598, "generation_token_f1": 0.22910216718266255, "prompt_id": "fc8d39645d6796a9", "reading_canonical_log_prob": -82.5196304321289, "reading_distances": [28, 32, 30, 26], "reading_log_probs": [-87.64574432373047, -87.7868881225586, -87.11932373046875, -86.14271545410156], "reading_logprob_std": 0.744918659018771, "tokenization_path_logprob_std": 139.58176302995062, "trajectory_balance_rmse": 2160.8158521294613} {"canonical_length": 340, "canonical_nll": 1.3153020522173713, "generation_rouge_l_f1": 0.21717990275526744, "generation_token_f1": 0.4246353322528363, "prompt_id": "7a3c4e3465a24fae", "reading_canonical_log_prob": -447.20269775390625, "reading_distances": [28, 28, 21, 20], "reading_log_probs": [-447.42462158203125, -446.1640625, -449.5672302246094, -450.9664001464844], "reading_logprob_std": 2.1472007027970417, "tokenization_path_logprob_std": 297.6731772400062, "trajectory_balance_rmse": 9922.389273592225} {"canonical_length": 22, "canonical_nll": 0.8835238543423739, "generation_rouge_l_f1": 0.5185185185185185, "generation_token_f1": 0.5185185185185185, "prompt_id": "9652191d0c5b1125", "reading_canonical_log_prob": -19.437524795532227, "reading_distances": [20, 17, 15, 23], "reading_log_probs": [-20.481592178344727, -21.595155715942383, -21.117694854736328, -20.61923599243164], "reading_logprob_std": 0.5076453490435812, "tokenization_path_logprob_std": 48.97558321183961, "trajectory_balance_rmse": 545.6501975629022} {"canonical_length": 229, "canonical_nll": 1.0252132499061817, "generation_rouge_l_f1": 0.2752688172043011, "generation_token_f1": 0.5462365591397849, "prompt_id": "1fb3dd895fb36ca1", "reading_canonical_log_prob": -234.77383422851562, "reading_distances": [46, 47, 47, 53], "reading_log_probs": [-235.50039672851562, -233.56210327148438, -232.13465881347656, -236.1489715576172], "reading_logprob_std": 1.8337124130551965, "tokenization_path_logprob_std": 270.70074830212366, "trajectory_balance_rmse": 6792.406746727771} {"canonical_length": 84, "canonical_nll": 0.15284280549912227, "generation_rouge_l_f1": 0.45751633986928103, "generation_token_f1": 0.47058823529411764, "prompt_id": "fba6cc8156910498", "reading_canonical_log_prob": -12.83879566192627, "reading_distances": [22, 14, 24, 27], "reading_log_probs": [-13.828207015991211, -12.609176635742188, -13.951604843139648, -14.007814407348633], "reading_logprob_std": 0.6642653542344367, "tokenization_path_logprob_std": 225.01644142908108, "trajectory_balance_rmse": 2020.3851071263105} {"canonical_length": 24, "canonical_nll": 0.5524690945943197, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "02b01e693f3f83a0", "reading_canonical_log_prob": -13.259258270263672, "reading_distances": [34, 32, 25, 24], "reading_log_probs": [-7.32963228225708, -6.3492326736450195, -7.146420478820801, -6.749795436859131], "reading_logprob_std": 0.43629415465867183, "tokenization_path_logprob_std": 73.77475948295165, "trajectory_balance_rmse": 500.6339634655336} {"canonical_length": 331, "canonical_nll": 0.9687049151187217, "generation_rouge_l_f1": 0.2836879432624113, "generation_token_f1": 0.6170212765957447, "prompt_id": "40bc9c44485ded59", "reading_canonical_log_prob": -320.6413269042969, "reading_distances": [27, 34, 28, 33], "reading_log_probs": [-327.65692138671875, -323.4745178222656, -325.6705017089844, -327.0595703125], "reading_logprob_std": 1.8574096925478363, "tokenization_path_logprob_std": 309.52316278056634, "trajectory_balance_rmse": 6582.701147990592} {"canonical_length": 11, "canonical_nll": 2.297130064530806, "generation_rouge_l_f1": 0.3, "generation_token_f1": 0.3, "prompt_id": "f0c9ccf7c1a3a5c6", "reading_canonical_log_prob": -25.268430709838867, "reading_distances": [22, 26, 27, 28], "reading_log_probs": [-21.97269058227539, -22.426952362060547, -22.035036087036133, -22.976318359375], "reading_logprob_std": 0.4617830590364818, "tokenization_path_logprob_std": 44.54856132818659, "trajectory_balance_rmse": 357.27794364727686} {"canonical_length": 149, "canonical_nll": 2.0415964830641777, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.4691358024691358, "prompt_id": "7b4b39937f27934d", "reading_canonical_log_prob": -304.1978759765625, "reading_distances": [27, 27, 23, 30], "reading_log_probs": [-300.35247802734375, -300.38922119140625, -301.4065856933594, -303.28155517578125], "reading_logprob_std": 1.3725925269411177, "tokenization_path_logprob_std": 162.0872196795787, "trajectory_balance_rmse": 4287.439069980983} {"canonical_length": 10, "canonical_nll": 0.8518041610717774, "generation_rouge_l_f1": 0.12121212121212123, "generation_token_f1": 0.12121212121212123, "prompt_id": "177155c261ac8893", "reading_canonical_log_prob": -8.518041610717773, "reading_distances": [18, 15, 11, 15], "reading_log_probs": [-6.856395244598389, -7.458509922027588, -7.958893775939941, -7.368221759796143], "reading_logprob_std": 0.4516120073503469, "tokenization_path_logprob_std": 56.35639314860203, "trajectory_balance_rmse": 265.16626203198643} {"canonical_length": 9, "canonical_nll": 0.18359246518876818, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "316f7da66f20e71c", "reading_canonical_log_prob": -1.6523321866989136, "reading_distances": [31, 31, 30, 26], "reading_log_probs": [-1.2522640228271484, -1.4224365949630737, -1.328964114189148, -1.3804283142089844], "reading_logprob_std": 0.07326780676926029, "tokenization_path_logprob_std": 40.00226985863474, "trajectory_balance_rmse": 298.4347188030155} {"canonical_length": 10, "canonical_nll": 1.9263570785522461, "generation_rouge_l_f1": 0.0958904109589041, "generation_token_f1": 0.1095890410958904, "prompt_id": "4c771fd22dee11bd", "reading_canonical_log_prob": -19.26357078552246, "reading_distances": [25, 31, 29, 30], "reading_log_probs": [-19.835529327392578, -20.654624938964844, -19.853519439697266, -20.599567413330078], "reading_logprob_std": 0.4524364037121794, "tokenization_path_logprob_std": 53.62167490443926, "trajectory_balance_rmse": 368.4520255208285} {"canonical_length": 150, "canonical_nll": 1.2521662394205728, "generation_rouge_l_f1": 0.22613065326633164, "generation_token_f1": 0.42211055276381915, "prompt_id": "fb7e151d45f5c282", "reading_canonical_log_prob": -187.82493591308594, "reading_distances": [37, 33, 38, 39], "reading_log_probs": [-192.64117431640625, -195.01840209960938, -194.0078582763672, -192.43875122070312], "reading_logprob_std": 1.21442395525584, "tokenization_path_logprob_std": 208.85252621000893, "trajectory_balance_rmse": 4184.1949580747} {"canonical_length": 186, "canonical_nll": 1.2759625014438425, "generation_rouge_l_f1": 0.30992736077481836, "generation_token_f1": 0.49878934624697346, "prompt_id": "c6ce7ec89de605a5", "reading_canonical_log_prob": -237.3290252685547, "reading_distances": [17, 19, 20, 19], "reading_log_probs": [-234.6563262939453, -232.34378051757812, -233.22720336914062, -232.94406127929688], "reading_logprob_std": 0.9807758048682084, "tokenization_path_logprob_std": 242.70848017010047, "trajectory_balance_rmse": 5631.175988308291} {"canonical_length": 220, "canonical_nll": 1.0287026145241478, "generation_rouge_l_f1": 0.29661016949152547, "generation_token_f1": 0.5550847457627119, "prompt_id": "320fa6e957218c02", "reading_canonical_log_prob": -226.3145751953125, "reading_distances": [14, 13, 14, 13], "reading_log_probs": [-230.02333068847656, -227.91725158691406, -227.264404296875, -228.59788513183594], "reading_logprob_std": 1.1813404037734994, "tokenization_path_logprob_std": 311.5624470488791, "trajectory_balance_rmse": 5610.845920638986} {"canonical_length": 58, "canonical_nll": 1.1829680081071525, "generation_rouge_l_f1": 0.08888888888888888, "generation_token_f1": 0.08888888888888888, "prompt_id": "d3de64d71666f0d5", "reading_canonical_log_prob": -68.61214447021484, "reading_distances": [22, 20, 23, 23], "reading_log_probs": [-70.93060302734375, -80.53036499023438, -74.70639038085938, -72.33984375], "reading_logprob_std": 4.232826288770994, "tokenization_path_logprob_std": 51.12760275360172, "trajectory_balance_rmse": 1189.393740407212} {"canonical_length": 11, "canonical_nll": 0.04382408207113093, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "bd4efb1cb33c7494", "reading_canonical_log_prob": -0.4820649027824402, "reading_distances": [44, 47, 47, 43], "reading_log_probs": [-1.3120410442352295, -0.9155309200286865, -0.897398829460144, -0.9121572971343994], "reading_logprob_std": 0.20199286754635415, "tokenization_path_logprob_std": 69.36759902323223, "trajectory_balance_rmse": 278.9587638753279} {"canonical_length": 38, "canonical_nll": 0.7031104439183286, "generation_rouge_l_f1": 0.34375, "generation_token_f1": 0.37500000000000006, "prompt_id": "d19ca6fde647262d", "reading_canonical_log_prob": -26.718196868896484, "reading_distances": [17, 15, 15, 17], "reading_log_probs": [-27.397388458251953, -24.246826171875, -26.262863159179688, -24.209888458251953], "reading_logprob_std": 1.5719897496509896, "tokenization_path_logprob_std": 76.70284893669135, "trajectory_balance_rmse": 1331.886378592337} {"canonical_length": 88, "canonical_nll": 1.3874861977317117, "generation_rouge_l_f1": 0.18279569892473116, "generation_token_f1": 0.3387096774193548, "prompt_id": "b0e33c086e3feda9", "reading_canonical_log_prob": -122.09878540039062, "reading_distances": [14, 17, 19, 20], "reading_log_probs": [-128.73355102539062, -136.07504272460938, -132.68055725097656, -134.22750854492188], "reading_logprob_std": 3.1223508801139825, "tokenization_path_logprob_std": 115.4291442647459, "trajectory_balance_rmse": 2814.931943123597} {"canonical_length": 53, "canonical_nll": 0.9095889397387235, "generation_rouge_l_f1": 0.6428571428571428, "generation_token_f1": 0.7857142857142858, "prompt_id": "85224f0bc07faa4c", "reading_canonical_log_prob": -48.208213806152344, "reading_distances": [18, 20, 17, 19], "reading_log_probs": [-55.597503662109375, -58.889957427978516, -54.91459655761719, -56.14815902709961], "reading_logprob_std": 1.7429014155406206, "tokenization_path_logprob_std": 137.55673820025055, "trajectory_balance_rmse": 1344.7878611907379} {"canonical_length": 26, "canonical_nll": 1.6142889169546275, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.2777777777777778, "prompt_id": "7327a63351d4e2c9", "reading_canonical_log_prob": -41.97151184082031, "reading_distances": [40, 29, 31, 36], "reading_log_probs": [-47.10228729248047, -48.29285430908203, -41.815460205078125, -49.61323547363281], "reading_logprob_std": 3.4178232738922096, "tokenization_path_logprob_std": 154.92115260299002, "trajectory_balance_rmse": 866.2746659576119} {"canonical_length": 84, "canonical_nll": 1.3819463820684523, "generation_rouge_l_f1": 0.2620689655172414, "generation_token_f1": 0.496551724137931, "prompt_id": "7fabfcdc4a3bff58", "reading_canonical_log_prob": -116.08349609375, "reading_distances": [29, 25, 32, 33], "reading_log_probs": [-121.95603942871094, -120.91352081298828, -122.47468566894531, -124.54529571533203], "reading_logprob_std": 1.5268305329835858, "tokenization_path_logprob_std": 150.19239578490584, "trajectory_balance_rmse": 2473.7105894664505} {"canonical_length": 13, "canonical_nll": 0.594976755288931, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "173c8d85c343f33f", "reading_canonical_log_prob": -7.7346978187561035, "reading_distances": [24, 32, 28, 32], "reading_log_probs": [-9.319028854370117, -9.286182403564453, -8.217105865478516, -9.38329792022705], "reading_logprob_std": 0.5576590693862992, "tokenization_path_logprob_std": 66.43418875245065, "trajectory_balance_rmse": 244.18185904102103} {"canonical_length": 118, "canonical_nll": 1.4819607492220603, "generation_rouge_l_f1": 0.3673469387755102, "generation_token_f1": 0.5306122448979592, "prompt_id": "e1f635b187b10d1b", "reading_canonical_log_prob": -174.87136840820312, "reading_distances": [29, 26, 23, 20], "reading_log_probs": [-165.13031005859375, -165.17674255371094, -165.13145446777344, -166.664306640625], "reading_logprob_std": 0.7593767459337318, "tokenization_path_logprob_std": 126.98354970911345, "trajectory_balance_rmse": 3171.84568836993} {"canonical_length": 242, "canonical_nll": 0.9513445767489347, "generation_rouge_l_f1": 0.35437881873727084, "generation_token_f1": 0.5824847250509164, "prompt_id": "5ff715a6459080e1", "reading_canonical_log_prob": -230.2253875732422, "reading_distances": [20, 30, 28, 26], "reading_log_probs": [-230.08587646484375, -235.19342041015625, -230.992919921875, -231.82305908203125], "reading_logprob_std": 2.2289797234590023, "tokenization_path_logprob_std": 193.29485464871104, "trajectory_balance_rmse": 4353.003977568659} {"canonical_length": 147, "canonical_nll": 0.7675235255234907, "generation_rouge_l_f1": 0.3573487031700288, "generation_token_f1": 0.409221902017291, "prompt_id": "3e82cf9b74d0c14b", "reading_canonical_log_prob": -112.82595825195312, "reading_distances": [24, 26, 25, 23], "reading_log_probs": [-117.79840087890625, -116.90641021728516, -118.19877624511719, -118.76036071777344], "reading_logprob_std": 0.7801727390971851, "tokenization_path_logprob_std": 169.80293393636256, "trajectory_balance_rmse": 2225.5254886608786} {"canonical_length": 42, "canonical_nll": 1.338357652936663, "generation_rouge_l_f1": 0.396551724137931, "generation_token_f1": 0.5344827586206896, "prompt_id": "b1354ebe20eb5cb8", "reading_canonical_log_prob": -56.211021423339844, "reading_distances": [46, 54, 47, 50], "reading_log_probs": [-65.63150024414062, -64.50822448730469, -64.55522155761719, -62.89373016357422], "reading_logprob_std": 1.128600296438737, "tokenization_path_logprob_std": 62.71231684230743, "trajectory_balance_rmse": 1257.1991425608294} {"canonical_length": 124, "canonical_nll": 1.2660564299552672, "generation_rouge_l_f1": 0.2081218274111675, "generation_token_f1": 0.416243654822335, "prompt_id": "751bafc67843f928", "reading_canonical_log_prob": -156.99099731445312, "reading_distances": [31, 31, 34, 36], "reading_log_probs": [-155.94371032714844, -154.49781799316406, -155.2750701904297, -157.44708251953125], "reading_logprob_std": 1.2522556982350976, "tokenization_path_logprob_std": 109.43896020760769, "trajectory_balance_rmse": 3436.740119171764} {"canonical_length": 283, "canonical_nll": 1.246165892260656, "generation_rouge_l_f1": 0.2559414990859232, "generation_token_f1": 0.45703839122486295, "prompt_id": "2c0f185491268c7a", "reading_canonical_log_prob": -352.6649475097656, "reading_distances": [23, 30, 38, 21], "reading_log_probs": [-353.6094970703125, -349.8527526855469, -349.579345703125, -348.72705078125], "reading_logprob_std": 2.1652923776732798, "tokenization_path_logprob_std": 206.3815494499234, "trajectory_balance_rmse": 6103.934350896174} {"canonical_length": 384, "canonical_nll": 1.4947118759155273, "generation_rouge_l_f1": 0.24620573355817876, "generation_token_f1": 0.4215851602023609, "prompt_id": "d93394062daf791b", "reading_canonical_log_prob": -573.9693603515625, "reading_distances": [25, 28, 29, 25], "reading_log_probs": [-575.63818359375, -564.5875854492188, -565.1670532226562, -569.64306640625], "reading_logprob_std": 5.112322411905707, "tokenization_path_logprob_std": 140.25952954338143, "trajectory_balance_rmse": 9011.273990025602} {"canonical_length": 13, "canonical_nll": 2.41601929297814, "generation_rouge_l_f1": 0.2727272727272727, "generation_token_f1": 0.2727272727272727, "prompt_id": "1bc766d97da83b42", "reading_canonical_log_prob": -31.40825080871582, "reading_distances": [20, 20, 19, 19], "reading_log_probs": [-30.28551483154297, -30.760051727294922, -33.32754135131836, -29.92641830444336], "reading_logprob_std": 1.5400932482961647, "tokenization_path_logprob_std": 32.80610599290916, "trajectory_balance_rmse": 360.95472341865525} {"canonical_length": 5, "canonical_nll": 0.6918039321899414, "generation_rouge_l_f1": 0.9090909090909091, "generation_token_f1": 0.9090909090909091, "prompt_id": "03cbf10fd71003f1", "reading_canonical_log_prob": -3.459019660949707, "reading_distances": [27, 20, 23, 23], "reading_log_probs": [-4.560503959655762, -4.248579025268555, -9.426944732666016, -2.917496681213379], "reading_logprob_std": 2.849549400339314, "tokenization_path_logprob_std": 29.187159316099095, "trajectory_balance_rmse": 140.0798493216335} {"canonical_length": 218, "canonical_nll": 1.3655070733586583, "generation_rouge_l_f1": 0.29729729729729726, "generation_token_f1": 0.48198198198198205, "prompt_id": "3b33c4e6381082ec", "reading_canonical_log_prob": -297.6805419921875, "reading_distances": [23, 20, 19, 25], "reading_log_probs": [-301.2023620605469, -303.05340576171875, -303.1020202636719, -301.49151611328125], "reading_logprob_std": 1.0064069919223653, "tokenization_path_logprob_std": 113.84928197550182, "trajectory_balance_rmse": 4772.222408035336} {"canonical_length": 14, "canonical_nll": 0.9676579066685268, "generation_rouge_l_f1": 0.7199999999999999, "generation_token_f1": 0.7199999999999999, "prompt_id": "275fffbea930b52f", "reading_canonical_log_prob": -13.547210693359375, "reading_distances": [12, 17, 19, 17], "reading_log_probs": [-21.163105010986328, -22.849712371826172, -28.2670841217041, -16.134422302246094], "reading_logprob_std": 5.002027368982892, "tokenization_path_logprob_std": 32.39010743465026, "trajectory_balance_rmse": 402.7188916800874} {"canonical_length": 181, "canonical_nll": 1.0809698789817852, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.42543859649122806, "prompt_id": "a7f0f685867c31e2", "reading_canonical_log_prob": -195.65554809570312, "reading_distances": [81, 74, 72, 75], "reading_log_probs": [-207.45343017578125, -206.48440551757812, -208.98062133789062, -206.5518798828125], "reading_logprob_std": 1.1625598848998382, "tokenization_path_logprob_std": 151.40870929842595, "trajectory_balance_rmse": 3850.031775005558} {"canonical_length": 128, "canonical_nll": 1.2958922386169434, "generation_rouge_l_f1": 0.2538860103626943, "generation_token_f1": 0.455958549222798, "prompt_id": "858a5416234d0d2a", "reading_canonical_log_prob": -165.87420654296875, "reading_distances": [19, 21, 21, 19], "reading_log_probs": [-167.75750732421875, -166.0908966064453, -165.33131408691406, -170.186279296875], "reading_logprob_std": 2.1502470861794856, "tokenization_path_logprob_std": 201.88207559478266, "trajectory_balance_rmse": 4032.2639910437897} {"canonical_length": 205, "canonical_nll": 0.8940399169921875, "generation_rouge_l_f1": 0.40105540897097625, "generation_token_f1": 0.5699208443271768, "prompt_id": "17197512e608a44f", "reading_canonical_log_prob": -183.27818298339844, "reading_distances": [16, 14, 12, 13], "reading_log_probs": [-211.0602569580078, -195.23895263671875, -191.4698486328125, -209.1503448486328], "reading_logprob_std": 9.823774129392518, "tokenization_path_logprob_std": 195.51665610688306, "trajectory_balance_rmse": 4811.807344474557} {"canonical_length": 160, "canonical_nll": 1.763667678833008, "generation_rouge_l_f1": 0.2885572139303482, "generation_token_f1": 0.46766169154228854, "prompt_id": "7c90d4dc6f1c3493", "reading_canonical_log_prob": -282.18682861328125, "reading_distances": [22, 24, 15, 17], "reading_log_probs": [-286.1904296875, -286.85662841796875, -283.23309326171875, -282.3802185058594], "reading_logprob_std": 2.1909476034245814, "tokenization_path_logprob_std": 130.64049488283644, "trajectory_balance_rmse": 4564.11710392492} {"canonical_length": 23, "canonical_nll": 5.349006321119226, "generation_rouge_l_f1": 0.5625000000000001, "generation_token_f1": 0.5625000000000001, "prompt_id": "d1e684b980513357", "reading_canonical_log_prob": -123.02714538574219, "reading_distances": [9, 19, 23, 13], "reading_log_probs": [-123.3920669555664, -122.15269470214844, -122.62747955322266, -122.9823989868164], "reading_logprob_std": 0.5266452776880366, "tokenization_path_logprob_std": 85.01328092904313, "trajectory_balance_rmse": 556.9900708190053} {"canonical_length": 392, "canonical_nll": 1.0701212980309311, "generation_rouge_l_f1": 0.2613803230543319, "generation_token_f1": 0.4258443465491924, "prompt_id": "5b2aa008c5db43fd", "reading_canonical_log_prob": -419.487548828125, "reading_distances": [7, 12, 10, 9], "reading_log_probs": [-428.20159912109375, -419.16400146484375, -422.3338317871094, -421.4599609375], "reading_logprob_std": 3.84752577113256, "tokenization_path_logprob_std": 277.61930664121087, "trajectory_balance_rmse": 10209.372931506961} {"canonical_length": 82, "canonical_nll": 1.4041783402605754, "generation_rouge_l_f1": 0.2885245901639344, "generation_token_f1": 0.360655737704918, "prompt_id": "180245122ca6704c", "reading_canonical_log_prob": -115.14262390136719, "reading_distances": [29, 24, 35, 31], "reading_log_probs": [-125.40786743164062, -120.51040649414062, -120.52293395996094, -118.56707763671875], "reading_logprob_std": 2.9189918581883303, "tokenization_path_logprob_std": 125.11235695560563, "trajectory_balance_rmse": 1738.101749593212} {"canonical_length": 99, "canonical_nll": 2.25469847399779, "generation_rouge_l_f1": 0.2336448598130841, "generation_token_f1": 0.39252336448598135, "prompt_id": "0280ead696ae0c9e", "reading_canonical_log_prob": -223.21514892578125, "reading_distances": [34, 24, 24, 32], "reading_log_probs": [-225.99818420410156, -224.2095947265625, -224.1767578125, -225.7633819580078], "reading_logprob_std": 0.9791360016443429, "tokenization_path_logprob_std": 213.70039678777422, "trajectory_balance_rmse": 3611.1608628267086} {"canonical_length": 148, "canonical_nll": 1.7589748485668286, "generation_rouge_l_f1": 0.17941952506596304, "generation_token_f1": 0.2955145118733509, "prompt_id": "1d5af645718b1db0", "reading_canonical_log_prob": -260.3282775878906, "reading_distances": [58, 46, 51, 52], "reading_log_probs": [-260.8916931152344, -256.30450439453125, -263.0611572265625, -262.8352355957031], "reading_logprob_std": 3.1342208011045747, "tokenization_path_logprob_std": 123.82788306091291, "trajectory_balance_rmse": 3346.3822277164445} {"canonical_length": 120, "canonical_nll": 1.5536703745524088, "generation_rouge_l_f1": 0.2666666666666666, "generation_token_f1": 0.42666666666666664, "prompt_id": "0211ac7bc9c6651a", "reading_canonical_log_prob": -186.44044494628906, "reading_distances": [17, 21, 20, 24], "reading_log_probs": [-194.44375610351562, -200.03665161132812, -193.52703857421875, -200.15872192382812], "reading_logprob_std": 3.5490711533765413, "tokenization_path_logprob_std": 103.52811102393348, "trajectory_balance_rmse": 3499.1835529843906} {"canonical_length": 115, "canonical_nll": 0.6717872951341712, "generation_rouge_l_f1": 0.3547400611620795, "generation_token_f1": 0.4587155963302752, "prompt_id": "0da3d3229c8dfac7", "reading_canonical_log_prob": -77.25553894042969, "reading_distances": [14, 15, 19, 15], "reading_log_probs": [-76.86417388916016, -77.4593276977539, -76.65248107910156, -76.34596252441406], "reading_logprob_std": 0.47011541312308386, "tokenization_path_logprob_std": 185.34917382173668, "trajectory_balance_rmse": 1997.80123743813} {"canonical_length": 196, "canonical_nll": 1.8098627207230549, "generation_rouge_l_f1": 0.2616407982261641, "generation_token_f1": 0.5232815964523282, "prompt_id": "dd26708537c260a2", "reading_canonical_log_prob": -354.73309326171875, "reading_distances": [35, 38, 37, 36], "reading_log_probs": [-351.488037109375, -354.71710205078125, -346.59234619140625, -349.5291748046875], "reading_logprob_std": 3.413002108088811, "tokenization_path_logprob_std": 364.46571331876396, "trajectory_balance_rmse": 6534.092465000544} {"canonical_length": 95, "canonical_nll": 1.8148999665912828, "generation_rouge_l_f1": 0.36046511627906974, "generation_token_f1": 0.5813953488372093, "prompt_id": "9fc2e86bc1fbb199", "reading_canonical_log_prob": -172.41549682617188, "reading_distances": [13, 14, 14, 17], "reading_log_probs": [-169.5731201171875, -171.6957244873047, -169.52362060546875, -169.66705322265625], "reading_logprob_std": 1.0555740235836248, "tokenization_path_logprob_std": 130.07973781748578, "trajectory_balance_rmse": 3194.0871389577783} {"canonical_length": 178, "canonical_nll": 1.284759350037307, "generation_rouge_l_f1": 0.34389140271493207, "generation_token_f1": 0.5294117647058824, "prompt_id": "0d0deb69605ca531", "reading_canonical_log_prob": -228.68716430664062, "reading_distances": [17, 19, 18, 14], "reading_log_probs": [-239.03323364257812, -226.1175537109375, -226.18316650390625, -244.547607421875], "reading_logprob_std": 9.306230918041612, "tokenization_path_logprob_std": 114.84565419692294, "trajectory_balance_rmse": 5952.736520956193} {"canonical_length": 20, "canonical_nll": 0.38579788208007815, "generation_rouge_l_f1": 0.5853658536585366, "generation_token_f1": 0.5853658536585366, "prompt_id": "5b98bbd785876818", "reading_canonical_log_prob": -7.7159576416015625, "reading_distances": [25, 35, 29, 29], "reading_log_probs": [-9.136632919311523, -11.554962158203125, -9.297542572021484, -11.276765823364258], "reading_logprob_std": 1.2762261469684946, "tokenization_path_logprob_std": 27.116719517199794, "trajectory_balance_rmse": 324.99582997310785} {"canonical_length": 322, "canonical_nll": 1.072064607039742, "generation_rouge_l_f1": 0.2344582593250444, "generation_token_f1": 0.49023090586145646, "prompt_id": "9d8ebc1fd91528b6", "reading_canonical_log_prob": -345.2048034667969, "reading_distances": [21, 16, 23, 24], "reading_log_probs": [-348.616943359375, -346.0535888671875, -350.9527893066406, -350.97149658203125], "reading_logprob_std": 2.34092600303288, "tokenization_path_logprob_std": 266.0126523261546, "trajectory_balance_rmse": 7682.821925733826} {"canonical_length": 202, "canonical_nll": 1.2381628055383664, "generation_rouge_l_f1": 0.27927927927927926, "generation_token_f1": 0.527027027027027, "prompt_id": "5f17b0be639454cf", "reading_canonical_log_prob": -250.10888671875, "reading_distances": [20, 15, 22, 24], "reading_log_probs": [-248.63397216796875, -249.6968994140625, -252.6700439453125, -248.86587524414062], "reading_logprob_std": 1.859107264270699, "tokenization_path_logprob_std": 323.3254104165037, "trajectory_balance_rmse": 5850.281828252283} {"canonical_length": 11, "canonical_nll": 0.4604782624678178, "generation_rouge_l_f1": 0.8421052631578948, "generation_token_f1": 0.8421052631578948, "prompt_id": "1932b722766373ae", "reading_canonical_log_prob": -5.065260887145996, "reading_distances": [39, 43, 40, 40], "reading_log_probs": [-4.31124210357666, -5.070369243621826, -4.280132293701172, -2.992809295654297], "reading_logprob_std": 0.8618505854530328, "tokenization_path_logprob_std": 10.803815136613053, "trajectory_balance_rmse": 252.38816027232795} {"canonical_length": 31, "canonical_nll": 2.040472215221774, "generation_rouge_l_f1": 0.13903743315508021, "generation_token_f1": 0.20320855614973263, "prompt_id": "3f8935161866e686", "reading_canonical_log_prob": -63.254638671875, "reading_distances": [29, 21, 32, 27], "reading_log_probs": [-54.25276184082031, -61.99882125854492, -52.97559356689453, -63.61017990112305], "reading_logprob_std": 5.372021726878532, "tokenization_path_logprob_std": 79.56126142097028, "trajectory_balance_rmse": 732.9880304167858} {"canonical_length": 45, "canonical_nll": 1.217963155110677, "generation_rouge_l_f1": 0.25333333333333335, "generation_token_f1": 0.37333333333333335, "prompt_id": "88f54467af0c5001", "reading_canonical_log_prob": -54.80834197998047, "reading_distances": [23, 28, 28, 26], "reading_log_probs": [-58.94496154785156, -67.1366195678711, -58.760074615478516, -61.02438735961914], "reading_logprob_std": 3.91699774606013, "tokenization_path_logprob_std": 121.35396728810487, "trajectory_balance_rmse": 1193.0044276540868} {"canonical_length": 272, "canonical_nll": 1.462099411908318, "generation_rouge_l_f1": 0.2758620689655172, "generation_token_f1": 0.5632183908045978, "prompt_id": "05940c01ebeedfd9", "reading_canonical_log_prob": -397.6910400390625, "reading_distances": [21, 29, 30, 24], "reading_log_probs": [-400.0385437011719, -404.178955078125, -403.0753173828125, -406.35400390625], "reading_logprob_std": 2.6291424021919956, "tokenization_path_logprob_std": 243.38038152205687, "trajectory_balance_rmse": 8923.951565920452} {"canonical_length": 26, "canonical_nll": 1.1506596345167894, "generation_rouge_l_f1": 0.2711864406779661, "generation_token_f1": 0.2711864406779661, "prompt_id": "cc3b11b7124977c4", "reading_canonical_log_prob": -29.917150497436523, "reading_distances": [22, 25, 18, 24], "reading_log_probs": [-30.837127685546875, -32.17002487182617, -30.57025146484375, -29.186603546142578], "reading_logprob_std": 1.2229277399026972, "tokenization_path_logprob_std": 77.66947527272043, "trajectory_balance_rmse": 741.1753156398931} {"canonical_length": 25, "canonical_nll": 1.1257044219970702, "generation_rouge_l_f1": 0.14207650273224043, "generation_token_f1": 0.15300546448087432, "prompt_id": "9335f10e0202a2ca", "reading_canonical_log_prob": -28.142610549926758, "reading_distances": [21, 24, 28, 25], "reading_log_probs": [-33.12351608276367, -39.545101165771484, -35.403892517089844, -32.32750701904297], "reading_logprob_std": 3.237513502693368, "tokenization_path_logprob_std": 25.530473615304405, "trajectory_balance_rmse": 307.33381381000146} {"canonical_length": 160, "canonical_nll": 1.3617704391479493, "generation_rouge_l_f1": 0.21897810218978103, "generation_token_f1": 0.44768856447688565, "prompt_id": "ee0b55cb338c65fa", "reading_canonical_log_prob": -217.88327026367188, "reading_distances": [31, 27, 24, 29], "reading_log_probs": [-237.530029296875, -244.92942810058594, -244.60992431640625, -236.8483123779297], "reading_logprob_std": 4.387386161638913, "tokenization_path_logprob_std": 260.00137756499294, "trajectory_balance_rmse": 4048.545774479846} {"canonical_length": 205, "canonical_nll": 0.5279657503453696, "generation_rouge_l_f1": 0.5200845665961946, "generation_token_f1": 0.718816067653277, "prompt_id": "9ed2cebe3982e4a6", "reading_canonical_log_prob": -108.23297882080078, "reading_distances": [11, 9, 12, 13], "reading_log_probs": [-107.11878204345703, -109.39852142333984, -113.4798583984375, -110.12615966796875], "reading_logprob_std": 2.6321551420934433, "tokenization_path_logprob_std": 337.57642930735767, "trajectory_balance_rmse": 2916.5608085736844} {"canonical_length": 67, "canonical_nll": 1.6349897242304106, "generation_rouge_l_f1": 0.3186813186813187, "generation_token_f1": 0.4835164835164836, "prompt_id": "57c5e7bb82e63751", "reading_canonical_log_prob": -109.5443115234375, "reading_distances": [25, 31, 31, 31], "reading_log_probs": [-108.04971313476562, -106.5525131225586, -108.39555358886719, -105.55931091308594], "reading_logprob_std": 1.3225870490339562, "tokenization_path_logprob_std": 178.47400916015212, "trajectory_balance_rmse": 1847.902766923778} {"canonical_length": 233, "canonical_nll": 1.3975824839055795, "generation_rouge_l_f1": 0.24847250509164973, "generation_token_f1": 0.5539714867617107, "prompt_id": "2049d32643760e75", "reading_canonical_log_prob": -325.63671875, "reading_distances": [33, 23, 35, 30], "reading_log_probs": [-327.52911376953125, -328.185791015625, -328.2901306152344, -330.30389404296875], "reading_logprob_std": 1.199384131551446, "tokenization_path_logprob_std": 293.1530461770801, "trajectory_balance_rmse": 6002.559768212386} {"canonical_length": 65, "canonical_nll": 1.4279872013972357, "generation_rouge_l_f1": 0.3263598326359833, "generation_token_f1": 0.401673640167364, "prompt_id": "9327dc9ed2ef5314", "reading_canonical_log_prob": -92.81916809082031, "reading_distances": [14, 17, 18, 19], "reading_log_probs": [-95.01461029052734, -95.74903106689453, -94.39468383789062, -95.69404602050781], "reading_logprob_std": 0.6397224299299027, "tokenization_path_logprob_std": 130.5091912126072, "trajectory_balance_rmse": 2047.750563601751} {"canonical_length": 27, "canonical_nll": 1.807416421395761, "generation_rouge_l_f1": 0.42553191489361697, "generation_token_f1": 0.5106382978723405, "prompt_id": "fb2aa3ee59bdbf8a", "reading_canonical_log_prob": -48.80024337768555, "reading_distances": [103, 96, 95, 98], "reading_log_probs": [-50.854942321777344, -52.851505279541016, -51.30710220336914, -50.79283905029297], "reading_logprob_std": 0.96100308336791, "tokenization_path_logprob_std": 182.60766097642164, "trajectory_balance_rmse": 910.2137471536518} {"canonical_length": 176, "canonical_nll": 0.847486062483354, "generation_rouge_l_f1": 0.44501278772378516, "generation_token_f1": 0.608695652173913, "prompt_id": "24055c0763c8403a", "reading_canonical_log_prob": -149.1575469970703, "reading_distances": [140, 136, 147, 136], "reading_log_probs": [-152.03689575195312, -150.9664764404297, -147.05624389648438, -146.6357421875], "reading_logprob_std": 2.728661907100829, "tokenization_path_logprob_std": 153.0461581880803, "trajectory_balance_rmse": 4390.584846556426} {"canonical_length": 73, "canonical_nll": 0.5697289035744864, "generation_rouge_l_f1": 0.36571428571428566, "generation_token_f1": 0.41142857142857137, "prompt_id": "c59968b8c3d9c69d", "reading_canonical_log_prob": -41.5902099609375, "reading_distances": [15, 15, 12, 15], "reading_log_probs": [-41.30736541748047, -42.848358154296875, -41.60076141357422, -41.680015563964844], "reading_logprob_std": 0.6786881471706576, "tokenization_path_logprob_std": 255.4015310742746, "trajectory_balance_rmse": 1659.9104917019943} {"canonical_length": 283, "canonical_nll": 1.7069456282437059, "generation_rouge_l_f1": 0.2732447817836812, "generation_token_f1": 0.5123339658444022, "prompt_id": "c3213521900f0885", "reading_canonical_log_prob": -483.06561279296875, "reading_distances": [39, 42, 35, 42], "reading_log_probs": [-478.76165771484375, -482.1243591308594, -476.99578857421875, -482.2994079589844], "reading_logprob_std": 2.6045311166332956, "tokenization_path_logprob_std": 219.18623474382449, "trajectory_balance_rmse": 6577.099183250214} {"canonical_length": 70, "canonical_nll": 1.7686713082449776, "generation_rouge_l_f1": 0.17808219178082188, "generation_token_f1": 0.2671232876712329, "prompt_id": "48bb22e3d5050293", "reading_canonical_log_prob": -123.80699157714844, "reading_distances": [16, 17, 20, 23], "reading_log_probs": [-130.0100860595703, -126.80400848388672, -130.41000366210938, -135.53170776367188], "reading_logprob_std": 3.6094201127168235, "tokenization_path_logprob_std": 87.38490073130069, "trajectory_balance_rmse": 1588.4384344333214} {"canonical_length": 183, "canonical_nll": 1.3904282199880464, "generation_rouge_l_f1": 0.2983682983682984, "generation_token_f1": 0.5594405594405594, "prompt_id": "5a342c18ce379992", "reading_canonical_log_prob": -254.4483642578125, "reading_distances": [30, 32, 26, 26], "reading_log_probs": [-250.80618286132812, -246.0463104248047, -246.37350463867188, -243.1443634033203], "reading_logprob_std": 3.161805983553699, "tokenization_path_logprob_std": 350.8452106264041, "trajectory_balance_rmse": 5128.220765522299} {"canonical_length": 280, "canonical_nll": 1.593996320452009, "generation_rouge_l_f1": 0.30158730158730157, "generation_token_f1": 0.5753968253968254, "prompt_id": "66f2043754e3c9e5", "reading_canonical_log_prob": -446.3189697265625, "reading_distances": [21, 15, 28, 21], "reading_log_probs": [-448.89007568359375, -447.7916259765625, -448.3170166015625, -451.34698486328125], "reading_logprob_std": 1.5723843618540778, "tokenization_path_logprob_std": 379.90486288509084, "trajectory_balance_rmse": 7449.073208799947} {"canonical_length": 61, "canonical_nll": 1.9804422347272028, "generation_rouge_l_f1": 0.1787709497206704, "generation_token_f1": 0.29050279329608936, "prompt_id": "b6ca714dbce654ce", "reading_canonical_log_prob": -120.80697631835938, "reading_distances": [44, 40, 44, 45], "reading_log_probs": [-129.64535522460938, -129.26748657226562, -127.89726257324219, -127.51943969726562], "reading_logprob_std": 1.0325559418217796, "tokenization_path_logprob_std": 196.9060011717569, "trajectory_balance_rmse": 2208.136914581426} {"canonical_length": 38, "canonical_nll": 1.127305081016139, "generation_rouge_l_f1": 0.42477876106194695, "generation_token_f1": 0.5486725663716814, "prompt_id": "4112128a24dc114d", "reading_canonical_log_prob": -42.83759307861328, "reading_distances": [28, 33, 29, 29], "reading_log_probs": [-43.158531188964844, -43.240055084228516, -43.60192108154297, -43.9038200378418], "reading_logprob_std": 0.34416203839563764, "tokenization_path_logprob_std": 199.99225517828842, "trajectory_balance_rmse": 1091.9902213053133} {"canonical_length": 24, "canonical_nll": 0.6264557043711344, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5384615384615385, "prompt_id": "07948a71ef97c7bb", "reading_canonical_log_prob": -15.034936904907227, "reading_distances": [22, 17, 20, 21], "reading_log_probs": [-12.406822204589844, -11.716981887817383, -11.816385269165039, -12.219442367553711], "reading_logprob_std": 0.32716066786416986, "tokenization_path_logprob_std": 120.79726352349765, "trajectory_balance_rmse": 832.2334533218972} {"canonical_length": 130, "canonical_nll": 1.6909156212439904, "generation_rouge_l_f1": 0.2065491183879093, "generation_token_f1": 0.3324937027707809, "prompt_id": "d894cb33b4a83ecb", "reading_canonical_log_prob": -219.81903076171875, "reading_distances": [22, 32, 25, 33], "reading_log_probs": [-223.5904541015625, -223.86224365234375, -231.6512451171875, -223.0457763671875], "reading_logprob_std": 4.0899893056717715, "tokenization_path_logprob_std": 83.15314129895397, "trajectory_balance_rmse": 3983.663307728082} {"canonical_length": 49, "canonical_nll": 0.4290499395253707, "generation_rouge_l_f1": 0.17647058823529413, "generation_token_f1": 0.21568627450980393, "prompt_id": "98f5a9670dc63974", "reading_canonical_log_prob": -21.023447036743164, "reading_distances": [14, 19, 21, 21], "reading_log_probs": [-23.312484741210938, -21.089065551757812, -24.63500213623047, -25.297008514404297], "reading_logprob_std": 1.8562884662963386, "tokenization_path_logprob_std": 92.56483989898217, "trajectory_balance_rmse": 1085.8784894084044} {"canonical_length": 155, "canonical_nll": 1.5032609508883568, "generation_rouge_l_f1": 0.22051282051282048, "generation_token_f1": 0.3641025641025641, "prompt_id": "9866aa5bc569efc5", "reading_canonical_log_prob": -233.0054473876953, "reading_distances": [11, 8, 10, 13], "reading_log_probs": [-234.35191345214844, -233.47479248046875, -233.09222412109375, -233.52197265625], "reading_logprob_std": 0.5305836767561133, "tokenization_path_logprob_std": 170.76590498205164, "trajectory_balance_rmse": 4442.1003803926105} {"canonical_length": 173, "canonical_nll": 2.030640706850614, "generation_rouge_l_f1": 0.182648401826484, "generation_token_f1": 0.37442922374429227, "prompt_id": "63dbcd016e93c526", "reading_canonical_log_prob": -351.30084228515625, "reading_distances": [23, 18, 27, 28], "reading_log_probs": [-354.2186584472656, -360.035888671875, -351.4203186035156, -354.89239501953125], "reading_logprob_std": 3.5923957132763804, "tokenization_path_logprob_std": 215.82219744500992, "trajectory_balance_rmse": 4730.036866341311} {"canonical_length": 45, "canonical_nll": 1.118312242296007, "generation_rouge_l_f1": 0.2807017543859649, "generation_token_f1": 0.3859649122807018, "prompt_id": "bb113220673ce1f4", "reading_canonical_log_prob": -50.32405090332031, "reading_distances": [24, 30, 27, 22], "reading_log_probs": [-52.64315414428711, -53.26518249511719, -51.18494415283203, -53.4941291809082], "reading_logprob_std": 1.0388133374555801, "tokenization_path_logprob_std": 18.530055455369407, "trajectory_balance_rmse": 739.484330474554} {"canonical_length": 16, "canonical_nll": 2.109546661376953, "generation_rouge_l_f1": 0.29629629629629634, "generation_token_f1": 0.3703703703703704, "prompt_id": "5d38e8c442eebce3", "reading_canonical_log_prob": -33.75274658203125, "reading_distances": [37, 40, 42, 45], "reading_log_probs": [-35.810855865478516, -40.16914367675781, -38.16136169433594, -37.867576599121094], "reading_logprob_std": 1.783357312711131, "tokenization_path_logprob_std": 62.12204693102883, "trajectory_balance_rmse": 459.42304881469914} {"canonical_length": 7, "canonical_nll": 1.1739278520856584, "generation_rouge_l_f1": 0.043668122270742356, "generation_token_f1": 0.043668122270742356, "prompt_id": "8ee1f7d9e89653dc", "reading_canonical_log_prob": -8.21749496459961, "reading_distances": [21, 23, 24, 19], "reading_log_probs": [-7.8923659324646, -9.256404876708984, -8.154012680053711, -10.278630256652832], "reading_logprob_std": 1.0953512304383817, "tokenization_path_logprob_std": 56.48473751819664, "trajectory_balance_rmse": 163.05007072946202} {"canonical_length": 220, "canonical_nll": 1.8032723166725853, "generation_rouge_l_f1": 0.18951612903225806, "generation_token_f1": 0.3870967741935483, "prompt_id": "8768fc8d898485e9", "reading_canonical_log_prob": -396.71990966796875, "reading_distances": [16, 15, 12, 16], "reading_log_probs": [-398.490966796875, -407.09881591796875, -397.7723083496094, -404.9754638671875], "reading_logprob_std": 4.655087319937099, "tokenization_path_logprob_std": 85.8333935924121, "trajectory_balance_rmse": 6572.878544562255} {"canonical_length": 238, "canonical_nll": 1.1067895969422925, "generation_rouge_l_f1": 0.2533936651583711, "generation_token_f1": 0.497737556561086, "prompt_id": "f7a53c2bee8bfe30", "reading_canonical_log_prob": -263.4159240722656, "reading_distances": [14, 15, 19, 24], "reading_log_probs": [-262.7723388671875, -263.5025329589844, -267.7378845214844, -268.5414733886719], "reading_logprob_std": 2.9218661097969925, "tokenization_path_logprob_std": 80.3967121519751, "trajectory_balance_rmse": 5452.672339035499} {"canonical_length": 218, "canonical_nll": 1.3761778665245126, "generation_rouge_l_f1": 0.24583333333333332, "generation_token_f1": 0.49166666666666664, "prompt_id": "90b0a660ae2019c3", "reading_canonical_log_prob": -300.00677490234375, "reading_distances": [31, 29, 40, 26], "reading_log_probs": [-302.23602294921875, -302.0235900878906, -304.29937744140625, -303.74688720703125], "reading_logprob_std": 1.1195042349509552, "tokenization_path_logprob_std": 273.5582622517298, "trajectory_balance_rmse": 5148.297630015796} {"canonical_length": 153, "canonical_nll": 1.3196527817670036, "generation_rouge_l_f1": 0.18918918918918917, "generation_token_f1": 0.2972972972972973, "prompt_id": "e11b424a8c471248", "reading_canonical_log_prob": -201.90687561035156, "reading_distances": [38, 51, 52, 57], "reading_log_probs": [-199.5303192138672, -196.99093627929688, -198.44496154785156, -198.13758850097656], "reading_logprob_std": 1.04441549758778, "tokenization_path_logprob_std": 307.6762799468445, "trajectory_balance_rmse": 3695.6455403931413} {"canonical_length": 69, "canonical_nll": 1.887952500495358, "generation_rouge_l_f1": 0.28346456692913385, "generation_token_f1": 0.4251968503937008, "prompt_id": "cc633ec6fcbbe3de", "reading_canonical_log_prob": -130.2687225341797, "reading_distances": [18, 20, 20, 22], "reading_log_probs": [-135.77235412597656, -136.552978515625, -141.9032440185547, -137.36126708984375], "reading_logprob_std": 2.7481820583701375, "tokenization_path_logprob_std": 173.31463998332978, "trajectory_balance_rmse": 1707.0247755294945} {"canonical_length": 51, "canonical_nll": 2.125023037779565, "generation_rouge_l_f1": 0.27499999999999997, "generation_token_f1": 0.5, "prompt_id": "66e12e5aa4182ef3", "reading_canonical_log_prob": -108.37617492675781, "reading_distances": [33, 39, 39, 49], "reading_log_probs": [-110.9524917602539, -108.21757507324219, -109.58361053466797, -110.22083282470703], "reading_logprob_std": 1.1609649951736183, "tokenization_path_logprob_std": 127.49439296237918, "trajectory_balance_rmse": 1284.0921213211732} {"canonical_length": 147, "canonical_nll": 1.4451996680019663, "generation_rouge_l_f1": 0.224390243902439, "generation_token_f1": 0.39024390243902435, "prompt_id": "831cd880037811e2", "reading_canonical_log_prob": -212.44435119628906, "reading_distances": [19, 19, 28, 19], "reading_log_probs": [-208.61654663085938, -210.23123168945312, -209.92877197265625, -209.72702026367188], "reading_logprob_std": 0.7040779827501912, "tokenization_path_logprob_std": 246.08231006267036, "trajectory_balance_rmse": 4919.849802678018} {"canonical_length": 239, "canonical_nll": 1.0361543918753269, "generation_rouge_l_f1": 0.23228346456692914, "generation_token_f1": 0.44488188976377957, "prompt_id": "61ee3d334ec189e4", "reading_canonical_log_prob": -247.64089965820312, "reading_distances": [22, 26, 31, 29], "reading_log_probs": [-247.82745361328125, -248.8636016845703, -247.744140625, -247.33026123046875], "reading_logprob_std": 0.6521353585513896, "tokenization_path_logprob_std": 371.21231803906284, "trajectory_balance_rmse": 7526.094376558148} {"canonical_length": 239, "canonical_nll": 1.5288711611695869, "generation_rouge_l_f1": 0.28632478632478636, "generation_token_f1": 0.5170940170940171, "prompt_id": "978f2098f6623381", "reading_canonical_log_prob": -365.40020751953125, "reading_distances": [16, 19, 23, 20], "reading_log_probs": [-372.083740234375, -372.192626953125, -370.27105712890625, -370.8479919433594], "reading_logprob_std": 0.9424293906390526, "tokenization_path_logprob_std": 451.82082599442754, "trajectory_balance_rmse": 6200.905529000783} {"canonical_length": 366, "canonical_nll": 1.9723057147583676, "generation_rouge_l_f1": 0.21122112211221125, "generation_token_f1": 0.44884488448844884, "prompt_id": "f4a2dbf6cf78f4f4", "reading_canonical_log_prob": -721.8638916015625, "reading_distances": [41, 37, 37, 36], "reading_log_probs": [-724.6332397460938, -722.70361328125, -724.3801879882812, -725.4208984375], "reading_logprob_std": 1.1432963784547072, "tokenization_path_logprob_std": 160.73702190839404, "trajectory_balance_rmse": 9848.861727180038} {"canonical_length": 323, "canonical_nll": 1.670924136513158, "generation_rouge_l_f1": 0.213768115942029, "generation_token_f1": 0.4239130434782608, "prompt_id": "607d603874554f5a", "reading_canonical_log_prob": -539.70849609375, "reading_distances": [21, 20, 22, 23], "reading_log_probs": [-552.3588256835938, -552.9492797851562, -544.5357666015625, -543.645263671875], "reading_logprob_std": 4.963365522955473, "tokenization_path_logprob_std": 372.1547112985782, "trajectory_balance_rmse": 7703.329291312019} {"canonical_length": 16, "canonical_nll": 1.3270922899246216, "generation_rouge_l_f1": 0.16, "generation_token_f1": 0.16, "prompt_id": "65ee5494d4aae49e", "reading_canonical_log_prob": -21.233476638793945, "reading_distances": [15, 17, 22, 18], "reading_log_probs": [-24.782024383544922, -27.68279266357422, -28.134742736816406, -27.41232681274414], "reading_logprob_std": 1.5103245581761815, "tokenization_path_logprob_std": 59.900940172007324, "trajectory_balance_rmse": 429.0764240276027} {"canonical_length": 120, "canonical_nll": 2.1417686462402346, "generation_rouge_l_f1": 0.2615384615384616, "generation_token_f1": 0.4307692307692308, "prompt_id": "bc654a75738c95db", "reading_canonical_log_prob": -257.0122375488281, "reading_distances": [27, 29, 26, 27], "reading_log_probs": [-244.7807159423828, -247.14801025390625, -244.67410278320312, -242.70578002929688], "reading_logprob_std": 1.8177075711839659, "tokenization_path_logprob_std": 182.78057109038775, "trajectory_balance_rmse": 3131.2320925599743} {"canonical_length": 166, "canonical_nll": 1.1323955489928463, "generation_rouge_l_f1": 0.29055690072639223, "generation_token_f1": 0.5036319612590798, "prompt_id": "1277e399b9abf4ff", "reading_canonical_log_prob": -187.9776611328125, "reading_distances": [28, 32, 24, 24], "reading_log_probs": [-188.63865661621094, -188.91925048828125, -186.15013122558594, -189.0054931640625], "reading_logprob_std": 1.3612069302499268, "tokenization_path_logprob_std": 160.4426242068016, "trajectory_balance_rmse": 5744.165460169887} {"canonical_length": 13, "canonical_nll": 1.9947207524226263, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "888761fd7cf1d9af", "reading_canonical_log_prob": -25.93136978149414, "reading_distances": [29, 23, 26, 29], "reading_log_probs": [-26.708301544189453, -28.68332862854004, -26.065589904785156, -27.264833450317383], "reading_logprob_std": 1.1152879925483372, "tokenization_path_logprob_std": 74.93894180656245, "trajectory_balance_rmse": 471.98603593831467} {"canonical_length": 196, "canonical_nll": 1.4346377314353476, "generation_rouge_l_f1": 0.2434077079107505, "generation_token_f1": 0.42596348884381335, "prompt_id": "4049144029e8d577", "reading_canonical_log_prob": -281.1889953613281, "reading_distances": [20, 24, 26, 23], "reading_log_probs": [-281.165771484375, -280.95904541015625, -283.15533447265625, -281.50347900390625], "reading_logprob_std": 0.9984931562337827, "tokenization_path_logprob_std": 165.90348340720024, "trajectory_balance_rmse": 7063.417689381709} {"canonical_length": 105, "canonical_nll": 1.2214794340587798, "generation_rouge_l_f1": 0.33155080213903737, "generation_token_f1": 0.427807486631016, "prompt_id": "35fc3a98ce123749", "reading_canonical_log_prob": -128.25534057617188, "reading_distances": [37, 34, 41, 33], "reading_log_probs": [-131.64540100097656, -128.6552276611328, -127.24372863769531, -127.029296875], "reading_logprob_std": 2.127326861210962, "tokenization_path_logprob_std": 294.90097084946956, "trajectory_balance_rmse": 2703.635087089197} {"canonical_length": 17, "canonical_nll": 5.713744219611673, "generation_rouge_l_f1": 0.030651340996168584, "generation_token_f1": 0.038314176245210725, "prompt_id": "30594b15ee7b3362", "reading_canonical_log_prob": -97.13365173339844, "reading_distances": [20, 22, 21, 20], "reading_log_probs": [-92.32402801513672, -92.19158935546875, -90.55841064453125, -91.3329849243164], "reading_logprob_std": 0.8226760547029687, "tokenization_path_logprob_std": 110.23932870892347, "trajectory_balance_rmse": 486.450262356391} {"canonical_length": 89, "canonical_nll": 1.2489773182386763, "generation_rouge_l_f1": 0.22527472527472528, "generation_token_f1": 0.3516483516483516, "prompt_id": "55622d0a348ebb7e", "reading_canonical_log_prob": -111.15898132324219, "reading_distances": [18, 15, 16, 18], "reading_log_probs": [-115.3720932006836, -113.0491943359375, -112.62723541259766, -112.18236541748047], "reading_logprob_std": 1.4210271450318284, "tokenization_path_logprob_std": 135.10031731204373, "trajectory_balance_rmse": 2414.8742139942046} {"canonical_length": 181, "canonical_nll": 1.3336233065273222, "generation_rouge_l_f1": 0.3295194508009153, "generation_token_f1": 0.5034324942791762, "prompt_id": "96bafc4191db5d29", "reading_canonical_log_prob": -241.3858184814453, "reading_distances": [35, 26, 23, 33], "reading_log_probs": [-255.33392333984375, -268.1103820800781, -253.44430541992188, -258.2286682128906], "reading_logprob_std": 6.524443160106474, "tokenization_path_logprob_std": 161.33989135414166, "trajectory_balance_rmse": 4744.995984324133} {"canonical_length": 186, "canonical_nll": 1.3883479948966735, "generation_rouge_l_f1": 0.2535885167464115, "generation_token_f1": 0.4641148325358852, "prompt_id": "a69ab9afafd56708", "reading_canonical_log_prob": -258.23272705078125, "reading_distances": [26, 30, 32, 31], "reading_log_probs": [-265.771240234375, -267.34808349609375, -267.0660400390625, -266.573486328125], "reading_logprob_std": 0.6909352625018134, "tokenization_path_logprob_std": 166.42036094235868, "trajectory_balance_rmse": 4663.877206800973} {"canonical_length": 7, "canonical_nll": 1.1519214085170202, "generation_rouge_l_f1": 0.13559322033898305, "generation_token_f1": 0.13559322033898305, "prompt_id": "7d277a000cf626ea", "reading_canonical_log_prob": -8.06344985961914, "reading_distances": [35, 27, 28, 28], "reading_log_probs": [-8.433649063110352, -8.90920639038086, -9.72574520111084, -8.311214447021484], "reading_logprob_std": 0.6413452453365067, "tokenization_path_logprob_std": 63.41862952586016, "trajectory_balance_rmse": 243.8324470314206} {"canonical_length": 8, "canonical_nll": 0.8435441851615906, "generation_rouge_l_f1": 0.13333333333333333, "generation_token_f1": 0.13333333333333333, "prompt_id": "4b1173c764ce0a81", "reading_canonical_log_prob": -6.748353481292725, "reading_distances": [45, 57, 50, 50], "reading_log_probs": [-6.237339973449707, -6.0357184410095215, -5.881245136260986, -6.458704948425293], "reading_logprob_std": 0.2504490383229182, "tokenization_path_logprob_std": 33.88823520481091, "trajectory_balance_rmse": 180.0608954914046} {"canonical_length": 48, "canonical_nll": 1.150264581044515, "generation_rouge_l_f1": 0.1839080459770115, "generation_token_f1": 0.20689655172413793, "prompt_id": "c86574795b337627", "reading_canonical_log_prob": -55.21269989013672, "reading_distances": [8, 10, 9, 8], "reading_log_probs": [-57.63299560546875, -51.602684020996094, -55.807769775390625, -55.27937698364258], "reading_logprob_std": 2.528432387194579, "tokenization_path_logprob_std": 22.192599810241255, "trajectory_balance_rmse": 697.0917870626229} {"canonical_length": 251, "canonical_nll": 1.3771432823393925, "generation_rouge_l_f1": 0.2064516129032258, "generation_token_f1": 0.36129032258064514, "prompt_id": "f0455cfad098aced", "reading_canonical_log_prob": -345.6629638671875, "reading_distances": [40, 40, 42, 42], "reading_log_probs": [-348.40032958984375, -353.51953125, -348.430419921875, -344.79803466796875], "reading_logprob_std": 3.586333307883592, "tokenization_path_logprob_std": 253.83716836577884, "trajectory_balance_rmse": 7200.878647271528} {"canonical_length": 92, "canonical_nll": 1.5493411188540251, "generation_rouge_l_f1": 0.2569444444444444, "generation_token_f1": 0.3263888888888889, "prompt_id": "765c339341bf24b5", "reading_canonical_log_prob": -142.5393829345703, "reading_distances": [73, 64, 58, 62], "reading_log_probs": [-140.29244995117188, -146.4362335205078, -148.46603393554688, -148.62144470214844], "reading_logprob_std": 3.9034716962159095, "tokenization_path_logprob_std": 159.1577228814825, "trajectory_balance_rmse": 1816.6130447768464} {"canonical_length": 25, "canonical_nll": 2.3539889526367186, "generation_rouge_l_f1": 0.24000000000000002, "generation_token_f1": 0.24000000000000002, "prompt_id": "ef02d7967a69b355", "reading_canonical_log_prob": -58.84972381591797, "reading_distances": [34, 29, 29, 42], "reading_log_probs": [-63.880287170410156, -67.13015747070312, -66.47453308105469, -67.28021240234375], "reading_logprob_std": 1.5798929127192252, "tokenization_path_logprob_std": 40.83440856792253, "trajectory_balance_rmse": 638.255958580273} {"canonical_length": 26, "canonical_nll": 1.2487041766826923, "generation_rouge_l_f1": 0.5142857142857143, "generation_token_f1": 0.5142857142857143, "prompt_id": "27392a7bee3d725d", "reading_canonical_log_prob": -32.46630859375, "reading_distances": [30, 32, 35, 30], "reading_log_probs": [-32.56504821777344, -29.154050827026367, -29.706466674804688, -28.19807243347168], "reading_logprob_std": 1.879077350795453, "tokenization_path_logprob_std": 43.72617665650952, "trajectory_balance_rmse": 792.4924248726037} {"canonical_length": 53, "canonical_nll": 1.2030954900777564, "generation_rouge_l_f1": 0.27802690582959644, "generation_token_f1": 0.3318385650224216, "prompt_id": "deecb5d2b0490e50", "reading_canonical_log_prob": -63.764060974121094, "reading_distances": [15, 23, 21, 17], "reading_log_probs": [-65.54649353027344, -66.96549224853516, -67.04351806640625, -65.97125244140625], "reading_logprob_std": 0.7404623752364472, "tokenization_path_logprob_std": 94.46995725939094, "trajectory_balance_rmse": 1519.4134756117496} {"canonical_length": 124, "canonical_nll": 1.3234364909510459, "generation_rouge_l_f1": 0.1994750656167979, "generation_token_f1": 0.37270341207349084, "prompt_id": "8c1c19dea7da1533", "reading_canonical_log_prob": -164.1061248779297, "reading_distances": [26, 29, 26, 20], "reading_log_probs": [-165.74502563476562, -164.27206420898438, -159.8192901611328, -155.31332397460938], "reading_logprob_std": 4.712521532878647, "tokenization_path_logprob_std": 331.22992048452295, "trajectory_balance_rmse": 2736.898661136352} {"canonical_length": 81, "canonical_nll": 1.448708051516686, "generation_rouge_l_f1": 0.18562874251497005, "generation_token_f1": 0.31736526946107785, "prompt_id": "8ddcb00e61289cea", "reading_canonical_log_prob": -117.34535217285156, "reading_distances": [17, 20, 20, 22], "reading_log_probs": [-119.83773803710938, -120.3514404296875, -119.38687133789062, -121.00566101074219], "reading_logprob_std": 0.6958264335426653, "tokenization_path_logprob_std": 75.16831872636531, "trajectory_balance_rmse": 2442.58454112619} {"canonical_length": 78, "canonical_nll": 0.9564993442633213, "generation_rouge_l_f1": 0.33540372670807456, "generation_token_f1": 0.4037267080745342, "prompt_id": "e509fb17129966c3", "reading_canonical_log_prob": -74.60694885253906, "reading_distances": [48, 40, 43, 52], "reading_log_probs": [-91.395263671875, -85.04899597167969, -88.80435180664062, -97.43986511230469], "reading_logprob_std": 5.210034949892118, "tokenization_path_logprob_std": 78.27145467277307, "trajectory_balance_rmse": 1758.36206479983} {"canonical_length": 170, "canonical_nll": 1.0870937571806065, "generation_rouge_l_f1": 0.2620320855614974, "generation_token_f1": 0.5668449197860964, "prompt_id": "07a3f091145dd3c4", "reading_canonical_log_prob": -184.80593872070312, "reading_distances": [27, 25, 30, 29], "reading_log_probs": [-183.99432373046875, -184.58172607421875, -186.45953369140625, -187.78977966308594], "reading_logprob_std": 1.7420082046967265, "tokenization_path_logprob_std": 215.20898431494547, "trajectory_balance_rmse": 5169.477474228167} {"canonical_length": 159, "canonical_nll": 1.7035493670769457, "generation_rouge_l_f1": 0.298932384341637, "generation_token_f1": 0.5480427046263346, "prompt_id": "1cfb5bc2275fa797", "reading_canonical_log_prob": -270.8643493652344, "reading_distances": [24, 27, 18, 22], "reading_log_probs": [-263.1465148925781, -265.2225341796875, -265.60711669921875, -264.2099609375], "reading_logprob_std": 1.1038035502064447, "tokenization_path_logprob_std": 143.44578232383628, "trajectory_balance_rmse": 4239.761866223266} {"canonical_length": 11, "canonical_nll": 0.033797537738626655, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "c9f832251ee1421d", "reading_canonical_log_prob": -0.3717729151248932, "reading_distances": [20, 22, 20, 18], "reading_log_probs": [-1.638607144355774, -1.9136884212493896, -1.34243905544281, -1.1307399272918701], "reading_logprob_std": 0.342231045121646, "tokenization_path_logprob_std": 45.01361713778783, "trajectory_balance_rmse": 187.4480293765562} {"canonical_length": 174, "canonical_nll": 1.19158856622104, "generation_rouge_l_f1": 0.33250620347394544, "generation_token_f1": 0.5508684863523573, "prompt_id": "13d2c1dd5e0685fa", "reading_canonical_log_prob": -207.33641052246094, "reading_distances": [54, 61, 58, 50], "reading_log_probs": [-204.33302307128906, -209.4002227783203, -203.69752502441406, -204.29090881347656], "reading_logprob_std": 2.662393963993489, "tokenization_path_logprob_std": 92.85962399922141, "trajectory_balance_rmse": 2571.8867526660906} {"canonical_length": 266, "canonical_nll": 1.2354518345424108, "generation_rouge_l_f1": 0.22137404580152673, "generation_token_f1": 0.5114503816793893, "prompt_id": "e2b16f584507a85d", "reading_canonical_log_prob": -328.63018798828125, "reading_distances": [32, 28, 27, 23], "reading_log_probs": [-330.61798095703125, -332.32177734375, -332.7574462890625, -333.678955078125], "reading_logprob_std": 1.2822738229826183, "tokenization_path_logprob_std": 688.0049261625567, "trajectory_balance_rmse": 8225.578657253247} {"canonical_length": 279, "canonical_nll": 1.301840614674339, "generation_rouge_l_f1": 0.2983425414364641, "generation_token_f1": 0.5524861878453038, "prompt_id": "76e96744ab3b125b", "reading_canonical_log_prob": -363.2135314941406, "reading_distances": [18, 25, 16, 24], "reading_log_probs": [-366.93804931640625, -369.25714111328125, -369.2007141113281, -369.0142517089844], "reading_logprob_std": 1.1145029762266156, "tokenization_path_logprob_std": 327.17455271766124, "trajectory_balance_rmse": 9190.726573147773} {"canonical_length": 157, "canonical_nll": 1.6815428521223128, "generation_rouge_l_f1": 0.2796208530805687, "generation_token_f1": 0.4691943127962085, "prompt_id": "2fad87f3c29a8aa9", "reading_canonical_log_prob": -264.0022277832031, "reading_distances": [27, 26, 28, 22], "reading_log_probs": [-261.20428466796875, -267.6319580078125, -262.70263671875, -262.21124267578125], "reading_logprob_std": 2.8649874788630525, "tokenization_path_logprob_std": 85.24212876520262, "trajectory_balance_rmse": 4834.155019568504} {"canonical_length": 135, "canonical_nll": 2.2026502821180554, "generation_rouge_l_f1": 0.13895781637717122, "generation_token_f1": 0.24317617866004965, "prompt_id": "057edeee1c342070", "reading_canonical_log_prob": -297.3577880859375, "reading_distances": [19, 19, 23, 21], "reading_log_probs": [-293.88690185546875, -296.2735290527344, -295.791015625, -297.67236328125], "reading_logprob_std": 1.5647252752110763, "tokenization_path_logprob_std": 49.756705257773774, "trajectory_balance_rmse": 4300.464728304849} {"canonical_length": 273, "canonical_nll": 1.0636028806805173, "generation_rouge_l_f1": 0.25047438330170774, "generation_token_f1": 0.5047438330170778, "prompt_id": "b42b291887a21049", "reading_canonical_log_prob": -290.36358642578125, "reading_distances": [31, 28, 27, 29], "reading_log_probs": [-290.55462646484375, -290.4910888671875, -290.0401611328125, -293.2672424316406], "reading_logprob_std": 1.4705842130425546, "tokenization_path_logprob_std": 157.05105830554254, "trajectory_balance_rmse": 8128.632880276692} {"canonical_length": 8, "canonical_nll": 1.0562931299209595, "generation_rouge_l_f1": 0.6666666666666665, "generation_token_f1": 0.6666666666666665, "prompt_id": "751d7c04030e09ea", "reading_canonical_log_prob": -8.450345039367676, "reading_distances": [22, 26, 26, 24], "reading_log_probs": [-3.2598764896392822, -1.3489962816238403, -0.6681495308876038, -5.170373439788818], "reading_logprob_std": 2.028032835932621, "tokenization_path_logprob_std": 94.07737657813318, "trajectory_balance_rmse": 291.95278925398696} {"canonical_length": 45, "canonical_nll": 1.3501882765028212, "generation_rouge_l_f1": 0.5542168674698795, "generation_token_f1": 0.674698795180723, "prompt_id": "bdda8d0e92726300", "reading_canonical_log_prob": -60.75847244262695, "reading_distances": [18, 23, 23, 24], "reading_log_probs": [-63.40357971191406, -64.14026641845703, -62.67647171020508, -63.24702453613281], "reading_logprob_std": 0.6029102619951631, "tokenization_path_logprob_std": 62.43026656036936, "trajectory_balance_rmse": 1370.9467298806683} {"canonical_length": 225, "canonical_nll": 0.8156623670789931, "generation_rouge_l_f1": 0.2744282744282745, "generation_token_f1": 0.5072765072765072, "prompt_id": "ce2f11428dfe632d", "reading_canonical_log_prob": -183.52403259277344, "reading_distances": [35, 33, 26, 26], "reading_log_probs": [-184.2701873779297, -184.76788330078125, -181.798828125, -185.43026733398438], "reading_logprob_std": 1.5848894616010563, "tokenization_path_logprob_std": 116.39648659095322, "trajectory_balance_rmse": 4678.478194122184} {"canonical_length": 371, "canonical_nll": 1.1681454393741577, "generation_rouge_l_f1": 0.2311733800350263, "generation_token_f1": 0.4238178633975481, "prompt_id": "3ce75f08c1522ce9", "reading_canonical_log_prob": -433.3819580078125, "reading_distances": [20, 17, 24, 20], "reading_log_probs": [-452.97698974609375, -460.1355895996094, -449.042236328125, -456.17327880859375], "reading_logprob_std": 4.713086926870308, "tokenization_path_logprob_std": 359.51308702283205, "trajectory_balance_rmse": 7842.235094887718} {"canonical_length": 3, "canonical_nll": 1.9780683517456055, "generation_rouge_l_f1": 0.01886792452830189, "generation_token_f1": 0.01886792452830189, "prompt_id": "cc1031d1c53d6dff", "reading_canonical_log_prob": -5.934205055236816, "reading_distances": [40, 29, 40, 40], "reading_log_probs": [-10.01876449584961, -7.515646934509277, -8.447148323059082, -7.895782470703125], "reading_logprob_std": 1.1014698424322986, "tokenization_path_logprob_std": 14.57411379824016, "trajectory_balance_rmse": 39.83504258351471} {"canonical_length": 14, "canonical_nll": 3.9511097499302457, "generation_rouge_l_f1": 0.07407407407407408, "generation_token_f1": 0.07407407407407408, "prompt_id": "6b9afce682e6088d", "reading_canonical_log_prob": -55.31553649902344, "reading_distances": [55, 50, 45, 53], "reading_log_probs": [-57.42233657836914, -56.47084045410156, -54.98879623413086, -57.309329986572266], "reading_logprob_std": 1.1226690902173133, "tokenization_path_logprob_std": 101.46653926806903, "trajectory_balance_rmse": 456.2586712944977} {"canonical_length": 15, "canonical_nll": 0.5675822575887044, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "16fc84421f512267", "reading_canonical_log_prob": -8.513733863830566, "reading_distances": [24, 30, 36, 24], "reading_log_probs": [-7.31300687789917, -6.02098274230957, -6.202228546142578, -7.080650329589844], "reading_logprob_std": 0.6379991314885923, "tokenization_path_logprob_std": 17.073572900417957, "trajectory_balance_rmse": 393.03696665908086} {"canonical_length": 12, "canonical_nll": 0.35519838333129883, "generation_rouge_l_f1": 0.7272727272727274, "generation_token_f1": 0.7272727272727274, "prompt_id": "7550a792cc616b59", "reading_canonical_log_prob": -4.262380599975586, "reading_distances": [24, 37, 26, 31], "reading_log_probs": [-6.759878158569336, -10.392529487609863, -9.364851951599121, -8.135542869567871], "reading_logprob_std": 1.568858610291912, "tokenization_path_logprob_std": 37.81137968929937, "trajectory_balance_rmse": 250.76689025241475} {"canonical_length": 242, "canonical_nll": 1.5934095146242253, "generation_rouge_l_f1": 0.2603550295857988, "generation_token_f1": 0.504930966469428, "prompt_id": "d8ba10df8798bb3c", "reading_canonical_log_prob": -385.6051025390625, "reading_distances": [33, 38, 35, 38], "reading_log_probs": [-393.6790771484375, -392.6826171875, -392.9651794433594, -393.59375], "reading_logprob_std": 0.484335324994803, "tokenization_path_logprob_std": 762.4880412089882, "trajectory_balance_rmse": 7383.048463163118} {"canonical_length": 15, "canonical_nll": 1.401537831624349, "generation_rouge_l_f1": 0.3, "generation_token_f1": 0.3, "prompt_id": "c62abdde12be0560", "reading_canonical_log_prob": -21.023067474365234, "reading_distances": [25, 28, 29, 29], "reading_log_probs": [-21.83765411376953, -21.87765121459961, -23.257543563842773, -20.67343521118164], "reading_logprob_std": 1.056919522650923, "tokenization_path_logprob_std": 84.27059305583474, "trajectory_balance_rmse": 375.3301949918058} {"canonical_length": 318, "canonical_nll": 1.6225196310559158, "generation_rouge_l_f1": 0.25177304964539005, "generation_token_f1": 0.5070921985815603, "prompt_id": "e65b179e41f3017d", "reading_canonical_log_prob": -515.9612426757812, "reading_distances": [21, 27, 27, 30], "reading_log_probs": [-521.927490234375, -521.3223876953125, -521.1786499023438, -520.1156005859375], "reading_logprob_std": 0.7537088603125942, "tokenization_path_logprob_std": 445.8155600312949, "trajectory_balance_rmse": 9449.171532697581} {"canonical_length": 83, "canonical_nll": 1.3788468694112388, "generation_rouge_l_f1": 0.1564245810055866, "generation_token_f1": 0.3240223463687151, "prompt_id": "688f5dc10c3a0a47", "reading_canonical_log_prob": -114.44429016113281, "reading_distances": [33, 34, 24, 33], "reading_log_probs": [-124.65873718261719, -122.553955078125, -118.24876403808594, -122.29601287841797], "reading_logprob_std": 2.678335254155825, "tokenization_path_logprob_std": 47.391094998851514, "trajectory_balance_rmse": 2368.461751966546} {"canonical_length": 324, "canonical_nll": 1.6970640582802854, "generation_rouge_l_f1": 0.3117338003502627, "generation_token_f1": 0.5499124343257442, "prompt_id": "d7836a3fff052985", "reading_canonical_log_prob": -549.8487548828125, "reading_distances": [22, 21, 22, 22], "reading_log_probs": [-553.1571044921875, -549.7899169921875, -553.2401733398438, -551.6590576171875], "reading_logprob_std": 1.619847703805898, "tokenization_path_logprob_std": 403.7004706189712, "trajectory_balance_rmse": 7812.388398082543} {"canonical_length": 18, "canonical_nll": 1.4533223046196833, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "97d1bc52cae659b0", "reading_canonical_log_prob": -26.159801483154297, "reading_distances": [9, 12, 13, 10], "reading_log_probs": [-28.08857536315918, -27.316295623779297, -28.68767547607422, -27.116220474243164], "reading_logprob_std": 0.7240507900389618, "tokenization_path_logprob_std": 37.70947877434971, "trajectory_balance_rmse": 300.40825184515535} {"canonical_length": 311, "canonical_nll": 1.077076804599578, "generation_rouge_l_f1": 0.2916666666666667, "generation_token_f1": 0.5277777777777778, "prompt_id": "d655055a06f3b252", "reading_canonical_log_prob": -334.97088623046875, "reading_distances": [30, 34, 35, 31], "reading_log_probs": [-335.4203186035156, -331.39630126953125, -332.2687072753906, -336.1236572265625], "reading_logprob_std": 2.3200158116562557, "tokenization_path_logprob_std": 588.437895383751, "trajectory_balance_rmse": 8231.213197083263} {"canonical_length": 270, "canonical_nll": 1.529180117006655, "generation_rouge_l_f1": 0.30550621669627, "generation_token_f1": 0.5008880994671403, "prompt_id": "62d551bf467199ab", "reading_canonical_log_prob": -412.8786315917969, "reading_distances": [18, 19, 24, 20], "reading_log_probs": [-412.86370849609375, -411.7811279296875, -411.08624267578125, -405.71783447265625], "reading_logprob_std": 3.181470842780278, "tokenization_path_logprob_std": 435.5679315880736, "trajectory_balance_rmse": 7108.952652961876} {"canonical_length": 11, "canonical_nll": 1.2443518205122515, "generation_rouge_l_f1": 0.7692307692307693, "generation_token_f1": 0.7692307692307693, "prompt_id": "1237b2c3cf7851d2", "reading_canonical_log_prob": -13.687870025634766, "reading_distances": [20, 20, 16, 18], "reading_log_probs": [-13.790509223937988, -14.411892890930176, -14.921001434326172, -13.681191444396973], "reading_logprob_std": 0.5778065997322365, "tokenization_path_logprob_std": 84.6285884653327, "trajectory_balance_rmse": 398.6332893366598} {"canonical_length": 3, "canonical_nll": 3.45923646291097, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "9af597ad9c807de8", "reading_canonical_log_prob": -10.37770938873291, "reading_distances": [17, 19, 21, 18], "reading_log_probs": [-10.09843921661377, -10.549546241760254, -11.391157150268555, -15.875813484191895], "reading_logprob_std": 2.6527068679043415, "tokenization_path_logprob_std": 18.515787911008587, "trajectory_balance_rmse": 49.69505947350739} {"canonical_length": 158, "canonical_nll": 1.4655736609350276, "generation_rouge_l_f1": 0.37463976945244953, "generation_token_f1": 0.6340057636887608, "prompt_id": "0366eeefae975865", "reading_canonical_log_prob": -231.56063842773438, "reading_distances": [31, 34, 31, 32], "reading_log_probs": [-226.90957641601562, -226.26736450195312, -231.07704162597656, -223.51271057128906], "reading_logprob_std": 3.1259547422328366, "tokenization_path_logprob_std": 239.44613125314774, "trajectory_balance_rmse": 3942.438743033398} {"canonical_length": 94, "canonical_nll": 1.8635477918259642, "generation_rouge_l_f1": 0.1875, "generation_token_f1": 0.3295454545454546, "prompt_id": "e99bd769bc8f61be", "reading_canonical_log_prob": -175.17349243164062, "reading_distances": [18, 16, 21, 22], "reading_log_probs": [-173.23666381835938, -176.35186767578125, -174.51858520507812, -176.22964477539062], "reading_logprob_std": 1.489108805966956, "tokenization_path_logprob_std": 404.34358877382147, "trajectory_balance_rmse": 3111.7796009709664} {"canonical_length": 287, "canonical_nll": 0.32638767787388395, "generation_rouge_l_f1": 0.3341772151898734, "generation_token_f1": 0.43544303797468353, "prompt_id": "e1ea57924db21e38", "reading_canonical_log_prob": -93.67326354980469, "reading_distances": [20, 19, 16, 21], "reading_log_probs": [-93.98686981201172, -94.23468780517578, -93.37142944335938, -94.14681243896484], "reading_logprob_std": 0.389435287534637, "tokenization_path_logprob_std": 123.10045262290512, "trajectory_balance_rmse": 3135.2971580377794} {"canonical_length": 248, "canonical_nll": 0.9325714726601878, "generation_rouge_l_f1": 0.38640429338103754, "generation_token_f1": 0.6189624329159212, "prompt_id": "d13dad8f3ef63531", "reading_canonical_log_prob": -231.27772521972656, "reading_distances": [29, 34, 32, 32], "reading_log_probs": [-232.5150146484375, -231.14028930664062, -233.3551788330078, -233.335693359375], "reading_logprob_std": 1.0406397640176108, "tokenization_path_logprob_std": 283.220163076113, "trajectory_balance_rmse": 4837.442060534594} {"canonical_length": 21, "canonical_nll": 1.4631824493408203, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.37499999999999994, "prompt_id": "146de30da91fe96b", "reading_canonical_log_prob": -30.726831436157227, "reading_distances": [19, 19, 18, 14], "reading_log_probs": [-29.286487579345703, -29.323150634765625, -27.752758026123047, -29.93866539001465], "reading_logprob_std": 0.9310472947848509, "tokenization_path_logprob_std": 78.41999555972434, "trajectory_balance_rmse": 665.2633376996476} {"canonical_length": 141, "canonical_nll": 1.235693207869293, "generation_rouge_l_f1": 0.18965517241379312, "generation_token_f1": 0.4080459770114942, "prompt_id": "e8d40d5d31440017", "reading_canonical_log_prob": -174.2327423095703, "reading_distances": [76, 71, 76, 68], "reading_log_probs": [-166.98033142089844, -171.63180541992188, -167.28701782226562, -170.13865661621094], "reading_logprob_std": 2.253586056740553, "tokenization_path_logprob_std": 153.6979758865932, "trajectory_balance_rmse": 3560.034177444792} {"canonical_length": 7, "canonical_nll": 0.046217156308037896, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "e77edca13c4c71bd", "reading_canonical_log_prob": -0.32352009415626526, "reading_distances": [13, 20, 17, 13], "reading_log_probs": [-0.9402287602424622, -0.8655890226364136, -0.47762468457221985, -0.8768082857131958], "reading_logprob_std": 0.21086834849015554, "tokenization_path_logprob_std": 39.63667749306926, "trajectory_balance_rmse": 280.6873544638189} {"canonical_length": 255, "canonical_nll": 1.4949201995251225, "generation_rouge_l_f1": 0.16538461538461538, "generation_token_f1": 0.35769230769230764, "prompt_id": "5dc83250d862853a", "reading_canonical_log_prob": -381.20465087890625, "reading_distances": [31, 29, 31, 27], "reading_log_probs": [-397.6844177246094, -392.69061279296875, -395.5118408203125, -388.09619140625], "reading_logprob_std": 4.139768046297359, "tokenization_path_logprob_std": 431.90669573823965, "trajectory_balance_rmse": 7448.528542825511} {"canonical_length": 25, "canonical_nll": 1.989072265625, "generation_rouge_l_f1": 0.09090909090909091, "generation_token_f1": 0.11188811188811189, "prompt_id": "a9f2944f6ac15039", "reading_canonical_log_prob": -49.726806640625, "reading_distances": [26, 27, 34, 28], "reading_log_probs": [-45.24297332763672, -48.05827331542969, -44.95805740356445, -48.314064025878906], "reading_logprob_std": 1.7883474032991022, "tokenization_path_logprob_std": 110.78227532638614, "trajectory_balance_rmse": 660.7124869283434} {"canonical_length": 51, "canonical_nll": 1.4262216605392157, "generation_rouge_l_f1": 0.17910447761194026, "generation_token_f1": 0.20895522388059706, "prompt_id": "0e9e30fa66e7e5e1", "reading_canonical_log_prob": -72.7373046875, "reading_distances": [14, 18, 21, 19], "reading_log_probs": [-83.3753433227539, -89.60940551757812, -80.9305419921875, -80.36357116699219], "reading_logprob_std": 4.233211323968983, "tokenization_path_logprob_std": 93.86604220656938, "trajectory_balance_rmse": 1197.5161243811826} {"canonical_length": 102, "canonical_nll": 2.462154313629749, "generation_rouge_l_f1": 0.13978494623655913, "generation_token_f1": 0.24193548387096772, "prompt_id": "f6e26ca6d4af6e82", "reading_canonical_log_prob": -251.13973999023438, "reading_distances": [27, 34, 25, 22], "reading_log_probs": [-246.57640075683594, -260.0491943359375, -258.22357177734375, -248.8182373046875], "reading_logprob_std": 6.708987757851941, "tokenization_path_logprob_std": 89.6961965126112, "trajectory_balance_rmse": 2698.3306722153184} {"canonical_length": 92, "canonical_nll": 1.9741219230320142, "generation_rouge_l_f1": 0.24431818181818185, "generation_token_f1": 0.3465909090909091, "prompt_id": "e5d1110bf220108e", "reading_canonical_log_prob": -181.6192169189453, "reading_distances": [30, 29, 24, 27], "reading_log_probs": [-182.21124267578125, -182.10012817382812, -187.46768188476562, -188.12582397460938], "reading_logprob_std": 3.2682508734861475, "tokenization_path_logprob_std": 202.15256307389203, "trajectory_balance_rmse": 2598.918281821177} {"canonical_length": 10, "canonical_nll": 1.242718505859375, "generation_rouge_l_f1": 0.12658227848101267, "generation_token_f1": 0.12658227848101267, "prompt_id": "dd04d981a73d4900", "reading_canonical_log_prob": -12.42718505859375, "reading_distances": [20, 14, 16, 15], "reading_log_probs": [-18.266441345214844, -18.302902221679688, -15.374570846557617, -15.734891891479492], "reading_logprob_std": 1.5830513775400437, "tokenization_path_logprob_std": 33.057025598739116, "trajectory_balance_rmse": 161.4372807267211} {"canonical_length": 299, "canonical_nll": 1.7134895452288879, "generation_rouge_l_f1": 0.2939068100358423, "generation_token_f1": 0.5412186379928315, "prompt_id": "d9940a473d55b366", "reading_canonical_log_prob": -512.3333740234375, "reading_distances": [26, 27, 16, 26], "reading_log_probs": [-514.93212890625, -522.3214111328125, -513.914306640625, -518.302001953125], "reading_logprob_std": 3.7978110426917735, "tokenization_path_logprob_std": 382.1623422243813, "trajectory_balance_rmse": 7910.688856616509} {"canonical_length": 416, "canonical_nll": 1.2975017841045673, "generation_rouge_l_f1": 0.14968152866242038, "generation_token_f1": 0.27070063694267515, "prompt_id": "7c0396dbe5594b7c", "reading_canonical_log_prob": -539.7607421875, "reading_distances": [25, 21, 23, 25], "reading_log_probs": [-541.1663818359375, -538.5032958984375, -538.136962890625, -537.5158081054688], "reading_logprob_std": 1.609639185742021, "tokenization_path_logprob_std": 69.35483086007167, "trajectory_balance_rmse": 7776.179530601982} {"canonical_length": 138, "canonical_nll": 1.5768516651098279, "generation_rouge_l_f1": 0.22982885085574573, "generation_token_f1": 0.45476772616136923, "prompt_id": "5ee117416302abcd", "reading_canonical_log_prob": -217.60552978515625, "reading_distances": [20, 17, 19, 22], "reading_log_probs": [-231.28086853027344, -232.32986450195312, -234.1114044189453, -236.18795776367188], "reading_logprob_std": 2.1517956183082263, "tokenization_path_logprob_std": 92.7660631544592, "trajectory_balance_rmse": 3444.6681483578136} {"canonical_length": 326, "canonical_nll": 1.6473867966353528, "generation_rouge_l_f1": 0.2175438596491228, "generation_token_f1": 0.48771929824561405, "prompt_id": "7d6d0adc840aad0a", "reading_canonical_log_prob": -537.048095703125, "reading_distances": [26, 24, 27, 19], "reading_log_probs": [-544.4224853515625, -545.9564208984375, -544.2930908203125, -542.2510375976562], "reading_logprob_std": 1.520728999342455, "tokenization_path_logprob_std": 479.06910588005974, "trajectory_balance_rmse": 9415.158511935277} {"canonical_length": 17, "canonical_nll": 1.903950859518612, "generation_rouge_l_f1": 0.21428571428571427, "generation_token_f1": 0.3571428571428571, "prompt_id": "305ea9936db72d06", "reading_canonical_log_prob": -32.367164611816406, "reading_distances": [62, 54, 57, 53], "reading_log_probs": [-33.43803024291992, -30.84552574157715, -31.563169479370117, -38.755008697509766], "reading_logprob_std": 3.574258977547039, "tokenization_path_logprob_std": 40.021800700952916, "trajectory_balance_rmse": 479.5210949780753} {"canonical_length": 8, "canonical_nll": 0.9098114371299744, "generation_rouge_l_f1": 0.5714285714285714, "generation_token_f1": 0.5714285714285714, "prompt_id": "f9a917228bf6b5fe", "reading_canonical_log_prob": -7.278491497039795, "reading_distances": [32, 38, 38, 37], "reading_log_probs": [-9.044631958007812, -7.657759666442871, -8.633465766906738, -7.814522743225098], "reading_logprob_std": 0.6616195912852774, "tokenization_path_logprob_std": 30.3096991242539, "trajectory_balance_rmse": 137.2446218282044} {"canonical_length": 206, "canonical_nll": 1.370093484526699, "generation_rouge_l_f1": 0.45714285714285713, "generation_token_f1": 0.6326530612244898, "prompt_id": "6fb0440b40107b60", "reading_canonical_log_prob": -282.2392578125, "reading_distances": [13, 11, 13, 12], "reading_log_probs": [-283.70440673828125, -285.9773864746094, -284.58477783203125, -285.8445129394531], "reading_logprob_std": 1.0826468893935581, "tokenization_path_logprob_std": 228.74281408678235, "trajectory_balance_rmse": 3946.3547847343452} {"canonical_length": 94, "canonical_nll": 2.2593932456158576, "generation_rouge_l_f1": 0.20809248554913296, "generation_token_f1": 0.3641618497109827, "prompt_id": "16cf88f25ee512b8", "reading_canonical_log_prob": -212.38296508789062, "reading_distances": [17, 17, 16, 22], "reading_log_probs": [-208.95144653320312, -210.1551055908203, -214.38121032714844, -208.61773681640625], "reading_logprob_std": 2.6533560990453395, "tokenization_path_logprob_std": 138.59992484355914, "trajectory_balance_rmse": 3140.6374615363593} {"canonical_length": 12, "canonical_nll": 1.7661177317301433, "generation_rouge_l_f1": 0.1946902654867257, "generation_token_f1": 0.1946902654867257, "prompt_id": "024090bc27e03da4", "reading_canonical_log_prob": -21.19341278076172, "reading_distances": [26, 19, 17, 29], "reading_log_probs": [-20.263742446899414, -19.22011375427246, -19.58370018005371, -18.142223358154297], "reading_logprob_std": 0.8862072650632296, "tokenization_path_logprob_std": 89.91280072746831, "trajectory_balance_rmse": 456.4228868554956} {"canonical_length": 58, "canonical_nll": 2.2339561725484915, "generation_rouge_l_f1": 0.21739130434782608, "generation_token_f1": 0.26956521739130435, "prompt_id": "d97dcaaab25ed5d3", "reading_canonical_log_prob": -129.5694580078125, "reading_distances": [11, 17, 10, 15], "reading_log_probs": [-138.54776000976562, -147.1885528564453, -145.19866943359375, -136.69961547851562], "reading_logprob_std": 5.070543173382126, "tokenization_path_logprob_std": 147.05369102943408, "trajectory_balance_rmse": 1976.2677505284373} {"canonical_length": 386, "canonical_nll": 0.9982244461929243, "generation_rouge_l_f1": 0.38525963149078724, "generation_token_f1": 0.5159128978224455, "prompt_id": "5a59813799d04978", "reading_canonical_log_prob": -385.31463623046875, "reading_distances": [9, 11, 12, 14], "reading_log_probs": [-386.7677307128906, -387.76983642578125, -387.33380126953125, -385.39007568359375], "reading_logprob_std": 1.0349735147827563, "tokenization_path_logprob_std": 405.42386240834753, "trajectory_balance_rmse": 7719.907717618157} {"canonical_length": 120, "canonical_nll": 1.2280291239420573, "generation_rouge_l_f1": 0.2828282828282828, "generation_token_f1": 0.4713804713804714, "prompt_id": "84eecbd6c7c38a00", "reading_canonical_log_prob": -147.36349487304688, "reading_distances": [20, 24, 27, 23], "reading_log_probs": [-148.3601531982422, -149.73117065429688, -149.46253967285156, -149.25662231445312], "reading_logprob_std": 0.5943036800925723, "tokenization_path_logprob_std": 240.64416298256157, "trajectory_balance_rmse": 2828.69003152111} {"canonical_length": 12, "canonical_nll": 0.030193301538626354, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1914d368729c6a5a", "reading_canonical_log_prob": -0.36231961846351624, "reading_distances": [23, 22, 23, 27], "reading_log_probs": [-0.21253114938735962, -0.1889391392469406, -0.19614076614379883, -0.23764505982398987], "reading_logprob_std": 0.021607623318839822, "tokenization_path_logprob_std": 38.82334108586749, "trajectory_balance_rmse": 366.16390890673955} {"canonical_length": 313, "canonical_nll": 1.2194477117861422, "generation_rouge_l_f1": 0.2045060658578856, "generation_token_f1": 0.4436741767764298, "prompt_id": "dc582c1cebaf17e1", "reading_canonical_log_prob": -381.6871337890625, "reading_distances": [43, 44, 34, 50], "reading_log_probs": [-384.13665771484375, -383.8995361328125, -384.318115234375, -381.261474609375], "reading_logprob_std": 1.438560053842586, "tokenization_path_logprob_std": 240.33883583931805, "trajectory_balance_rmse": 9352.022754535405} {"canonical_length": 34, "canonical_nll": 1.3949475007898666, "generation_rouge_l_f1": 0.07766990291262137, "generation_token_f1": 0.0970873786407767, "prompt_id": "c32c06c0fe76b404", "reading_canonical_log_prob": -47.42821502685547, "reading_distances": [19, 19, 20, 22], "reading_log_probs": [-51.00932312011719, -61.5677490234375, -55.61579132080078, -53.670108795166016], "reading_logprob_std": 4.484825241932591, "tokenization_path_logprob_std": 64.19281814413064, "trajectory_balance_rmse": 745.4137073249747} {"canonical_length": 124, "canonical_nll": 1.7316163586032005, "generation_rouge_l_f1": 0.3183673469387755, "generation_token_f1": 0.5306122448979592, "prompt_id": "3415a33aecea8511", "reading_canonical_log_prob": -214.72042846679688, "reading_distances": [22, 20, 19, 23], "reading_log_probs": [-217.1907958984375, -220.82308959960938, -216.5859375, -221.57952880859375], "reading_logprob_std": 2.5212761338886485, "tokenization_path_logprob_std": 146.11005360326047, "trajectory_balance_rmse": 4130.179152991742} {"canonical_length": 333, "canonical_nll": 1.5303051464550488, "generation_rouge_l_f1": 0.22635135135135134, "generation_token_f1": 0.4797297297297297, "prompt_id": "f7e2c3707f2870f0", "reading_canonical_log_prob": -509.59161376953125, "reading_distances": [58, 53, 59, 53], "reading_log_probs": [-505.70147705078125, -507.286376953125, -509.5386962890625, -504.86077880859375], "reading_logprob_std": 2.0571483786496096, "tokenization_path_logprob_std": 231.93870907256527, "trajectory_balance_rmse": 8896.286472064969} {"canonical_length": 39, "canonical_nll": 1.6601269061748798, "generation_rouge_l_f1": 0.0915032679738562, "generation_token_f1": 0.14379084967320263, "prompt_id": "ec9f4061e1384314", "reading_canonical_log_prob": -64.74494934082031, "reading_distances": [19, 18, 19, 16], "reading_log_probs": [-66.88594055175781, -66.23625183105469, -67.95659637451172, -67.08214569091797], "reading_logprob_std": 0.7098521882308153, "tokenization_path_logprob_std": 96.86026864575578, "trajectory_balance_rmse": 1159.066130898302} {"canonical_length": 14, "canonical_nll": 0.45946008818490164, "generation_rouge_l_f1": 0.2758620689655172, "generation_token_f1": 0.2758620689655172, "prompt_id": "a05868d3246d849c", "reading_canonical_log_prob": -6.432441234588623, "reading_distances": [10, 9, 8, 9], "reading_log_probs": [-6.487002372741699, -6.439847469329834, -6.884183883666992, -6.755086898803711], "reading_logprob_std": 0.21317497295890975, "tokenization_path_logprob_std": 114.1832133023861, "trajectory_balance_rmse": 494.4430726328018} {"canonical_length": 127, "canonical_nll": 1.5681709830216535, "generation_rouge_l_f1": 0.2074074074074074, "generation_token_f1": 0.4, "prompt_id": "420ee5406fba3002", "reading_canonical_log_prob": -199.15771484375, "reading_distances": [21, 24, 21, 25], "reading_log_probs": [-198.00442504882812, -197.20570373535156, -196.90994262695312, -194.38018798828125], "reading_logprob_std": 1.5663549560870558, "tokenization_path_logprob_std": 229.87917097358266, "trajectory_balance_rmse": 3856.145981639965} {"canonical_length": 9, "canonical_nll": 1.8325962490505643, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "0a4e72c4105acd5f", "reading_canonical_log_prob": -16.493366241455078, "reading_distances": [16, 8, 15, 16], "reading_log_probs": [-15.511476516723633, -15.638402938842773, -16.39031982421875, -17.123790740966797], "reading_logprob_std": 0.7470926172160833, "tokenization_path_logprob_std": 28.974607298037874, "trajectory_balance_rmse": 157.68765432008863} {"canonical_length": 16, "canonical_nll": 1.16411292552948, "generation_rouge_l_f1": 0.45454545454545453, "generation_token_f1": 0.45454545454545453, "prompt_id": "0ce850d0751da7a2", "reading_canonical_log_prob": -18.62580680847168, "reading_distances": [61, 60, 73, 64], "reading_log_probs": [-27.251842498779297, -23.92989158630371, -25.698495864868164, -29.655719757080078], "reading_logprob_std": 2.428984008652274, "tokenization_path_logprob_std": 32.99756808216646, "trajectory_balance_rmse": 459.29601721432914} {"canonical_length": 9, "canonical_nll": 5.749874114990234, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.22222222222222224, "prompt_id": "eeee24292d24cae3", "reading_canonical_log_prob": -51.74886703491211, "reading_distances": [19, 19, 26, 23], "reading_log_probs": [-56.227169036865234, -55.50053787231445, -54.933494567871094, -54.61928176879883], "reading_logprob_std": 0.7061496571804121, "tokenization_path_logprob_std": 43.951500395367965, "trajectory_balance_rmse": 230.53702558125656} {"canonical_length": 20, "canonical_nll": 3.088736915588379, "generation_rouge_l_f1": 0.09150326797385622, "generation_token_f1": 0.09803921568627451, "prompt_id": "aa417b16bf366ab6", "reading_canonical_log_prob": -61.77473831176758, "reading_distances": [31, 33, 27, 38], "reading_log_probs": [-54.238731384277344, -52.94245910644531, -53.6170654296875, -55.679386138916016], "reading_logprob_std": 1.166951387147118, "tokenization_path_logprob_std": 65.49383371883152, "trajectory_balance_rmse": 554.9019486792471} {"canonical_length": 277, "canonical_nll": 1.4328509719793547, "generation_rouge_l_f1": 0.25384615384615383, "generation_token_f1": 0.4961538461538461, "prompt_id": "31ffef5f718912de", "reading_canonical_log_prob": -396.89971923828125, "reading_distances": [15, 18, 20, 15], "reading_log_probs": [-398.48883056640625, -402.97412109375, -408.0948791503906, -406.6878662109375], "reading_logprob_std": 4.29740170815182, "tokenization_path_logprob_std": 288.2339165836029, "trajectory_balance_rmse": 8377.799954632375} {"canonical_length": 170, "canonical_nll": 1.1454381606158088, "generation_rouge_l_f1": 0.14163090128755365, "generation_token_f1": 0.27467811158798283, "prompt_id": "fe5c1f86ece4d243", "reading_canonical_log_prob": -194.7244873046875, "reading_distances": [18, 14, 18, 19], "reading_log_probs": [-194.9680633544922, -186.5213623046875, -196.51708984375, -182.9265594482422], "reading_logprob_std": 6.5592412611046305, "tokenization_path_logprob_std": 109.63686202605942, "trajectory_balance_rmse": 4349.333809984887} {"canonical_length": 166, "canonical_nll": 1.7529083619634789, "generation_rouge_l_f1": 0.4031413612565445, "generation_token_f1": 0.5811518324607329, "prompt_id": "8db7426fa0df6c62", "reading_canonical_log_prob": -290.9827880859375, "reading_distances": [18, 17, 24, 22], "reading_log_probs": [-295.71417236328125, -294.7696228027344, -293.27862548828125, -294.1065368652344], "reading_logprob_std": 1.0310491237705848, "tokenization_path_logprob_std": 171.00203070998384, "trajectory_balance_rmse": 2988.52167876416} {"canonical_length": 39, "canonical_nll": 0.6707228636130308, "generation_rouge_l_f1": 0.3211678832116789, "generation_token_f1": 0.43795620437956206, "prompt_id": "6c3aa6c41d3147aa", "reading_canonical_log_prob": -26.158191680908203, "reading_distances": [49, 49, 41, 43], "reading_log_probs": [-25.761075973510742, -28.907930374145508, -25.078371047973633, -25.392520904541016], "reading_logprob_std": 1.7707561938582088, "tokenization_path_logprob_std": 116.29067315211648, "trajectory_balance_rmse": 994.9699783153987} {"canonical_length": 8, "canonical_nll": 2.2146124839782715, "generation_rouge_l_f1": 0.42857142857142855, "generation_token_f1": 0.42857142857142855, "prompt_id": "e07ed69174530f07", "reading_canonical_log_prob": -17.716899871826172, "reading_distances": [35, 38, 33, 38], "reading_log_probs": [-14.806170463562012, -15.911725997924805, -19.71355438232422, -14.665568351745605], "reading_logprob_std": 2.359615255149416, "tokenization_path_logprob_std": 61.73987554591213, "trajectory_balance_rmse": 158.05529244591256} {"canonical_length": 57, "canonical_nll": 2.23722103185821, "generation_rouge_l_f1": 0.3030303030303031, "generation_token_f1": 0.40909090909090906, "prompt_id": "93c96f6a3b9968fc", "reading_canonical_log_prob": -127.52159881591797, "reading_distances": [25, 18, 29, 22], "reading_log_probs": [-133.04751586914062, -132.59100341796875, -133.33009338378906, -131.00648498535156], "reading_logprob_std": 1.037230662574217, "tokenization_path_logprob_std": 170.15252558962595, "trajectory_balance_rmse": 1672.029230201861} {"canonical_length": 264, "canonical_nll": 1.322316025242661, "generation_rouge_l_f1": 0.30952380952380953, "generation_token_f1": 0.5555555555555556, "prompt_id": "0a9cdb872ed6cdad", "reading_canonical_log_prob": -349.0914306640625, "reading_distances": [23, 20, 21, 19], "reading_log_probs": [-348.60552978515625, -347.07958984375, -348.67816162109375, -349.1464538574219], "reading_logprob_std": 0.8978225111004194, "tokenization_path_logprob_std": 359.2827697708878, "trajectory_balance_rmse": 6587.972950675301} {"canonical_length": 141, "canonical_nll": 1.6367739413646942, "generation_rouge_l_f1": 0.30985915492957744, "generation_token_f1": 0.5633802816901408, "prompt_id": "c6291fd5f11c978c", "reading_canonical_log_prob": -230.78512573242188, "reading_distances": [23, 17, 21, 19], "reading_log_probs": [-235.06077575683594, -235.7842254638672, -237.7539825439453, -234.85525512695312], "reading_logprob_std": 1.3217579986974848, "tokenization_path_logprob_std": 280.80733509565147, "trajectory_balance_rmse": 4067.498269602222} {"canonical_length": 9, "canonical_nll": 0.014259359902805753, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "c5b68b2a56c5748c", "reading_canonical_log_prob": -0.12833423912525177, "reading_distances": [48, 43, 38, 41], "reading_log_probs": [-0.13944309949874878, -0.16877374053001404, -0.09895531833171844, -0.3355766534805298], "reading_logprob_std": 0.10394525723479858, "tokenization_path_logprob_std": 61.436859254018636, "trajectory_balance_rmse": 154.4172109802258} {"canonical_length": 101, "canonical_nll": 2.256680932375464, "generation_rouge_l_f1": 0.26424870466321243, "generation_token_f1": 0.4196891191709845, "prompt_id": "4478e3a9a8e3ebd2", "reading_canonical_log_prob": -227.92477416992188, "reading_distances": [28, 21, 23, 19], "reading_log_probs": [-218.82540893554688, -219.7420654296875, -221.154296875, -220.4685821533203], "reading_logprob_std": 0.9981827350559014, "tokenization_path_logprob_std": 173.93555597768216, "trajectory_balance_rmse": 1671.2829680645095} {"canonical_length": 153, "canonical_nll": 1.7903051438674429, "generation_rouge_l_f1": 0.23529411764705885, "generation_token_f1": 0.4611764705882353, "prompt_id": "1679ec014dbdb2b7", "reading_canonical_log_prob": -273.91668701171875, "reading_distances": [25, 25, 20, 28], "reading_log_probs": [-278.3105773925781, -269.10748291015625, -270.1353454589844, -275.8572082519531], "reading_logprob_std": 4.4432101082357915, "tokenization_path_logprob_std": 203.24397411359104, "trajectory_balance_rmse": 4192.8824906574} {"canonical_length": 6, "canonical_nll": 0.3378804524739583, "generation_rouge_l_f1": 0.888888888888889, "generation_token_f1": 0.888888888888889, "prompt_id": "d608b8ecb5eb0aa8", "reading_canonical_log_prob": -2.02728271484375, "reading_distances": [6, 12, 17, 13], "reading_log_probs": [-2.0751194953918457, -1.9851750135421753, -2.067903995513916, -1.6767314672470093], "reading_logprob_std": 0.1874950103342237, "tokenization_path_logprob_std": 42.4615574379623, "trajectory_balance_rmse": 185.80401945716108} {"canonical_length": 8, "canonical_nll": 0.8439986109733582, "generation_rouge_l_f1": 0.7692307692307692, "generation_token_f1": 0.7692307692307692, "prompt_id": "c7c5da8841670598", "reading_canonical_log_prob": -6.751988887786865, "reading_distances": [23, 23, 28, 23], "reading_log_probs": [-7.23712682723999, -3.9626002311706543, -2.8545331954956055, -7.018877029418945], "reading_logprob_std": 2.1963546376859817, "tokenization_path_logprob_std": 54.43200319805313, "trajectory_balance_rmse": 310.35303869983375} {"canonical_length": 72, "canonical_nll": 2.0420843760172525, "generation_rouge_l_f1": 0.17575757575757575, "generation_token_f1": 0.26666666666666666, "prompt_id": "8b986ced31f57ea9", "reading_canonical_log_prob": -147.0300750732422, "reading_distances": [31, 31, 35, 33], "reading_log_probs": [-141.57711791992188, -137.79856872558594, -144.6361541748047, -138.57984924316406], "reading_logprob_std": 3.117966214683692, "tokenization_path_logprob_std": 181.48014579948995, "trajectory_balance_rmse": 1973.26325048122} {"canonical_length": 288, "canonical_nll": 1.3540495766533747, "generation_rouge_l_f1": 0.25047438330170774, "generation_token_f1": 0.47058823529411764, "prompt_id": "3d6847202e1bf0e3", "reading_canonical_log_prob": -389.9662780761719, "reading_distances": [24, 20, 25, 23], "reading_log_probs": [-401.9652099609375, -399.00830078125, -399.81768798828125, -400.78753662109375], "reading_logprob_std": 1.2748683334171855, "tokenization_path_logprob_std": 343.0701349647848, "trajectory_balance_rmse": 7424.08759026731} {"canonical_length": 6, "canonical_nll": 1.880492369333903, "generation_rouge_l_f1": 0.8000000000000002, "generation_token_f1": 0.8000000000000002, "prompt_id": "81930576a5952ad4", "reading_canonical_log_prob": -11.282954216003418, "reading_distances": [48, 43, 45, 39], "reading_log_probs": [-9.822592735290527, -11.69556999206543, -10.305477142333984, -9.603570938110352], "reading_logprob_std": 0.9394449688846799, "tokenization_path_logprob_std": 32.50023543408963, "trajectory_balance_rmse": 129.50289382325045} {"canonical_length": 271, "canonical_nll": 1.1479729796687614, "generation_rouge_l_f1": 0.2379182156133829, "generation_token_f1": 0.5353159851301116, "prompt_id": "52dc7c85e884efbf", "reading_canonical_log_prob": -311.1006774902344, "reading_distances": [16, 11, 11, 19], "reading_log_probs": [-306.4927978515625, -305.34478759765625, -304.58233642578125, -306.151611328125], "reading_logprob_std": 0.8553322301360664, "tokenization_path_logprob_std": 325.2681377057469, "trajectory_balance_rmse": 8006.799488756622} {"canonical_length": 220, "canonical_nll": 1.18441162109375, "generation_rouge_l_f1": 0.22559652928416488, "generation_token_f1": 0.351409978308026, "prompt_id": "02a68089e544097d", "reading_canonical_log_prob": -260.570556640625, "reading_distances": [27, 21, 23, 25], "reading_log_probs": [-262.1172180175781, -259.08868408203125, -258.9256591796875, -259.31365966796875], "reading_logprob_std": 1.5123309973376595, "tokenization_path_logprob_std": 470.3142503737304, "trajectory_balance_rmse": 5301.7076258153065} {"canonical_length": 116, "canonical_nll": 1.8532483331088363, "generation_rouge_l_f1": 0.28782287822878233, "generation_token_f1": 0.4575645756457565, "prompt_id": "4fcd7cfb63238642", "reading_canonical_log_prob": -214.976806640625, "reading_distances": [28, 26, 36, 35], "reading_log_probs": [-213.7278289794922, -213.98611450195312, -211.51878356933594, -214.16110229492188], "reading_logprob_std": 1.2326972359906723, "tokenization_path_logprob_std": 99.56666913615737, "trajectory_balance_rmse": 4189.34109440559} {"canonical_length": 260, "canonical_nll": 1.5980019202599158, "generation_rouge_l_f1": 0.1851851851851852, "generation_token_f1": 0.3777777777777778, "prompt_id": "0b49b7e13d1bd9b0", "reading_canonical_log_prob": -415.4804992675781, "reading_distances": [35, 29, 27, 24], "reading_log_probs": [-418.06427001953125, -413.67584228515625, -421.04364013671875, -415.338134765625], "reading_logprob_std": 3.229644081387602, "tokenization_path_logprob_std": 354.9348468629579, "trajectory_balance_rmse": 6497.215120693825} {"canonical_length": 169, "canonical_nll": 1.4186732726689626, "generation_rouge_l_f1": 0.27567567567567564, "generation_token_f1": 0.5621621621621621, "prompt_id": "112cdc2afc089b69", "reading_canonical_log_prob": -239.7557830810547, "reading_distances": [23, 27, 25, 31], "reading_log_probs": [-243.99996948242188, -243.2045440673828, -244.72802734375, -245.64669799804688], "reading_logprob_std": 1.040975405168911, "tokenization_path_logprob_std": 152.1434427479493, "trajectory_balance_rmse": 4195.389813197872} {"canonical_length": 250, "canonical_nll": 1.2015927734375, "generation_rouge_l_f1": 0.2151898734177215, "generation_token_f1": 0.46835443037974683, "prompt_id": "845212ac338f2658", "reading_canonical_log_prob": -300.398193359375, "reading_distances": [41, 42, 28, 40], "reading_log_probs": [-315.2860107421875, -311.7793884277344, -304.49810791015625, -308.577880859375], "reading_logprob_std": 4.596969092940049, "tokenization_path_logprob_std": 292.6636143641556, "trajectory_balance_rmse": 6940.995245164307} {"canonical_length": 30, "canonical_nll": 0.6731420516967773, "generation_rouge_l_f1": 0.4074074074074074, "generation_token_f1": 0.5925925925925927, "prompt_id": "9ca4c748ffa8cd48", "reading_canonical_log_prob": -20.19426155090332, "reading_distances": [46, 53, 45, 52], "reading_log_probs": [-19.689321517944336, -19.85922622680664, -20.061447143554688, -19.815296173095703], "reading_logprob_std": 0.15454992562204403, "tokenization_path_logprob_std": 167.61962352332, "trajectory_balance_rmse": 1003.07061687974} {"canonical_length": 501, "canonical_nll": 0.5862794065189932, "generation_rouge_l_f1": 0.2874251497005988, "generation_token_f1": 0.4383233532934132, "prompt_id": "585bee96b6b0ece7", "reading_canonical_log_prob": -293.7259826660156, "reading_distances": [19, 22, 22, 17], "reading_log_probs": [-300.4315185546875, -299.4825439453125, -302.54833984375, -302.06298828125], "reading_logprob_std": 1.4240945149089734, "tokenization_path_logprob_std": 180.1761421966659, "trajectory_balance_rmse": 10565.92107810645} {"canonical_length": 168, "canonical_nll": 1.227817626226516, "generation_rouge_l_f1": 0.15873015873015872, "generation_token_f1": 0.32653061224489793, "prompt_id": "b3cb4c0e99b9a63e", "reading_canonical_log_prob": -206.2733612060547, "reading_distances": [21, 25, 25, 26], "reading_log_probs": [-202.6831817626953, -203.96603393554688, -202.94552612304688, -204.87713623046875], "reading_logprob_std": 1.0054282529408827, "tokenization_path_logprob_std": 201.78726661423818, "trajectory_balance_rmse": 4828.695504357081} {"canonical_length": 430, "canonical_nll": 1.0637790413789971, "generation_rouge_l_f1": 0.21406727828746178, "generation_token_f1": 0.3792048929663608, "prompt_id": "2fa618fc412d053e", "reading_canonical_log_prob": -457.42498779296875, "reading_distances": [27, 29, 26, 25], "reading_log_probs": [-454.7247619628906, -454.9542541503906, -450.7821350097656, -450.75897216796875], "reading_logprob_std": 2.3510984703251117, "tokenization_path_logprob_std": 396.26145624102105, "trajectory_balance_rmse": 9396.558242550602} {"canonical_length": 160, "canonical_nll": 1.1620952606201171, "generation_rouge_l_f1": 0.2632794457274827, "generation_token_f1": 0.5034642032332564, "prompt_id": "72f894b5c3eed051", "reading_canonical_log_prob": -185.93524169921875, "reading_distances": [27, 26, 28, 25], "reading_log_probs": [-193.35302734375, -191.02127075195312, -190.9383544921875, -192.81553649902344], "reading_logprob_std": 1.2351351867865157, "tokenization_path_logprob_std": 346.4764640358277, "trajectory_balance_rmse": 4328.267053710259} {"canonical_length": 98, "canonical_nll": 1.5430593685227998, "generation_rouge_l_f1": 0.17983651226158037, "generation_token_f1": 0.34877384196185285, "prompt_id": "9c00b56f3aa02b5b", "reading_canonical_log_prob": -151.21981811523438, "reading_distances": [37, 31, 42, 34], "reading_log_probs": [-149.644287109375, -147.70925903320312, -149.60011291503906, -152.19537353515625], "reading_logprob_std": 1.841427353536816, "tokenization_path_logprob_std": 265.71080981025665, "trajectory_balance_rmse": 2884.203087611154} {"canonical_length": 10, "canonical_nll": 0.7949313163757324, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "363eae7c1280c1a6", "reading_canonical_log_prob": -7.949313163757324, "reading_distances": [65, 53, 54, 43], "reading_log_probs": [-4.689387321472168, -8.28824234008789, -6.792126655578613, -7.857433319091797], "reading_logprob_std": 1.6064614954556595, "tokenization_path_logprob_std": 82.8700703725971, "trajectory_balance_rmse": 295.74196371621866} {"canonical_length": 7, "canonical_nll": 1.6510639190673828, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "86c615edc3ae2a5f", "reading_canonical_log_prob": -11.55744743347168, "reading_distances": [42, 39, 44, 43], "reading_log_probs": [-11.488239288330078, -11.899846076965332, -9.091191291809082, -9.914527893066406], "reading_logprob_std": 1.3197139827165036, "tokenization_path_logprob_std": 36.43474625824124, "trajectory_balance_rmse": 171.11979965616595} {"canonical_length": 28, "canonical_nll": 0.7508647782461983, "generation_rouge_l_f1": 0.16058394160583941, "generation_token_f1": 0.17518248175182483, "prompt_id": "b424089c85e11874", "reading_canonical_log_prob": -21.024213790893555, "reading_distances": [23, 24, 20, 16], "reading_log_probs": [-20.145591735839844, -20.530120849609375, -19.695499420166016, -20.772098541259766], "reading_logprob_std": 0.4705642278588589, "tokenization_path_logprob_std": 103.42179388617075, "trajectory_balance_rmse": 766.9979637951636} {"canonical_length": 7, "canonical_nll": 0.6060974257332938, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.33333333333333337, "prompt_id": "69e0fa630dc3c290", "reading_canonical_log_prob": -4.242681980133057, "reading_distances": [41, 32, 25, 23], "reading_log_probs": [-6.8506693840026855, -7.127790451049805, -6.624690532684326, -6.463390350341797], "reading_logprob_std": 0.28844444113521905, "tokenization_path_logprob_std": 13.546436921344727, "trajectory_balance_rmse": 109.67604847227851} {"canonical_length": 158, "canonical_nll": 1.5365376532832278, "generation_rouge_l_f1": 0.19900497512437812, "generation_token_f1": 0.41791044776119407, "prompt_id": "edfb4a427bd0504a", "reading_canonical_log_prob": -242.77294921875, "reading_distances": [31, 25, 27, 23], "reading_log_probs": [-244.06300354003906, -239.8060302734375, -237.9270477294922, -239.08721923828125], "reading_logprob_std": 2.6758716205142594, "tokenization_path_logprob_std": 111.49603861063989, "trajectory_balance_rmse": 4079.007816228517} {"canonical_length": 9, "canonical_nll": 0.3239419725206163, "generation_rouge_l_f1": 0.6956521739130436, "generation_token_f1": 0.6956521739130436, "prompt_id": "9a5bfd989c2d5e52", "reading_canonical_log_prob": -2.915477752685547, "reading_distances": [37, 40, 36, 44], "reading_log_probs": [-2.9191198348999023, -4.516558647155762, -3.0115785598754883, -3.2249715328216553], "reading_logprob_std": 0.7434493669025701, "tokenization_path_logprob_std": 54.97264608741489, "trajectory_balance_rmse": 181.09570547377854} {"canonical_length": 272, "canonical_nll": 0.8457434037152458, "generation_rouge_l_f1": 0.469964664310954, "generation_token_f1": 0.6678445229681979, "prompt_id": "a2cdcfce41950902", "reading_canonical_log_prob": -230.04220581054688, "reading_distances": [29, 18, 27, 24], "reading_log_probs": [-242.84730529785156, -238.78744506835938, -246.27801513671875, -242.09310913085938], "reading_logprob_std": 3.0736858988951337, "tokenization_path_logprob_std": 196.7693328118316, "trajectory_balance_rmse": 5040.087954335611} {"canonical_length": 72, "canonical_nll": 1.6244210137261286, "generation_rouge_l_f1": 0.2059800664451827, "generation_token_f1": 0.29900332225913623, "prompt_id": "41dfa00e84da9369", "reading_canonical_log_prob": -116.95831298828125, "reading_distances": [44, 37, 39, 41], "reading_log_probs": [-123.80579376220703, -119.50511169433594, -126.79550170898438, -127.28697204589844], "reading_logprob_std": 3.5765614572256963, "tokenization_path_logprob_std": 72.50924996710084, "trajectory_balance_rmse": 1118.2841804319432} {"canonical_length": 13, "canonical_nll": 1.2618274688720703, "generation_rouge_l_f1": 0.08088235294117647, "generation_token_f1": 0.08823529411764706, "prompt_id": "398d73f7e39625a8", "reading_canonical_log_prob": -16.403757095336914, "reading_distances": [27, 27, 24, 31], "reading_log_probs": [-14.351842880249023, -13.996187210083008, -17.95219612121582, -14.957347869873047], "reading_logprob_std": 1.8027438132696054, "tokenization_path_logprob_std": 33.79908396523476, "trajectory_balance_rmse": 177.24752425335154} {"canonical_length": 89, "canonical_nll": 1.6327298625131672, "generation_rouge_l_f1": 0.264026402640264, "generation_token_f1": 0.33663366336633666, "prompt_id": "d769bfcaa2a347d7", "reading_canonical_log_prob": -145.31295776367188, "reading_distances": [67, 74, 63, 79], "reading_log_probs": [-147.1447296142578, -148.61251831054688, -148.1628875732422, -145.6727294921875], "reading_logprob_std": 1.3039442412316473, "tokenization_path_logprob_std": 126.40864223914912, "trajectory_balance_rmse": 2613.503243690495} {"canonical_length": 351, "canonical_nll": 1.4027464776976495, "generation_rouge_l_f1": 0.215625, "generation_token_f1": 0.48125, "prompt_id": "026d0e5d489e80b2", "reading_canonical_log_prob": -492.364013671875, "reading_distances": [7, 12, 10, 9], "reading_log_probs": [-488.1650390625, -486.5191345214844, -486.6364440917969, -487.4977722167969], "reading_logprob_std": 0.7748241090111366, "tokenization_path_logprob_std": 114.40118704697666, "trajectory_balance_rmse": 9952.302460850911} {"canonical_length": 55, "canonical_nll": 0.9774387706409801, "generation_rouge_l_f1": 0.2444444444444445, "generation_token_f1": 0.26666666666666666, "prompt_id": "4c12c24de5859788", "reading_canonical_log_prob": -53.759132385253906, "reading_distances": [20, 25, 24, 27], "reading_log_probs": [-43.261077880859375, -44.051456451416016, -45.073692321777344, -44.86311721801758], "reading_logprob_std": 0.8278939486753355, "tokenization_path_logprob_std": 169.25789979656147, "trajectory_balance_rmse": 1763.7840543189461} {"canonical_length": 298, "canonical_nll": 1.845225289363989, "generation_rouge_l_f1": 0.17486338797814208, "generation_token_f1": 0.31693989071038253, "prompt_id": "1259351bdd54ab2c", "reading_canonical_log_prob": -549.8771362304688, "reading_distances": [18, 17, 17, 15], "reading_log_probs": [-541.917236328125, -544.1654052734375, -543.457275390625, -545.3009033203125], "reading_logprob_std": 1.4161259926063958, "tokenization_path_logprob_std": 163.55711915897967, "trajectory_balance_rmse": 6917.190435276353} {"canonical_length": 17, "canonical_nll": 1.125426124123966, "generation_rouge_l_f1": 0.6153846153846153, "generation_token_f1": 0.6153846153846153, "prompt_id": "1e2afb46510ad456", "reading_canonical_log_prob": -19.132244110107422, "reading_distances": [26, 30, 36, 28], "reading_log_probs": [-34.26898193359375, -26.12100601196289, -36.41695022583008, -43.485111236572266], "reading_logprob_std": 7.149696009349488, "tokenization_path_logprob_std": 73.91976902313723, "trajectory_balance_rmse": 371.64370245274824} {"canonical_length": 23, "canonical_nll": 1.2042015739109204, "generation_rouge_l_f1": 0.12075471698113208, "generation_token_f1": 0.12830188679245283, "prompt_id": "dfa52b61eae824fc", "reading_canonical_log_prob": -27.696636199951172, "reading_distances": [14, 16, 20, 16], "reading_log_probs": [-20.61693000793457, -20.430906295776367, -19.90804672241211, -20.08909797668457], "reading_logprob_std": 0.3212892783152983, "tokenization_path_logprob_std": 37.85491341890058, "trajectory_balance_rmse": 367.9906885950782} {"canonical_length": 5, "canonical_nll": 1.50463809967041, "generation_rouge_l_f1": 0.031007751937984496, "generation_token_f1": 0.031007751937984496, "prompt_id": "30fd5bfb1635c7a5", "reading_canonical_log_prob": -7.523190498352051, "reading_distances": [19, 19, 28, 23], "reading_log_probs": [-12.720200538635254, -11.337101936340332, -11.322691917419434, -9.648968696594238], "reading_logprob_std": 1.2566426176297356, "tokenization_path_logprob_std": 11.703084033415093, "trajectory_balance_rmse": 18.48504905606795} {"canonical_length": 11, "canonical_nll": 0.14091261950406161, "generation_rouge_l_f1": 0.8888888888888888, "generation_token_f1": 0.8888888888888888, "prompt_id": "bd172f5e24a2f430", "reading_canonical_log_prob": -1.5500388145446777, "reading_distances": [36, 40, 44, 39], "reading_log_probs": [-3.6803321838378906, -3.0859739780426025, -3.3171584606170654, -4.634052276611328], "reading_logprob_std": 0.6818443646635106, "tokenization_path_logprob_std": 44.76593024055535, "trajectory_balance_rmse": 262.31169113367457} {"canonical_length": 8, "canonical_nll": 0.19301477074623108, "generation_rouge_l_f1": 0.8750000000000001, "generation_token_f1": 0.8750000000000001, "prompt_id": "74eb19d7e1933dab", "reading_canonical_log_prob": -1.5441181659698486, "reading_distances": [27, 23, 22, 23], "reading_log_probs": [-2.337235450744629, -2.5735793113708496, -2.3191416263580322, -2.1956608295440674], "reading_logprob_std": 0.15786006512823184, "tokenization_path_logprob_std": 53.23007382621081, "trajectory_balance_rmse": 229.26557778747272} {"canonical_length": 2, "canonical_nll": 0.3892258405685425, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "69770982bd8468eb", "reading_canonical_log_prob": -0.778451681137085, "reading_distances": [27, 24, 27, 30], "reading_log_probs": [-0.556615948677063, -0.7340447902679443, -0.7379156947135925, -0.5130842328071594], "reading_logprob_std": 0.11748522607309761, "tokenization_path_logprob_std": 21.31873180007448, "trajectory_balance_rmse": 30.599707701830553} {"canonical_length": 33, "canonical_nll": 0.9813641634854403, "generation_rouge_l_f1": 0.1444043321299639, "generation_token_f1": 0.1732851985559567, "prompt_id": "decae069198bb200", "reading_canonical_log_prob": -32.38501739501953, "reading_distances": [16, 19, 18, 15], "reading_log_probs": [-36.22148132324219, -36.82946014404297, -34.75617218017578, -34.39397430419922], "reading_logprob_std": 1.1625350387015414, "tokenization_path_logprob_std": 113.92587661059028, "trajectory_balance_rmse": 724.1292797322386} {"canonical_length": 296, "canonical_nll": 1.6645272744668496, "generation_rouge_l_f1": 0.26194690265486725, "generation_token_f1": 0.5415929203539823, "prompt_id": "78378b5209c8e413", "reading_canonical_log_prob": -492.7000732421875, "reading_distances": [15, 20, 20, 20], "reading_log_probs": [-491.68377685546875, -490.4465026855469, -488.718505859375, -491.04644775390625], "reading_logprob_std": 1.2745943281475778, "tokenization_path_logprob_std": 234.65213013735507, "trajectory_balance_rmse": 7236.034163887868} {"canonical_length": 46, "canonical_nll": 1.1157974574876868, "generation_rouge_l_f1": 0.19475655430711614, "generation_token_f1": 0.2546816479400749, "prompt_id": "6c87b1ea5d43b53e", "reading_canonical_log_prob": -51.326683044433594, "reading_distances": [22, 18, 15, 22], "reading_log_probs": [-53.813453674316406, -52.81407928466797, -54.590293884277344, -53.490333557128906], "reading_logprob_std": 0.7376091647351973, "tokenization_path_logprob_std": 131.26429534481088, "trajectory_balance_rmse": 844.8716269472113} {"canonical_length": 120, "canonical_nll": 2.3300636291503904, "generation_rouge_l_f1": 0.17553191489361702, "generation_token_f1": 0.2978723404255319, "prompt_id": "c823f34b5de71522", "reading_canonical_log_prob": -279.6076354980469, "reading_distances": [21, 26, 20, 22], "reading_log_probs": [-282.7594299316406, -283.6065673828125, -284.5736999511719, -285.21417236328125], "reading_logprob_std": 1.0787694401233041, "tokenization_path_logprob_std": 214.62268257828214, "trajectory_balance_rmse": 3851.1592413288727} {"canonical_length": 23, "canonical_nll": 1.4271100914996604, "generation_rouge_l_f1": 0.7333333333333333, "generation_token_f1": 0.8666666666666667, "prompt_id": "95ee759ef85d853e", "reading_canonical_log_prob": -32.82353210449219, "reading_distances": [29, 27, 29, 26], "reading_log_probs": [-50.555824279785156, -38.22775650024414, -48.04344177246094, -36.93284606933594], "reading_logprob_std": 6.863846489809417, "tokenization_path_logprob_std": 69.74772264190275, "trajectory_balance_rmse": 277.91064713654083} {"canonical_length": 11, "canonical_nll": 0.9105586138638583, "generation_rouge_l_f1": 0.823529411764706, "generation_token_f1": 0.823529411764706, "prompt_id": "036acca12e6542a8", "reading_canonical_log_prob": -10.016144752502441, "reading_distances": [18, 20, 25, 21], "reading_log_probs": [-11.757416725158691, -13.138381004333496, -12.674121856689453, -12.593778610229492], "reading_logprob_std": 0.5748547064561557, "tokenization_path_logprob_std": 48.52815690485839, "trajectory_balance_rmse": 284.621100758444} {"canonical_length": 13, "canonical_nll": 2.0689337803767276, "generation_rouge_l_f1": 0.5555555555555556, "generation_token_f1": 0.5555555555555556, "prompt_id": "5e5a1ac1eccda1ac", "reading_canonical_log_prob": -26.89613914489746, "reading_distances": [20, 36, 27, 27], "reading_log_probs": [-25.54581642150879, -26.030120849609375, -25.36336326599121, -25.688112258911133], "reading_logprob_std": 0.2821178090509391, "tokenization_path_logprob_std": 141.78848801575515, "trajectory_balance_rmse": 494.3735155531223} {"canonical_length": 8, "canonical_nll": 0.00238543888553977, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "05b4960bf40b9d90", "reading_canonical_log_prob": -0.01908351108431816, "reading_distances": [24, 30, 24, 25], "reading_log_probs": [-0.08560144901275635, -0.11839371919631958, -0.07850855588912964, -0.03909884765744209], "reading_logprob_std": 0.03255733667544756, "tokenization_path_logprob_std": 57.438189961946605, "trajectory_balance_rmse": 204.52908323944212} {"canonical_length": 199, "canonical_nll": 1.311223629131988, "generation_rouge_l_f1": 0.23893805309734512, "generation_token_f1": 0.4601769911504425, "prompt_id": "55f01821168c15c2", "reading_canonical_log_prob": -260.9335021972656, "reading_distances": [29, 33, 37, 34], "reading_log_probs": [-257.9971923828125, -256.1515808105469, -256.49139404296875, -259.3795471191406], "reading_logprob_std": 1.484960242081479, "tokenization_path_logprob_std": 110.82529019567977, "trajectory_balance_rmse": 5393.391583219733} {"canonical_length": 181, "canonical_nll": 0.8625210925360411, "generation_rouge_l_f1": 0.2798165137614679, "generation_token_f1": 0.43577981651376146, "prompt_id": "28aed3c6f636dfb4", "reading_canonical_log_prob": -156.11631774902344, "reading_distances": [13, 12, 11, 15], "reading_log_probs": [-154.86273193359375, -153.90069580078125, -154.861328125, -152.94677734375], "reading_logprob_std": 0.9171821457069163, "tokenization_path_logprob_std": 107.6698420436056, "trajectory_balance_rmse": 4229.683266785592} {"canonical_length": 102, "canonical_nll": 2.161356308880974, "generation_rouge_l_f1": 0.2214765100671141, "generation_token_f1": 0.5100671140939598, "prompt_id": "33e9d73a4126f71a", "reading_canonical_log_prob": -220.45834350585938, "reading_distances": [18, 20, 12, 20], "reading_log_probs": [-218.28668212890625, -218.39833068847656, -220.6134796142578, -217.6795654296875], "reading_logprob_std": 1.2853794293326246, "tokenization_path_logprob_std": 293.0267591484805, "trajectory_balance_rmse": 3364.30033604458} {"canonical_length": 169, "canonical_nll": 1.992987096662352, "generation_rouge_l_f1": 0.2328767123287671, "generation_token_f1": 0.4383561643835616, "prompt_id": "ba9c16c03160cc50", "reading_canonical_log_prob": -336.8148193359375, "reading_distances": [23, 21, 23, 16], "reading_log_probs": [-333.22552490234375, -334.06103515625, -331.8346862792969, -331.04937744140625], "reading_logprob_std": 1.3543617413023863, "tokenization_path_logprob_std": 363.3951348661052, "trajectory_balance_rmse": 4275.212623173587} {"canonical_length": 278, "canonical_nll": 0.8008434377985892, "generation_rouge_l_f1": 0.2746615087040619, "generation_token_f1": 0.40618955512572535, "prompt_id": "d04a82b0b7e2a434", "reading_canonical_log_prob": -222.6344757080078, "reading_distances": [23, 27, 24, 28], "reading_log_probs": [-225.40914916992188, -231.09744262695312, -232.04776000976562, -229.92050170898438], "reading_logprob_std": 2.9381635278776974, "tokenization_path_logprob_std": 239.4936742144218, "trajectory_balance_rmse": 4848.868933104072} {"canonical_length": 38, "canonical_nll": 0.8842188182630037, "generation_rouge_l_f1": 0.41379310344827586, "generation_token_f1": 0.5172413793103449, "prompt_id": "e1a492712f46f15d", "reading_canonical_log_prob": -33.60031509399414, "reading_distances": [19, 21, 19, 17], "reading_log_probs": [-32.716346740722656, -32.963409423828125, -31.27658462524414, -31.37924575805664], "reading_logprob_std": 0.8797388668261954, "tokenization_path_logprob_std": 110.1522620101818, "trajectory_balance_rmse": 823.9450713302319} {"canonical_length": 25, "canonical_nll": 0.7318704223632813, "generation_rouge_l_f1": 0.1228668941979522, "generation_token_f1": 0.1501706484641638, "prompt_id": "fcc6cd8702ae37ce", "reading_canonical_log_prob": -18.29676055908203, "reading_distances": [23, 26, 26, 22], "reading_log_probs": [-25.858280181884766, -25.101383209228516, -22.509687423706055, -21.356332778930664], "reading_logprob_std": 2.1237940423744455, "tokenization_path_logprob_std": 28.42278128180951, "trajectory_balance_rmse": 421.56492699072174} {"canonical_length": 342, "canonical_nll": 1.3228122644257128, "generation_rouge_l_f1": 0.23294509151414308, "generation_token_f1": 0.43594009983361065, "prompt_id": "38014139c1d9865d", "reading_canonical_log_prob": -452.40179443359375, "reading_distances": [32, 30, 25, 38], "reading_log_probs": [-469.3147277832031, -470.5014953613281, -465.52978515625, -471.85711669921875], "reading_logprob_std": 2.7201207908804994, "tokenization_path_logprob_std": 268.467485263915, "trajectory_balance_rmse": 7784.107212558617} {"canonical_length": 6, "canonical_nll": 1.165693203608195, "generation_rouge_l_f1": 0.888888888888889, "generation_token_f1": 0.888888888888889, "prompt_id": "bd70ea68bde60a24", "reading_canonical_log_prob": -6.99415922164917, "reading_distances": [25, 25, 19, 25], "reading_log_probs": [-5.770493984222412, -6.496870994567871, -5.987557411193848, -6.217196941375732], "reading_logprob_std": 0.3115332834648753, "tokenization_path_logprob_std": 17.351038678872737, "trajectory_balance_rmse": 137.55128516843024} {"canonical_length": 11, "canonical_nll": 2.62217885797674, "generation_rouge_l_f1": 0.064, "generation_token_f1": 0.08, "prompt_id": "57765c15fcf7839d", "reading_canonical_log_prob": -28.84396743774414, "reading_distances": [10, 12, 12, 14], "reading_log_probs": [-29.48360824584961, -30.689878463745117, -30.061206817626953, -30.91443634033203], "reading_logprob_std": 0.6461186605187667, "tokenization_path_logprob_std": 49.99690733891964, "trajectory_balance_rmse": 270.8866442268389} {"canonical_length": 50, "canonical_nll": 1.5405154418945313, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.3214285714285714, "prompt_id": "253987951dace9e0", "reading_canonical_log_prob": -77.02577209472656, "reading_distances": [32, 23, 29, 26], "reading_log_probs": [-75.77831268310547, -79.27361297607422, -76.30828857421875, -77.75801849365234], "reading_logprob_std": 1.570804622961055, "tokenization_path_logprob_std": 188.31967208003752, "trajectory_balance_rmse": 1631.956144129716} {"canonical_length": 38, "canonical_nll": 1.2039248817845394, "generation_rouge_l_f1": 0.16077170418006428, "generation_token_f1": 0.17363344051446944, "prompt_id": "e17273b705c2c377", "reading_canonical_log_prob": -45.7491455078125, "reading_distances": [29, 22, 28, 29], "reading_log_probs": [-46.695716857910156, -46.903839111328125, -45.69662857055664, -46.427978515625], "reading_logprob_std": 0.5269289513040446, "tokenization_path_logprob_std": 140.39805429032478, "trajectory_balance_rmse": 1071.524440988629} {"canonical_length": 344, "canonical_nll": 1.7238226602243822, "generation_rouge_l_f1": 0.23389830508474577, "generation_token_f1": 0.46779661016949153, "prompt_id": "3e31d044adc4af42", "reading_canonical_log_prob": -592.9949951171875, "reading_distances": [23, 27, 31, 26], "reading_log_probs": [-590.9593505859375, -590.8020629882812, -590.104248046875, -589.78076171875], "reading_logprob_std": 0.5612231314003422, "tokenization_path_logprob_std": 133.56083620207602, "trajectory_balance_rmse": 10627.171801717754} {"canonical_length": 23, "canonical_nll": 0.718435867972996, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.711111111111111, "prompt_id": "31f5a769dd1b152b", "reading_canonical_log_prob": -16.524024963378906, "reading_distances": [24, 19, 24, 19], "reading_log_probs": [-17.28329849243164, -19.346988677978516, -19.316774368286133, -18.53234100341797], "reading_logprob_std": 0.9675505135801784, "tokenization_path_logprob_std": 61.93502821633008, "trajectory_balance_rmse": 680.9840917360804} {"canonical_length": 22, "canonical_nll": 0.8479523225264116, "generation_rouge_l_f1": 0.7333333333333333, "generation_token_f1": 0.8000000000000002, "prompt_id": "74fb912570f58841", "reading_canonical_log_prob": -18.654951095581055, "reading_distances": [26, 24, 17, 29], "reading_log_probs": [-22.164318084716797, -38.144832611083984, -20.48429298400879, -42.10202407836914], "reading_logprob_std": 10.994664505462351, "tokenization_path_logprob_std": 41.86601725441686, "trajectory_balance_rmse": 281.2317328438565} {"canonical_length": 10, "canonical_nll": 0.40674724578857424, "generation_rouge_l_f1": 0.16666666666666669, "generation_token_f1": 0.16666666666666669, "prompt_id": "434a6990c55f121a", "reading_canonical_log_prob": -4.067472457885742, "reading_distances": [14, 20, 19, 15], "reading_log_probs": [-3.5637447834014893, -3.5417423248291016, -3.074151039123535, -3.3545970916748047], "reading_logprob_std": 0.2266134441300293, "tokenization_path_logprob_std": 52.298215975931804, "trajectory_balance_rmse": 317.9338338386821} {"canonical_length": 222, "canonical_nll": 1.1894274187517595, "generation_rouge_l_f1": 0.18235294117647058, "generation_token_f1": 0.3470588235294118, "prompt_id": "2e39c9a4bb908fa5", "reading_canonical_log_prob": -264.0528869628906, "reading_distances": [16, 23, 18, 23], "reading_log_probs": [-258.11505126953125, -260.29400634765625, -260.35498046875, -261.1396179199219], "reading_logprob_std": 1.2989599742688005, "tokenization_path_logprob_std": 116.72972702707956, "trajectory_balance_rmse": 7561.313909584518} {"canonical_length": 226, "canonical_nll": 1.5640268241409707, "generation_rouge_l_f1": 0.2030905077262693, "generation_token_f1": 0.423841059602649, "prompt_id": "1a576b185bace9d2", "reading_canonical_log_prob": -353.4700622558594, "reading_distances": [10, 16, 12, 16], "reading_log_probs": [-352.5402526855469, -351.396484375, -351.54339599609375, -358.0652160644531], "reading_logprob_std": 3.1603645876195685, "tokenization_path_logprob_std": 374.8802460186224, "trajectory_balance_rmse": 5638.931572481904} {"canonical_length": 270, "canonical_nll": 1.264342357494213, "generation_rouge_l_f1": 0.24535315985130113, "generation_token_f1": 0.5353159851301116, "prompt_id": "c2df8f1e4778ecd3", "reading_canonical_log_prob": -341.3724365234375, "reading_distances": [35, 28, 35, 35], "reading_log_probs": [-341.70623779296875, -341.7767333984375, -338.2265930175781, -339.32550048828125], "reading_logprob_std": 1.770132872602905, "tokenization_path_logprob_std": 345.4270895072584, "trajectory_balance_rmse": 7321.847158356517} {"canonical_length": 2, "canonical_nll": 0.8624330163002014, "generation_rouge_l_f1": 0.025974025974025976, "generation_token_f1": 0.025974025974025976, "prompt_id": "42c831675952234b", "reading_canonical_log_prob": -1.7248660326004028, "reading_distances": [68, 67, 67, 49], "reading_log_probs": [-2.2551844120025635, -1.3271546363830566, -2.287294864654541, -2.2193357944488525], "reading_logprob_std": 0.46422250241330787, "tokenization_path_logprob_std": 7.023034385485826, "trajectory_balance_rmse": 28.95891211311165} {"canonical_length": 16, "canonical_nll": 2.8310418128967285, "generation_rouge_l_f1": 0.2758620689655172, "generation_token_f1": 0.2758620689655172, "prompt_id": "a9f4c15e6cb406fd", "reading_canonical_log_prob": -45.296669006347656, "reading_distances": [21, 18, 20, 19], "reading_log_probs": [-48.40736389160156, -48.41735076904297, -46.87660217285156, -47.192893981933594], "reading_logprob_std": 0.8057868570253134, "tokenization_path_logprob_std": 77.31596808042843, "trajectory_balance_rmse": 569.2158051002127} {"canonical_length": 161, "canonical_nll": 1.8531442962077833, "generation_rouge_l_f1": 0.16216216216216214, "generation_token_f1": 0.3243243243243243, "prompt_id": "25e2da0cacffb5d9", "reading_canonical_log_prob": -298.3562316894531, "reading_distances": [23, 21, 19, 23], "reading_log_probs": [-294.76812744140625, -295.17718505859375, -293.59759521484375, -293.53155517578125], "reading_logprob_std": 0.8303687869533448, "tokenization_path_logprob_std": 393.97767919563836, "trajectory_balance_rmse": 4712.542836557266} {"canonical_length": 116, "canonical_nll": 1.287500052616514, "generation_rouge_l_f1": 0.2081218274111675, "generation_token_f1": 0.30456852791878175, "prompt_id": "d71e3178e4fc08e6", "reading_canonical_log_prob": -149.35000610351562, "reading_distances": [36, 35, 42, 35], "reading_log_probs": [-150.13145446777344, -150.681884765625, -148.68792724609375, -149.45361328125], "reading_logprob_std": 0.8620230836104883, "tokenization_path_logprob_std": 79.11089729192082, "trajectory_balance_rmse": 4207.952140078129} {"canonical_length": 9, "canonical_nll": 0.6392218801710341, "generation_rouge_l_f1": 0.875, "generation_token_f1": 0.875, "prompt_id": "c69244953386b7ea", "reading_canonical_log_prob": -5.752996921539307, "reading_distances": [16, 19, 15, 5], "reading_log_probs": [-6.0398712158203125, -6.393109321594238, -6.234692573547363, -4.9810709953308105], "reading_logprob_std": 0.6373320425789782, "tokenization_path_logprob_std": 20.01748523361788, "trajectory_balance_rmse": 279.2925103728255} {"canonical_length": 16, "canonical_nll": 2.1069142818450928, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "44b99fb81cd83d68", "reading_canonical_log_prob": -33.710628509521484, "reading_distances": [25, 21, 24, 30], "reading_log_probs": [-37.028194427490234, -37.04859924316406, -34.91409683227539, -36.91498947143555], "reading_logprob_std": 1.0432387002066932, "tokenization_path_logprob_std": 100.54165378319269, "trajectory_balance_rmse": 519.0981280544006} {"canonical_length": 23, "canonical_nll": 0.3943954965342646, "generation_rouge_l_f1": 0.08411214953271028, "generation_token_f1": 0.11214953271028036, "prompt_id": "f3f998d8d70ee428", "reading_canonical_log_prob": -9.071096420288086, "reading_distances": [51, 41, 40, 42], "reading_log_probs": [-13.642406463623047, -15.612318992614746, -10.675378799438477, -14.723899841308594], "reading_logprob_std": 2.148771473496827, "tokenization_path_logprob_std": 33.84410738397084, "trajectory_balance_rmse": 376.31926901361396} {"canonical_length": 45, "canonical_nll": 0.8787812127007378, "generation_rouge_l_f1": 0.34328358208955223, "generation_token_f1": 0.4477611940298507, "prompt_id": "8565123aabdca8e0", "reading_canonical_log_prob": -39.5451545715332, "reading_distances": [21, 22, 24, 22], "reading_log_probs": [-45.3309440612793, -45.84242248535156, -44.36427688598633, -47.074989318847656], "reading_logprob_std": 1.128784256481038, "tokenization_path_logprob_std": 77.58616461522301, "trajectory_balance_rmse": 1349.1698530509648} {"canonical_length": 293, "canonical_nll": 1.754860731521971, "generation_rouge_l_f1": 0.19780219780219777, "generation_token_f1": 0.43956043956043955, "prompt_id": "c754674f366c507b", "reading_canonical_log_prob": -514.1741943359375, "reading_distances": [27, 30, 36, 32], "reading_log_probs": [-503.69989013671875, -503.90240478515625, -510.6671142578125, -509.4899597167969], "reading_logprob_std": 3.6569099781043373, "tokenization_path_logprob_std": 108.87108839346467, "trajectory_balance_rmse": 8354.542377586391} {"canonical_length": 15, "canonical_nll": 0.9478977839152019, "generation_rouge_l_f1": 0.10218978102189781, "generation_token_f1": 0.11678832116788321, "prompt_id": "f7c03b54285f64b4", "reading_canonical_log_prob": -14.218466758728027, "reading_distances": [26, 26, 29, 25], "reading_log_probs": [-23.222835540771484, -32.70891571044922, -30.677940368652344, -21.951824188232422], "reading_logprob_std": 5.347622191911175, "tokenization_path_logprob_std": 100.840288515818, "trajectory_balance_rmse": 326.8509941947341} {"canonical_length": 33, "canonical_nll": 0.5144463741418087, "generation_rouge_l_f1": 0.16455696202531644, "generation_token_f1": 0.2658227848101266, "prompt_id": "278cabb1c777cfb5", "reading_canonical_log_prob": -16.976730346679688, "reading_distances": [58, 44, 50, 56], "reading_log_probs": [-23.0026798248291, -25.428253173828125, -17.094602584838867, -14.482053756713867], "reading_logprob_std": 5.078420657839409, "tokenization_path_logprob_std": 98.02992905384276, "trajectory_balance_rmse": 597.4398402397428} {"canonical_length": 82, "canonical_nll": 0.5389006079697027, "generation_rouge_l_f1": 0.36363636363636365, "generation_token_f1": 0.4031620553359684, "prompt_id": "26496139eab8348f", "reading_canonical_log_prob": -44.189849853515625, "reading_distances": [17, 20, 18, 17], "reading_log_probs": [-45.75945281982422, -45.5903434753418, -45.24505615234375, -46.384979248046875], "reading_logprob_std": 0.47736946078783593, "tokenization_path_logprob_std": 40.11585760742956, "trajectory_balance_rmse": 1163.4352165793784} {"canonical_length": 130, "canonical_nll": 2.2853834885817306, "generation_rouge_l_f1": 0.1804511278195489, "generation_token_f1": 0.3659147869674185, "prompt_id": "5e25b104005e6f72", "reading_canonical_log_prob": -297.099853515625, "reading_distances": [25, 23, 23, 19], "reading_log_probs": [-307.090087890625, -308.10308837890625, -303.3800354003906, -304.8518371582031], "reading_logprob_std": 2.137843347921534, "tokenization_path_logprob_std": 269.82675502517014, "trajectory_balance_rmse": 2309.800443808791} {"canonical_length": 107, "canonical_nll": 2.572059702650409, "generation_rouge_l_f1": 0.2177650429799427, "generation_token_f1": 0.34383954154727797, "prompt_id": "14f8273006e150ca", "reading_canonical_log_prob": -275.21038818359375, "reading_distances": [25, 25, 34, 30], "reading_log_probs": [-286.460205078125, -277.6080017089844, -283.2407531738281, -281.4627685546875], "reading_logprob_std": 3.690631073676341, "tokenization_path_logprob_std": 163.74712364762104, "trajectory_balance_rmse": 2701.689312504256} {"canonical_length": 185, "canonical_nll": 1.657695873363598, "generation_rouge_l_f1": 0.28418230563002683, "generation_token_f1": 0.49865951742627346, "prompt_id": "c0e6a5cad32d62b6", "reading_canonical_log_prob": -306.6737365722656, "reading_distances": [13, 17, 17, 12], "reading_log_probs": [-307.3057861328125, -309.01361083984375, -309.6821594238281, -309.0513610839844], "reading_logprob_std": 1.0188693076963355, "tokenization_path_logprob_std": 336.36888753366657, "trajectory_balance_rmse": 4533.3968054065235} {"canonical_length": 16, "canonical_nll": 1.0815364122390747, "generation_rouge_l_f1": 0.43749999999999994, "generation_token_f1": 0.5625, "prompt_id": "937bfa51bfbd0542", "reading_canonical_log_prob": -17.304582595825195, "reading_distances": [31, 34, 37, 39], "reading_log_probs": [-22.289058685302734, -20.319568634033203, -19.96839714050293, -17.732791900634766], "reading_logprob_std": 1.8671858092567326, "tokenization_path_logprob_std": 44.4790213924533, "trajectory_balance_rmse": 554.1060871842171} {"canonical_length": 75, "canonical_nll": 0.5907782491048177, "generation_rouge_l_f1": 0.3059701492537314, "generation_token_f1": 0.3656716417910448, "prompt_id": "19b52394b9bc5537", "reading_canonical_log_prob": -44.30836868286133, "reading_distances": [13, 13, 17, 12], "reading_log_probs": [-43.98762893676758, -43.37644958496094, -43.61372375488281, -42.918827056884766], "reading_logprob_std": 0.4476123298830166, "tokenization_path_logprob_std": 87.90140272896952, "trajectory_balance_rmse": 836.0721886819352} {"canonical_length": 339, "canonical_nll": 1.4523132684308167, "generation_rouge_l_f1": 0.2391304347826087, "generation_token_f1": 0.43115942028985504, "prompt_id": "c03437022c1ce69c", "reading_canonical_log_prob": -492.3341979980469, "reading_distances": [29, 29, 24, 28], "reading_log_probs": [-497.46368408203125, -496.13616943359375, -497.520263671875, -490.7701416015625], "reading_logprob_std": 3.1995198448144793, "tokenization_path_logprob_std": 234.0054302441362, "trajectory_balance_rmse": 7465.205297097463} {"canonical_length": 22, "canonical_nll": 1.2031055797230115, "generation_rouge_l_f1": 0.29508196721311475, "generation_token_f1": 0.32786885245901637, "prompt_id": "01ef5d8df6df3761", "reading_canonical_log_prob": -26.46832275390625, "reading_distances": [21, 20, 21, 26], "reading_log_probs": [-25.364551544189453, -28.027772903442383, -27.80172348022461, -29.606170654296875], "reading_logprob_std": 1.7517218941911565, "tokenization_path_logprob_std": 53.65103073088451, "trajectory_balance_rmse": 483.32464974701895} {"canonical_length": 110, "canonical_nll": 1.6895303899591618, "generation_rouge_l_f1": 0.20163487738419617, "generation_token_f1": 0.35967302452316074, "prompt_id": "1275798cfa3be44f", "reading_canonical_log_prob": -185.8483428955078, "reading_distances": [36, 34, 37, 38], "reading_log_probs": [-187.3026123046875, -189.899169921875, -188.89590454101562, -190.7333221435547], "reading_logprob_std": 1.4756048592498854, "tokenization_path_logprob_std": 154.18123684903526, "trajectory_balance_rmse": 3535.763234580744} {"canonical_length": 4, "canonical_nll": 0.06994910538196564, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1f2fdcd7eec76cc5", "reading_canonical_log_prob": -0.27979642152786255, "reading_distances": [29, 29, 25, 33], "reading_log_probs": [-4.937194347381592, -0.23512816429138184, -0.3167191743850708, -3.759092330932617], "reading_logprob_std": 2.4000183170345735, "tokenization_path_logprob_std": 38.153237207210026, "trajectory_balance_rmse": 67.60521147878217} {"canonical_length": 128, "canonical_nll": 1.1559526920318604, "generation_rouge_l_f1": 0.21114369501466276, "generation_token_f1": 0.3460410557184751, "prompt_id": "b8b97c6b0ef87a98", "reading_canonical_log_prob": -147.96194458007812, "reading_distances": [32, 28, 24, 26], "reading_log_probs": [-143.87335205078125, -147.42172241210938, -145.560791015625, -143.33457946777344], "reading_logprob_std": 1.8450943820274188, "tokenization_path_logprob_std": 97.9789035598191, "trajectory_balance_rmse": 2094.1191982786404} {"canonical_length": 168, "canonical_nll": 1.129118419828869, "generation_rouge_l_f1": 0.2729528535980149, "generation_token_f1": 0.49627791563275425, "prompt_id": "7ec56cfda55797c0", "reading_canonical_log_prob": -189.69189453125, "reading_distances": [27, 35, 24, 27], "reading_log_probs": [-188.78509521484375, -191.5623321533203, -192.30130004882812, -192.6102294921875], "reading_logprob_std": 1.7427945344344324, "tokenization_path_logprob_std": 109.12849497241683, "trajectory_balance_rmse": 4496.054975233228} {"canonical_length": 32, "canonical_nll": 1.370691180229187, "generation_rouge_l_f1": 0.46511627906976744, "generation_token_f1": 0.5116279069767442, "prompt_id": "c6a2889afcb5222e", "reading_canonical_log_prob": -43.862117767333984, "reading_distances": [15, 16, 13, 9], "reading_log_probs": [-42.74580383300781, -43.73125457763672, -44.32268524169922, -42.15925598144531], "reading_logprob_std": 0.97052831447116, "tokenization_path_logprob_std": 48.24652869815215, "trajectory_balance_rmse": 624.4485410237158} {"canonical_length": 32, "canonical_nll": 2.168635368347168, "generation_rouge_l_f1": 0.1935483870967742, "generation_token_f1": 0.1935483870967742, "prompt_id": "5371fa87b29b339c", "reading_canonical_log_prob": -69.39633178710938, "reading_distances": [12, 17, 19, 16], "reading_log_probs": [-72.68032836914062, -71.31114959716797, -73.64910125732422, -72.70770263671875], "reading_logprob_std": 0.96248508600607, "tokenization_path_logprob_std": 127.25407846292312, "trajectory_balance_rmse": 925.4730850075107} {"canonical_length": 63, "canonical_nll": 1.2960902622767858, "generation_rouge_l_f1": 0.2962962962962963, "generation_token_f1": 0.4232804232804233, "prompt_id": "c89f55288eb64d0c", "reading_canonical_log_prob": -81.6536865234375, "reading_distances": [22, 29, 25, 31], "reading_log_probs": [-91.94915771484375, -91.89044189453125, -91.60111236572266, -91.8634262084961], "reading_logprob_std": 0.154159814106323, "tokenization_path_logprob_std": 76.86956509433061, "trajectory_balance_rmse": 2035.9176325139724} {"canonical_length": 221, "canonical_nll": 1.559541866250707, "generation_rouge_l_f1": 0.2887473460721869, "generation_token_f1": 0.5307855626326964, "prompt_id": "0e1e524655779cac", "reading_canonical_log_prob": -344.65875244140625, "reading_distances": [18, 24, 18, 17], "reading_log_probs": [-342.921630859375, -342.08892822265625, -340.2497253417969, -343.22039794921875], "reading_logprob_std": 1.3357120844207662, "tokenization_path_logprob_std": 219.73437032292918, "trajectory_balance_rmse": 6211.433202176536} {"canonical_length": 9, "canonical_nll": 0.24909573131137425, "generation_rouge_l_f1": 0.8750000000000001, "generation_token_f1": 0.8750000000000001, "prompt_id": "5f85bd9588d13639", "reading_canonical_log_prob": -2.241861581802368, "reading_distances": [21, 13, 20, 14], "reading_log_probs": [-2.6309051513671875, -2.678919553756714, -1.9843734502792358, -2.0096018314361572], "reading_logprob_std": 0.3804978551395689, "tokenization_path_logprob_std": 57.057186331318974, "trajectory_balance_rmse": 266.12289757808674} {"canonical_length": 26, "canonical_nll": 0.8730433537409856, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "570dbf2f8a32567a", "reading_canonical_log_prob": -22.699127197265625, "reading_distances": [20, 14, 21, 23], "reading_log_probs": [-23.461456298828125, -22.467041015625, -23.920658111572266, -24.709516525268555], "reading_logprob_std": 0.9363661614168671, "tokenization_path_logprob_std": 146.81745113664337, "trajectory_balance_rmse": 728.9907471620041} {"canonical_length": 26, "canonical_nll": 1.1712859227107122, "generation_rouge_l_f1": 0.13907284768211922, "generation_token_f1": 0.15231788079470196, "prompt_id": "b4dad9a8bbd784f6", "reading_canonical_log_prob": -30.453433990478516, "reading_distances": [40, 29, 38, 37], "reading_log_probs": [-29.162567138671875, -29.684673309326172, -29.590530395507812, -29.077049255371094], "reading_logprob_std": 0.30342399809566284, "tokenization_path_logprob_std": 67.25068721164517, "trajectory_balance_rmse": 819.1867225293093} {"canonical_length": 10, "canonical_nll": 2.793928527832031, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.39999999999999997, "prompt_id": "9e76d45f4e1b2541", "reading_canonical_log_prob": -27.939285278320312, "reading_distances": [28, 41, 31, 40], "reading_log_probs": [-31.979137420654297, -31.75067901611328, -32.033878326416016, -35.47392654418945], "reading_logprob_std": 1.7805768394127257, "tokenization_path_logprob_std": 32.32662678381499, "trajectory_balance_rmse": 441.2044900292947} {"canonical_length": 256, "canonical_nll": 1.5155577659606934, "generation_rouge_l_f1": 0.22393822393822396, "generation_token_f1": 0.4633204633204633, "prompt_id": "6a9973c4375e58ef", "reading_canonical_log_prob": -387.9827880859375, "reading_distances": [32, 35, 29, 33], "reading_log_probs": [-385.3514099121094, -380.6202087402344, -388.4294128417969, -381.87664794921875], "reading_logprob_std": 3.5288498266173307, "tokenization_path_logprob_std": 507.3587475774036, "trajectory_balance_rmse": 8057.715170334842} {"canonical_length": 277, "canonical_nll": 1.211331914908619, "generation_rouge_l_f1": 0.24029574861367836, "generation_token_f1": 0.5138632162661737, "prompt_id": "4a7ad2175af88b68", "reading_canonical_log_prob": -335.5389404296875, "reading_distances": [30, 31, 28, 36], "reading_log_probs": [-338.103759765625, -338.368408203125, -335.01055908203125, -341.6004943847656], "reading_logprob_std": 2.6927968942000264, "tokenization_path_logprob_std": 391.7040168139021, "trajectory_balance_rmse": 7318.836206733959} {"canonical_length": 24, "canonical_nll": 0.10095630089441936, "generation_rouge_l_f1": 0.37037037037037035, "generation_token_f1": 0.37037037037037035, "prompt_id": "7271cafab8ca2bf4", "reading_canonical_log_prob": -2.4229512214660645, "reading_distances": [28, 34, 34, 26], "reading_log_probs": [-2.3568496704101562, -3.1973795890808105, -8.390498161315918, -7.497861385345459], "reading_logprob_std": 3.024908135081731, "tokenization_path_logprob_std": 85.79265744559015, "trajectory_balance_rmse": 702.8587367128913} {"canonical_length": 11, "canonical_nll": 0.4665397730740634, "generation_rouge_l_f1": 0.12162162162162163, "generation_token_f1": 0.12162162162162163, "prompt_id": "7798c57a62239065", "reading_canonical_log_prob": -5.131937503814697, "reading_distances": [14, 14, 17, 15], "reading_log_probs": [-6.962700843811035, -7.63696813583374, -5.449221134185791, -6.639776706695557], "reading_logprob_std": 0.9150404122073436, "tokenization_path_logprob_std": 72.39830534832915, "trajectory_balance_rmse": 327.215013261187} {"canonical_length": 97, "canonical_nll": 1.2440051835836823, "generation_rouge_l_f1": 0.2601626016260163, "generation_token_f1": 0.3794037940379404, "prompt_id": "fd06dfbeb5280c45", "reading_canonical_log_prob": -120.66850280761719, "reading_distances": [39, 44, 39, 45], "reading_log_probs": [-122.08619689941406, -121.59681701660156, -120.8660888671875, -121.50679779052734], "reading_logprob_std": 0.5013676812181255, "tokenization_path_logprob_std": 373.2844676505417, "trajectory_balance_rmse": 2892.8398573025297} {"canonical_length": 210, "canonical_nll": 1.148390125093006, "generation_rouge_l_f1": 0.2795180722891566, "generation_token_f1": 0.4481927710843374, "prompt_id": "3efd48165ac7ab7c", "reading_canonical_log_prob": -241.16192626953125, "reading_distances": [16, 19, 17, 19], "reading_log_probs": [-243.96763610839844, -251.99807739257812, -240.57565307617188, -245.1590576171875], "reading_logprob_std": 4.792916537801654, "tokenization_path_logprob_std": 172.45037581836124, "trajectory_balance_rmse": 2998.810357342136} {"canonical_length": 69, "canonical_nll": 1.5714018448539402, "generation_rouge_l_f1": 0.17971014492753623, "generation_token_f1": 0.25507246376811593, "prompt_id": "18e11e5d4d3c48a7", "reading_canonical_log_prob": -108.42672729492188, "reading_distances": [32, 40, 35, 39], "reading_log_probs": [-112.53741455078125, -120.11968994140625, -112.91699981689453, -108.34197235107422], "reading_logprob_std": 4.88842999084723, "tokenization_path_logprob_std": 165.76216910683007, "trajectory_balance_rmse": 2627.753144425771} {"canonical_length": 17, "canonical_nll": 0.16889039207907283, "generation_rouge_l_f1": 0.7692307692307693, "generation_token_f1": 1.0, "prompt_id": "b05a564949c24a61", "reading_canonical_log_prob": -2.8711366653442383, "reading_distances": [70, 64, 51, 67], "reading_log_probs": [-2.2406063079833984, -2.3397369384765625, -2.5460991859436035, -2.6599597930908203], "reading_logprob_std": 0.19085387462554673, "tokenization_path_logprob_std": 71.72270884601836, "trajectory_balance_rmse": 185.37177746914247} {"canonical_length": 129, "canonical_nll": 0.8191247984420421, "generation_rouge_l_f1": 0.45689655172413796, "generation_token_f1": 0.5862068965517241, "prompt_id": "977aa631a6e3fb74", "reading_canonical_log_prob": -105.66709899902344, "reading_distances": [9, 15, 14, 11], "reading_log_probs": [-112.1969985961914, -111.4139175415039, -112.26338958740234, -108.88816833496094], "reading_logprob_std": 1.5826957939332078, "tokenization_path_logprob_std": 140.31794401201444, "trajectory_balance_rmse": 1832.20020172484} {"canonical_length": 85, "canonical_nll": 1.319609608369715, "generation_rouge_l_f1": 0.29859154929577464, "generation_token_f1": 0.3605633802816901, "prompt_id": "308d6723913890bf", "reading_canonical_log_prob": -112.16681671142578, "reading_distances": [24, 23, 23, 21], "reading_log_probs": [-114.40404510498047, -110.22114562988281, -112.63314819335938, -113.04449462890625], "reading_logprob_std": 1.7425881903530804, "tokenization_path_logprob_std": 102.51022498993419, "trajectory_balance_rmse": 2506.7780028585835} {"canonical_length": 189, "canonical_nll": 1.219981037119709, "generation_rouge_l_f1": 0.29357798165137616, "generation_token_f1": 0.41743119266055045, "prompt_id": "ff140f7a02786106", "reading_canonical_log_prob": -230.576416015625, "reading_distances": [15, 15, 20, 21], "reading_log_probs": [-235.68545532226562, -234.0327606201172, -230.21047973632812, -228.47427368164062], "reading_logprob_std": 3.332027883078307, "tokenization_path_logprob_std": 216.3910691551028, "trajectory_balance_rmse": 4640.753247601087} {"canonical_length": 27, "canonical_nll": 1.8215970639829282, "generation_rouge_l_f1": 0.2962962962962963, "generation_token_f1": 0.4074074074074074, "prompt_id": "536f5c5bf3146493", "reading_canonical_log_prob": -49.18312072753906, "reading_distances": [31, 42, 32, 33], "reading_log_probs": [-58.38268280029297, -61.46198654174805, -53.26192092895508, -60.45954132080078], "reading_logprob_std": 3.6522751672703495, "tokenization_path_logprob_std": 80.56713281800917, "trajectory_balance_rmse": 1125.2977064862664} {"canonical_length": 210, "canonical_nll": 1.6259515671502975, "generation_rouge_l_f1": 0.2661122661122661, "generation_token_f1": 0.5280665280665281, "prompt_id": "18e930a2b5a84e85", "reading_canonical_log_prob": -341.4498291015625, "reading_distances": [14, 19, 17, 14], "reading_log_probs": [-339.2505798339844, -347.2860107421875, -342.5994873046875, -340.9964599609375], "reading_logprob_std": 3.4511239252657075, "tokenization_path_logprob_std": 393.697121165924, "trajectory_balance_rmse": 6031.898752894531} {"canonical_length": 21, "canonical_nll": 1.631134033203125, "generation_rouge_l_f1": 0.2727272727272727, "generation_token_f1": 0.36363636363636365, "prompt_id": "2631f3bcf5083818", "reading_canonical_log_prob": -34.253814697265625, "reading_distances": [32, 21, 35, 28], "reading_log_probs": [-36.80852508544922, -33.357398986816406, -34.57080841064453, -35.578025817871094], "reading_logprob_std": 1.4677025292081747, "tokenization_path_logprob_std": 86.16320254581447, "trajectory_balance_rmse": 376.7216874809502} {"canonical_length": 169, "canonical_nll": 1.0583290235530696, "generation_rouge_l_f1": 0.2548076923076923, "generation_token_f1": 0.4423076923076923, "prompt_id": "62ad42126bc04505", "reading_canonical_log_prob": -178.85760498046875, "reading_distances": [43, 45, 43, 38], "reading_log_probs": [-187.71002197265625, -184.7518768310547, -187.17288208007812, -188.15916442871094], "reading_logprob_std": 1.5188948545180363, "tokenization_path_logprob_std": 309.9958365185374, "trajectory_balance_rmse": 3966.701003500696} {"canonical_length": 221, "canonical_nll": 1.3761549729567308, "generation_rouge_l_f1": 0.27814569536423844, "generation_token_f1": 0.4172185430463576, "prompt_id": "44a0d1ef8b9a47d0", "reading_canonical_log_prob": -304.1302490234375, "reading_distances": [15, 7, 13, 19], "reading_log_probs": [-310.52484130859375, -308.5699157714844, -304.5550537109375, -304.16790771484375], "reading_logprob_std": 3.102651780471547, "tokenization_path_logprob_std": 246.13014251095476, "trajectory_balance_rmse": 6816.298066063384} {"canonical_length": 84, "canonical_nll": 1.301637467883882, "generation_rouge_l_f1": 0.024390243902439022, "generation_token_f1": 0.024390243902439022, "prompt_id": "1bc849b3b651f6ed", "reading_canonical_log_prob": -109.3375473022461, "reading_distances": [23, 20, 21, 26], "reading_log_probs": [-110.2777099609375, -111.9029541015625, -107.2197494506836, -109.29582214355469], "reading_logprob_std": 1.9578105407018132, "tokenization_path_logprob_std": 228.71896217394507, "trajectory_balance_rmse": 2797.696043586647} {"canonical_length": 225, "canonical_nll": 1.3718492296006946, "generation_rouge_l_f1": 0.21095334685598377, "generation_token_f1": 0.4340770791075051, "prompt_id": "a8febb0397bb60c3", "reading_canonical_log_prob": -308.66607666015625, "reading_distances": [16, 15, 18, 12], "reading_log_probs": [-305.3916320800781, -307.3394775390625, -306.51220703125, -304.8951416015625], "reading_logprob_std": 1.1019054791918916, "tokenization_path_logprob_std": 117.38621623507457, "trajectory_balance_rmse": 5921.371353109639} {"canonical_length": 7, "canonical_nll": 2.843714850289481, "generation_rouge_l_f1": 0.27272727272727276, "generation_token_f1": 0.27272727272727276, "prompt_id": "37af528e072ecf7b", "reading_canonical_log_prob": -19.906003952026367, "reading_distances": [9, 15, 22, 20], "reading_log_probs": [-22.800979614257812, -24.877437591552734, -25.956148147583008, -24.573373794555664], "reading_logprob_std": 1.3094618329437377, "tokenization_path_logprob_std": 64.76951397077737, "trajectory_balance_rmse": 176.62155335522414} {"canonical_length": 252, "canonical_nll": 1.6624425252278645, "generation_rouge_l_f1": 0.176, "generation_token_f1": 0.4760000000000001, "prompt_id": "192d126d4e011ab4", "reading_canonical_log_prob": -418.9355163574219, "reading_distances": [21, 25, 16, 20], "reading_log_probs": [-416.86907958984375, -413.3475341796875, -413.09844970703125, -413.693603515625], "reading_logprob_std": 1.7615948078035517, "tokenization_path_logprob_std": 363.2480837360542, "trajectory_balance_rmse": 8345.621021340077} {"canonical_length": 182, "canonical_nll": 1.4984300215165693, "generation_rouge_l_f1": 0.18765432098765433, "generation_token_f1": 0.37530864197530867, "prompt_id": "55712d2518889f49", "reading_canonical_log_prob": -272.7142639160156, "reading_distances": [23, 25, 23, 19], "reading_log_probs": [-281.35516357421875, -275.1662902832031, -278.1728515625, -262.9039611816406], "reading_logprob_std": 8.06958937032502, "tokenization_path_logprob_std": 91.31921290675707, "trajectory_balance_rmse": 4663.208348618325} {"canonical_length": 12, "canonical_nll": 1.3590707778930664, "generation_rouge_l_f1": 0.20512820512820515, "generation_token_f1": 0.20512820512820515, "prompt_id": "587eb61a8e257f2b", "reading_canonical_log_prob": -16.308849334716797, "reading_distances": [21, 25, 23, 29], "reading_log_probs": [-18.574764251708984, -16.67640495300293, -15.207565307617188, -16.58261489868164], "reading_logprob_std": 1.3834589397857382, "tokenization_path_logprob_std": 44.69192245529288, "trajectory_balance_rmse": 434.96586904679964} {"canonical_length": 109, "canonical_nll": 0.957702356740969, "generation_rouge_l_f1": 0.10029498525073745, "generation_token_f1": 0.1415929203539823, "prompt_id": "fff5e81d18ab40f8", "reading_canonical_log_prob": -104.38955688476562, "reading_distances": [15, 15, 14, 16], "reading_log_probs": [-104.8338851928711, -109.72712707519531, -106.66273498535156, -109.29305267333984], "reading_logprob_std": 2.3034425849809543, "tokenization_path_logprob_std": 163.00704562287294, "trajectory_balance_rmse": 1492.4046884835263} {"canonical_length": 15, "canonical_nll": 0.5508763631184895, "generation_rouge_l_f1": 0.4210526315789473, "generation_token_f1": 0.4210526315789473, "prompt_id": "e02d75efb09e53b0", "reading_canonical_log_prob": -8.263145446777344, "reading_distances": [25, 28, 20, 18], "reading_log_probs": [-7.291224479675293, -8.262349128723145, -7.421272277832031, -7.650353908538818], "reading_logprob_std": 0.4304451448433185, "tokenization_path_logprob_std": 54.762010832578554, "trajectory_balance_rmse": 344.7481585396276} {"canonical_length": 27, "canonical_nll": 1.5483483208550348, "generation_rouge_l_f1": 0.20168067226890754, "generation_token_f1": 0.25210084033613445, "prompt_id": "93316a155b82ba8e", "reading_canonical_log_prob": -41.80540466308594, "reading_distances": [27, 27, 28, 24], "reading_log_probs": [-45.256126403808594, -43.01604461669922, -47.13862609863281, -41.628135681152344], "reading_logprob_std": 2.4326179678028796, "tokenization_path_logprob_std": 182.70636637814258, "trajectory_balance_rmse": 794.5450570906063} {"canonical_length": 110, "canonical_nll": 1.4598570390181107, "generation_rouge_l_f1": 0.43668122270742354, "generation_token_f1": 0.5851528384279476, "prompt_id": "37f1255c888facef", "reading_canonical_log_prob": -160.5842742919922, "reading_distances": [14, 19, 18, 12], "reading_log_probs": [-160.97731018066406, -159.61346435546875, -161.323974609375, -160.4043426513672], "reading_logprob_std": 0.7475292745937364, "tokenization_path_logprob_std": 259.467794673736, "trajectory_balance_rmse": 3481.8143480481103} {"canonical_length": 150, "canonical_nll": 0.9665211995442708, "generation_rouge_l_f1": 0.40909090909090906, "generation_token_f1": 0.5909090909090909, "prompt_id": "cf1ad5289e6ad557", "reading_canonical_log_prob": -144.97817993164062, "reading_distances": [19, 21, 21, 19], "reading_log_probs": [-141.55746459960938, -144.12603759765625, -144.36111450195312, -143.91424560546875], "reading_logprob_std": 1.3010331004090836, "tokenization_path_logprob_std": 121.59475266484297, "trajectory_balance_rmse": 3866.3092986022284} {"canonical_length": 314, "canonical_nll": 1.0498700961945162, "generation_rouge_l_f1": 0.2880258899676375, "generation_token_f1": 0.4919093851132686, "prompt_id": "ff9bcfc5d56ada93", "reading_canonical_log_prob": -329.6592102050781, "reading_distances": [41, 34, 36, 40], "reading_log_probs": [-353.7966003417969, -359.07647705078125, -355.6094970703125, -357.1241149902344], "reading_logprob_std": 2.2427992477721315, "tokenization_path_logprob_std": 236.31832943795052, "trajectory_balance_rmse": 7523.871011421788} {"canonical_length": 12, "canonical_nll": 1.9527463912963867, "generation_rouge_l_f1": 0.10526315789473685, "generation_token_f1": 0.10526315789473685, "prompt_id": "6f26ffa7e9fb10c0", "reading_canonical_log_prob": -23.43295669555664, "reading_distances": [22, 21, 20, 22], "reading_log_probs": [-29.348392486572266, -24.473392486572266, -24.42774772644043, -24.080259323120117], "reading_logprob_std": 2.516760189245503, "tokenization_path_logprob_std": 46.68736305877527, "trajectory_balance_rmse": 318.2151246168474} {"canonical_length": 41, "canonical_nll": 1.1156634819216844, "generation_rouge_l_f1": 0.20359281437125748, "generation_token_f1": 0.311377245508982, "prompt_id": "07e67bead15cc6c8", "reading_canonical_log_prob": -45.74220275878906, "reading_distances": [31, 38, 35, 24], "reading_log_probs": [-43.12353515625, -42.868736267089844, -43.77098083496094, -44.79620361328125], "reading_logprob_std": 0.8593692223572996, "tokenization_path_logprob_std": 164.13855133716754, "trajectory_balance_rmse": 1063.1326562348522} {"canonical_length": 413, "canonical_nll": 1.2104081345528148, "generation_rouge_l_f1": 0.23903177004538576, "generation_token_f1": 0.4266263237518911, "prompt_id": "52cb3482c2bf1893", "reading_canonical_log_prob": -499.8985595703125, "reading_distances": [26, 24, 22, 33], "reading_log_probs": [-500.3878173828125, -496.9914855957031, -497.97955322265625, -501.88330078125], "reading_logprob_std": 2.230781905417009, "tokenization_path_logprob_std": 137.2289239868633, "trajectory_balance_rmse": 9484.360979976536} {"canonical_length": 265, "canonical_nll": 1.7090728183962265, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.4392156862745098, "prompt_id": "3b670d936478ce43", "reading_canonical_log_prob": -452.904296875, "reading_distances": [31, 26, 30, 29], "reading_log_probs": [-456.4993896484375, -456.39044189453125, -458.79278564453125, -452.64556884765625], "reading_logprob_std": 2.54471795343923, "tokenization_path_logprob_std": 471.2910970496348, "trajectory_balance_rmse": 8660.991369039708} {"canonical_length": 78, "canonical_nll": 2.439855722280649, "generation_rouge_l_f1": 0.17857142857142855, "generation_token_f1": 0.26785714285714285, "prompt_id": "d940482038c5842a", "reading_canonical_log_prob": -190.30874633789062, "reading_distances": [20, 30, 28, 26], "reading_log_probs": [-190.46786499023438, -194.251708984375, -188.46620178222656, -188.02737426757812], "reading_logprob_std": 2.83853148326889, "tokenization_path_logprob_std": 295.7780593895706, "trajectory_balance_rmse": 2498.1376428865506} {"canonical_length": 291, "canonical_nll": 1.0223000647685783, "generation_rouge_l_f1": 0.26713532513181015, "generation_token_f1": 0.492091388400703, "prompt_id": "07d10dd03fee840c", "reading_canonical_log_prob": -297.48931884765625, "reading_distances": [26, 27, 19, 24], "reading_log_probs": [-305.1890563964844, -306.0377502441406, -300.94659423828125, -299.83807373046875], "reading_logprob_std": 3.067795869584115, "tokenization_path_logprob_std": 515.304057662882, "trajectory_balance_rmse": 7946.309118051637} {"canonical_length": 4, "canonical_nll": 3.3054556846618652, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "7ef3a945dcbdd086", "reading_canonical_log_prob": -13.221822738647461, "reading_distances": [21, 23, 17, 24], "reading_log_probs": [-13.887290000915527, -13.502843856811523, -13.825895309448242, -13.4866304397583], "reading_logprob_std": 0.21051959480304716, "tokenization_path_logprob_std": 21.483425290123378, "trajectory_balance_rmse": 62.76688245744207} {"canonical_length": 156, "canonical_nll": 1.1315561930338542, "generation_rouge_l_f1": 0.28641975308641976, "generation_token_f1": 0.43950617283950616, "prompt_id": "ef63cfbdbb1633a2", "reading_canonical_log_prob": -176.52276611328125, "reading_distances": [25, 27, 34, 30], "reading_log_probs": [-182.086181640625, -176.1637420654297, -180.71653747558594, -181.27224731445312], "reading_logprob_std": 2.6574939651709415, "tokenization_path_logprob_std": 296.66838362336523, "trajectory_balance_rmse": 3889.8948379833655} {"canonical_length": 152, "canonical_nll": 1.6007090116802014, "generation_rouge_l_f1": 0.22377622377622378, "generation_token_f1": 0.4009324009324009, "prompt_id": "1d1550150b11ce18", "reading_canonical_log_prob": -243.30776977539062, "reading_distances": [24, 30, 29, 27], "reading_log_probs": [-244.98489379882812, -245.91078186035156, -243.22555541992188, -242.88088989257812], "reading_logprob_std": 1.440163572017058, "tokenization_path_logprob_std": 329.66210136375975, "trajectory_balance_rmse": 5474.804523406552} {"canonical_length": 436, "canonical_nll": 0.9250557575750789, "generation_rouge_l_f1": 0.2866894197952219, "generation_token_f1": 0.41979522184300344, "prompt_id": "becce7dcacc67b08", "reading_canonical_log_prob": -403.3243103027344, "reading_distances": [22, 25, 20, 29], "reading_log_probs": [-410.9693908691406, -407.2158508300781, -404.80328369140625, -406.6345520019531], "reading_logprob_std": 2.5886402593514903, "tokenization_path_logprob_std": 59.98892761161782, "trajectory_balance_rmse": 7567.186247059396} {"canonical_length": 289, "canonical_nll": 1.2188857979427985, "generation_rouge_l_f1": 0.14393939393939392, "generation_token_f1": 0.30303030303030304, "prompt_id": "11042f17f722f10a", "reading_canonical_log_prob": -352.25799560546875, "reading_distances": [23, 29, 28, 26], "reading_log_probs": [-352.50982666015625, -357.25531005859375, -355.4390869140625, -353.56939697265625], "reading_logprob_std": 2.0937054439419156, "tokenization_path_logprob_std": 308.11255934891886, "trajectory_balance_rmse": 8650.561772082458} {"canonical_length": 31, "canonical_nll": 1.3474977554813508, "generation_rouge_l_f1": 0.13513513513513514, "generation_token_f1": 0.16216216216216217, "prompt_id": "b29cd52aed559d3c", "reading_canonical_log_prob": -41.772430419921875, "reading_distances": [21, 19, 19, 18], "reading_log_probs": [-41.368438720703125, -40.796058654785156, -42.936920166015625, -41.94683837890625], "reading_logprob_std": 0.9133337778728203, "tokenization_path_logprob_std": 70.43486443091369, "trajectory_balance_rmse": 675.6050759406252} {"canonical_length": 192, "canonical_nll": 0.9843354225158691, "generation_rouge_l_f1": 0.26008968609865474, "generation_token_f1": 0.4753363228699552, "prompt_id": "3d6b176b83ba1bb0", "reading_canonical_log_prob": -188.99240112304688, "reading_distances": [20, 22, 20, 16], "reading_log_probs": [-196.29441833496094, -195.67050170898438, -194.2567138671875, -192.9864501953125], "reading_logprob_std": 1.480446527595229, "tokenization_path_logprob_std": 111.73017543478643, "trajectory_balance_rmse": 2999.807116499311} {"canonical_length": 146, "canonical_nll": 2.0109085710081334, "generation_rouge_l_f1": 0.256578947368421, "generation_token_f1": 0.47368421052631576, "prompt_id": "ed74cc90058602c2", "reading_canonical_log_prob": -293.5926513671875, "reading_distances": [28, 29, 22, 30], "reading_log_probs": [-295.1475830078125, -295.8374328613281, -295.7591247558594, -295.4111022949219], "reading_logprob_std": 0.3199385473844629, "tokenization_path_logprob_std": 378.1161060616375, "trajectory_balance_rmse": 4768.943964527302} {"canonical_length": 217, "canonical_nll": 1.719277799404162, "generation_rouge_l_f1": 0.23651452282157676, "generation_token_f1": 0.4979253112033195, "prompt_id": "b2e99fb7bff1b5b7", "reading_canonical_log_prob": -373.0832824707031, "reading_distances": [26, 37, 39, 36], "reading_log_probs": [-374.45977783203125, -372.55340576171875, -375.08843994140625, -368.5781555175781], "reading_logprob_std": 2.9330838100732732, "tokenization_path_logprob_std": 337.17105605833274, "trajectory_balance_rmse": 6687.726095790776} {"canonical_length": 318, "canonical_nll": 0.8098637802795794, "generation_rouge_l_f1": 0.2704507512520868, "generation_token_f1": 0.5442404006677797, "prompt_id": "5a32b0efbf18f769", "reading_canonical_log_prob": -257.53668212890625, "reading_distances": [26, 29, 29, 25], "reading_log_probs": [-253.05029296875, -254.3463897705078, -254.12857055664062, -250.6560516357422], "reading_logprob_std": 1.6906450803265864, "tokenization_path_logprob_std": 249.6466258842845, "trajectory_balance_rmse": 7284.032675882697} {"canonical_length": 25, "canonical_nll": 0.5884130096435547, "generation_rouge_l_f1": 0.09691629955947137, "generation_token_f1": 0.1145374449339207, "prompt_id": "3697c04353d3a1f1", "reading_canonical_log_prob": -14.710325241088867, "reading_distances": [29, 29, 32, 26], "reading_log_probs": [-26.373634338378906, -18.113685607910156, -20.73302459716797, -26.551532745361328], "reading_logprob_std": 4.203055607254228, "tokenization_path_logprob_std": 64.42983294212851, "trajectory_balance_rmse": 334.3009396116412} {"canonical_length": 83, "canonical_nll": 2.0115885906908884, "generation_rouge_l_f1": 0.3622047244094488, "generation_token_f1": 0.5511811023622047, "prompt_id": "4983a8fb02a849d4", "reading_canonical_log_prob": -166.96185302734375, "reading_distances": [18, 19, 13, 10], "reading_log_probs": [-168.94451904296875, -167.1266326904297, -168.0951385498047, -168.51336669921875], "reading_logprob_std": 0.7771743665449602, "tokenization_path_logprob_std": 184.97783051357038, "trajectory_balance_rmse": 1993.9523401927013} {"canonical_length": 244, "canonical_nll": 1.8526683869909069, "generation_rouge_l_f1": 0.20039292730844793, "generation_token_f1": 0.44793713163064836, "prompt_id": "fe89f27185bbc3b2", "reading_canonical_log_prob": -452.05108642578125, "reading_distances": [43, 39, 29, 41], "reading_log_probs": [-452.3775634765625, -450.1120300292969, -449.035888671875, -450.8381652832031], "reading_logprob_std": 1.40246012155249, "tokenization_path_logprob_std": 78.4835172521902, "trajectory_balance_rmse": 8587.604061734337} {"canonical_length": 87, "canonical_nll": 1.7576039632161458, "generation_rouge_l_f1": 0.17424242424242423, "generation_token_f1": 0.303030303030303, "prompt_id": "6b767f6a353fd065", "reading_canonical_log_prob": -152.9115447998047, "reading_distances": [20, 17, 23, 18], "reading_log_probs": [-156.279052734375, -154.19918823242188, -156.61973571777344, -161.6242218017578], "reading_logprob_std": 3.149706179418019, "tokenization_path_logprob_std": 160.98282391190824, "trajectory_balance_rmse": 2226.3424709058877} {"canonical_length": 136, "canonical_nll": 1.553294350119198, "generation_rouge_l_f1": 0.241042345276873, "generation_token_f1": 0.5146579804560261, "prompt_id": "e77b6aefa4b6596d", "reading_canonical_log_prob": -211.24803161621094, "reading_distances": [31, 25, 29, 28], "reading_log_probs": [-210.7879638671875, -208.53204345703125, -206.42311096191406, -206.70095825195312], "reading_logprob_std": 2.014993819146643, "tokenization_path_logprob_std": 185.13525853661568, "trajectory_balance_rmse": 3348.3776405343488} {"canonical_length": 9, "canonical_nll": 2.676209979587131, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.75, "prompt_id": "383239282305a169", "reading_canonical_log_prob": -24.08588981628418, "reading_distances": [31, 34, 35, 32], "reading_log_probs": [-24.520971298217773, -23.676067352294922, -25.228618621826172, -23.189651489257812], "reading_logprob_std": 0.9033014746937384, "tokenization_path_logprob_std": 26.557800849270176, "trajectory_balance_rmse": 178.9637154726581} {"canonical_length": 10, "canonical_nll": 0.24541890621185303, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.8571428571428571, "prompt_id": "29653e40ee0ea781", "reading_canonical_log_prob": -2.4541890621185303, "reading_distances": [30, 24, 20, 25], "reading_log_probs": [-1.890744924545288, -2.4453580379486084, -5.919450283050537, -3.4484410285949707], "reading_logprob_std": 1.782925004802171, "tokenization_path_logprob_std": 34.458371753993646, "trajectory_balance_rmse": 188.40565394005253} {"canonical_length": 251, "canonical_nll": 1.735769203459599, "generation_rouge_l_f1": 0.20647773279352227, "generation_token_f1": 0.4412955465587044, "prompt_id": "b8e0545bd0cc139e", "reading_canonical_log_prob": -435.6780700683594, "reading_distances": [48, 52, 34, 45], "reading_log_probs": [-435.2398986816406, -436.114501953125, -436.4029235839844, -436.25640869140625], "reading_logprob_std": 0.5224654357930417, "tokenization_path_logprob_std": 123.05706472972366, "trajectory_balance_rmse": 7944.846277979179} {"canonical_length": 430, "canonical_nll": 1.2705455691315406, "generation_rouge_l_f1": 0.20625000000000002, "generation_token_f1": 0.35000000000000003, "prompt_id": "a7d20d598bd1e5dd", "reading_canonical_log_prob": -546.3345947265625, "reading_distances": [36, 35, 38, 29], "reading_log_probs": [-632.9423828125, -625.146484375, -636.0604248046875, -650.5675659179688], "reading_logprob_std": 10.633860711628717, "tokenization_path_logprob_std": 333.04057688880584, "trajectory_balance_rmse": 9845.017634698865} {"canonical_length": 47, "canonical_nll": 3.7514414685837765, "generation_rouge_l_f1": 0.10135135135135134, "generation_token_f1": 0.16891891891891891, "prompt_id": "ecbe9c8ff4f546fd", "reading_canonical_log_prob": -176.3177490234375, "reading_distances": [12, 14, 10, 13], "reading_log_probs": [-186.24209594726562, -160.82647705078125, -165.89605712890625, -168.62881469726562], "reading_logprob_std": 11.046069190466264, "tokenization_path_logprob_std": 107.02430112011083, "trajectory_balance_rmse": 1103.4023484910376} {"canonical_length": 12, "canonical_nll": 1.1085563500722249, "generation_rouge_l_f1": 0.5263157894736842, "generation_token_f1": 0.5263157894736842, "prompt_id": "641a36ce0ce7d08c", "reading_canonical_log_prob": -13.3026762008667, "reading_distances": [25, 23, 26, 29], "reading_log_probs": [-17.23637580871582, -19.215606689453125, -21.147260665893555, -18.781700134277344], "reading_logprob_std": 1.6102753679780326, "tokenization_path_logprob_std": 72.2507035869124, "trajectory_balance_rmse": 423.99535926583366} {"canonical_length": 105, "canonical_nll": 1.906858898344494, "generation_rouge_l_f1": 0.23966942148760328, "generation_token_f1": 0.3884297520661157, "prompt_id": "32d37daf5bb72f67", "reading_canonical_log_prob": -200.22018432617188, "reading_distances": [19, 20, 24, 20], "reading_log_probs": [-198.97535705566406, -201.2123260498047, -199.0021514892578, -199.8232421875], "reading_logprob_std": 1.0492963085861517, "tokenization_path_logprob_std": 139.83551667482573, "trajectory_balance_rmse": 3321.5995600151305} {"canonical_length": 19, "canonical_nll": 2.6809365121941817, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.28571428571428575, "prompt_id": "b21a0da1ea5ee3b7", "reading_canonical_log_prob": -50.93779373168945, "reading_distances": [20, 26, 23, 24], "reading_log_probs": [-50.82268524169922, -57.79888153076172, -51.145469665527344, -48.156005859375], "reading_logprob_std": 4.103580935054029, "tokenization_path_logprob_std": 74.16092301721284, "trajectory_balance_rmse": 538.9702422590486} {"canonical_length": 281, "canonical_nll": 1.5847975978647686, "generation_rouge_l_f1": 0.20817843866171, "generation_token_f1": 0.4646840148698885, "prompt_id": "11a80e6e30ed7f8f", "reading_canonical_log_prob": -445.328125, "reading_distances": [19, 17, 19, 21], "reading_log_probs": [-443.5399169921875, -445.33660888671875, -447.2111511230469, -445.58331298828125], "reading_logprob_std": 1.5029459268055323, "tokenization_path_logprob_std": 127.45226776257526, "trajectory_balance_rmse": 8516.646146477276} {"canonical_length": 119, "canonical_nll": 2.0717138723165047, "generation_rouge_l_f1": 0.22454308093994776, "generation_token_f1": 0.4073107049608355, "prompt_id": "e2dff42d7f8d6988", "reading_canonical_log_prob": -246.53395080566406, "reading_distances": [23, 21, 26, 21], "reading_log_probs": [-248.97216796875, -244.74024963378906, -245.78118896484375, -246.13519287109375], "reading_logprob_std": 1.8095728684266372, "tokenization_path_logprob_std": 229.6073505175535, "trajectory_balance_rmse": 3525.0894461171115} {"canonical_length": 338, "canonical_nll": 1.7735144259661613, "generation_rouge_l_f1": 0.19047619047619047, "generation_token_f1": 0.39506172839506176, "prompt_id": "bd63a203fd3cb2b2", "reading_canonical_log_prob": -599.4478759765625, "reading_distances": [29, 29, 31, 30], "reading_log_probs": [-593.8118896484375, -594.7446899414062, -591.2733764648438, -592.7337646484375], "reading_logprob_std": 1.4917297152051652, "tokenization_path_logprob_std": 242.6605023280268, "trajectory_balance_rmse": 7624.876237667277} {"canonical_length": 17, "canonical_nll": 0.5548389098223518, "generation_rouge_l_f1": 0.12403100775193798, "generation_token_f1": 0.12403100775193798, "prompt_id": "f5560d8544cd62d8", "reading_canonical_log_prob": -9.43226146697998, "reading_distances": [25, 21, 22, 18], "reading_log_probs": [-8.77860164642334, -9.49732494354248, -8.599262237548828, -8.99456787109375], "reading_logprob_std": 0.3884708325925587, "tokenization_path_logprob_std": 57.47564352481375, "trajectory_balance_rmse": 542.0541792659791} {"canonical_length": 143, "canonical_nll": 1.5431602851494208, "generation_rouge_l_f1": 0.2303921568627451, "generation_token_f1": 0.5, "prompt_id": "d5ad06fe63008db3", "reading_canonical_log_prob": -220.6719207763672, "reading_distances": [21, 24, 21, 26], "reading_log_probs": [-222.249755859375, -223.1080322265625, -223.20156860351562, -221.77835083007812], "reading_logprob_std": 0.6872141238741882, "tokenization_path_logprob_std": 371.3995463275156, "trajectory_balance_rmse": 4657.2183279907695} {"canonical_length": 68, "canonical_nll": 1.3334601907169117, "generation_rouge_l_f1": 0.182370820668693, "generation_token_f1": 0.243161094224924, "prompt_id": "32cf7dbccdbc97bc", "reading_canonical_log_prob": -90.67529296875, "reading_distances": [16, 20, 16, 23], "reading_log_probs": [-91.86520385742188, -88.23741912841797, -87.15010070800781, -93.20365142822266], "reading_logprob_std": 2.8820661363940383, "tokenization_path_logprob_std": 29.759860477211646, "trajectory_balance_rmse": 2458.155005160457} {"canonical_length": 47, "canonical_nll": 1.902157722635472, "generation_rouge_l_f1": 0.11805555555555554, "generation_token_f1": 0.15972222222222224, "prompt_id": "c180433ec97360b0", "reading_canonical_log_prob": -89.40141296386719, "reading_distances": [32, 24, 28, 36], "reading_log_probs": [-91.13951110839844, -91.1876220703125, -90.5046615600586, -91.98893737792969], "reading_logprob_std": 0.6081727814636084, "tokenization_path_logprob_std": 26.601366457891302, "trajectory_balance_rmse": 1181.0502585410277} {"canonical_length": 122, "canonical_nll": 0.9790184146068135, "generation_rouge_l_f1": 0.3082437275985663, "generation_token_f1": 0.4444444444444444, "prompt_id": "aaf9ee6d37011f66", "reading_canonical_log_prob": -119.44024658203125, "reading_distances": [35, 34, 46, 32], "reading_log_probs": [-133.2849578857422, -133.6953125, -134.0884552001953, -139.48619079589844], "reading_logprob_std": 2.91681433861679, "tokenization_path_logprob_std": 256.34545982864375, "trajectory_balance_rmse": 2787.5799498535107} {"canonical_length": 19, "canonical_nll": 0.20795061713770815, "generation_rouge_l_f1": 0.0772200772200772, "generation_token_f1": 0.10038610038610038, "prompt_id": "f459f20b3852c4a3", "reading_canonical_log_prob": -3.951061725616455, "reading_distances": [15, 21, 20, 22], "reading_log_probs": [-1.723508358001709, -1.3035509586334229, -2.8012025356292725, -2.0469629764556885], "reading_logprob_std": 0.63291065950093, "tokenization_path_logprob_std": 104.12875743733207, "trajectory_balance_rmse": 364.28630029778975} {"canonical_length": 19, "canonical_nll": 1.5921457190262644, "generation_rouge_l_f1": 0.3243243243243243, "generation_token_f1": 0.3243243243243243, "prompt_id": "a05d7928d7de0d6c", "reading_canonical_log_prob": -30.250768661499023, "reading_distances": [26, 28, 28, 20], "reading_log_probs": [-36.812660217285156, -34.234493255615234, -40.85084533691406, -35.7148551940918], "reading_logprob_std": 2.835857320273009, "tokenization_path_logprob_std": 37.306230293680024, "trajectory_balance_rmse": 525.612547516032} {"canonical_length": 15, "canonical_nll": 2.0820847829182942, "generation_rouge_l_f1": 0.8181818181818181, "generation_token_f1": 0.8181818181818181, "prompt_id": "2c22dccbca7a08c7", "reading_canonical_log_prob": -31.231271743774414, "reading_distances": [26, 28, 26, 28], "reading_log_probs": [-33.324161529541016, -28.8726749420166, -31.871976852416992, -32.885398864746094], "reading_logprob_std": 2.005032735731341, "tokenization_path_logprob_std": 60.45123743790078, "trajectory_balance_rmse": 526.8076395482735} {"canonical_length": 166, "canonical_nll": 1.9402642307511295, "generation_rouge_l_f1": 0.30594900849858353, "generation_token_f1": 0.49858356940509907, "prompt_id": "6e818f0ad7674362", "reading_canonical_log_prob": -322.0838623046875, "reading_distances": [25, 24, 28, 23], "reading_log_probs": [-321.65545654296875, -319.1225891113281, -318.5035400390625, -316.56219482421875], "reading_logprob_std": 2.1015663188853244, "tokenization_path_logprob_std": 339.04240155923685, "trajectory_balance_rmse": 4331.175801266959} {"canonical_length": 70, "canonical_nll": 1.2140599931989398, "generation_rouge_l_f1": 0.33488372093023255, "generation_token_f1": 0.5023255813953489, "prompt_id": "8ebd94ecdb15c9f4", "reading_canonical_log_prob": -84.98419952392578, "reading_distances": [22, 16, 20, 23], "reading_log_probs": [-82.73712158203125, -82.44970703125, -84.571533203125, -83.86375427246094], "reading_logprob_std": 0.9882476603172982, "tokenization_path_logprob_std": 111.45528839901694, "trajectory_balance_rmse": 2427.3267871570224} {"canonical_length": 3, "canonical_nll": 0.08052436014016469, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "2148a4bb4107090b", "reading_canonical_log_prob": -0.24157308042049408, "reading_distances": [15, 22, 21, 16], "reading_log_probs": [-0.30058562755584717, -0.18929632008075714, -0.30875489115715027, -0.4293152391910553], "reading_logprob_std": 0.09808057444025721, "tokenization_path_logprob_std": 26.6881842463468, "trajectory_balance_rmse": 64.66796087312449} {"canonical_length": 368, "canonical_nll": 1.0188870637313179, "generation_rouge_l_f1": 0.19405320813771518, "generation_token_f1": 0.42879499217527384, "prompt_id": "5d2176dd2ade1ce9", "reading_canonical_log_prob": -374.950439453125, "reading_distances": [25, 27, 20, 22], "reading_log_probs": [-369.89117431640625, -367.9089050292969, -370.05377197265625, -369.10955810546875], "reading_logprob_std": 0.9789584289208159, "tokenization_path_logprob_std": 384.87326826592033, "trajectory_balance_rmse": 9892.83895234217} {"canonical_length": 12, "canonical_nll": 1.151366392771403, "generation_rouge_l_f1": 0.07751937984496124, "generation_token_f1": 0.07751937984496124, "prompt_id": "9d6ee475e283101d", "reading_canonical_log_prob": -13.816396713256836, "reading_distances": [30, 28, 33, 33], "reading_log_probs": [-12.732880592346191, -12.50379467010498, -13.044198989868164, -13.073148727416992], "reading_logprob_std": 0.27114277709505447, "tokenization_path_logprob_std": 44.45968191471248, "trajectory_balance_rmse": 413.54844219259905} {"canonical_length": 12, "canonical_nll": 0.37487781047821045, "generation_rouge_l_f1": 0.75, "generation_token_f1": 0.75, "prompt_id": "a41c0de97fa957ce", "reading_canonical_log_prob": -4.498533725738525, "reading_distances": [20, 17, 24, 23], "reading_log_probs": [-5.820149898529053, -5.485653877258301, -5.951563835144043, -5.4354047775268555], "reading_logprob_std": 0.25219094167561235, "tokenization_path_logprob_std": 13.653856608354182, "trajectory_balance_rmse": 221.81949045279742} {"canonical_length": 9, "canonical_nll": 0.0591153966055976, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "916ba9265d9ad9c6", "reading_canonical_log_prob": -0.5320385694503784, "reading_distances": [28, 36, 39, 32], "reading_log_probs": [-0.2886458933353424, -0.28116706013679504, -0.21895605325698853, -0.22123511135578156], "reading_logprob_std": 0.03755447404633385, "tokenization_path_logprob_std": 25.205390952096362, "trajectory_balance_rmse": 245.10753495588546} {"canonical_length": 99, "canonical_nll": 0.9138649063880997, "generation_rouge_l_f1": 0.4371584699453551, "generation_token_f1": 0.6666666666666666, "prompt_id": "5baddab4b656da75", "reading_canonical_log_prob": -90.47262573242188, "reading_distances": [20, 25, 26, 30], "reading_log_probs": [-98.69290924072266, -98.12792205810547, -101.40213012695312, -101.84288787841797], "reading_logprob_std": 1.8774346563089641, "tokenization_path_logprob_std": 179.55909776739637, "trajectory_balance_rmse": 2515.8475048526466} {"canonical_length": 54, "canonical_nll": 0.765057952315719, "generation_rouge_l_f1": 0.288135593220339, "generation_token_f1": 0.2994350282485876, "prompt_id": "22a09297862230a9", "reading_canonical_log_prob": -41.31312942504883, "reading_distances": [31, 27, 27, 34], "reading_log_probs": [-44.84749984741211, -48.85493850708008, -44.559669494628906, -46.36651611328125], "reading_logprob_std": 1.9654536783649108, "tokenization_path_logprob_std": 94.98794324536185, "trajectory_balance_rmse": 1111.0332992248848} {"canonical_length": 40, "canonical_nll": 0.18543335199356079, "generation_rouge_l_f1": 0.43902439024390244, "generation_token_f1": 0.43902439024390244, "prompt_id": "c6508f3625213115", "reading_canonical_log_prob": -7.417334079742432, "reading_distances": [10, 15, 15, 12], "reading_log_probs": [-6.278135776519775, -7.9395928382873535, -7.778934001922607, -9.318768501281738], "reading_logprob_std": 1.2435603893953597, "tokenization_path_logprob_std": 61.28467639596381, "trajectory_balance_rmse": 352.0958749880674} {"canonical_length": 272, "canonical_nll": 1.6473844191607308, "generation_rouge_l_f1": 0.27151051625239003, "generation_token_f1": 0.5086042065009561, "prompt_id": "a4b74ba3880275a6", "reading_canonical_log_prob": -448.08856201171875, "reading_distances": [20, 15, 19, 14], "reading_log_probs": [-452.31658935546875, -454.97314453125, -459.0230407714844, -451.0003356933594], "reading_logprob_std": 3.5392438704006577, "tokenization_path_logprob_std": 245.28824091154675, "trajectory_balance_rmse": 8128.563406858081} {"canonical_length": 17, "canonical_nll": 0.5325914270737592, "generation_rouge_l_f1": 0.20689655172413793, "generation_token_f1": 0.20689655172413793, "prompt_id": "5f38035dc4f5627c", "reading_canonical_log_prob": -9.054054260253906, "reading_distances": [20, 17, 20, 22], "reading_log_probs": [-9.89117431640625, -9.630431175231934, -11.149369239807129, -10.52027416229248], "reading_logprob_std": 0.6795563546182607, "tokenization_path_logprob_std": 44.54100903751316, "trajectory_balance_rmse": 599.5683610710051} {"canonical_length": 34, "canonical_nll": 0.8983030880198759, "generation_rouge_l_f1": 0.23577235772357721, "generation_token_f1": 0.25203252032520324, "prompt_id": "0e027c7a776ba5aa", "reading_canonical_log_prob": -30.54230499267578, "reading_distances": [18, 25, 24, 27], "reading_log_probs": [-34.5955810546875, -37.805030822753906, -40.560813903808594, -36.27117156982422], "reading_logprob_std": 2.533776336543395, "tokenization_path_logprob_std": 108.33552555905597, "trajectory_balance_rmse": 740.4783865992875} {"canonical_length": 23, "canonical_nll": 2.306236930515455, "generation_rouge_l_f1": 0.2777777777777778, "generation_token_f1": 0.5, "prompt_id": "9d315888010aca18", "reading_canonical_log_prob": -53.04344940185547, "reading_distances": [19, 24, 26, 19], "reading_log_probs": [-51.253231048583984, -56.013427734375, -50.831478118896484, -50.87826919555664], "reading_logprob_std": 2.519963439770399, "tokenization_path_logprob_std": 79.60399473442367, "trajectory_balance_rmse": 529.5010623628965} {"canonical_length": 390, "canonical_nll": 1.069138433994391, "generation_rouge_l_f1": 0.26249999999999996, "generation_token_f1": 0.478125, "prompt_id": "b3017655183fdef3", "reading_canonical_log_prob": -416.9639892578125, "reading_distances": [23, 22, 20, 23], "reading_log_probs": [-417.58270263671875, -416.20703125, -417.6279602050781, -418.0130615234375], "reading_logprob_std": 0.7910339358745703, "tokenization_path_logprob_std": 262.8886002180054, "trajectory_balance_rmse": 10306.773164868304} {"canonical_length": 24, "canonical_nll": 1.2554266452789307, "generation_rouge_l_f1": 0.4736842105263159, "generation_token_f1": 0.4736842105263159, "prompt_id": "70f7bc293fb9b6e2", "reading_canonical_log_prob": -30.130239486694336, "reading_distances": [36, 38, 32, 32], "reading_log_probs": [-38.287593841552734, -36.121803283691406, -40.6838264465332, -24.324960708618164], "reading_logprob_std": 7.262791828635364, "tokenization_path_logprob_std": 67.56003351131059, "trajectory_balance_rmse": 615.9297038004731} {"canonical_length": 85, "canonical_nll": 1.5387831744025735, "generation_rouge_l_f1": 0.1527777777777778, "generation_token_f1": 0.20833333333333334, "prompt_id": "6a1786b00d1d1907", "reading_canonical_log_prob": -130.79656982421875, "reading_distances": [30, 26, 28, 25], "reading_log_probs": [-124.61785888671875, -123.87496948242188, -121.45733642578125, -123.8885269165039], "reading_logprob_std": 1.3792670945906111, "tokenization_path_logprob_std": 130.4796696407849, "trajectory_balance_rmse": 1637.1573577857498} {"canonical_length": 108, "canonical_nll": 1.7462717692057292, "generation_rouge_l_f1": 0.18289085545722714, "generation_token_f1": 0.32448377581120946, "prompt_id": "7988aedbe67fdb66", "reading_canonical_log_prob": -188.59735107421875, "reading_distances": [15, 15, 17, 16], "reading_log_probs": [-184.8016357421875, -190.62713623046875, -187.45831298828125, -185.70916748046875], "reading_logprob_std": 2.5675112975825773, "tokenization_path_logprob_std": 50.826771878105745, "trajectory_balance_rmse": 3025.0494804372593} {"canonical_length": 98, "canonical_nll": 2.6438579948580996, "generation_rouge_l_f1": 0.16292134831460675, "generation_token_f1": 0.25842696629213485, "prompt_id": "0b1fe757c0710416", "reading_canonical_log_prob": -259.09808349609375, "reading_distances": [12, 14, 16, 18], "reading_log_probs": [-256.3865051269531, -252.63336181640625, -252.1403045654297, -251.2147216796875], "reading_logprob_std": 2.2725902422900655, "tokenization_path_logprob_std": 191.30717007786077, "trajectory_balance_rmse": 2876.9567289927736} {"canonical_length": 13, "canonical_nll": 1.2415208082932692, "generation_rouge_l_f1": 0.054421768707482984, "generation_token_f1": 0.054421768707482984, "prompt_id": "78fa2b118c48dd30", "reading_canonical_log_prob": -16.1397705078125, "reading_distances": [22, 33, 29, 27], "reading_log_probs": [-20.044662475585938, -18.148895263671875, -18.908103942871094, -18.85250473022461], "reading_logprob_std": 0.7842987357251751, "tokenization_path_logprob_std": 37.33233455019194, "trajectory_balance_rmse": 289.4614390029197} {"canonical_length": 424, "canonical_nll": 1.2038955688476562, "generation_rouge_l_f1": 0.1867088607594937, "generation_token_f1": 0.449367088607595, "prompt_id": "995817411c15ddee", "reading_canonical_log_prob": -510.45172119140625, "reading_distances": [18, 16, 10, 10], "reading_log_probs": [-512.5947875976562, -513.5443725585938, -514.267822265625, -514.84619140625], "reading_logprob_std": 0.9713477904934622, "tokenization_path_logprob_std": 305.7572568554434, "trajectory_balance_rmse": 10315.31887119873} {"canonical_length": 64, "canonical_nll": 1.4343147277832031, "generation_rouge_l_f1": 0.1898305084745763, "generation_token_f1": 0.3322033898305085, "prompt_id": "fd32715308ec2a93", "reading_canonical_log_prob": -91.796142578125, "reading_distances": [37, 47, 42, 46], "reading_log_probs": [-97.19049835205078, -100.12593078613281, -99.55134582519531, -99.4415283203125], "reading_logprob_std": 1.2931909752449942, "tokenization_path_logprob_std": 254.795349955169, "trajectory_balance_rmse": 2124.103893568534} {"canonical_length": 333, "canonical_nll": 1.2689625049854543, "generation_rouge_l_f1": 0.1784386617100372, "generation_token_f1": 0.36802973977695164, "prompt_id": "6c1da32736122296", "reading_canonical_log_prob": -422.56451416015625, "reading_distances": [35, 35, 36, 41], "reading_log_probs": [-424.1241149902344, -422.67376708984375, -419.66064453125, -424.8663330078125], "reading_logprob_std": 2.301492460136313, "tokenization_path_logprob_std": 222.21046130023797, "trajectory_balance_rmse": 6914.442391241098} {"canonical_length": 53, "canonical_nll": 1.167872518863318, "generation_rouge_l_f1": 0.2771739130434782, "generation_token_f1": 0.3206521739130435, "prompt_id": "ffb0593302e5d193", "reading_canonical_log_prob": -61.89724349975586, "reading_distances": [40, 37, 35, 40], "reading_log_probs": [-68.5721435546875, -69.40448760986328, -69.62345886230469, -66.04182434082031], "reading_logprob_std": 1.642765309423343, "tokenization_path_logprob_std": 118.07146858720806, "trajectory_balance_rmse": 1180.9046656775008} {"canonical_length": 142, "canonical_nll": 1.4819568043023768, "generation_rouge_l_f1": 0.2686567164179105, "generation_token_f1": 0.41293532338308453, "prompt_id": "780311bcf262bb9d", "reading_canonical_log_prob": -210.4378662109375, "reading_distances": [35, 31, 31, 37], "reading_log_probs": [-224.04632568359375, -224.0973358154297, -216.59371948242188, -218.99386596679688], "reading_logprob_std": 3.7547931160106773, "tokenization_path_logprob_std": 211.97801611437345, "trajectory_balance_rmse": 4571.589534142584} {"canonical_length": 32, "canonical_nll": 1.8218086957931519, "generation_rouge_l_f1": 0.3943661971830986, "generation_token_f1": 0.4507042253521127, "prompt_id": "ddc773449a2312dd", "reading_canonical_log_prob": -58.29787826538086, "reading_distances": [34, 28, 33, 31], "reading_log_probs": [-60.84368896484375, -58.29002380371094, -59.04997634887695, -59.650169372558594], "reading_logprob_std": 1.0782259822685123, "tokenization_path_logprob_std": 139.31086953023842, "trajectory_balance_rmse": 1140.887508589381} {"canonical_length": 259, "canonical_nll": 0.7635804990083555, "generation_rouge_l_f1": 0.40551181102362205, "generation_token_f1": 0.6141732283464567, "prompt_id": "fa8dcc0f054c9984", "reading_canonical_log_prob": -197.76734924316406, "reading_distances": [21, 28, 27, 21], "reading_log_probs": [-193.7538604736328, -197.22467041015625, -193.51368713378906, -193.96144104003906], "reading_logprob_std": 1.7504247317966755, "tokenization_path_logprob_std": 260.1640349877694, "trajectory_balance_rmse": 4065.6903306273484} {"canonical_length": 79, "canonical_nll": 1.1799680492545985, "generation_rouge_l_f1": 0.16363636363636364, "generation_token_f1": 0.296969696969697, "prompt_id": "2d9359e9dd847731", "reading_canonical_log_prob": -93.21747589111328, "reading_distances": [21, 13, 17, 16], "reading_log_probs": [-93.4471435546875, -96.28142547607422, -96.18191528320312, -96.55458068847656], "reading_logprob_std": 1.4546388873266376, "tokenization_path_logprob_std": 198.10851846151698, "trajectory_balance_rmse": 2319.084657861633} {"canonical_length": 91, "canonical_nll": 1.9278656676575379, "generation_rouge_l_f1": 0.17415730337078653, "generation_token_f1": 0.31460674157303375, "prompt_id": "5cb4ccbc96ab0eb7", "reading_canonical_log_prob": -175.43577575683594, "reading_distances": [27, 35, 32, 31], "reading_log_probs": [-174.01779174804688, -175.65347290039062, -174.5238800048828, -175.13623046875], "reading_logprob_std": 0.7130320388475687, "tokenization_path_logprob_std": 137.41512191015335, "trajectory_balance_rmse": 2831.322040127051} {"canonical_length": 76, "canonical_nll": 0.5187962180689761, "generation_rouge_l_f1": 0.3559322033898305, "generation_token_f1": 0.44067796610169485, "prompt_id": "23716ea232c0bbd4", "reading_canonical_log_prob": -39.42851257324219, "reading_distances": [4, 4, 2, 5], "reading_log_probs": [-40.630462646484375, -42.53456115722656, -42.727108001708984, -40.68994903564453], "reading_logprob_std": 1.1407138238151273, "tokenization_path_logprob_std": 54.72696787478457, "trajectory_balance_rmse": 1445.1534457759378} {"canonical_length": 7, "canonical_nll": 3.066666466849191, "generation_rouge_l_f1": 0.30769230769230765, "generation_token_f1": 0.30769230769230765, "prompt_id": "d26a8398a22a7919", "reading_canonical_log_prob": -21.466665267944336, "reading_distances": [39, 35, 40, 40], "reading_log_probs": [-23.441570281982422, -19.77726936340332, -20.35761070251465, -20.025623321533203], "reading_logprob_std": 1.7106356743869655, "tokenization_path_logprob_std": 32.74165118528281, "trajectory_balance_rmse": 253.87085553886172} {"canonical_length": 177, "canonical_nll": 2.0322044932909606, "generation_rouge_l_f1": 0.17090069284064663, "generation_token_f1": 0.3233256351039261, "prompt_id": "cf319e0486bfdfdc", "reading_canonical_log_prob": -359.7001953125, "reading_distances": [31, 27, 24, 18], "reading_log_probs": [-357.82574462890625, -357.8926696777344, -357.719482421875, -358.7660217285156], "reading_logprob_std": 0.4819987719031185, "tokenization_path_logprob_std": 272.05624879072576, "trajectory_balance_rmse": 6240.019436846859} {"canonical_length": 8, "canonical_nll": 2.0241763591766357, "generation_rouge_l_f1": 0.14285714285714285, "generation_token_f1": 0.14285714285714285, "prompt_id": "2145da07c0ca4001", "reading_canonical_log_prob": -16.193410873413086, "reading_distances": [22, 22, 21, 25], "reading_log_probs": [-19.15461540222168, -18.028900146484375, -17.467042922973633, -18.440290451049805], "reading_logprob_std": 0.7104888058460072, "tokenization_path_logprob_std": 32.12980979360573, "trajectory_balance_rmse": 290.1533602431686} {"canonical_length": 272, "canonical_nll": 1.3453438702751608, "generation_rouge_l_f1": 0.22556390977443608, "generation_token_f1": 0.4548872180451128, "prompt_id": "924cae405c1c753d", "reading_canonical_log_prob": -365.93353271484375, "reading_distances": [27, 22, 24, 21], "reading_log_probs": [-375.0506591796875, -373.0770263671875, -372.437255859375, -375.743896484375], "reading_logprob_std": 1.5721815019615837, "tokenization_path_logprob_std": 145.71260189568682, "trajectory_balance_rmse": 8141.409559274945} {"canonical_length": 147, "canonical_nll": 1.5348966429833653, "generation_rouge_l_f1": 0.22277227722772278, "generation_token_f1": 0.4405940594059406, "prompt_id": "eb989912613f5027", "reading_canonical_log_prob": -225.6298065185547, "reading_distances": [30, 30, 34, 36], "reading_log_probs": [-226.74264526367188, -229.0662384033203, -224.78553771972656, -227.6729736328125], "reading_logprob_std": 1.7957759953813235, "tokenization_path_logprob_std": 168.13069504022712, "trajectory_balance_rmse": 4313.570802660805} {"canonical_length": 7, "canonical_nll": 2.7529773712158203, "generation_rouge_l_f1": 0.20000000000000004, "generation_token_f1": 0.8000000000000002, "prompt_id": "1a61687e59394e25", "reading_canonical_log_prob": -19.270841598510742, "reading_distances": [19, 19, 20, 19], "reading_log_probs": [-22.27920150756836, -21.172073364257812, -21.846712112426758, -20.278656005859375], "reading_logprob_std": 0.8721187838243004, "tokenization_path_logprob_std": 20.17151633101696, "trajectory_balance_rmse": 100.53848355990772} {"canonical_length": 12, "canonical_nll": 1.9865719477335613, "generation_rouge_l_f1": 0.32, "generation_token_f1": 0.32, "prompt_id": "687a259eb6971a8a", "reading_canonical_log_prob": -23.838863372802734, "reading_distances": [22, 21, 20, 22], "reading_log_probs": [-25.8656005859375, -23.700502395629883, -23.34493064880371, -22.45797348022461], "reading_logprob_std": 1.4465577528765103, "tokenization_path_logprob_std": 42.37705800573739, "trajectory_balance_rmse": 309.3749076532473} {"canonical_length": 177, "canonical_nll": 1.487876762778072, "generation_rouge_l_f1": 0.25853658536585367, "generation_token_f1": 0.41951219512195126, "prompt_id": "ae453e9c5b7b5add", "reading_canonical_log_prob": -263.35418701171875, "reading_distances": [27, 28, 24, 25], "reading_log_probs": [-260.8603515625, -255.56472778320312, -257.5447692871094, -255.25637817382812], "reading_logprob_std": 2.5770398474447895, "tokenization_path_logprob_std": 228.63869456542463, "trajectory_balance_rmse": 4498.266960996204} {"canonical_length": 5, "canonical_nll": 4.257653045654297, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "ffdd99b122d4c9bc", "reading_canonical_log_prob": -21.288265228271484, "reading_distances": [42, 43, 43, 40], "reading_log_probs": [-15.787700653076172, -16.080585479736328, -17.58852195739746, -17.15322494506836], "reading_logprob_std": 0.8567042865927851, "tokenization_path_logprob_std": 39.91917594923226, "trajectory_balance_rmse": 96.40058769559974} {"canonical_length": 350, "canonical_nll": 1.0891448102678571, "generation_rouge_l_f1": 0.2596491228070175, "generation_token_f1": 0.5578947368421053, "prompt_id": "56f7657e17072dd3", "reading_canonical_log_prob": -381.20068359375, "reading_distances": [16, 12, 11, 15], "reading_log_probs": [-382.410400390625, -380.1054382324219, -379.66778564453125, -379.38543701171875], "reading_logprob_std": 1.377644485341908, "tokenization_path_logprob_std": 391.2648164578221, "trajectory_balance_rmse": 9642.895534777412} {"canonical_length": 172, "canonical_nll": 1.2630961218545602, "generation_rouge_l_f1": 0.29999999999999993, "generation_token_f1": 0.5533333333333332, "prompt_id": "f0c44899ac8f3454", "reading_canonical_log_prob": -217.25253295898438, "reading_distances": [26, 32, 27, 32], "reading_log_probs": [-214.52740478515625, -219.39378356933594, -217.93222045898438, -216.93807983398438], "reading_logprob_std": 2.046149423888235, "tokenization_path_logprob_std": 319.461230648123, "trajectory_balance_rmse": 5066.421768674041} {"canonical_length": 316, "canonical_nll": 1.2467326876483387, "generation_rouge_l_f1": 0.2060390763765542, "generation_token_f1": 0.47246891651865014, "prompt_id": "fb86a83720d2950d", "reading_canonical_log_prob": -393.967529296875, "reading_distances": [25, 22, 27, 26], "reading_log_probs": [-396.5528259277344, -393.9244384765625, -393.8325500488281, -392.7677001953125], "reading_logprob_std": 1.6102766430563036, "tokenization_path_logprob_std": 234.5582904641233, "trajectory_balance_rmse": 8039.413464739387} {"canonical_length": 60, "canonical_nll": 1.6965432484944662, "generation_rouge_l_f1": 0.1316614420062696, "generation_token_f1": 0.2068965517241379, "prompt_id": "4f164749bc77ca14", "reading_canonical_log_prob": -101.79259490966797, "reading_distances": [26, 25, 19, 23], "reading_log_probs": [-101.900146484375, -104.24234771728516, -103.82427978515625, -108.061767578125], "reading_logprob_std": 2.5799357508533958, "tokenization_path_logprob_std": 115.05415780566811, "trajectory_balance_rmse": 2113.797231232525} {"canonical_length": 161, "canonical_nll": 1.6181307016692548, "generation_rouge_l_f1": 0.23232323232323232, "generation_token_f1": 0.41414141414141414, "prompt_id": "8d15f97380395bd9", "reading_canonical_log_prob": -260.51904296875, "reading_distances": [26, 31, 27, 30], "reading_log_probs": [-262.372802734375, -261.4172668457031, -263.9488525390625, -265.66522216796875], "reading_logprob_std": 1.8627261114184175, "tokenization_path_logprob_std": 329.2812884644395, "trajectory_balance_rmse": 4406.147877750657} {"canonical_length": 9, "canonical_nll": 0.48354821734958225, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.34782608695652173, "prompt_id": "a75869a5f1810362", "reading_canonical_log_prob": -4.35193395614624, "reading_distances": [44, 44, 46, 48], "reading_log_probs": [-5.9109625816345215, -4.919294834136963, -6.550559997558594, -4.687310695648193], "reading_logprob_std": 0.869691245449331, "tokenization_path_logprob_std": 28.948236534859255, "trajectory_balance_rmse": 203.19947471761301} {"canonical_length": 228, "canonical_nll": 1.3333394903885691, "generation_rouge_l_f1": 0.18661257606490872, "generation_token_f1": 0.4827586206896552, "prompt_id": "f3ee3068b8bfca4b", "reading_canonical_log_prob": -304.00140380859375, "reading_distances": [22, 24, 28, 26], "reading_log_probs": [-305.79156494140625, -306.67095947265625, -306.5281677246094, -304.9420166015625], "reading_logprob_std": 0.7938912133443995, "tokenization_path_logprob_std": 177.95899097142777, "trajectory_balance_rmse": 6160.289112819695} {"canonical_length": 126, "canonical_nll": 1.6110038151816717, "generation_rouge_l_f1": 0.08450704225352113, "generation_token_f1": 0.09859154929577463, "prompt_id": "6a31e531aca13031", "reading_canonical_log_prob": -202.98648071289062, "reading_distances": [22, 30, 22, 21], "reading_log_probs": [-202.24729919433594, -197.68031311035156, -199.95068359375, -197.39761352539062], "reading_logprob_std": 2.26207729305226, "tokenization_path_logprob_std": 184.17259852450988, "trajectory_balance_rmse": 3132.6242024278095} {"canonical_length": 11, "canonical_nll": 0.8106113780628551, "generation_rouge_l_f1": 0.13636363636363638, "generation_token_f1": 0.15909090909090906, "prompt_id": "01a0ba7b191885ca", "reading_canonical_log_prob": -8.916725158691406, "reading_distances": [12, 6, 7, 9], "reading_log_probs": [-9.091907501220703, -7.778144836425781, -9.473491668701172, -8.247739791870117], "reading_logprob_std": 0.7735947423786594, "tokenization_path_logprob_std": 34.6018834988867, "trajectory_balance_rmse": 196.67428277966724} {"canonical_length": 9, "canonical_nll": 2.4666004180908203, "generation_rouge_l_f1": 0.03610108303249098, "generation_token_f1": 0.03610108303249098, "prompt_id": "8b612a049e9d2b42", "reading_canonical_log_prob": -22.199403762817383, "reading_distances": [16, 11, 14, 13], "reading_log_probs": [-21.896244049072266, -21.15887451171875, -22.6585750579834, -19.134321212768555], "reading_logprob_std": 1.51441306314054, "tokenization_path_logprob_std": 47.538536280228776, "trajectory_balance_rmse": 228.48153921063948} {"canonical_length": 168, "canonical_nll": 0.7194642566499256, "generation_rouge_l_f1": 0.2850356294536817, "generation_token_f1": 0.43230403800475053, "prompt_id": "1792c7fe245165d5", "reading_canonical_log_prob": -120.8699951171875, "reading_distances": [53, 51, 52, 52], "reading_log_probs": [-122.73313903808594, -121.63540649414062, -122.42485046386719, -121.17706298828125], "reading_logprob_std": 0.7136587297590997, "tokenization_path_logprob_std": 150.8271535766848, "trajectory_balance_rmse": 3294.728376630262} {"canonical_length": 240, "canonical_nll": 1.5400101979573568, "generation_rouge_l_f1": 0.2827868852459016, "generation_token_f1": 0.5983606557377049, "prompt_id": "9148af04e8535f5f", "reading_canonical_log_prob": -369.6024475097656, "reading_distances": [24, 22, 23, 22], "reading_log_probs": [-365.94842529296875, -366.9688720703125, -367.70343017578125, -368.3435974121094], "reading_logprob_std": 1.0286505201213838, "tokenization_path_logprob_std": 191.10124594188093, "trajectory_balance_rmse": 5777.644018759024} {"canonical_length": 11, "canonical_nll": 1.1094932556152344, "generation_rouge_l_f1": 0.08620689655172413, "generation_token_f1": 0.08620689655172413, "prompt_id": "4f964e48306df1f7", "reading_canonical_log_prob": -12.204425811767578, "reading_distances": [19, 16, 11, 12], "reading_log_probs": [-12.454566955566406, -13.031984329223633, -15.237493515014648, -11.247251510620117], "reading_logprob_std": 1.6710095476071987, "tokenization_path_logprob_std": 38.98879029096098, "trajectory_balance_rmse": 235.18512149607207} {"canonical_length": 13, "canonical_nll": 1.8858972696157603, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.4444444444444444, "prompt_id": "65511a559eb975a4", "reading_canonical_log_prob": -24.516664505004883, "reading_distances": [22, 21, 18, 18], "reading_log_probs": [-21.41493797302246, -23.236339569091797, -22.82948112487793, -22.262393951416016], "reading_logprob_std": 0.7891086458817059, "tokenization_path_logprob_std": 7.892805535775057, "trajectory_balance_rmse": 429.11344408819895} {"canonical_length": 273, "canonical_nll": 1.1957019624255953, "generation_rouge_l_f1": 0.29574861367837335, "generation_token_f1": 0.5693160813308688, "prompt_id": "dfd43bd8f8a6af69", "reading_canonical_log_prob": -326.4266357421875, "reading_distances": [16, 9, 16, 19], "reading_log_probs": [-325.02081298828125, -327.9248046875, -325.3238525390625, -324.8158264160156], "reading_logprob_std": 1.450741441642858, "tokenization_path_logprob_std": 400.06531165211703, "trajectory_balance_rmse": 6363.162989754609} {"canonical_length": 137, "canonical_nll": 2.055836475678604, "generation_rouge_l_f1": 0.15763546798029557, "generation_token_f1": 0.3103448275862069, "prompt_id": "9ebcabae9c8c462a", "reading_canonical_log_prob": -281.64959716796875, "reading_distances": [13, 16, 16, 21], "reading_log_probs": [-281.4571838378906, -274.9755554199219, -278.41522216796875, -273.318359375], "reading_logprob_std": 3.629291916555649, "tokenization_path_logprob_std": 292.0709735432109, "trajectory_balance_rmse": 4555.63384887714} {"canonical_length": 224, "canonical_nll": 1.102937834603446, "generation_rouge_l_f1": 0.24705882352941178, "generation_token_f1": 0.4, "prompt_id": "4e90bb5af9e629db", "reading_canonical_log_prob": -247.05807495117188, "reading_distances": [57, 54, 44, 47], "reading_log_probs": [-241.86325073242188, -241.77247619628906, -244.1220245361328, -242.85211181640625], "reading_logprob_std": 1.0949434591708118, "tokenization_path_logprob_std": 228.16209434627416, "trajectory_balance_rmse": 6142.748560149469} {"canonical_length": 17, "canonical_nll": 0.7115841472850126, "generation_rouge_l_f1": 0.18300653594771243, "generation_token_f1": 0.19607843137254904, "prompt_id": "3779413f4fba856c", "reading_canonical_log_prob": -12.096930503845215, "reading_distances": [12, 16, 21, 18], "reading_log_probs": [-12.483013153076172, -13.055320739746094, -14.624526977539062, -13.462516784667969], "reading_logprob_std": 0.9060688587929288, "tokenization_path_logprob_std": 81.62741081694647, "trajectory_balance_rmse": 605.9840215379983} {"canonical_length": 125, "canonical_nll": 1.3886068115234376, "generation_rouge_l_f1": 0.19576719576719576, "generation_token_f1": 0.37566137566137564, "prompt_id": "f749a091b4c13176", "reading_canonical_log_prob": -173.5758514404297, "reading_distances": [18, 22, 15, 19], "reading_log_probs": [-178.78652954101562, -176.24075317382812, -178.30868530273438, -172.2498779296875], "reading_logprob_std": 2.976990958493323, "tokenization_path_logprob_std": 136.03119404078004, "trajectory_balance_rmse": 3730.5633763559863} {"canonical_length": 45, "canonical_nll": 2.114601304796007, "generation_rouge_l_f1": 0.23214285714285715, "generation_token_f1": 0.41071428571428575, "prompt_id": "6378420685c9b069", "reading_canonical_log_prob": -95.15705871582031, "reading_distances": [23, 19, 22, 21], "reading_log_probs": [-99.41844940185547, -99.04605865478516, -99.84131622314453, -101.37947082519531], "reading_logprob_std": 1.0249497051363718, "tokenization_path_logprob_std": 201.0739908659948, "trajectory_balance_rmse": 1605.0203065076412} {"canonical_length": 119, "canonical_nll": 1.0018617646033023, "generation_rouge_l_f1": 0.26535626535626533, "generation_token_f1": 0.41769041769041776, "prompt_id": "634524f69adfb3f0", "reading_canonical_log_prob": -119.22154998779297, "reading_distances": [28, 25, 28, 25], "reading_log_probs": [-123.05546569824219, -126.13903045654297, -122.38734436035156, -122.25186157226562], "reading_logprob_std": 1.8212661213467058, "tokenization_path_logprob_std": 171.1286171103346, "trajectory_balance_rmse": 2809.93535938461} {"canonical_length": 162, "canonical_nll": 1.8671372025101274, "generation_rouge_l_f1": 0.13023255813953488, "generation_token_f1": 0.24186046511627904, "prompt_id": "c5c8a284983e4726", "reading_canonical_log_prob": -302.4762268066406, "reading_distances": [20, 22, 23, 19], "reading_log_probs": [-302.7508239746094, -303.6509704589844, -302.1409912109375, -306.34619140625], "reading_logprob_std": 1.8559990263164283, "tokenization_path_logprob_std": 37.36515565202926, "trajectory_balance_rmse": 3788.5239732920463} {"canonical_length": 88, "canonical_nll": 1.591382460160689, "generation_rouge_l_f1": 0.24223602484472048, "generation_token_f1": 0.31677018633540377, "prompt_id": "626ba72af7ac67f2", "reading_canonical_log_prob": -140.04165649414062, "reading_distances": [31, 26, 29, 27], "reading_log_probs": [-140.46029663085938, -145.8385467529297, -145.2705535888672, -145.36386108398438], "reading_logprob_std": 2.527609867887906, "tokenization_path_logprob_std": 68.2042177962301, "trajectory_balance_rmse": 1952.8402047554234} {"canonical_length": 297, "canonical_nll": 0.8452278419777199, "generation_rouge_l_f1": 0.29347826086956524, "generation_token_f1": 0.5181159420289855, "prompt_id": "c0257d516383b045", "reading_canonical_log_prob": -251.0326690673828, "reading_distances": [26, 29, 34, 23], "reading_log_probs": [-254.87771606445312, -254.80368041992188, -254.07846069335938, -254.47964477539062], "reading_logprob_std": 0.36453653160059196, "tokenization_path_logprob_std": 341.04455676725246, "trajectory_balance_rmse": 8240.603905019483} {"canonical_length": 38, "canonical_nll": 0.38735688360113846, "generation_rouge_l_f1": 0.14615384615384616, "generation_token_f1": 0.2153846153846154, "prompt_id": "dae9d23ce12cb989", "reading_canonical_log_prob": -14.719561576843262, "reading_distances": [21, 14, 18, 13], "reading_log_probs": [-17.496654510498047, -15.270444869995117, -15.646533966064453, -12.62568473815918], "reading_logprob_std": 2.0076319204539916, "tokenization_path_logprob_std": 54.66475142621006, "trajectory_balance_rmse": 357.07091836136544} {"canonical_length": 10, "canonical_nll": 0.6115030765533447, "generation_rouge_l_f1": 0.11594202898550725, "generation_token_f1": 0.11594202898550725, "prompt_id": "cde09f27bb3893ac", "reading_canonical_log_prob": -6.115030765533447, "reading_distances": [29, 28, 22, 26], "reading_log_probs": [-6.975586891174316, -6.515083312988281, -5.804986000061035, -6.93012809753418], "reading_logprob_std": 0.5421317183296017, "tokenization_path_logprob_std": 42.00105492664214, "trajectory_balance_rmse": 271.35674862371303} {"canonical_length": 174, "canonical_nll": 1.7325523639547413, "generation_rouge_l_f1": 0.2559241706161137, "generation_token_f1": 0.5450236966824644, "prompt_id": "64bb51344f196fcb", "reading_canonical_log_prob": -301.464111328125, "reading_distances": [27, 18, 20, 18], "reading_log_probs": [-311.2955322265625, -302.52886962890625, -307.7615966796875, -312.8957824707031], "reading_logprob_std": 4.592730237274822, "tokenization_path_logprob_std": 221.8279386038052, "trajectory_balance_rmse": 4540.686771529196} {"canonical_length": 312, "canonical_nll": 1.4580751076722755, "generation_rouge_l_f1": 0.24381625441696117, "generation_token_f1": 0.4770318021201414, "prompt_id": "bc3436553628cc88", "reading_canonical_log_prob": -454.91943359375, "reading_distances": [37, 30, 30, 31], "reading_log_probs": [-464.00970458984375, -457.9768981933594, -458.6801452636719, -467.3702087402344], "reading_logprob_std": 4.475283850141638, "tokenization_path_logprob_std": 685.395723395664, "trajectory_balance_rmse": 9949.525010458437} {"canonical_length": 39, "canonical_nll": 1.7631452511518428, "generation_rouge_l_f1": 0.21052631578947367, "generation_token_f1": 0.26900584795321636, "prompt_id": "b3b9c2d52d21969e", "reading_canonical_log_prob": -68.76266479492188, "reading_distances": [27, 30, 29, 26], "reading_log_probs": [-65.45175170898438, -66.01314544677734, -65.75162506103516, -67.38262939453125], "reading_logprob_std": 0.8532989864756076, "tokenization_path_logprob_std": 35.051416356834345, "trajectory_balance_rmse": 1242.8741207592184} {"canonical_length": 21, "canonical_nll": 0.27987561907087055, "generation_rouge_l_f1": 0.10606060606060606, "generation_token_f1": 0.12121212121212122, "prompt_id": "cc11564e8f134b4b", "reading_canonical_log_prob": -5.877388000488281, "reading_distances": [15, 14, 14, 14], "reading_log_probs": [-7.622470855712891, -7.251763820648193, -7.377096176147461, -7.607714653015137], "reading_logprob_std": 0.18107186902047911, "tokenization_path_logprob_std": 17.557116848267277, "trajectory_balance_rmse": 205.69322898136897} {"canonical_length": 20, "canonical_nll": 1.4219404220581056, "generation_rouge_l_f1": 0.1595744680851064, "generation_token_f1": 0.18085106382978722, "prompt_id": "e07efb6f1797853d", "reading_canonical_log_prob": -28.43880844116211, "reading_distances": [27, 28, 25, 24], "reading_log_probs": [-32.630516052246094, -32.52804946899414, -32.69900131225586, -35.99052047729492], "reading_logprob_std": 1.687128917167839, "tokenization_path_logprob_std": 68.5365305474216, "trajectory_balance_rmse": 482.35629354798976} {"canonical_length": 10, "canonical_nll": 0.25610086917877195, "generation_rouge_l_f1": 0.6428571428571429, "generation_token_f1": 0.6428571428571429, "prompt_id": "302fe191c6bb7b5b", "reading_canonical_log_prob": -2.5610086917877197, "reading_distances": [24, 30, 26, 22], "reading_log_probs": [-1.3362170457839966, -1.191478967666626, -1.1896060705184937, -1.0666399002075195], "reading_logprob_std": 0.1102363771974068, "tokenization_path_logprob_std": 46.81395206733904, "trajectory_balance_rmse": 230.82100445480174} {"canonical_length": 22, "canonical_nll": 1.9924605976451526, "generation_rouge_l_f1": 0.3448275862068965, "generation_token_f1": 0.3448275862068965, "prompt_id": "de0e448f061c6724", "reading_canonical_log_prob": -43.83413314819336, "reading_distances": [32, 25, 34, 28], "reading_log_probs": [-53.94844055175781, -51.20323181152344, -54.44601821899414, -51.34628677368164], "reading_logprob_std": 1.700475969811629, "tokenization_path_logprob_std": 65.85140633006148, "trajectory_balance_rmse": 868.3042691063655} {"canonical_length": 79, "canonical_nll": 1.366497377806072, "generation_rouge_l_f1": 0.31034482758620696, "generation_token_f1": 0.4068965517241379, "prompt_id": "f979fa68d4ceeb01", "reading_canonical_log_prob": -107.95329284667969, "reading_distances": [26, 27, 25, 24], "reading_log_probs": [-112.2449951171875, -114.87619018554688, -112.1592025756836, -110.26313018798828], "reading_logprob_std": 1.8955160062221559, "tokenization_path_logprob_std": 113.36510317844419, "trajectory_balance_rmse": 2239.541771627594} {"canonical_length": 193, "canonical_nll": 1.545172656755991, "generation_rouge_l_f1": 0.189010989010989, "generation_token_f1": 0.3428571428571428, "prompt_id": "e2f73d0c042e6f14", "reading_canonical_log_prob": -298.21832275390625, "reading_distances": [12, 13, 12, 16], "reading_log_probs": [-293.92401123046875, -298.6372375488281, -296.627197265625, -291.923828125], "reading_logprob_std": 2.9545763772658535, "tokenization_path_logprob_std": 410.0900243787445, "trajectory_balance_rmse": 5850.603573898997} {"canonical_length": 216, "canonical_nll": 1.8656669899269387, "generation_rouge_l_f1": 0.2184873949579832, "generation_token_f1": 0.4327731092436975, "prompt_id": "409538e18660e490", "reading_canonical_log_prob": -402.98406982421875, "reading_distances": [28, 38, 30, 31], "reading_log_probs": [-404.61956787109375, -406.90087890625, -405.3619689941406, -408.3448791503906], "reading_logprob_std": 1.6579258381385642, "tokenization_path_logprob_std": 179.5155887874328, "trajectory_balance_rmse": 6690.195178082858} {"canonical_length": 64, "canonical_nll": 2.1981277465820312, "generation_rouge_l_f1": 0.29629629629629634, "generation_token_f1": 0.34567901234567905, "prompt_id": "cc38da16fec412a8", "reading_canonical_log_prob": -140.68017578125, "reading_distances": [32, 31, 26, 30], "reading_log_probs": [-139.4287567138672, -145.23077392578125, -136.7415313720703, -134.91165161132812], "reading_logprob_std": 4.501747870619086, "tokenization_path_logprob_std": 96.4600414985366, "trajectory_balance_rmse": 1523.0691852698787} {"canonical_length": 53, "canonical_nll": 0.942612701991819, "generation_rouge_l_f1": 0.36904761904761907, "generation_token_f1": 0.4523809523809524, "prompt_id": "0e93241c78cc22bd", "reading_canonical_log_prob": -49.958473205566406, "reading_distances": [32, 28, 26, 35], "reading_log_probs": [-52.13933181762695, -51.868743896484375, -52.01123809814453, -50.88179016113281], "reading_logprob_std": 0.5730817101068809, "tokenization_path_logprob_std": 200.46408185772452, "trajectory_balance_rmse": 1371.9442223700814} {"canonical_length": 19, "canonical_nll": 0.221563038073088, "generation_rouge_l_f1": 0.38888888888888884, "generation_token_f1": 0.41666666666666663, "prompt_id": "32e3b4bcc84e208d", "reading_canonical_log_prob": -4.209697723388672, "reading_distances": [31, 28, 36, 29], "reading_log_probs": [-3.0123467445373535, -3.323132276535034, -3.4165329933166504, -3.054868698120117], "reading_logprob_std": 0.1985892727375518, "tokenization_path_logprob_std": 21.542896510843498, "trajectory_balance_rmse": 520.1229465803334} {"canonical_length": 28, "canonical_nll": 1.496912956237793, "generation_rouge_l_f1": 0.24242424242424246, "generation_token_f1": 0.33333333333333337, "prompt_id": "08748791c0916706", "reading_canonical_log_prob": -41.9135627746582, "reading_distances": [22, 19, 23, 21], "reading_log_probs": [-59.462913513183594, -48.01475524902344, -62.2241096496582, -48.874420166015625], "reading_logprob_std": 7.2552263269911395, "tokenization_path_logprob_std": 159.2020120917028, "trajectory_balance_rmse": 1170.8233588210587} {"canonical_length": 268, "canonical_nll": 1.5716140519327193, "generation_rouge_l_f1": 0.241635687732342, "generation_token_f1": 0.4535315985130111, "prompt_id": "340800ab7642e87c", "reading_canonical_log_prob": -421.19256591796875, "reading_distances": [20, 20, 25, 20], "reading_log_probs": [-423.78753662109375, -425.8511962890625, -425.35498046875, -420.933349609375], "reading_logprob_std": 2.2144265717569604, "tokenization_path_logprob_std": 443.9497008390042, "trajectory_balance_rmse": 6583.259603214528} {"canonical_length": 335, "canonical_nll": 1.3276793522621269, "generation_rouge_l_f1": 0.2579617834394905, "generation_token_f1": 0.49681528662420377, "prompt_id": "ba9377f6ac227fec", "reading_canonical_log_prob": -444.7725830078125, "reading_distances": [21, 25, 30, 26], "reading_log_probs": [-445.00048828125, -443.84130859375, -445.2016296386719, -444.7044677734375], "reading_logprob_std": 0.599615888826835, "tokenization_path_logprob_std": 81.19908843027945, "trajectory_balance_rmse": 9548.329852364797} {"canonical_length": 12, "canonical_nll": 0.007012542337179184, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "39897df6fcbea057", "reading_canonical_log_prob": -0.08415050804615021, "reading_distances": [22, 20, 26, 18], "reading_log_probs": [-0.1677108258008957, -0.160405695438385, -0.292665034532547, -0.17133966088294983], "reading_logprob_std": 0.06325347047303738, "tokenization_path_logprob_std": 89.7483160406232, "trajectory_balance_rmse": 299.01575937985984} {"canonical_length": 55, "canonical_nll": 0.7165596008300781, "generation_rouge_l_f1": 0.2244224422442244, "generation_token_f1": 0.2772277227722772, "prompt_id": "d9f80153d939b659", "reading_canonical_log_prob": -39.4107780456543, "reading_distances": [21, 17, 22, 21], "reading_log_probs": [-44.783878326416016, -43.1125373840332, -43.663482666015625, -46.17283248901367], "reading_logprob_std": 1.3522710531673772, "tokenization_path_logprob_std": 77.96915724698304, "trajectory_balance_rmse": 1475.7336877436617} {"canonical_length": 105, "canonical_nll": 1.5851422991071429, "generation_rouge_l_f1": 0.23295454545454547, "generation_token_f1": 0.3409090909090909, "prompt_id": "ea105e2e33c793e5", "reading_canonical_log_prob": -166.43994140625, "reading_distances": [34, 44, 32, 30], "reading_log_probs": [-177.92333984375, -180.03150939941406, -175.23927307128906, -174.4365692138672], "reading_logprob_std": 2.5612396238123853, "tokenization_path_logprob_std": 127.70542665594301, "trajectory_balance_rmse": 3329.2589814560733} {"canonical_length": 311, "canonical_nll": 1.5127564457835108, "generation_rouge_l_f1": 0.2103321033210332, "generation_token_f1": 0.40959409594095936, "prompt_id": "2fb44fa5d32cf514", "reading_canonical_log_prob": -470.4672546386719, "reading_distances": [33, 30, 30, 31], "reading_log_probs": [-481.0587158203125, -479.9197998046875, -479.1737365722656, -481.5015869140625], "reading_logprob_std": 1.0616009640261233, "tokenization_path_logprob_std": 184.11510751206285, "trajectory_balance_rmse": 8564.040090355667} {"canonical_length": 134, "canonical_nll": 1.0897063070268773, "generation_rouge_l_f1": 0.23738872403560832, "generation_token_f1": 0.35608308605341243, "prompt_id": "f67931e5043d993b", "reading_canonical_log_prob": -146.02064514160156, "reading_distances": [27, 26, 23, 29], "reading_log_probs": [-147.9490509033203, -152.1824493408203, -147.12835693359375, -147.49916076660156], "reading_logprob_std": 2.3525188534571506, "tokenization_path_logprob_std": 298.3731965864544, "trajectory_balance_rmse": 3824.783711883037} {"canonical_length": 42, "canonical_nll": 1.4444049653552828, "generation_rouge_l_f1": 0.13680781758957655, "generation_token_f1": 0.21498371335504887, "prompt_id": "f4dd2f85181ef8ef", "reading_canonical_log_prob": -60.665008544921875, "reading_distances": [23, 21, 17, 16], "reading_log_probs": [-56.749427795410156, -57.445621490478516, -57.46732711791992, -60.19758605957031], "reading_logprob_std": 1.5252856160205437, "tokenization_path_logprob_std": 78.6248989004128, "trajectory_balance_rmse": 1409.7618740176777} {"canonical_length": 386, "canonical_nll": 0.9902116844691143, "generation_rouge_l_f1": 0.22496147919876733, "generation_token_f1": 0.3667180277349769, "prompt_id": "044b1de06d0cdd9b", "reading_canonical_log_prob": -382.2217102050781, "reading_distances": [44, 47, 39, 43], "reading_log_probs": [-385.6833801269531, -384.4171142578125, -382.923583984375, -381.31414794921875], "reading_logprob_std": 1.8876663450782007, "tokenization_path_logprob_std": 466.3396586722431, "trajectory_balance_rmse": 9177.197606237543} {"canonical_length": 120, "canonical_nll": 0.8567750295003255, "generation_rouge_l_f1": 0.34523809523809523, "generation_token_f1": 0.5357142857142858, "prompt_id": "c7dd76b068a7ae00", "reading_canonical_log_prob": -102.81300354003906, "reading_distances": [41, 43, 39, 47], "reading_log_probs": [-101.95511627197266, -98.84548950195312, -99.88981628417969, -100.5848617553711], "reading_logprob_std": 1.3042227123120242, "tokenization_path_logprob_std": 305.13193303293605, "trajectory_balance_rmse": 2874.324644133024} {"canonical_length": 35, "canonical_nll": 2.951233128138951, "generation_rouge_l_f1": 0.16931216931216933, "generation_token_f1": 0.1798941798941799, "prompt_id": "2fa90c11cfdc9fd0", "reading_canonical_log_prob": -103.29315948486328, "reading_distances": [20, 19, 24, 27], "reading_log_probs": [-96.07989501953125, -100.87049865722656, -100.5199203491211, -100.30606079101562], "reading_logprob_std": 2.254835848321059, "tokenization_path_logprob_std": 46.99222613318468, "trajectory_balance_rmse": 934.897925218856} {"canonical_length": 33, "canonical_nll": 1.4795754172585227, "generation_rouge_l_f1": 0.17567567567567566, "generation_token_f1": 0.20945945945945948, "prompt_id": "acac7a68437d400e", "reading_canonical_log_prob": -48.82598876953125, "reading_distances": [24, 27, 34, 27], "reading_log_probs": [-54.225059509277344, -53.038734436035156, -52.520111083984375, -54.78173828125], "reading_logprob_std": 1.0426767529063008, "tokenization_path_logprob_std": 107.94530490704872, "trajectory_balance_rmse": 744.0949988548612} {"canonical_length": 41, "canonical_nll": 0.5175782645620951, "generation_rouge_l_f1": 0.46511627906976744, "generation_token_f1": 0.46511627906976744, "prompt_id": "98a669e9058b6ddc", "reading_canonical_log_prob": -21.2207088470459, "reading_distances": [12, 12, 9, 12], "reading_log_probs": [-21.710092544555664, -26.956710815429688, -21.499441146850586, -26.711559295654297], "reading_logprob_std": 3.022059370339758, "tokenization_path_logprob_std": 84.7124947736821, "trajectory_balance_rmse": 754.851265911858} {"canonical_length": 133, "canonical_nll": 1.5330734826568375, "generation_rouge_l_f1": 0.2538860103626943, "generation_token_f1": 0.3989637305699482, "prompt_id": "610adc3864696d51", "reading_canonical_log_prob": -203.89877319335938, "reading_distances": [21, 19, 21, 21], "reading_log_probs": [-207.23165893554688, -212.83421325683594, -210.45648193359375, -204.45379638671875], "reading_logprob_std": 3.667672349609064, "tokenization_path_logprob_std": 252.89398850341263, "trajectory_balance_rmse": 4538.9681439401675} {"canonical_length": 34, "canonical_nll": 0.5338410209206974, "generation_rouge_l_f1": 0.17869415807560135, "generation_token_f1": 0.17869415807560135, "prompt_id": "74b66fda46123ad6", "reading_canonical_log_prob": -18.15059471130371, "reading_distances": [38, 42, 44, 39], "reading_log_probs": [-21.161378860473633, -19.963163375854492, -21.38262176513672, -20.650959014892578], "reading_logprob_std": 0.6303737336748567, "tokenization_path_logprob_std": 109.34181013743138, "trajectory_balance_rmse": 906.7056614989223} {"canonical_length": 23, "canonical_nll": 0.5627729996390964, "generation_rouge_l_f1": 0.3, "generation_token_f1": 0.3142857142857143, "prompt_id": "9ce2e67b60ca246c", "reading_canonical_log_prob": -12.943778991699219, "reading_distances": [19, 18, 17, 19], "reading_log_probs": [-12.827088356018066, -12.876254081726074, -13.418224334716797, -12.433186531066895], "reading_logprob_std": 0.4049030486620368, "tokenization_path_logprob_std": 88.6833554527289, "trajectory_balance_rmse": 563.6916265810695} {"canonical_length": 9, "canonical_nll": 1.7622669008043077, "generation_rouge_l_f1": 0.8, "generation_token_f1": 0.8, "prompt_id": "f37797ee8ed6a36e", "reading_canonical_log_prob": -15.86040210723877, "reading_distances": [42, 32, 42, 48], "reading_log_probs": [-13.66134262084961, -12.940547943115234, -14.404041290283203, -15.067707061767578], "reading_logprob_std": 0.9199673078982125, "tokenization_path_logprob_std": 26.153270101926882, "trajectory_balance_rmse": 231.94229860360147} {"canonical_length": 25, "canonical_nll": 1.6187466430664061, "generation_rouge_l_f1": 0.1509433962264151, "generation_token_f1": 0.2012578616352201, "prompt_id": "b9be7345d7f0f09d", "reading_canonical_log_prob": -40.468666076660156, "reading_distances": [29, 23, 24, 28], "reading_log_probs": [-41.200111389160156, -42.91707229614258, -41.64012908935547, -42.631961822509766], "reading_logprob_std": 0.8107285870420164, "tokenization_path_logprob_std": 19.369166068864033, "trajectory_balance_rmse": 781.8255284091607} {"canonical_length": 102, "canonical_nll": 0.9885613684560738, "generation_rouge_l_f1": 0.419672131147541, "generation_token_f1": 0.4983606557377049, "prompt_id": "101280c93f2c70da", "reading_canonical_log_prob": -100.83325958251953, "reading_distances": [18, 16, 10, 18], "reading_log_probs": [-101.26164245605469, -103.06481170654297, -101.57679748535156, -103.7141342163086], "reading_logprob_std": 1.175069934733148, "tokenization_path_logprob_std": 130.22832390181316, "trajectory_balance_rmse": 1199.1614004700457} {"canonical_length": 309, "canonical_nll": 1.1904217865088997, "generation_rouge_l_f1": 0.26804123711340205, "generation_token_f1": 0.5085910652920962, "prompt_id": "bc039a3de3397314", "reading_canonical_log_prob": -367.84033203125, "reading_distances": [35, 38, 27, 35], "reading_log_probs": [-369.37713623046875, -373.0538330078125, -370.8656311035156, -367.70269775390625], "reading_logprob_std": 2.272379146860687, "tokenization_path_logprob_std": 441.94844478605864, "trajectory_balance_rmse": 9483.57669972538} {"canonical_length": 144, "canonical_nll": 1.1920795440673828, "generation_rouge_l_f1": 0.29041095890410956, "generation_token_f1": 0.4109589041095891, "prompt_id": "3bf14cf65dcd8fce", "reading_canonical_log_prob": -171.65945434570312, "reading_distances": [35, 36, 35, 39], "reading_log_probs": [-171.2618408203125, -171.36666870117188, -172.58969116210938, -173.57894897460938], "reading_logprob_std": 1.0996880080775646, "tokenization_path_logprob_std": 188.84292040414152, "trajectory_balance_rmse": 2855.4777997485394} {"canonical_length": 38, "canonical_nll": 1.8937237388209294, "generation_rouge_l_f1": 0.0959409594095941, "generation_token_f1": 0.14022140221402213, "prompt_id": "ab60c5be348f9273", "reading_canonical_log_prob": -71.96150207519531, "reading_distances": [15, 20, 25, 22], "reading_log_probs": [-74.04188537597656, -77.73545837402344, -73.50012969970703, -75.66899108886719], "reading_logprob_std": 1.9038493780384798, "tokenization_path_logprob_std": 186.46000184193477, "trajectory_balance_rmse": 1489.9005460813141} {"canonical_length": 12, "canonical_nll": 3.1334222157796225, "generation_rouge_l_f1": 0.37499999999999994, "generation_token_f1": 0.37499999999999994, "prompt_id": "8392d28ac48cc769", "reading_canonical_log_prob": -37.60106658935547, "reading_distances": [38, 35, 33, 32], "reading_log_probs": [-44.23250961303711, -45.73395919799805, -41.739479064941406, -44.72257995605469], "reading_logprob_std": 1.6977281766624204, "tokenization_path_logprob_std": 97.08865082022048, "trajectory_balance_rmse": 467.71042469014844} {"canonical_length": 90, "canonical_nll": 2.0982842339409724, "generation_rouge_l_f1": 0.15254237288135591, "generation_token_f1": 0.23728813559322032, "prompt_id": "2a39144255d482e3", "reading_canonical_log_prob": -188.8455810546875, "reading_distances": [29, 29, 25, 23], "reading_log_probs": [-202.0773468017578, -199.0127716064453, -194.27037048339844, -216.37399291992188], "reading_logprob_std": 9.518434260751325, "tokenization_path_logprob_std": 204.11952301418486, "trajectory_balance_rmse": 2498.9864948089103} {"canonical_length": 292, "canonical_nll": 1.4825037080947667, "generation_rouge_l_f1": 0.21621621621621623, "generation_token_f1": 0.47567567567567565, "prompt_id": "7938846720b7f660", "reading_canonical_log_prob": -432.8910827636719, "reading_distances": [20, 13, 22, 17], "reading_log_probs": [-439.698486328125, -437.99127197265625, -437.3089599609375, -436.583251953125], "reading_logprob_std": 1.3324108630450096, "tokenization_path_logprob_std": 340.36330128975777, "trajectory_balance_rmse": 7535.782913020067} {"canonical_length": 196, "canonical_nll": 1.6668839746591997, "generation_rouge_l_f1": 0.22869955156950675, "generation_token_f1": 0.4349775784753363, "prompt_id": "2a7c37e2ef87098f", "reading_canonical_log_prob": -326.7092590332031, "reading_distances": [27, 30, 28, 31], "reading_log_probs": [-336.8318786621094, -336.941650390625, -341.7069091796875, -340.17706298828125], "reading_logprob_std": 2.423569095516908, "tokenization_path_logprob_std": 162.82725987858743, "trajectory_balance_rmse": 6031.474067269407} {"canonical_length": 284, "canonical_nll": 1.3599965270136443, "generation_rouge_l_f1": 0.2647058823529412, "generation_token_f1": 0.46691176470588236, "prompt_id": "f9f1a4c90d1c6936", "reading_canonical_log_prob": -386.239013671875, "reading_distances": [27, 25, 28, 28], "reading_log_probs": [-389.3250732421875, -387.5199279785156, -391.68841552734375, -386.11187744140625], "reading_logprob_std": 2.408755474265072, "tokenization_path_logprob_std": 262.2788476547987, "trajectory_balance_rmse": 8671.44551056639} {"canonical_length": 149, "canonical_nll": 1.4121059955366506, "generation_rouge_l_f1": 0.28297362110311747, "generation_token_f1": 0.42685851318944845, "prompt_id": "4c7fd497c60d26bd", "reading_canonical_log_prob": -210.40379333496094, "reading_distances": [20, 22, 27, 23], "reading_log_probs": [-212.5127716064453, -209.57949829101562, -211.39315795898438, -211.607421875], "reading_logprob_std": 1.2289911064983225, "tokenization_path_logprob_std": 135.91645664463545, "trajectory_balance_rmse": 3999.983425956277} {"canonical_length": 25, "canonical_nll": 1.5449465942382812, "generation_rouge_l_f1": 0.3636363636363636, "generation_token_f1": 0.40404040404040403, "prompt_id": "c2e4f5be5a2ad0b2", "reading_canonical_log_prob": -38.62366485595703, "reading_distances": [13, 17, 20, 16], "reading_log_probs": [-37.165771484375, -43.90552520751953, -43.08154296875, -42.68169021606445], "reading_logprob_std": 3.0711370451150395, "tokenization_path_logprob_std": 144.71712709436355, "trajectory_balance_rmse": 676.8870598866399} {"canonical_length": 14, "canonical_nll": 1.196266038077218, "generation_rouge_l_f1": 0.3703703703703704, "generation_token_f1": 0.4444444444444445, "prompt_id": "619e2fb6adbffc57", "reading_canonical_log_prob": -16.747724533081055, "reading_distances": [36, 30, 30, 30], "reading_log_probs": [-18.96427345275879, -16.418346405029297, -17.12778663635254, -17.41425323486328], "reading_logprob_std": 1.0737081385969403, "tokenization_path_logprob_std": 91.65456853234487, "trajectory_balance_rmse": 505.3273819587537} {"canonical_length": 136, "canonical_nll": 1.4910986283246208, "generation_rouge_l_f1": 0.25545171339563866, "generation_token_f1": 0.5046728971962617, "prompt_id": "237ba288f6603bf1", "reading_canonical_log_prob": -202.78941345214844, "reading_distances": [18, 17, 18, 21], "reading_log_probs": [-204.2732391357422, -193.68234252929688, -198.9010009765625, -197.04107666015625], "reading_logprob_std": 4.428195806932457, "tokenization_path_logprob_std": 289.38293829057017, "trajectory_balance_rmse": 3892.831406268929} {"canonical_length": 9, "canonical_nll": 0.6818849775526259, "generation_rouge_l_f1": 0.7142857142857143, "generation_token_f1": 0.7142857142857143, "prompt_id": "52afc3f6a37f4218", "reading_canonical_log_prob": -6.136964797973633, "reading_distances": [27, 26, 24, 22], "reading_log_probs": [-6.774018287658691, -8.067147254943848, -5.092263698577881, -5.534509658813477], "reading_logprob_std": 1.338426611159612, "tokenization_path_logprob_std": 38.20319758781104, "trajectory_balance_rmse": 266.9564094477646} {"canonical_length": 115, "canonical_nll": 1.7870537799337636, "generation_rouge_l_f1": 0.20155038759689922, "generation_token_f1": 0.3307493540051679, "prompt_id": "0096b48ae77be6ff", "reading_canonical_log_prob": -205.5111846923828, "reading_distances": [35, 31, 30, 25], "reading_log_probs": [-224.80426025390625, -230.90328979492188, -229.32003784179688, -235.11837768554688], "reading_logprob_std": 4.260925413721139, "tokenization_path_logprob_std": 246.35744675048835, "trajectory_balance_rmse": 4018.587510102749} {"canonical_length": 252, "canonical_nll": 1.1388980926029266, "generation_rouge_l_f1": 0.3254901960784313, "generation_token_f1": 0.5058823529411766, "prompt_id": "2061c66b7116786e", "reading_canonical_log_prob": -287.0023193359375, "reading_distances": [31, 38, 33, 29], "reading_log_probs": [-283.3919372558594, -284.06646728515625, -281.20672607421875, -284.7718811035156], "reading_logprob_std": 1.5416561041907235, "tokenization_path_logprob_std": 267.4476170601632, "trajectory_balance_rmse": 6890.097839610233} {"canonical_length": 94, "canonical_nll": 1.493427520102643, "generation_rouge_l_f1": 0.2160493827160494, "generation_token_f1": 0.3518518518518518, "prompt_id": "5622ae2789dbdaf4", "reading_canonical_log_prob": -140.38218688964844, "reading_distances": [18, 18, 16, 17], "reading_log_probs": [-139.0891876220703, -137.37342834472656, -138.14578247070312, -134.8793182373047], "reading_logprob_std": 1.803786483256181, "tokenization_path_logprob_std": 151.59661481300157, "trajectory_balance_rmse": 2630.8428777063364} {"canonical_length": 279, "canonical_nll": 1.3448096928203406, "generation_rouge_l_f1": 0.24474187380497128, "generation_token_f1": 0.4435946462715105, "prompt_id": "531c2e72ca7ad121", "reading_canonical_log_prob": -375.201904296875, "reading_distances": [27, 16, 22, 25], "reading_log_probs": [-370.01605224609375, -373.19775390625, -372.70660400390625, -376.07794189453125], "reading_logprob_std": 2.4834691893384138, "tokenization_path_logprob_std": 164.79498501927802, "trajectory_balance_rmse": 7780.443133864592} {"canonical_length": 33, "canonical_nll": 1.3031655513879024, "generation_rouge_l_f1": 0.5490196078431373, "generation_token_f1": 0.6274509803921569, "prompt_id": "50acd3d956ee080e", "reading_canonical_log_prob": -43.00446319580078, "reading_distances": [26, 23, 37, 36], "reading_log_probs": [-44.56771469116211, -46.06035232543945, -48.57398986816406, -45.247459411621094], "reading_logprob_std": 1.750838709732035, "tokenization_path_logprob_std": 73.94548384320933, "trajectory_balance_rmse": 1011.033689604614} {"canonical_length": 26, "canonical_nll": 1.2079071631798377, "generation_rouge_l_f1": 0.8571428571428572, "generation_token_f1": 0.8571428571428572, "prompt_id": "3fcc8e674911ba73", "reading_canonical_log_prob": -31.40558624267578, "reading_distances": [16, 15, 15, 15], "reading_log_probs": [-35.0815315246582, -37.131065368652344, -34.253173828125, -33.32804870605469], "reading_logprob_std": 1.6217922029024145, "tokenization_path_logprob_std": 60.353671163388555, "trajectory_balance_rmse": 695.5993017660168} {"canonical_length": 347, "canonical_nll": 1.029594927081457, "generation_rouge_l_f1": 0.26198083067092653, "generation_token_f1": 0.5175718849840256, "prompt_id": "915d77693c098234", "reading_canonical_log_prob": -357.2694396972656, "reading_distances": [17, 13, 11, 17], "reading_log_probs": [-353.72265625, -354.390869140625, -361.36541748046875, -358.191162109375], "reading_logprob_std": 3.5588891134557676, "tokenization_path_logprob_std": 219.52381100957643, "trajectory_balance_rmse": 7786.250259712856} {"canonical_length": 16, "canonical_nll": 0.082561194896698, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "d4ba094ba07a6bba", "reading_canonical_log_prob": -1.320979118347168, "reading_distances": [21, 21, 28, 21], "reading_log_probs": [-1.6397901773452759, -1.3068861961364746, -1.6792441606521606, -0.9311889410018921], "reading_logprob_std": 0.3480760073306257, "tokenization_path_logprob_std": 50.01098254494593, "trajectory_balance_rmse": 350.5263737975298} {"canonical_length": 35, "canonical_nll": 1.165032196044922, "generation_rouge_l_f1": 0.16, "generation_token_f1": 0.2, "prompt_id": "40f13ac2314179d4", "reading_canonical_log_prob": -40.776126861572266, "reading_distances": [20, 25, 28, 30], "reading_log_probs": [-40.1981315612793, -39.892765045166016, -39.408363342285156, -40.836151123046875], "reading_logprob_std": 0.5977214902360611, "tokenization_path_logprob_std": 144.49991309660803, "trajectory_balance_rmse": 1130.360089205641} {"canonical_length": 228, "canonical_nll": 1.312334696451823, "generation_rouge_l_f1": 0.2677824267782427, "generation_token_f1": 0.48953974895397484, "prompt_id": "c60f4d0722caa17f", "reading_canonical_log_prob": -299.2123107910156, "reading_distances": [62, 53, 67, 59], "reading_log_probs": [-292.3644104003906, -285.97955322265625, -286.9371032714844, -286.6175537109375], "reading_logprob_std": 2.953451149380987, "tokenization_path_logprob_std": 300.9550838530163, "trajectory_balance_rmse": 6660.378349351334} {"canonical_length": 20, "canonical_nll": 2.2699905395507813, "generation_rouge_l_f1": 0.15, "generation_token_f1": 0.19999999999999998, "prompt_id": "2ca8b957d3b2b5bc", "reading_canonical_log_prob": -45.399810791015625, "reading_distances": [24, 23, 30, 30], "reading_log_probs": [-49.59615707397461, -48.51786422729492, -49.34708786010742, -48.36847686767578], "reading_logprob_std": 0.6054996226513905, "tokenization_path_logprob_std": 83.20551802748133, "trajectory_balance_rmse": 614.4814975509165} {"canonical_length": 159, "canonical_nll": 1.434269167342276, "generation_rouge_l_f1": 0.3186274509803922, "generation_token_f1": 0.5245098039215687, "prompt_id": "762bfb8c99f50dd6", "reading_canonical_log_prob": -228.04879760742188, "reading_distances": [22, 20, 31, 20], "reading_log_probs": [-232.46597290039062, -233.9202880859375, -230.03611755371094, -231.1585693359375], "reading_logprob_std": 1.6758655208077597, "tokenization_path_logprob_std": 153.25503168796502, "trajectory_balance_rmse": 4058.5969996637227} {"canonical_length": 231, "canonical_nll": 1.1657134877654898, "generation_rouge_l_f1": 0.21643286573146292, "generation_token_f1": 0.4128256513026052, "prompt_id": "9f2ef447478c22f4", "reading_canonical_log_prob": -269.2798156738281, "reading_distances": [29, 27, 19, 24], "reading_log_probs": [-269.0779724121094, -268.9787292480469, -265.93359375, -268.2672119140625], "reading_logprob_std": 1.46569607511252, "tokenization_path_logprob_std": 290.02881173038975, "trajectory_balance_rmse": 6917.586098829949} {"canonical_length": 361, "canonical_nll": 1.2736725107124307, "generation_rouge_l_f1": 0.23287671232876714, "generation_token_f1": 0.42808219178082196, "prompt_id": "dc0376ea25a8c1bc", "reading_canonical_log_prob": -459.7957763671875, "reading_distances": [27, 26, 30, 30], "reading_log_probs": [-458.0426330566406, -456.28326416015625, -454.46392822265625, -456.54034423828125], "reading_logprob_std": 1.46762153915951, "tokenization_path_logprob_std": 345.1788596451481, "trajectory_balance_rmse": 7956.732504669331} {"canonical_length": 18, "canonical_nll": 2.9327420128716364, "generation_rouge_l_f1": 0.30303030303030304, "generation_token_f1": 0.3636363636363636, "prompt_id": "9cf805580809e94d", "reading_canonical_log_prob": -52.78935623168945, "reading_distances": [31, 22, 23, 20], "reading_log_probs": [-56.373695373535156, -55.89521408081055, -56.2470703125, -54.418968200683594], "reading_logprob_std": 0.8995831725236204, "tokenization_path_logprob_std": 53.74899823296087, "trajectory_balance_rmse": 497.91757792242987} {"canonical_length": 280, "canonical_nll": 1.7173218863351005, "generation_rouge_l_f1": 0.22950819672131148, "generation_token_f1": 0.43715846994535523, "prompt_id": "d3ad116b6d046506", "reading_canonical_log_prob": -480.8501281738281, "reading_distances": [24, 22, 22, 24], "reading_log_probs": [-480.9259338378906, -479.43621826171875, -480.66619873046875, -480.07330322265625], "reading_logprob_std": 0.6635729094640462, "tokenization_path_logprob_std": 115.71906525825358, "trajectory_balance_rmse": 9357.207768625152} {"canonical_length": 141, "canonical_nll": 1.6808296365940825, "generation_rouge_l_f1": 0.336, "generation_token_f1": 0.5439999999999999, "prompt_id": "d4c38137aba16167", "reading_canonical_log_prob": -236.99697875976562, "reading_distances": [12, 14, 9, 16], "reading_log_probs": [-247.97891235351562, -258.1357116699219, -252.2718963623047, -250.93389892578125], "reading_logprob_std": 4.265783472386154, "tokenization_path_logprob_std": 94.11787667376669, "trajectory_balance_rmse": 1834.22967722115} {"canonical_length": 41, "canonical_nll": 1.6480124868997714, "generation_rouge_l_f1": 0.2535211267605634, "generation_token_f1": 0.28169014084507044, "prompt_id": "84547b634d24f63a", "reading_canonical_log_prob": -67.56851196289062, "reading_distances": [24, 28, 29, 24], "reading_log_probs": [-65.30892181396484, -64.26128387451172, -64.75505065917969, -65.0141830444336], "reading_logprob_std": 0.4443161263247898, "tokenization_path_logprob_std": 248.65694254955315, "trajectory_balance_rmse": 1325.6124415355416} {"canonical_length": 6, "canonical_nll": 0.3522101243336995, "generation_rouge_l_f1": 0.8333333333333333, "generation_token_f1": 0.8333333333333333, "prompt_id": "58c9afa198f60472", "reading_canonical_log_prob": -2.1132607460021973, "reading_distances": [43, 40, 52, 40], "reading_log_probs": [-5.712743282318115, -6.840003490447998, -7.116794586181641, -5.7019500732421875], "reading_logprob_std": 0.742504546616954, "tokenization_path_logprob_std": 15.80195210167499, "trajectory_balance_rmse": 90.33671822734568} {"canonical_length": 238, "canonical_nll": 1.6570762826615022, "generation_rouge_l_f1": 0.22003929273084477, "generation_token_f1": 0.49508840864440085, "prompt_id": "74608c6aad72d41d", "reading_canonical_log_prob": -394.3841552734375, "reading_distances": [37, 39, 36, 32], "reading_log_probs": [-400.1669006347656, -402.70428466796875, -399.56365966796875, -399.79541015625], "reading_logprob_std": 1.4525554256772943, "tokenization_path_logprob_std": 269.05072379339515, "trajectory_balance_rmse": 6935.183352934382} {"canonical_length": 296, "canonical_nll": 1.520014273153769, "generation_rouge_l_f1": 0.24860853432282004, "generation_token_f1": 0.5269016697588126, "prompt_id": "5c242c11a7bd0e44", "reading_canonical_log_prob": -449.9242248535156, "reading_distances": [26, 35, 28, 36], "reading_log_probs": [-456.22540283203125, -453.3045654296875, -455.4638977050781, -452.400634765625], "reading_logprob_std": 1.793586219196757, "tokenization_path_logprob_std": 248.36028402777518, "trajectory_balance_rmse": 9586.088304507171} {"canonical_length": 42, "canonical_nll": 0.8069747743152437, "generation_rouge_l_f1": 0.64, "generation_token_f1": 0.7200000000000001, "prompt_id": "c3513f19426dd701", "reading_canonical_log_prob": -33.892940521240234, "reading_distances": [23, 27, 26, 33], "reading_log_probs": [-36.96336364746094, -39.61703109741211, -36.686317443847656, -41.40909194946289], "reading_logprob_std": 2.254409924951157, "tokenization_path_logprob_std": 48.735991284688986, "trajectory_balance_rmse": 545.1234526656842} {"canonical_length": 356, "canonical_nll": 1.2163253098391416, "generation_rouge_l_f1": 0.20477815699658702, "generation_token_f1": 0.4436860068259387, "prompt_id": "80a15e1bafbe7e14", "reading_canonical_log_prob": -433.0118103027344, "reading_distances": [20, 20, 17, 16], "reading_log_probs": [-427.721923828125, -428.17352294921875, -430.063232421875, -428.2410583496094], "reading_logprob_std": 1.034853033838092, "tokenization_path_logprob_std": 382.49025146095795, "trajectory_balance_rmse": 9910.165781393227} {"canonical_length": 51, "canonical_nll": 1.7993645761527268, "generation_rouge_l_f1": 0.10344827586206896, "generation_token_f1": 0.13793103448275865, "prompt_id": "74f44ff0ad2c7451", "reading_canonical_log_prob": -91.76759338378906, "reading_distances": [23, 25, 25, 22], "reading_log_probs": [-87.64388275146484, -85.96308898925781, -85.5214614868164, -86.36476135253906], "reading_logprob_std": 0.9143951302527906, "tokenization_path_logprob_std": 149.48702434516207, "trajectory_balance_rmse": 1669.2967271306125} {"canonical_length": 21, "canonical_nll": 1.3163957141694569, "generation_rouge_l_f1": 0.24489795918367344, "generation_token_f1": 0.3673469387755102, "prompt_id": "0b30260636e7066b", "reading_canonical_log_prob": -27.644309997558594, "reading_distances": [32, 34, 23, 31], "reading_log_probs": [-36.0040397644043, -39.06479263305664, -27.68860626220703, -31.882177352905273], "reading_logprob_std": 4.950572435684123, "tokenization_path_logprob_std": 97.88396737733873, "trajectory_balance_rmse": 688.9783897358931} {"canonical_length": 142, "canonical_nll": 1.2874334630831865, "generation_rouge_l_f1": 0.3379501385041551, "generation_token_f1": 0.5207756232686981, "prompt_id": "270a3b8ee410411a", "reading_canonical_log_prob": -182.8155517578125, "reading_distances": [24, 31, 31, 29], "reading_log_probs": [-182.52346801757812, -185.0261688232422, -184.55136108398438, -184.02960205078125], "reading_logprob_std": 1.085323359932582, "tokenization_path_logprob_std": 185.98467432422007, "trajectory_balance_rmse": 3390.145788113911} {"canonical_length": 2, "canonical_nll": 2.5878608226776123, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "aa51f5831004ecdf", "reading_canonical_log_prob": -5.175721645355225, "reading_distances": [35, 33, 39, 28], "reading_log_probs": [-3.464860200881958, -4.7654924392700195, -3.6438698768615723, -4.599212646484375], "reading_logprob_std": 0.658837998798892, "tokenization_path_logprob_std": 33.946506900962376, "trajectory_balance_rmse": 59.714136984671995} {"canonical_length": 53, "canonical_nll": 0.9517011822394605, "generation_rouge_l_f1": 0.3174603174603175, "generation_token_f1": 0.42857142857142855, "prompt_id": "d23e5c49d3b87636", "reading_canonical_log_prob": -50.440162658691406, "reading_distances": [20, 16, 18, 13], "reading_log_probs": [-47.997684478759766, -53.73944091796875, -46.975830078125, -49.949607849121094], "reading_logprob_std": 2.9829317064049423, "tokenization_path_logprob_std": 144.13859772523813, "trajectory_balance_rmse": 1116.8736570867625} {"canonical_length": 78, "canonical_nll": 1.8234033829126604, "generation_rouge_l_f1": 0.19166666666666668, "generation_token_f1": 0.3166666666666667, "prompt_id": "1cf9ae29e6e4ee3d", "reading_canonical_log_prob": -142.2254638671875, "reading_distances": [19, 22, 27, 21], "reading_log_probs": [-147.07635498046875, -145.5025634765625, -146.14503479003906, -148.27813720703125], "reading_logprob_std": 1.2060666910193867, "tokenization_path_logprob_std": 114.11720074499706, "trajectory_balance_rmse": 2239.4807748999347} {"canonical_length": 113, "canonical_nll": 1.5442049819811257, "generation_rouge_l_f1": 0.14937759336099585, "generation_token_f1": 0.2987551867219917, "prompt_id": "4b26a788bce312c3", "reading_canonical_log_prob": -174.4951629638672, "reading_distances": [27, 37, 31, 34], "reading_log_probs": [-172.0522918701172, -172.81085205078125, -174.10711669921875, -171.2587432861328], "reading_logprob_std": 1.2120922935701959, "tokenization_path_logprob_std": 286.69076942595336, "trajectory_balance_rmse": 3375.724473458256} {"canonical_length": 320, "canonical_nll": 1.5626283645629884, "generation_rouge_l_f1": 0.2604166666666667, "generation_token_f1": 0.48958333333333337, "prompt_id": "c59752dc85c30f5b", "reading_canonical_log_prob": -500.04107666015625, "reading_distances": [35, 38, 41, 35], "reading_log_probs": [-507.37091064453125, -509.1111755371094, -515.852294921875, -499.7572021484375], "reading_logprob_std": 6.613889918957227, "tokenization_path_logprob_std": 562.5692125418714, "trajectory_balance_rmse": 10550.911613270648} {"canonical_length": 9, "canonical_nll": 2.0807217491997614, "generation_rouge_l_f1": 0.7692307692307693, "generation_token_f1": 0.7692307692307693, "prompt_id": "b4e868a234198084", "reading_canonical_log_prob": -18.72649574279785, "reading_distances": [25, 32, 30, 31], "reading_log_probs": [-21.380334854125977, -27.203075408935547, -21.020509719848633, -24.87247085571289], "reading_logprob_std": 2.9541247466472873, "tokenization_path_logprob_std": 50.757512217904406, "trajectory_balance_rmse": 356.7855162571227} {"canonical_length": 68, "canonical_nll": 2.1878974016974952, "generation_rouge_l_f1": 0.20600858369098712, "generation_token_f1": 0.34334763948497854, "prompt_id": "466e7627e6013e86", "reading_canonical_log_prob": -148.7770233154297, "reading_distances": [33, 36, 37, 37], "reading_log_probs": [-154.1643524169922, -152.16590881347656, -153.40090942382812, -156.3785400390625], "reading_logprob_std": 1.7705230229394706, "tokenization_path_logprob_std": 131.6158407737203, "trajectory_balance_rmse": 2058.1838011219556} {"canonical_length": 14, "canonical_nll": 1.1560651234218053, "generation_rouge_l_f1": 0.4166666666666667, "generation_token_f1": 0.5416666666666666, "prompt_id": "ca45cac53e43d02f", "reading_canonical_log_prob": -16.184911727905273, "reading_distances": [24, 24, 23, 29], "reading_log_probs": [-20.105327606201172, -18.822132110595703, -17.781530380249023, -20.43649673461914], "reading_logprob_std": 1.221139108313708, "tokenization_path_logprob_std": 65.27327393474944, "trajectory_balance_rmse": 244.76974382578277} {"canonical_length": 20, "canonical_nll": 2.3487874984741213, "generation_rouge_l_f1": 0.11188811188811189, "generation_token_f1": 0.12587412587412586, "prompt_id": "685a9c22a80a27a9", "reading_canonical_log_prob": -46.97574996948242, "reading_distances": [24, 21, 20, 16], "reading_log_probs": [-56.49580383300781, -53.78896713256836, -54.88489532470703, -57.56671142578125], "reading_logprob_std": 1.6766380638157374, "tokenization_path_logprob_std": 72.28196616522322, "trajectory_balance_rmse": 597.5179947630721} {"canonical_length": 12, "canonical_nll": 1.540136496225993, "generation_rouge_l_f1": 0.3255813953488372, "generation_token_f1": 0.37209302325581395, "prompt_id": "426d94bf331d4f8d", "reading_canonical_log_prob": -18.481637954711914, "reading_distances": [23, 26, 23, 31], "reading_log_probs": [-20.948387145996094, -18.7147159576416, -22.08856773376465, -14.917396545410156], "reading_logprob_std": 3.16083260541176, "tokenization_path_logprob_std": 62.69271496175725, "trajectory_balance_rmse": 234.2069817801767} {"canonical_length": 123, "canonical_nll": 1.1839738551194106, "generation_rouge_l_f1": 0.24778761061946902, "generation_token_f1": 0.415929203539823, "prompt_id": "efba508a45047679", "reading_canonical_log_prob": -145.6287841796875, "reading_distances": [29, 34, 36, 30], "reading_log_probs": [-142.79330444335938, -143.90237426757812, -147.0382080078125, -144.79327392578125], "reading_logprob_std": 1.8008332459624194, "tokenization_path_logprob_std": 134.10396480599434, "trajectory_balance_rmse": 3378.568078298708} {"canonical_length": 138, "canonical_nll": 1.339428666709126, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.4264705882352941, "prompt_id": "79eb49a82dc5be50", "reading_canonical_log_prob": -184.84115600585938, "reading_distances": [20, 28, 20, 15], "reading_log_probs": [-189.62661743164062, -188.3463897705078, -186.88034057617188, -189.71487426757812], "reading_logprob_std": 1.3305838319748526, "tokenization_path_logprob_std": 279.40405008583065, "trajectory_balance_rmse": 3787.9273260960845} {"canonical_length": 355, "canonical_nll": 1.3042834644586268, "generation_rouge_l_f1": 0.2747603833865815, "generation_token_f1": 0.5367412140575081, "prompt_id": "144d91c0bdbf405c", "reading_canonical_log_prob": -463.0206298828125, "reading_distances": [30, 28, 30, 35], "reading_log_probs": [-461.38177490234375, -460.1920471191406, -460.5662841796875, -463.816162109375], "reading_logprob_std": 1.6289743140091617, "tokenization_path_logprob_std": 300.10739623921376, "trajectory_balance_rmse": 8862.343454807058} {"canonical_length": 194, "canonical_nll": 1.280319528481395, "generation_rouge_l_f1": 0.24778761061946902, "generation_token_f1": 0.47787610619469023, "prompt_id": "6bebc48cdc1b16ab", "reading_canonical_log_prob": -248.38198852539062, "reading_distances": [58, 61, 53, 54], "reading_log_probs": [-253.89756774902344, -252.38296508789062, -252.38819885253906, -251.75091552734375], "reading_logprob_std": 0.9122305777360277, "tokenization_path_logprob_std": 184.09373092257343, "trajectory_balance_rmse": 4952.059997060358} {"canonical_length": 136, "canonical_nll": 2.887205011704389, "generation_rouge_l_f1": 0.22382671480144406, "generation_token_f1": 0.34657039711191334, "prompt_id": "f19548b23e650d6d", "reading_canonical_log_prob": -392.6598815917969, "reading_distances": [20, 12, 22, 19], "reading_log_probs": [-433.4057312011719, -407.99261474609375, -419.2738037109375, -417.6055908203125], "reading_logprob_std": 10.478707642627539, "tokenization_path_logprob_std": 242.03050190083695, "trajectory_balance_rmse": 3659.881421413718} {"canonical_length": 283, "canonical_nll": 1.1085069204808966, "generation_rouge_l_f1": 0.18246869409660108, "generation_token_f1": 0.4221824686940966, "prompt_id": "b85ad6ca4cfd648f", "reading_canonical_log_prob": -313.70745849609375, "reading_distances": [25, 23, 24, 23], "reading_log_probs": [-311.58966064453125, -309.6329345703125, -310.9952392578125, -312.5929260253906], "reading_logprob_std": 1.23688727666002, "tokenization_path_logprob_std": 457.143574169099, "trajectory_balance_rmse": 7164.844134675457} {"canonical_length": 146, "canonical_nll": 1.6469344047650898, "generation_rouge_l_f1": 0.21195652173913043, "generation_token_f1": 0.4293478260869565, "prompt_id": "6b7b364a0fc8e714", "reading_canonical_log_prob": -240.45242309570312, "reading_distances": [33, 22, 34, 27], "reading_log_probs": [-236.94126892089844, -246.39146423339844, -242.20703125, -239.1857452392578], "reading_logprob_std": 4.088929699693596, "tokenization_path_logprob_std": 95.09324355642961, "trajectory_balance_rmse": 4303.23326908641} {"canonical_length": 112, "canonical_nll": 2.7381559099469865, "generation_rouge_l_f1": 0.1701030927835052, "generation_token_f1": 0.3402061855670104, "prompt_id": "1d91ffc1f9553c07", "reading_canonical_log_prob": -306.6734619140625, "reading_distances": [12, 16, 19, 16], "reading_log_probs": [-308.904052734375, -309.3714599609375, -309.1974182128906, -310.094970703125], "reading_logprob_std": 0.5068018370600273, "tokenization_path_logprob_std": 218.58945307653002, "trajectory_balance_rmse": 3023.127982211028} {"canonical_length": 37, "canonical_nll": 0.9467334231814822, "generation_rouge_l_f1": 0.26436781609195403, "generation_token_f1": 0.29885057471264365, "prompt_id": "1ac7a2e92a59b9ce", "reading_canonical_log_prob": -35.029136657714844, "reading_distances": [13, 20, 20, 13], "reading_log_probs": [-35.856048583984375, -38.14719009399414, -39.669349670410156, -38.01081466674805], "reading_logprob_std": 1.5684365319657925, "tokenization_path_logprob_std": 127.57446096827074, "trajectory_balance_rmse": 1222.4640082401627} {"canonical_length": 43, "canonical_nll": 2.607514492301054, "generation_rouge_l_f1": 0.19266055045871558, "generation_token_f1": 0.24770642201834864, "prompt_id": "66c94013abca2b3a", "reading_canonical_log_prob": -112.12312316894531, "reading_distances": [36, 24, 28, 26], "reading_log_probs": [-112.81989288330078, -111.7939453125, -116.93016052246094, -116.33550262451172], "reading_logprob_std": 2.5440526700026447, "tokenization_path_logprob_std": 171.31120775228916, "trajectory_balance_rmse": 1168.614573692035} {"canonical_length": 15, "canonical_nll": 0.6556199391682943, "generation_rouge_l_f1": 0.5806451612903226, "generation_token_f1": 0.7096774193548386, "prompt_id": "785090c8f7a0ea91", "reading_canonical_log_prob": -9.834299087524414, "reading_distances": [17, 16, 18, 13], "reading_log_probs": [-8.81997299194336, -8.907716751098633, -9.111028671264648, -8.479856491088867], "reading_logprob_std": 0.263133373959298, "tokenization_path_logprob_std": 70.65853096979833, "trajectory_balance_rmse": 493.3548734956112} {"canonical_length": 217, "canonical_nll": 1.274874498217886, "generation_rouge_l_f1": 0.1917808219178082, "generation_token_f1": 0.29680365296803657, "prompt_id": "30ed32c8494397b3", "reading_canonical_log_prob": -276.64776611328125, "reading_distances": [23, 23, 22, 15], "reading_log_probs": [-280.3376159667969, -277.53302001953125, -278.04071044921875, -280.1668701171875], "reading_logprob_std": 1.4400855985055059, "tokenization_path_logprob_std": 151.8058943733956, "trajectory_balance_rmse": 4843.387244896888} {"canonical_length": 16, "canonical_nll": 2.327626943588257, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.2105263157894737, "prompt_id": "e8d7fb5246616f5e", "reading_canonical_log_prob": -37.24203109741211, "reading_distances": [79, 78, 81, 78], "reading_log_probs": [-21.929834365844727, -18.479087829589844, -22.943038940429688, -21.408222198486328], "reading_logprob_std": 1.9163477529522097, "tokenization_path_logprob_std": 42.90401131150958, "trajectory_balance_rmse": 439.17390090369514} {"canonical_length": 6, "canonical_nll": 2.7304821014404297, "generation_rouge_l_f1": 0.6666666666666665, "generation_token_f1": 0.6666666666666665, "prompt_id": "96f83a49dbeabc97", "reading_canonical_log_prob": -16.382892608642578, "reading_distances": [31, 28, 32, 30], "reading_log_probs": [-13.925288200378418, -12.824983596801758, -12.793139457702637, -13.176776885986328], "reading_logprob_std": 0.526359385136691, "tokenization_path_logprob_std": 6.988519458146333, "trajectory_balance_rmse": 140.01400627952154} {"canonical_length": 80, "canonical_nll": 1.270307731628418, "generation_rouge_l_f1": 0.27848101265822783, "generation_token_f1": 0.5063291139240507, "prompt_id": "dea31d77a681f384", "reading_canonical_log_prob": -101.62461853027344, "reading_distances": [13, 18, 15, 14], "reading_log_probs": [-100.58824157714844, -102.68019104003906, -100.60606384277344, -100.4669418334961], "reading_logprob_std": 1.0650162138991088, "tokenization_path_logprob_std": 143.22041572411183, "trajectory_balance_rmse": 2020.1825329436385} {"canonical_length": 9, "canonical_nll": 1.8787379794650607, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "428bb97daf254741", "reading_canonical_log_prob": -16.908641815185547, "reading_distances": [21, 22, 25, 22], "reading_log_probs": [-13.591882705688477, -13.965075492858887, -14.030838012695312, -14.985502243041992], "reading_logprob_std": 0.5937949482668533, "tokenization_path_logprob_std": 83.06228126881595, "trajectory_balance_rmse": 339.7373794593097} {"canonical_length": 249, "canonical_nll": 1.5120148563002007, "generation_rouge_l_f1": 0.23170731707317074, "generation_token_f1": 0.45528455284552843, "prompt_id": "ed06fac728bea8f5", "reading_canonical_log_prob": -376.49169921875, "reading_distances": [21, 18, 19, 17], "reading_log_probs": [-375.613525390625, -378.00726318359375, -377.8033447265625, -376.05194091796875], "reading_logprob_std": 1.2127716438140204, "tokenization_path_logprob_std": 182.28132200569382, "trajectory_balance_rmse": 7418.432963033966} {"canonical_length": 42, "canonical_nll": 2.312636239188058, "generation_rouge_l_f1": 0.18367346938775508, "generation_token_f1": 0.28571428571428575, "prompt_id": "2455341860ece852", "reading_canonical_log_prob": -97.13072204589844, "reading_distances": [18, 16, 24, 16], "reading_log_probs": [-100.47020721435547, -98.98399353027344, -105.7564697265625, -98.7498779296875], "reading_logprob_std": 3.267599881949288, "tokenization_path_logprob_std": 179.28497844009874, "trajectory_balance_rmse": 1376.0704871139803} {"canonical_length": 56, "canonical_nll": 1.8670433589390345, "generation_rouge_l_f1": 0.3779527559055118, "generation_token_f1": 0.4881889763779528, "prompt_id": "2b713b1aba655ea9", "reading_canonical_log_prob": -104.55442810058594, "reading_distances": [19, 18, 10, 18], "reading_log_probs": [-108.87213134765625, -107.01129913330078, -108.19438171386719, -105.85177612304688], "reading_logprob_std": 1.3315581642321639, "tokenization_path_logprob_std": 178.1940883821597, "trajectory_balance_rmse": 1653.7132223358317} {"canonical_length": 21, "canonical_nll": 2.096569606236049, "generation_rouge_l_f1": 0.17142857142857143, "generation_token_f1": 0.22857142857142856, "prompt_id": "12ac8fa81a314c1b", "reading_canonical_log_prob": -44.02796173095703, "reading_distances": [27, 26, 32, 32], "reading_log_probs": [-43.16441345214844, -44.287391662597656, -44.597618103027344, -45.365753173828125], "reading_logprob_std": 0.9133358127823029, "tokenization_path_logprob_std": 55.45982521051078, "trajectory_balance_rmse": 567.4457803194997} {"canonical_length": 174, "canonical_nll": 1.3604811745128413, "generation_rouge_l_f1": 0.2544642857142857, "generation_token_f1": 0.45535714285714285, "prompt_id": "46239be323c13eb4", "reading_canonical_log_prob": -236.72372436523438, "reading_distances": [21, 15, 22, 21], "reading_log_probs": [-251.43321228027344, -239.82962036132812, -242.14852905273438, -247.5916748046875], "reading_logprob_std": 5.250878978888909, "tokenization_path_logprob_std": 256.6850155640857, "trajectory_balance_rmse": 3916.77048425461} {"canonical_length": 298, "canonical_nll": 1.311368289409868, "generation_rouge_l_f1": 0.2307692307692308, "generation_token_f1": 0.4578754578754579, "prompt_id": "3fde994857b8ea58", "reading_canonical_log_prob": -390.7877502441406, "reading_distances": [39, 36, 35, 35], "reading_log_probs": [-389.8282775878906, -389.4739685058594, -390.45355224609375, -390.6064147949219], "reading_logprob_std": 0.531302385444209, "tokenization_path_logprob_std": 268.886940495607, "trajectory_balance_rmse": 8609.298063250155} {"canonical_length": 111, "canonical_nll": 1.217918602195946, "generation_rouge_l_f1": 0.2985915492957747, "generation_token_f1": 0.4507042253521127, "prompt_id": "366758412bdbb007", "reading_canonical_log_prob": -135.18896484375, "reading_distances": [39, 36, 33, 39], "reading_log_probs": [-137.627197265625, -136.87472534179688, -130.68173217773438, -135.65818786621094], "reading_logprob_std": 3.1262455687991872, "tokenization_path_logprob_std": 147.98881546728947, "trajectory_balance_rmse": 3131.2949894193675} {"canonical_length": 17, "canonical_nll": 1.242495144114775, "generation_rouge_l_f1": 0.5142857142857142, "generation_token_f1": 0.5714285714285714, "prompt_id": "f77e57c7031a2b6c", "reading_canonical_log_prob": -21.122417449951172, "reading_distances": [10, 15, 16, 13], "reading_log_probs": [-20.92931365966797, -20.839860916137695, -21.17284393310547, -21.34317398071289], "reading_logprob_std": 0.22945646072476436, "tokenization_path_logprob_std": 29.844833717482732, "trajectory_balance_rmse": 672.678518099564} {"canonical_length": 8, "canonical_nll": 2.259507179260254, "generation_rouge_l_f1": 0.044444444444444446, "generation_token_f1": 0.06666666666666667, "prompt_id": "1f37960a14b0dc02", "reading_canonical_log_prob": -18.07605743408203, "reading_distances": [24, 23, 21, 30], "reading_log_probs": [-16.12492561340332, -15.464574813842773, -12.490511894226074, -18.818090438842773], "reading_logprob_std": 2.5985176956112896, "tokenization_path_logprob_std": 85.76161216211801, "trajectory_balance_rmse": 246.37895076173243} {"canonical_length": 28, "canonical_nll": 2.1271890912737166, "generation_rouge_l_f1": 0.23999999999999996, "generation_token_f1": 0.28, "prompt_id": "8d8a8351f2dceabb", "reading_canonical_log_prob": -59.56129455566406, "reading_distances": [20, 20, 21, 19], "reading_log_probs": [-76.77873229980469, -62.746734619140625, -69.70989227294922, -69.79066467285156], "reading_logprob_std": 5.728638484900937, "tokenization_path_logprob_std": 114.3911142823625, "trajectory_balance_rmse": 867.4775528737099} {"canonical_length": 2, "canonical_nll": 0.43566808104515076, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "7a9b64f9112cd0c5", "reading_canonical_log_prob": -0.8713361620903015, "reading_distances": [45, 42, 52, 41], "reading_log_probs": [-0.7870564460754395, -3.366713285446167, -1.8012471199035645, -3.220580577850342], "reading_logprob_std": 1.22786394401884, "tokenization_path_logprob_std": 0.0, "trajectory_balance_rmse": 0.8634589314460754} {"canonical_length": 5, "canonical_nll": 1.81256103515625, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "2797e2b7991dcf3e", "reading_canonical_log_prob": -9.06280517578125, "reading_distances": [28, 20, 20, 23], "reading_log_probs": [-10.795236587524414, -9.528938293457031, -9.383278846740723, -9.270211219787598], "reading_logprob_std": 0.708506590470984, "tokenization_path_logprob_std": 64.72352756163471, "trajectory_balance_rmse": 101.93362559920804} {"canonical_length": 313, "canonical_nll": 1.7532586533421526, "generation_rouge_l_f1": 0.20905923344947733, "generation_token_f1": 0.48780487804878053, "prompt_id": "d353ae8508e0bb05", "reading_canonical_log_prob": -548.7699584960938, "reading_distances": [42, 41, 40, 45], "reading_log_probs": [-550.5436401367188, -555.893310546875, -554.83056640625, -549.61181640625], "reading_logprob_std": 3.104928221285238, "tokenization_path_logprob_std": 317.1069036118831, "trajectory_balance_rmse": 8831.052486914059} {"canonical_length": 7, "canonical_nll": 0.004434865234153611, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1099a8369d87bee2", "reading_canonical_log_prob": -0.03104405663907528, "reading_distances": [20, 19, 25, 19], "reading_log_probs": [-0.035442911088466644, -0.04496409744024277, -0.033923789858818054, -0.037044957280159], "reading_logprob_std": 0.004914859520673748, "tokenization_path_logprob_std": 37.00337326181806, "trajectory_balance_rmse": 243.41298074194214} {"canonical_length": 264, "canonical_nll": 1.3848763668175899, "generation_rouge_l_f1": 0.248565965583174, "generation_token_f1": 0.508604206500956, "prompt_id": "d8d494771b6c2d81", "reading_canonical_log_prob": -365.60736083984375, "reading_distances": [25, 25, 31, 25], "reading_log_probs": [-364.7420654296875, -365.0410461425781, -366.5221862792969, -363.17230224609375], "reading_logprob_std": 1.373258868823144, "tokenization_path_logprob_std": 523.3436010329414, "trajectory_balance_rmse": 8270.215441618739} {"canonical_length": 372, "canonical_nll": 1.2857967909946237, "generation_rouge_l_f1": 0.2090032154340836, "generation_token_f1": 0.39871382636655944, "prompt_id": "27da9cdd8cf02e41", "reading_canonical_log_prob": -478.31640625, "reading_distances": [33, 34, 34, 31], "reading_log_probs": [-483.82916259765625, -472.529296875, -476.02301025390625, -479.1268615722656], "reading_logprob_std": 4.796721109364886, "tokenization_path_logprob_std": 309.35008993131567, "trajectory_balance_rmse": 9990.776692285148} {"canonical_length": 51, "canonical_nll": 1.1472601049086626, "generation_rouge_l_f1": 0.2992125984251968, "generation_token_f1": 0.4094488188976378, "prompt_id": "47574188e8957081", "reading_canonical_log_prob": -58.5102653503418, "reading_distances": [33, 28, 28, 33], "reading_log_probs": [-55.208251953125, -54.40907287597656, -50.06690979003906, -53.02992630004883], "reading_logprob_std": 2.261140594690518, "tokenization_path_logprob_std": 214.3575420930009, "trajectory_balance_rmse": 1454.3409992605523} {"canonical_length": 238, "canonical_nll": 2.119700199415704, "generation_rouge_l_f1": 0.18930041152263377, "generation_token_f1": 0.374485596707819, "prompt_id": "b7c16abbc7061a65", "reading_canonical_log_prob": -504.4886474609375, "reading_distances": [27, 30, 34, 27], "reading_log_probs": [-512.4869384765625, -522.0736083984375, -520.0156860351562, -519.947021484375], "reading_logprob_std": 4.213087245440845, "tokenization_path_logprob_std": 279.19931049058545, "trajectory_balance_rmse": 7268.248030062625} {"canonical_length": 334, "canonical_nll": 1.3915437252935534, "generation_rouge_l_f1": 0.23489932885906042, "generation_token_f1": 0.5234899328859061, "prompt_id": "01fef5f51abd900f", "reading_canonical_log_prob": -464.7756042480469, "reading_distances": [20, 31, 34, 33], "reading_log_probs": [-466.2173156738281, -471.678466796875, -467.3522644042969, -466.31475830078125], "reading_logprob_std": 2.5768777969108507, "tokenization_path_logprob_std": 543.1816207028563, "trajectory_balance_rmse": 10188.784990500453} {"canonical_length": 60, "canonical_nll": 1.1549455006917317, "generation_rouge_l_f1": 0.28181818181818186, "generation_token_f1": 0.3363636363636363, "prompt_id": "1d0839cb7b1b8564", "reading_canonical_log_prob": -69.2967300415039, "reading_distances": [29, 34, 29, 30], "reading_log_probs": [-71.4222640991211, -72.2933349609375, -73.24679565429688, -71.68498229980469], "reading_logprob_std": 0.8100998859948799, "tokenization_path_logprob_std": 95.9927963829377, "trajectory_balance_rmse": 1449.3498687079978} {"canonical_length": 198, "canonical_nll": 1.6703181411280776, "generation_rouge_l_f1": 0.2926829268292683, "generation_token_f1": 0.5658536585365854, "prompt_id": "b900859423a57fd5", "reading_canonical_log_prob": -330.7229919433594, "reading_distances": [19, 22, 22, 25], "reading_log_probs": [-348.3460693359375, -360.2449951171875, -336.12225341796875, -350.5205078125], "reading_logprob_std": 9.914284068404921, "tokenization_path_logprob_std": 273.58710466553475, "trajectory_balance_rmse": 5838.059990248693} {"canonical_length": 48, "canonical_nll": 1.5816073417663574, "generation_rouge_l_f1": 0.22916666666666666, "generation_token_f1": 0.3333333333333333, "prompt_id": "5fa2bbf4741331df", "reading_canonical_log_prob": -75.91715240478516, "reading_distances": [18, 16, 18, 16], "reading_log_probs": [-71.81155395507812, -73.0764389038086, -70.99237060546875, -75.76524353027344], "reading_logprob_std": 2.086783479795874, "tokenization_path_logprob_std": 20.93690085226958, "trajectory_balance_rmse": 1579.5589278516263} {"canonical_length": 501, "canonical_nll": 0.6540223995368638, "generation_rouge_l_f1": 0.21664050235478807, "generation_token_f1": 0.38932496075353223, "prompt_id": "97c9524265f7d241", "reading_canonical_log_prob": -327.66522216796875, "reading_distances": [22, 18, 21, 19], "reading_log_probs": [-322.61175537109375, -322.88397216796875, -321.3633728027344, -320.25653076171875], "reading_logprob_std": 1.2117622861861788, "tokenization_path_logprob_std": 293.25628622850803, "trajectory_balance_rmse": 9386.37595351217} {"canonical_length": 11, "canonical_nll": 1.7383495677601208, "generation_rouge_l_f1": 0.588235294117647, "generation_token_f1": 0.588235294117647, "prompt_id": "990e2b768a18fc67", "reading_canonical_log_prob": -19.121845245361328, "reading_distances": [25, 26, 23, 31], "reading_log_probs": [-22.158836364746094, -20.40327262878418, -21.00479507446289, -24.954513549804688], "reading_logprob_std": 2.018776970140656, "tokenization_path_logprob_std": 67.75702358186481, "trajectory_balance_rmse": 295.37244178965403} {"canonical_length": 30, "canonical_nll": 0.8866849263509115, "generation_rouge_l_f1": 0.2773722627737226, "generation_token_f1": 0.29197080291970806, "prompt_id": "4a59564b57b1ad24", "reading_canonical_log_prob": -26.600547790527344, "reading_distances": [31, 30, 32, 32], "reading_log_probs": [-27.614704132080078, -27.051532745361328, -27.25368881225586, -28.75539779663086], "reading_logprob_std": 0.7609111408034327, "tokenization_path_logprob_std": 37.13030291053252, "trajectory_balance_rmse": 764.9791927754173} {"canonical_length": 118, "canonical_nll": 1.712634199756687, "generation_rouge_l_f1": 0.22356495468277948, "generation_token_f1": 0.3202416918429003, "prompt_id": "fc01869b83f5b227", "reading_canonical_log_prob": -202.09083557128906, "reading_distances": [26, 26, 28, 33], "reading_log_probs": [-207.88827514648438, -210.00997924804688, -206.52285766601562, -212.39169311523438], "reading_logprob_std": 2.5645430716283744, "tokenization_path_logprob_std": 226.34302540547526, "trajectory_balance_rmse": 3309.905559190577} {"canonical_length": 10, "canonical_nll": 0.15234644412994386, "generation_rouge_l_f1": 0.3636363636363636, "generation_token_f1": 0.3636363636363636, "prompt_id": "e35706a135002934", "reading_canonical_log_prob": -1.5234644412994385, "reading_distances": [32, 30, 33, 36], "reading_log_probs": [-1.6978039741516113, -1.8953514099121094, -2.083930015563965, -2.6382927894592285], "reading_logprob_std": 0.4049155663000285, "tokenization_path_logprob_std": 13.164598937767932, "trajectory_balance_rmse": 294.59364608284034} {"canonical_length": 123, "canonical_nll": 1.0353151026780043, "generation_rouge_l_f1": 0.2558746736292428, "generation_token_f1": 0.38642297650130547, "prompt_id": "5df9638e97e9c3b6", "reading_canonical_log_prob": -127.34375762939453, "reading_distances": [11, 11, 10, 14], "reading_log_probs": [-131.6328887939453, -129.5624542236328, -131.561767578125, -130.43997192382812], "reading_logprob_std": 0.9891294526796985, "tokenization_path_logprob_std": 130.25403788072734, "trajectory_balance_rmse": 3107.96809957553} {"canonical_length": 25, "canonical_nll": 0.7769578552246094, "generation_rouge_l_f1": 0.12186379928315413, "generation_token_f1": 0.15053763440860216, "prompt_id": "6f6b5ba97837e4b7", "reading_canonical_log_prob": -19.423946380615234, "reading_distances": [27, 27, 33, 26], "reading_log_probs": [-19.24685287475586, -17.575103759765625, -18.852684020996094, -17.628143310546875], "reading_logprob_std": 0.8517070755932281, "tokenization_path_logprob_std": 76.32252511301867, "trajectory_balance_rmse": 603.6822296254411} {"canonical_length": 211, "canonical_nll": 0.9249015193414915, "generation_rouge_l_f1": 0.38658777120315585, "generation_token_f1": 0.5798816568047337, "prompt_id": "111777e66227bbb0", "reading_canonical_log_prob": -195.1542205810547, "reading_distances": [40, 43, 44, 36], "reading_log_probs": [-186.50582885742188, -182.2146759033203, -187.07852172851562, -184.33872985839844], "reading_logprob_std": 2.2194847333442373, "tokenization_path_logprob_std": 154.11163328655564, "trajectory_balance_rmse": 4929.080585856788} {"canonical_length": 264, "canonical_nll": 1.1496949629350142, "generation_rouge_l_f1": 0.22529644268774704, "generation_token_f1": 0.46640316205533594, "prompt_id": "7eae8a9735d3d72c", "reading_canonical_log_prob": -303.51947021484375, "reading_distances": [31, 30, 35, 31], "reading_log_probs": [-308.4244384765625, -305.19696044921875, -308.9067687988281, -309.3944396972656], "reading_logprob_std": 1.89757503713572, "tokenization_path_logprob_std": 146.5355769767535, "trajectory_balance_rmse": 7854.27230260914} {"canonical_length": 6, "canonical_nll": 0.004262791946530342, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "88814d5eacafd109", "reading_canonical_log_prob": -0.025576751679182053, "reading_distances": [38, 42, 36, 41], "reading_log_probs": [-0.024695327505469322, -0.04372430965304375, -0.019869137555360794, -0.07035457342863083], "reading_logprob_std": 0.022907750560964536, "tokenization_path_logprob_std": 19.30487321204902, "trajectory_balance_rmse": 151.17750578100234} {"canonical_length": 150, "canonical_nll": 1.4328563435872397, "generation_rouge_l_f1": 0.20183486238532108, "generation_token_f1": 0.4097859327217125, "prompt_id": "c6b7c5355ef042f7", "reading_canonical_log_prob": -214.92845153808594, "reading_distances": [40, 40, 38, 41], "reading_log_probs": [-220.84124755859375, -219.15679931640625, -220.12985229492188, -221.46531677246094], "reading_logprob_std": 0.991314859170896, "tokenization_path_logprob_std": 131.62419438296715, "trajectory_balance_rmse": 4232.35130020995} {"canonical_length": 21, "canonical_nll": 1.018840335664295, "generation_rouge_l_f1": 0.47368421052631576, "generation_token_f1": 0.47368421052631576, "prompt_id": "62bbc871380d57bb", "reading_canonical_log_prob": -21.395647048950195, "reading_distances": [13, 17, 17, 13], "reading_log_probs": [-22.94411849975586, -22.680713653564453, -22.809310913085938, -21.136316299438477], "reading_logprob_std": 0.844408837282748, "tokenization_path_logprob_std": 43.17533853606918, "trajectory_balance_rmse": 529.8552485283082} {"canonical_length": 28, "canonical_nll": 1.990302222115653, "generation_rouge_l_f1": 0.1283422459893048, "generation_token_f1": 0.1497326203208556, "prompt_id": "085c601e3edc073f", "reading_canonical_log_prob": -55.72846221923828, "reading_distances": [15, 11, 13, 11], "reading_log_probs": [-53.569297790527344, -53.294612884521484, -54.014888763427734, -54.48681640625], "reading_logprob_std": 0.5227097507166549, "tokenization_path_logprob_std": 75.17653576872068, "trajectory_balance_rmse": 796.6810779305987} {"canonical_length": 314, "canonical_nll": 1.5442012130834495, "generation_rouge_l_f1": 0.26943005181347146, "generation_token_f1": 0.5319516407599308, "prompt_id": "575d828d632e226f", "reading_canonical_log_prob": -484.8791809082031, "reading_distances": [37, 32, 28, 34], "reading_log_probs": [-493.7498779296875, -495.6769714355469, -496.2938232421875, -494.8124694824219], "reading_logprob_std": 1.1044126146638253, "tokenization_path_logprob_std": 488.7107597085075, "trajectory_balance_rmse": 9064.682313679454} {"canonical_length": 382, "canonical_nll": 1.0788560637628846, "generation_rouge_l_f1": 0.20415224913494812, "generation_token_f1": 0.38408304498269896, "prompt_id": "acf1157605893744", "reading_canonical_log_prob": -412.1230163574219, "reading_distances": [20, 17, 14, 14], "reading_log_probs": [-414.1856384277344, -414.2799987792969, -414.7330322265625, -414.193603515625], "reading_logprob_std": 0.2601752726012459, "tokenization_path_logprob_std": 323.1106725333693, "trajectory_balance_rmse": 10313.474259356315} {"canonical_length": 19, "canonical_nll": 1.6951450548673932, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.28571428571428575, "prompt_id": "c0621d0076f956e9", "reading_canonical_log_prob": -32.20775604248047, "reading_distances": [34, 34, 32, 34], "reading_log_probs": [-35.59221267700195, -35.00503158569336, -35.00621032714844, -34.80936813354492], "reading_logprob_std": 0.338877853162137, "tokenization_path_logprob_std": 50.74339966442059, "trajectory_balance_rmse": 401.30757527411674} {"canonical_length": 252, "canonical_nll": 1.2677224779885912, "generation_rouge_l_f1": 0.3173996175908222, "generation_token_f1": 0.508604206500956, "prompt_id": "71ec6348f5f044b8", "reading_canonical_log_prob": -319.466064453125, "reading_distances": [19, 20, 17, 19], "reading_log_probs": [-317.5160217285156, -319.48834228515625, -320.16571044921875, -319.154296875], "reading_logprob_std": 1.1250265009546352, "tokenization_path_logprob_std": 195.99999756715735, "trajectory_balance_rmse": 7522.880586107701} {"canonical_length": 13, "canonical_nll": 0.5290975937476525, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.3157894736842105, "prompt_id": "73a9260fbda0ad11", "reading_canonical_log_prob": -6.878268718719482, "reading_distances": [40, 30, 36, 38], "reading_log_probs": [-5.485265254974365, -4.765415668487549, -5.457995414733887, -6.6263227462768555], "reading_logprob_std": 0.7707447632277934, "tokenization_path_logprob_std": 58.597994652185534, "trajectory_balance_rmse": 229.73723242444677} {"canonical_length": 142, "canonical_nll": 1.0629313294316682, "generation_rouge_l_f1": 0.3395225464190981, "generation_token_f1": 0.4986737400530504, "prompt_id": "0e246b28f0fca81c", "reading_canonical_log_prob": -150.93624877929688, "reading_distances": [10, 7, 10, 8], "reading_log_probs": [-153.11843872070312, -150.16993713378906, -155.3941650390625, -153.67333984375], "reading_logprob_std": 2.173862427563908, "tokenization_path_logprob_std": 123.31995414290722, "trajectory_balance_rmse": 3836.432782484827} {"canonical_length": 38, "canonical_nll": 1.446101138466283, "generation_rouge_l_f1": 0.05128205128205128, "generation_token_f1": 0.05982905982905983, "prompt_id": "7dd4910bb043bf96", "reading_canonical_log_prob": -54.95184326171875, "reading_distances": [14, 9, 11, 12], "reading_log_probs": [-57.07174301147461, -56.46403121948242, -56.49977111816406, -57.2548942565918], "reading_logprob_std": 0.40072460259137377, "tokenization_path_logprob_std": 106.83539600688825, "trajectory_balance_rmse": 715.1901595414104} {"canonical_length": 187, "canonical_nll": 1.3717719909341577, "generation_rouge_l_f1": 0.23640661938534277, "generation_token_f1": 0.43498817966903075, "prompt_id": "b7fe1fa68d367af3", "reading_canonical_log_prob": -256.5213623046875, "reading_distances": [17, 13, 15, 15], "reading_log_probs": [-254.9174346923828, -253.179931640625, -259.44378662109375, -255.5321044921875], "reading_logprob_std": 2.645033884751136, "tokenization_path_logprob_std": 369.8663846272787, "trajectory_balance_rmse": 5330.302265139762} {"canonical_length": 96, "canonical_nll": 1.5339976946512859, "generation_rouge_l_f1": 0.28248587570621475, "generation_token_f1": 0.42372881355932207, "prompt_id": "b435d26419c9a382", "reading_canonical_log_prob": -147.26377868652344, "reading_distances": [41, 46, 48, 52], "reading_log_probs": [-147.94680786132812, -144.98611450195312, -147.8798065185547, -146.89646911621094], "reading_logprob_std": 1.3803159817496773, "tokenization_path_logprob_std": 303.0880197519214, "trajectory_balance_rmse": 3230.753077644222} {"canonical_length": 113, "canonical_nll": 1.5499254074771847, "generation_rouge_l_f1": 0.20496894409937888, "generation_token_f1": 0.37267080745341613, "prompt_id": "fc2765baa4652639", "reading_canonical_log_prob": -175.14157104492188, "reading_distances": [19, 26, 24, 21], "reading_log_probs": [-182.37522888183594, -184.31210327148438, -186.0296630859375, -179.90542602539062], "reading_logprob_std": 2.6312448502614583, "tokenization_path_logprob_std": 204.5858373245982, "trajectory_balance_rmse": 2921.248133620476} {"canonical_length": 216, "canonical_nll": 1.6369722154405382, "generation_rouge_l_f1": 0.22269807280513917, "generation_token_f1": 0.43683083511777304, "prompt_id": "6389a5bbf9eaba60", "reading_canonical_log_prob": -353.58599853515625, "reading_distances": [34, 29, 31, 26], "reading_log_probs": [-365.9650573730469, -361.1938781738281, -357.6417236328125, -359.8412170410156], "reading_logprob_std": 3.521703200522, "tokenization_path_logprob_std": 195.5361673271667, "trajectory_balance_rmse": 6149.3204444184585} {"canonical_length": 13, "canonical_nll": 0.5475153923034668, "generation_rouge_l_f1": 0.14285714285714288, "generation_token_f1": 0.14285714285714288, "prompt_id": "286cf1906a688fcb", "reading_canonical_log_prob": -7.117700099945068, "reading_distances": [25, 25, 24, 24], "reading_log_probs": [-6.3650665283203125, -7.007743835449219, -6.28806734085083, -7.753907203674316], "reading_logprob_std": 0.6813725088554726, "tokenization_path_logprob_std": 22.55011631687379, "trajectory_balance_rmse": 330.69320019064685} {"canonical_length": 93, "canonical_nll": 1.4081990231749832, "generation_rouge_l_f1": 0.3076923076923077, "generation_token_f1": 0.4253393665158371, "prompt_id": "33333486aa163f12", "reading_canonical_log_prob": -130.96250915527344, "reading_distances": [34, 26, 32, 28], "reading_log_probs": [-126.56620788574219, -127.75242614746094, -129.3839874267578, -129.42507934570312], "reading_logprob_std": 1.3838831479534242, "tokenization_path_logprob_std": 178.84709110696315, "trajectory_balance_rmse": 2850.379664576001} {"canonical_length": 60, "canonical_nll": 2.899652862548828, "generation_rouge_l_f1": 0.2550335570469799, "generation_token_f1": 0.2751677852348993, "prompt_id": "c5556f6cb2c9f8dd", "reading_canonical_log_prob": -173.9791717529297, "reading_distances": [25, 23, 27, 23], "reading_log_probs": [-173.2148895263672, -166.6995849609375, -169.52505493164062, -174.95652770996094], "reading_logprob_std": 3.7053863624113927, "tokenization_path_logprob_std": 64.54059764737286, "trajectory_balance_rmse": 504.5217802975881} {"canonical_length": 11, "canonical_nll": 2.582017031582919, "generation_rouge_l_f1": 0.5333333333333333, "generation_token_f1": 0.5333333333333333, "prompt_id": "b1950a5b21116b51", "reading_canonical_log_prob": -28.40218734741211, "reading_distances": [17, 18, 12, 19], "reading_log_probs": [-27.70709800720215, -28.866546630859375, -32.4766845703125, -25.600379943847656], "reading_logprob_std": 2.879757573383758, "tokenization_path_logprob_std": 43.351945864968684, "trajectory_balance_rmse": 296.81544571187214} {"canonical_length": 263, "canonical_nll": 1.607501592019665, "generation_rouge_l_f1": 0.2239089184060721, "generation_token_f1": 0.4629981024667932, "prompt_id": "355011c0ec753a64", "reading_canonical_log_prob": -422.7729187011719, "reading_distances": [18, 21, 20, 18], "reading_log_probs": [-419.24591064453125, -421.54852294921875, -420.310791015625, -420.4259033203125], "reading_logprob_std": 0.9413590511715068, "tokenization_path_logprob_std": 326.29329531543294, "trajectory_balance_rmse": 8805.689820723977} {"canonical_length": 29, "canonical_nll": 1.1601895299451104, "generation_rouge_l_f1": 0.2413793103448276, "generation_token_f1": 0.2931034482758621, "prompt_id": "c2e4ff98064c8812", "reading_canonical_log_prob": -33.6454963684082, "reading_distances": [25, 22, 22, 29], "reading_log_probs": [-29.723649978637695, -27.84381866455078, -27.341276168823242, -28.420631408691406], "reading_logprob_std": 1.0270344755809646, "tokenization_path_logprob_std": 165.9305849549563, "trajectory_balance_rmse": 871.2439552326344} {"canonical_length": 57, "canonical_nll": 2.157245033665707, "generation_rouge_l_f1": 0.14482758620689656, "generation_token_f1": 0.23448275862068962, "prompt_id": "e95bb93274990ecf", "reading_canonical_log_prob": -122.96296691894531, "reading_distances": [28, 30, 27, 28], "reading_log_probs": [-124.64385986328125, -125.90652465820312, -124.37325286865234, -125.09164428710938], "reading_logprob_std": 0.6707644923294069, "tokenization_path_logprob_std": 198.94989935752085, "trajectory_balance_rmse": 1491.182688454082} {"canonical_length": 40, "canonical_nll": 1.504625415802002, "generation_rouge_l_f1": 0.35616438356164376, "generation_token_f1": 0.4383561643835617, "prompt_id": "0eb7b4a395752869", "reading_canonical_log_prob": -60.18501663208008, "reading_distances": [30, 29, 36, 29], "reading_log_probs": [-73.94828796386719, -64.59696197509766, -70.30781555175781, -74.46699523925781], "reading_logprob_std": 4.548805477372619, "tokenization_path_logprob_std": 107.65246085183828, "trajectory_balance_rmse": 865.0942967380669} {"canonical_length": 124, "canonical_nll": 2.044554187405494, "generation_rouge_l_f1": 0.32669322709163345, "generation_token_f1": 0.43027888446215135, "prompt_id": "5d63d37b660c138f", "reading_canonical_log_prob": -253.52471923828125, "reading_distances": [38, 28, 37, 39], "reading_log_probs": [-253.46849060058594, -254.65464782714844, -252.7894287109375, -254.63401794433594], "reading_logprob_std": 0.9178110457605677, "tokenization_path_logprob_std": 92.41080653633996, "trajectory_balance_rmse": 3695.846908269589} {"canonical_length": 241, "canonical_nll": 1.2233302959268024, "generation_rouge_l_f1": 0.27504911591355596, "generation_token_f1": 0.5304518664047152, "prompt_id": "17f90ddfcc762d15", "reading_canonical_log_prob": -294.8226013183594, "reading_distances": [23, 25, 31, 31], "reading_log_probs": [-288.8666687011719, -286.8943786621094, -289.54583740234375, -287.5746765136719], "reading_logprob_std": 1.2041233257171373, "tokenization_path_logprob_std": 164.80715536352562, "trajectory_balance_rmse": 7300.570259397494} {"canonical_length": 61, "canonical_nll": 1.690095495005123, "generation_rouge_l_f1": 0.3312101910828026, "generation_token_f1": 0.4713375796178344, "prompt_id": "9e46c646f8e1ae11", "reading_canonical_log_prob": -103.0958251953125, "reading_distances": [21, 19, 23, 26], "reading_log_probs": [-105.80960083007812, -104.69085693359375, -106.9386978149414, -104.46111297607422], "reading_logprob_std": 1.1397684321534918, "tokenization_path_logprob_std": 88.2667485174789, "trajectory_balance_rmse": 1876.5605378283785} {"canonical_length": 317, "canonical_nll": 1.143769562056388, "generation_rouge_l_f1": 0.28130360205831906, "generation_token_f1": 0.5214408233276158, "prompt_id": "c405755d9e3e8c9b", "reading_canonical_log_prob": -362.574951171875, "reading_distances": [23, 26, 23, 21], "reading_log_probs": [-356.1932373046875, -355.42083740234375, -357.51617431640625, -353.04718017578125], "reading_logprob_std": 1.8761892549803296, "tokenization_path_logprob_std": 256.02453810777405, "trajectory_balance_rmse": 8127.47027291127} {"canonical_length": 225, "canonical_nll": 0.5559396023220486, "generation_rouge_l_f1": 0.28378378378378377, "generation_token_f1": 0.46396396396396394, "prompt_id": "d5c04061128f24af", "reading_canonical_log_prob": -125.08641052246094, "reading_distances": [27, 26, 23, 22], "reading_log_probs": [-126.56523132324219, -125.50274658203125, -125.66580200195312, -128.04689025878906], "reading_logprob_std": 1.16554692224227, "tokenization_path_logprob_std": 247.04252035919228, "trajectory_balance_rmse": 4492.190145379174} {"canonical_length": 61, "canonical_nll": 0.9115680632044058, "generation_rouge_l_f1": 0.4732824427480916, "generation_token_f1": 0.5038167938931297, "prompt_id": "1419b413999c869f", "reading_canonical_log_prob": -55.60565185546875, "reading_distances": [17, 13, 12, 12], "reading_log_probs": [-57.31779479980469, -62.17238235473633, -58.672332763671875, -54.189456939697266], "reading_logprob_std": 3.307339214612949, "tokenization_path_logprob_std": 28.01669746834577, "trajectory_balance_rmse": 913.9761818979688} {"canonical_length": 116, "canonical_nll": 1.572930960819639, "generation_rouge_l_f1": 0.26881720430107525, "generation_token_f1": 0.4139784946236559, "prompt_id": "3501affcfd13e96c", "reading_canonical_log_prob": -182.45999145507812, "reading_distances": [24, 21, 25, 18], "reading_log_probs": [-192.1694793701172, -188.31573486328125, -190.0960693359375, -194.71795654296875], "reading_logprob_std": 2.756280501256156, "tokenization_path_logprob_std": 169.60816799999762, "trajectory_balance_rmse": 3301.514113091162} {"canonical_length": 26, "canonical_nll": 2.1419991713303785, "generation_rouge_l_f1": 0.21621621621621623, "generation_token_f1": 0.3243243243243243, "prompt_id": "6fdd3868e470d9e0", "reading_canonical_log_prob": -55.691978454589844, "reading_distances": [57, 60, 47, 39], "reading_log_probs": [-70.28951263427734, -67.04226684570312, -58.721458435058594, -56.540138244628906], "reading_logprob_std": 6.568222075566607, "tokenization_path_logprob_std": 92.6235413078303, "trajectory_balance_rmse": 830.5662332740577} {"canonical_length": 304, "canonical_nll": 0.9282157295628598, "generation_rouge_l_f1": 0.34024896265560167, "generation_token_f1": 0.5518672199170125, "prompt_id": "1d8b10c0b2c4c1d8", "reading_canonical_log_prob": -282.1775817871094, "reading_distances": [27, 19, 19, 23], "reading_log_probs": [-287.47918701171875, -283.7999267578125, -283.9637451171875, -289.4630432128906], "reading_logprob_std": 2.771446330299651, "tokenization_path_logprob_std": 256.11496170432474, "trajectory_balance_rmse": 6233.537782972331} {"canonical_length": 31, "canonical_nll": 0.9184464485414566, "generation_rouge_l_f1": 0.16521739130434782, "generation_token_f1": 0.1826086956521739, "prompt_id": "6f83d657c17ec17d", "reading_canonical_log_prob": -28.471839904785156, "reading_distances": [16, 19, 23, 24], "reading_log_probs": [-27.09270477294922, -28.555370330810547, -27.672069549560547, -27.33980941772461], "reading_logprob_std": 0.6392912861643528, "tokenization_path_logprob_std": 110.181672139445, "trajectory_balance_rmse": 672.6810121119431} {"canonical_length": 39, "canonical_nll": 1.624630854679988, "generation_rouge_l_f1": 0.09756097560975609, "generation_token_f1": 0.09756097560975609, "prompt_id": "6587efe9dc00b7ac", "reading_canonical_log_prob": -63.36060333251953, "reading_distances": [24, 22, 21, 23], "reading_log_probs": [-66.19291687011719, -69.49569702148438, -70.58963775634766, -65.05659484863281], "reading_logprob_std": 2.630710253702297, "tokenization_path_logprob_std": 140.6196055936146, "trajectory_balance_rmse": 1286.8230926545887} {"canonical_length": 15, "canonical_nll": 0.7587185541788737, "generation_rouge_l_f1": 0.6153846153846153, "generation_token_f1": 0.6153846153846153, "prompt_id": "ea95653fa930225e", "reading_canonical_log_prob": -11.380778312683105, "reading_distances": [15, 23, 15, 16], "reading_log_probs": [-11.470185279846191, -12.944284439086914, -11.460390090942383, -12.098749160766602], "reading_logprob_std": 0.7007455241124703, "tokenization_path_logprob_std": 93.18669187723513, "trajectory_balance_rmse": 620.5284511158872} {"canonical_length": 218, "canonical_nll": 1.7287303679579988, "generation_rouge_l_f1": 0.29237288135593226, "generation_token_f1": 0.5169491525423728, "prompt_id": "dbfaa5f0bb7fb12c", "reading_canonical_log_prob": -376.86322021484375, "reading_distances": [38, 42, 40, 45], "reading_log_probs": [-380.13275146484375, -381.4317626953125, -381.7730712890625, -387.3833312988281], "reading_logprob_std": 3.214053321058496, "tokenization_path_logprob_std": 279.0553719470622, "trajectory_balance_rmse": 6949.153570170122} {"canonical_length": 236, "canonical_nll": 1.3416344594147245, "generation_rouge_l_f1": 0.23434343434343435, "generation_token_f1": 0.5373737373737374, "prompt_id": "6ef66ee37f69a6e9", "reading_canonical_log_prob": -316.625732421875, "reading_distances": [37, 45, 46, 48], "reading_log_probs": [-316.43426513671875, -319.687744140625, -314.5747985839844, -315.3641357421875], "reading_logprob_std": 2.248087583706187, "tokenization_path_logprob_std": 314.3089734867492, "trajectory_balance_rmse": 7606.170425720737} {"canonical_length": 254, "canonical_nll": 0.43768658975916586, "generation_rouge_l_f1": 0.18972332015810275, "generation_token_f1": 0.3557312252964428, "prompt_id": "e2bfc09d77b30452", "reading_canonical_log_prob": -111.17239379882812, "reading_distances": [21, 26, 24, 23], "reading_log_probs": [-115.17080688476562, -117.7619400024414, -118.98574829101562, -113.92615509033203], "reading_logprob_std": 2.320692117156292, "tokenization_path_logprob_std": 54.599049821521234, "trajectory_balance_rmse": 4405.89948407052} {"canonical_length": 315, "canonical_nll": 0.933257572234623, "generation_rouge_l_f1": 0.2078364565587734, "generation_token_f1": 0.4190800681431005, "prompt_id": "3024b6f11a619ed8", "reading_canonical_log_prob": -293.97613525390625, "reading_distances": [16, 13, 15, 12], "reading_log_probs": [-293.08526611328125, -291.68194580078125, -292.8960876464844, -292.9388427734375], "reading_logprob_std": 0.6507876387159849, "tokenization_path_logprob_std": 142.78590417513564, "trajectory_balance_rmse": 8004.942638543664} {"canonical_length": 147, "canonical_nll": 1.7602559822757229, "generation_rouge_l_f1": 0.21307506053268763, "generation_token_f1": 0.42615012106537525, "prompt_id": "2027b774c4dd5ef1", "reading_canonical_log_prob": -258.75762939453125, "reading_distances": [79, 81, 68, 82], "reading_log_probs": [-255.42172241210938, -254.62777709960938, -256.2268371582031, -268.9966735839844], "reading_logprob_std": 6.81694432904321, "tokenization_path_logprob_std": 53.60095880291985, "trajectory_balance_rmse": 4187.676901893961} {"canonical_length": 320, "canonical_nll": 1.0962640762329101, "generation_rouge_l_f1": 0.12432432432432433, "generation_token_f1": 0.1783783783783784, "prompt_id": "ea4a6ac8af4f585f", "reading_canonical_log_prob": -350.80450439453125, "reading_distances": [29, 33, 37, 34], "reading_log_probs": [-356.5821533203125, -359.04010009765625, -353.9737548828125, -355.3620300292969], "reading_logprob_std": 2.149753992909485, "tokenization_path_logprob_std": 247.7413439138289, "trajectory_balance_rmse": 9078.533705738446} {"canonical_length": 142, "canonical_nll": 1.44507050850022, "generation_rouge_l_f1": 0.24258760107816713, "generation_token_f1": 0.44204851752021557, "prompt_id": "e9a730206e7c88d2", "reading_canonical_log_prob": -205.20001220703125, "reading_distances": [33, 22, 21, 22], "reading_log_probs": [-214.50460815429688, -216.94161987304688, -212.94918823242188, -215.23086547851562], "reading_logprob_std": 1.6572580017025464, "tokenization_path_logprob_std": 52.3198347375079, "trajectory_balance_rmse": 3535.821132198824} {"canonical_length": 324, "canonical_nll": 1.0437579684787326, "generation_rouge_l_f1": 0.20872865275142316, "generation_token_f1": 0.46299810246679324, "prompt_id": "21922a55d95c3a47", "reading_canonical_log_prob": -338.1775817871094, "reading_distances": [18, 19, 18, 19], "reading_log_probs": [-333.2730712890625, -330.8292541503906, -330.7243957519531, -332.7673034667969], "reading_logprob_std": 1.3122592686749313, "tokenization_path_logprob_std": 113.55014457805414, "trajectory_balance_rmse": 8938.043700666878} {"canonical_length": 17, "canonical_nll": 0.5711383819580078, "generation_rouge_l_f1": 0.34146341463414637, "generation_token_f1": 0.34146341463414637, "prompt_id": "732a8e545760e4c1", "reading_canonical_log_prob": -9.709352493286133, "reading_distances": [24, 22, 16, 17], "reading_log_probs": [-8.473152160644531, -8.972457885742188, -9.742500305175781, -9.412049293518066], "reading_logprob_std": 0.5505665680219405, "tokenization_path_logprob_std": 72.79131900912505, "trajectory_balance_rmse": 446.734694446601} {"canonical_length": 8, "canonical_nll": 0.8304320573806763, "generation_rouge_l_f1": 0.875, "generation_token_f1": 0.875, "prompt_id": "530dfa24f61035c2", "reading_canonical_log_prob": -6.64345645904541, "reading_distances": [34, 32, 26, 35], "reading_log_probs": [-6.239013195037842, -6.987544059753418, -6.070883274078369, -7.117207050323486], "reading_logprob_std": 0.5253303329865381, "tokenization_path_logprob_std": 45.238897600030185, "trajectory_balance_rmse": 140.89421702018126} {"canonical_length": 8, "canonical_nll": 1.6756858825683594, "generation_rouge_l_f1": 0.5714285714285714, "generation_token_f1": 0.5714285714285714, "prompt_id": "ab024353313a6e65", "reading_canonical_log_prob": -13.405487060546875, "reading_distances": [27, 29, 36, 33], "reading_log_probs": [-11.900307655334473, -12.984047889709473, -12.362205505371094, -11.80147647857666], "reading_logprob_std": 0.5398449048761668, "tokenization_path_logprob_std": 12.853284879514975, "trajectory_balance_rmse": 160.17172116224498} {"canonical_length": 12, "canonical_nll": 1.1973281701405842, "generation_rouge_l_f1": 0.19354838709677422, "generation_token_f1": 0.21505376344086022, "prompt_id": "2321d8df80c63fef", "reading_canonical_log_prob": -14.367938041687012, "reading_distances": [20, 20, 20, 16], "reading_log_probs": [-14.16826057434082, -15.597452163696289, -15.006377220153809, -15.088689804077148], "reading_logprob_std": 0.5921150062098969, "tokenization_path_logprob_std": 59.462453616386085, "trajectory_balance_rmse": 307.0075010725644} {"canonical_length": 16, "canonical_nll": 0.38307514786720276, "generation_rouge_l_f1": 0.5263157894736842, "generation_token_f1": 0.5263157894736842, "prompt_id": "f3f8000f2b6cf3b2", "reading_canonical_log_prob": -6.129202365875244, "reading_distances": [5, 7, 4, 12], "reading_log_probs": [-6.250858783721924, -5.5841217041015625, -7.064619541168213, -4.692526817321777], "reading_logprob_std": 1.0061801776778054, "tokenization_path_logprob_std": 27.286862782081084, "trajectory_balance_rmse": 273.04671622422086} {"canonical_length": 8, "canonical_nll": 0.0324983224272728, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "30a8ec8b58b46182", "reading_canonical_log_prob": -0.2599865794181824, "reading_distances": [28, 24, 26, 26], "reading_log_probs": [-0.43533089756965637, -0.525145411491394, -0.8539330363273621, -0.6314018964767456], "reading_logprob_std": 0.1804277521950456, "tokenization_path_logprob_std": 31.43061078217761, "trajectory_balance_rmse": 72.57919263243646} {"canonical_length": 267, "canonical_nll": 1.5321542118372542, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.48496240601503765, "prompt_id": "f1f8d8a792fc708b", "reading_canonical_log_prob": -409.0851745605469, "reading_distances": [24, 28, 29, 30], "reading_log_probs": [-413.04193115234375, -418.66217041015625, -413.26171875, -411.18963623046875], "reading_logprob_std": 3.21925927749399, "tokenization_path_logprob_std": 353.846494667126, "trajectory_balance_rmse": 8036.752763224568} {"canonical_length": 194, "canonical_nll": 1.113485277313547, "generation_rouge_l_f1": 0.23504273504273504, "generation_token_f1": 0.46153846153846156, "prompt_id": "1d19c88736fcd9d4", "reading_canonical_log_prob": -216.01614379882812, "reading_distances": [52, 48, 29, 47], "reading_log_probs": [-224.91656494140625, -219.64385986328125, -216.35110473632812, -223.35508728027344], "reading_logprob_std": 3.843587263161852, "tokenization_path_logprob_std": 237.4444609534832, "trajectory_balance_rmse": 4992.383075226764} {"canonical_length": 15, "canonical_nll": 2.6931299845377605, "generation_rouge_l_f1": 0.10810810810810811, "generation_token_f1": 0.10810810810810811, "prompt_id": "1dc2cb66ddb80668", "reading_canonical_log_prob": -40.396949768066406, "reading_distances": [15, 15, 15, 13], "reading_log_probs": [-44.536407470703125, -45.414573669433594, -52.016788482666016, -49.578819274902344], "reading_logprob_std": 3.5240491450483318, "tokenization_path_logprob_std": 50.430872695814536, "trajectory_balance_rmse": 418.56359689708927} {"canonical_length": 16, "canonical_nll": 1.7614198923110962, "generation_rouge_l_f1": 0.07627118644067797, "generation_token_f1": 0.1016949152542373, "prompt_id": "bf66528e05d6425b", "reading_canonical_log_prob": -28.18271827697754, "reading_distances": [29, 37, 31, 36], "reading_log_probs": [-32.001014709472656, -35.24544906616211, -34.632049560546875, -34.013675689697266], "reading_logprob_std": 1.4075810738153733, "tokenization_path_logprob_std": 111.82056569874517, "trajectory_balance_rmse": 742.3168110460895} {"canonical_length": 118, "canonical_nll": 1.2832307977191473, "generation_rouge_l_f1": 0.2909090909090909, "generation_token_f1": 0.4727272727272727, "prompt_id": "471e2be2592f5add", "reading_canonical_log_prob": -151.42123413085938, "reading_distances": [20, 22, 21, 20], "reading_log_probs": [-153.18228149414062, -147.592041015625, -150.6733856201172, -149.17042541503906], "reading_logprob_std": 2.3784660248387315, "tokenization_path_logprob_std": 177.74310242869274, "trajectory_balance_rmse": 3631.157740383482} {"canonical_length": 19, "canonical_nll": 7.253118414627878, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "57103ee2e4cc89e4", "reading_canonical_log_prob": -137.8092498779297, "reading_distances": [36, 32, 40, 39], "reading_log_probs": [-137.5208740234375, -140.0544891357422, -137.63551330566406, -136.42945861816406], "reading_logprob_std": 1.5294441669637127, "tokenization_path_logprob_std": 49.29525101435895, "trajectory_balance_rmse": 398.6252335418076} {"canonical_length": 23, "canonical_nll": 1.8061687635338826, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.25, "prompt_id": "aa0821e4d154fd43", "reading_canonical_log_prob": -41.5418815612793, "reading_distances": [20, 22, 25, 21], "reading_log_probs": [-48.31523132324219, -48.62266540527344, -41.23455047607422, -46.63677215576172], "reading_logprob_std": 3.4249062691936447, "tokenization_path_logprob_std": 53.66896917079467, "trajectory_balance_rmse": 675.6748121824938} {"canonical_length": 24, "canonical_nll": 1.6701297760009766, "generation_rouge_l_f1": 0.31578947368421045, "generation_token_f1": 0.38596491228070173, "prompt_id": "2aaa42d342162266", "reading_canonical_log_prob": -40.08311462402344, "reading_distances": [31, 32, 27, 28], "reading_log_probs": [-40.643463134765625, -41.12293243408203, -40.14564895629883, -42.05662155151367], "reading_logprob_std": 0.8141151177667202, "tokenization_path_logprob_std": 95.4297412609509, "trajectory_balance_rmse": 927.9267358057176} {"canonical_length": 10, "canonical_nll": 0.39380435943603515, "generation_rouge_l_f1": 0.375, "generation_token_f1": 0.625, "prompt_id": "b5604eaaebf6e52c", "reading_canonical_log_prob": -3.9380435943603516, "reading_distances": [19, 20, 12, 15], "reading_log_probs": [-4.450503349304199, -3.5026750564575195, -3.1816868782043457, -5.260514259338379], "reading_logprob_std": 0.943351593112956, "tokenization_path_logprob_std": 62.88461448134764, "trajectory_balance_rmse": 290.1013459529093} {"canonical_length": 223, "canonical_nll": 1.3562868400539518, "generation_rouge_l_f1": 0.17610062893081757, "generation_token_f1": 0.4528301886792453, "prompt_id": "5dbcfc1c200ae9f7", "reading_canonical_log_prob": -302.45196533203125, "reading_distances": [23, 24, 25, 24], "reading_log_probs": [-294.6158142089844, -297.4414978027344, -295.57720947265625, -294.793212890625], "reading_logprob_std": 1.2924096845058712, "tokenization_path_logprob_std": 389.56438654387665, "trajectory_balance_rmse": 7171.381446430729} {"canonical_length": 163, "canonical_nll": 1.3521293218881807, "generation_rouge_l_f1": 0.22356495468277943, "generation_token_f1": 0.4652567975830816, "prompt_id": "d526cee0379fafdd", "reading_canonical_log_prob": -220.39707946777344, "reading_distances": [34, 29, 22, 30], "reading_log_probs": [-220.32997131347656, -219.241455078125, -221.55218505859375, -219.0914306640625], "reading_logprob_std": 1.1412693784877248, "tokenization_path_logprob_std": 172.0112264627939, "trajectory_balance_rmse": 5644.6702791184125} {"canonical_length": 85, "canonical_nll": 1.6593128877527574, "generation_rouge_l_f1": 0.25233644859813087, "generation_token_f1": 0.47663551401869164, "prompt_id": "b641694c0de50f89", "reading_canonical_log_prob": -141.04159545898438, "reading_distances": [35, 25, 31, 37], "reading_log_probs": [-152.8653106689453, -144.1995391845703, -151.937744140625, -146.00064086914062], "reading_logprob_std": 4.295856967480018, "tokenization_path_logprob_std": 73.09994156904055, "trajectory_balance_rmse": 2445.8082818144744} {"canonical_length": 13, "canonical_nll": 0.555191443516658, "generation_rouge_l_f1": 0.16296296296296298, "generation_token_f1": 0.17777777777777776, "prompt_id": "2e5bfbd760499e1d", "reading_canonical_log_prob": -7.217488765716553, "reading_distances": [22, 24, 25, 19], "reading_log_probs": [-11.202801704406738, -5.926815509796143, -4.375053405761719, -5.222113609313965], "reading_logprob_std": 3.080109422172348, "tokenization_path_logprob_std": 43.805881070881476, "trajectory_balance_rmse": 455.74721274286077} {"canonical_length": 363, "canonical_nll": 1.4474309075305614, "generation_rouge_l_f1": 0.24065040650406502, "generation_token_f1": 0.48780487804878053, "prompt_id": "1d0a6d1714584f14", "reading_canonical_log_prob": -525.4174194335938, "reading_distances": [36, 39, 33, 28], "reading_log_probs": [-523.486572265625, -521.331787109375, -521.8941040039062, -524.5657958984375], "reading_logprob_std": 1.4792094904799271, "tokenization_path_logprob_std": 277.9226184644224, "trajectory_balance_rmse": 9747.007270204967} {"canonical_length": 37, "canonical_nll": 2.334625244140625, "generation_rouge_l_f1": 0.05405405405405406, "generation_token_f1": 0.05405405405405406, "prompt_id": "c7b4cbd5453e19a7", "reading_canonical_log_prob": -86.38113403320312, "reading_distances": [24, 21, 16, 21], "reading_log_probs": [-82.35124969482422, -82.872802734375, -79.07067108154297, -81.31842803955078], "reading_logprob_std": 1.6838851732980094, "tokenization_path_logprob_std": 79.5917686220901, "trajectory_balance_rmse": 1141.5939594301942} {"canonical_length": 10, "canonical_nll": 1.235848617553711, "generation_rouge_l_f1": 0.7142857142857143, "generation_token_f1": 0.7142857142857143, "prompt_id": "88beef2f88591814", "reading_canonical_log_prob": -12.35848617553711, "reading_distances": [62, 63, 57, 70], "reading_log_probs": [-15.50097942352295, -17.498821258544922, -14.71098518371582, -16.464725494384766], "reading_logprob_std": 1.2062772792297698, "tokenization_path_logprob_std": 38.0642254456188, "trajectory_balance_rmse": 390.7137295434329} {"canonical_length": 65, "canonical_nll": 0.7862985464242789, "generation_rouge_l_f1": 0.728476821192053, "generation_token_f1": 0.7417218543046358, "prompt_id": "cfc50376a0f13c13", "reading_canonical_log_prob": -51.109405517578125, "reading_distances": [28, 24, 30, 17], "reading_log_probs": [-50.33262634277344, -52.819854736328125, -52.594688415527344, -51.977779388427734], "reading_logprob_std": 1.1236149235689932, "tokenization_path_logprob_std": 62.863642634459595, "trajectory_balance_rmse": 906.7491690681159} {"canonical_length": 90, "canonical_nll": 1.7596037122938368, "generation_rouge_l_f1": 0.1853035143769968, "generation_token_f1": 0.3067092651757189, "prompt_id": "6a7a68d7cf446159", "reading_canonical_log_prob": -158.3643341064453, "reading_distances": [21, 18, 14, 18], "reading_log_probs": [-159.21632385253906, -160.61688232421875, -162.1537322998047, -160.59017944335938], "reading_logprob_std": 1.2001642258906489, "tokenization_path_logprob_std": 116.19194295054616, "trajectory_balance_rmse": 2973.9458445334235} {"canonical_length": 216, "canonical_nll": 1.7005221048990886, "generation_rouge_l_f1": 0.21473684210526317, "generation_token_f1": 0.44631578947368417, "prompt_id": "a929bd728f13cca9", "reading_canonical_log_prob": -367.3127746582031, "reading_distances": [35, 36, 29, 38], "reading_log_probs": [-365.6174011230469, -365.37841796875, -368.1173400878906, -366.8245544433594], "reading_logprob_std": 1.2592433811401293, "tokenization_path_logprob_std": 389.4287081119471, "trajectory_balance_rmse": 7739.844392010355} {"canonical_length": 321, "canonical_nll": 1.0386455215026285, "generation_rouge_l_f1": 0.33993399339933994, "generation_token_f1": 0.5907590759075908, "prompt_id": "167bbbea26df8184", "reading_canonical_log_prob": -333.40521240234375, "reading_distances": [31, 34, 34, 37], "reading_log_probs": [-343.01251220703125, -341.01171875, -339.5489501953125, -340.7105712890625], "reading_logprob_std": 1.4398528124949588, "tokenization_path_logprob_std": 277.40307672139625, "trajectory_balance_rmse": 8538.693970591286} {"canonical_length": 195, "canonical_nll": 1.7736187274639423, "generation_rouge_l_f1": 0.24545454545454548, "generation_token_f1": 0.4590909090909091, "prompt_id": "00b0a2b88466f41c", "reading_canonical_log_prob": -345.85565185546875, "reading_distances": [21, 26, 16, 25], "reading_log_probs": [-357.14300537109375, -352.56982421875, -352.13262939453125, -355.27093505859375], "reading_logprob_std": 2.360413135769217, "tokenization_path_logprob_std": 171.57375631574126, "trajectory_balance_rmse": 5489.081022750879} {"canonical_length": 89, "canonical_nll": 1.312098985307672, "generation_rouge_l_f1": 0.3469387755102041, "generation_token_f1": 0.45918367346938777, "prompt_id": "a26fac121253e753", "reading_canonical_log_prob": -116.77680969238281, "reading_distances": [29, 27, 29, 31], "reading_log_probs": [-119.49263763427734, -118.36814880371094, -118.80928039550781, -117.02085876464844], "reading_logprob_std": 1.0428080075192516, "tokenization_path_logprob_std": 173.97684967096058, "trajectory_balance_rmse": 3019.3745080475846} {"canonical_length": 223, "canonical_nll": 1.1398955888277746, "generation_rouge_l_f1": 0.23556581986143188, "generation_token_f1": 0.45265588914549654, "prompt_id": "109439f51e6651d1", "reading_canonical_log_prob": -254.19671630859375, "reading_distances": [38, 33, 38, 41], "reading_log_probs": [-248.55166625976562, -253.31524658203125, -249.98411560058594, -252.11895751953125], "reading_logprob_std": 2.1321793758222496, "tokenization_path_logprob_std": 345.19978927915844, "trajectory_balance_rmse": 4834.798474789142} {"canonical_length": 174, "canonical_nll": 1.0931021153241738, "generation_rouge_l_f1": 0.2830626450116009, "generation_token_f1": 0.5011600928074246, "prompt_id": "25f04ac4bbf2d3c3", "reading_canonical_log_prob": -190.19976806640625, "reading_distances": [22, 20, 16, 20], "reading_log_probs": [-191.8394012451172, -191.56883239746094, -190.82806396484375, -190.96963500976562], "reading_logprob_std": 0.48134547891767376, "tokenization_path_logprob_std": 157.25139421380405, "trajectory_balance_rmse": 4907.445102478592} {"canonical_length": 13, "canonical_nll": 1.9732573582575872, "generation_rouge_l_f1": 0.09248554913294797, "generation_token_f1": 0.09248554913294797, "prompt_id": "68cd67de83285a28", "reading_canonical_log_prob": -25.652345657348633, "reading_distances": [10, 9, 11, 10], "reading_log_probs": [-29.0097599029541, -24.663301467895508, -27.207849502563477, -28.012733459472656], "reading_logprob_std": 1.8590746196865442, "tokenization_path_logprob_std": 29.677089392695432, "trajectory_balance_rmse": 256.52440489214723} {"canonical_length": 12, "canonical_nll": 0.2024287780125936, "generation_rouge_l_f1": 0.125, "generation_token_f1": 0.125, "prompt_id": "41752c3a8cc3cb46", "reading_canonical_log_prob": -2.429145336151123, "reading_distances": [20, 21, 21, 19], "reading_log_probs": [-3.1470727920532227, -2.7630863189697266, -3.4784631729125977, -2.661698579788208], "reading_logprob_std": 0.374388528065464, "tokenization_path_logprob_std": 45.254750662714194, "trajectory_balance_rmse": 307.7537826468537} {"canonical_length": 33, "canonical_nll": 0.47842548832748877, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.375, "prompt_id": "6595c3b5c5d331ea", "reading_canonical_log_prob": -15.788041114807129, "reading_distances": [24, 21, 22, 23], "reading_log_probs": [-18.265731811523438, -17.811222076416016, -20.09032440185547, -21.984375], "reading_logprob_std": 1.9052769874084328, "tokenization_path_logprob_std": 75.92169698360526, "trajectory_balance_rmse": 462.33978197900933} {"canonical_length": 7, "canonical_nll": 0.6084461893354144, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.8571428571428571, "prompt_id": "77f51b75ebf88292", "reading_canonical_log_prob": -4.2591233253479, "reading_distances": [19, 22, 19, 26], "reading_log_probs": [-2.988649368286133, -2.5023550987243652, -2.7567479610443115, -2.8601341247558594], "reading_logprob_std": 0.2061930337134288, "tokenization_path_logprob_std": 44.07019933898815, "trajectory_balance_rmse": 187.9509778762476} {"canonical_length": 250, "canonical_nll": 1.0575968017578126, "generation_rouge_l_f1": 0.21073558648111332, "generation_token_f1": 0.46520874751491054, "prompt_id": "3d42fc04af88620e", "reading_canonical_log_prob": -264.3992004394531, "reading_distances": [29, 25, 33, 29], "reading_log_probs": [-264.437255859375, -265.336181640625, -263.9645080566406, -263.11004638671875], "reading_logprob_std": 0.9291714214173582, "tokenization_path_logprob_std": 93.43762099206123, "trajectory_balance_rmse": 6511.271946431379} {"canonical_length": 311, "canonical_nll": 0.6556120256135701, "generation_rouge_l_f1": 0.3920792079207921, "generation_token_f1": 0.5782178217821783, "prompt_id": "391687d0204ea44e", "reading_canonical_log_prob": -203.8953399658203, "reading_distances": [21, 24, 23, 26], "reading_log_probs": [-205.16407775878906, -208.67837524414062, -207.86422729492188, -208.669921875], "reading_logprob_std": 1.6644340729717413, "tokenization_path_logprob_std": 210.0863815303776, "trajectory_balance_rmse": 5966.899445280469} {"canonical_length": 45, "canonical_nll": 1.9355989244249132, "generation_rouge_l_f1": 0.17194570135746606, "generation_token_f1": 0.20814479638009054, "prompt_id": "9043379228e84740", "reading_canonical_log_prob": -87.1019515991211, "reading_distances": [32, 31, 36, 33], "reading_log_probs": [-86.28743743896484, -91.09664916992188, -90.0876693725586, -88.77146911621094], "reading_logprob_std": 2.0796149327923374, "tokenization_path_logprob_std": 83.22540446164668, "trajectory_balance_rmse": 1357.0825468883222} {"canonical_length": 14, "canonical_nll": 1.279369626726423, "generation_rouge_l_f1": 0.46153846153846156, "generation_token_f1": 0.46153846153846156, "prompt_id": "82a1a337beb39341", "reading_canonical_log_prob": -17.911174774169922, "reading_distances": [26, 23, 25, 34], "reading_log_probs": [-14.677828788757324, -16.0057430267334, -14.881717681884766, -15.185565948486328], "reading_logprob_std": 0.5838929062900969, "tokenization_path_logprob_std": 93.21669121083845, "trajectory_balance_rmse": 510.2317368622434} {"canonical_length": 15, "canonical_nll": 1.211831283569336, "generation_rouge_l_f1": 0.05494505494505494, "generation_token_f1": 0.08791208791208792, "prompt_id": "32a6cf563f35ecba", "reading_canonical_log_prob": -18.17746925354004, "reading_distances": [33, 25, 30, 33], "reading_log_probs": [-30.275691986083984, -22.73658561706543, -18.88101577758789, -20.07073211669922], "reading_logprob_std": 5.117010808043832, "tokenization_path_logprob_std": 37.478813503301, "trajectory_balance_rmse": 249.5243137216937} {"canonical_length": 147, "canonical_nll": 1.628528828523597, "generation_rouge_l_f1": 0.1842818428184282, "generation_token_f1": 0.3306233062330623, "prompt_id": "2655c987f350c6c7", "reading_canonical_log_prob": -239.39373779296875, "reading_distances": [14, 17, 14, 15], "reading_log_probs": [-238.6561279296875, -240.369140625, -238.34237670898438, -240.9034423828125], "reading_logprob_std": 1.2594834357566411, "tokenization_path_logprob_std": 9.49187942653814, "trajectory_balance_rmse": 3936.459334615902} {"canonical_length": 392, "canonical_nll": 1.08443326366191, "generation_rouge_l_f1": 0.30597014925373134, "generation_token_f1": 0.49253731343283585, "prompt_id": "dd04ff95cd33338f", "reading_canonical_log_prob": -425.09783935546875, "reading_distances": [25, 30, 24, 23], "reading_log_probs": [-437.4377136230469, -463.07891845703125, -434.5567626953125, -432.24334716796875], "reading_logprob_std": 14.324950331143546, "tokenization_path_logprob_std": 163.42379495745317, "trajectory_balance_rmse": 5722.916925115879} {"canonical_length": 197, "canonical_nll": 0.6851939864570115, "generation_rouge_l_f1": 0.4225941422594142, "generation_token_f1": 0.6192468619246863, "prompt_id": "f51cb0e119af1f71", "reading_canonical_log_prob": -134.98321533203125, "reading_distances": [38, 44, 38, 39], "reading_log_probs": [-139.1444854736328, -136.4645538330078, -135.55291748046875, -137.73260498046875], "reading_logprob_std": 1.5616463781435046, "tokenization_path_logprob_std": 292.2469822195078, "trajectory_balance_rmse": 4690.830375901713} {"canonical_length": 11, "canonical_nll": 2.412445761940696, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.18181818181818182, "prompt_id": "40f09143e7b23121", "reading_canonical_log_prob": -26.536903381347656, "reading_distances": [42, 47, 49, 42], "reading_log_probs": [-23.77438735961914, -26.799196243286133, -25.374202728271484, -24.47726821899414], "reading_logprob_std": 1.304778630034026, "tokenization_path_logprob_std": 99.23872429617488, "trajectory_balance_rmse": 387.3827109588509} {"canonical_length": 98, "canonical_nll": 1.3115763761559311, "generation_rouge_l_f1": 0.23323615160349853, "generation_token_f1": 0.3556851311953353, "prompt_id": "09e3575b08c40649", "reading_canonical_log_prob": -128.53448486328125, "reading_distances": [16, 17, 13, 13], "reading_log_probs": [-131.20143127441406, -135.4156494140625, -129.82557678222656, -128.4119110107422], "reading_logprob_std": 3.0239839613009125, "tokenization_path_logprob_std": 143.8496371050097, "trajectory_balance_rmse": 2715.354378838788} {"canonical_length": 11, "canonical_nll": 0.8040638836947355, "generation_rouge_l_f1": 0.1032258064516129, "generation_token_f1": 0.1032258064516129, "prompt_id": "fc90d5db5297a0af", "reading_canonical_log_prob": -8.84470272064209, "reading_distances": [18, 22, 20, 20], "reading_log_probs": [-8.760710716247559, -15.467121124267578, -12.746158599853516, -17.902490615844727], "reading_logprob_std": 3.919548194683112, "tokenization_path_logprob_std": 36.178248974709824, "trajectory_balance_rmse": 322.3961680687041} {"canonical_length": 244, "canonical_nll": 1.1256986524237962, "generation_rouge_l_f1": 0.3181818181818182, "generation_token_f1": 0.6239669421487604, "prompt_id": "e15ab7087e38fd8b", "reading_canonical_log_prob": -274.67047119140625, "reading_distances": [21, 16, 16, 15], "reading_log_probs": [-275.8404846191406, -275.2168884277344, -274.90350341796875, -273.7694091796875], "reading_logprob_std": 0.8677430272631937, "tokenization_path_logprob_std": 44.59084738048726, "trajectory_balance_rmse": 6212.326482157479} {"canonical_length": 83, "canonical_nll": 1.0446089779038028, "generation_rouge_l_f1": 0.39779005524861877, "generation_token_f1": 0.5414364640883977, "prompt_id": "514750b55e58033f", "reading_canonical_log_prob": -86.70254516601562, "reading_distances": [17, 18, 16, 13], "reading_log_probs": [-89.93553161621094, -87.41206359863281, -88.73863220214844, -87.96804809570312], "reading_logprob_std": 1.0929376824455992, "tokenization_path_logprob_std": 322.3951866870716, "trajectory_balance_rmse": 2582.5277438790235} {"canonical_length": 145, "canonical_nll": 1.3938809098868534, "generation_rouge_l_f1": 0.16666666666666669, "generation_token_f1": 0.27619047619047615, "prompt_id": "612827ad6442cf24", "reading_canonical_log_prob": -202.11273193359375, "reading_distances": [27, 26, 28, 29], "reading_log_probs": [-197.93243408203125, -198.50567626953125, -197.68124389648438, -198.394287109375], "reading_logprob_std": 0.3878936967659274, "tokenization_path_logprob_std": 331.80022038260626, "trajectory_balance_rmse": 5082.543168989163} {"canonical_length": 184, "canonical_nll": 0.870036913001019, "generation_rouge_l_f1": 0.22573363431151242, "generation_token_f1": 0.38374717832957117, "prompt_id": "637d5437d0c071a7", "reading_canonical_log_prob": -160.0867919921875, "reading_distances": [249, 215, 215, 227], "reading_log_probs": [-177.2099609375, -175.32408142089844, -174.52392578125, -173.84201049804688], "reading_logprob_std": 1.4553409765369443, "tokenization_path_logprob_std": 140.3710726040386, "trajectory_balance_rmse": 3862.962078994973} {"canonical_length": 65, "canonical_nll": 1.0547997107872595, "generation_rouge_l_f1": 0.15434083601286172, "generation_token_f1": 0.20578778135048234, "prompt_id": "c97d2c584d8df586", "reading_canonical_log_prob": -68.56198120117188, "reading_distances": [15, 19, 16, 17], "reading_log_probs": [-69.31504821777344, -69.73275756835938, -68.00505065917969, -69.1548843383789], "reading_logprob_std": 0.7392142955602319, "tokenization_path_logprob_std": 56.34093931031481, "trajectory_balance_rmse": 1161.2129589646418} {"canonical_length": 17, "canonical_nll": 0.8005392972160789, "generation_rouge_l_f1": 0.43478260869565216, "generation_token_f1": 0.5217391304347827, "prompt_id": "a17a07dc87573da6", "reading_canonical_log_prob": -13.60916805267334, "reading_distances": [29, 25, 21, 24], "reading_log_probs": [-12.862805366516113, -13.440656661987305, -13.73403263092041, -13.851875305175781], "reading_logprob_std": 0.4416130977269805, "tokenization_path_logprob_std": 135.8295315349838, "trajectory_balance_rmse": 617.3760151361403} {"canonical_length": 13, "canonical_nll": 0.6650098653940054, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.22727272727272727, "prompt_id": "4b95b246a8b6c8d6", "reading_canonical_log_prob": -8.64512825012207, "reading_distances": [10, 12, 14, 12], "reading_log_probs": [-7.982721328735352, -8.480009078979492, -9.911049842834473, -13.443355560302734], "reading_logprob_std": 2.4654976140550464, "tokenization_path_logprob_std": 34.407327345941866, "trajectory_balance_rmse": 419.8542233446613} {"canonical_length": 99, "canonical_nll": 1.3583837952276674, "generation_rouge_l_f1": 0.3083700440528634, "generation_token_f1": 0.4317180616740088, "prompt_id": "0c93e5e51c22b30d", "reading_canonical_log_prob": -134.47999572753906, "reading_distances": [17, 18, 18, 15], "reading_log_probs": [-129.69212341308594, -134.20660400390625, -132.87803649902344, -133.47470092773438], "reading_logprob_std": 1.989456626376375, "tokenization_path_logprob_std": 209.80761912249727, "trajectory_balance_rmse": 2893.2562702814857} {"canonical_length": 306, "canonical_nll": 1.2324758292802798, "generation_rouge_l_f1": 0.21416234887737481, "generation_token_f1": 0.4594127806563039, "prompt_id": "6c10022ee752e034", "reading_canonical_log_prob": -377.1376037597656, "reading_distances": [29, 34, 30, 36], "reading_log_probs": [-386.6563720703125, -382.2271728515625, -383.42352294921875, -386.532470703125], "reading_logprob_std": 2.230785922630421, "tokenization_path_logprob_std": 310.3890519250873, "trajectory_balance_rmse": 8832.67769503955} {"canonical_length": 69, "canonical_nll": 1.2823676510133606, "generation_rouge_l_f1": 0.2911877394636015, "generation_token_f1": 0.38314176245210724, "prompt_id": "7884c89879ae1c04", "reading_canonical_log_prob": -88.48336791992188, "reading_distances": [35, 28, 26, 37], "reading_log_probs": [-93.26986694335938, -91.28398132324219, -92.73155975341797, -96.88427734375], "reading_logprob_std": 2.380499616384161, "tokenization_path_logprob_std": 213.76887442776746, "trajectory_balance_rmse": 1776.5399708614273} {"canonical_length": 46, "canonical_nll": 1.1671100284742273, "generation_rouge_l_f1": 0.5789473684210527, "generation_token_f1": 0.6578947368421053, "prompt_id": "803eb754dd6450de", "reading_canonical_log_prob": -53.68706130981445, "reading_distances": [14, 15, 17, 13], "reading_log_probs": [-71.41921997070312, -63.57912063598633, -71.82292175292969, -73.33534240722656], "reading_logprob_std": 4.38494768345673, "tokenization_path_logprob_std": 86.08806598573146, "trajectory_balance_rmse": 1554.5627500652877} {"canonical_length": 12, "canonical_nll": 1.0663685003916423, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "7f58078c29e97e0d", "reading_canonical_log_prob": -12.796422004699707, "reading_distances": [39, 46, 39, 39], "reading_log_probs": [-19.09832191467285, -14.960725784301758, -12.88833999633789, -23.40404510498047], "reading_logprob_std": 4.658212347498899, "tokenization_path_logprob_std": 16.663110463419862, "trajectory_balance_rmse": 251.53823401332943} {"canonical_length": 135, "canonical_nll": 1.4341625072337962, "generation_rouge_l_f1": 0.20304568527918782, "generation_token_f1": 0.3807106598984772, "prompt_id": "979bfe699ee39dca", "reading_canonical_log_prob": -193.6119384765625, "reading_distances": [22, 21, 20, 29], "reading_log_probs": [-192.38754272460938, -194.33383178710938, -191.08157348632812, -192.10372924804688], "reading_logprob_std": 1.3592121077598265, "tokenization_path_logprob_std": 287.646272840904, "trajectory_balance_rmse": 3374.464598307589} {"canonical_length": 13, "canonical_nll": 1.5821615365835338, "generation_rouge_l_f1": 0.09142857142857141, "generation_token_f1": 0.11428571428571428, "prompt_id": "f30d075da952702b", "reading_canonical_log_prob": -20.568099975585938, "reading_distances": [34, 26, 31, 29], "reading_log_probs": [-42.35581970214844, -35.64642333984375, -24.49188804626465, -29.089534759521484], "reading_logprob_std": 7.792545426570344, "tokenization_path_logprob_std": 101.15913798852097, "trajectory_balance_rmse": 274.715790988205} {"canonical_length": 117, "canonical_nll": 1.271087646484375, "generation_rouge_l_f1": 0.18749999999999997, "generation_token_f1": 0.32812499999999994, "prompt_id": "ad627413d7298d7b", "reading_canonical_log_prob": -148.71725463867188, "reading_distances": [24, 17, 22, 29], "reading_log_probs": [-148.2800750732422, -146.15811157226562, -147.30494689941406, -149.83056640625], "reading_logprob_std": 1.5555957352185983, "tokenization_path_logprob_std": 139.12408072123853, "trajectory_balance_rmse": 3041.5645364968564} {"canonical_length": 314, "canonical_nll": 1.5413618148512143, "generation_rouge_l_f1": 0.29850746268656714, "generation_token_f1": 0.541044776119403, "prompt_id": "802f028e684a8238", "reading_canonical_log_prob": -483.98760986328125, "reading_distances": [24, 22, 23, 19], "reading_log_probs": [-483.4017333984375, -481.7630920410156, -483.99908447265625, -481.1697692871094], "reading_logprob_std": 1.3348019847591517, "tokenization_path_logprob_std": 361.83187389401587, "trajectory_balance_rmse": 7810.591978459877} {"canonical_length": 190, "canonical_nll": 1.5195781506990131, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.5, "prompt_id": "2e2423f516c40a85", "reading_canonical_log_prob": -288.7198486328125, "reading_distances": [8, 16, 21, 16], "reading_log_probs": [-288.31494140625, -283.4872131347656, -285.20428466796875, -286.22540283203125], "reading_logprob_std": 2.017383140630016, "tokenization_path_logprob_std": 243.8309757336547, "trajectory_balance_rmse": 5666.569982490204} {"canonical_length": 297, "canonical_nll": 1.6122360486374159, "generation_rouge_l_f1": 0.28471001757469244, "generation_token_f1": 0.5202108963093146, "prompt_id": "8d1b09dca5de1a98", "reading_canonical_log_prob": -478.8341064453125, "reading_distances": [26, 21, 28, 23], "reading_log_probs": [-476.090087890625, -474.6958312988281, -478.1820983886719, -473.9561462402344], "reading_logprob_std": 1.8581807882867858, "tokenization_path_logprob_std": 288.3526100420029, "trajectory_balance_rmse": 9916.586100417457} {"canonical_length": 82, "canonical_nll": 0.7689574637064119, "generation_rouge_l_f1": 0.28030303030303033, "generation_token_f1": 0.4090909090909091, "prompt_id": "0609bdeeadeb9baf", "reading_canonical_log_prob": -63.05451202392578, "reading_distances": [35, 36, 39, 34], "reading_log_probs": [-61.65336990356445, -63.436622619628906, -63.583805084228516, -61.86152648925781], "reading_logprob_std": 1.0172976234758, "tokenization_path_logprob_std": 143.46795543809773, "trajectory_balance_rmse": 1419.934841399667} {"canonical_length": 222, "canonical_nll": 1.1389404846741273, "generation_rouge_l_f1": 0.3952095808383233, "generation_token_f1": 0.6067864271457086, "prompt_id": "e9f992cd65ad5461", "reading_canonical_log_prob": -252.84478759765625, "reading_distances": [25, 32, 29, 24], "reading_log_probs": [-253.463623046875, -254.44796752929688, -252.62548828125, -253.57241821289062], "reading_logprob_std": 0.7454268265022118, "tokenization_path_logprob_std": 187.8165483109301, "trajectory_balance_rmse": 4447.82300613661} {"canonical_length": 6, "canonical_nll": 3.023989995320638, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "15eeee68ceb20913", "reading_canonical_log_prob": -18.143939971923828, "reading_distances": [32, 38, 36, 31], "reading_log_probs": [-17.315364837646484, -21.754295349121094, -17.411169052124023, -18.279394149780273], "reading_logprob_std": 2.088343327554823, "tokenization_path_logprob_std": 56.53352670237865, "trajectory_balance_rmse": 196.74546893628087} {"canonical_length": 105, "canonical_nll": 1.4702671595982142, "generation_rouge_l_f1": 0.21727019498607242, "generation_token_f1": 0.3342618384401114, "prompt_id": "728a7ca9e481ba81", "reading_canonical_log_prob": -154.3780517578125, "reading_distances": [29, 30, 30, 24], "reading_log_probs": [-148.3016357421875, -143.75155639648438, -146.31890869140625, -146.94015502929688], "reading_logprob_std": 1.9068399775856024, "tokenization_path_logprob_std": 128.07042947251605, "trajectory_balance_rmse": 2667.2595831052345} {"canonical_length": 123, "canonical_nll": 1.5954421128683942, "generation_rouge_l_f1": 0.08465608465608465, "generation_token_f1": 0.12698412698412698, "prompt_id": "92cf4663b97aed32", "reading_canonical_log_prob": -196.2393798828125, "reading_distances": [26, 24, 30, 22], "reading_log_probs": [-198.87222290039062, -202.94009399414062, -202.72328186035156, -198.69308471679688], "reading_logprob_std": 2.3405288782025653, "tokenization_path_logprob_std": 142.13319926323817, "trajectory_balance_rmse": 2367.955666399428} {"canonical_length": 66, "canonical_nll": 1.1441337123061672, "generation_rouge_l_f1": 0.265625, "generation_token_f1": 0.5468750000000001, "prompt_id": "3226fde653ee5231", "reading_canonical_log_prob": -75.51282501220703, "reading_distances": [8, 13, 12, 12], "reading_log_probs": [-73.72956085205078, -74.8189697265625, -73.97723388671875, -73.86773681640625], "reading_logprob_std": 0.49096835496639585, "tokenization_path_logprob_std": 61.22253297071187, "trajectory_balance_rmse": 2000.976948800595} {"canonical_length": 259, "canonical_nll": 1.2602732301218629, "generation_rouge_l_f1": 0.31606217616580307, "generation_token_f1": 0.4922279792746114, "prompt_id": "d40a7f2ec7ebb044", "reading_canonical_log_prob": -326.4107666015625, "reading_distances": [22, 24, 26, 20], "reading_log_probs": [-331.680419921875, -338.2571105957031, -337.89959716796875, -333.05657958984375], "reading_logprob_std": 3.3472998960011884, "tokenization_path_logprob_std": 160.10001559167833, "trajectory_balance_rmse": 6396.285844571848} {"canonical_length": 13, "canonical_nll": 1.1533602201021635, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "6844823da97b3fe4", "reading_canonical_log_prob": -14.993682861328125, "reading_distances": [20, 24, 22, 21], "reading_log_probs": [-15.272372245788574, -15.831414222717285, -16.635787963867188, -16.149932861328125], "reading_logprob_std": 0.5719899041559193, "tokenization_path_logprob_std": 111.81265122699715, "trajectory_balance_rmse": 380.83325273654805} {"canonical_length": 118, "canonical_nll": 0.9505891315007614, "generation_rouge_l_f1": 0.27089337175792505, "generation_token_f1": 0.4610951008645533, "prompt_id": "49597fd7a90f1c14", "reading_canonical_log_prob": -112.16951751708984, "reading_distances": [24, 25, 31, 24], "reading_log_probs": [-122.41641235351562, -119.75900268554688, -127.30413818359375, -123.52364349365234], "reading_logprob_std": 3.1301142275446643, "tokenization_path_logprob_std": 220.41171142322435, "trajectory_balance_rmse": 2623.325827832172} {"canonical_length": 258, "canonical_nll": 1.6089546144470688, "generation_rouge_l_f1": 0.18323586744639375, "generation_token_f1": 0.3157894736842105, "prompt_id": "06dc86d1f57bb10a", "reading_canonical_log_prob": -415.11029052734375, "reading_distances": [29, 34, 39, 30], "reading_log_probs": [-421.29461669921875, -423.8793029785156, -418.0647888183594, -427.5927734375], "reading_logprob_std": 4.032783346867966, "tokenization_path_logprob_std": 255.1403990031687, "trajectory_balance_rmse": 6856.098791311751} {"canonical_length": 250, "canonical_nll": 1.364917236328125, "generation_rouge_l_f1": 0.2285714285714286, "generation_token_f1": 0.48380952380952386, "prompt_id": "104414445caa5ddc", "reading_canonical_log_prob": -341.22930908203125, "reading_distances": [37, 32, 36, 35], "reading_log_probs": [-341.180908203125, -341.1237487792969, -339.60296630859375, -340.5443115234375], "reading_logprob_std": 0.7321817183950622, "tokenization_path_logprob_std": 466.6737749260283, "trajectory_balance_rmse": 7778.405195654529} {"canonical_length": 274, "canonical_nll": 0.9843293350108349, "generation_rouge_l_f1": 0.3481624758220503, "generation_token_f1": 0.5454545454545454, "prompt_id": "96cee7006ba2cc41", "reading_canonical_log_prob": -269.70623779296875, "reading_distances": [20, 20, 21, 18], "reading_log_probs": [-278.7591247558594, -282.234130859375, -278.708984375, -276.65008544921875], "reading_logprob_std": 2.3161302845441436, "tokenization_path_logprob_std": 152.13521712892015, "trajectory_balance_rmse": 6831.5362468997055} {"canonical_length": 150, "canonical_nll": 2.198420003255208, "generation_rouge_l_f1": 0.23300970873786406, "generation_token_f1": 0.4514563106796116, "prompt_id": "c61662a19e060e22", "reading_canonical_log_prob": -329.76300048828125, "reading_distances": [28, 29, 33, 33], "reading_log_probs": [-329.5984191894531, -329.58221435546875, -330.099853515625, -329.7799072265625], "reading_logprob_std": 0.24049173357504067, "tokenization_path_logprob_std": 330.5918692812274, "trajectory_balance_rmse": 4316.0157900664735} {"canonical_length": 80, "canonical_nll": 1.3872719764709474, "generation_rouge_l_f1": 0.3548387096774194, "generation_token_f1": 0.4731182795698925, "prompt_id": "448e404116b3a283", "reading_canonical_log_prob": -110.98175811767578, "reading_distances": [18, 27, 26, 24], "reading_log_probs": [-107.12256622314453, -107.94475555419922, -106.05359649658203, -108.97296905517578], "reading_logprob_std": 1.2382488957483722, "tokenization_path_logprob_std": 107.3429733861094, "trajectory_balance_rmse": 1805.1175757086503} {"canonical_length": 40, "canonical_nll": 2.619934844970703, "generation_rouge_l_f1": 0.2682926829268293, "generation_token_f1": 0.3902439024390244, "prompt_id": "62e94a0db3b86cbd", "reading_canonical_log_prob": -104.79739379882812, "reading_distances": [16, 17, 15, 16], "reading_log_probs": [-105.79118347167969, -104.98834991455078, -106.20927429199219, -104.36278533935547], "reading_logprob_std": 0.824174926026456, "tokenization_path_logprob_std": 90.59260875955144, "trajectory_balance_rmse": 1479.3945892256986} {"canonical_length": 203, "canonical_nll": 1.146665676473984, "generation_rouge_l_f1": 0.236144578313253, "generation_token_f1": 0.4915662650602409, "prompt_id": "6996ebbe576a8626", "reading_canonical_log_prob": -232.77313232421875, "reading_distances": [33, 32, 37, 29], "reading_log_probs": [-230.0089569091797, -232.058349609375, -229.93655395507812, -231.0164031982422], "reading_logprob_std": 0.9989137501535489, "tokenization_path_logprob_std": 458.38600686879823, "trajectory_balance_rmse": 5507.546078536484} {"canonical_length": 97, "canonical_nll": 1.6803412683231314, "generation_rouge_l_f1": 0.25210084033613445, "generation_token_f1": 0.4789915966386555, "prompt_id": "539a3dd889cf333b", "reading_canonical_log_prob": -162.99310302734375, "reading_distances": [39, 25, 33, 33], "reading_log_probs": [-164.1464385986328, -162.21905517578125, -169.71255493164062, -168.83462524414062], "reading_logprob_std": 3.621286815150658, "tokenization_path_logprob_std": 253.13281529228593, "trajectory_balance_rmse": 3522.481579330384} {"canonical_length": 83, "canonical_nll": 1.9769288947783321, "generation_rouge_l_f1": 0.17045454545454547, "generation_token_f1": 0.27840909090909094, "prompt_id": "22dd62a830e20aab", "reading_canonical_log_prob": -164.08509826660156, "reading_distances": [24, 37, 29, 27], "reading_log_probs": [-159.552978515625, -161.20838928222656, -160.40994262695312, -161.3267822265625], "reading_logprob_std": 0.8222570918523685, "tokenization_path_logprob_std": 146.95275294973194, "trajectory_balance_rmse": 2439.368052284248} {"canonical_length": 26, "canonical_nll": 1.4060990260197566, "generation_rouge_l_f1": 0.10256410256410256, "generation_token_f1": 0.10256410256410256, "prompt_id": "23f859fe909ddbe9", "reading_canonical_log_prob": -36.55857467651367, "reading_distances": [17, 14, 14, 16], "reading_log_probs": [-35.4285774230957, -39.748313903808594, -35.818138122558594, -36.49097442626953], "reading_logprob_std": 1.967439516833373, "tokenization_path_logprob_std": 129.13295237416108, "trajectory_balance_rmse": 754.3049271287786} {"canonical_length": 97, "canonical_nll": 0.7754119401125564, "generation_rouge_l_f1": 0.33935018050541516, "generation_token_f1": 0.4693140794223826, "prompt_id": "9658518adf176971", "reading_canonical_log_prob": -75.21495819091797, "reading_distances": [16, 13, 18, 14], "reading_log_probs": [-74.10813903808594, -73.09858703613281, -74.45561981201172, -76.1053466796875], "reading_logprob_std": 1.2494170888664229, "tokenization_path_logprob_std": 89.30005658742455, "trajectory_balance_rmse": 1572.550365212264} {"canonical_length": 243, "canonical_nll": 1.4113257137345678, "generation_rouge_l_f1": 0.29045643153526973, "generation_token_f1": 0.5062240663900415, "prompt_id": "c496b50e51f7e82e", "reading_canonical_log_prob": -342.9521484375, "reading_distances": [30, 37, 26, 33], "reading_log_probs": [-343.9645080566406, -341.5355529785156, -343.0588073730469, -343.02496337890625], "reading_logprob_std": 1.0059263313177362, "tokenization_path_logprob_std": 185.25582009471276, "trajectory_balance_rmse": 7385.072510491155} {"canonical_length": 157, "canonical_nll": 1.1694831605170184, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.4888888888888889, "prompt_id": "9433be6b41e84522", "reading_canonical_log_prob": -183.60885620117188, "reading_distances": [41, 34, 38, 35], "reading_log_probs": [-185.3135986328125, -189.47134399414062, -188.71493530273438, -183.0316162109375], "reading_logprob_std": 3.0056300689279123, "tokenization_path_logprob_std": 298.08463696990214, "trajectory_balance_rmse": 4567.472070790504} {"canonical_length": 14, "canonical_nll": 0.8129113742283413, "generation_rouge_l_f1": 0.6, "generation_token_f1": 0.6, "prompt_id": "22dd55d64faca0d8", "reading_canonical_log_prob": -11.380759239196777, "reading_distances": [40, 42, 49, 49], "reading_log_probs": [-16.72719955444336, -19.36486053466797, -19.978933334350586, -16.017560958862305], "reading_logprob_std": 1.9431200406628668, "tokenization_path_logprob_std": 50.5336521427027, "trajectory_balance_rmse": 531.3859003199965} {"canonical_length": 52, "canonical_nll": 1.4086732130784254, "generation_rouge_l_f1": 0.24827586206896549, "generation_token_f1": 0.4, "prompt_id": "9ce142b4fe68303a", "reading_canonical_log_prob": -73.25100708007812, "reading_distances": [31, 27, 33, 38], "reading_log_probs": [-75.05064392089844, -74.64585876464844, -74.80215454101562, -72.73756408691406], "reading_logprob_std": 1.0608364232594536, "tokenization_path_logprob_std": 149.47415343743583, "trajectory_balance_rmse": 1723.8738053225438} {"canonical_length": 28, "canonical_nll": 0.9680006844656808, "generation_rouge_l_f1": 0.5333333333333333, "generation_token_f1": 0.5333333333333333, "prompt_id": "e62a412929959b77", "reading_canonical_log_prob": -27.104019165039062, "reading_distances": [27, 24, 28, 24], "reading_log_probs": [-30.422889709472656, -31.23674774169922, -29.51559066772461, -31.421329498291016], "reading_logprob_std": 0.8713366670505069, "tokenization_path_logprob_std": 108.21900832058813, "trajectory_balance_rmse": 1036.7446732130518} {"canonical_length": 21, "canonical_nll": 1.332124528430757, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4210526315789474, "prompt_id": "273ee8648fccd69e", "reading_canonical_log_prob": -27.9746150970459, "reading_distances": [31, 32, 30, 28], "reading_log_probs": [-35.28618621826172, -30.96426010131836, -31.42757797241211, -40.20978546142578], "reading_logprob_std": 4.2878831556364885, "tokenization_path_logprob_std": 51.987895075197144, "trajectory_balance_rmse": 552.4627416722773} {"canonical_length": 109, "canonical_nll": 1.3416304282092173, "generation_rouge_l_f1": 0.2776203966005666, "generation_token_f1": 0.4589235127478754, "prompt_id": "0ffd3304b685d298", "reading_canonical_log_prob": -146.2377166748047, "reading_distances": [22, 24, 24, 19], "reading_log_probs": [-148.94924926757812, -150.74362182617188, -149.8104248046875, -152.1886749267578], "reading_logprob_std": 1.386553989021267, "tokenization_path_logprob_std": 222.48016637027465, "trajectory_balance_rmse": 3151.8046870854146} {"canonical_length": 288, "canonical_nll": 1.2983231014675565, "generation_rouge_l_f1": 0.32954545454545453, "generation_token_f1": 0.5833333333333334, "prompt_id": "6b90af69594f1bfc", "reading_canonical_log_prob": -373.91705322265625, "reading_distances": [30, 33, 33, 33], "reading_log_probs": [-379.9326477050781, -383.8104248046875, -381.7914733886719, -384.5145263671875], "reading_logprob_std": 2.0710895197312036, "tokenization_path_logprob_std": 245.0373266078234, "trajectory_balance_rmse": 9601.985980463354} {"canonical_length": 256, "canonical_nll": 1.3082644939422607, "generation_rouge_l_f1": 0.305019305019305, "generation_token_f1": 0.5559845559845559, "prompt_id": "7021b5d065291a9c", "reading_canonical_log_prob": -334.91571044921875, "reading_distances": [18, 24, 19, 23], "reading_log_probs": [-330.60797119140625, -332.6458740234375, -331.4045104980469, -335.6677551269531], "reading_logprob_std": 2.221803694188121, "tokenization_path_logprob_std": 210.00611459684683, "trajectory_balance_rmse": 7267.850716752844} {"canonical_length": 68, "canonical_nll": 0.8910352482515222, "generation_rouge_l_f1": 0.3314285714285714, "generation_token_f1": 0.5028571428571429, "prompt_id": "56589cd90846fa45", "reading_canonical_log_prob": -60.590396881103516, "reading_distances": [39, 45, 32, 27], "reading_log_probs": [-70.93836975097656, -73.07951354980469, -72.51436614990234, -65.37310791015625], "reading_logprob_std": 3.520724942947841, "tokenization_path_logprob_std": 66.87552467486488, "trajectory_balance_rmse": 1468.8748632650997} {"canonical_length": 187, "canonical_nll": 1.8207377877464905, "generation_rouge_l_f1": 0.21649484536082475, "generation_token_f1": 0.4536082474226804, "prompt_id": "b2517684dcd5d425", "reading_canonical_log_prob": -340.47796630859375, "reading_distances": [47, 36, 38, 36], "reading_log_probs": [-329.30828857421875, -331.54925537109375, -331.33258056640625, -329.250244140625], "reading_logprob_std": 1.2513853992524009, "tokenization_path_logprob_std": 126.00647265502246, "trajectory_balance_rmse": 6350.353996002769} {"canonical_length": 2, "canonical_nll": 9.3614501953125, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.33333333333333337, "prompt_id": "44fca97732ef92a8", "reading_canonical_log_prob": -18.722900390625, "reading_distances": [22, 24, 22, 18], "reading_log_probs": [-17.852441787719727, -19.385557174682617, -17.80194091796875, -17.24335479736328], "reading_logprob_std": 0.9189155754413085, "tokenization_path_logprob_std": 28.8403390037306, "trajectory_balance_rmse": 47.963075868358956} {"canonical_length": 14, "canonical_nll": 0.27360498905181885, "generation_rouge_l_f1": 0.625, "generation_token_f1": 0.625, "prompt_id": "907693271b754ffe", "reading_canonical_log_prob": -3.830469846725464, "reading_distances": [5, 6, 7, 8], "reading_log_probs": [-3.7649009227752686, -5.372032642364502, -5.090995788574219, -4.882526397705078], "reading_logprob_std": 0.704304871433442, "tokenization_path_logprob_std": 25.067807697601616, "trajectory_balance_rmse": 267.80383011689355} {"canonical_length": 220, "canonical_nll": 1.2374447909268467, "generation_rouge_l_f1": 0.2260127931769723, "generation_token_f1": 0.42643923240938164, "prompt_id": "3d885c9990d3f7a8", "reading_canonical_log_prob": -272.23785400390625, "reading_distances": [43, 52, 46, 50], "reading_log_probs": [-284.8681945800781, -283.8075256347656, -279.6515197753906, -293.8213195800781], "reading_logprob_std": 5.9639810688994706, "tokenization_path_logprob_std": 136.99400858760487, "trajectory_balance_rmse": 5617.726832872457} {"canonical_length": 12, "canonical_nll": 1.3795452117919922, "generation_rouge_l_f1": 0.14285714285714288, "generation_token_f1": 0.14285714285714288, "prompt_id": "946f59f054cd3b7a", "reading_canonical_log_prob": -16.554542541503906, "reading_distances": [27, 25, 33, 27], "reading_log_probs": [-16.947847366333008, -17.685611724853516, -18.77094078063965, -18.458845138549805], "reading_logprob_std": 0.8177363508917493, "tokenization_path_logprob_std": 51.974595651494056, "trajectory_balance_rmse": 383.47205276369067} {"canonical_length": 114, "canonical_nll": 1.1939317134388707, "generation_rouge_l_f1": 0.20178041543026706, "generation_token_f1": 0.3679525222551928, "prompt_id": "1e094a0368e082c2", "reading_canonical_log_prob": -136.10821533203125, "reading_distances": [30, 31, 29, 21], "reading_log_probs": [-140.5198974609375, -135.4933319091797, -136.54241943359375, -136.63827514648438], "reading_logprob_std": 2.209342897492484, "tokenization_path_logprob_std": 277.7729161310415, "trajectory_balance_rmse": 3433.8956642674802} {"canonical_length": 8, "canonical_nll": 0.024775946512818336, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "f01addec53ef2695", "reading_canonical_log_prob": -0.1982075721025467, "reading_distances": [24, 25, 23, 24], "reading_log_probs": [-7.761085033416748, -0.07584034651517868, -0.06099817156791687, -7.888463020324707], "reading_logprob_std": 4.47843953497493, "tokenization_path_logprob_std": 60.650988344712324, "trajectory_balance_rmse": 211.12976010788122} {"canonical_length": 43, "canonical_nll": 1.3338610626930414, "generation_rouge_l_f1": 0.2962962962962963, "generation_token_f1": 0.4444444444444445, "prompt_id": "bd70996cafa5e523", "reading_canonical_log_prob": -57.35602569580078, "reading_distances": [32, 29, 25, 18], "reading_log_probs": [-62.36918258666992, -68.11476135253906, -68.40443420410156, -65.64125061035156], "reading_logprob_std": 2.798507279448813, "tokenization_path_logprob_std": 168.6779212328999, "trajectory_balance_rmse": 1106.7672618218587} {"canonical_length": 280, "canonical_nll": 1.0950750078473772, "generation_rouge_l_f1": 0.223463687150838, "generation_token_f1": 0.4878957169459963, "prompt_id": "40b3b0fca502cfb8", "reading_canonical_log_prob": -306.6210021972656, "reading_distances": [29, 29, 26, 34], "reading_log_probs": [-300.65533447265625, -301.19024658203125, -302.8232421875, -300.6287536621094], "reading_logprob_std": 1.0321656116711664, "tokenization_path_logprob_std": 217.89437117006838, "trajectory_balance_rmse": 7416.002850786569} {"canonical_length": 74, "canonical_nll": 1.1933385488149282, "generation_rouge_l_f1": 0.3285714285714285, "generation_token_f1": 0.5857142857142857, "prompt_id": "88f331fe7934b488", "reading_canonical_log_prob": -88.30705261230469, "reading_distances": [15, 10, 13, 16], "reading_log_probs": [-90.60882568359375, -88.64768981933594, -89.48806762695312, -89.21220397949219], "reading_logprob_std": 0.8243035713312531, "tokenization_path_logprob_std": 142.77911490741124, "trajectory_balance_rmse": 2022.3436848724343} {"canonical_length": 33, "canonical_nll": 0.5631790161132812, "generation_rouge_l_f1": 0.18390804597701152, "generation_token_f1": 0.3218390804597701, "prompt_id": "8b4c34e2f1db7af4", "reading_canonical_log_prob": -18.58490753173828, "reading_distances": [18, 17, 20, 21], "reading_log_probs": [-18.646591186523438, -18.660396575927734, -20.0334529876709, -19.058773040771484], "reading_logprob_std": 0.6511181947035553, "tokenization_path_logprob_std": 113.75590865968422, "trajectory_balance_rmse": 1198.3260485751168} {"canonical_length": 155, "canonical_nll": 2.2000687137726813, "generation_rouge_l_f1": 0.10576923076923077, "generation_token_f1": 0.1971153846153846, "prompt_id": "7a99289ed340c72d", "reading_canonical_log_prob": -341.0106506347656, "reading_distances": [13, 21, 17, 17], "reading_log_probs": [-341.19903564453125, -337.2374267578125, -342.7707214355469, -343.0615234375], "reading_logprob_std": 2.6810355571634608, "tokenization_path_logprob_std": 172.83569244709727, "trajectory_balance_rmse": 3304.1967949409864} {"canonical_length": 21, "canonical_nll": 1.0298875172932942, "generation_rouge_l_f1": 0.6190476190476191, "generation_token_f1": 0.6666666666666666, "prompt_id": "321ea8e8622b4791", "reading_canonical_log_prob": -21.62763786315918, "reading_distances": [26, 33, 35, 27], "reading_log_probs": [-21.003528594970703, -23.297657012939453, -20.639665603637695, -20.456209182739258], "reading_logprob_std": 1.3186919142228024, "tokenization_path_logprob_std": 77.85231982217644, "trajectory_balance_rmse": 599.0308822642021} {"canonical_length": 176, "canonical_nll": 2.0061173872514204, "generation_rouge_l_f1": 0.22168674698795182, "generation_token_f1": 0.4626506024096385, "prompt_id": "e9ba47f9f768bdc0", "reading_canonical_log_prob": -353.07666015625, "reading_distances": [25, 29, 29, 20], "reading_log_probs": [-354.10845947265625, -355.98358154296875, -355.792236328125, -354.541259765625], "reading_logprob_std": 0.9228737197718808, "tokenization_path_logprob_std": 148.35187658873028, "trajectory_balance_rmse": 5887.142218855205} {"canonical_length": 7, "canonical_nll": 1.0713036400931222, "generation_rouge_l_f1": 0.04580152671755725, "generation_token_f1": 0.04580152671755725, "prompt_id": "cd4e165565532ef7", "reading_canonical_log_prob": -7.4991254806518555, "reading_distances": [31, 32, 36, 37], "reading_log_probs": [-13.152931213378906, -9.285134315490723, -9.807794570922852, -7.262522220611572], "reading_logprob_std": 2.444198145069067, "tokenization_path_logprob_std": 13.69738821659776, "trajectory_balance_rmse": 215.05410537506256} {"canonical_length": 270, "canonical_nll": 1.067232033058449, "generation_rouge_l_f1": 0.2786596119929453, "generation_token_f1": 0.525573192239859, "prompt_id": "df9cf174d34cff3e", "reading_canonical_log_prob": -288.15264892578125, "reading_distances": [32, 36, 42, 30], "reading_log_probs": [-290.8424987792969, -291.78631591796875, -289.768310546875, -290.5347900390625], "reading_logprob_std": 0.8349405834159855, "tokenization_path_logprob_std": 284.3045604193944, "trajectory_balance_rmse": 5544.832883816592} {"canonical_length": 143, "canonical_nll": 1.7382761281687062, "generation_rouge_l_f1": 0.43971631205673756, "generation_token_f1": 0.6524822695035462, "prompt_id": "1fba2562017bea7b", "reading_canonical_log_prob": -248.573486328125, "reading_distances": [29, 34, 33, 29], "reading_log_probs": [-242.22332763671875, -244.75120544433594, -245.1510467529297, -245.87115478515625], "reading_logprob_std": 1.5864284707766036, "tokenization_path_logprob_std": 179.0941731171099, "trajectory_balance_rmse": 4292.154702195886} {"canonical_length": 28, "canonical_nll": 1.184464999607631, "generation_rouge_l_f1": 0.38532110091743116, "generation_token_f1": 0.4403669724770642, "prompt_id": "56997da805bc232c", "reading_canonical_log_prob": -33.16501998901367, "reading_distances": [25, 26, 29, 36], "reading_log_probs": [-31.05646324157715, -30.797605514526367, -32.65960693359375, -32.612998962402344], "reading_logprob_std": 0.9926712878428035, "tokenization_path_logprob_std": 38.052989006985335, "trajectory_balance_rmse": 905.5471477744104} {"canonical_length": 110, "canonical_nll": 1.6410519686612215, "generation_rouge_l_f1": 0.32298136645962733, "generation_token_f1": 0.4347826086956522, "prompt_id": "02d66047fd43a021", "reading_canonical_log_prob": -180.51571655273438, "reading_distances": [22, 19, 16, 18], "reading_log_probs": [-181.72708129882812, -178.30508422851562, -182.18991088867188, -182.03091430664062], "reading_logprob_std": 1.8487734118129509, "tokenization_path_logprob_std": 229.14218741337413, "trajectory_balance_rmse": 2937.561674072806} {"canonical_length": 79, "canonical_nll": 1.7271801912331883, "generation_rouge_l_f1": 0.14746543778801843, "generation_token_f1": 0.23963133640552994, "prompt_id": "83a9bce5dad053f6", "reading_canonical_log_prob": -136.44723510742188, "reading_distances": [20, 21, 22, 24], "reading_log_probs": [-152.28553771972656, -154.78936767578125, -152.33804321289062, -156.17286682128906], "reading_logprob_std": 1.915119283382909, "tokenization_path_logprob_std": 182.58240688304895, "trajectory_balance_rmse": 1920.583484382122} {"canonical_length": 21, "canonical_nll": 7.180714925130208, "generation_rouge_l_f1": 0.017167381974248927, "generation_token_f1": 0.025751072961373394, "prompt_id": "35f4df3b6373154b", "reading_canonical_log_prob": -150.79501342773438, "reading_distances": [20, 17, 23, 20], "reading_log_probs": [-146.2571258544922, -144.83621215820312, -145.75958251953125, -146.46926879882812], "reading_logprob_std": 0.7265610298974835, "tokenization_path_logprob_std": 36.46519719171495, "trajectory_balance_rmse": 313.76869012703577} {"canonical_length": 228, "canonical_nll": 1.3161795097484923, "generation_rouge_l_f1": 0.21132075471698117, "generation_token_f1": 0.42641509433962266, "prompt_id": "7c0757fa3b0cff4b", "reading_canonical_log_prob": -300.08892822265625, "reading_distances": [21, 29, 32, 26], "reading_log_probs": [-296.11407470703125, -292.3525390625, -296.7280578613281, -296.2605285644531], "reading_logprob_std": 2.0245103276451775, "tokenization_path_logprob_std": 333.4264811032544, "trajectory_balance_rmse": 3987.5292515207316} {"canonical_length": 211, "canonical_nll": 1.619594772845083, "generation_rouge_l_f1": 0.2516268980477223, "generation_token_f1": 0.5075921908893709, "prompt_id": "5fa3ea0df830456f", "reading_canonical_log_prob": -341.7344970703125, "reading_distances": [20, 27, 23, 19], "reading_log_probs": [-348.2915344238281, -346.409912109375, -344.65216064453125, -344.12518310546875], "reading_logprob_std": 1.8870481690505327, "tokenization_path_logprob_std": 153.15833944589488, "trajectory_balance_rmse": 6178.38004567884} {"canonical_length": 145, "canonical_nll": 1.3267165611530172, "generation_rouge_l_f1": 0.36893203883495146, "generation_token_f1": 0.4611650485436893, "prompt_id": "70669892361b488c", "reading_canonical_log_prob": -192.3739013671875, "reading_distances": [35, 36, 29, 32], "reading_log_probs": [-194.40072631835938, -197.38388061523438, -192.72532653808594, -191.57986450195312], "reading_logprob_std": 2.522634994391266, "tokenization_path_logprob_std": 203.19868332968014, "trajectory_balance_rmse": 3214.0230766622562} {"canonical_length": 12, "canonical_nll": 1.4886536598205566, "generation_rouge_l_f1": 0.5454545454545454, "generation_token_f1": 0.7272727272727273, "prompt_id": "e90b400c44335d9a", "reading_canonical_log_prob": -17.86384391784668, "reading_distances": [19, 18, 21, 25], "reading_log_probs": [-17.44660186767578, -17.388219833374023, -16.36460304260254, -16.78132438659668], "reading_logprob_std": 0.5169215077769472, "tokenization_path_logprob_std": 68.23589805713853, "trajectory_balance_rmse": 318.96620302890267} {"canonical_length": 232, "canonical_nll": 1.5024863933694774, "generation_rouge_l_f1": 0.2926829268292683, "generation_token_f1": 0.540650406504065, "prompt_id": "9f0a3bc3aaf621a1", "reading_canonical_log_prob": -348.57684326171875, "reading_distances": [20, 31, 25, 26], "reading_log_probs": [-346.85345458984375, -347.32525634765625, -345.8122863769531, -346.5954284667969], "reading_logprob_std": 0.6329984220094445, "tokenization_path_logprob_std": 325.3427062663205, "trajectory_balance_rmse": 6838.588749670427} {"canonical_length": 93, "canonical_nll": 0.6144190962596606, "generation_rouge_l_f1": 0.18731117824773413, "generation_token_f1": 0.32024169184290036, "prompt_id": "f2be334fa1e04283", "reading_canonical_log_prob": -57.14097595214844, "reading_distances": [25, 15, 23, 24], "reading_log_probs": [-54.88502502441406, -55.21425247192383, -56.40540313720703, -54.03786849975586], "reading_logprob_std": 0.9808819973300401, "tokenization_path_logprob_std": 98.82563078314949, "trajectory_balance_rmse": 2190.5859915908773} {"canonical_length": 54, "canonical_nll": 1.7763441580313224, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "1a42eb844c9543a9", "reading_canonical_log_prob": -95.9225845336914, "reading_distances": [45, 31, 40, 36], "reading_log_probs": [-103.48684692382812, -102.7252426147461, -100.23927307128906, -101.65653991699219], "reading_logprob_std": 1.4085339934867322, "tokenization_path_logprob_std": 49.61717859236925, "trajectory_balance_rmse": 1451.6220828671655} {"canonical_length": 40, "canonical_nll": 0.6399684429168702, "generation_rouge_l_f1": 0.15282392026578073, "generation_token_f1": 0.21262458471760798, "prompt_id": "3afed29700596cb5", "reading_canonical_log_prob": -25.598737716674805, "reading_distances": [11, 9, 16, 14], "reading_log_probs": [-28.54022979736328, -26.274658203125, -30.326486587524414, -31.99148941040039], "reading_logprob_std": 2.4512994444254246, "tokenization_path_logprob_std": 48.18972702346907, "trajectory_balance_rmse": 610.1994831402643} {"canonical_length": 28, "canonical_nll": 0.536959035055978, "generation_rouge_l_f1": 0.22608695652173913, "generation_token_f1": 0.4, "prompt_id": "e24f8c298fcc6b27", "reading_canonical_log_prob": -15.034852981567383, "reading_distances": [13, 12, 13, 12], "reading_log_probs": [-13.701958656311035, -14.011092185974121, -13.488139152526855, -14.042688369750977], "reading_logprob_std": 0.2644778007896218, "tokenization_path_logprob_std": 30.473593619985163, "trajectory_balance_rmse": 832.6468156329609} {"canonical_length": 232, "canonical_nll": 1.5666353949185075, "generation_rouge_l_f1": 0.24, "generation_token_f1": 0.508, "prompt_id": "d2492eb6d00f3204", "reading_canonical_log_prob": -363.45941162109375, "reading_distances": [31, 27, 20, 34], "reading_log_probs": [-361.8553771972656, -367.116943359375, -362.4358215332031, -363.2752380371094], "reading_logprob_std": 2.370183191083111, "tokenization_path_logprob_std": 406.71063375836917, "trajectory_balance_rmse": 7835.1727333625095} {"canonical_length": 83, "canonical_nll": 1.6758816270943147, "generation_rouge_l_f1": 0.26332288401253917, "generation_token_f1": 0.3448275862068965, "prompt_id": "d1d32462252d2ba1", "reading_canonical_log_prob": -139.09817504882812, "reading_distances": [18, 20, 14, 14], "reading_log_probs": [-140.86863708496094, -142.16259765625, -140.61395263671875, -142.69100952148438], "reading_logprob_std": 1.002161215574421, "tokenization_path_logprob_std": 153.78769511888373, "trajectory_balance_rmse": 2308.8342140365166} {"canonical_length": 171, "canonical_nll": 1.426010042603253, "generation_rouge_l_f1": 0.3023255813953488, "generation_token_f1": 0.4930232558139535, "prompt_id": "981d7ddcc229e43f", "reading_canonical_log_prob": -243.84771728515625, "reading_distances": [22, 19, 20, 21], "reading_log_probs": [-216.09625244140625, -235.8660888671875, -223.34011840820312, -216.41209411621094], "reading_logprob_std": 9.250113290799089, "tokenization_path_logprob_std": 40.03258280703022, "trajectory_balance_rmse": 4275.36481937363} {"canonical_length": 14, "canonical_nll": 2.161798749651228, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.42105263157894735, "prompt_id": "a353a5c04a14d88b", "reading_canonical_log_prob": -30.265182495117188, "reading_distances": [35, 31, 31, 35], "reading_log_probs": [-27.066242218017578, -26.801197052001953, -26.588991165161133, -28.572185516357422], "reading_logprob_std": 0.8981635912238943, "tokenization_path_logprob_std": 63.379951661585345, "trajectory_balance_rmse": 464.00864388844997} {"canonical_length": 37, "canonical_nll": 0.6629251531652502, "generation_rouge_l_f1": 0.17687074829931973, "generation_token_f1": 0.19047619047619047, "prompt_id": "104ca723f309949c", "reading_canonical_log_prob": -24.528230667114258, "reading_distances": [17, 15, 21, 14], "reading_log_probs": [-25.017803192138672, -25.142627716064453, -24.771005630493164, -24.588302612304688], "reading_logprob_std": 0.24828086150617865, "tokenization_path_logprob_std": 89.740754610437, "trajectory_balance_rmse": 732.3974660153415} {"canonical_length": 270, "canonical_nll": 0.9884454797815394, "generation_rouge_l_f1": 0.30103092783505153, "generation_token_f1": 0.5443298969072164, "prompt_id": "5768cefb5b534d32", "reading_canonical_log_prob": -266.8802795410156, "reading_distances": [22, 20, 25, 20], "reading_log_probs": [-268.00543212890625, -266.22100830078125, -264.271240234375, -265.72589111328125], "reading_logprob_std": 1.5407630811581108, "tokenization_path_logprob_std": 233.06799462366257, "trajectory_balance_rmse": 5870.55576333559} {"canonical_length": 460, "canonical_nll": 0.8060503088909646, "generation_rouge_l_f1": 0.2927536231884058, "generation_token_f1": 0.49565217391304345, "prompt_id": "2c20880d9a9f5257", "reading_canonical_log_prob": -370.78314208984375, "reading_distances": [31, 27, 35, 21], "reading_log_probs": [-368.0451965332031, -366.8426208496094, -366.6129150390625, -367.0001220703125], "reading_logprob_std": 0.633594235636803, "tokenization_path_logprob_std": 273.1189813551616, "trajectory_balance_rmse": 10163.174199306968} {"canonical_length": 43, "canonical_nll": 1.6056738565134447, "generation_rouge_l_f1": 0.3300970873786408, "generation_token_f1": 0.5631067961165049, "prompt_id": "1a2898fe9876d2bc", "reading_canonical_log_prob": -69.04397583007812, "reading_distances": [19, 17, 12, 16], "reading_log_probs": [-68.68097686767578, -74.83079528808594, -82.90351867675781, -73.90345001220703], "reading_logprob_std": 5.876544165846868, "tokenization_path_logprob_std": 107.43831389837044, "trajectory_balance_rmse": 1305.1595939474184} {"canonical_length": 85, "canonical_nll": 1.1639292997472426, "generation_rouge_l_f1": 0.3756345177664974, "generation_token_f1": 0.5583756345177665, "prompt_id": "52568183284a12f5", "reading_canonical_log_prob": -98.93399047851562, "reading_distances": [17, 20, 15, 17], "reading_log_probs": [-100.05067443847656, -102.37017822265625, -103.33229064941406, -100.27651977539062], "reading_logprob_std": 1.6033015553064243, "tokenization_path_logprob_std": 152.92302109082692, "trajectory_balance_rmse": 2486.1822220206} {"canonical_length": 202, "canonical_nll": 1.7385030312113243, "generation_rouge_l_f1": 0.25787965616045844, "generation_token_f1": 0.5214899713467049, "prompt_id": "ea255680cda4b618", "reading_canonical_log_prob": -351.1776123046875, "reading_distances": [29, 17, 25, 26], "reading_log_probs": [-351.5609130859375, -351.71337890625, -355.1105041503906, -347.7361755371094], "reading_logprob_std": 3.0137295406351634, "tokenization_path_logprob_std": 380.594259849917, "trajectory_balance_rmse": 5692.573122272973} {"canonical_length": 19, "canonical_nll": 1.4395869405646073, "generation_rouge_l_f1": 0.41379310344827586, "generation_token_f1": 0.41379310344827586, "prompt_id": "559aadd3a6bc5deb", "reading_canonical_log_prob": -27.35215187072754, "reading_distances": [32, 35, 35, 39], "reading_log_probs": [-29.309629440307617, -27.8243465423584, -29.06523895263672, -28.918594360351562], "reading_logprob_std": 0.656848367121128, "tokenization_path_logprob_std": 108.72783402955066, "trajectory_balance_rmse": 544.095700637376} {"canonical_length": 24, "canonical_nll": 0.7915045420328776, "generation_rouge_l_f1": 0.5945945945945946, "generation_token_f1": 0.5945945945945946, "prompt_id": "4196e10bb50daa8f", "reading_canonical_log_prob": -18.996109008789062, "reading_distances": [25, 23, 20, 18], "reading_log_probs": [-20.40125274658203, -20.915138244628906, -20.62680435180664, -23.197010040283203], "reading_logprob_std": 1.2918734227094215, "tokenization_path_logprob_std": 21.22808024807305, "trajectory_balance_rmse": 689.9100951748355} {"canonical_length": 293, "canonical_nll": 1.4747930012465338, "generation_rouge_l_f1": 0.2616822429906542, "generation_token_f1": 0.5046728971962617, "prompt_id": "0d2dbbf4ae30ce19", "reading_canonical_log_prob": -432.1143493652344, "reading_distances": [18, 13, 20, 21], "reading_log_probs": [-422.98883056640625, -426.75048828125, -427.7334289550781, -425.55328369140625], "reading_logprob_std": 2.049191040237445, "tokenization_path_logprob_std": 242.72241805050317, "trajectory_balance_rmse": 9469.027769843808} {"canonical_length": 227, "canonical_nll": 1.1481807221400055, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.4098765432098765, "prompt_id": "cd0a37baf57f0d68", "reading_canonical_log_prob": -260.63702392578125, "reading_distances": [23, 29, 26, 27], "reading_log_probs": [-255.90237426757812, -257.990966796875, -260.4930419921875, -259.8726501464844], "reading_logprob_std": 2.069330367116375, "tokenization_path_logprob_std": 339.2760960360585, "trajectory_balance_rmse": 5771.883073741255} {"canonical_length": 27, "canonical_nll": 1.9674148559570312, "generation_rouge_l_f1": 0.21875, "generation_token_f1": 0.265625, "prompt_id": "f0dd5c7b12d6ff90", "reading_canonical_log_prob": -53.120201110839844, "reading_distances": [23, 23, 19, 24], "reading_log_probs": [-68.58638000488281, -67.95358276367188, -68.40018463134766, -66.19001007080078], "reading_logprob_std": 1.0943873242566309, "tokenization_path_logprob_std": 117.86540097936376, "trajectory_balance_rmse": 714.1785344780986} {"canonical_length": 29, "canonical_nll": 0.9892083529768319, "generation_rouge_l_f1": 0.24242424242424243, "generation_token_f1": 0.30303030303030304, "prompt_id": "c737e978a37038bd", "reading_canonical_log_prob": -28.687042236328125, "reading_distances": [23, 22, 27, 26], "reading_log_probs": [-35.09766387939453, -34.44581604003906, -34.0384521484375, -33.969451904296875], "reading_logprob_std": 0.5177939198430181, "tokenization_path_logprob_std": 64.76880270963024, "trajectory_balance_rmse": 616.9104156630029} {"canonical_length": 90, "canonical_nll": 1.4670162624782985, "generation_rouge_l_f1": 0.3076923076923077, "generation_token_f1": 0.5274725274725275, "prompt_id": "6fe0c1612b081574", "reading_canonical_log_prob": -132.03146362304688, "reading_distances": [20, 25, 24, 27], "reading_log_probs": [-134.2584228515625, -136.83934020996094, -133.88040161132812, -131.17791748046875], "reading_logprob_std": 2.316678535467045, "tokenization_path_logprob_std": 294.7714503592176, "trajectory_balance_rmse": 2689.3327192732436} {"canonical_length": 276, "canonical_nll": 1.4117326598236526, "generation_rouge_l_f1": 0.2626262626262626, "generation_token_f1": 0.4646464646464647, "prompt_id": "a45d6c8378dedb7b", "reading_canonical_log_prob": -389.6382141113281, "reading_distances": [13, 14, 15, 14], "reading_log_probs": [-384.6635437011719, -382.7995300292969, -385.3968200683594, -386.82781982421875], "reading_logprob_std": 1.6762340443326629, "tokenization_path_logprob_std": 256.3723880156069, "trajectory_balance_rmse": 7448.880591751029} {"canonical_length": 103, "canonical_nll": 1.3876538322967233, "generation_rouge_l_f1": 0.27717391304347827, "generation_token_f1": 0.35869565217391297, "prompt_id": "2e7570c8d7724f9e", "reading_canonical_log_prob": -142.9283447265625, "reading_distances": [25, 22, 17, 25], "reading_log_probs": [-132.98171997070312, -135.58053588867188, -134.86117553710938, -136.11203002929688], "reading_logprob_std": 1.367773152550673, "tokenization_path_logprob_std": 87.28938643862774, "trajectory_balance_rmse": 1893.5991877348358} {"canonical_length": 30, "canonical_nll": 1.70821533203125, "generation_rouge_l_f1": 0.11971830985915494, "generation_token_f1": 0.1267605633802817, "prompt_id": "7aa76ccd6b778b98", "reading_canonical_log_prob": -51.2464599609375, "reading_distances": [40, 36, 27, 35], "reading_log_probs": [-53.743080139160156, -50.67623519897461, -51.726295471191406, -49.49544143676758], "reading_logprob_std": 1.8025214524201678, "tokenization_path_logprob_std": 139.3231172651829, "trajectory_balance_rmse": 1042.8093277623968} {"canonical_length": 252, "canonical_nll": 1.472929939391121, "generation_rouge_l_f1": 0.2875816993464052, "generation_token_f1": 0.5010893246187363, "prompt_id": "0667a89ba9993054", "reading_canonical_log_prob": -371.1783447265625, "reading_distances": [28, 27, 25, 26], "reading_log_probs": [-377.8800354003906, -377.68072509765625, -380.93450927734375, -377.87310791015625], "reading_logprob_std": 1.5643391633911967, "tokenization_path_logprob_std": 141.1389794750837, "trajectory_balance_rmse": 6110.4439596762295} {"canonical_length": 248, "canonical_nll": 1.2060146947060861, "generation_rouge_l_f1": 0.31517509727626464, "generation_token_f1": 0.5719844357976653, "prompt_id": "45c2d1c36166f6b6", "reading_canonical_log_prob": -299.0916442871094, "reading_distances": [24, 24, 22, 32], "reading_log_probs": [-295.4586181640625, -291.07958984375, -299.7442626953125, -291.9736633300781], "reading_logprob_std": 3.936429069090655, "tokenization_path_logprob_std": 160.42063820897417, "trajectory_balance_rmse": 5812.684129698311} {"canonical_length": 20, "canonical_nll": 0.8436938285827636, "generation_rouge_l_f1": 0.6, "generation_token_f1": 0.6, "prompt_id": "e5f7552dff198384", "reading_canonical_log_prob": -16.873876571655273, "reading_distances": [29, 20, 21, 27], "reading_log_probs": [-16.72589111328125, -17.33480453491211, -17.588376998901367, -17.504749298095703], "reading_logprob_std": 0.38959929621746586, "tokenization_path_logprob_std": 75.75848628307014, "trajectory_balance_rmse": 594.8009546091914} {"canonical_length": 117, "canonical_nll": 1.8303707807491987, "generation_rouge_l_f1": 0.2195121951219512, "generation_token_f1": 0.3719512195121951, "prompt_id": "4a930bccec9843cd", "reading_canonical_log_prob": -214.15338134765625, "reading_distances": [34, 32, 26, 27], "reading_log_probs": [-234.67034912109375, -219.60658264160156, -223.8926544189453, -227.78875732421875], "reading_logprob_std": 6.396158228226356, "tokenization_path_logprob_std": 316.8783579118931, "trajectory_balance_rmse": 3528.6562223402125} {"canonical_length": 49, "canonical_nll": 0.9794446205606266, "generation_rouge_l_f1": 0.15730337078651685, "generation_token_f1": 0.21722846441947566, "prompt_id": "35925631dfc295f8", "reading_canonical_log_prob": -47.9927864074707, "reading_distances": [33, 29, 36, 28], "reading_log_probs": [-47.659393310546875, -50.722599029541016, -49.47209167480469, -49.30529022216797], "reading_logprob_std": 1.2575915110723153, "tokenization_path_logprob_std": 175.1752950800172, "trajectory_balance_rmse": 1284.0312371658652} {"canonical_length": 295, "canonical_nll": 1.233795993610964, "generation_rouge_l_f1": 0.23272727272727273, "generation_token_f1": 0.5781818181818182, "prompt_id": "f6597a51ed7cc783", "reading_canonical_log_prob": -363.9698181152344, "reading_distances": [21, 21, 19, 23], "reading_log_probs": [-364.21942138671875, -364.0071716308594, -364.70635986328125, -362.6754455566406], "reading_logprob_std": 0.8685712331521658, "tokenization_path_logprob_std": 478.7884939768764, "trajectory_balance_rmse": 8015.149622996552} {"canonical_length": 213, "canonical_nll": 1.2947232958296655, "generation_rouge_l_f1": 0.1940928270042194, "generation_token_f1": 0.4852320675105486, "prompt_id": "87270c0f1888627a", "reading_canonical_log_prob": -275.77606201171875, "reading_distances": [13, 15, 9, 10], "reading_log_probs": [-279.0616760253906, -279.4869689941406, -278.61822509765625, -281.52520751953125], "reading_logprob_std": 1.284724007402467, "tokenization_path_logprob_std": 648.9688266275224, "trajectory_balance_rmse": 6434.154129854844} {"canonical_length": 14, "canonical_nll": 2.0345593861171176, "generation_rouge_l_f1": 0.10169491525423728, "generation_token_f1": 0.11299435028248588, "prompt_id": "a53e27e32808bf25", "reading_canonical_log_prob": -28.48383140563965, "reading_distances": [25, 20, 27, 27], "reading_log_probs": [-26.19059944152832, -25.748441696166992, -25.57733154296875, -25.900299072265625], "reading_logprob_std": 0.2602118790146402, "tokenization_path_logprob_std": 35.10681236440539, "trajectory_balance_rmse": 361.5127688873213} {"canonical_length": 128, "canonical_nll": 1.2343499660491943, "generation_rouge_l_f1": 0.32156862745098036, "generation_token_f1": 0.5647058823529412, "prompt_id": "1e1844cf1025dc97", "reading_canonical_log_prob": -157.99679565429688, "reading_distances": [15, 16, 18, 16], "reading_log_probs": [-157.63369750976562, -164.90512084960938, -159.8009796142578, -158.03305053710938], "reading_logprob_std": 3.3433188762604353, "tokenization_path_logprob_std": 214.97977479111236, "trajectory_balance_rmse": 3919.00677564755} {"canonical_length": 155, "canonical_nll": 1.6504840481665826, "generation_rouge_l_f1": 0.20821917808219179, "generation_token_f1": 0.35616438356164387, "prompt_id": "87981277aeac45ee", "reading_canonical_log_prob": -255.8250274658203, "reading_distances": [21, 21, 23, 18], "reading_log_probs": [-256.06427001953125, -253.24566650390625, -257.29022216796875, -254.0323486328125], "reading_logprob_std": 1.8521885305647925, "tokenization_path_logprob_std": 126.56264556585975, "trajectory_balance_rmse": 4042.0103999588387} {"canonical_length": 26, "canonical_nll": 0.3246425481942984, "generation_rouge_l_f1": 0.2159090909090909, "generation_token_f1": 0.2159090909090909, "prompt_id": "40b80040395896ae", "reading_canonical_log_prob": -8.440706253051758, "reading_distances": [39, 46, 45, 47], "reading_log_probs": [-2.9159443378448486, -3.5696725845336914, -3.261305809020996, -4.007434844970703], "reading_logprob_std": 0.46380860785349226, "tokenization_path_logprob_std": 126.78500145271583, "trajectory_balance_rmse": 417.12777419593465} {"canonical_length": 356, "canonical_nll": 1.5431933456592346, "generation_rouge_l_f1": 0.19826086956521738, "generation_token_f1": 0.40695652173913044, "prompt_id": "63d17b19976fa874", "reading_canonical_log_prob": -549.3768310546875, "reading_distances": [29, 29, 27, 20], "reading_log_probs": [-539.3985595703125, -537.7896728515625, -535.369873046875, -537.4855346679688], "reading_logprob_std": 1.6558595723304053, "tokenization_path_logprob_std": 329.91453855274926, "trajectory_balance_rmse": 11031.648601155044} {"canonical_length": 19, "canonical_nll": 0.7932738253944799, "generation_rouge_l_f1": 0.11764705882352941, "generation_token_f1": 0.11764705882352941, "prompt_id": "53bd5d756df134c2", "reading_canonical_log_prob": -15.072202682495117, "reading_distances": [27, 21, 25, 25], "reading_log_probs": [-11.062670707702637, -10.680763244628906, -11.477583885192871, -15.021162033081055], "reading_logprob_std": 2.000387268641561, "tokenization_path_logprob_std": 105.38469207093601, "trajectory_balance_rmse": 333.4672549745012} {"canonical_length": 114, "canonical_nll": 1.240139877587034, "generation_rouge_l_f1": 0.18343195266272191, "generation_token_f1": 0.34319526627218927, "prompt_id": "1cf0a058de57af94", "reading_canonical_log_prob": -141.37594604492188, "reading_distances": [38, 28, 33, 36], "reading_log_probs": [-159.9130859375, -156.7777099609375, -159.97457885742188, -159.30252075195312], "reading_logprob_std": 1.5070240019176888, "tokenization_path_logprob_std": 71.98367117608146, "trajectory_balance_rmse": 2106.653077197181} {"canonical_length": 257, "canonical_nll": 1.6103242510487596, "generation_rouge_l_f1": 0.23387096774193547, "generation_token_f1": 0.49193548387096775, "prompt_id": "a5e93d67499f0604", "reading_canonical_log_prob": -413.85333251953125, "reading_distances": [15, 19, 20, 19], "reading_log_probs": [-420.2448425292969, -422.32598876953125, -425.8037109375, -423.33056640625], "reading_logprob_std": 2.308932853722464, "tokenization_path_logprob_std": 367.7089372360126, "trajectory_balance_rmse": 6271.740416454652} {"canonical_length": 118, "canonical_nll": 1.5934754711086467, "generation_rouge_l_f1": 0.3056768558951965, "generation_token_f1": 0.5065502183406113, "prompt_id": "9493ef83593c3f2d", "reading_canonical_log_prob": -188.0301055908203, "reading_distances": [32, 36, 37, 32], "reading_log_probs": [-188.150634765625, -184.3378143310547, -188.90155029296875, -184.5992431640625], "reading_logprob_std": 2.365018211687537, "tokenization_path_logprob_std": 52.452615066084036, "trajectory_balance_rmse": 3404.53562586616} {"canonical_length": 22, "canonical_nll": 2.2668602683327417, "generation_rouge_l_f1": 0.21428571428571427, "generation_token_f1": 0.21428571428571427, "prompt_id": "be453082a5e24aed", "reading_canonical_log_prob": -49.87092590332031, "reading_distances": [35, 24, 25, 37], "reading_log_probs": [-54.669769287109375, -45.06504821777344, -56.84749221801758, -52.28733825683594], "reading_logprob_std": 5.119013461062509, "tokenization_path_logprob_std": 57.585481831170505, "trajectory_balance_rmse": 495.00870826441576} {"canonical_length": 280, "canonical_nll": 1.2457615443638392, "generation_rouge_l_f1": 0.32326820603907636, "generation_token_f1": 0.5719360568383659, "prompt_id": "6b6295e1afd98591", "reading_canonical_log_prob": -348.813232421875, "reading_distances": [14, 17, 17, 20], "reading_log_probs": [-352.2510986328125, -352.1757507324219, -351.4599609375, -352.3682861328125], "reading_logprob_std": 0.41026298888144797, "tokenization_path_logprob_std": 365.83560771456007, "trajectory_balance_rmse": 8214.838091678457} {"canonical_length": 200, "canonical_nll": 1.4481640625, "generation_rouge_l_f1": 0.24411134903640253, "generation_token_f1": 0.3854389721627409, "prompt_id": "900ff807cdf33dd9", "reading_canonical_log_prob": -289.6328125, "reading_distances": [23, 20, 27, 21], "reading_log_probs": [-295.6395263671875, -292.8954162597656, -287.0989685058594, -288.8958435058594], "reading_logprob_std": 3.859756856997798, "tokenization_path_logprob_std": 294.2309367969735, "trajectory_balance_rmse": 4957.7306422515085} {"canonical_length": 120, "canonical_nll": 1.5945579528808593, "generation_rouge_l_f1": 0.20543806646525678, "generation_token_f1": 0.3564954682779456, "prompt_id": "6b5f709d21a7a023", "reading_canonical_log_prob": -191.34695434570312, "reading_distances": [13, 19, 16, 18], "reading_log_probs": [-185.13519287109375, -193.49163818359375, -194.9622039794922, -182.38571166992188], "reading_logprob_std": 6.175437223123116, "tokenization_path_logprob_std": 127.4244266159221, "trajectory_balance_rmse": 3212.7458444226627} {"canonical_length": 52, "canonical_nll": 1.1877689361572266, "generation_rouge_l_f1": 0.3612903225806452, "generation_token_f1": 0.46451612903225814, "prompt_id": "e072285f618dcbd9", "reading_canonical_log_prob": -61.76398468017578, "reading_distances": [28, 34, 21, 27], "reading_log_probs": [-59.983543395996094, -60.079429626464844, -59.24606704711914, -57.70852279663086], "reading_logprob_std": 1.0957690970462735, "tokenization_path_logprob_std": 60.05271248641231, "trajectory_balance_rmse": 1730.117803878983} {"canonical_length": 147, "canonical_nll": 1.5076964501620962, "generation_rouge_l_f1": 0.16019417475728154, "generation_token_f1": 0.2961165048543689, "prompt_id": "21874077bcbf7b19", "reading_canonical_log_prob": -221.63137817382812, "reading_distances": [27, 20, 22, 23], "reading_log_probs": [-222.64064025878906, -223.08303833007812, -222.34542846679688, -222.78753662109375], "reading_logprob_std": 0.30704149928363794, "tokenization_path_logprob_std": 197.25369998891816, "trajectory_balance_rmse": 5956.306148716317} {"canonical_length": 108, "canonical_nll": 1.2151490670663339, "generation_rouge_l_f1": 0.18384401114206128, "generation_token_f1": 0.3899721448467967, "prompt_id": "8ba54ff004576b32", "reading_canonical_log_prob": -131.23609924316406, "reading_distances": [20, 23, 22, 21], "reading_log_probs": [-129.40707397460938, -128.37002563476562, -128.6155548095703, -128.83084106445312], "reading_logprob_std": 0.44281296107986035, "tokenization_path_logprob_std": 238.08208677516365, "trajectory_balance_rmse": 3770.6244893843987} {"canonical_length": 8, "canonical_nll": 0.03245231881737709, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "bccef35e52f5dc4b", "reading_canonical_log_prob": -0.2596185505390167, "reading_distances": [34, 34, 34, 34], "reading_log_probs": [-0.7941974401473999, -0.8928654193878174, -0.7317632436752319, -0.6646268367767334], "reading_logprob_std": 0.09702920480334663, "tokenization_path_logprob_std": 55.83622007420032, "trajectory_balance_rmse": 220.01807349578522} {"canonical_length": 53, "canonical_nll": 1.2015524090461012, "generation_rouge_l_f1": 0.15503875968992248, "generation_token_f1": 0.2248062015503876, "prompt_id": "492f8c42a0de57b2", "reading_canonical_log_prob": -63.68227767944336, "reading_distances": [47, 39, 37, 45], "reading_log_probs": [-61.124691009521484, -62.78502655029297, -60.16804885864258, -59.45600128173828], "reading_logprob_std": 1.4403239493670348, "tokenization_path_logprob_std": 118.06830584185607, "trajectory_balance_rmse": 1341.4406227222878} {"canonical_length": 211, "canonical_nll": 1.347064990002962, "generation_rouge_l_f1": 0.20361990950226244, "generation_token_f1": 0.416289592760181, "prompt_id": "6aa6f6caad65a47f", "reading_canonical_log_prob": -284.230712890625, "reading_distances": [21, 21, 29, 27], "reading_log_probs": [-290.02301025390625, -287.89605712890625, -290.1646423339844, -293.3114013671875], "reading_logprob_std": 2.2310690602786587, "tokenization_path_logprob_std": 312.139087449862, "trajectory_balance_rmse": 5154.828109546727} {"canonical_length": 107, "canonical_nll": 1.6097072708272488, "generation_rouge_l_f1": 0.3010752688172043, "generation_token_f1": 0.4157706093189964, "prompt_id": "83dae6df424d2709", "reading_canonical_log_prob": -172.23867797851562, "reading_distances": [31, 32, 29, 25], "reading_log_probs": [-177.05960083007812, -175.61309814453125, -181.07003784179688, -176.39608764648438], "reading_logprob_std": 2.4299056675291717, "tokenization_path_logprob_std": 127.99088355580338, "trajectory_balance_rmse": 2830.1148404893993} {"canonical_length": 104, "canonical_nll": 1.361977797288161, "generation_rouge_l_f1": 0.19302949061662197, "generation_token_f1": 0.3914209115281502, "prompt_id": "d7aad76280327bd4", "reading_canonical_log_prob": -141.64569091796875, "reading_distances": [28, 32, 36, 32], "reading_log_probs": [-143.6541748046875, -145.37083435058594, -146.24407958984375, -147.21334838867188], "reading_logprob_std": 1.511597957638526, "tokenization_path_logprob_std": 136.56994548615762, "trajectory_balance_rmse": 3232.521808542889} {"canonical_length": 362, "canonical_nll": 1.5462727414968922, "generation_rouge_l_f1": 0.13290113452188007, "generation_token_f1": 0.2301458670988655, "prompt_id": "117db47d424e536b", "reading_canonical_log_prob": -559.750732421875, "reading_distances": [27, 15, 15, 23], "reading_log_probs": [-565.614013671875, -565.082763671875, -565.614990234375, -562.4707641601562], "reading_logprob_std": 1.5042774893045576, "tokenization_path_logprob_std": 333.7487539988439, "trajectory_balance_rmse": 9297.20101446373} {"canonical_length": 207, "canonical_nll": 1.621801993697162, "generation_rouge_l_f1": 0.2796610169491526, "generation_token_f1": 0.5042372881355932, "prompt_id": "f593668e2032bc26", "reading_canonical_log_prob": -335.7130126953125, "reading_distances": [32, 31, 31, 33], "reading_log_probs": [-350.6189880371094, -344.89727783203125, -348.9000244140625, -349.7297058105469], "reading_logprob_std": 2.525640810994134, "tokenization_path_logprob_std": 314.2790035665548, "trajectory_balance_rmse": 6081.836434506001} {"canonical_length": 243, "canonical_nll": 1.1144145447530864, "generation_rouge_l_f1": 0.289532293986637, "generation_token_f1": 0.5523385300668151, "prompt_id": "62a2987fac290e0c", "reading_canonical_log_prob": -270.802734375, "reading_distances": [13, 16, 18, 14], "reading_log_probs": [-272.9127197265625, -278.1134033203125, -277.3869934082031, -275.93438720703125], "reading_logprob_std": 2.3018571585620053, "tokenization_path_logprob_std": 348.3847076005127, "trajectory_balance_rmse": 6857.15085173511} {"canonical_length": 28, "canonical_nll": 0.8242286273411342, "generation_rouge_l_f1": 0.2142857142857143, "generation_token_f1": 0.25, "prompt_id": "0597c4917add4bdd", "reading_canonical_log_prob": -23.078401565551758, "reading_distances": [17, 16, 17, 20], "reading_log_probs": [-18.493404388427734, -22.708904266357422, -19.7042236328125, -22.339427947998047], "reading_logprob_std": 2.044042969345484, "tokenization_path_logprob_std": 60.22635121628196, "trajectory_balance_rmse": 470.7529390916948} {"canonical_length": 51, "canonical_nll": 2.039363786285999, "generation_rouge_l_f1": 0.09302325581395349, "generation_token_f1": 0.13372093023255813, "prompt_id": "ea67625f5161437c", "reading_canonical_log_prob": -104.00755310058594, "reading_distances": [19, 21, 20, 20], "reading_log_probs": [-102.20330047607422, -102.46389770507812, -99.78424072265625, -102.49703979492188], "reading_logprob_std": 1.3085291533453047, "tokenization_path_logprob_std": 171.70839374869286, "trajectory_balance_rmse": 1532.6019849507186} {"canonical_length": 98, "canonical_nll": 0.9575740269252232, "generation_rouge_l_f1": 0.23577235772357724, "generation_token_f1": 0.34959349593495936, "prompt_id": "f9bf7abf6deb13c3", "reading_canonical_log_prob": -93.84225463867188, "reading_distances": [17, 20, 14, 24], "reading_log_probs": [-93.12052154541016, -93.36497497558594, -92.55534362792969, -92.41490173339844], "reading_logprob_std": 0.4523041829900527, "tokenization_path_logprob_std": 188.34407568283356, "trajectory_balance_rmse": 3033.9490137938183} {"canonical_length": 144, "canonical_nll": 1.7059672673543294, "generation_rouge_l_f1": 0.3146666666666667, "generation_token_f1": 0.43200000000000005, "prompt_id": "dd61adf3859cf501", "reading_canonical_log_prob": -245.65928649902344, "reading_distances": [40, 41, 39, 36], "reading_log_probs": [-275.7322998046875, -267.44580078125, -270.9098815917969, -266.26641845703125], "reading_logprob_std": 4.247325056098666, "tokenization_path_logprob_std": 191.9202115048972, "trajectory_balance_rmse": 3317.44944279461} {"canonical_length": 38, "canonical_nll": 2.818751485724198, "generation_rouge_l_f1": 0.1806451612903226, "generation_token_f1": 0.2967741935483871, "prompt_id": "932b65e0c79cc6cd", "reading_canonical_log_prob": -107.11255645751953, "reading_distances": [31, 23, 29, 24], "reading_log_probs": [-118.26783752441406, -107.90312957763672, -118.44366455078125, -114.4741439819336], "reading_logprob_std": 4.931940101923322, "tokenization_path_logprob_std": 65.85949781061306, "trajectory_balance_rmse": 805.1625173934156} {"canonical_length": 246, "canonical_nll": 1.4278940340367758, "generation_rouge_l_f1": 0.2357563850687623, "generation_token_f1": 0.39685658153241654, "prompt_id": "5a38778847ffa7e8", "reading_canonical_log_prob": -351.2619323730469, "reading_distances": [28, 29, 34, 30], "reading_log_probs": [-355.203125, -352.4476318359375, -353.09857177734375, -349.399169921875], "reading_logprob_std": 2.399829826087226, "tokenization_path_logprob_std": 344.7944731579202, "trajectory_balance_rmse": 7620.891190780785} {"canonical_length": 163, "canonical_nll": 0.9053670497028374, "generation_rouge_l_f1": 0.3052631578947368, "generation_token_f1": 0.4947368421052632, "prompt_id": "bea3ed585e39be34", "reading_canonical_log_prob": -147.5748291015625, "reading_distances": [19, 15, 16, 14], "reading_log_probs": [-149.48953247070312, -147.71226501464844, -149.54054260253906, -150.19187927246094], "reading_logprob_std": 1.0634024770109216, "tokenization_path_logprob_std": 294.9106149640125, "trajectory_balance_rmse": 3912.8478756589598} {"canonical_length": 256, "canonical_nll": 0.7242609262466431, "generation_rouge_l_f1": 0.35714285714285715, "generation_token_f1": 0.6260504201680673, "prompt_id": "e77970903b119833", "reading_canonical_log_prob": -185.41079711914062, "reading_distances": [21, 26, 26, 33], "reading_log_probs": [-196.34661865234375, -202.620361328125, -201.88656616210938, -204.87924194335938], "reading_logprob_std": 3.6223108543958666, "tokenization_path_logprob_std": 215.5602549689199, "trajectory_balance_rmse": 3268.8422748268563} {"canonical_length": 28, "canonical_nll": 1.381157738821847, "generation_rouge_l_f1": 0.08264462809917356, "generation_token_f1": 0.13223140495867766, "prompt_id": "4ea6cdd31bfaf581", "reading_canonical_log_prob": -38.67241668701172, "reading_distances": [24, 17, 19, 19], "reading_log_probs": [-41.24201202392578, -42.3705940246582, -40.54737091064453, -42.97454071044922], "reading_logprob_std": 1.0930815240296592, "tokenization_path_logprob_std": 6.521815551109264, "trajectory_balance_rmse": 271.2249342782591} {"canonical_length": 92, "canonical_nll": 0.8514808986497961, "generation_rouge_l_f1": 0.5714285714285714, "generation_token_f1": 0.6938775510204082, "prompt_id": "c7f44052e9719f7b", "reading_canonical_log_prob": -78.33624267578125, "reading_distances": [30, 36, 40, 40], "reading_log_probs": [-78.0851821899414, -77.03396606445312, -75.5035171508789, -78.59991455078125], "reading_logprob_std": 1.3667828563429458, "tokenization_path_logprob_std": 231.9748382627612, "trajectory_balance_rmse": 2454.998307951595} {"canonical_length": 22, "canonical_nll": 2.230008905584162, "generation_rouge_l_f1": 0.5128205128205129, "generation_token_f1": 0.5641025641025641, "prompt_id": "4d4bec1ba05bf153", "reading_canonical_log_prob": -49.06019592285156, "reading_distances": [41, 31, 45, 38], "reading_log_probs": [-52.33565139770508, -61.89569854736328, -62.522178649902344, -55.21706008911133], "reading_logprob_std": 5.015175314088314, "tokenization_path_logprob_std": 82.91504749124103, "trajectory_balance_rmse": 978.9392889431485} {"canonical_length": 56, "canonical_nll": 2.272982597351074, "generation_rouge_l_f1": 0.18, "generation_token_f1": 0.2, "prompt_id": "2dc7b73ea6e1f7f9", "reading_canonical_log_prob": -127.28702545166016, "reading_distances": [45, 47, 42, 49], "reading_log_probs": [-130.29612731933594, -135.96788024902344, -133.19839477539062, -135.4237823486328], "reading_logprob_std": 2.5788115147999835, "tokenization_path_logprob_std": 184.89135498771682, "trajectory_balance_rmse": 1743.810423111268} {"canonical_length": 14, "canonical_nll": 2.1219125475202287, "generation_rouge_l_f1": 0.13793103448275862, "generation_token_f1": 0.20689655172413793, "prompt_id": "cc99639a6604cf81", "reading_canonical_log_prob": -29.706775665283203, "reading_distances": [37, 35, 37, 33], "reading_log_probs": [-39.91273498535156, -31.01329231262207, -35.53297424316406, -33.91248321533203], "reading_logprob_std": 3.717573459961664, "tokenization_path_logprob_std": 20.359494900748345, "trajectory_balance_rmse": 395.26973363028907} {"canonical_length": 41, "canonical_nll": 1.190271051918588, "generation_rouge_l_f1": 0.232258064516129, "generation_token_f1": 0.29677419354838713, "prompt_id": "34af9dd76e319507", "reading_canonical_log_prob": -48.80111312866211, "reading_distances": [29, 35, 37, 35], "reading_log_probs": [-55.62872314453125, -49.77668762207031, -58.16607666015625, -50.253501892089844], "reading_logprob_std": 4.110915610428623, "tokenization_path_logprob_std": 128.47651875918893, "trajectory_balance_rmse": 1141.5158422317727} {"canonical_length": 319, "canonical_nll": 1.40345309744808, "generation_rouge_l_f1": 0.3159722222222222, "generation_token_f1": 0.5069444444444445, "prompt_id": "761d4b8b4ce8e2b9", "reading_canonical_log_prob": -447.7015380859375, "reading_distances": [29, 23, 27, 31], "reading_log_probs": [-450.85797119140625, -446.49603271484375, -447.47222900390625, -448.9224548339844], "reading_logprob_std": 1.8969189812395209, "tokenization_path_logprob_std": 338.1763970610984, "trajectory_balance_rmse": 9552.383791424872} {"canonical_length": 30, "canonical_nll": 0.4548694610595703, "generation_rouge_l_f1": 0.7234042553191491, "generation_token_f1": 0.8936170212765958, "prompt_id": "667ddd428394372e", "reading_canonical_log_prob": -13.64608383178711, "reading_distances": [32, 33, 31, 34], "reading_log_probs": [-15.7443208694458, -14.793878555297852, -13.003620147705078, -14.793564796447754], "reading_logprob_std": 1.1448306009689624, "tokenization_path_logprob_std": 82.64121258359279, "trajectory_balance_rmse": 842.0437529282623} {"canonical_length": 15, "canonical_nll": 0.7098648071289062, "generation_rouge_l_f1": 0.14218009478672988, "generation_token_f1": 0.14218009478672988, "prompt_id": "eaab92bbea496d12", "reading_canonical_log_prob": -10.647972106933594, "reading_distances": [28, 24, 20, 23], "reading_log_probs": [-10.77435302734375, -11.329453468322754, -11.460079193115234, -10.840595245361328], "reading_logprob_std": 0.3443049410413636, "tokenization_path_logprob_std": 47.98327480157624, "trajectory_balance_rmse": 447.6909899656664} {"canonical_length": 77, "canonical_nll": 0.8896266392299107, "generation_rouge_l_f1": 0.6888888888888889, "generation_token_f1": 0.8222222222222222, "prompt_id": "7cb40862b1b15473", "reading_canonical_log_prob": -68.50125122070312, "reading_distances": [33, 25, 25, 25], "reading_log_probs": [-67.96267700195312, -68.01518249511719, -72.84544372558594, -67.90121459960938], "reading_logprob_std": 2.4433201321334104, "tokenization_path_logprob_std": 155.72185067606748, "trajectory_balance_rmse": 952.596382477334} {"canonical_length": 37, "canonical_nll": 1.5539119823558911, "generation_rouge_l_f1": 0.20224719101123592, "generation_token_f1": 0.29213483146067415, "prompt_id": "b218ec9e88427665", "reading_canonical_log_prob": -57.49474334716797, "reading_distances": [28, 35, 29, 27], "reading_log_probs": [-70.42164611816406, -67.47008514404297, -67.50563049316406, -70.60387420654297], "reading_logprob_std": 1.7480722204038965, "tokenization_path_logprob_std": 153.51936830411748, "trajectory_balance_rmse": 1299.2118700009823} {"canonical_length": 68, "canonical_nll": 1.702455857220818, "generation_rouge_l_f1": 0.2035928143712575, "generation_token_f1": 0.29341317365269465, "prompt_id": "605c5bbf05c7cb07", "reading_canonical_log_prob": -115.76699829101562, "reading_distances": [34, 31, 35, 28], "reading_log_probs": [-113.53433227539062, -110.54438781738281, -111.18231201171875, -110.58213806152344], "reading_logprob_std": 1.4129106774147275, "tokenization_path_logprob_std": 168.89434412211605, "trajectory_balance_rmse": 2761.008763277369} {"canonical_length": 163, "canonical_nll": 2.2851356553153757, "generation_rouge_l_f1": 0.16786570743405277, "generation_token_f1": 0.35011990407673865, "prompt_id": "371fd9b264246fc1", "reading_canonical_log_prob": -372.47711181640625, "reading_distances": [18, 15, 18, 17], "reading_log_probs": [-373.60821533203125, -370.3698425292969, -371.241455078125, -370.93463134765625], "reading_logprob_std": 1.4262275261642678, "tokenization_path_logprob_std": 271.030885894276, "trajectory_balance_rmse": 4607.075358618949} {"canonical_length": 9, "canonical_nll": 2.2317220899793835, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.3157894736842105, "prompt_id": "ed3c13f24a96d3b3", "reading_canonical_log_prob": -20.085498809814453, "reading_distances": [28, 24, 28, 29], "reading_log_probs": [-25.656030654907227, -22.744449615478516, -23.662919998168945, -21.978008270263672], "reading_logprob_std": 1.5876500089616339, "tokenization_path_logprob_std": 38.279839093946016, "trajectory_balance_rmse": 232.52035873541624} {"canonical_length": 27, "canonical_nll": 0.9520060927779587, "generation_rouge_l_f1": 0.4482758620689655, "generation_token_f1": 0.48275862068965514, "prompt_id": "9e311e757597dd6a", "reading_canonical_log_prob": -25.704164505004883, "reading_distances": [32, 34, 33, 34], "reading_log_probs": [-22.485631942749023, -25.368392944335938, -25.6094970703125, -24.577083587646484], "reading_logprob_std": 1.419894174670529, "tokenization_path_logprob_std": 82.07851707965744, "trajectory_balance_rmse": 820.7359787849341} {"canonical_length": 23, "canonical_nll": 1.3642053189484968, "generation_rouge_l_f1": 0.31818181818181823, "generation_token_f1": 0.31818181818181823, "prompt_id": "6d2468c671bba59a", "reading_canonical_log_prob": -31.37672233581543, "reading_distances": [38, 41, 45, 45], "reading_log_probs": [-22.600666046142578, -24.587081909179688, -25.319400787353516, -23.74864387512207], "reading_logprob_std": 1.1676816854858276, "tokenization_path_logprob_std": 75.87551558722018, "trajectory_balance_rmse": 610.0649928267253} {"canonical_length": 34, "canonical_nll": 2.0659260469324447, "generation_rouge_l_f1": 0.19704433497536947, "generation_token_f1": 0.2561576354679803, "prompt_id": "d8ebdfdb02245d42", "reading_canonical_log_prob": -70.24148559570312, "reading_distances": [20, 20, 21, 20], "reading_log_probs": [-68.23048400878906, -67.49307250976562, -68.95790100097656, -71.81498718261719], "reading_logprob_std": 1.890969429507736, "tokenization_path_logprob_std": 97.6243444068196, "trajectory_balance_rmse": 1262.8434045760348} {"canonical_length": 282, "canonical_nll": 1.278599407655973, "generation_rouge_l_f1": 0.20530973451327433, "generation_token_f1": 0.4530973451327434, "prompt_id": "0659b1d0d1c9b272", "reading_canonical_log_prob": -360.5650329589844, "reading_distances": [22, 21, 19, 21], "reading_log_probs": [-361.817138671875, -362.56683349609375, -365.890869140625, -362.2742919921875], "reading_logprob_std": 1.8614675032524193, "tokenization_path_logprob_std": 181.43050901640396, "trajectory_balance_rmse": 7195.651388021359} {"canonical_length": 126, "canonical_nll": 1.6287792145259796, "generation_rouge_l_f1": 0.26294820717131473, "generation_token_f1": 0.5258964143426295, "prompt_id": "11e44461a25c9604", "reading_canonical_log_prob": -205.22618103027344, "reading_distances": [13, 16, 12, 13], "reading_log_probs": [-206.72991943359375, -206.37332153320312, -206.150634765625, -205.97984313964844], "reading_logprob_std": 0.32389963183033144, "tokenization_path_logprob_std": 236.52159054023227, "trajectory_balance_rmse": 4475.544307476844} {"canonical_length": 4, "canonical_nll": 1.2132052183151245, "generation_rouge_l_f1": 0.011363636363636362, "generation_token_f1": 0.011363636363636362, "prompt_id": "72cbdeacc12e1d9a", "reading_canonical_log_prob": -4.852820873260498, "reading_distances": [20, 24, 19, 21], "reading_log_probs": [-5.993967056274414, -5.63615608215332, -6.2016682624816895, -4.918857097625732], "reading_logprob_std": 0.5632462643336812, "tokenization_path_logprob_std": 33.924726753428004, "trajectory_balance_rmse": 91.22684155850327} {"canonical_length": 9, "canonical_nll": 1.2341384887695312, "generation_rouge_l_f1": 0.75, "generation_token_f1": 0.75, "prompt_id": "e46df451982e05b1", "reading_canonical_log_prob": -11.107246398925781, "reading_distances": [16, 22, 15, 22], "reading_log_probs": [-16.02937126159668, -14.53587532043457, -14.047643661499023, -15.018574714660645], "reading_logprob_std": 0.8462446471776072, "tokenization_path_logprob_std": 31.843152197472516, "trajectory_balance_rmse": 135.64640518154795} {"canonical_length": 112, "canonical_nll": 1.1285957608904158, "generation_rouge_l_f1": 0.37260273972602737, "generation_token_f1": 0.536986301369863, "prompt_id": "ad4b8c7e2797f458", "reading_canonical_log_prob": -126.40272521972656, "reading_distances": [36, 35, 37, 30], "reading_log_probs": [-137.86215209960938, -138.2747344970703, -139.96087646484375, -142.1182098388672], "reading_logprob_std": 1.9355955242289602, "tokenization_path_logprob_std": 161.39048482343327, "trajectory_balance_rmse": 2436.4821205795856} {"canonical_length": 229, "canonical_nll": 1.3984574896799946, "generation_rouge_l_f1": 0.2274678111587983, "generation_token_f1": 0.49785407725321895, "prompt_id": "9d2ca51ff7ce1141", "reading_canonical_log_prob": -320.24676513671875, "reading_distances": [29, 33, 32, 26], "reading_log_probs": [-322.94451904296875, -322.9765625, -316.0874328613281, -315.5308837890625], "reading_logprob_std": 4.1351202207263755, "tokenization_path_logprob_std": 254.52671731976278, "trajectory_balance_rmse": 7348.626003374761} {"canonical_length": 3, "canonical_nll": 2.0100577672322593, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "f22d1f7967fa63cf", "reading_canonical_log_prob": -6.030173301696777, "reading_distances": [24, 31, 24, 32], "reading_log_probs": [-6.7859930992126465, -7.588469505310059, -7.083093166351318, -7.425004482269287], "reading_logprob_std": 0.3581901014827414, "tokenization_path_logprob_std": 44.05487331346218, "trajectory_balance_rmse": 73.97286592465188} {"canonical_length": 341, "canonical_nll": 1.476322207632652, "generation_rouge_l_f1": 0.1943048576214405, "generation_token_f1": 0.3685092127303183, "prompt_id": "a546d753c3fe436f", "reading_canonical_log_prob": -503.4258728027344, "reading_distances": [20, 22, 25, 22], "reading_log_probs": [-505.29046630859375, -499.83197021484375, -498.1436462402344, -500.3932800292969], "reading_logprob_std": 3.0697585258596454, "tokenization_path_logprob_std": 195.31304324366997, "trajectory_balance_rmse": 9878.766558002482} {"canonical_length": 175, "canonical_nll": 1.4094131905691965, "generation_rouge_l_f1": 0.23218997361477572, "generation_token_f1": 0.4274406332453826, "prompt_id": "55985253cc61ade7", "reading_canonical_log_prob": -246.64730834960938, "reading_distances": [25, 30, 26, 24], "reading_log_probs": [-248.24984741210938, -251.5251922607422, -252.69338989257812, -253.27487182617188], "reading_logprob_std": 2.245143082725562, "tokenization_path_logprob_std": 132.0845968848129, "trajectory_balance_rmse": 4112.407715880095} {"canonical_length": 134, "canonical_nll": 1.5824056596898322, "generation_rouge_l_f1": 0.2606516290726817, "generation_token_f1": 0.45614035087719296, "prompt_id": "e4db8e939e6495ed", "reading_canonical_log_prob": -212.0423583984375, "reading_distances": [17, 15, 20, 15], "reading_log_probs": [-209.3638916015625, -210.15283203125, -208.25555419921875, -208.2999267578125], "reading_logprob_std": 0.9136807445697961, "tokenization_path_logprob_std": 243.87136359801536, "trajectory_balance_rmse": 4214.134693725769} {"canonical_length": 36, "canonical_nll": 0.7499369515313042, "generation_rouge_l_f1": 0.40816326530612246, "generation_token_f1": 0.4489795918367347, "prompt_id": "8ba984bcebc3168b", "reading_canonical_log_prob": -26.997730255126953, "reading_distances": [16, 17, 15, 15], "reading_log_probs": [-31.012813568115234, -30.28150749206543, -30.33684539794922, -28.67055892944336], "reading_logprob_std": 0.9938404670222613, "tokenization_path_logprob_std": 74.80488301947982, "trajectory_balance_rmse": 1018.3820894099015} {"canonical_length": 4, "canonical_nll": 3.121133804321289, "generation_rouge_l_f1": 0.015151515151515154, "generation_token_f1": 0.015151515151515154, "prompt_id": "fe48fe173961fdd5", "reading_canonical_log_prob": -12.484535217285156, "reading_distances": [18, 27, 24, 23], "reading_log_probs": [-12.221600532531738, -14.194250106811523, -12.914987564086914, -13.304048538208008], "reading_logprob_std": 0.8228084796888285, "tokenization_path_logprob_std": 26.03747619099873, "trajectory_balance_rmse": 57.57772442778603} {"canonical_length": 267, "canonical_nll": 1.7908041736159879, "generation_rouge_l_f1": 0.186219739292365, "generation_token_f1": 0.33519553072625696, "prompt_id": "51d00b55119644b0", "reading_canonical_log_prob": -478.14471435546875, "reading_distances": [25, 28, 26, 25], "reading_log_probs": [-487.6377258300781, -489.400390625, -488.4499816894531, -486.33941650390625], "reading_logprob_std": 1.2967805964511492, "tokenization_path_logprob_std": 272.8588449184326, "trajectory_balance_rmse": 7989.880407641107} {"canonical_length": 7, "canonical_nll": 0.007913623537336077, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "e708a1adc69b9f0e", "reading_canonical_log_prob": -0.05539536476135254, "reading_distances": [22, 28, 24, 25], "reading_log_probs": [-0.019852736964821815, -0.022157570347189903, -0.022145526483654976, -0.024210922420024872], "reading_logprob_std": 0.001780570632403434, "tokenization_path_logprob_std": 21.4005091448111, "trajectory_balance_rmse": 159.36403809195437} {"canonical_length": 22, "canonical_nll": 1.10869685086337, "generation_rouge_l_f1": 0.07874015748031497, "generation_token_f1": 0.10236220472440945, "prompt_id": "d7a89d7235636e3d", "reading_canonical_log_prob": -24.39133071899414, "reading_distances": [14, 16, 15, 14], "reading_log_probs": [-23.803546905517578, -25.21104621887207, -27.116004943847656, -25.56934356689453], "reading_logprob_std": 1.3607864322558723, "tokenization_path_logprob_std": 69.98865091158595, "trajectory_balance_rmse": 686.8340206714925} {"canonical_length": 12, "canonical_nll": 0.5671176910400391, "generation_rouge_l_f1": 0.8571428571428572, "generation_token_f1": 0.8571428571428572, "prompt_id": "17e2c9ee9612ad40", "reading_canonical_log_prob": -6.805412292480469, "reading_distances": [21, 22, 23, 29], "reading_log_probs": [-6.78049898147583, -9.051919937133789, -9.93136978149414, -9.236984252929688], "reading_logprob_std": 1.3666111351650738, "tokenization_path_logprob_std": 41.93270042978074, "trajectory_balance_rmse": 397.32103958991604} {"canonical_length": 324, "canonical_nll": 1.266379839108314, "generation_rouge_l_f1": 0.2057142857142857, "generation_token_f1": 0.419047619047619, "prompt_id": "36541d1f10617d25", "reading_canonical_log_prob": -410.30706787109375, "reading_distances": [11, 11, 15, 17], "reading_log_probs": [-420.1129150390625, -424.192626953125, -417.948974609375, -420.44940185546875], "reading_logprob_std": 2.593051227109055, "tokenization_path_logprob_std": 291.99421766839697, "trajectory_balance_rmse": 8427.03103740409} {"canonical_length": 315, "canonical_nll": 1.1987388005332342, "generation_rouge_l_f1": 0.2617801047120419, "generation_token_f1": 0.5828970331588132, "prompt_id": "e15e866d282a743f", "reading_canonical_log_prob": -377.60272216796875, "reading_distances": [24, 27, 30, 33], "reading_log_probs": [-378.07855224609375, -384.2868347167969, -379.43414306640625, -389.048095703125], "reading_logprob_std": 4.994643706834226, "tokenization_path_logprob_std": 356.9424798540475, "trajectory_balance_rmse": 7895.034496139567} {"canonical_length": 10, "canonical_nll": 2.203978729248047, "generation_rouge_l_f1": 0.053571428571428575, "generation_token_f1": 0.053571428571428575, "prompt_id": "718643fa581e9e44", "reading_canonical_log_prob": -22.03978729248047, "reading_distances": [18, 21, 22, 21], "reading_log_probs": [-22.04310417175293, -21.454130172729492, -20.539234161376953, -22.193647384643555], "reading_logprob_std": 0.7501212218816509, "tokenization_path_logprob_std": 29.093542708793354, "trajectory_balance_rmse": 163.96034209084976} {"canonical_length": 159, "canonical_nll": 1.0732194432672464, "generation_rouge_l_f1": 0.339712918660287, "generation_token_f1": 0.5071770334928231, "prompt_id": "482bbea97f7ae0fa", "reading_canonical_log_prob": -170.6418914794922, "reading_distances": [12, 10, 15, 12], "reading_log_probs": [-171.1019287109375, -173.06588745117188, -172.95831298828125, -172.6119384765625], "reading_logprob_std": 0.9092599211161813, "tokenization_path_logprob_std": 244.3251218730141, "trajectory_balance_rmse": 4160.988011378665} {"canonical_length": 11, "canonical_nll": 1.217668359929865, "generation_rouge_l_f1": 0.09424083769633508, "generation_token_f1": 0.09424083769633508, "prompt_id": "f9bf285092a25c3b", "reading_canonical_log_prob": -13.394351959228516, "reading_distances": [47, 46, 43, 42], "reading_log_probs": [-12.9588623046875, -14.812809944152832, -14.924650192260742, -13.585066795349121], "reading_logprob_std": 0.9577717669477288, "tokenization_path_logprob_std": 15.030744882181727, "trajectory_balance_rmse": 290.99049910400004} {"canonical_length": 150, "canonical_nll": 1.0797141520182292, "generation_rouge_l_f1": 0.3583535108958838, "generation_token_f1": 0.5230024213075061, "prompt_id": "9d4eb4020bd4213f", "reading_canonical_log_prob": -161.95712280273438, "reading_distances": [31, 18, 28, 24], "reading_log_probs": [-159.88829040527344, -161.02352905273438, -158.9796142578125, -159.57290649414062], "reading_logprob_std": 0.8586721975014512, "tokenization_path_logprob_std": 114.25394711747003, "trajectory_balance_rmse": 3833.2574765490444} {"canonical_length": 134, "canonical_nll": 1.2137029847102379, "generation_rouge_l_f1": 0.26111111111111107, "generation_token_f1": 0.4222222222222222, "prompt_id": "13778e3d12464b89", "reading_canonical_log_prob": -162.63619995117188, "reading_distances": [18, 19, 21, 19], "reading_log_probs": [-163.30718994140625, -162.9487762451172, -163.02682495117188, -162.8203125], "reading_logprob_std": 0.2060251754847707, "tokenization_path_logprob_std": 157.7715013430405, "trajectory_balance_rmse": 2071.37657634716} {"canonical_length": 395, "canonical_nll": 1.4721651874011077, "generation_rouge_l_f1": 0.2875399361022364, "generation_token_f1": 0.45047923322683703, "prompt_id": "513a1cb695ee9eb0", "reading_canonical_log_prob": -581.5052490234375, "reading_distances": [25, 24, 16, 27], "reading_log_probs": [-595.2223510742188, -589.349853515625, -585.8934326171875, -600.678955078125], "reading_logprob_std": 6.520458422614652, "tokenization_path_logprob_std": 474.92152947561107, "trajectory_balance_rmse": 7469.627914023336} {"canonical_length": 193, "canonical_nll": 1.2084866854810963, "generation_rouge_l_f1": 0.30222222222222217, "generation_token_f1": 0.48888888888888893, "prompt_id": "444facd57b3cdbe1", "reading_canonical_log_prob": -233.23793029785156, "reading_distances": [24, 22, 22, 14], "reading_log_probs": [-242.26480102539062, -235.75624084472656, -241.74288940429688, -239.37908935546875], "reading_logprob_std": 2.9652658985032274, "tokenization_path_logprob_std": 225.1807024539722, "trajectory_balance_rmse": 5369.604613776312} {"canonical_length": 226, "canonical_nll": 1.4023904716019082, "generation_rouge_l_f1": 0.24411134903640258, "generation_token_f1": 0.6381156316916489, "prompt_id": "5870502ae5fa3ae7", "reading_canonical_log_prob": -316.94024658203125, "reading_distances": [27, 20, 23, 23], "reading_log_probs": [-317.9397888183594, -316.82135009765625, -320.9659423828125, -319.7582702636719], "reading_logprob_std": 1.847904275876725, "tokenization_path_logprob_std": 306.5053391015807, "trajectory_balance_rmse": 3802.62063133383} {"canonical_length": 17, "canonical_nll": 1.117763407090131, "generation_rouge_l_f1": 0.7222222222222222, "generation_token_f1": 0.7777777777777778, "prompt_id": "8c61a7a56532124f", "reading_canonical_log_prob": -19.001977920532227, "reading_distances": [29, 40, 31, 29], "reading_log_probs": [-15.426322937011719, -15.516752243041992, -14.303860664367676, -13.724052429199219], "reading_logprob_std": 0.8749704799023567, "tokenization_path_logprob_std": 103.1049005247508, "trajectory_balance_rmse": 491.10923159187547} {"canonical_length": 12, "canonical_nll": 0.2763262192408244, "generation_rouge_l_f1": 0.8750000000000001, "generation_token_f1": 0.8750000000000001, "prompt_id": "fb0ad86c22711d8d", "reading_canonical_log_prob": -3.3159146308898926, "reading_distances": [9, 12, 14, 16], "reading_log_probs": [-3.750234842300415, -4.560441017150879, -4.817698955535889, -3.618429660797119], "reading_logprob_std": 0.5919670717438115, "tokenization_path_logprob_std": 56.44376643949447, "trajectory_balance_rmse": 241.26040676050263} {"canonical_length": 219, "canonical_nll": 1.5306409025845462, "generation_rouge_l_f1": 0.23481781376518218, "generation_token_f1": 0.4736842105263158, "prompt_id": "a18ee8ea214b4b09", "reading_canonical_log_prob": -335.2103576660156, "reading_distances": [36, 42, 52, 45], "reading_log_probs": [-331.807373046875, -331.10162353515625, -330.708740234375, -332.78643798828125], "reading_logprob_std": 0.9116573541067242, "tokenization_path_logprob_std": 275.28542212589355, "trajectory_balance_rmse": 6771.436664691401} {"canonical_length": 283, "canonical_nll": 1.3889175253285115, "generation_rouge_l_f1": 0.19030520646319568, "generation_token_f1": 0.36983842010771995, "prompt_id": "6933d659ab69515c", "reading_canonical_log_prob": -393.06365966796875, "reading_distances": [32, 27, 25, 23], "reading_log_probs": [-397.6386413574219, -396.186279296875, -394.2424621582031, -384.9481201171875], "reading_logprob_std": 5.70929183326722, "tokenization_path_logprob_std": 626.4358483497342, "trajectory_balance_rmse": 7275.180909945341} {"canonical_length": 49, "canonical_nll": 1.6414821391202965, "generation_rouge_l_f1": 0.27826086956521734, "generation_token_f1": 0.3043478260869565, "prompt_id": "e748f45af7778fcd", "reading_canonical_log_prob": -80.43262481689453, "reading_distances": [21, 31, 38, 33], "reading_log_probs": [-77.61343383789062, -80.8984375, -83.21208190917969, -76.65364074707031], "reading_logprob_std": 3.0199574012733548, "tokenization_path_logprob_std": 77.4220667481173, "trajectory_balance_rmse": 1166.1954140967778} {"canonical_length": 8, "canonical_nll": 1.0590777397155762, "generation_rouge_l_f1": 0.631578947368421, "generation_token_f1": 0.631578947368421, "prompt_id": "3f21812c287e05df", "reading_canonical_log_prob": -8.47262191772461, "reading_distances": [41, 27, 42, 32], "reading_log_probs": [-12.828817367553711, -9.916221618652344, -11.769142150878906, -10.317731857299805], "reading_logprob_std": 1.3420380276515116, "tokenization_path_logprob_std": 24.658404788113266, "trajectory_balance_rmse": 236.68712920070624} {"canonical_length": 38, "canonical_nll": 1.620858343024003, "generation_rouge_l_f1": 0.15286624203821655, "generation_token_f1": 0.20382165605095542, "prompt_id": "cc2b221229d6b7ba", "reading_canonical_log_prob": -61.59261703491211, "reading_distances": [27, 19, 21, 26], "reading_log_probs": [-66.78034973144531, -67.88853454589844, -62.91902160644531, -66.12014770507812], "reading_logprob_std": 2.1339371741971553, "tokenization_path_logprob_std": 247.20153306514288, "trajectory_balance_rmse": 1258.5320364270942} {"canonical_length": 57, "canonical_nll": 1.3924735889100193, "generation_rouge_l_f1": 0.240343347639485, "generation_token_f1": 0.33476394849785407, "prompt_id": "6c3587030700c60e", "reading_canonical_log_prob": -79.3709945678711, "reading_distances": [33, 25, 30, 23], "reading_log_probs": [-83.5706558227539, -82.34430694580078, -82.28876495361328, -80.82305145263672], "reading_logprob_std": 1.1240597848276104, "tokenization_path_logprob_std": 108.25787662400879, "trajectory_balance_rmse": 1494.7435165579652} {"canonical_length": 108, "canonical_nll": 1.8445616828070746, "generation_rouge_l_f1": 0.224, "generation_token_f1": 0.448, "prompt_id": "3f751ca73a73cf7d", "reading_canonical_log_prob": -199.21266174316406, "reading_distances": [26, 21, 13, 23], "reading_log_probs": [-205.64590454101562, -209.93785095214844, -203.3290252685547, -205.62802124023438], "reading_logprob_std": 2.758708105123164, "tokenization_path_logprob_std": 52.79167308334699, "trajectory_balance_rmse": 2440.169977448562} {"canonical_length": 77, "canonical_nll": 0.8275868304364093, "generation_rouge_l_f1": 0.1992619926199262, "generation_token_f1": 0.30996309963099633, "prompt_id": "7c7baba7466fe6f6", "reading_canonical_log_prob": -63.724185943603516, "reading_distances": [8, 7, 7, 11], "reading_log_probs": [-62.83380126953125, -62.559146881103516, -62.59383773803711, -63.99346923828125], "reading_logprob_std": 0.6767138577718088, "tokenization_path_logprob_std": 197.81985697000445, "trajectory_balance_rmse": 2107.505049851867} {"canonical_length": 294, "canonical_nll": 0.8465046266309258, "generation_rouge_l_f1": 0.2642857142857143, "generation_token_f1": 0.5, "prompt_id": "fd3d2e71f4e4243f", "reading_canonical_log_prob": -248.8723602294922, "reading_distances": [15, 19, 17, 20], "reading_log_probs": [-245.63658142089844, -245.31895446777344, -246.33517456054688, -247.44100952148438], "reading_logprob_std": 0.9400164854524153, "tokenization_path_logprob_std": 331.84893849025127, "trajectory_balance_rmse": 6647.817847314823} {"canonical_length": 399, "canonical_nll": 1.0240008896753603, "generation_rouge_l_f1": 0.2922374429223744, "generation_token_f1": 0.4414003044140031, "prompt_id": "5bc59064e72854cb", "reading_canonical_log_prob": -408.57635498046875, "reading_distances": [22, 20, 21, 24], "reading_log_probs": [-408.0718688964844, -415.67291259765625, -407.4168395996094, -407.702880859375], "reading_logprob_std": 3.980232905235116, "tokenization_path_logprob_std": 163.00523890592027, "trajectory_balance_rmse": 10589.92197969669} {"canonical_length": 280, "canonical_nll": 1.2529390607561384, "generation_rouge_l_f1": 0.23747680890538034, "generation_token_f1": 0.43784786641929496, "prompt_id": "cd25f5705dfa5cbb", "reading_canonical_log_prob": -350.82293701171875, "reading_distances": [22, 18, 25, 28], "reading_log_probs": [-359.8885192871094, -357.780517578125, -361.06085205078125, -359.2265930175781], "reading_logprob_std": 1.36846679201363, "tokenization_path_logprob_std": 246.88555675147614, "trajectory_balance_rmse": 8360.828374658686} {"canonical_length": 365, "canonical_nll": 0.8891844867026969, "generation_rouge_l_f1": 0.23240589198036007, "generation_token_f1": 0.45499181669394434, "prompt_id": "ee3200ffbd24e2bc", "reading_canonical_log_prob": -324.5523376464844, "reading_distances": [26, 28, 28, 31], "reading_log_probs": [-332.3160400390625, -324.2970886230469, -329.85174560546875, -331.5474853515625], "reading_logprob_std": 3.6201404827167805, "tokenization_path_logprob_std": 377.48600889901013, "trajectory_balance_rmse": 7117.801837718532} {"canonical_length": 115, "canonical_nll": 2.2610452403192935, "generation_rouge_l_f1": 0.2836879432624113, "generation_token_f1": 0.44680851063829785, "prompt_id": "4180e393ac001d5f", "reading_canonical_log_prob": -260.02020263671875, "reading_distances": [23, 25, 20, 17], "reading_log_probs": [-260.33013916015625, -267.19329833984375, -266.1255798339844, -264.42462158203125], "reading_logprob_std": 3.0159813992094224, "tokenization_path_logprob_std": 151.03441461806665, "trajectory_balance_rmse": 3446.608156305146} {"canonical_length": 9, "canonical_nll": 1.5495096842447917, "generation_rouge_l_f1": 0.6153846153846153, "generation_token_f1": 0.6153846153846153, "prompt_id": "79896777ff1e0459", "reading_canonical_log_prob": -13.945587158203125, "reading_distances": [36, 51, 52, 45], "reading_log_probs": [-9.658981323242188, -8.953950881958008, -7.7666015625, -8.337007522583008], "reading_logprob_std": 0.8135090803644943, "tokenization_path_logprob_std": 49.011305311935836, "trajectory_balance_rmse": 234.9154320088434} {"canonical_length": 147, "canonical_nll": 1.5011639238214818, "generation_rouge_l_f1": 0.18004866180048662, "generation_token_f1": 0.3309002433090024, "prompt_id": "4be1dda9c6193970", "reading_canonical_log_prob": -220.6710968017578, "reading_distances": [43, 40, 45, 43], "reading_log_probs": [-218.75704956054688, -217.95318603515625, -217.68121337890625, -218.67242431640625], "reading_logprob_std": 0.5310801448235118, "tokenization_path_logprob_std": 246.10064049075135, "trajectory_balance_rmse": 5470.16175805288} {"canonical_length": 9, "canonical_nll": 1.1537663141886394, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "e275be5afb765439", "reading_canonical_log_prob": -10.383896827697754, "reading_distances": [32, 31, 30, 24], "reading_log_probs": [-15.020917892456055, -17.707115173339844, -13.338099479675293, -20.602365493774414], "reading_logprob_std": 3.1812003007452896, "tokenization_path_logprob_std": 38.55995474956669, "trajectory_balance_rmse": 139.76710918411746} {"canonical_length": 8, "canonical_nll": 1.014936089515686, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "b4100d452af762d4", "reading_canonical_log_prob": -8.119488716125488, "reading_distances": [17, 18, 23, 23], "reading_log_probs": [-13.906920433044434, -14.265753746032715, -13.600224494934082, -15.175079345703125], "reading_logprob_std": 0.6819714561471673, "tokenization_path_logprob_std": 11.966461033433916, "trajectory_balance_rmse": 135.10793695783062} {"canonical_length": 65, "canonical_nll": 0.6729059659517729, "generation_rouge_l_f1": 0.30418250950570347, "generation_token_f1": 0.3650190114068442, "prompt_id": "660009a4d953661a", "reading_canonical_log_prob": -43.738887786865234, "reading_distances": [24, 31, 30, 34], "reading_log_probs": [-43.725440979003906, -45.886775970458984, -43.213096618652344, -47.39299011230469], "reading_logprob_std": 1.9423682767371435, "tokenization_path_logprob_std": 62.13320741392437, "trajectory_balance_rmse": 1236.7426051429286} {"canonical_length": 245, "canonical_nll": 0.9223530672034439, "generation_rouge_l_f1": 0.29203539823008845, "generation_token_f1": 0.5663716814159292, "prompt_id": "9c1c5be112145501", "reading_canonical_log_prob": -225.97650146484375, "reading_distances": [26, 30, 16, 16], "reading_log_probs": [-225.4754638671875, -227.0082550048828, -223.54061889648438, -228.34893798828125], "reading_logprob_std": 2.0674418731818696, "tokenization_path_logprob_std": 118.4314816575537, "trajectory_balance_rmse": 5429.200733264212} {"canonical_length": 41, "canonical_nll": 0.6855270106617998, "generation_rouge_l_f1": 0.6551724137931035, "generation_token_f1": 0.7241379310344828, "prompt_id": "8a1926c471e546ee", "reading_canonical_log_prob": -28.10660743713379, "reading_distances": [21, 15, 27, 24], "reading_log_probs": [-26.904766082763672, -25.9981746673584, -28.274873733520508, -26.94040298461914], "reading_logprob_std": 0.9377427675588695, "tokenization_path_logprob_std": 55.48527584499842, "trajectory_balance_rmse": 655.9379565276743} {"canonical_length": 311, "canonical_nll": 1.2340463718034063, "generation_rouge_l_f1": 0.19277108433734938, "generation_token_f1": 0.39242685025817553, "prompt_id": "a1e93e4a3a2191cd", "reading_canonical_log_prob": -383.7884216308594, "reading_distances": [37, 35, 36, 42], "reading_log_probs": [-387.18658447265625, -388.31671142578125, -384.97650146484375, -386.5469055175781], "reading_logprob_std": 1.3942210316718173, "tokenization_path_logprob_std": 413.10353685326817, "trajectory_balance_rmse": 8253.742495032608} {"canonical_length": 130, "canonical_nll": 1.7116971529447116, "generation_rouge_l_f1": 0.20855614973262032, "generation_token_f1": 0.4331550802139038, "prompt_id": "c4f0abd89b25e9b2", "reading_canonical_log_prob": -222.5206298828125, "reading_distances": [22, 21, 18, 21], "reading_log_probs": [-225.11485290527344, -224.47552490234375, -224.4210662841797, -222.00485229492188], "reading_logprob_std": 1.3695334632448177, "tokenization_path_logprob_std": 198.52680640319335, "trajectory_balance_rmse": 4329.09824506055} {"canonical_length": 226, "canonical_nll": 1.1851893062085177, "generation_rouge_l_f1": 0.25388601036269426, "generation_token_f1": 0.40932642487046633, "prompt_id": "8d709bdbfc4890e5", "reading_canonical_log_prob": -267.852783203125, "reading_distances": [23, 27, 30, 20], "reading_log_probs": [-259.416015625, -258.9437255859375, -259.7110595703125, -259.72528076171875], "reading_logprob_std": 0.365785218639889, "tokenization_path_logprob_std": 160.81928953466692, "trajectory_balance_rmse": 4908.444561751525} {"canonical_length": 410, "canonical_nll": 1.2483295719798018, "generation_rouge_l_f1": 0.24796084828711254, "generation_token_f1": 0.41761827079934744, "prompt_id": "fbcfdeca84611d04", "reading_canonical_log_prob": -511.81512451171875, "reading_distances": [30, 19, 29, 28], "reading_log_probs": [-514.8486328125, -511.6173095703125, -517.6077270507812, -511.03057861328125], "reading_logprob_std": 3.056684275956314, "tokenization_path_logprob_std": 183.43972536174473, "trajectory_balance_rmse": 9842.626222111698} {"canonical_length": 45, "canonical_nll": 1.380354224310981, "generation_rouge_l_f1": 0.27659574468085113, "generation_token_f1": 0.4468085106382979, "prompt_id": "8c6898d598b2732f", "reading_canonical_log_prob": -62.11594009399414, "reading_distances": [31, 31, 33, 26], "reading_log_probs": [-71.27528381347656, -70.3680191040039, -71.06938171386719, -69.03695678710938], "reading_logprob_std": 1.0111882568510342, "tokenization_path_logprob_std": 26.421225560393967, "trajectory_balance_rmse": 1317.7343832087552} {"canonical_length": 93, "canonical_nll": 0.6176952649188298, "generation_rouge_l_f1": 0.768361581920904, "generation_token_f1": 0.7796610169491526, "prompt_id": "f5259298cfe12698", "reading_canonical_log_prob": -57.44565963745117, "reading_distances": [19, 16, 17, 21], "reading_log_probs": [-54.22949981689453, -56.043907165527344, -56.24847412109375, -55.645416259765625], "reading_logprob_std": 0.9100124571664004, "tokenization_path_logprob_std": 84.19551499818847, "trajectory_balance_rmse": 1671.3271824163846} {"canonical_length": 183, "canonical_nll": 0.44577730939687926, "generation_rouge_l_f1": 0.3109540636042402, "generation_token_f1": 0.49469964664310956, "prompt_id": "af8c4385ce2b0480", "reading_canonical_log_prob": -81.5772476196289, "reading_distances": [31, 31, 37, 35], "reading_log_probs": [-87.94419860839844, -84.69611358642578, -91.71993255615234, -81.88906860351562], "reading_logprob_std": 4.23605794578954, "tokenization_path_logprob_std": 64.75689423033876, "trajectory_balance_rmse": 3466.106343805918} {"canonical_length": 164, "canonical_nll": 1.7946075811618711, "generation_rouge_l_f1": 0.17874396135265697, "generation_token_f1": 0.3719806763285024, "prompt_id": "8db3810d387bcbf4", "reading_canonical_log_prob": -294.3156433105469, "reading_distances": [24, 22, 19, 25], "reading_log_probs": [-306.9429626464844, -307.29254150390625, -298.807861328125, -301.8472900390625], "reading_logprob_std": 4.114472893023561, "tokenization_path_logprob_std": 445.00703376814886, "trajectory_balance_rmse": 5001.895159453326} {"canonical_length": 170, "canonical_nll": 0.8495649449965533, "generation_rouge_l_f1": 0.37864077669902907, "generation_token_f1": 0.5194174757281553, "prompt_id": "d92a9462a284aee7", "reading_canonical_log_prob": -144.42604064941406, "reading_distances": [12, 16, 13, 14], "reading_log_probs": [-142.4058837890625, -142.9735107421875, -144.0831298828125, -142.083984375], "reading_logprob_std": 0.8783539817113143, "tokenization_path_logprob_std": 113.43957083767411, "trajectory_balance_rmse": 4351.1142088228125} {"canonical_length": 76, "canonical_nll": 0.9606409574809828, "generation_rouge_l_f1": 0.3358778625954199, "generation_token_f1": 0.4656488549618321, "prompt_id": "32c20ba8cc10f798", "reading_canonical_log_prob": -73.00871276855469, "reading_distances": [33, 33, 28, 28], "reading_log_probs": [-71.97734069824219, -73.09771728515625, -73.6941909790039, -71.9288558959961], "reading_logprob_std": 0.8681203415486011, "tokenization_path_logprob_std": 158.96916249507086, "trajectory_balance_rmse": 2539.643889142992} {"canonical_length": 21, "canonical_nll": 0.7161559150332496, "generation_rouge_l_f1": 0.4137931034482759, "generation_token_f1": 0.4137931034482759, "prompt_id": "2e575375b1e838f0", "reading_canonical_log_prob": -15.039274215698242, "reading_distances": [25, 23, 27, 29], "reading_log_probs": [-18.010726928710938, -18.68186378479004, -17.080476760864258, -17.718870162963867], "reading_logprob_std": 0.6645997340504364, "tokenization_path_logprob_std": 88.79420229057601, "trajectory_balance_rmse": 273.1048417924477} {"canonical_length": 9, "canonical_nll": 0.0007758475840091705, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "f19b5cf2d17aa856", "reading_canonical_log_prob": -0.006982628256082535, "reading_distances": [37, 36, 30, 42], "reading_log_probs": [-0.013475785031914711, -0.015175092034041882, -0.013575544580817223, -0.013363425619900227], "reading_logprob_std": 0.0008561494513893191, "tokenization_path_logprob_std": 28.046238014979107, "trajectory_balance_rmse": 258.8016793277742} {"canonical_length": 3, "canonical_nll": 3.159466107686361, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "745badf44c39e597", "reading_canonical_log_prob": -9.478398323059082, "reading_distances": [24, 26, 19, 20], "reading_log_probs": [-10.83647346496582, -9.78775691986084, -9.659423828125, -9.274264335632324], "reading_logprob_std": 0.6679721460314499, "tokenization_path_logprob_std": 14.064776273975264, "trajectory_balance_rmse": 58.96273294326162} {"canonical_length": 272, "canonical_nll": 1.3726376926197725, "generation_rouge_l_f1": 0.2533589251439539, "generation_token_f1": 0.46833013435700577, "prompt_id": "8f529e5554e9c902", "reading_canonical_log_prob": -373.3574523925781, "reading_distances": [21, 22, 18, 19], "reading_log_probs": [-375.41326904296875, -375.598388671875, -383.9258728027344, -376.0125732421875], "reading_logprob_std": 4.13316573901425, "tokenization_path_logprob_std": 99.67681078508633, "trajectory_balance_rmse": 7437.668998096555} {"canonical_length": 324, "canonical_nll": 1.157350328233507, "generation_rouge_l_f1": 0.20326678765880218, "generation_token_f1": 0.4246823956442832, "prompt_id": "642f20171ea337c8", "reading_canonical_log_prob": -374.98150634765625, "reading_distances": [26, 21, 25, 27], "reading_log_probs": [-382.63262939453125, -381.2840270996094, -381.78143310546875, -378.50128173828125], "reading_logprob_std": 1.787963790682595, "tokenization_path_logprob_std": 376.2812920237357, "trajectory_balance_rmse": 7005.69282827551} {"canonical_length": 14, "canonical_nll": 2.1050352369035994, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4210526315789474, "prompt_id": "c076cc444fb03f81", "reading_canonical_log_prob": -29.47049331665039, "reading_distances": [196, 181, 192, 186], "reading_log_probs": [-34.09446716308594, -31.77227783203125, -36.05169677734375, -32.826595306396484], "reading_logprob_std": 1.8406769914294852, "tokenization_path_logprob_std": 63.29332235406951, "trajectory_balance_rmse": 512.704677648833} {"canonical_length": 183, "canonical_nll": 1.332242538368767, "generation_rouge_l_f1": 0.17297297297297298, "generation_token_f1": 0.2378378378378378, "prompt_id": "8082ddc02c38399e", "reading_canonical_log_prob": -243.80038452148438, "reading_distances": [18, 26, 25, 25], "reading_log_probs": [-243.95523071289062, -238.7117919921875, -241.250244140625, -241.11026000976562], "reading_logprob_std": 2.143214952155023, "tokenization_path_logprob_std": 204.6667297644672, "trajectory_balance_rmse": 2092.7948554844093} {"canonical_length": 7, "canonical_nll": 0.2984889234815325, "generation_rouge_l_f1": 0.05128205128205127, "generation_token_f1": 0.05128205128205127, "prompt_id": "3af881f54016c0d1", "reading_canonical_log_prob": -2.0894224643707275, "reading_distances": [35, 38, 36, 40], "reading_log_probs": [-3.7249372005462646, -3.117990493774414, -4.237984657287598, -3.447582483291626], "reading_logprob_std": 0.47401485906987606, "tokenization_path_logprob_std": 13.853980782113544, "trajectory_balance_rmse": 99.18522524686674} {"canonical_length": 241, "canonical_nll": 0.747593606656023, "generation_rouge_l_f1": 0.3176470588235294, "generation_token_f1": 0.5254901960784314, "prompt_id": "edd1c37dcb8ffb27", "reading_canonical_log_prob": -180.17005920410156, "reading_distances": [14, 14, 15, 16], "reading_log_probs": [-182.64512634277344, -182.80685424804688, -182.91339111328125, -178.98513793945312], "reading_logprob_std": 1.9048551051544371, "tokenization_path_logprob_std": 79.32100070940486, "trajectory_balance_rmse": 4930.972215465802} {"canonical_length": 277, "canonical_nll": 1.7209516724954874, "generation_rouge_l_f1": 0.21052631578947367, "generation_token_f1": 0.4548872180451128, "prompt_id": "b7a747a09567587c", "reading_canonical_log_prob": -476.70361328125, "reading_distances": [36, 29, 37, 27], "reading_log_probs": [-484.7396240234375, -487.14459228515625, -485.84722900390625, -485.3544921875], "reading_logprob_std": 1.0214011375400838, "tokenization_path_logprob_std": 539.9165217859742, "trajectory_balance_rmse": 7527.734486304022} {"canonical_length": 203, "canonical_nll": 1.5693948679956897, "generation_rouge_l_f1": 0.19870410367170624, "generation_token_f1": 0.5226781857451404, "prompt_id": "85ba85bd1384cec6", "reading_canonical_log_prob": -318.587158203125, "reading_distances": [19, 22, 16, 16], "reading_log_probs": [-313.8575744628906, -313.33819580078125, -312.97235107421875, -316.7799072265625], "reading_logprob_std": 1.7337365039055068, "tokenization_path_logprob_std": 319.5795293245978, "trajectory_balance_rmse": 6212.018862757357} {"canonical_length": 25, "canonical_nll": 0.7482810974121094, "generation_rouge_l_f1": 0.7586206896551724, "generation_token_f1": 0.7586206896551724, "prompt_id": "ff4cb1405a9ba9bb", "reading_canonical_log_prob": -18.707027435302734, "reading_distances": [16, 15, 16, 22], "reading_log_probs": [-20.231590270996094, -20.52956199645996, -19.166156768798828, -20.17005157470703], "reading_logprob_std": 0.5932712313530062, "tokenization_path_logprob_std": 110.5751081540135, "trajectory_balance_rmse": 483.9300531885732} {"canonical_length": 103, "canonical_nll": 1.5171533973471631, "generation_rouge_l_f1": 0.32631578947368417, "generation_token_f1": 0.5157894736842105, "prompt_id": "ea60f11b47ae285e", "reading_canonical_log_prob": -156.2667999267578, "reading_distances": [22, 29, 23, 29], "reading_log_probs": [-153.988037109375, -150.8957061767578, -151.93289184570312, -151.77078247070312], "reading_logprob_std": 1.3092664252002424, "tokenization_path_logprob_std": 281.5460984219437, "trajectory_balance_rmse": 3523.9034990605414} {"canonical_length": 22, "canonical_nll": 0.6218102195046165, "generation_rouge_l_f1": 0.30303030303030304, "generation_token_f1": 0.30303030303030304, "prompt_id": "6245bcac7beebc0c", "reading_canonical_log_prob": -13.679824829101562, "reading_distances": [25, 30, 21, 25], "reading_log_probs": [-12.119598388671875, -11.89146614074707, -12.121817588806152, -12.478416442871094], "reading_logprob_std": 0.24247591511299318, "tokenization_path_logprob_std": 98.93810968540406, "trajectory_balance_rmse": 667.1816367277943} {"canonical_length": 156, "canonical_nll": 1.4638052720289965, "generation_rouge_l_f1": 0.16705336426914152, "generation_token_f1": 0.3897911832946636, "prompt_id": "5016da9f4e182301", "reading_canonical_log_prob": -228.35362243652344, "reading_distances": [14, 16, 20, 18], "reading_log_probs": [-230.35458374023438, -235.29058837890625, -238.0770263671875, -238.26324462890625], "reading_logprob_std": 3.687623196312448, "tokenization_path_logprob_std": 269.0164987291737, "trajectory_balance_rmse": 5419.582741402382} {"canonical_length": 103, "canonical_nll": 1.5954012083775788, "generation_rouge_l_f1": 0.18716577540106952, "generation_token_f1": 0.3636363636363636, "prompt_id": "d6b153dd541c60bc", "reading_canonical_log_prob": -164.32632446289062, "reading_distances": [30, 33, 31, 34], "reading_log_probs": [-160.8553924560547, -161.39346313476562, -163.36932373046875, -164.61944580078125], "reading_logprob_std": 1.747647526807603, "tokenization_path_logprob_std": 209.96402603314468, "trajectory_balance_rmse": 2325.1431234841048} {"canonical_length": 50, "canonical_nll": 1.2273199462890625, "generation_rouge_l_f1": 0.16153846153846155, "generation_token_f1": 0.2307692307692308, "prompt_id": "f50b16ccb1ed30ed", "reading_canonical_log_prob": -61.365997314453125, "reading_distances": [14, 17, 13, 16], "reading_log_probs": [-73.5284194946289, -71.76050567626953, -65.37647247314453, -67.02255249023438], "reading_logprob_std": 3.8494558570382074, "tokenization_path_logprob_std": 73.30032058425884, "trajectory_balance_rmse": 1219.4548506524452} {"canonical_length": 26, "canonical_nll": 0.36870721670297474, "generation_rouge_l_f1": 0.7499999999999999, "generation_token_f1": 0.7499999999999999, "prompt_id": "ac38f5f5447a0903", "reading_canonical_log_prob": -9.586387634277344, "reading_distances": [35, 43, 36, 36], "reading_log_probs": [-10.310501098632812, -11.169370651245117, -11.167236328125, -11.520219802856445], "reading_logprob_std": 0.5150055437534186, "tokenization_path_logprob_std": 119.1963482612681, "trajectory_balance_rmse": 860.2743230762388} {"canonical_length": 10, "canonical_nll": 1.400993251800537, "generation_rouge_l_f1": 0.08759124087591241, "generation_token_f1": 0.08759124087591241, "prompt_id": "0027f65734731d27", "reading_canonical_log_prob": -14.009932518005371, "reading_distances": [25, 23, 18, 21], "reading_log_probs": [-22.064697265625, -15.831814765930176, -14.238792419433594, -18.686662673950195], "reading_logprob_std": 3.4396724068164426, "tokenization_path_logprob_std": 43.415968492849366, "trajectory_balance_rmse": 356.4599386356046} {"canonical_length": 146, "canonical_nll": 1.460101401969178, "generation_rouge_l_f1": 0.21836228287841192, "generation_token_f1": 0.38213399503722084, "prompt_id": "488c8f3398b00ac7", "reading_canonical_log_prob": -213.1748046875, "reading_distances": [15, 14, 19, 12], "reading_log_probs": [-211.3332977294922, -212.04888916015625, -210.59185791015625, -212.6575927734375], "reading_logprob_std": 0.893321617262818, "tokenization_path_logprob_std": 251.33524855945495, "trajectory_balance_rmse": 3137.1163088440057} {"canonical_length": 4, "canonical_nll": 6.671800136566162, "generation_rouge_l_f1": 0.6, "generation_token_f1": 0.6, "prompt_id": "91adc07e055d938f", "reading_canonical_log_prob": -26.68720054626465, "reading_distances": [31, 31, 34, 30], "reading_log_probs": [-26.65005111694336, -25.78838348388672, -28.488237380981445, -27.639644622802734], "reading_logprob_std": 1.1739244615585485, "tokenization_path_logprob_std": 8.116022678263924, "trajectory_balance_rmse": 47.705734175385544} {"canonical_length": 263, "canonical_nll": 1.2978940318292538, "generation_rouge_l_f1": 0.2601941747572815, "generation_token_f1": 0.4660194174757281, "prompt_id": "5f3070e7c5d93f6b", "reading_canonical_log_prob": -341.34613037109375, "reading_distances": [19, 21, 16, 20], "reading_log_probs": [-345.5799560546875, -345.65545654296875, -345.67144775390625, -346.78961181640625], "reading_logprob_std": 0.578373645836312, "tokenization_path_logprob_std": 316.5349293221071, "trajectory_balance_rmse": 8015.569165143439} {"canonical_length": 17, "canonical_nll": 0.08968475285698385, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.34782608695652173, "prompt_id": "f969cf0286bd838d", "reading_canonical_log_prob": -1.5246407985687256, "reading_distances": [18, 22, 21, 15], "reading_log_probs": [-2.1550066471099854, -6.995261192321777, -1.7085949182510376, -1.1275571584701538], "reading_logprob_std": 2.698755820533777, "tokenization_path_logprob_std": 64.56649238291061, "trajectory_balance_rmse": 322.17174397842336} {"canonical_length": 291, "canonical_nll": 0.6455400080205649, "generation_rouge_l_f1": 0.3160270880361174, "generation_token_f1": 0.4469525959367946, "prompt_id": "60292acfc5a172b6", "reading_canonical_log_prob": -187.85214233398438, "reading_distances": [25, 36, 33, 23], "reading_log_probs": [-192.31027221679688, -191.7617950439453, -189.946044921875, -191.14959716796875], "reading_logprob_std": 1.014800359446416, "tokenization_path_logprob_std": 301.30379569518544, "trajectory_balance_rmse": 6242.995892162952} {"canonical_length": 174, "canonical_nll": 0.8991301087127335, "generation_rouge_l_f1": 0.3671497584541063, "generation_token_f1": 0.5024154589371981, "prompt_id": "2b8f8cb7903f03af", "reading_canonical_log_prob": -156.44863891601562, "reading_distances": [37, 33, 41, 42], "reading_log_probs": [-151.11639404296875, -149.78314208984375, -151.91952514648438, -150.4932403564453], "reading_logprob_std": 0.9089169962195879, "tokenization_path_logprob_std": 417.49545311012747, "trajectory_balance_rmse": 4858.5855108346} {"canonical_length": 71, "canonical_nll": 1.2737654565085827, "generation_rouge_l_f1": 0.21052631578947367, "generation_token_f1": 0.3026315789473684, "prompt_id": "2a8ed34f73d4ba31", "reading_canonical_log_prob": -90.43734741210938, "reading_distances": [19, 14, 21, 19], "reading_log_probs": [-92.0851058959961, -92.03836059570312, -91.73933410644531, -92.0584716796875], "reading_logprob_std": 0.16179274877107297, "tokenization_path_logprob_std": 26.196236563751714, "trajectory_balance_rmse": 1537.5452420942327} {"canonical_length": 370, "canonical_nll": 0.7859743685335726, "generation_rouge_l_f1": 0.346875, "generation_token_f1": 0.5625000000000001, "prompt_id": "a2f2bd1ddbd7ac6c", "reading_canonical_log_prob": -290.8105163574219, "reading_distances": [22, 19, 21, 22], "reading_log_probs": [-297.7287292480469, -294.1496887207031, -297.552001953125, -296.00604248046875], "reading_logprob_std": 1.6638356448362708, "tokenization_path_logprob_std": 234.98762355215905, "trajectory_balance_rmse": 10350.098806180968} {"canonical_length": 292, "canonical_nll": 1.4867625824392658, "generation_rouge_l_f1": 0.196969696969697, "generation_token_f1": 0.3636363636363636, "prompt_id": "439bf3be7d027444", "reading_canonical_log_prob": -434.1346740722656, "reading_distances": [22, 23, 22, 22], "reading_log_probs": [-432.82293701171875, -432.69989013671875, -437.19683837890625, -432.3917541503906], "reading_logprob_std": 2.2865257139261943, "tokenization_path_logprob_std": 201.39505839104203, "trajectory_balance_rmse": 7386.579770924501} {"canonical_length": 152, "canonical_nll": 1.2172719051963405, "generation_rouge_l_f1": 0.3510638297872341, "generation_token_f1": 0.5053191489361702, "prompt_id": "53583cdc2f7841b7", "reading_canonical_log_prob": -185.02532958984375, "reading_distances": [23, 21, 25, 25], "reading_log_probs": [-186.0840301513672, -187.7272186279297, -185.0433349609375, -187.53012084960938], "reading_logprob_std": 1.2682164735268204, "tokenization_path_logprob_std": 152.39071124708784, "trajectory_balance_rmse": 3411.7806054230878} {"canonical_length": 73, "canonical_nll": 1.1998699658537564, "generation_rouge_l_f1": 0.20253164556962028, "generation_token_f1": 0.3481012658227848, "prompt_id": "11b8b10c9b5e1733", "reading_canonical_log_prob": -87.59050750732422, "reading_distances": [17, 16, 12, 20], "reading_log_probs": [-88.951171875, -89.76335906982422, -90.20159149169922, -88.58299255371094], "reading_logprob_std": 0.7395904587387172, "tokenization_path_logprob_std": 243.09866094779792, "trajectory_balance_rmse": 2377.276443775796} {"canonical_length": 17, "canonical_nll": 0.9950355080997243, "generation_rouge_l_f1": 0.10843373493975904, "generation_token_f1": 0.12048192771084336, "prompt_id": "d62e67048e61fa64", "reading_canonical_log_prob": -16.915603637695312, "reading_distances": [15, 9, 17, 13], "reading_log_probs": [-40.2542724609375, -26.97515106201172, -33.97856903076172, -26.027202606201172], "reading_logprob_std": 6.6539445247411235, "tokenization_path_logprob_std": 101.47038446382021, "trajectory_balance_rmse": 386.3591863114591} {"canonical_length": 17, "canonical_nll": 1.4157534206614775, "generation_rouge_l_f1": 0.1415929203539823, "generation_token_f1": 0.17699115044247787, "prompt_id": "15d2843edae95663", "reading_canonical_log_prob": -24.067808151245117, "reading_distances": [21, 22, 23, 26], "reading_log_probs": [-26.220245361328125, -29.588062286376953, -23.968338012695312, -27.938621520996094], "reading_logprob_std": 2.405396728549143, "tokenization_path_logprob_std": 44.587490186461324, "trajectory_balance_rmse": 446.68624423321916} {"canonical_length": 243, "canonical_nll": 1.4842718932854295, "generation_rouge_l_f1": 0.19918699186991873, "generation_token_f1": 0.4634146341463415, "prompt_id": "960951e022fd8e15", "reading_canonical_log_prob": -360.6780700683594, "reading_distances": [33, 34, 33, 30], "reading_log_probs": [-374.41632080078125, -375.6747741699219, -376.71063232421875, -376.4573669433594], "reading_logprob_std": 1.0313032419797092, "tokenization_path_logprob_std": 408.5358607459158, "trajectory_balance_rmse": 7968.51260575053} {"canonical_length": 192, "canonical_nll": 1.4298473993937175, "generation_rouge_l_f1": 0.2772727272727273, "generation_token_f1": 0.46818181818181814, "prompt_id": "e17ce1bfa59fa95a", "reading_canonical_log_prob": -274.53070068359375, "reading_distances": [18, 25, 15, 25], "reading_log_probs": [-273.4163513183594, -277.47479248046875, -271.3899841308594, -271.36346435546875], "reading_logprob_std": 2.8746794158750033, "tokenization_path_logprob_std": 154.63921536143638, "trajectory_balance_rmse": 5864.734313292394} {"canonical_length": 37, "canonical_nll": 1.0709673907305743, "generation_rouge_l_f1": 0.6376811594202898, "generation_token_f1": 0.6376811594202898, "prompt_id": "305cbe322d895404", "reading_canonical_log_prob": -39.62579345703125, "reading_distances": [13, 24, 14, 13], "reading_log_probs": [-39.200340270996094, -41.99188995361328, -40.38966751098633, -43.39148712158203], "reading_logprob_std": 1.832800016374761, "tokenization_path_logprob_std": 39.16694715284172, "trajectory_balance_rmse": 723.6176008046297} {"canonical_length": 25, "canonical_nll": 1.1304251098632812, "generation_rouge_l_f1": 0.14179104477611942, "generation_token_f1": 0.14925373134328357, "prompt_id": "fee5a1b64fc319ac", "reading_canonical_log_prob": -28.26062774658203, "reading_distances": [27, 26, 24, 24], "reading_log_probs": [-41.75897979736328, -33.087623596191406, -34.67704391479492, -31.89533233642578], "reading_logprob_std": 4.418931904776369, "tokenization_path_logprob_std": 78.09243050923763, "trajectory_balance_rmse": 711.2705908816781} {"canonical_length": 12, "canonical_nll": 0.9191360473632812, "generation_rouge_l_f1": 0.36363636363636365, "generation_token_f1": 0.36363636363636365, "prompt_id": "dce2143d59fcc65f", "reading_canonical_log_prob": -11.029632568359375, "reading_distances": [11, 11, 13, 20], "reading_log_probs": [-11.699007034301758, -11.762088775634766, -12.6060791015625, -12.008511543273926], "reading_logprob_std": 0.41359026823636086, "tokenization_path_logprob_std": 58.76261828400595, "trajectory_balance_rmse": 326.11116708843593} {"canonical_length": 30, "canonical_nll": 0.35786558787027994, "generation_rouge_l_f1": 0.14965986394557826, "generation_token_f1": 0.14965986394557826, "prompt_id": "bc27bf2b83eeb772", "reading_canonical_log_prob": -10.735967636108398, "reading_distances": [35, 42, 39, 37], "reading_log_probs": [-12.586774826049805, -12.83676528930664, -12.691485404968262, -14.168824195861816], "reading_logprob_std": 0.7390510144764796, "tokenization_path_logprob_std": 44.25883756665301, "trajectory_balance_rmse": 519.6648594930307} {"canonical_length": 171, "canonical_nll": 2.231220803065607, "generation_rouge_l_f1": 0.2572178477690289, "generation_token_f1": 0.42519685039370075, "prompt_id": "8e8e4001993842cc", "reading_canonical_log_prob": -381.53875732421875, "reading_distances": [23, 25, 30, 23], "reading_log_probs": [-375.3848876953125, -375.7803955078125, -374.8510437011719, -377.8553771972656], "reading_logprob_std": 1.3146608487525082, "tokenization_path_logprob_std": 212.66900556825294, "trajectory_balance_rmse": 3452.1480577531515} {"canonical_length": 93, "canonical_nll": 1.9746466195711525, "generation_rouge_l_f1": 0.24680851063829787, "generation_token_f1": 0.3829787234042554, "prompt_id": "ae4b947d6478d1ef", "reading_canonical_log_prob": -183.6421356201172, "reading_distances": [21, 24, 18, 27], "reading_log_probs": [-187.0338134765625, -191.2236785888672, -186.1963653564453, -188.54794311523438], "reading_logprob_std": 2.2081710794469873, "tokenization_path_logprob_std": 89.15279118722677, "trajectory_balance_rmse": 2665.8110679037436} {"canonical_length": 153, "canonical_nll": 2.1798319099775325, "generation_rouge_l_f1": 0.3218390804597701, "generation_token_f1": 0.5210727969348659, "prompt_id": "e8d78f3e32f3fc0a", "reading_canonical_log_prob": -333.5142822265625, "reading_distances": [18, 20, 23, 24], "reading_log_probs": [-341.987548828125, -342.363525390625, -343.1634521484375, -337.9218444824219], "reading_logprob_std": 2.3433758030800864, "tokenization_path_logprob_std": 141.50365612575695, "trajectory_balance_rmse": 5734.94861810131} {"canonical_length": 14, "canonical_nll": 2.286100115094866, "generation_rouge_l_f1": 0.5333333333333333, "generation_token_f1": 0.5333333333333333, "prompt_id": "0652d72e6ba6a111", "reading_canonical_log_prob": -32.005401611328125, "reading_distances": [35, 27, 25, 28], "reading_log_probs": [-44.58148193359375, -42.79730987548828, -41.49090576171875, -42.730125427246094], "reading_logprob_std": 1.2717875464174726, "tokenization_path_logprob_std": 29.854138986730025, "trajectory_balance_rmse": 264.3685324741564} {"canonical_length": 35, "canonical_nll": 1.7976063319614954, "generation_rouge_l_f1": 0.047619047619047616, "generation_token_f1": 0.047619047619047616, "prompt_id": "abb413900ce0e656", "reading_canonical_log_prob": -62.916221618652344, "reading_distances": [28, 24, 23, 27], "reading_log_probs": [-64.69404602050781, -65.94024658203125, -66.1701889038086, -65.21906280517578], "reading_logprob_std": 0.676096433117367, "tokenization_path_logprob_std": 129.97852259135027, "trajectory_balance_rmse": 1227.7163462876586} {"canonical_length": 294, "canonical_nll": 0.9484598587970344, "generation_rouge_l_f1": 0.19964973730297722, "generation_token_f1": 0.46935201401050786, "prompt_id": "7607920df7068b2c", "reading_canonical_log_prob": -278.8471984863281, "reading_distances": [14, 8, 13, 9], "reading_log_probs": [-280.0959167480469, -278.7209777832031, -279.5699462890625, -279.9255065917969], "reading_logprob_std": 0.6119778049949283, "tokenization_path_logprob_std": 264.5568415266993, "trajectory_balance_rmse": 7261.878147303779} {"canonical_length": 50, "canonical_nll": 2.8105282592773437, "generation_rouge_l_f1": 0.23188405797101447, "generation_token_f1": 0.3768115942028985, "prompt_id": "6e3d23fca9ec801f", "reading_canonical_log_prob": -140.5264129638672, "reading_distances": [28, 32, 27, 24], "reading_log_probs": [-142.9237823486328, -143.54000854492188, -143.649169921875, -142.31369018554688], "reading_logprob_std": 0.6176172867539336, "tokenization_path_logprob_std": 145.4900060231126, "trajectory_balance_rmse": 1759.4645532931197} {"canonical_length": 18, "canonical_nll": 2.185856501261393, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "4afc9f7951fa42e2", "reading_canonical_log_prob": -39.34541702270508, "reading_distances": [18, 20, 24, 21], "reading_log_probs": [-37.83903503417969, -37.715396881103516, -36.879913330078125, -37.253623962402344], "reading_logprob_std": 0.44053316071988585, "tokenization_path_logprob_std": 70.25864417105821, "trajectory_balance_rmse": 562.8519494813748} {"canonical_length": 501, "canonical_nll": 0.6510000019492266, "generation_rouge_l_f1": 0.1297297297297297, "generation_token_f1": 0.16756756756756758, "prompt_id": "031d7ff67f65087b", "reading_canonical_log_prob": -326.1510009765625, "reading_distances": [34, 27, 22, 24], "reading_log_probs": [-321.7852783203125, -322.61346435546875, -324.2449645996094, -322.1760559082031], "reading_logprob_std": 1.080977882688221, "tokenization_path_logprob_std": 160.43314466396325, "trajectory_balance_rmse": 7765.073020572214} {"canonical_length": 128, "canonical_nll": 1.1182044744491577, "generation_rouge_l_f1": 0.24501424501424499, "generation_token_f1": 0.39316239316239315, "prompt_id": "81ff4f6d7865cbd6", "reading_canonical_log_prob": -143.1301727294922, "reading_distances": [20, 19, 18, 19], "reading_log_probs": [-154.4864959716797, -162.48541259765625, -156.22901916503906, -162.19692993164062], "reading_logprob_std": 4.095846331651579, "tokenization_path_logprob_std": 125.54209796295274, "trajectory_balance_rmse": 3110.504997975257} {"canonical_length": 112, "canonical_nll": 1.4684231621878487, "generation_rouge_l_f1": 0.16842105263157894, "generation_token_f1": 0.26315789473684215, "prompt_id": "fa3e1b2e7d03f031", "reading_canonical_log_prob": -164.46339416503906, "reading_distances": [26, 29, 30, 28], "reading_log_probs": [-164.036376953125, -167.95236206054688, -165.84707641601562, -164.592529296875], "reading_logprob_std": 1.7372744375164537, "tokenization_path_logprob_std": 208.03467786066017, "trajectory_balance_rmse": 3758.902743493423} {"canonical_length": 136, "canonical_nll": 1.3187596938189339, "generation_rouge_l_f1": 0.22055137844611528, "generation_token_f1": 0.3959899749373434, "prompt_id": "c4484ce7220f53d7", "reading_canonical_log_prob": -179.351318359375, "reading_distances": [19, 15, 13, 15], "reading_log_probs": [-185.25335693359375, -185.43560791015625, -187.32415771484375, -188.43771362304688], "reading_logprob_std": 1.5351662411771096, "tokenization_path_logprob_std": 166.59892225849973, "trajectory_balance_rmse": 3757.9408330422566} {"canonical_length": 5, "canonical_nll": 0.009143863618373872, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "e265516b7f250f1e", "reading_canonical_log_prob": -0.045719318091869354, "reading_distances": [36, 36, 40, 41], "reading_log_probs": [-0.061348095536231995, -0.057616133242845535, -0.07010287046432495, -0.03680254891514778], "reading_logprob_std": 0.014115868139000226, "tokenization_path_logprob_std": 51.16652958607905, "trajectory_balance_rmse": 150.33017317299962} {"canonical_length": 20, "canonical_nll": 0.6334090232849121, "generation_rouge_l_f1": 0.32, "generation_token_f1": 0.32, "prompt_id": "d3f0dad3a37f19c3", "reading_canonical_log_prob": -12.668180465698242, "reading_distances": [16, 14, 16, 17], "reading_log_probs": [-11.310752868652344, -10.711729049682617, -11.406270027160645, -10.988821029663086], "reading_logprob_std": 0.3168890411083623, "tokenization_path_logprob_std": 72.36391390835801, "trajectory_balance_rmse": 683.7901198021127} {"canonical_length": 23, "canonical_nll": 1.6062205770741338, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.35714285714285715, "prompt_id": "448ac2a15956c368", "reading_canonical_log_prob": -36.94307327270508, "reading_distances": [29, 35, 41, 31], "reading_log_probs": [-40.710601806640625, -42.003807067871094, -41.8422737121582, -39.374794006347656], "reading_logprob_std": 1.2166793068975044, "tokenization_path_logprob_std": 40.80540985308814, "trajectory_balance_rmse": 648.8686707683696} {"canonical_length": 165, "canonical_nll": 1.4990105831261837, "generation_rouge_l_f1": 0.3505154639175258, "generation_token_f1": 0.4948453608247423, "prompt_id": "e00cf91341eb5691", "reading_canonical_log_prob": -247.3367462158203, "reading_distances": [14, 14, 9, 17], "reading_log_probs": [-249.78744506835938, -249.53436279296875, -248.17660522460938, -248.68359375], "reading_logprob_std": 0.7473115323813555, "tokenization_path_logprob_std": 108.62268101207235, "trajectory_balance_rmse": 5362.027560481841} {"canonical_length": 224, "canonical_nll": 1.207282611301967, "generation_rouge_l_f1": 0.29769392033542974, "generation_token_f1": 0.6037735849056604, "prompt_id": "bf8645b4ff7492c2", "reading_canonical_log_prob": -270.4313049316406, "reading_distances": [52, 55, 60, 48], "reading_log_probs": [-267.99835205078125, -269.08251953125, -270.35760498046875, -267.77679443359375], "reading_logprob_std": 1.182579289166934, "tokenization_path_logprob_std": 270.69610387209786, "trajectory_balance_rmse": 6414.344085624697} {"canonical_length": 261, "canonical_nll": 1.4313598866663673, "generation_rouge_l_f1": 0.2037037037037037, "generation_token_f1": 0.39629629629629626, "prompt_id": "93af13c34520a089", "reading_canonical_log_prob": -373.5849304199219, "reading_distances": [25, 28, 26, 22], "reading_log_probs": [-368.0283203125, -368.5145263671875, -364.4793701171875, -365.6822204589844], "reading_logprob_std": 1.9167446361693286, "tokenization_path_logprob_std": 391.44757130091097, "trajectory_balance_rmse": 7869.424376922168} {"canonical_length": 107, "canonical_nll": 1.3012314555800963, "generation_rouge_l_f1": 0.2598870056497175, "generation_token_f1": 0.4067796610169492, "prompt_id": "8f7942ab9e010ef0", "reading_canonical_log_prob": -139.2317657470703, "reading_distances": [34, 31, 34, 35], "reading_log_probs": [-145.36416625976562, -145.84103393554688, -147.46847534179688, -145.39987182617188], "reading_logprob_std": 0.9907532665762404, "tokenization_path_logprob_std": 152.14446424265856, "trajectory_balance_rmse": 2586.2767123789426} {"canonical_length": 39, "canonical_nll": 2.7145499204977965, "generation_rouge_l_f1": 0.2807017543859649, "generation_token_f1": 0.3157894736842105, "prompt_id": "ba9caab6c5bf7b20", "reading_canonical_log_prob": -105.86744689941406, "reading_distances": [38, 37, 40, 36], "reading_log_probs": [-124.32608032226562, -124.429443359375, -126.59393310546875, -127.96251678466797], "reading_logprob_std": 1.7658368008123597, "tokenization_path_logprob_std": 36.48504793777236, "trajectory_balance_rmse": 922.8069348225082} {"canonical_length": 292, "canonical_nll": 1.4696886245518515, "generation_rouge_l_f1": 0.22535211267605634, "generation_token_f1": 0.4014084507042254, "prompt_id": "90e4d685d3873c08", "reading_canonical_log_prob": -429.1490783691406, "reading_distances": [30, 30, 22, 20], "reading_log_probs": [-432.7176513671875, -432.13134765625, -432.1130676269531, -434.2511291503906], "reading_logprob_std": 1.0052339061286715, "tokenization_path_logprob_std": 405.6028475177215, "trajectory_balance_rmse": 9191.441173171483} {"canonical_length": 15, "canonical_nll": 2.5888031005859373, "generation_rouge_l_f1": 0.0849056603773585, "generation_token_f1": 0.09433962264150943, "prompt_id": "ec455005996bdab4", "reading_canonical_log_prob": -38.83204650878906, "reading_distances": [42, 34, 48, 44], "reading_log_probs": [-39.96156311035156, -39.00547790527344, -41.40170669555664, -40.43851089477539], "reading_logprob_std": 0.9974481510947931, "tokenization_path_logprob_std": 99.85510711085739, "trajectory_balance_rmse": 381.1709259354619} {"canonical_length": 243, "canonical_nll": 1.760327751253858, "generation_rouge_l_f1": 0.20982142857142858, "generation_token_f1": 0.40625, "prompt_id": "ab90737e4575d4fd", "reading_canonical_log_prob": -427.7596435546875, "reading_distances": [11, 17, 14, 17], "reading_log_probs": [-432.9112854003906, -439.1810607910156, -435.76727294921875, -438.18768310546875], "reading_logprob_std": 2.7959234170711627, "tokenization_path_logprob_std": 265.763117257904, "trajectory_balance_rmse": 6576.786186674119} {"canonical_length": 240, "canonical_nll": 1.4299448649088542, "generation_rouge_l_f1": 0.26452905811623245, "generation_token_f1": 0.5811623246492986, "prompt_id": "ec5a98de132967a0", "reading_canonical_log_prob": -343.186767578125, "reading_distances": [33, 32, 29, 37], "reading_log_probs": [-350.1507568359375, -342.24700927734375, -344.6874084472656, -346.34185791015625], "reading_logprob_std": 3.320211446231734, "tokenization_path_logprob_std": 382.5764297298226, "trajectory_balance_rmse": 7143.642593506251} {"canonical_length": 37, "canonical_nll": 1.6758966188173037, "generation_rouge_l_f1": 0.21621621621621623, "generation_token_f1": 0.21621621621621623, "prompt_id": "71251a192c44eeb3", "reading_canonical_log_prob": -62.008174896240234, "reading_distances": [22, 24, 25, 22], "reading_log_probs": [-66.07292938232422, -73.4636459350586, -70.05944061279297, -67.47500610351562], "reading_logprob_std": 3.2482367617736143, "tokenization_path_logprob_std": 157.85091433457305, "trajectory_balance_rmse": 980.0689863282571} {"canonical_length": 221, "canonical_nll": 1.897069042084983, "generation_rouge_l_f1": 0.31533477321814257, "generation_token_f1": 0.5226781857451404, "prompt_id": "646766dc50567beb", "reading_canonical_log_prob": -419.25225830078125, "reading_distances": [44, 41, 39, 54], "reading_log_probs": [-416.15887451171875, -408.48388671875, -413.98309326171875, -415.3769226074219], "reading_logprob_std": 3.4634867674594, "tokenization_path_logprob_std": 313.9599017878128, "trajectory_balance_rmse": 5166.638676038381} {"canonical_length": 87, "canonical_nll": 1.2643361584893589, "generation_rouge_l_f1": 0.2492917847025496, "generation_token_f1": 0.3796033994334277, "prompt_id": "3175f6135d7796a6", "reading_canonical_log_prob": -109.99724578857422, "reading_distances": [40, 40, 42, 47], "reading_log_probs": [-110.7519760131836, -109.97686767578125, -106.97050476074219, -109.51832580566406], "reading_logprob_std": 1.637123526768417, "tokenization_path_logprob_std": 80.11642219764433, "trajectory_balance_rmse": 2070.6736474723743} {"canonical_length": 315, "canonical_nll": 1.033370341951885, "generation_rouge_l_f1": 0.22807017543859648, "generation_token_f1": 0.4701754385964912, "prompt_id": "364270b26598f5f6", "reading_canonical_log_prob": -325.51165771484375, "reading_distances": [25, 17, 25, 26], "reading_log_probs": [-330.93890380859375, -327.78826904296875, -327.5458068847656, -328.3182373046875], "reading_logprob_std": 1.561084081009182, "tokenization_path_logprob_std": 337.1512699317455, "trajectory_balance_rmse": 7539.24862489823} {"canonical_length": 132, "canonical_nll": 1.872573505748402, "generation_rouge_l_f1": 0.16541353383458648, "generation_token_f1": 0.3558897243107769, "prompt_id": "3b49b131c410b62c", "reading_canonical_log_prob": -247.17970275878906, "reading_distances": [18, 13, 10, 23], "reading_log_probs": [-246.47161865234375, -249.34078979492188, -240.83731079101562, -244.79803466796875], "reading_logprob_std": 3.5521303985212334, "tokenization_path_logprob_std": 295.4482137399308, "trajectory_balance_rmse": 3617.1676230714643} {"canonical_length": 9, "canonical_nll": 0.2249838246239556, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.8571428571428571, "prompt_id": "300deaab60201ff0", "reading_canonical_log_prob": -2.0248544216156006, "reading_distances": [31, 19, 23, 20], "reading_log_probs": [-2.173396587371826, -2.28267765045166, -2.345137596130371, -2.7330358028411865], "reading_logprob_std": 0.24355391349346459, "tokenization_path_logprob_std": 80.04935839913395, "trajectory_balance_rmse": 228.0468261797262} {"canonical_length": 239, "canonical_nll": 1.5338852375621077, "generation_rouge_l_f1": 0.21951219512195125, "generation_token_f1": 0.5528455284552846, "prompt_id": "166e7bed707f11c1", "reading_canonical_log_prob": -366.59857177734375, "reading_distances": [21, 26, 23, 26], "reading_log_probs": [-361.022705078125, -356.36102294921875, -363.5630187988281, -362.739013671875], "reading_logprob_std": 3.2191702293300866, "tokenization_path_logprob_std": 341.02560404785623, "trajectory_balance_rmse": 7523.1370952264415} {"canonical_length": 278, "canonical_nll": 1.3314543799530687, "generation_rouge_l_f1": 0.22889305816135083, "generation_token_f1": 0.46529080675422146, "prompt_id": "af085ada31a636ce", "reading_canonical_log_prob": -370.1443176269531, "reading_distances": [29, 31, 32, 32], "reading_log_probs": [-370.16632080078125, -372.62066650390625, -370.91864013671875, -373.73919677734375], "reading_logprob_std": 1.6191248522927044, "tokenization_path_logprob_std": 86.72937637427148, "trajectory_balance_rmse": 7490.474648876664} {"canonical_length": 157, "canonical_nll": 1.658355712890625, "generation_rouge_l_f1": 0.21002386634844866, "generation_token_f1": 0.4391408114558472, "prompt_id": "f0f7445e4deb0e6e", "reading_canonical_log_prob": -260.3618469238281, "reading_distances": [28, 29, 13, 35], "reading_log_probs": [-254.3138885498047, -260.4070129394531, -256.27685546875, -263.4751892089844], "reading_logprob_std": 4.1149802420546795, "tokenization_path_logprob_std": 219.98929282434955, "trajectory_balance_rmse": 5586.355211036047} {"canonical_length": 20, "canonical_nll": 0.9656076431274414, "generation_rouge_l_f1": 0.21487603305785125, "generation_token_f1": 0.2809917355371901, "prompt_id": "b5d6efcaf5d67694", "reading_canonical_log_prob": -19.312152862548828, "reading_distances": [41, 35, 40, 28], "reading_log_probs": [-23.243921279907227, -24.00221061706543, -25.189659118652344, -23.32708740234375], "reading_logprob_std": 0.8992039885950683, "tokenization_path_logprob_std": 85.19830037014162, "trajectory_balance_rmse": 604.102028689628} {"canonical_length": 91, "canonical_nll": 2.271465594951923, "generation_rouge_l_f1": 0.27108433734939763, "generation_token_f1": 0.3855421686746988, "prompt_id": "b7652a39606177d7", "reading_canonical_log_prob": -206.703369140625, "reading_distances": [24, 13, 22, 24], "reading_log_probs": [-216.22520446777344, -213.27078247070312, -210.01251220703125, -210.14251708984375], "reading_logprob_std": 2.9544389043557646, "tokenization_path_logprob_std": 161.83405508735484, "trajectory_balance_rmse": 2576.4723060350516} {"canonical_length": 18, "canonical_nll": 1.1029116312662761, "generation_rouge_l_f1": 0.0956175298804781, "generation_token_f1": 0.0956175298804781, "prompt_id": "f3723ab195d4cd31", "reading_canonical_log_prob": -19.85240936279297, "reading_distances": [25, 26, 22, 20], "reading_log_probs": [-18.614215850830078, -17.19088363647461, -18.774927139282227, -19.033161163330078], "reading_logprob_std": 0.8264927405127539, "tokenization_path_logprob_std": 87.19176264957457, "trajectory_balance_rmse": 536.9257062183406} {"canonical_length": 7, "canonical_nll": 0.9536158697945731, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "ff3fea8e176d412d", "reading_canonical_log_prob": -6.675311088562012, "reading_distances": [15, 23, 26, 19], "reading_log_probs": [-12.062801361083984, -11.324942588806152, -13.87537956237793, -15.146733283996582], "reading_logprob_std": 1.733677799945164, "tokenization_path_logprob_std": 41.192356703542124, "trajectory_balance_rmse": 174.93874251401417} {"canonical_length": 69, "canonical_nll": 1.2487574038298235, "generation_rouge_l_f1": 0.1257861635220126, "generation_token_f1": 0.18238993710691823, "prompt_id": "f5b5083636173713", "reading_canonical_log_prob": -86.16426086425781, "reading_distances": [16, 25, 19, 24], "reading_log_probs": [-89.03072357177734, -96.37937927246094, -99.82803344726562, -93.84342956542969], "reading_logprob_std": 4.545020354669482, "tokenization_path_logprob_std": 176.08326164364772, "trajectory_balance_rmse": 1429.9325046843965} {"canonical_length": 291, "canonical_nll": 0.9819498488173861, "generation_rouge_l_f1": 0.2730627306273063, "generation_token_f1": 0.5202952029520295, "prompt_id": "640dfdad1fff517f", "reading_canonical_log_prob": -285.7474060058594, "reading_distances": [22, 25, 24, 22], "reading_log_probs": [-289.2578125, -287.62640380859375, -287.37884521484375, -285.12579345703125], "reading_logprob_std": 1.6994159133754037, "tokenization_path_logprob_std": 539.7114312222352, "trajectory_balance_rmse": 8586.021166820128} {"canonical_length": 17, "canonical_nll": 2.0651319167193245, "generation_rouge_l_f1": 0.13333333333333333, "generation_token_f1": 0.13333333333333333, "prompt_id": "f39da6ed7768e21d", "reading_canonical_log_prob": -35.107242584228516, "reading_distances": [34, 32, 27, 33], "reading_log_probs": [-32.59518814086914, -34.55430221557617, -33.1286506652832, -33.286651611328125], "reading_logprob_std": 0.8299205107470067, "tokenization_path_logprob_std": 47.74666730880567, "trajectory_balance_rmse": 512.2552207929} {"canonical_length": 16, "canonical_nll": 2.4031333923339844, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.2962962962962963, "prompt_id": "968e82be4a52a789", "reading_canonical_log_prob": -38.45013427734375, "reading_distances": [34, 29, 30, 27], "reading_log_probs": [-38.93620681762695, -39.011478424072266, -38.308082580566406, -36.51071548461914], "reading_logprob_std": 1.1641278404887938, "tokenization_path_logprob_std": 60.97427587646133, "trajectory_balance_rmse": 543.8458051338357} {"canonical_length": 15, "canonical_nll": 2.1705474853515625, "generation_rouge_l_f1": 0.19999999999999998, "generation_token_f1": 0.3, "prompt_id": "e36a9df5c4b484e7", "reading_canonical_log_prob": -32.55821228027344, "reading_distances": [23, 25, 23, 26], "reading_log_probs": [-34.633323669433594, -30.42106056213379, -32.29900360107422, -34.434627532958984], "reading_logprob_std": 1.9880469957254732, "tokenization_path_logprob_std": 80.12715346546445, "trajectory_balance_rmse": 380.83056456098313} {"canonical_length": 243, "canonical_nll": 1.2307338636107896, "generation_rouge_l_f1": 0.25984251968503935, "generation_token_f1": 0.5118110236220472, "prompt_id": "87e83988cc685281", "reading_canonical_log_prob": -299.0683288574219, "reading_distances": [28, 33, 37, 39], "reading_log_probs": [-306.2471923828125, -305.32147216796875, -303.9765319824219, -305.07916259765625], "reading_logprob_std": 0.9336542210271002, "tokenization_path_logprob_std": 296.9229840544288, "trajectory_balance_rmse": 7126.457676807599} {"canonical_length": 106, "canonical_nll": 0.8775906832713001, "generation_rouge_l_f1": 0.23569023569023567, "generation_token_f1": 0.39057239057239057, "prompt_id": "753de9bc3063e720", "reading_canonical_log_prob": -93.02461242675781, "reading_distances": [21, 16, 16, 21], "reading_log_probs": [-92.63435363769531, -95.0679931640625, -91.48457336425781, -91.16880798339844], "reading_logprob_std": 1.7686310713091766, "tokenization_path_logprob_std": 103.80830643134131, "trajectory_balance_rmse": 3282.6187626161177} {"canonical_length": 18, "canonical_nll": 2.0237918429904513, "generation_rouge_l_f1": 0.48, "generation_token_f1": 0.48, "prompt_id": "52785eaaf4af8786", "reading_canonical_log_prob": -36.428253173828125, "reading_distances": [20, 24, 24, 27], "reading_log_probs": [-34.44669723510742, -27.53115463256836, -28.438777923583984, -30.1376895904541], "reading_logprob_std": 3.0685430337524955, "tokenization_path_logprob_std": 80.54366522660091, "trajectory_balance_rmse": 530.8648324239308} {"canonical_length": 293, "canonical_nll": 1.8087741474242747, "generation_rouge_l_f1": 0.2821869488536155, "generation_token_f1": 0.4691358024691358, "prompt_id": "9296c91d93887868", "reading_canonical_log_prob": -529.9708251953125, "reading_distances": [19, 19, 23, 17], "reading_log_probs": [-526.24365234375, -528.140625, -525.8818359375, -524.735595703125], "reading_logprob_std": 1.4146221644568722, "tokenization_path_logprob_std": 400.6558073516591, "trajectory_balance_rmse": 8058.899059722957} {"canonical_length": 130, "canonical_nll": 1.9062004676231972, "generation_rouge_l_f1": 0.26262626262626265, "generation_token_f1": 0.4292929292929292, "prompt_id": "dd7d0916ee5ef6b2", "reading_canonical_log_prob": -247.80606079101562, "reading_distances": [54, 60, 62, 57], "reading_log_probs": [-241.70326232910156, -241.59088134765625, -240.56195068359375, -241.15562438964844], "reading_logprob_std": 0.5176649873948278, "tokenization_path_logprob_std": 202.2773871137198, "trajectory_balance_rmse": 3541.2372617227725} {"canonical_length": 172, "canonical_nll": 1.9977771847747092, "generation_rouge_l_f1": 0.19523809523809524, "generation_token_f1": 0.42857142857142855, "prompt_id": "0d595f7cde9d737e", "reading_canonical_log_prob": -343.61767578125, "reading_distances": [17, 18, 19, 20], "reading_log_probs": [-343.8453674316406, -345.3921203613281, -342.645263671875, -344.46356201171875], "reading_logprob_std": 1.1521179530267878, "tokenization_path_logprob_std": 368.8593545468694, "trajectory_balance_rmse": 4806.434508778393} {"canonical_length": 154, "canonical_nll": 1.0484985747894684, "generation_rouge_l_f1": 0.19431279620853079, "generation_token_f1": 0.45023696682464454, "prompt_id": "8c2391cc4e076841", "reading_canonical_log_prob": -161.46878051757812, "reading_distances": [233, 246, 247, 230], "reading_log_probs": [-140.78897094726562, -146.3244171142578, -143.29171752929688, -152.40087890625], "reading_logprob_std": 5.006985011320019, "tokenization_path_logprob_std": 147.00579154481466, "trajectory_balance_rmse": 3557.8283250368754} {"canonical_length": 395, "canonical_nll": 0.8933286256428006, "generation_rouge_l_f1": 0.23297491039426524, "generation_token_f1": 0.39784946236559143, "prompt_id": "a3d283a16147427b", "reading_canonical_log_prob": -352.86480712890625, "reading_distances": [14, 17, 17, 14], "reading_log_probs": [-353.564208984375, -351.60467529296875, -353.5983581542969, -352.1296691894531], "reading_logprob_std": 1.0126812634095934, "tokenization_path_logprob_std": 259.800969008893, "trajectory_balance_rmse": 7732.629678335078} {"canonical_length": 11, "canonical_nll": 1.4663274938409978, "generation_rouge_l_f1": 0.64, "generation_token_f1": 0.64, "prompt_id": "977c41664864ae25", "reading_canonical_log_prob": -16.129602432250977, "reading_distances": [30, 32, 35, 27], "reading_log_probs": [-16.3009090423584, -16.501251220703125, -15.15776252746582, -15.037166595458984], "reading_logprob_std": 0.7586729067138458, "tokenization_path_logprob_std": 42.21613805506221, "trajectory_balance_rmse": 355.7580590657867} {"canonical_length": 20, "canonical_nll": 0.9060344696044922, "generation_rouge_l_f1": 0.11458333333333334, "generation_token_f1": 0.14583333333333337, "prompt_id": "4c8a56abb260b4fb", "reading_canonical_log_prob": -18.120689392089844, "reading_distances": [28, 29, 25, 28], "reading_log_probs": [-32.0014533996582, -19.581485748291016, -19.54166030883789, -20.642545700073242], "reading_logprob_std": 6.061258171000821, "tokenization_path_logprob_std": 58.317268235727354, "trajectory_balance_rmse": 286.1831769640429} {"canonical_length": 244, "canonical_nll": 0.7720198709456647, "generation_rouge_l_f1": 0.38610038610038616, "generation_token_f1": 0.5714285714285715, "prompt_id": "7a98c8dec132eec9", "reading_canonical_log_prob": -188.3728485107422, "reading_distances": [32, 35, 31, 33], "reading_log_probs": [-201.37274169921875, -201.68243408203125, -204.0130157470703, -202.4130859375], "reading_logprob_std": 1.1787886077354803, "tokenization_path_logprob_std": 286.3358375173268, "trajectory_balance_rmse": 6595.450008886129} {"canonical_length": 419, "canonical_nll": 1.290238919861277, "generation_rouge_l_f1": 0.21726190476190477, "generation_token_f1": 0.39285714285714285, "prompt_id": "7f79088a335cbb01", "reading_canonical_log_prob": -540.610107421875, "reading_distances": [23, 13, 20, 16], "reading_log_probs": [-523.4749145507812, -535.4227294921875, -528.42333984375, -530.9262084960938], "reading_logprob_std": 4.9852573004767855, "tokenization_path_logprob_std": 117.03404778437019, "trajectory_balance_rmse": 10153.723277411566} {"canonical_length": 12, "canonical_nll": 3.5876073837280273, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.5714285714285715, "prompt_id": "48b64f1a7aabf6e4", "reading_canonical_log_prob": -43.05128860473633, "reading_distances": [30, 23, 21, 27], "reading_log_probs": [-41.56292724609375, -45.85846710205078, -44.00601577758789, -41.677181243896484], "reading_logprob_std": 2.0569312868378016, "tokenization_path_logprob_std": 69.56209075967968, "trajectory_balance_rmse": 477.5202736725375} {"canonical_length": 319, "canonical_nll": 1.5714430853864616, "generation_rouge_l_f1": 0.21575342465753425, "generation_token_f1": 0.4178082191780822, "prompt_id": "735c6c6b04d8c3fe", "reading_canonical_log_prob": -501.29034423828125, "reading_distances": [35, 38, 37, 39], "reading_log_probs": [-500.94305419921875, -502.4707336425781, -505.8501892089844, -503.84197998046875], "reading_logprob_std": 2.0846960194257376, "tokenization_path_logprob_std": 315.37424913259787, "trajectory_balance_rmse": 8873.158872037075} {"canonical_length": 57, "canonical_nll": 1.2867056863349782, "generation_rouge_l_f1": 0.2299651567944251, "generation_token_f1": 0.28571428571428575, "prompt_id": "eab393eed1fb7651", "reading_canonical_log_prob": -73.34222412109375, "reading_distances": [35, 35, 37, 35], "reading_log_probs": [-72.84507751464844, -73.94119262695312, -72.63619995117188, -72.59315490722656], "reading_logprob_std": 0.634470095797365, "tokenization_path_logprob_std": 124.21842407469207, "trajectory_balance_rmse": 1713.7408541988332} {"canonical_length": 20, "canonical_nll": 0.0007693677209317685, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "691dfdc26548dcb9", "reading_canonical_log_prob": -0.015387354418635368, "reading_distances": [30, 31, 33, 31], "reading_log_probs": [-0.018508046865463257, -0.024027444422245026, -0.018566513434052467, -0.03085702657699585], "reading_logprob_std": 0.005848690313711633, "tokenization_path_logprob_std": 48.62883357092704, "trajectory_balance_rmse": 420.29742469093077} {"canonical_length": 230, "canonical_nll": 1.9071830417798914, "generation_rouge_l_f1": 0.2601279317697228, "generation_token_f1": 0.4818763326226013, "prompt_id": "2b073d802d00719d", "reading_canonical_log_prob": -438.652099609375, "reading_distances": [24, 28, 25, 21], "reading_log_probs": [-437.14080810546875, -435.85479736328125, -435.057861328125, -431.32391357421875], "reading_logprob_std": 2.498918660757355, "tokenization_path_logprob_std": 57.02897503552553, "trajectory_balance_rmse": 5328.322983060905} {"canonical_length": 11, "canonical_nll": 6.728311018510298, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "11a956113e5f3fb8", "reading_canonical_log_prob": -74.01142120361328, "reading_distances": [29, 30, 27, 26], "reading_log_probs": [-78.10589599609375, -78.17178344726562, -72.44727325439453, -73.41435241699219], "reading_logprob_std": 3.0327839617878434, "tokenization_path_logprob_std": 0.0, "trajectory_balance_rmse": 74.18000793457031} {"canonical_length": 162, "canonical_nll": 1.3327415372118538, "generation_rouge_l_f1": 0.3046683046683047, "generation_token_f1": 0.37837837837837834, "prompt_id": "b48c6b78ff2012e6", "reading_canonical_log_prob": -215.9041290283203, "reading_distances": [30, 20, 32, 22], "reading_log_probs": [-211.57009887695312, -211.40174865722656, -213.69664001464844, -213.60009765625], "reading_logprob_std": 1.2509995717285889, "tokenization_path_logprob_std": 264.95485261662554, "trajectory_balance_rmse": 3964.2678436172177} {"canonical_length": 142, "canonical_nll": 0.9518344503053477, "generation_rouge_l_f1": 0.2751322751322751, "generation_token_f1": 0.4814814814814815, "prompt_id": "c057418a3f03722b", "reading_canonical_log_prob": -135.16049194335938, "reading_distances": [34, 36, 34, 30], "reading_log_probs": [-142.46469116210938, -135.64776611328125, -132.8502197265625, -133.51565551757812], "reading_logprob_std": 4.395155792114534, "tokenization_path_logprob_std": 127.81863726179944, "trajectory_balance_rmse": 3729.3951814077727} {"canonical_length": 239, "canonical_nll": 1.358164767341135, "generation_rouge_l_f1": 0.2744282744282744, "generation_token_f1": 0.5446985446985447, "prompt_id": "14f504daadaa6ae8", "reading_canonical_log_prob": -324.60137939453125, "reading_distances": [23, 17, 20, 23], "reading_log_probs": [-333.30078125, -331.4067687988281, -329.263671875, -329.79339599609375], "reading_logprob_std": 1.8180513941263936, "tokenization_path_logprob_std": 434.1918968170068, "trajectory_balance_rmse": 8093.728054829483} {"canonical_length": 97, "canonical_nll": 1.5834910599226804, "generation_rouge_l_f1": 0.1701149425287356, "generation_token_f1": 0.3218390804597701, "prompt_id": "20abb31273f6c931", "reading_canonical_log_prob": -153.5986328125, "reading_distances": [20, 23, 30, 29], "reading_log_probs": [-152.79869079589844, -151.38780212402344, -149.96498107910156, -152.7948760986328], "reading_logprob_std": 1.3550250573597287, "tokenization_path_logprob_std": 318.1602189561438, "trajectory_balance_rmse": 3199.709748510843} {"canonical_length": 5, "canonical_nll": 0.45745134353637695, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "609097a0b2afbf5d", "reading_canonical_log_prob": -2.2872567176818848, "reading_distances": [26, 26, 23, 33], "reading_log_probs": [-6.686130046844482, -4.153532981872559, -6.763201713562012, -5.363585948944092], "reading_logprob_std": 1.2383670089762513, "tokenization_path_logprob_std": 39.67441233483111, "trajectory_balance_rmse": 90.45478687169606} {"canonical_length": 70, "canonical_nll": 2.5032383510044642, "generation_rouge_l_f1": 0.08074534161490683, "generation_token_f1": 0.13664596273291926, "prompt_id": "b13cf3764c885d75", "reading_canonical_log_prob": -175.2266845703125, "reading_distances": [30, 22, 25, 35], "reading_log_probs": [-174.21018981933594, -177.32521057128906, -174.6039581298828, -172.0904998779297], "reading_logprob_std": 2.1501235718854645, "tokenization_path_logprob_std": 113.63420682296223, "trajectory_balance_rmse": 2011.9567803764887} {"canonical_length": 42, "canonical_nll": 0.9938864935012091, "generation_rouge_l_f1": 0.17224880382775118, "generation_token_f1": 0.22009569377990432, "prompt_id": "d289e388118f2de0", "reading_canonical_log_prob": -41.74323272705078, "reading_distances": [21, 22, 26, 20], "reading_log_probs": [-42.275291442871094, -44.80739974975586, -45.707366943359375, -41.51091003417969], "reading_logprob_std": 2.0012918669754245, "tokenization_path_logprob_std": 152.33808355108675, "trajectory_balance_rmse": 996.9158057927575} {"canonical_length": 10, "canonical_nll": 2.4615333557128904, "generation_rouge_l_f1": 0.35294117647058826, "generation_token_f1": 0.35294117647058826, "prompt_id": "58e5e1069ca06eb2", "reading_canonical_log_prob": -24.615333557128906, "reading_distances": [23, 12, 20, 22], "reading_log_probs": [-29.08696746826172, -28.04248809814453, -26.450225830078125, -27.353483200073242], "reading_logprob_std": 1.1133362806415505, "tokenization_path_logprob_std": 79.93247988051661, "trajectory_balance_rmse": 367.1404290560177} {"canonical_length": 113, "canonical_nll": 1.786265685495022, "generation_rouge_l_f1": 0.2702702702702703, "generation_token_f1": 0.36756756756756753, "prompt_id": "ec3868510227235c", "reading_canonical_log_prob": -201.8480224609375, "reading_distances": [34, 28, 26, 26], "reading_log_probs": [-207.11756896972656, -206.18991088867188, -204.6951904296875, -205.79562377929688], "reading_logprob_std": 1.0031870681793058, "tokenization_path_logprob_std": 158.69019112196287, "trajectory_balance_rmse": 3465.1680172112847} {"canonical_length": 16, "canonical_nll": 2.9343984127044678, "generation_rouge_l_f1": 0.125, "generation_token_f1": 0.18750000000000003, "prompt_id": "d625c7ada7581821", "reading_canonical_log_prob": -46.950374603271484, "reading_distances": [11, 10, 11, 10], "reading_log_probs": [-44.06721496582031, -46.58985137939453, -48.00336456298828, -46.70335006713867], "reading_logprob_std": 1.645881616857437, "tokenization_path_logprob_std": 38.29540036255459, "trajectory_balance_rmse": 532.713140525592} {"canonical_length": 7, "canonical_nll": 0.950664928981236, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.5714285714285715, "prompt_id": "34ca0227975643ca", "reading_canonical_log_prob": -6.654654502868652, "reading_distances": [23, 19, 21, 16], "reading_log_probs": [-7.671399116516113, -6.838437080383301, -7.159346103668213, -6.097151279449463], "reading_logprob_std": 0.6592309521077735, "tokenization_path_logprob_std": 20.95542577269105, "trajectory_balance_rmse": 82.67579704427372} {"canonical_length": 92, "canonical_nll": 0.5894963223001232, "generation_rouge_l_f1": 0.40384615384615385, "generation_token_f1": 0.49358974358974356, "prompt_id": "03d8c4a193a46caa", "reading_canonical_log_prob": -54.23366165161133, "reading_distances": [29, 29, 32, 23], "reading_log_probs": [-55.14442443847656, -55.94329071044922, -54.857425689697266, -55.345184326171875], "reading_logprob_std": 0.4596740432510605, "tokenization_path_logprob_std": 168.29846670991026, "trajectory_balance_rmse": 1801.3849762460725} {"canonical_length": 22, "canonical_nll": 1.180427377874201, "generation_rouge_l_f1": 0.358974358974359, "generation_token_f1": 0.5128205128205129, "prompt_id": "cfb9b15522adbd99", "reading_canonical_log_prob": -25.969402313232422, "reading_distances": [22, 14, 16, 16], "reading_log_probs": [-26.555892944335938, -27.42025375366211, -26.79940414428711, -27.365863800048828], "reading_logprob_std": 0.4254173584479282, "tokenization_path_logprob_std": 45.008195648775626, "trajectory_balance_rmse": 724.9004553001017} {"canonical_length": 115, "canonical_nll": 1.403217481530231, "generation_rouge_l_f1": 0.26525198938992045, "generation_token_f1": 0.40318302387267907, "prompt_id": "7e152e3b35975991", "reading_canonical_log_prob": -161.37001037597656, "reading_distances": [25, 29, 33, 23], "reading_log_probs": [-166.7274627685547, -164.2747039794922, -167.99307250976562, -168.1383056640625], "reading_logprob_std": 1.788459452136243, "tokenization_path_logprob_std": 117.4814344486195, "trajectory_balance_rmse": 4498.856619159743} {"canonical_length": 10, "canonical_nll": 4.112678527832031, "generation_rouge_l_f1": 0.06153846153846155, "generation_token_f1": 0.07692307692307693, "prompt_id": "6df3b4a03782fc55", "reading_canonical_log_prob": -41.12678527832031, "reading_distances": [26, 26, 24, 27], "reading_log_probs": [-39.32568359375, -39.03411865234375, -45.44929122924805, -40.22166442871094], "reading_logprob_std": 3.003874558706511, "tokenization_path_logprob_std": 62.89789219287203, "trajectory_balance_rmse": 392.80964184459873} {"canonical_length": 150, "canonical_nll": 1.6075439453125, "generation_rouge_l_f1": 0.21307506053268765, "generation_token_f1": 0.4067796610169491, "prompt_id": "68f61446f4cfbca0", "reading_canonical_log_prob": -241.131591796875, "reading_distances": [20, 26, 25, 32], "reading_log_probs": [-245.3811492919922, -243.97036743164062, -246.099365234375, -243.8845977783203], "reading_logprob_std": 1.087466052816549, "tokenization_path_logprob_std": 367.60750731315613, "trajectory_balance_rmse": 4032.239050739624} {"canonical_length": 206, "canonical_nll": 1.5014550662735133, "generation_rouge_l_f1": 0.22632794457274827, "generation_token_f1": 0.44341801385681295, "prompt_id": "6a128630a7f46740", "reading_canonical_log_prob": -309.29974365234375, "reading_distances": [26, 28, 25, 28], "reading_log_probs": [-309.3092041015625, -310.796630859375, -310.989013671875, -312.0106201171875], "reading_logprob_std": 1.1137888727065421, "tokenization_path_logprob_std": 212.1469259434055, "trajectory_balance_rmse": 5697.52927534781} {"canonical_length": 324, "canonical_nll": 1.3152298162012925, "generation_rouge_l_f1": 0.21052631578947367, "generation_token_f1": 0.39649122807017545, "prompt_id": "fc8ef010d8778b0f", "reading_canonical_log_prob": -426.13446044921875, "reading_distances": [41, 46, 39, 39], "reading_log_probs": [-428.4352111816406, -428.3450622558594, -429.59735107421875, -427.6031494140625], "reading_logprob_std": 0.8239399732680379, "tokenization_path_logprob_std": 425.7787147884195, "trajectory_balance_rmse": 9321.618214186003} {"canonical_length": 54, "canonical_nll": 0.31652390515362777, "generation_rouge_l_f1": 0.21498371335504882, "generation_token_f1": 0.254071661237785, "prompt_id": "68eb421daa21c33e", "reading_canonical_log_prob": -17.0922908782959, "reading_distances": [20, 21, 24, 17], "reading_log_probs": [-20.730680465698242, -19.005807876586914, -21.170303344726562, -22.519046783447266], "reading_logprob_std": 1.449531675020663, "tokenization_path_logprob_std": 43.53799824834568, "trajectory_balance_rmse": 977.2182996938376} {"canonical_length": 133, "canonical_nll": 1.091004565246123, "generation_rouge_l_f1": 0.4383561643835616, "generation_token_f1": 0.5863013698630137, "prompt_id": "a12c87820d75347d", "reading_canonical_log_prob": -145.10360717773438, "reading_distances": [22, 28, 23, 20], "reading_log_probs": [-147.98599243164062, -148.02366638183594, -149.45169067382812, -144.92864990234375], "reading_logprob_std": 1.9055469952326076, "tokenization_path_logprob_std": 85.010326274708, "trajectory_balance_rmse": 3693.222325655606} {"canonical_length": 220, "canonical_nll": 0.029653347622264514, "generation_rouge_l_f1": 0.9027027027027027, "generation_token_f1": 0.9135135135135135, "prompt_id": "c386599fdaa650e0", "reading_canonical_log_prob": -6.523736476898193, "reading_distances": [14, 10, 11, 12], "reading_log_probs": [-7.63262939453125, -9.4688720703125, -8.953997611999512, -7.726490497589111], "reading_logprob_std": 0.9098705954639944, "tokenization_path_logprob_std": 188.27583678566796, "trajectory_balance_rmse": 3178.281265456334} {"canonical_length": 13, "canonical_nll": 0.6373811868520883, "generation_rouge_l_f1": 0.380952380952381, "generation_token_f1": 0.380952380952381, "prompt_id": "28aa44aad6506016", "reading_canonical_log_prob": -8.285955429077148, "reading_distances": [24, 28, 30, 33], "reading_log_probs": [-10.86003303527832, -11.048595428466797, -12.925698280334473, -11.549322128295898], "reading_logprob_std": 0.9330193723793946, "tokenization_path_logprob_std": 56.43498350010302, "trajectory_balance_rmse": 380.3869059030664} {"canonical_length": 150, "canonical_nll": 1.5361335245768228, "generation_rouge_l_f1": 0.27842227378190254, "generation_token_f1": 0.4686774941995359, "prompt_id": "764869f6f55c420d", "reading_canonical_log_prob": -230.42002868652344, "reading_distances": [11, 13, 15, 13], "reading_log_probs": [-232.80545043945312, -229.8839111328125, -232.1669158935547, -231.21661376953125], "reading_logprob_std": 1.2701312738142594, "tokenization_path_logprob_std": 152.16374778546904, "trajectory_balance_rmse": 3356.3517745517083} {"canonical_length": 266, "canonical_nll": 0.8310038631123707, "generation_rouge_l_f1": 0.33199999999999996, "generation_token_f1": 0.5519999999999999, "prompt_id": "8fb6a1df0bc7d753", "reading_canonical_log_prob": -221.04702758789062, "reading_distances": [44, 32, 42, 44], "reading_log_probs": [-224.69888305664062, -226.55728149414062, -230.30718994140625, -234.2076416015625], "reading_logprob_std": 4.214326619017943, "tokenization_path_logprob_std": 172.2254681285461, "trajectory_balance_rmse": 6217.7410074422} {"canonical_length": 81, "canonical_nll": 1.8771592128423997, "generation_rouge_l_f1": 0.2176470588235294, "generation_token_f1": 0.30588235294117644, "prompt_id": "f5aeb1160ca2b326", "reading_canonical_log_prob": -152.04989624023438, "reading_distances": [52, 48, 48, 53], "reading_log_probs": [-145.22227478027344, -146.7095184326172, -147.65927124023438, -148.44178771972656], "reading_logprob_std": 1.3853768534807, "tokenization_path_logprob_std": 132.02853503247948, "trajectory_balance_rmse": 2391.6992292376813} {"canonical_length": 29, "canonical_nll": 0.8601179451778017, "generation_rouge_l_f1": 0.591549295774648, "generation_token_f1": 0.647887323943662, "prompt_id": "7d27494a4e6427f6", "reading_canonical_log_prob": -24.94342041015625, "reading_distances": [38, 36, 37, 27], "reading_log_probs": [-24.929668426513672, -26.578847885131836, -25.752887725830078, -26.5523738861084], "reading_logprob_std": 0.7827699261781991, "tokenization_path_logprob_std": 54.28496180296787, "trajectory_balance_rmse": 798.0781634882374} {"canonical_length": 13, "canonical_nll": 0.49779730576735276, "generation_rouge_l_f1": 0.8421052631578948, "generation_token_f1": 0.8421052631578948, "prompt_id": "b015ce761d144f7f", "reading_canonical_log_prob": -6.471364974975586, "reading_distances": [29, 23, 31, 25], "reading_log_probs": [-6.884650707244873, -9.095794677734375, -7.572760581970215, -8.46835708618164], "reading_logprob_std": 0.9740885756747599, "tokenization_path_logprob_std": 80.94122338804904, "trajectory_balance_rmse": 270.0628444567755} {"canonical_length": 246, "canonical_nll": 1.3544993827013465, "generation_rouge_l_f1": 0.2054263565891473, "generation_token_f1": 0.4728682170542635, "prompt_id": "e7d47f51d6085f91", "reading_canonical_log_prob": -333.20684814453125, "reading_distances": [24, 16, 22, 18], "reading_log_probs": [-333.5026550292969, -332.21856689453125, -335.1287536621094, -334.48211669921875], "reading_logprob_std": 1.266998038889259, "tokenization_path_logprob_std": 195.37052963878614, "trajectory_balance_rmse": 7556.286555056577} {"canonical_length": 205, "canonical_nll": 1.6850274807069359, "generation_rouge_l_f1": 0.242152466367713, "generation_token_f1": 0.4573991031390135, "prompt_id": "321f893720a9d001", "reading_canonical_log_prob": -345.4306335449219, "reading_distances": [29, 23, 37, 23], "reading_log_probs": [-348.9571228027344, -348.58819580078125, -347.00799560546875, -350.5098876953125], "reading_logprob_std": 1.4375750235262166, "tokenization_path_logprob_std": 593.7029194286224, "trajectory_balance_rmse": 6616.434867153464} {"canonical_length": 159, "canonical_nll": 1.4306122401975236, "generation_rouge_l_f1": 0.31932773109243695, "generation_token_f1": 0.46218487394957974, "prompt_id": "46617b87eb38bf31", "reading_canonical_log_prob": -227.46734619140625, "reading_distances": [34, 38, 37, 34], "reading_log_probs": [-225.793701171875, -233.40524291992188, -230.92266845703125, -233.36764526367188], "reading_logprob_std": 3.5794436652536183, "tokenization_path_logprob_std": 161.09966023865505, "trajectory_balance_rmse": 2805.7902920236534} {"canonical_length": 4, "canonical_nll": 0.6197775602340698, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "03d95aafdc590416", "reading_canonical_log_prob": -2.4791102409362793, "reading_distances": [38, 34, 23, 35], "reading_log_probs": [-3.5362396240234375, -3.288079023361206, -2.9789183139801025, -3.0655016899108887], "reading_logprob_std": 0.24939971657636623, "tokenization_path_logprob_std": 20.584055846497332, "trajectory_balance_rmse": 82.49106926701523} {"canonical_length": 153, "canonical_nll": 0.7041756125057445, "generation_rouge_l_f1": 0.4893111638954869, "generation_token_f1": 0.684085510688836, "prompt_id": "3e79c37d02563d09", "reading_canonical_log_prob": -107.7388687133789, "reading_distances": [39, 36, 39, 30], "reading_log_probs": [-110.44168853759766, -111.70391845703125, -105.68446350097656, -115.46803283691406], "reading_logprob_std": 4.037420540164917, "tokenization_path_logprob_std": 194.79994233744839, "trajectory_balance_rmse": 2812.7805734991043} {"canonical_length": 385, "canonical_nll": 1.0474644252232144, "generation_rouge_l_f1": 0.22758620689655173, "generation_token_f1": 0.4137931034482758, "prompt_id": "8a6a50a6d59a29d8", "reading_canonical_log_prob": -403.2738037109375, "reading_distances": [35, 25, 33, 31], "reading_log_probs": [-418.5518493652344, -411.9247741699219, -404.33837890625, -404.9059143066406], "reading_logprob_std": 6.703806723694666, "tokenization_path_logprob_std": 311.8456949957713, "trajectory_balance_rmse": 8854.984187043005} {"canonical_length": 28, "canonical_nll": 1.083698000226702, "generation_rouge_l_f1": 0.1487603305785124, "generation_token_f1": 0.1652892561983471, "prompt_id": "280d0259825e43a6", "reading_canonical_log_prob": -30.343544006347656, "reading_distances": [32, 33, 36, 36], "reading_log_probs": [-30.639469146728516, -31.554948806762695, -28.1276798248291, -29.7010440826416], "reading_logprob_std": 1.4630573210135247, "tokenization_path_logprob_std": 116.7758587163317, "trajectory_balance_rmse": 579.752900356475} {"canonical_length": 245, "canonical_nll": 1.6953326789700256, "generation_rouge_l_f1": 0.11928429423459244, "generation_token_f1": 0.2544731610337972, "prompt_id": "fcecb04380951f2a", "reading_canonical_log_prob": -415.35650634765625, "reading_distances": [26, 23, 25, 26], "reading_log_probs": [-416.31134033203125, -414.44854736328125, -417.490478515625, -419.02215576171875], "reading_logprob_std": 1.930590559438652, "tokenization_path_logprob_std": 254.0682710941888, "trajectory_balance_rmse": 7152.525119955554} {"canonical_length": 126, "canonical_nll": 1.5395461793929812, "generation_rouge_l_f1": 0.22837370242214533, "generation_token_f1": 0.35986159169550175, "prompt_id": "1ae061f91192abf8", "reading_canonical_log_prob": -193.98281860351562, "reading_distances": [29, 32, 30, 21], "reading_log_probs": [-208.91702270507812, -210.99771118164062, -213.07672119140625, -224.5193328857422], "reading_logprob_std": 6.971097101151556, "tokenization_path_logprob_std": 221.57085125149052, "trajectory_balance_rmse": 3057.633267897257} {"canonical_length": 205, "canonical_nll": 1.4380118021150914, "generation_rouge_l_f1": 0.20472440944881892, "generation_token_f1": 0.3346456692913386, "prompt_id": "eedaf90695c8e20b", "reading_canonical_log_prob": -294.79241943359375, "reading_distances": [16, 17, 16, 21], "reading_log_probs": [-293.6681213378906, -293.3002624511719, -292.0479431152344, -292.8202209472656], "reading_logprob_std": 0.6996659512827028, "tokenization_path_logprob_std": 211.71011637971836, "trajectory_balance_rmse": 6716.312844582372} {"canonical_length": 119, "canonical_nll": 1.6117022858948267, "generation_rouge_l_f1": 0.16666666666666666, "generation_token_f1": 0.35714285714285715, "prompt_id": "a0b38c7913178787", "reading_canonical_log_prob": -191.79257202148438, "reading_distances": [48, 44, 45, 41], "reading_log_probs": [-172.31399536132812, -173.17660522460938, -173.75173950195312, -172.33338928222656], "reading_logprob_std": 0.6991119380186357, "tokenization_path_logprob_std": 104.35258911477274, "trajectory_balance_rmse": 2155.854522326762} {"canonical_length": 68, "canonical_nll": 1.7116814781637752, "generation_rouge_l_f1": 0.2649842271293375, "generation_token_f1": 0.334384858044164, "prompt_id": "4101000b96f3bbde", "reading_canonical_log_prob": -116.39434051513672, "reading_distances": [18, 19, 23, 15], "reading_log_probs": [-120.86672973632812, -130.6025848388672, -126.42025756835938, -125.53274536132812], "reading_logprob_std": 3.9935683648294846, "tokenization_path_logprob_std": 58.980509455838366, "trajectory_balance_rmse": 1550.4061173612754} {"canonical_length": 138, "canonical_nll": 1.364782250445822, "generation_rouge_l_f1": 0.18581907090464547, "generation_token_f1": 0.3374083129584352, "prompt_id": "5539256ae94ad40f", "reading_canonical_log_prob": -188.33995056152344, "reading_distances": [63, 57, 65, 62], "reading_log_probs": [-191.74957275390625, -192.79830932617188, -192.30133056640625, -191.45570373535156], "reading_logprob_std": 0.5954904476765535, "tokenization_path_logprob_std": 313.3706577023257, "trajectory_balance_rmse": 3534.1568404947775} {"canonical_length": 86, "canonical_nll": 1.4264655002327853, "generation_rouge_l_f1": 0.2367601246105919, "generation_token_f1": 0.3302180685358255, "prompt_id": "3b0f78eeffb87a9e", "reading_canonical_log_prob": -122.67603302001953, "reading_distances": [32, 33, 33, 35], "reading_log_probs": [-122.44449615478516, -127.30348205566406, -124.3630142211914, -122.34700775146484], "reading_logprob_std": 2.3197922005462743, "tokenization_path_logprob_std": 111.15535930258008, "trajectory_balance_rmse": 2303.1440200750867} {"canonical_length": 302, "canonical_nll": 1.4369143050237996, "generation_rouge_l_f1": 0.3190661478599221, "generation_token_f1": 0.48249027237354086, "prompt_id": "3a8d8459792a6678", "reading_canonical_log_prob": -433.9481201171875, "reading_distances": [14, 16, 18, 17], "reading_log_probs": [-440.12640380859375, -436.324951171875, -434.95623779296875, -430.62335205078125], "reading_logprob_std": 3.9226389841120626, "tokenization_path_logprob_std": 134.7607359100867, "trajectory_balance_rmse": 7984.364360887868} {"canonical_length": 289, "canonical_nll": 1.1851567991052119, "generation_rouge_l_f1": 0.30612244897959184, "generation_token_f1": 0.5836734693877551, "prompt_id": "a788a16b80a319d9", "reading_canonical_log_prob": -342.51031494140625, "reading_distances": [12, 20, 16, 20], "reading_log_probs": [-344.5230712890625, -345.24530029296875, -344.46917724609375, -343.54229736328125], "reading_logprob_std": 0.698100343493001, "tokenization_path_logprob_std": 225.04403092812146, "trajectory_balance_rmse": 6980.271327270764} {"canonical_length": 252, "canonical_nll": 1.3810033646840898, "generation_rouge_l_f1": 0.15970961887477314, "generation_token_f1": 0.3557168784029038, "prompt_id": "79833cd8d7978d26", "reading_canonical_log_prob": -348.0128479003906, "reading_distances": [30, 30, 26, 31], "reading_log_probs": [-350.5531005859375, -345.681884765625, -348.324462890625, -349.6141357421875], "reading_logprob_std": 2.1151496691781184, "tokenization_path_logprob_std": 292.02986244687077, "trajectory_balance_rmse": 5679.753725856289} {"canonical_length": 26, "canonical_nll": 2.0140066880446215, "generation_rouge_l_f1": 0.1794871794871795, "generation_token_f1": 0.1794871794871795, "prompt_id": "7c29635f759f46ec", "reading_canonical_log_prob": -52.364173889160156, "reading_distances": [18, 22, 22, 16], "reading_log_probs": [-45.708717346191406, -48.50112533569336, -49.24708557128906, -49.26309585571289], "reading_logprob_std": 1.6854403663147755, "tokenization_path_logprob_std": 47.885241366627376, "trajectory_balance_rmse": 774.6105008839564} {"canonical_length": 346, "canonical_nll": 1.3236048703937862, "generation_rouge_l_f1": 0.3516819571865443, "generation_token_f1": 0.5443425076452599, "prompt_id": "88945c2caccedba1", "reading_canonical_log_prob": -457.96728515625, "reading_distances": [39, 36, 53, 44], "reading_log_probs": [-461.06256103515625, -461.73516845703125, -459.6956481933594, -465.67718505859375], "reading_logprob_std": 2.567319084859364, "tokenization_path_logprob_std": 240.672157338615, "trajectory_balance_rmse": 10153.748180656652} {"canonical_length": 17, "canonical_nll": 1.4444149241727942, "generation_rouge_l_f1": 0.4666666666666667, "generation_token_f1": 0.4666666666666667, "prompt_id": "f3be16dc1ce038fa", "reading_canonical_log_prob": -24.5550537109375, "reading_distances": [34, 49, 48, 47], "reading_log_probs": [-25.094940185546875, -25.134586334228516, -24.88580894470215, -25.214689254760742], "reading_logprob_std": 0.14027249027687508, "tokenization_path_logprob_std": 96.43408877581423, "trajectory_balance_rmse": 652.1725190958797} {"canonical_length": 56, "canonical_nll": 1.1589130674089705, "generation_rouge_l_f1": 0.3076923076923077, "generation_token_f1": 0.49230769230769234, "prompt_id": "08f613a2192f5ff2", "reading_canonical_log_prob": -64.89913177490234, "reading_distances": [23, 24, 20, 20], "reading_log_probs": [-71.47360229492188, -68.59788513183594, -66.95806884765625, -67.73230743408203], "reading_logprob_std": 1.9726197964595344, "tokenization_path_logprob_std": 33.960578370924985, "trajectory_balance_rmse": 1536.8013740047832} {"canonical_length": 180, "canonical_nll": 1.8335890028211805, "generation_rouge_l_f1": 0.17972350230414746, "generation_token_f1": 0.33640552995391704, "prompt_id": "ddaf5127b3985530", "reading_canonical_log_prob": -330.0460205078125, "reading_distances": [26, 27, 31, 27], "reading_log_probs": [-328.62603759765625, -330.5607604980469, -330.20721435546875, -331.5132751464844], "reading_logprob_std": 1.2013009310865745, "tokenization_path_logprob_std": 117.20713592102534, "trajectory_balance_rmse": 4909.170131245016} {"canonical_length": 184, "canonical_nll": 1.0784160779870076, "generation_rouge_l_f1": 0.2579075425790754, "generation_token_f1": 0.5352798053527981, "prompt_id": "6fc1da282516cfc1", "reading_canonical_log_prob": -198.42855834960938, "reading_distances": [18, 12, 9, 14], "reading_log_probs": [-206.88690185546875, -204.6075439453125, -204.24484252929688, -205.19256591796875], "reading_logprob_std": 1.1697117480321704, "tokenization_path_logprob_std": 257.1684090987446, "trajectory_balance_rmse": 4808.834450913784} {"canonical_length": 80, "canonical_nll": 1.3681278228759766, "generation_rouge_l_f1": 0.1411764705882353, "generation_token_f1": 0.23529411764705882, "prompt_id": "9d2634370eff9e7e", "reading_canonical_log_prob": -109.45022583007812, "reading_distances": [39, 26, 33, 34], "reading_log_probs": [-110.09564971923828, -116.38504028320312, -112.86278533935547, -116.1373291015625], "reading_logprob_std": 2.984761862843094, "tokenization_path_logprob_std": 169.69679810619346, "trajectory_balance_rmse": 2515.894299042217} {"canonical_length": 132, "canonical_nll": 0.9308261293353457, "generation_rouge_l_f1": 0.44599303135888496, "generation_token_f1": 0.5923344947735192, "prompt_id": "8a81f9867fa3fdd7", "reading_canonical_log_prob": -122.86904907226562, "reading_distances": [33, 32, 31, 34], "reading_log_probs": [-120.02166748046875, -117.53226470947266, -118.02497100830078, -118.67643737792969], "reading_logprob_std": 1.07896037963614, "tokenization_path_logprob_std": 83.67986377374828, "trajectory_balance_rmse": 2467.565455137816} {"canonical_length": 33, "canonical_nll": 1.496804208466501, "generation_rouge_l_f1": 0.631578947368421, "generation_token_f1": 0.7017543859649122, "prompt_id": "32094cae1c9e7173", "reading_canonical_log_prob": -49.39453887939453, "reading_distances": [44, 46, 39, 44], "reading_log_probs": [-46.505104064941406, -46.91450881958008, -46.302284240722656, -47.87031555175781], "reading_logprob_std": 0.6963989621960243, "tokenization_path_logprob_std": 62.10069231602638, "trajectory_balance_rmse": 825.9736586129824} {"canonical_length": 23, "canonical_nll": 1.9315639993418818, "generation_rouge_l_f1": 0.48275862068965514, "generation_token_f1": 0.48275862068965514, "prompt_id": "896dee39b103c9ac", "reading_canonical_log_prob": -44.42597198486328, "reading_distances": [27, 29, 30, 26], "reading_log_probs": [-52.86595153808594, -60.85826110839844, -61.97734832763672, -64.28797912597656], "reading_logprob_std": 4.964134823188215, "tokenization_path_logprob_std": 129.57079951176868, "trajectory_balance_rmse": 867.8281035466966} {"canonical_length": 105, "canonical_nll": 1.9032162620907738, "generation_rouge_l_f1": 0.23243243243243247, "generation_token_f1": 0.39459459459459456, "prompt_id": "1057782952469d89", "reading_canonical_log_prob": -199.83770751953125, "reading_distances": [11, 10, 10, 9], "reading_log_probs": [-197.65982055664062, -196.64122009277344, -196.65621948242188, -196.33358764648438], "reading_logprob_std": 0.5775385815687277, "tokenization_path_logprob_std": 256.339587275303, "trajectory_balance_rmse": 3657.986884139661} {"canonical_length": 25, "canonical_nll": 0.5380147171020507, "generation_rouge_l_f1": 0.7272727272727274, "generation_token_f1": 0.7272727272727274, "prompt_id": "03db11358469fd8f", "reading_canonical_log_prob": -13.45036792755127, "reading_distances": [19, 20, 25, 24], "reading_log_probs": [-9.187347412109375, -10.620555877685547, -9.365469932556152, -9.329235076904297], "reading_logprob_std": 0.6677081044180321, "tokenization_path_logprob_std": 49.97382724849796, "trajectory_balance_rmse": 302.8889981928467} {"canonical_length": 240, "canonical_nll": 1.2636385599772135, "generation_rouge_l_f1": 0.2236024844720497, "generation_token_f1": 0.5383022774327123, "prompt_id": "85339fcc838b71e5", "reading_canonical_log_prob": -303.27325439453125, "reading_distances": [21, 19, 22, 25], "reading_log_probs": [-302.29864501953125, -300.68768310546875, -301.98236083984375, -299.87213134765625], "reading_logprob_std": 1.1320183370839656, "tokenization_path_logprob_std": 254.97597450241304, "trajectory_balance_rmse": 7799.195985576778} {"canonical_length": 28, "canonical_nll": 0.632936954498291, "generation_rouge_l_f1": 0.1515151515151515, "generation_token_f1": 0.18181818181818182, "prompt_id": "5af4615607f58b8b", "reading_canonical_log_prob": -17.72223472595215, "reading_distances": [47, 45, 38, 39], "reading_log_probs": [-15.126167297363281, -16.9318904876709, -15.672028541564941, -17.404415130615234], "reading_logprob_std": 1.0630423667870357, "tokenization_path_logprob_std": 127.8853380267367, "trajectory_balance_rmse": 860.1132608075909} {"canonical_length": 65, "canonical_nll": 2.1842181865985575, "generation_rouge_l_f1": 0.2604166666666667, "generation_token_f1": 0.3645833333333333, "prompt_id": "13f18463e2c1c5f7", "reading_canonical_log_prob": -141.97418212890625, "reading_distances": [25, 18, 20, 18], "reading_log_probs": [-143.5923614501953, -155.58380126953125, -149.88356018066406, -146.64212036132812], "reading_logprob_std": 5.128581495410075, "tokenization_path_logprob_std": 29.80255642815785, "trajectory_balance_rmse": 1882.8038685511317} {"canonical_length": 64, "canonical_nll": 2.060919761657715, "generation_rouge_l_f1": 0.17575757575757575, "generation_token_f1": 0.2545454545454545, "prompt_id": "d02d88b60737f21f", "reading_canonical_log_prob": -131.89886474609375, "reading_distances": [21, 23, 16, 19], "reading_log_probs": [-134.39093017578125, -132.8111572265625, -138.01553344726562, -129.86419677734375], "reading_logprob_std": 3.395329438478207, "tokenization_path_logprob_std": 49.9686749441646, "trajectory_balance_rmse": 2156.759555679328} {"canonical_length": 236, "canonical_nll": 1.4787644208487818, "generation_rouge_l_f1": 0.224, "generation_token_f1": 0.468, "prompt_id": "559f28b026670bdd", "reading_canonical_log_prob": -348.9884033203125, "reading_distances": [19, 13, 16, 14], "reading_log_probs": [-354.4485778808594, -350.31304931640625, -352.7046813964844, -353.0063171386719], "reading_logprob_std": 1.7148485645957607, "tokenization_path_logprob_std": 58.87369589549156, "trajectory_balance_rmse": 6920.52666720802} {"canonical_length": 50, "canonical_nll": 0.8980606079101563, "generation_rouge_l_f1": 0.3381642512077294, "generation_token_f1": 0.37681159420289856, "prompt_id": "44cafb7b8f27398a", "reading_canonical_log_prob": -44.90303039550781, "reading_distances": [30, 23, 28, 26], "reading_log_probs": [-46.78028106689453, -45.457244873046875, -46.89531326293945, -46.37434768676758], "reading_logprob_std": 0.652492814900278, "tokenization_path_logprob_std": 95.19182365437814, "trajectory_balance_rmse": 1441.6806389745013} {"canonical_length": 274, "canonical_nll": 1.1407261312442973, "generation_rouge_l_f1": 0.28044280442804426, "generation_token_f1": 0.5276752767527676, "prompt_id": "55f3c7490c4e7162", "reading_canonical_log_prob": -312.5589599609375, "reading_distances": [14, 17, 22, 21], "reading_log_probs": [-308.46588134765625, -309.2979736328125, -304.3115234375, -306.6746520996094], "reading_logprob_std": 2.2077599493290148, "tokenization_path_logprob_std": 266.1419481195193, "trajectory_balance_rmse": 6687.6653959060695} {"canonical_length": 308, "canonical_nll": 1.198254275631595, "generation_rouge_l_f1": 0.20557491289198607, "generation_token_f1": 0.41463414634146334, "prompt_id": "41cd8f9fa13781b8", "reading_canonical_log_prob": -369.06231689453125, "reading_distances": [28, 27, 31, 17], "reading_log_probs": [-375.66131591796875, -377.7158508300781, -374.85931396484375, -378.7391052246094], "reading_logprob_std": 1.7934310806312075, "tokenization_path_logprob_std": 199.6840449504411, "trajectory_balance_rmse": 8792.474416449142} {"canonical_length": 373, "canonical_nll": 1.2581553932169487, "generation_rouge_l_f1": 0.19093851132686088, "generation_token_f1": 0.4110032362459547, "prompt_id": "28de6d5b6a032c7b", "reading_canonical_log_prob": -469.2919616699219, "reading_distances": [26, 21, 23, 25], "reading_log_probs": [-459.5745849609375, -460.621826171875, -459.25457763671875, -462.61700439453125], "reading_logprob_std": 1.5168897248990159, "tokenization_path_logprob_std": 277.41890868285265, "trajectory_balance_rmse": 8403.574732685202} {"canonical_length": 132, "canonical_nll": 1.8942232998934658, "generation_rouge_l_f1": 0.19047619047619047, "generation_token_f1": 0.4126984126984127, "prompt_id": "dc17c375b90b9957", "reading_canonical_log_prob": -250.0374755859375, "reading_distances": [23, 24, 22, 18], "reading_log_probs": [-248.7384033203125, -247.1573944091797, -245.07955932617188, -246.61671447753906], "reading_logprob_std": 1.509990934744346, "tokenization_path_logprob_std": 335.55593307756686, "trajectory_balance_rmse": 3935.8136652688863} {"canonical_length": 24, "canonical_nll": 0.3302619655927022, "generation_rouge_l_f1": 0.8444444444444443, "generation_token_f1": 0.8444444444444443, "prompt_id": "c1a810543bd499fe", "reading_canonical_log_prob": -7.9262871742248535, "reading_distances": [28, 25, 24, 27], "reading_log_probs": [-8.909167289733887, -8.2446928024292, -9.368144035339355, -8.311203956604004], "reading_logprob_std": 0.5317764945168317, "tokenization_path_logprob_std": 69.11604019451846, "trajectory_balance_rmse": 532.7889622017764} {"canonical_length": 15, "canonical_nll": 1.3221256256103515, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.43478260869565216, "prompt_id": "f21bbbf6325c37df", "reading_canonical_log_prob": -19.831884384155273, "reading_distances": [42, 39, 38, 36], "reading_log_probs": [-19.32849884033203, -18.039079666137695, -19.286645889282227, -18.455886840820312], "reading_logprob_std": 0.6354863935458401, "tokenization_path_logprob_std": 98.78034134668643, "trajectory_balance_rmse": 451.5226510728567} {"canonical_length": 141, "canonical_nll": 1.820033296625665, "generation_rouge_l_f1": 0.2573099415204678, "generation_token_f1": 0.4736842105263158, "prompt_id": "ff640b48674add94", "reading_canonical_log_prob": -256.62469482421875, "reading_distances": [48, 54, 50, 41], "reading_log_probs": [-271.9002685546875, -278.3616027832031, -274.15167236328125, -272.2440490722656], "reading_logprob_std": 2.968204400158149, "tokenization_path_logprob_std": 107.02374959639762, "trajectory_balance_rmse": 4065.0336959094934} {"canonical_length": 133, "canonical_nll": 2.138323561589521, "generation_rouge_l_f1": 0.1534653465346535, "generation_token_f1": 0.297029702970297, "prompt_id": "294ad70d174d121a", "reading_canonical_log_prob": -284.39703369140625, "reading_distances": [29, 22, 27, 28], "reading_log_probs": [-285.712890625, -288.7873840332031, -287.786376953125, -288.97662353515625], "reading_logprob_std": 1.4960593088830019, "tokenization_path_logprob_std": 473.12531722016996, "trajectory_balance_rmse": 4832.602792957428} {"canonical_length": 49, "canonical_nll": 3.010111906090561, "generation_rouge_l_f1": 0.27027027027027023, "generation_token_f1": 0.4864864864864865, "prompt_id": "fdd72207f77ed671", "reading_canonical_log_prob": -147.4954833984375, "reading_distances": [19, 23, 20, 23], "reading_log_probs": [-145.25344848632812, -145.6863250732422, -146.93572998046875, -144.15206909179688], "reading_logprob_std": 1.1508769018176397, "tokenization_path_logprob_std": 126.92568605487672, "trajectory_balance_rmse": 1299.3585987823872} {"canonical_length": 9, "canonical_nll": 1.8290672302246094, "generation_rouge_l_f1": 0.631578947368421, "generation_token_f1": 0.7368421052631577, "prompt_id": "bead008713b25f77", "reading_canonical_log_prob": -16.461605072021484, "reading_distances": [36, 42, 33, 40], "reading_log_probs": [-17.250263214111328, -26.009855270385742, -23.075403213500977, -27.146778106689453], "reading_logprob_std": 4.42609131509399, "tokenization_path_logprob_std": 84.90387499039767, "trajectory_balance_rmse": 313.2047020419347} {"canonical_length": 353, "canonical_nll": 1.0674313985592245, "generation_rouge_l_f1": 0.2401263823064771, "generation_token_f1": 0.3886255924170616, "prompt_id": "5e76d06be1ed709a", "reading_canonical_log_prob": -376.80328369140625, "reading_distances": [31, 31, 23, 25], "reading_log_probs": [-375.4794921875, -374.09747314453125, -377.78240966796875, -377.8924560546875], "reading_logprob_std": 1.8490664955300362, "tokenization_path_logprob_std": 467.3950361469335, "trajectory_balance_rmse": 8490.51563905368} {"canonical_length": 238, "canonical_nll": 1.5692641314338236, "generation_rouge_l_f1": 0.21132075471698114, "generation_token_f1": 0.39999999999999997, "prompt_id": "f6f871799dc3c062", "reading_canonical_log_prob": -373.48486328125, "reading_distances": [30, 29, 31, 29], "reading_log_probs": [-374.3087463378906, -371.62554931640625, -375.2182312011719, -372.4993896484375], "reading_logprob_std": 1.642242877951567, "tokenization_path_logprob_std": 200.45703605403418, "trajectory_balance_rmse": 5100.083850190702} {"canonical_length": 108, "canonical_nll": 1.0403770164207176, "generation_rouge_l_f1": 0.313953488372093, "generation_token_f1": 0.4011627906976744, "prompt_id": "e5243c35b3e882de", "reading_canonical_log_prob": -112.3607177734375, "reading_distances": [26, 27, 23, 32], "reading_log_probs": [-108.5975341796875, -110.36285400390625, -111.80597686767578, -112.18907165527344], "reading_logprob_std": 1.629775959265755, "tokenization_path_logprob_std": 109.92171081136675, "trajectory_balance_rmse": 2013.7304511474401} {"canonical_length": 147, "canonical_nll": 1.699557557397959, "generation_rouge_l_f1": 0.196078431372549, "generation_token_f1": 0.3872549019607843, "prompt_id": "26ddb945cf9f0d7d", "reading_canonical_log_prob": -249.8349609375, "reading_distances": [33, 39, 29, 28], "reading_log_probs": [-253.9288330078125, -253.90884399414062, -254.55398559570312, -252.60321044921875], "reading_logprob_std": 0.8203099413483694, "tokenization_path_logprob_std": 330.7837534745024, "trajectory_balance_rmse": 5102.476373972762} {"canonical_length": 48, "canonical_nll": 1.9330886205037434, "generation_rouge_l_f1": 0.10289389067524116, "generation_token_f1": 0.16077170418006434, "prompt_id": "e4581a40938db0d9", "reading_canonical_log_prob": -92.78825378417969, "reading_distances": [19, 20, 15, 22], "reading_log_probs": [-100.10465240478516, -93.44194793701172, -95.50798034667969, -100.66093444824219], "reading_logprob_std": 3.5209570986506886, "tokenization_path_logprob_std": 77.58268836792332, "trajectory_balance_rmse": 1395.0413010751815} {"canonical_length": 10, "canonical_nll": 0.0187471404671669, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "267c9641a2f8cfb9", "reading_canonical_log_prob": -0.187471404671669, "reading_distances": [22, 25, 25, 23], "reading_log_probs": [-0.4520403742790222, -1.1878244876861572, -3.9046175479888916, -3.3693408966064453], "reading_logprob_std": 1.668302760491167, "tokenization_path_logprob_std": 83.23777953385991, "trajectory_balance_rmse": 270.1405372775245} {"canonical_length": 63, "canonical_nll": 0.9296898009285094, "generation_rouge_l_f1": 0.3181818181818182, "generation_token_f1": 0.35795454545454547, "prompt_id": "df9c5775b476f764", "reading_canonical_log_prob": -58.570457458496094, "reading_distances": [29, 29, 30, 25], "reading_log_probs": [-57.775726318359375, -59.934104919433594, -55.89329528808594, -56.45207977294922], "reading_logprob_std": 1.796270827867026, "tokenization_path_logprob_std": 99.11084979603386, "trajectory_balance_rmse": 1113.8465478301605} {"canonical_length": 106, "canonical_nll": 2.006520685159935, "generation_rouge_l_f1": 0.1806020066889632, "generation_token_f1": 0.35451505016722407, "prompt_id": "518ad8fe3c294839", "reading_canonical_log_prob": -212.69119262695312, "reading_distances": [18, 25, 21, 17], "reading_log_probs": [-217.88333129882812, -230.41693115234375, -213.20849609375, -228.25338745117188], "reading_logprob_std": 8.234294083054891, "tokenization_path_logprob_std": 38.563306486977105, "trajectory_balance_rmse": 2898.7505053532554} {"canonical_length": 203, "canonical_nll": 0.7746349015259391, "generation_rouge_l_f1": 0.3779904306220096, "generation_token_f1": 0.5789473684210528, "prompt_id": "6369662548b008b0", "reading_canonical_log_prob": -157.25088500976562, "reading_distances": [23, 23, 26, 23], "reading_log_probs": [-164.61891174316406, -166.2440948486328, -163.78839111328125, -162.96572875976562], "reading_logprob_std": 1.3999683632111708, "tokenization_path_logprob_std": 213.37080512818108, "trajectory_balance_rmse": 2767.601910327487} {"canonical_length": 68, "canonical_nll": 0.5304584503173828, "generation_rouge_l_f1": 0.2846975088967972, "generation_token_f1": 0.3345195729537367, "prompt_id": "3f0899fb36d7151b", "reading_canonical_log_prob": -36.07117462158203, "reading_distances": [32, 35, 29, 37], "reading_log_probs": [-34.70488739013672, -34.963340759277344, -35.4661979675293, -36.73707962036133], "reading_logprob_std": 0.9032511783630457, "tokenization_path_logprob_std": 115.86489824830242, "trajectory_balance_rmse": 1723.1894143804038} {"canonical_length": 59, "canonical_nll": 0.8738391035694187, "generation_rouge_l_f1": 0.3738317757009346, "generation_token_f1": 0.5794392523364486, "prompt_id": "ca44a0ae267d5905", "reading_canonical_log_prob": -51.5565071105957, "reading_distances": [16, 19, 18, 20], "reading_log_probs": [-56.56989288330078, -57.630340576171875, -55.955116271972656, -56.147216796875], "reading_logprob_std": 0.7485591654945792, "tokenization_path_logprob_std": 172.62647673895466, "trajectory_balance_rmse": 1625.5091403352192} {"canonical_length": 144, "canonical_nll": 2.420779334174262, "generation_rouge_l_f1": 0.20875420875420875, "generation_token_f1": 0.3232323232323232, "prompt_id": "b3b2b51b9f629425", "reading_canonical_log_prob": -348.59222412109375, "reading_distances": [9, 12, 15, 11], "reading_log_probs": [-366.3157043457031, -378.053466796875, -366.9896240234375, -359.99957275390625], "reading_logprob_std": 7.501862419636405, "tokenization_path_logprob_std": 131.16069760413313, "trajectory_balance_rmse": 3781.683831448428} {"canonical_length": 120, "canonical_nll": 1.1926626841227213, "generation_rouge_l_f1": 0.15864022662889518, "generation_token_f1": 0.37393767705382436, "prompt_id": "4412ea6c558d478d", "reading_canonical_log_prob": -143.11952209472656, "reading_distances": [24, 26, 31, 24], "reading_log_probs": [-144.8946533203125, -142.3056182861328, -142.27999877929688, -145.31124877929688], "reading_logprob_std": 1.631359865689205, "tokenization_path_logprob_std": 99.91251534914016, "trajectory_balance_rmse": 3508.2462571291703} {"canonical_length": 30, "canonical_nll": 1.0021069844563801, "generation_rouge_l_f1": 0.14193548387096774, "generation_token_f1": 0.24516129032258063, "prompt_id": "7938d8548e06b8bf", "reading_canonical_log_prob": -30.063209533691406, "reading_distances": [8, 9, 6, 8], "reading_log_probs": [-29.365020751953125, -29.76859474182129, -31.26668930053711, -31.151947021484375], "reading_logprob_std": 0.9636491993899076, "tokenization_path_logprob_std": 97.34900229687304, "trajectory_balance_rmse": 421.11324205174424} {"canonical_length": 332, "canonical_nll": 1.2889856545321912, "generation_rouge_l_f1": 0.2110912343470483, "generation_token_f1": 0.5008944543828265, "prompt_id": "8829bdef994324cc", "reading_canonical_log_prob": -427.9432373046875, "reading_distances": [34, 33, 33, 35], "reading_log_probs": [-421.6782531738281, -424.6676330566406, -426.0228271484375, -422.08050537109375], "reading_logprob_std": 2.0825705994176458, "tokenization_path_logprob_std": 592.7701609674319, "trajectory_balance_rmse": 9554.857247687525} {"canonical_length": 14, "canonical_nll": 1.0520412581307548, "generation_rouge_l_f1": 0.06086956521739131, "generation_token_f1": 0.07826086956521738, "prompt_id": "75996f4d9b7d5f4d", "reading_canonical_log_prob": -14.728577613830566, "reading_distances": [10, 18, 19, 23], "reading_log_probs": [-11.826838493347168, -11.32159423828125, -11.671429634094238, -11.965132713317871], "reading_logprob_std": 0.27708858900031863, "tokenization_path_logprob_std": 69.28941871728141, "trajectory_balance_rmse": 314.9400580073425} {"canonical_length": 164, "canonical_nll": 1.245537734613186, "generation_rouge_l_f1": 0.15308641975308643, "generation_token_f1": 0.32592592592592595, "prompt_id": "f5d8be75fb333542", "reading_canonical_log_prob": -204.2681884765625, "reading_distances": [18, 26, 19, 19], "reading_log_probs": [-204.29840087890625, -206.79962158203125, -204.45947265625, -202.34158325195312], "reading_logprob_std": 1.8245328768497187, "tokenization_path_logprob_std": 246.84807064120673, "trajectory_balance_rmse": 5339.606076758174} {"canonical_length": 45, "canonical_nll": 1.4469550238715279, "generation_rouge_l_f1": 0.303030303030303, "generation_token_f1": 0.4444444444444444, "prompt_id": "61f6f6b3601f9e9d", "reading_canonical_log_prob": -65.11297607421875, "reading_distances": [19, 15, 27, 22], "reading_log_probs": [-67.27603149414062, -67.5071792602539, -69.16983795166016, -69.21549224853516], "reading_logprob_std": 1.0442817192940461, "tokenization_path_logprob_std": 79.85305617422927, "trajectory_balance_rmse": 1197.0326139516092} {"canonical_length": 26, "canonical_nll": 1.5909238962026744, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.42424242424242425, "prompt_id": "4478590e9673bdfe", "reading_canonical_log_prob": -41.36402130126953, "reading_distances": [35, 29, 29, 31], "reading_log_probs": [-37.50718688964844, -40.24830627441406, -38.04296875, -38.767173767089844], "reading_logprob_std": 1.1891910633691143, "tokenization_path_logprob_std": 129.0492365735097, "trajectory_balance_rmse": 1161.4838516659943} {"canonical_length": 90, "canonical_nll": 1.0054502699110244, "generation_rouge_l_f1": 0.13450292397660818, "generation_token_f1": 0.24561403508771928, "prompt_id": "b3f4ac8bd91cf595", "reading_canonical_log_prob": -90.49052429199219, "reading_distances": [30, 32, 23, 32], "reading_log_probs": [-105.53482055664062, -104.88590240478516, -108.11355590820312, -107.3900375366211], "reading_logprob_std": 1.5199977948079268, "tokenization_path_logprob_std": 106.35786140894446, "trajectory_balance_rmse": 2502.152468047445} {"canonical_length": 116, "canonical_nll": 1.0950765280887997, "generation_rouge_l_f1": 0.3285024154589372, "generation_token_f1": 0.5603864734299517, "prompt_id": "47d49f308f57d626", "reading_canonical_log_prob": -127.02887725830078, "reading_distances": [32, 35, 39, 36], "reading_log_probs": [-130.43797302246094, -135.2226104736328, -133.86097717285156, -130.64256286621094], "reading_logprob_std": 2.377685182731129, "tokenization_path_logprob_std": 149.384582815744, "trajectory_balance_rmse": 3879.2631544383526} {"canonical_length": 264, "canonical_nll": 1.2420841563831677, "generation_rouge_l_f1": 0.26037735849056604, "generation_token_f1": 0.558490566037736, "prompt_id": "d8021c0fb98991e7", "reading_canonical_log_prob": -327.91021728515625, "reading_distances": [15, 24, 20, 24], "reading_log_probs": [-338.3248291015625, -332.8499755859375, -333.92596435546875, -335.2956237792969], "reading_logprob_std": 2.371972294869482, "tokenization_path_logprob_std": 294.94558548169687, "trajectory_balance_rmse": 6742.551173873815} {"canonical_length": 165, "canonical_nll": 1.6466591574928977, "generation_rouge_l_f1": 0.14485981308411214, "generation_token_f1": 0.2757009345794392, "prompt_id": "fb0e2e105f1d060e", "reading_canonical_log_prob": -271.6987609863281, "reading_distances": [35, 27, 23, 27], "reading_log_probs": [-273.8316650390625, -270.1261291503906, -266.197021484375, -270.8905334472656], "reading_logprob_std": 3.1453707915450044, "tokenization_path_logprob_std": 209.8446885188559, "trajectory_balance_rmse": 4320.116433697484} {"canonical_length": 12, "canonical_nll": 1.484463055928548, "generation_rouge_l_f1": 0.5263157894736842, "generation_token_f1": 0.5263157894736842, "prompt_id": "d8467aeeb0933718", "reading_canonical_log_prob": -17.813556671142578, "reading_distances": [41, 42, 45, 36], "reading_log_probs": [-21.389602661132812, -22.166358947753906, -21.50271224975586, -21.368072509765625], "reading_logprob_std": 0.3777591485420005, "tokenization_path_logprob_std": 33.96883890882028, "trajectory_balance_rmse": 338.041352905656} {"canonical_length": 62, "canonical_nll": 0.45353763334212766, "generation_rouge_l_f1": 0.819672131147541, "generation_token_f1": 0.8524590163934426, "prompt_id": "b80e624242a44cf6", "reading_canonical_log_prob": -28.119333267211914, "reading_distances": [23, 29, 22, 21], "reading_log_probs": [-33.39171600341797, -32.045528411865234, -30.335269927978516, -30.35953140258789], "reading_logprob_std": 1.4752509874983903, "tokenization_path_logprob_std": 174.72712554904527, "trajectory_balance_rmse": 1813.5600598678127} {"canonical_length": 225, "canonical_nll": 1.3339959038628473, "generation_rouge_l_f1": 0.17316017316017318, "generation_token_f1": 0.3203463203463204, "prompt_id": "aad9f0c226359f45", "reading_canonical_log_prob": -300.1490783691406, "reading_distances": [13, 20, 17, 19], "reading_log_probs": [-305.129150390625, -314.1514892578125, -302.78082275390625, -303.6234130859375], "reading_logprob_std": 5.2442578855827096, "tokenization_path_logprob_std": 320.50751018165096, "trajectory_balance_rmse": 6115.073456574512} {"canonical_length": 221, "canonical_nll": 1.0518301709205318, "generation_rouge_l_f1": 0.3073770491803279, "generation_token_f1": 0.4836065573770492, "prompt_id": "2fd9ba0f00f7f678", "reading_canonical_log_prob": -232.4544677734375, "reading_distances": [36, 26, 21, 28], "reading_log_probs": [-235.66043090820312, -238.0565185546875, -234.0395050048828, -236.82235717773438], "reading_logprob_std": 1.7108118888688142, "tokenization_path_logprob_std": 347.2182618368258, "trajectory_balance_rmse": 5722.390232733822} {"canonical_length": 8, "canonical_nll": 1.7646355628967285, "generation_rouge_l_f1": 0.04152249134948097, "generation_token_f1": 0.04844290657439446, "prompt_id": "b1a346c0437945a9", "reading_canonical_log_prob": -14.117084503173828, "reading_distances": [28, 31, 24, 31], "reading_log_probs": [-13.673637390136719, -12.716997146606445, -13.278036117553711, -14.26661491394043], "reading_logprob_std": 0.6529836173654154, "tokenization_path_logprob_std": 63.16582151284347, "trajectory_balance_rmse": 205.70917260918281} {"canonical_length": 21, "canonical_nll": 2.499814896356492, "generation_rouge_l_f1": 0.15060240963855423, "generation_token_f1": 0.16265060240963855, "prompt_id": "c2ce9d8e95420539", "reading_canonical_log_prob": -52.49611282348633, "reading_distances": [19, 16, 19, 22], "reading_log_probs": [-66.44365692138672, -62.76315689086914, -60.48361587524414, -65.4840316772461], "reading_logprob_std": 2.7017382582562393, "tokenization_path_logprob_std": 63.937662515491084, "trajectory_balance_rmse": 435.2932444899014} {"canonical_length": 86, "canonical_nll": 1.4003168150436047, "generation_rouge_l_f1": 0.19373219373219375, "generation_token_f1": 0.3247863247863248, "prompt_id": "ecdc1eebf63862d2", "reading_canonical_log_prob": -120.42724609375, "reading_distances": [32, 35, 28, 36], "reading_log_probs": [-124.21062469482422, -122.31002807617188, -124.42948150634766, -124.46199035644531], "reading_logprob_std": 1.034707359206914, "tokenization_path_logprob_std": 111.02201165394622, "trajectory_balance_rmse": 2701.139467819517} {"canonical_length": 149, "canonical_nll": 1.0268730829226091, "generation_rouge_l_f1": 0.2992125984251969, "generation_token_f1": 0.46719160104986884, "prompt_id": "a473b04dd30c4a84", "reading_canonical_log_prob": -153.00408935546875, "reading_distances": [50, 50, 49, 47], "reading_log_probs": [-159.4148406982422, -161.10238647460938, -160.05929565429688, -164.48696899414062], "reading_logprob_std": 2.257161692684561, "tokenization_path_logprob_std": 146.5570623891822, "trajectory_balance_rmse": 4720.421471178889} {"canonical_length": 147, "canonical_nll": 1.5943892083200468, "generation_rouge_l_f1": 0.21848739495798322, "generation_token_f1": 0.39215686274509803, "prompt_id": "30cb3266dd228382", "reading_canonical_log_prob": -234.37521362304688, "reading_distances": [17, 20, 16, 12], "reading_log_probs": [-236.77267456054688, -234.1385040283203, -235.3594970703125, -236.53872680664062], "reading_logprob_std": 1.212199541072453, "tokenization_path_logprob_std": 301.039383283929, "trajectory_balance_rmse": 3240.9436318361486} {"canonical_length": 151, "canonical_nll": 1.1380068545309914, "generation_rouge_l_f1": 0.24671916010498687, "generation_token_f1": 0.4881889763779527, "prompt_id": "d49ee101a0892105", "reading_canonical_log_prob": -171.8390350341797, "reading_distances": [21, 17, 24, 23], "reading_log_probs": [-170.78445434570312, -171.23443603515625, -169.92218017578125, -171.3087615966797], "reading_logprob_std": 0.6371189138565285, "tokenization_path_logprob_std": 222.1567580091107, "trajectory_balance_rmse": 3391.030447327022} {"canonical_length": 18, "canonical_nll": 0.9522167841593424, "generation_rouge_l_f1": 0.18487394957983194, "generation_token_f1": 0.21848739495798317, "prompt_id": "c17572e82dc23485", "reading_canonical_log_prob": -17.139902114868164, "reading_distances": [26, 29, 24, 21], "reading_log_probs": [-17.151958465576172, -16.912269592285156, -22.39234733581543, -16.842235565185547], "reading_logprob_std": 2.7150038681511024, "tokenization_path_logprob_std": 80.64897544366494, "trajectory_balance_rmse": 656.0536584092544} {"canonical_length": 267, "canonical_nll": 1.371514938297343, "generation_rouge_l_f1": 0.22350674373795762, "generation_token_f1": 0.5356454720616571, "prompt_id": "50c4d271f8ba07ce", "reading_canonical_log_prob": -366.1944885253906, "reading_distances": [13, 16, 12, 13], "reading_log_probs": [-367.9200439453125, -366.6336975097656, -368.27239990234375, -368.90777587890625], "reading_logprob_std": 0.9580769296429051, "tokenization_path_logprob_std": 263.1287517148757, "trajectory_balance_rmse": 7358.227013198055} {"canonical_length": 27, "canonical_nll": 2.201921674940321, "generation_rouge_l_f1": 0.21428571428571427, "generation_token_f1": 0.21428571428571427, "prompt_id": "71a578aa99112e47", "reading_canonical_log_prob": -59.45188522338867, "reading_distances": [10, 13, 11, 16], "reading_log_probs": [-60.23261260986328, -59.69001007080078, -60.15245056152344, -61.08103561401367], "reading_logprob_std": 0.5796362747736367, "tokenization_path_logprob_std": 69.95767909463589, "trajectory_balance_rmse": 1020.5345857641476} {"canonical_length": 260, "canonical_nll": 1.1917167076697717, "generation_rouge_l_f1": 0.2314990512333966, "generation_token_f1": 0.5199240986717268, "prompt_id": "c22b91c04ec53dee", "reading_canonical_log_prob": -309.8463439941406, "reading_distances": [38, 39, 38, 43], "reading_log_probs": [-297.8944396972656, -296.046630859375, -302.464599609375, -304.217041015625], "reading_logprob_std": 3.8220088787859887, "tokenization_path_logprob_std": 276.79298835514254, "trajectory_balance_rmse": 7621.144021855482} {"canonical_length": 42, "canonical_nll": 1.9237405686151414, "generation_rouge_l_f1": 0.12056737588652483, "generation_token_f1": 0.14184397163120568, "prompt_id": "ce56aa7776fbfa4a", "reading_canonical_log_prob": -80.79710388183594, "reading_distances": [18, 19, 16, 16], "reading_log_probs": [-92.9992446899414, -89.54104614257812, -82.69798278808594, -86.9524154663086], "reading_logprob_std": 4.342311211484718, "tokenization_path_logprob_std": 85.55190528073267, "trajectory_balance_rmse": 1079.6004514141418} {"canonical_length": 92, "canonical_nll": 1.1979192650836448, "generation_rouge_l_f1": 0.41237113402061853, "generation_token_f1": 0.5257731958762887, "prompt_id": "61cafca95080c624", "reading_canonical_log_prob": -110.20857238769531, "reading_distances": [28, 27, 23, 27], "reading_log_probs": [-112.80801391601562, -120.60188293457031, -114.90505981445312, -113.98414611816406], "reading_logprob_std": 3.459547164545816, "tokenization_path_logprob_std": 57.11889505304758, "trajectory_balance_rmse": 1157.9595364138618} {"canonical_length": 9, "canonical_nll": 2.305250803629557, "generation_rouge_l_f1": 0.588235294117647, "generation_token_f1": 0.588235294117647, "prompt_id": "33cbcf32b9a54c98", "reading_canonical_log_prob": -20.747257232666016, "reading_distances": [24, 22, 17, 21], "reading_log_probs": [-13.863359451293945, -13.401268005371094, -16.210742950439453, -17.385639190673828], "reading_logprob_std": 1.8990992764873242, "tokenization_path_logprob_std": 32.466885273609314, "trajectory_balance_rmse": 218.70510324150544} {"canonical_length": 13, "canonical_nll": 0.8914020978487455, "generation_rouge_l_f1": 0.787878787878788, "generation_token_f1": 0.787878787878788, "prompt_id": "5e0db48c41c73148", "reading_canonical_log_prob": -11.588227272033691, "reading_distances": [39, 37, 34, 28], "reading_log_probs": [-13.242795944213867, -12.87391471862793, -12.821950912475586, -12.319622993469238], "reading_logprob_std": 0.3794409361435154, "tokenization_path_logprob_std": 44.96337772731928, "trajectory_balance_rmse": 352.93308872272786} {"canonical_length": 45, "canonical_nll": 1.6786214192708333, "generation_rouge_l_f1": 0.14107883817427386, "generation_token_f1": 0.19087136929460582, "prompt_id": "c70b50c5f56eb753", "reading_canonical_log_prob": -75.5379638671875, "reading_distances": [26, 33, 32, 25], "reading_log_probs": [-100.2837142944336, -112.63587188720703, -111.07305908203125, -104.85604858398438], "reading_logprob_std": 5.711911917387979, "tokenization_path_logprob_std": 109.90560780537139, "trajectory_balance_rmse": 1110.8763111512935} {"canonical_length": 28, "canonical_nll": 1.057253292628697, "generation_rouge_l_f1": 0.25641025641025644, "generation_token_f1": 0.30769230769230765, "prompt_id": "a944937cb2c4cc95", "reading_canonical_log_prob": -29.603092193603516, "reading_distances": [16, 13, 12, 14], "reading_log_probs": [-27.393646240234375, -26.255718231201172, -28.833385467529297, -29.328432083129883], "reading_logprob_std": 1.3976806801183934, "tokenization_path_logprob_std": 69.42871151977856, "trajectory_balance_rmse": 884.3991626492354} {"canonical_length": 158, "canonical_nll": 1.462818387188489, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.4019607843137255, "prompt_id": "b725cba739a4a837", "reading_canonical_log_prob": -231.12530517578125, "reading_distances": [35, 37, 30, 31], "reading_log_probs": [-225.93356323242188, -229.02330017089844, -225.49981689453125, -228.79689025878906], "reading_logprob_std": 1.8545021908629047, "tokenization_path_logprob_std": 135.86343961232498, "trajectory_balance_rmse": 4088.5861910257013} {"canonical_length": 501, "canonical_nll": 0.720566922795035, "generation_rouge_l_f1": 0.2358346094946401, "generation_token_f1": 0.3950995405819296, "prompt_id": "c3741a95aac9caea", "reading_canonical_log_prob": -361.0040283203125, "reading_distances": [33, 35, 25, 30], "reading_log_probs": [-362.76580810546875, -367.21038818359375, -359.1617431640625, -364.97705078125], "reading_logprob_std": 3.4304938375009892, "tokenization_path_logprob_std": 151.63093783269147, "trajectory_balance_rmse": 7060.041822391915} {"canonical_length": 14, "canonical_nll": 0.5981057712009975, "generation_rouge_l_f1": 0.8461538461538461, "generation_token_f1": 0.8461538461538461, "prompt_id": "77d6fb3ebd2588d3", "reading_canonical_log_prob": -8.373480796813965, "reading_distances": [23, 18, 19, 26], "reading_log_probs": [-10.037578582763672, -10.97244644165039, -9.359664916992188, -9.968511581420898], "reading_logprob_std": 0.6657050656473473, "tokenization_path_logprob_std": 58.47259694101632, "trajectory_balance_rmse": 291.4102426597079} {"canonical_length": 242, "canonical_nll": 1.500351204359827, "generation_rouge_l_f1": 0.23750000000000002, "generation_token_f1": 0.38749999999999996, "prompt_id": "8d10c80bdd9627ab", "reading_canonical_log_prob": -363.0849914550781, "reading_distances": [16, 20, 18, 18], "reading_log_probs": [-368.3583984375, -368.3965148925781, -367.23626708984375, -366.1682434082031], "reading_logprob_std": 1.0610314963228338, "tokenization_path_logprob_std": 420.8254041585802, "trajectory_balance_rmse": 7308.91244018643} {"canonical_length": 233, "canonical_nll": 1.4399288324839057, "generation_rouge_l_f1": 0.20374220374220373, "generation_token_f1": 0.4158004158004158, "prompt_id": "a741547771fa6f03", "reading_canonical_log_prob": -335.50341796875, "reading_distances": [22, 28, 26, 20], "reading_log_probs": [-330.73931884765625, -330.99169921875, -332.1202392578125, -330.65972900390625], "reading_logprob_std": 0.6766257347556555, "tokenization_path_logprob_std": 216.11741674926088, "trajectory_balance_rmse": 6524.286951981516} {"canonical_length": 177, "canonical_nll": 0.9946123543432204, "generation_rouge_l_f1": 0.25821596244131456, "generation_token_f1": 0.44131455399061037, "prompt_id": "51a6d546f1967143", "reading_canonical_log_prob": -176.04638671875, "reading_distances": [27, 33, 28, 28], "reading_log_probs": [-174.36212158203125, -178.09805297851562, -178.00149536132812, -178.02513122558594], "reading_logprob_std": 1.8401780747845309, "tokenization_path_logprob_std": 328.53794924412455, "trajectory_balance_rmse": 4291.056085357261} {"canonical_length": 163, "canonical_nll": 1.2371356238616755, "generation_rouge_l_f1": 0.2638888888888889, "generation_token_f1": 0.5185185185185186, "prompt_id": "715a0325621f71c4", "reading_canonical_log_prob": -201.65310668945312, "reading_distances": [46, 42, 32, 36], "reading_log_probs": [-188.54847717285156, -191.39950561523438, -196.759765625, -195.5458526611328], "reading_logprob_std": 3.785003862050313, "tokenization_path_logprob_std": 260.24990880410456, "trajectory_balance_rmse": 3726.604108885754} {"canonical_length": 264, "canonical_nll": 1.3980993791060015, "generation_rouge_l_f1": 0.3107569721115538, "generation_token_f1": 0.44223107569721115, "prompt_id": "0857b1f32ec14a13", "reading_canonical_log_prob": -369.0982360839844, "reading_distances": [17, 23, 21, 17], "reading_log_probs": [-383.267822265625, -382.6848449707031, -377.9938049316406, -378.9693298339844], "reading_logprob_std": 2.6362017620811997, "tokenization_path_logprob_std": 93.1637655236118, "trajectory_balance_rmse": 5928.571653174937} {"canonical_length": 2, "canonical_nll": 4.568577289581299, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "7c6d4e136a035811", "reading_canonical_log_prob": -9.137154579162598, "reading_distances": [65, 71, 83, 67], "reading_log_probs": [-12.90255355834961, -11.709558486938477, -10.529441833496094, -12.301719665527344], "reading_logprob_std": 1.0124309341415851, "tokenization_path_logprob_std": 0.0, "trajectory_balance_rmse": 9.203330039978027} {"canonical_length": 13, "canonical_nll": 1.3507088881272535, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "d3b27b0b1b980505", "reading_canonical_log_prob": -17.559215545654297, "reading_distances": [31, 27, 34, 37], "reading_log_probs": [-19.080636978149414, -19.21405601501465, -19.302860260009766, -18.02969741821289], "reading_logprob_std": 0.5918328090571752, "tokenization_path_logprob_std": 41.464467526047635, "trajectory_balance_rmse": 305.29697599441204} {"canonical_length": 131, "canonical_nll": 0.5930517648012583, "generation_rouge_l_f1": 0.3086053412462908, "generation_token_f1": 0.45103857566765576, "prompt_id": "37ec341c729bc100", "reading_canonical_log_prob": -77.68978118896484, "reading_distances": [9, 8, 14, 7], "reading_log_probs": [-71.38346862792969, -77.05773162841797, -79.21101379394531, -73.48585510253906], "reading_logprob_std": 3.5125978142446166, "tokenization_path_logprob_std": 149.68180123889857, "trajectory_balance_rmse": 1884.008437961911} {"canonical_length": 12, "canonical_nll": 4.204797744750977, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "b0f40cc1f3136930", "reading_canonical_log_prob": -50.45757293701172, "reading_distances": [10, 12, 9, 13], "reading_log_probs": [-51.40790557861328, -50.368404388427734, -51.85115051269531, -54.70328903198242], "reading_logprob_std": 1.8542976203018156, "tokenization_path_logprob_std": 60.15321479031332, "trajectory_balance_rmse": 371.53276521502585} {"canonical_length": 264, "canonical_nll": 1.3340278394294507, "generation_rouge_l_f1": 0.28460038986354774, "generation_token_f1": 0.5575048732943468, "prompt_id": "dc37089a4982f612", "reading_canonical_log_prob": -352.183349609375, "reading_distances": [16, 16, 14, 14], "reading_log_probs": [-364.5092468261719, -366.19384765625, -356.38995361328125, -355.0579833984375], "reading_logprob_std": 5.62720429729124, "tokenization_path_logprob_std": 342.84829984406747, "trajectory_balance_rmse": 7266.2121666448} {"canonical_length": 3, "canonical_nll": 1.5098930994669597, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "9ba6dc4bdecfc908", "reading_canonical_log_prob": -4.529679298400879, "reading_distances": [24, 25, 31, 15], "reading_log_probs": [-4.461397647857666, -4.821296691894531, -4.013057708740234, -3.8750431537628174], "reading_logprob_std": 0.43224623049777144, "tokenization_path_logprob_std": 17.187684297561646, "trajectory_balance_rmse": 19.425751254921792} {"canonical_length": 261, "canonical_nll": 1.0437066673775743, "generation_rouge_l_f1": 0.5078125, "generation_token_f1": 0.7226562500000001, "prompt_id": "5cf08716fd03e235", "reading_canonical_log_prob": -272.4074401855469, "reading_distances": [30, 30, 30, 31], "reading_log_probs": [-270.86639404296875, -270.27294921875, -270.0768737792969, -274.1243896484375], "reading_logprob_std": 1.8895431115247494, "tokenization_path_logprob_std": 144.88099637676447, "trajectory_balance_rmse": 5863.407751714564} {"canonical_length": 102, "canonical_nll": 0.9486220864688649, "generation_rouge_l_f1": 0.2574850299401198, "generation_token_f1": 0.4251497005988024, "prompt_id": "8cfc1971edf74b1e", "reading_canonical_log_prob": -96.75945281982422, "reading_distances": [38, 31, 32, 39], "reading_log_probs": [-95.7044677734375, -102.20841979980469, -94.42713165283203, -97.6245346069336], "reading_logprob_std": 3.408365572503415, "tokenization_path_logprob_std": 113.51704996616586, "trajectory_balance_rmse": 2578.599896447856} {"canonical_length": 42, "canonical_nll": 3.386619204566592, "generation_rouge_l_f1": 0.3617021276595745, "generation_token_f1": 0.4893617021276596, "prompt_id": "73cb687de2d9ad9d", "reading_canonical_log_prob": -142.23800659179688, "reading_distances": [24, 27, 23, 23], "reading_log_probs": [-140.66905212402344, -142.2307586669922, -141.69235229492188, -141.18824768066406], "reading_logprob_std": 0.6699792678610982, "tokenization_path_logprob_std": 61.95369693026898, "trajectory_balance_rmse": 1059.3200919691972} {"canonical_length": 50, "canonical_nll": 1.3182969665527344, "generation_rouge_l_f1": 0.34196891191709844, "generation_token_f1": 0.4145077720207254, "prompt_id": "5b7c88eabb38a156", "reading_canonical_log_prob": -65.91484832763672, "reading_distances": [29, 26, 30, 26], "reading_log_probs": [-66.28300476074219, -66.70042419433594, -65.68106842041016, -65.09162902832031], "reading_logprob_std": 0.7030107487404987, "tokenization_path_logprob_std": 149.0728946209293, "trajectory_balance_rmse": 1341.1856041761382} {"canonical_length": 297, "canonical_nll": 2.002240832807239, "generation_rouge_l_f1": 0.19424460431654678, "generation_token_f1": 0.37769784172661874, "prompt_id": "7564700da05daae9", "reading_canonical_log_prob": -594.66552734375, "reading_distances": [25, 22, 21, 30], "reading_log_probs": [-599.5916748046875, -603.981201171875, -600.0426635742188, -598.4886474609375], "reading_logprob_std": 2.3941455099812528, "tokenization_path_logprob_std": 380.4567935768871, "trajectory_balance_rmse": 8016.749866499582} {"canonical_length": 240, "canonical_nll": 1.5002432505289713, "generation_rouge_l_f1": 0.21386138613861386, "generation_token_f1": 0.499009900990099, "prompt_id": "ca2d7d9e3b473a6f", "reading_canonical_log_prob": -360.0583801269531, "reading_distances": [17, 14, 21, 13], "reading_log_probs": [-365.19085693359375, -366.21746826171875, -366.6143493652344, -365.67816162109375], "reading_logprob_std": 0.6219955208615053, "tokenization_path_logprob_std": 160.09558109485008, "trajectory_balance_rmse": 7683.967920234767} {"canonical_length": 282, "canonical_nll": 1.3431020966658356, "generation_rouge_l_f1": 0.2748091603053435, "generation_token_f1": 0.4122137404580153, "prompt_id": "9e92f8d560ad3563", "reading_canonical_log_prob": -378.7547912597656, "reading_distances": [22, 25, 24, 15], "reading_log_probs": [-384.9652099609375, -385.70556640625, -392.1661071777344, -397.9388427734375], "reading_logprob_std": 6.092565898566216, "tokenization_path_logprob_std": 254.77221863858955, "trajectory_balance_rmse": 5682.129813868903} {"canonical_length": 16, "canonical_nll": 2.096230983734131, "generation_rouge_l_f1": 0.2580645161290323, "generation_token_f1": 0.2580645161290323, "prompt_id": "8efbf41d41230763", "reading_canonical_log_prob": -33.539695739746094, "reading_distances": [37, 36, 42, 28], "reading_log_probs": [-31.345090866088867, -31.480558395385742, -32.498687744140625, -32.788360595703125], "reading_logprob_std": 0.7224385237887463, "tokenization_path_logprob_std": 49.161686599110126, "trajectory_balance_rmse": 499.02923773112343} {"canonical_length": 39, "canonical_nll": 1.7723353459284856, "generation_rouge_l_f1": 0.37037037037037035, "generation_token_f1": 0.4444444444444445, "prompt_id": "3f9b43f9811260d6", "reading_canonical_log_prob": -69.12107849121094, "reading_distances": [32, 31, 28, 35], "reading_log_probs": [-69.64512634277344, -72.46505737304688, -70.60104370117188, -70.19026184082031], "reading_logprob_std": 1.2240964403042356, "tokenization_path_logprob_std": 91.40334973517909, "trajectory_balance_rmse": 901.8176839997265} {"canonical_length": 281, "canonical_nll": 1.0830142201053714, "generation_rouge_l_f1": 0.2039355992844365, "generation_token_f1": 0.4150268336314848, "prompt_id": "9e7206bd2d909c4b", "reading_canonical_log_prob": -304.3269958496094, "reading_distances": [28, 37, 33, 32], "reading_log_probs": [-305.2725830078125, -310.1572570800781, -306.872314453125, -308.42840576171875], "reading_logprob_std": 2.0932350132332203, "tokenization_path_logprob_std": 64.20884657603514, "trajectory_balance_rmse": 5981.343082514386} {"canonical_length": 193, "canonical_nll": 1.4974838019652688, "generation_rouge_l_f1": 0.29247311827956984, "generation_token_f1": 0.4903225806451613, "prompt_id": "59cc1c6a894af7a8", "reading_canonical_log_prob": -289.0143737792969, "reading_distances": [20, 23, 27, 23], "reading_log_probs": [-287.77862548828125, -284.828857421875, -284.6187744140625, -287.8118591308594], "reading_logprob_std": 1.775413543526491, "tokenization_path_logprob_std": 248.01133671139857, "trajectory_balance_rmse": 5086.918485793896} {"canonical_length": 85, "canonical_nll": 1.2191111845128677, "generation_rouge_l_f1": 0.2620320855614973, "generation_token_f1": 0.36898395721925137, "prompt_id": "bb9e06240071dbbd", "reading_canonical_log_prob": -103.62445068359375, "reading_distances": [25, 25, 20, 24], "reading_log_probs": [-103.68146514892578, -103.30980682373047, -105.63966369628906, -106.013671875], "reading_logprob_std": 1.362927640979656, "tokenization_path_logprob_std": 92.07061648506799, "trajectory_balance_rmse": 2314.402613076219} {"canonical_length": 59, "canonical_nll": 0.9696010007696637, "generation_rouge_l_f1": 0.23076923076923078, "generation_token_f1": 0.2991452991452992, "prompt_id": "fbd2f42b1f0643a2", "reading_canonical_log_prob": -57.206459045410156, "reading_distances": [17, 17, 13, 17], "reading_log_probs": [-55.77810287475586, -54.69649124145508, -58.615360260009766, -58.09430694580078], "reading_logprob_std": 1.8654511741070559, "tokenization_path_logprob_std": 51.63718475836173, "trajectory_balance_rmse": 1334.0078798481284} {"canonical_length": 73, "canonical_nll": 0.9984292853368472, "generation_rouge_l_f1": 0.2619808306709265, "generation_token_f1": 0.3322683706070288, "prompt_id": "1a0ff319dcddb777", "reading_canonical_log_prob": -72.88533782958984, "reading_distances": [48, 48, 55, 41], "reading_log_probs": [-74.18575286865234, -73.22503662109375, -72.9202880859375, -71.60697174072266], "reading_logprob_std": 1.0649841305452858, "tokenization_path_logprob_std": 142.6238827608123, "trajectory_balance_rmse": 919.6843598068457} {"canonical_length": 15, "canonical_nll": 0.6598589579264323, "generation_rouge_l_f1": 0.41666666666666663, "generation_token_f1": 0.4999999999999999, "prompt_id": "9f8d22c249bf7888", "reading_canonical_log_prob": -9.897884368896484, "reading_distances": [20, 21, 24, 21], "reading_log_probs": [-8.998525619506836, -8.891656875610352, -8.879876136779785, -9.078511238098145], "reading_logprob_std": 0.09416544368765355, "tokenization_path_logprob_std": 63.73632520815566, "trajectory_balance_rmse": 466.56225832960894} {"canonical_length": 168, "canonical_nll": 1.168555668422154, "generation_rouge_l_f1": 0.2227979274611399, "generation_token_f1": 0.3626943005181347, "prompt_id": "9edad123b1e31f1d", "reading_canonical_log_prob": -196.31735229492188, "reading_distances": [28, 29, 28, 31], "reading_log_probs": [-202.23597717285156, -202.7459716796875, -209.07135009765625, -202.26260375976562], "reading_logprob_std": 3.3364927913846394, "tokenization_path_logprob_std": 295.11689101397207, "trajectory_balance_rmse": 4080.5342856796465} {"canonical_length": 82, "canonical_nll": 1.3274591957650534, "generation_rouge_l_f1": 0.17699115044247787, "generation_token_f1": 0.2713864306784661, "prompt_id": "e027ef71856c6f38", "reading_canonical_log_prob": -108.85165405273438, "reading_distances": [29, 27, 25, 26], "reading_log_probs": [-114.4966049194336, -113.17857360839844, -123.32293701171875, -111.70443725585938], "reading_logprob_std": 5.224208619678578, "tokenization_path_logprob_std": 115.38248397243046, "trajectory_balance_rmse": 2907.8298262538465} {"canonical_length": 27, "canonical_nll": 1.9715337400083188, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.23684210526315788, "prompt_id": "59a16e60dba35dc9", "reading_canonical_log_prob": -53.23141098022461, "reading_distances": [15, 14, 18, 15], "reading_log_probs": [-54.412353515625, -53.96113967895508, -54.420989990234375, -54.21183776855469], "reading_logprob_std": 0.21639692940392044, "tokenization_path_logprob_std": 97.89916432113814, "trajectory_balance_rmse": 865.3720610921296} {"canonical_length": 16, "canonical_nll": 0.5799188017845154, "generation_rouge_l_f1": 0.375, "generation_token_f1": 0.375, "prompt_id": "ca87800b07a107f6", "reading_canonical_log_prob": -9.278700828552246, "reading_distances": [27, 32, 30, 26], "reading_log_probs": [-9.35546588897705, -9.171019554138184, -10.035873413085938, -9.28689193725586], "reading_logprob_std": 0.3898770205458949, "tokenization_path_logprob_std": 15.58557120322133, "trajectory_balance_rmse": 305.79119502782515} {"canonical_length": 100, "canonical_nll": 1.4760662841796874, "generation_rouge_l_f1": 0.24641833810888247, "generation_token_f1": 0.37822349570200575, "prompt_id": "911e3e27940b6a5a", "reading_canonical_log_prob": -147.60662841796875, "reading_distances": [29, 27, 23, 30], "reading_log_probs": [-144.32901000976562, -144.9940185546875, -146.8046875, -141.8467559814453], "reading_logprob_std": 2.051374927918525, "tokenization_path_logprob_std": 247.80292425751324, "trajectory_balance_rmse": 2763.8764608249494} {"canonical_length": 190, "canonical_nll": 2.2231006823087993, "generation_rouge_l_f1": 0.20568927789934358, "generation_token_f1": 0.4201312910284464, "prompt_id": "1e23f4d5f326fadb", "reading_canonical_log_prob": -422.3891296386719, "reading_distances": [23, 25, 31, 29], "reading_log_probs": [-423.67681884765625, -426.26995849609375, -422.2052917480469, -426.60589599609375], "reading_logprob_std": 2.1108636800133596, "tokenization_path_logprob_std": 264.76542047402853, "trajectory_balance_rmse": 5469.04092743999} {"canonical_length": 151, "canonical_nll": 1.5276871232796978, "generation_rouge_l_f1": 0.1954022988505747, "generation_token_f1": 0.5172413793103449, "prompt_id": "75c88d19daf68f00", "reading_canonical_log_prob": -230.68075561523438, "reading_distances": [18, 19, 24, 20], "reading_log_probs": [-230.2217254638672, -233.2774658203125, -229.14039611816406, -232.20787048339844], "reading_logprob_std": 1.8735076995893951, "tokenization_path_logprob_std": 282.15204308029985, "trajectory_balance_rmse": 5222.286015398893} {"canonical_length": 165, "canonical_nll": 1.4781292539654356, "generation_rouge_l_f1": 0.23404255319148937, "generation_token_f1": 0.46099290780141844, "prompt_id": "0efac862acbbc3cf", "reading_canonical_log_prob": -243.89132690429688, "reading_distances": [28, 33, 24, 31], "reading_log_probs": [-243.51675415039062, -244.70704650878906, -243.51275634765625, -244.10641479492188], "reading_logprob_std": 0.5703821345736071, "tokenization_path_logprob_std": 146.88988429539006, "trajectory_balance_rmse": 5338.298604154463} {"canonical_length": 225, "canonical_nll": 1.021934814453125, "generation_rouge_l_f1": 0.23834196891191708, "generation_token_f1": 0.4196891191709845, "prompt_id": "867d14d7a7b70c9c", "reading_canonical_log_prob": -229.93533325195312, "reading_distances": [14, 16, 20, 20], "reading_log_probs": [-236.24530029296875, -229.31607055664062, -223.3035888671875, -230.43930053710938], "reading_logprob_std": 5.3036285945201564, "tokenization_path_logprob_std": 76.43511304042184, "trajectory_balance_rmse": 4874.576087785215} {"canonical_length": 263, "canonical_nll": 1.666250018565827, "generation_rouge_l_f1": 0.22348484848484848, "generation_token_f1": 0.4242424242424242, "prompt_id": "67b31fbc0903ce67", "reading_canonical_log_prob": -438.2237548828125, "reading_distances": [27, 20, 20, 14], "reading_log_probs": [-446.2166748046875, -443.5705261230469, -443.1123046875, -443.921142578125], "reading_logprob_std": 1.3812977710863985, "tokenization_path_logprob_std": 396.160942283458, "trajectory_balance_rmse": 7441.176245139736} {"canonical_length": 38, "canonical_nll": 0.6920829070241827, "generation_rouge_l_f1": 0.48192771084337344, "generation_token_f1": 0.48192771084337344, "prompt_id": "c50601d0ada030f4", "reading_canonical_log_prob": -26.299150466918945, "reading_distances": [28, 26, 24, 26], "reading_log_probs": [-25.276988983154297, -27.062061309814453, -26.85540008544922, -28.435958862304688], "reading_logprob_std": 1.2937485880869866, "tokenization_path_logprob_std": 72.86295692908612, "trajectory_balance_rmse": 344.9838332957605} {"canonical_length": 35, "canonical_nll": 0.5365291050502232, "generation_rouge_l_f1": 0.2097902097902098, "generation_token_f1": 0.27972027972027974, "prompt_id": "d0d4fbc6a94b1aa0", "reading_canonical_log_prob": -18.778518676757812, "reading_distances": [21, 19, 20, 21], "reading_log_probs": [-17.826677322387695, -19.09792709350586, -17.741376876831055, -17.61951446533203], "reading_logprob_std": 0.6896286097581965, "tokenization_path_logprob_std": 92.65870589979174, "trajectory_balance_rmse": 760.7396536564743} {"canonical_length": 74, "canonical_nll": 0.46773075413059545, "generation_rouge_l_f1": 0.2760942760942761, "generation_token_f1": 0.30976430976430974, "prompt_id": "a83eef42db047836", "reading_canonical_log_prob": -34.61207580566406, "reading_distances": [20, 25, 26, 29], "reading_log_probs": [-33.916587829589844, -39.50990295410156, -33.68746566772461, -33.21343231201172], "reading_logprob_std": 2.9665232856531767, "tokenization_path_logprob_std": 126.2576588235165, "trajectory_balance_rmse": 1507.351328373771} {"canonical_length": 107, "canonical_nll": 1.2797389520662967, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.4722222222222222, "prompt_id": "8e875ab98664b402", "reading_canonical_log_prob": -136.93206787109375, "reading_distances": [18, 17, 18, 18], "reading_log_probs": [-135.3494415283203, -137.187255859375, -133.0866241455078, -134.52615356445312], "reading_logprob_std": 1.71134931948589, "tokenization_path_logprob_std": 108.2304213228403, "trajectory_balance_rmse": 2824.3861749430907} {"canonical_length": 31, "canonical_nll": 0.8311244595435358, "generation_rouge_l_f1": 0.35398230088495575, "generation_token_f1": 0.42477876106194684, "prompt_id": "54f5b26b0d406b0a", "reading_canonical_log_prob": -25.76485824584961, "reading_distances": [29, 31, 27, 34], "reading_log_probs": [-31.78421974182129, -31.810810089111328, -29.478940963745117, -28.702835083007812], "reading_logprob_std": 1.5945064884707476, "tokenization_path_logprob_std": 93.92964820631096, "trajectory_balance_rmse": 847.9049871395104} {"canonical_length": 262, "canonical_nll": 1.4165668050751432, "generation_rouge_l_f1": 0.23463687150837992, "generation_token_f1": 0.5102420856610801, "prompt_id": "652a0c6a273ec7b9", "reading_canonical_log_prob": -371.1405029296875, "reading_distances": [16, 17, 13, 13], "reading_log_probs": [-374.9752197265625, -371.4667053222656, -372.773193359375, -373.33837890625], "reading_logprob_std": 1.4539414539015474, "tokenization_path_logprob_std": 281.96117912451876, "trajectory_balance_rmse": 5842.740110531471} {"canonical_length": 304, "canonical_nll": 1.7083945023386102, "generation_rouge_l_f1": 0.15942028985507248, "generation_token_f1": 0.36231884057971014, "prompt_id": "13a435b70dd47ea2", "reading_canonical_log_prob": -519.3519287109375, "reading_distances": [23, 14, 17, 11], "reading_log_probs": [-519.0323486328125, -519.06103515625, -519.443115234375, -517.341552734375], "reading_logprob_std": 0.9375283468782237, "tokenization_path_logprob_std": 647.1319480072538, "trajectory_balance_rmse": 9392.533351594087} {"canonical_length": 15, "canonical_nll": 0.4499899546305339, "generation_rouge_l_f1": 0.11842105263157894, "generation_token_f1": 0.14473684210526316, "prompt_id": "1f8b04daa461e185", "reading_canonical_log_prob": -6.749849319458008, "reading_distances": [24, 18, 25, 23], "reading_log_probs": [-5.614482879638672, -5.526101112365723, -4.015684127807617, -4.874772071838379], "reading_logprob_std": 0.7390738543764453, "tokenization_path_logprob_std": 27.076549366163327, "trajectory_balance_rmse": 290.3409599213985} {"canonical_length": 18, "canonical_nll": 0.19395609696706137, "generation_rouge_l_f1": 0.3516483516483517, "generation_token_f1": 0.3516483516483517, "prompt_id": "713c6185b310a4ca", "reading_canonical_log_prob": -3.4912097454071045, "reading_distances": [14, 27, 24, 30], "reading_log_probs": [-6.054012298583984, -16.270830154418945, -10.41731071472168, -6.097369194030762], "reading_logprob_std": 4.829152034550219, "tokenization_path_logprob_std": 92.3278391321855, "trajectory_balance_rmse": 516.4584757808225} {"canonical_length": 227, "canonical_nll": 0.9261972704647922, "generation_rouge_l_f1": 0.22978723404255322, "generation_token_f1": 0.49787234042553186, "prompt_id": "a7de6528410fa4b7", "reading_canonical_log_prob": -210.2467803955078, "reading_distances": [21, 28, 25, 25], "reading_log_probs": [-215.9215545654297, -213.2548370361328, -213.7742919921875, -212.1774444580078], "reading_logprob_std": 1.5737696173569182, "tokenization_path_logprob_std": 150.2569840360326, "trajectory_balance_rmse": 4439.208353245612} {"canonical_length": 154, "canonical_nll": 1.8087689288250812, "generation_rouge_l_f1": 0.23076923076923078, "generation_token_f1": 0.4182692307692308, "prompt_id": "b7245c913ba925d6", "reading_canonical_log_prob": -278.5504150390625, "reading_distances": [22, 28, 21, 28], "reading_log_probs": [-280.0207214355469, -282.81292724609375, -282.67877197265625, -287.94427490234375], "reading_logprob_std": 3.313087595641326, "tokenization_path_logprob_std": 179.7839915685107, "trajectory_balance_rmse": 3879.4267776866277} {"canonical_length": 9, "canonical_nll": 0.728341261545817, "generation_rouge_l_f1": 0.8750000000000001, "generation_token_f1": 0.8750000000000001, "prompt_id": "402312ec99016f32", "reading_canonical_log_prob": -6.5550713539123535, "reading_distances": [23, 21, 22, 20], "reading_log_probs": [-7.5443034172058105, -5.93550443649292, -5.501809120178223, -5.414986610412598], "reading_logprob_std": 0.9899737269362165, "tokenization_path_logprob_std": 50.7466365734214, "trajectory_balance_rmse": 219.91271416644958} {"canonical_length": 290, "canonical_nll": 1.5141114333580281, "generation_rouge_l_f1": 0.2811387900355872, "generation_token_f1": 0.5444839857651246, "prompt_id": "083ca3a9f1344c38", "reading_canonical_log_prob": -439.0923156738281, "reading_distances": [30, 35, 32, 31], "reading_log_probs": [-444.3560791015625, -446.63214111328125, -445.57366943359375, -446.42449951171875], "reading_logprob_std": 1.0339542480302757, "tokenization_path_logprob_std": 338.12460760210155, "trajectory_balance_rmse": 7937.725865150179} {"canonical_length": 58, "canonical_nll": 1.8807288860452587, "generation_rouge_l_f1": 0.15966386554621848, "generation_token_f1": 0.2773109243697479, "prompt_id": "c06dd3bfbe8f1823", "reading_canonical_log_prob": -109.082275390625, "reading_distances": [35, 29, 32, 34], "reading_log_probs": [-112.10086059570312, -114.46334838867188, -113.57261657714844, -114.98221588134766], "reading_logprob_std": 1.26158694231949, "tokenization_path_logprob_std": 219.56543975076906, "trajectory_balance_rmse": 1577.1709853476489} {"canonical_length": 117, "canonical_nll": 0.9347431150257078, "generation_rouge_l_f1": 0.22589531680440775, "generation_token_f1": 0.3856749311294766, "prompt_id": "22135eb61f4e8a41", "reading_canonical_log_prob": -109.36494445800781, "reading_distances": [23, 16, 22, 21], "reading_log_probs": [-128.32867431640625, -117.92112731933594, -120.29826354980469, -124.0872802734375], "reading_logprob_std": 4.55359584694576, "tokenization_path_logprob_std": 234.9351909719673, "trajectory_balance_rmse": 3191.1944309067653} {"canonical_length": 21, "canonical_nll": 1.6383470807756697, "generation_rouge_l_f1": 0.5000000000000001, "generation_token_f1": 0.5000000000000001, "prompt_id": "b36f1ed6b55c0058", "reading_canonical_log_prob": -34.40528869628906, "reading_distances": [27, 26, 24, 28], "reading_log_probs": [-44.146202087402344, -45.36219787597656, -40.61002731323242, -43.266845703125], "reading_logprob_std": 2.0163654427466016, "tokenization_path_logprob_std": 103.12276691558961, "trajectory_balance_rmse": 597.3525111989301} {"canonical_length": 25, "canonical_nll": 1.7219673156738282, "generation_rouge_l_f1": 0.375, "generation_token_f1": 0.4583333333333333, "prompt_id": "017529d62a606652", "reading_canonical_log_prob": -43.0491828918457, "reading_distances": [25, 19, 26, 27], "reading_log_probs": [-41.55651092529297, -41.432029724121094, -44.13275909423828, -44.283935546875], "reading_logprob_std": 1.5690112474674986, "tokenization_path_logprob_std": 142.77519993920578, "trajectory_balance_rmse": 696.6445947636424} {"canonical_length": 231, "canonical_nll": 1.774810923126353, "generation_rouge_l_f1": 0.133587786259542, "generation_token_f1": 0.3244274809160305, "prompt_id": "47cd1ce3b4c7133a", "reading_canonical_log_prob": -409.9813232421875, "reading_distances": [15, 12, 12, 9], "reading_log_probs": [-410.16162109375, -407.758056640625, -408.26385498046875, -413.0794677734375], "reading_logprob_std": 2.409284861461693, "tokenization_path_logprob_std": 166.20912688306464, "trajectory_balance_rmse": 7050.293272188724} {"canonical_length": 34, "canonical_nll": 0.7728038675644818, "generation_rouge_l_f1": 0.15384615384615383, "generation_token_f1": 0.19047619047619047, "prompt_id": "549cddbffb2c458b", "reading_canonical_log_prob": -26.275331497192383, "reading_distances": [11, 13, 17, 10], "reading_log_probs": [-27.88546371459961, -28.58652687072754, -29.216581344604492, -28.273456573486328], "reading_logprob_std": 0.5626105918914012, "tokenization_path_logprob_std": 130.2319577332543, "trajectory_balance_rmse": 610.7685183896025} {"canonical_length": 356, "canonical_nll": 1.4674024260445926, "generation_rouge_l_f1": 0.23247863247863249, "generation_token_f1": 0.4478632478632478, "prompt_id": "6b4ba85e93ef4f52", "reading_canonical_log_prob": -522.395263671875, "reading_distances": [29, 24, 24, 28], "reading_log_probs": [-521.4708251953125, -517.562255859375, -520.4067993164062, -521.1810913085938], "reading_logprob_std": 1.7860556315186438, "tokenization_path_logprob_std": 383.90434900397184, "trajectory_balance_rmse": 9978.082093767849} {"canonical_length": 420, "canonical_nll": 1.3252825055803572, "generation_rouge_l_f1": 0.2571428571428571, "generation_token_f1": 0.4380952380952381, "prompt_id": "71f47b6572a134fd", "reading_canonical_log_prob": -556.61865234375, "reading_distances": [28, 37, 36, 36], "reading_log_probs": [-570.36962890625, -553.9730224609375, -572.7635498046875, -556.6444091796875], "reading_logprob_std": 9.500035684488749, "tokenization_path_logprob_std": 347.0278271896341, "trajectory_balance_rmse": 9885.877943594282} {"canonical_length": 144, "canonical_nll": 1.4755240546332464, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.5311653116531165, "prompt_id": "c1009daeb59a17f6", "reading_canonical_log_prob": -212.4754638671875, "reading_distances": [18, 17, 20, 19], "reading_log_probs": [-220.19630432128906, -219.32980346679688, -224.73822021484375, -222.3174591064453], "reading_logprob_std": 2.413784435679843, "tokenization_path_logprob_std": 165.78647859425496, "trajectory_balance_rmse": 3348.6416988108026} {"canonical_length": 19, "canonical_nll": 2.7250050996479236, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "231e10798735d9e9", "reading_canonical_log_prob": -51.77509689331055, "reading_distances": [17, 20, 16, 21], "reading_log_probs": [-56.46030044555664, -55.630489349365234, -50.263282775878906, -55.70732879638672], "reading_logprob_std": 2.8593272596687043, "tokenization_path_logprob_std": 27.44185440586152, "trajectory_balance_rmse": 563.9431603979172} {"canonical_length": 133, "canonical_nll": 1.8957782257768445, "generation_rouge_l_f1": 0.14775725593667546, "generation_token_f1": 0.2269129287598945, "prompt_id": "948eaf298aee4503", "reading_canonical_log_prob": -252.1385040283203, "reading_distances": [24, 19, 19, 19], "reading_log_probs": [-252.78884887695312, -249.58663940429688, -251.19595336914062, -245.266845703125], "reading_logprob_std": 3.237499293770703, "tokenization_path_logprob_std": 329.5270664641309, "trajectory_balance_rmse": 3700.1041990759345} {"canonical_length": 43, "canonical_nll": 0.4679219445516897, "generation_rouge_l_f1": 0.6415094339622641, "generation_token_f1": 0.660377358490566, "prompt_id": "cc672b4406958c5b", "reading_canonical_log_prob": -20.120643615722656, "reading_distances": [44, 54, 40, 46], "reading_log_probs": [-35.65658950805664, -29.53664779663086, -27.9277286529541, -35.26409149169922], "reading_logprob_std": 3.942899558641438, "tokenization_path_logprob_std": 72.30977297889798, "trajectory_balance_rmse": 1098.5842537092217} {"canonical_length": 189, "canonical_nll": 1.2825102629484955, "generation_rouge_l_f1": 0.1078167115902965, "generation_token_f1": 0.1940700808625337, "prompt_id": "f0c1c49e728adae3", "reading_canonical_log_prob": -242.39443969726562, "reading_distances": [17, 19, 17, 21], "reading_log_probs": [-226.04666137695312, -230.8714599609375, -230.13583374023438, -241.00140380859375], "reading_logprob_std": 6.356504899137019, "tokenization_path_logprob_std": 75.03403891329711, "trajectory_balance_rmse": 4848.711071201398} {"canonical_length": 202, "canonical_nll": 1.7572124216816214, "generation_rouge_l_f1": 0.2280701754385965, "generation_token_f1": 0.5263157894736843, "prompt_id": "91ba22c36f60a796", "reading_canonical_log_prob": -354.9569091796875, "reading_distances": [28, 31, 29, 23], "reading_log_probs": [-358.9928894042969, -358.39263916015625, -355.7383117675781, -358.4037170410156], "reading_logprob_std": 1.4562984447100364, "tokenization_path_logprob_std": 262.2662383744218, "trajectory_balance_rmse": 6583.233007797785} {"canonical_length": 91, "canonical_nll": 1.4079412104009272, "generation_rouge_l_f1": 0.2956521739130435, "generation_token_f1": 0.49565217391304345, "prompt_id": "f1222deb29a5be25", "reading_canonical_log_prob": -128.12265014648438, "reading_distances": [13, 13, 9, 15], "reading_log_probs": [-125.92463684082031, -126.66124725341797, -128.49351501464844, -124.71788024902344], "reading_logprob_std": 1.5808038511511862, "tokenization_path_logprob_std": 193.19025460484812, "trajectory_balance_rmse": 3180.6542421154672} {"canonical_length": 9, "canonical_nll": 0.031075845162073772, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "6b02223ad9dd90ef", "reading_canonical_log_prob": -0.27968260645866394, "reading_distances": [25, 29, 27, 25], "reading_log_probs": [-0.6294581294059753, -0.5773528218269348, -1.0798134803771973, -0.4036664068698883], "reading_logprob_std": 0.2881442573226163, "tokenization_path_logprob_std": 68.11952753055665, "trajectory_balance_rmse": 266.5370041912771} {"canonical_length": 170, "canonical_nll": 1.4261398315429688, "generation_rouge_l_f1": 0.20046620046620042, "generation_token_f1": 0.44289044289044294, "prompt_id": "a027618ad6b82f3e", "reading_canonical_log_prob": -242.4437713623047, "reading_distances": [23, 19, 18, 19], "reading_log_probs": [-246.26223754882812, -247.0335693359375, -244.78683471679688, -245.37222290039062], "reading_logprob_std": 0.9880308729254098, "tokenization_path_logprob_std": 263.39712889986146, "trajectory_balance_rmse": 5069.261128678213} {"canonical_length": 38, "canonical_nll": 1.9682607148822986, "generation_rouge_l_f1": 0.05405405405405406, "generation_token_f1": 0.05405405405405406, "prompt_id": "252eb77faa4d024a", "reading_canonical_log_prob": -74.79390716552734, "reading_distances": [20, 24, 22, 13], "reading_log_probs": [-75.09661102294922, -75.29239654541016, -74.85167694091797, -75.1251220703125], "reading_logprob_std": 0.18168753253871586, "tokenization_path_logprob_std": 37.76111225606146, "trajectory_balance_rmse": 977.9724861422983} {"canonical_length": 273, "canonical_nll": 0.9508557441907052, "generation_rouge_l_f1": 0.25995807127882603, "generation_token_f1": 0.4779874213836478, "prompt_id": "75be9357ba7c860a", "reading_canonical_log_prob": -259.5836181640625, "reading_distances": [24, 18, 17, 16], "reading_log_probs": [-255.33755493164062, -254.15924072265625, -249.986083984375, -244.8572998046875], "reading_logprob_std": 4.744366680302775, "tokenization_path_logprob_std": 119.37719357349202, "trajectory_balance_rmse": 6107.529814622929} {"canonical_length": 347, "canonical_nll": 1.1571029091431015, "generation_rouge_l_f1": 0.2642740619902121, "generation_token_f1": 0.49918433931484507, "prompt_id": "8d16eaea1c875841", "reading_canonical_log_prob": -401.51470947265625, "reading_distances": [41, 42, 47, 53], "reading_log_probs": [-409.32183837890625, -410.39984130859375, -409.209228515625, -413.66357421875], "reading_logprob_std": 2.080387551540833, "tokenization_path_logprob_std": 237.044574477202, "trajectory_balance_rmse": 10510.877531448206} {"canonical_length": 355, "canonical_nll": 1.052713915327905, "generation_rouge_l_f1": 0.2593192868719611, "generation_token_f1": 0.4991896272285251, "prompt_id": "f4746ccf1632be92", "reading_canonical_log_prob": -373.71343994140625, "reading_distances": [23, 26, 23, 20], "reading_log_probs": [-358.0547180175781, -364.4764709472656, -364.32122802734375, -361.15789794921875], "reading_logprob_std": 3.0438663512505206, "tokenization_path_logprob_std": 403.1352059002424, "trajectory_balance_rmse": 10304.794837486332} {"canonical_length": 450, "canonical_nll": 1.1529302300347222, "generation_rouge_l_f1": 0.2836879432624113, "generation_token_f1": 0.4652482269503546, "prompt_id": "06190df6c1e1573b", "reading_canonical_log_prob": -518.818603515625, "reading_distances": [16, 15, 17, 12], "reading_log_probs": [-512.9597778320312, -519.0897216796875, -515.22509765625, -520.742431640625], "reading_logprob_std": 3.5518226440623364, "tokenization_path_logprob_std": 368.51161161869265, "trajectory_balance_rmse": 10037.674461914039} {"canonical_length": 102, "canonical_nll": 0.9710786856857001, "generation_rouge_l_f1": 0.31718061674008813, "generation_token_f1": 0.4581497797356828, "prompt_id": "56fa8783ee42ce19", "reading_canonical_log_prob": -99.0500259399414, "reading_distances": [13, 19, 16, 19], "reading_log_probs": [-98.81341552734375, -99.23828125, -96.2749252319336, -96.06187438964844], "reading_logprob_std": 1.6611204197846565, "tokenization_path_logprob_std": 38.90338032025813, "trajectory_balance_rmse": 2741.7146857919715} {"canonical_length": 202, "canonical_nll": 2.378444256168781, "generation_rouge_l_f1": 0.24215246636771304, "generation_token_f1": 0.42152466367713, "prompt_id": "11f8f65e51ee2cf5", "reading_canonical_log_prob": -480.44573974609375, "reading_distances": [22, 23, 23, 20], "reading_log_probs": [-486.4466857910156, -480.5998229980469, -480.63916015625, -483.93780517578125], "reading_logprob_std": 2.831848216499947, "tokenization_path_logprob_std": 205.3162894343917, "trajectory_balance_rmse": 4447.35838064277} {"canonical_length": 11, "canonical_nll": 4.262051669034091, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.25, "prompt_id": "8980941d8fb125ec", "reading_canonical_log_prob": -46.882568359375, "reading_distances": [15, 18, 16, 21], "reading_log_probs": [-41.35615158081055, -41.828514099121094, -43.499656677246094, -44.709110260009766], "reading_logprob_std": 1.5441664979036405, "tokenization_path_logprob_std": 55.34171548740018, "trajectory_balance_rmse": 321.16729324772865} {"canonical_length": 69, "canonical_nll": 0.7151405223901721, "generation_rouge_l_f1": 0.17777777777777776, "generation_token_f1": 0.3111111111111111, "prompt_id": "2d0e3ff402d06b28", "reading_canonical_log_prob": -49.344696044921875, "reading_distances": [16, 14, 13, 14], "reading_log_probs": [-50.643089294433594, -54.68278503417969, -56.077720642089844, -55.92793273925781], "reading_logprob_std": 2.538086621263875, "tokenization_path_logprob_std": 170.06555833615084, "trajectory_balance_rmse": 1236.1508547127762} {"canonical_length": 53, "canonical_nll": 0.6876652555645637, "generation_rouge_l_f1": 0.23030303030303031, "generation_token_f1": 0.36363636363636365, "prompt_id": "54b6e30181a1bec1", "reading_canonical_log_prob": -36.446258544921875, "reading_distances": [19, 24, 19, 16], "reading_log_probs": [-32.54566192626953, -32.677650451660156, -36.59547424316406, -36.58112335205078], "reading_logprob_std": 2.2965553840717563, "tokenization_path_logprob_std": 33.75798832714257, "trajectory_balance_rmse": 662.1878587827993} {"canonical_length": 9, "canonical_nll": 0.5592784881591797, "generation_rouge_l_f1": 0.823529411764706, "generation_token_f1": 0.823529411764706, "prompt_id": "2996ca79ae58108b", "reading_canonical_log_prob": -5.033506393432617, "reading_distances": [30, 28, 26, 36], "reading_log_probs": [-4.586468696594238, -4.195788383483887, -3.9587507247924805, -5.933286190032959], "reading_logprob_std": 0.88197005347407, "tokenization_path_logprob_std": 58.57723426927629, "trajectory_balance_rmse": 204.48351107498652} {"canonical_length": 11, "canonical_nll": 1.3999912955544211, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.23529411764705882, "prompt_id": "e6cc80b8ba91ddd7", "reading_canonical_log_prob": -15.399904251098633, "reading_distances": [21, 22, 27, 25], "reading_log_probs": [-14.419485092163086, -14.0665283203125, -15.669503211975098, -14.021061897277832], "reading_logprob_std": 0.7710825340498713, "tokenization_path_logprob_std": 46.71133742888259, "trajectory_balance_rmse": 293.58308864019307} {"canonical_length": 218, "canonical_nll": 1.909194806300172, "generation_rouge_l_f1": 0.18333333333333335, "generation_token_f1": 0.4041666666666667, "prompt_id": "1826061f4f79f3ac", "reading_canonical_log_prob": -416.2044677734375, "reading_distances": [14, 16, 13, 20], "reading_log_probs": [-416.77374267578125, -415.40728759765625, -416.4274597167969, -415.1546325683594], "reading_logprob_std": 0.7817343738163594, "tokenization_path_logprob_std": 165.08637844382622, "trajectory_balance_rmse": 6177.4420448227065} {"canonical_length": 182, "canonical_nll": 1.958449478987809, "generation_rouge_l_f1": 0.21126760563380279, "generation_token_f1": 0.49295774647887325, "prompt_id": "d48fbaa3dac1814d", "reading_canonical_log_prob": -356.43780517578125, "reading_distances": [29, 25, 30, 24], "reading_log_probs": [-363.4030456542969, -360.21221923828125, -363.251953125, -366.4541931152344], "reading_logprob_std": 2.549023733635186, "tokenization_path_logprob_std": 93.07355833994023, "trajectory_balance_rmse": 4808.466375843402} {"canonical_length": 105, "canonical_nll": 0.6756785074869792, "generation_rouge_l_f1": 0.31974921630094044, "generation_token_f1": 0.41379310344827586, "prompt_id": "7fbf1f51feeeeb73", "reading_canonical_log_prob": -70.94624328613281, "reading_distances": [17, 20, 17, 20], "reading_log_probs": [-78.28628540039062, -76.57060241699219, -75.0420150756836, -78.63929748535156], "reading_logprob_std": 1.662076884808982, "tokenization_path_logprob_std": 75.40649048346899, "trajectory_balance_rmse": 1248.8259778553784} {"canonical_length": 18, "canonical_nll": 0.4129434691535102, "generation_rouge_l_f1": 0.23762376237623764, "generation_token_f1": 0.25742574257425743, "prompt_id": "8cc0717e876412c6", "reading_canonical_log_prob": -7.432982444763184, "reading_distances": [20, 21, 29, 26], "reading_log_probs": [-10.34348201751709, -12.167996406555176, -9.339597702026367, -10.007770538330078], "reading_logprob_std": 1.2097592268938095, "tokenization_path_logprob_std": 119.32829206806373, "trajectory_balance_rmse": 482.6280453212522} {"canonical_length": 152, "canonical_nll": 2.2555598208778784, "generation_rouge_l_f1": 0.3313253012048193, "generation_token_f1": 0.4819277108433735, "prompt_id": "3f19d2fd29921f16", "reading_canonical_log_prob": -342.8450927734375, "reading_distances": [19, 17, 16, 12], "reading_log_probs": [-347.8451232910156, -344.17059326171875, -355.5398864746094, -347.2367248535156], "reading_logprob_std": 4.836374193300241, "tokenization_path_logprob_std": 346.11371746966046, "trajectory_balance_rmse": 3903.0832890764846} {"canonical_length": 257, "canonical_nll": 2.04142933411357, "generation_rouge_l_f1": 0.1900647948164147, "generation_token_f1": 0.35853131749460043, "prompt_id": "774504b35575be36", "reading_canonical_log_prob": -524.6473388671875, "reading_distances": [31, 24, 28, 29], "reading_log_probs": [-525.2196655273438, -523.4805908203125, -525.1160888671875, -525.20703125], "reading_logprob_std": 0.8514198690869302, "tokenization_path_logprob_std": 433.46377884948174, "trajectory_balance_rmse": 7331.927609652254} {"canonical_length": 9, "canonical_nll": 0.6426234245300293, "generation_rouge_l_f1": 0.07999999999999999, "generation_token_f1": 0.07999999999999999, "prompt_id": "18b528d0286abbc1", "reading_canonical_log_prob": -5.783610820770264, "reading_distances": [34, 30, 33, 31], "reading_log_probs": [-1.263230562210083, -2.20412278175354, -2.0735342502593994, -2.5342676639556885], "reading_logprob_std": 0.53973217401287, "tokenization_path_logprob_std": 24.205626455310732, "trajectory_balance_rmse": 253.19605656611074} {"canonical_length": 4, "canonical_nll": 1.2317899465560913, "generation_rouge_l_f1": 0.8, "generation_token_f1": 0.8, "prompt_id": "ad4fd52fcf4632b3", "reading_canonical_log_prob": -4.927159786224365, "reading_distances": [22, 29, 20, 25], "reading_log_probs": [-4.6390485763549805, -5.124692916870117, -5.109724044799805, -5.207507610321045], "reading_logprob_std": 0.25774267878969315, "tokenization_path_logprob_std": 47.576782698320464, "trajectory_balance_rmse": 139.81292618255532} {"canonical_length": 349, "canonical_nll": 1.5869640798486748, "generation_rouge_l_f1": 0.18675721561969438, "generation_token_f1": 0.400679117147708, "prompt_id": "ba9050b431bd2ba0", "reading_canonical_log_prob": -553.8504638671875, "reading_distances": [18, 15, 21, 19], "reading_log_probs": [-549.0341186523438, -550.9190673828125, -551.4598388671875, -552.532958984375], "reading_logprob_std": 1.4642318985297813, "tokenization_path_logprob_std": 229.93774428289402, "trajectory_balance_rmse": 8224.170166689835} {"canonical_length": 35, "canonical_nll": 0.6273050580705916, "generation_rouge_l_f1": 0.13100436681222707, "generation_token_f1": 0.1834061135371179, "prompt_id": "da7f7e74060fc4e6", "reading_canonical_log_prob": -21.955677032470703, "reading_distances": [27, 33, 26, 26], "reading_log_probs": [-22.361417770385742, -22.952983856201172, -23.367107391357422, -21.54442024230957], "reading_logprob_std": 0.7909164098559757, "tokenization_path_logprob_std": 92.42651028868256, "trajectory_balance_rmse": 957.7118921109634} {"canonical_length": 366, "canonical_nll": 1.141753650102459, "generation_rouge_l_f1": 0.22910216718266255, "generation_token_f1": 0.47368421052631576, "prompt_id": "378c4d815b07ba59", "reading_canonical_log_prob": -417.8818359375, "reading_distances": [29, 25, 29, 34], "reading_log_probs": [-419.25506591796875, -419.4009094238281, -417.7583312988281, -424.048095703125], "reading_logprob_std": 2.724734831784921, "tokenization_path_logprob_std": 560.1007502259805, "trajectory_balance_rmse": 9674.691362308937} {"canonical_length": 158, "canonical_nll": 0.6283046867274031, "generation_rouge_l_f1": 0.4584980237154149, "generation_token_f1": 0.6245059288537549, "prompt_id": "e2790ed28464c480", "reading_canonical_log_prob": -99.27214050292969, "reading_distances": [14, 24, 22, 19], "reading_log_probs": [-101.16445922851562, -99.07308197021484, -95.41170501708984, -97.56082153320312], "reading_logprob_std": 2.428400693095983, "tokenization_path_logprob_std": 171.61478327282694, "trajectory_balance_rmse": 2377.8514687134843} {"canonical_length": 253, "canonical_nll": 1.3293976915683672, "generation_rouge_l_f1": 0.3116370808678501, "generation_token_f1": 0.5956607495069033, "prompt_id": "2481a184df55ff40", "reading_canonical_log_prob": -336.3376159667969, "reading_distances": [47, 39, 37, 41], "reading_log_probs": [-354.13360595703125, -367.86578369140625, -354.832763671875, -345.3292236328125], "reading_logprob_std": 9.285526812948719, "tokenization_path_logprob_std": 336.48191789117146, "trajectory_balance_rmse": 6199.105376095302} {"canonical_length": 342, "canonical_nll": 1.6340537266424524, "generation_rouge_l_f1": 0.19661016949152546, "generation_token_f1": 0.4542372881355932, "prompt_id": "130ff8d17ea30994", "reading_canonical_log_prob": -558.8463745117188, "reading_distances": [18, 20, 17, 21], "reading_log_probs": [-563.496337890625, -561.1611938476562, -558.4644775390625, -558.8604736328125], "reading_logprob_std": 2.3271255333958316, "tokenization_path_logprob_std": 394.7579189526552, "trajectory_balance_rmse": 9753.437734893849} {"canonical_length": 16, "canonical_nll": 1.1576988697052002, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.7777777777777778, "prompt_id": "16bf564574acdadf", "reading_canonical_log_prob": -18.523181915283203, "reading_distances": [27, 28, 22, 31], "reading_log_probs": [-21.07418441772461, -20.480121612548828, -18.71076202392578, -21.025814056396484], "reading_logprob_std": 1.1078844415532694, "tokenization_path_logprob_std": 37.079903654455805, "trajectory_balance_rmse": 278.24012587924676} {"canonical_length": 50, "canonical_nll": 0.5708674621582032, "generation_rouge_l_f1": 0.18681318681318684, "generation_token_f1": 0.3076923076923077, "prompt_id": "effedc7b02b1c679", "reading_canonical_log_prob": -28.543373107910156, "reading_distances": [29, 29, 27, 26], "reading_log_probs": [-34.404579162597656, -40.077613830566406, -33.877952575683594, -34.70311737060547], "reading_logprob_std": 2.8947033782966836, "tokenization_path_logprob_std": 84.22120083136895, "trajectory_balance_rmse": 1057.8758136896438} {"canonical_length": 176, "canonical_nll": 1.4862091758034446, "generation_rouge_l_f1": 0.22802850356294538, "generation_token_f1": 0.3990498812351544, "prompt_id": "33cc463f90c8e3a1", "reading_canonical_log_prob": -261.57281494140625, "reading_distances": [25, 17, 25, 28], "reading_log_probs": [-258.5277099609375, -256.7308349609375, -262.81817626953125, -257.9906311035156], "reading_logprob_std": 2.64375549594354, "tokenization_path_logprob_std": 102.94887680077012, "trajectory_balance_rmse": 4567.217063294761} {"canonical_length": 283, "canonical_nll": 1.2798166443518109, "generation_rouge_l_f1": 0.22950819672131148, "generation_token_f1": 0.49544626593806923, "prompt_id": "e32f1e4a8a806d4a", "reading_canonical_log_prob": -362.1881103515625, "reading_distances": [15, 14, 15, 15], "reading_log_probs": [-357.6654357910156, -358.8204650878906, -357.371826171875, -358.583984375], "reading_logprob_std": 0.7004662811417461, "tokenization_path_logprob_std": 518.9495608353888, "trajectory_balance_rmse": 8202.220745469887} {"canonical_length": 193, "canonical_nll": 1.2876477809767648, "generation_rouge_l_f1": 0.2724014336917563, "generation_token_f1": 0.41577060931899634, "prompt_id": "f16b4d6e1a4f11c6", "reading_canonical_log_prob": -248.51602172851562, "reading_distances": [11, 8, 9, 13], "reading_log_probs": [-250.4691162109375, -252.52896118164062, -251.039306640625, -250.7634735107422], "reading_logprob_std": 0.9159160229241494, "tokenization_path_logprob_std": 224.22645531346288, "trajectory_balance_rmse": 5441.980597401811} {"canonical_length": 11, "canonical_nll": 1.1692393042824485, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "7cf86e055cecbc26", "reading_canonical_log_prob": -12.861632347106934, "reading_distances": [9, 16, 13, 16], "reading_log_probs": [-13.808557510375977, -13.9271879196167, -15.707712173461914, -14.567587852478027], "reading_logprob_std": 0.8697341944711259, "tokenization_path_logprob_std": 50.86435760291838, "trajectory_balance_rmse": 282.8072155480853} {"canonical_length": 251, "canonical_nll": 1.6500292774215637, "generation_rouge_l_f1": 0.18473895582329314, "generation_token_f1": 0.39357429718875503, "prompt_id": "279ec9f263f853e8", "reading_canonical_log_prob": -414.1573486328125, "reading_distances": [29, 20, 27, 24], "reading_log_probs": [-414.26983642578125, -414.72393798828125, -412.7384948730469, -413.0966796875], "reading_logprob_std": 0.9418847335426476, "tokenization_path_logprob_std": 207.09973759147974, "trajectory_balance_rmse": 6982.026832924424} {"canonical_length": 102, "canonical_nll": 1.1412350523705577, "generation_rouge_l_f1": 0.30057803468208094, "generation_token_f1": 0.4277456647398844, "prompt_id": "cf98e6698c25bbf0", "reading_canonical_log_prob": -116.40597534179688, "reading_distances": [110, 109, 100, 109], "reading_log_probs": [-106.39302825927734, -116.52886962890625, -115.67184448242188, -111.6302261352539], "reading_logprob_std": 4.63074716261748, "tokenization_path_logprob_std": 192.08940483546053, "trajectory_balance_rmse": 2442.483716392076} {"canonical_length": 125, "canonical_nll": 1.498604248046875, "generation_rouge_l_f1": 0.21483375959079282, "generation_token_f1": 0.4040920716112532, "prompt_id": "588a316f7d854f31", "reading_canonical_log_prob": -187.32553100585938, "reading_distances": [34, 25, 33, 35], "reading_log_probs": [-189.1034393310547, -187.57530212402344, -190.0097198486328, -188.46658325195312], "reading_logprob_std": 1.0273013039209267, "tokenization_path_logprob_std": 143.50589108005002, "trajectory_balance_rmse": 4162.527305224932} {"canonical_length": 116, "canonical_nll": 1.8502192003973599, "generation_rouge_l_f1": 0.2271468144044321, "generation_token_f1": 0.38227146814404434, "prompt_id": "6bdc8fa7262e4a79", "reading_canonical_log_prob": -214.62542724609375, "reading_distances": [18, 15, 18, 11], "reading_log_probs": [-219.51620483398438, -229.2003936767578, -217.78106689453125, -223.1748046875], "reading_logprob_std": 5.049593418497613, "tokenization_path_logprob_std": 242.9357138711516, "trajectory_balance_rmse": 3309.9160447459117} {"canonical_length": 7, "canonical_nll": 2.1782331466674805, "generation_rouge_l_f1": 0.07058823529411765, "generation_token_f1": 0.09411764705882351, "prompt_id": "73284d596fde1232", "reading_canonical_log_prob": -15.247632026672363, "reading_distances": [26, 25, 31, 27], "reading_log_probs": [-13.151677131652832, -9.473527908325195, -11.683866500854492, -9.432269096374512], "reading_logprob_std": 1.813703295858417, "tokenization_path_logprob_std": 34.97745698381266, "trajectory_balance_rmse": 242.23269538450285} {"canonical_length": 47, "canonical_nll": 1.6357858536091257, "generation_rouge_l_f1": 0.20216606498194947, "generation_token_f1": 0.26714801444043323, "prompt_id": "08b3c472b8ed1840", "reading_canonical_log_prob": -76.8819351196289, "reading_distances": [37, 51, 48, 51], "reading_log_probs": [-78.58626556396484, -81.08661651611328, -75.23213195800781, -77.55896759033203], "reading_logprob_std": 2.4271170947866083, "tokenization_path_logprob_std": 97.36134881239107, "trajectory_balance_rmse": 936.0844235646322} {"canonical_length": 149, "canonical_nll": 0.9247377126808934, "generation_rouge_l_f1": 0.3171355498721228, "generation_token_f1": 0.516624040920716, "prompt_id": "51fa8b33eba8ebe6", "reading_canonical_log_prob": -137.78591918945312, "reading_distances": [40, 36, 42, 45], "reading_log_probs": [-142.15577697753906, -144.28367614746094, -141.97384643554688, -142.802490234375], "reading_logprob_std": 1.0486168345544784, "tokenization_path_logprob_std": 182.6646039400103, "trajectory_balance_rmse": 3081.1269275141417} {"canonical_length": 25, "canonical_nll": 0.8052664947509766, "generation_rouge_l_f1": 0.5625, "generation_token_f1": 0.5625, "prompt_id": "6e11957de8d37c71", "reading_canonical_log_prob": -20.131662368774414, "reading_distances": [26, 22, 27, 19], "reading_log_probs": [-18.672630310058594, -20.05086326599121, -20.762256622314453, -19.36626434326172], "reading_logprob_std": 0.8977166875517433, "tokenization_path_logprob_std": 49.66067396083709, "trajectory_balance_rmse": 692.8466251782012} {"canonical_length": 124, "canonical_nll": 1.1279867849042338, "generation_rouge_l_f1": 0.453125, "generation_token_f1": 0.65625, "prompt_id": "613e08997d787f9d", "reading_canonical_log_prob": -139.870361328125, "reading_distances": [21, 18, 24, 20], "reading_log_probs": [-139.27764892578125, -141.2631072998047, -143.32278442382812, -142.54998779296875], "reading_logprob_std": 1.7679770385439575, "tokenization_path_logprob_std": 263.22834876144697, "trajectory_balance_rmse": 3606.114923821054} {"canonical_length": 7, "canonical_nll": 0.44972429956708637, "generation_rouge_l_f1": 0.9090909090909091, "generation_token_f1": 0.9090909090909091, "prompt_id": "82a7bb73a8e6bdca", "reading_canonical_log_prob": -3.1480700969696045, "reading_distances": [33, 34, 31, 40], "reading_log_probs": [-2.294247627258301, -2.732832431793213, -2.8619558811187744, -2.075118064880371], "reading_logprob_std": 0.3686735073261852, "tokenization_path_logprob_std": 25.08151816922622, "trajectory_balance_rmse": 151.34741860624726} {"canonical_length": 500, "canonical_nll": 0.8500213623046875, "generation_rouge_l_f1": 0.16032608695652176, "generation_token_f1": 0.2989130434782609, "prompt_id": "04c1fbf027769e8a", "reading_canonical_log_prob": -425.01068115234375, "reading_distances": [24, 24, 28, 22], "reading_log_probs": [-428.971435546875, -426.5018310546875, -436.31988525390625, -428.31256103515625], "reading_logprob_std": 4.323607170060074, "tokenization_path_logprob_std": 327.34390873895285, "trajectory_balance_rmse": 10957.367571307093} {"canonical_length": 47, "canonical_nll": 2.0567096142058676, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "5950f3bf48416566", "reading_canonical_log_prob": -96.66535186767578, "reading_distances": [32, 35, 36, 48], "reading_log_probs": [-99.03800201416016, -93.15045166015625, -94.93891906738281, -95.1308364868164], "reading_logprob_std": 2.4814149971641726, "tokenization_path_logprob_std": 88.42015415670738, "trajectory_balance_rmse": 1583.747246405311} {"canonical_length": 310, "canonical_nll": 1.2497745636970765, "generation_rouge_l_f1": 0.281195079086116, "generation_token_f1": 0.5026362038664324, "prompt_id": "f93b23d1ff15b3d1", "reading_canonical_log_prob": -387.43011474609375, "reading_distances": [22, 25, 23, 24], "reading_log_probs": [-384.6653747558594, -385.00274658203125, -384.6202392578125, -385.7057800292969], "reading_logprob_std": 0.5014368307577995, "tokenization_path_logprob_std": 252.32706371697333, "trajectory_balance_rmse": 7678.821720685976} {"canonical_length": 228, "canonical_nll": 1.622291966488487, "generation_rouge_l_f1": 0.2138364779874214, "generation_token_f1": 0.46960167714884693, "prompt_id": "cb117bc3f0e8e883", "reading_canonical_log_prob": -369.882568359375, "reading_distances": [25, 19, 19, 23], "reading_log_probs": [-363.48236083984375, -359.31597900390625, -363.4093017578125, -365.25933837890625], "reading_logprob_std": 2.516995095579661, "tokenization_path_logprob_std": 470.1727301733588, "trajectory_balance_rmse": 6375.841986332313} {"canonical_length": 318, "canonical_nll": 2.0170194038055227, "generation_rouge_l_f1": 0.22797927461139894, "generation_token_f1": 0.48013816925734026, "prompt_id": "0ff4ea021b8e685b", "reading_canonical_log_prob": -641.4121704101562, "reading_distances": [16, 20, 14, 14], "reading_log_probs": [-642.2459716796875, -649.5689697265625, -643.9491577148438, -641.712158203125], "reading_logprob_std": 3.595460576314188, "tokenization_path_logprob_std": 278.31964750511634, "trajectory_balance_rmse": 8530.762275136663} {"canonical_length": 70, "canonical_nll": 1.54550290788923, "generation_rouge_l_f1": 0.2100456621004566, "generation_token_f1": 0.3926940639269407, "prompt_id": "d8931b93e350f362", "reading_canonical_log_prob": -108.1852035522461, "reading_distances": [14, 16, 14, 11], "reading_log_probs": [-115.80210876464844, -116.19941711425781, -112.39180755615234, -110.15953063964844], "reading_logprob_std": 2.8807967005553827, "tokenization_path_logprob_std": 173.50961100011915, "trajectory_balance_rmse": 2046.778325119221} {"canonical_length": 43, "canonical_nll": 1.9380683011786883, "generation_rouge_l_f1": 0.3692307692307692, "generation_token_f1": 0.5230769230769231, "prompt_id": "cfd710328af17995", "reading_canonical_log_prob": -83.3369369506836, "reading_distances": [32, 31, 27, 30], "reading_log_probs": [-86.3503189086914, -86.78677368164062, -86.32916259765625, -87.68622589111328], "reading_logprob_std": 0.6347985761752393, "tokenization_path_logprob_std": 138.92572590337778, "trajectory_balance_rmse": 1139.2590857744833} {"canonical_length": 237, "canonical_nll": 0.8136704199424776, "generation_rouge_l_f1": 0.2774566473988439, "generation_token_f1": 0.4624277456647399, "prompt_id": "3f921f925896c713", "reading_canonical_log_prob": -192.8398895263672, "reading_distances": [37, 35, 42, 39], "reading_log_probs": [-205.42884826660156, -201.506591796875, -208.46380615234375, -203.50839233398438], "reading_logprob_std": 2.961550540696039, "tokenization_path_logprob_std": 211.1712209189227, "trajectory_balance_rmse": 3182.4131899076197} {"canonical_length": 242, "canonical_nll": 1.0413577497498063, "generation_rouge_l_f1": 0.2785862785862786, "generation_token_f1": 0.4906444906444906, "prompt_id": "74c9b8835b365ad8", "reading_canonical_log_prob": -252.00857543945312, "reading_distances": [21, 15, 17, 19], "reading_log_probs": [-254.74972534179688, -253.258056640625, -253.93338012695312, -252.527587890625], "reading_logprob_std": 0.9484763260702259, "tokenization_path_logprob_std": 118.60845837787276, "trajectory_balance_rmse": 6020.371279241632} {"canonical_length": 243, "canonical_nll": 1.239944018454218, "generation_rouge_l_f1": 0.19844357976653695, "generation_token_f1": 0.5058365758754864, "prompt_id": "d4cbdb2804902b19", "reading_canonical_log_prob": -301.306396484375, "reading_distances": [26, 16, 23, 24], "reading_log_probs": [-304.774169921875, -309.0561828613281, -308.23089599609375, -306.4653015136719], "reading_logprob_std": 1.907345547930448, "tokenization_path_logprob_std": 102.41213825111811, "trajectory_balance_rmse": 5699.855422688687} {"canonical_length": 140, "canonical_nll": 1.3500404357910156, "generation_rouge_l_f1": 0.2324159021406728, "generation_token_f1": 0.41590214067278286, "prompt_id": "ad65facd3056e4a9", "reading_canonical_log_prob": -189.0056610107422, "reading_distances": [18, 13, 19, 17], "reading_log_probs": [-198.6278076171875, -190.0364990234375, -196.57138061523438, -196.3056640625], "reading_logprob_std": 3.7138203493638398, "tokenization_path_logprob_std": 136.90694916961388, "trajectory_balance_rmse": 2648.0280289783873} {"canonical_length": 245, "canonical_nll": 1.1585683394451531, "generation_rouge_l_f1": 0.24421052631578946, "generation_token_f1": 0.5136842105263157, "prompt_id": "4eba5e8b52e3bb5e", "reading_canonical_log_prob": -283.8492431640625, "reading_distances": [19, 24, 22, 25], "reading_log_probs": [-289.8837890625, -289.7600402832031, -291.3746643066406, -291.6416931152344], "reading_logprob_std": 0.9809511504385466, "tokenization_path_logprob_std": 363.643459289736, "trajectory_balance_rmse": 5887.028932985429} {"canonical_length": 118, "canonical_nll": 1.8056279845156913, "generation_rouge_l_f1": 0.15873015873015872, "generation_token_f1": 0.2751322751322751, "prompt_id": "4873d3597b3d71b4", "reading_canonical_log_prob": -213.06410217285156, "reading_distances": [21, 22, 20, 20], "reading_log_probs": [-214.98727416992188, -216.919677734375, -216.69265747070312, -214.26019287109375], "reading_logprob_std": 1.2978329669135913, "tokenization_path_logprob_std": 230.42483599691676, "trajectory_balance_rmse": 4105.452047566248} {"canonical_length": 38, "canonical_nll": 1.3723343297054893, "generation_rouge_l_f1": 0.19512195121951217, "generation_token_f1": 0.19512195121951217, "prompt_id": "2089d8bccf6a73c9", "reading_canonical_log_prob": -52.148704528808594, "reading_distances": [12, 10, 9, 11], "reading_log_probs": [-56.678009033203125, -59.00999069213867, -61.472808837890625, -63.55268096923828], "reading_logprob_std": 2.9821236016778587, "tokenization_path_logprob_std": 157.03689425651842, "trajectory_balance_rmse": 958.0513891412523} {"canonical_length": 277, "canonical_nll": 1.6811126819156137, "generation_rouge_l_f1": 0.21093750000000003, "generation_token_f1": 0.42578125, "prompt_id": "9b806d8ec328a266", "reading_canonical_log_prob": -465.668212890625, "reading_distances": [31, 23, 30, 30], "reading_log_probs": [-474.0312805175781, -469.6107177734375, -476.2025146484375, -474.1153564453125], "reading_logprob_std": 2.77432403715753, "tokenization_path_logprob_std": 410.1437113119198, "trajectory_balance_rmse": 8041.734403722451} {"canonical_length": 400, "canonical_nll": 0.9584629821777344, "generation_rouge_l_f1": 0.24793388429752064, "generation_token_f1": 0.41983471074380163, "prompt_id": "ce03043d2a4264da", "reading_canonical_log_prob": -383.38519287109375, "reading_distances": [20, 23, 24, 18], "reading_log_probs": [-389.4862976074219, -396.04168701171875, -393.459228515625, -387.1894226074219], "reading_logprob_std": 3.9620605760882763, "tokenization_path_logprob_std": 322.2689137176091, "trajectory_balance_rmse": 8878.781908645686} {"canonical_length": 12, "canonical_nll": 1.1111074288686116, "generation_rouge_l_f1": 0.32, "generation_token_f1": 0.32, "prompt_id": "59741a0906fe97c9", "reading_canonical_log_prob": -13.33328914642334, "reading_distances": [27, 27, 30, 23], "reading_log_probs": [-15.390155792236328, -13.289369583129883, -13.921890258789062, -13.23780632019043], "reading_logprob_std": 1.0030132365846092, "tokenization_path_logprob_std": 77.73324257900802, "trajectory_balance_rmse": 290.87309577816444} {"canonical_length": 103, "canonical_nll": 0.5742677855260164, "generation_rouge_l_f1": 0.32799999999999996, "generation_token_f1": 0.504, "prompt_id": "995c665a4ec8066d", "reading_canonical_log_prob": -59.14958190917969, "reading_distances": [31, 27, 34, 26], "reading_log_probs": [-65.54862976074219, -67.92144012451172, -62.03514099121094, -70.06434631347656], "reading_logprob_std": 3.4408723563118593, "tokenization_path_logprob_std": 241.12899617178914, "trajectory_balance_rmse": 1833.240171806049} {"canonical_length": 63, "canonical_nll": 1.1041996062748016, "generation_rouge_l_f1": 0.6115702479338843, "generation_token_f1": 0.7272727272727273, "prompt_id": "e7b51e88b3b34da7", "reading_canonical_log_prob": -69.5645751953125, "reading_distances": [29, 30, 26, 28], "reading_log_probs": [-70.93414306640625, -70.78575897216797, -70.68514251708984, -70.56136322021484], "reading_logprob_std": 0.15779268190974557, "tokenization_path_logprob_std": 139.96213698094968, "trajectory_balance_rmse": 1480.9420236500885} {"canonical_length": 90, "canonical_nll": 1.072786119249132, "generation_rouge_l_f1": 0.4516129032258064, "generation_token_f1": 0.5913978494623656, "prompt_id": "6b580b925c517c44", "reading_canonical_log_prob": -96.55075073242188, "reading_distances": [26, 23, 27, 20], "reading_log_probs": [-96.74630737304688, -97.01100158691406, -97.37583923339844, -97.20914459228516], "reading_logprob_std": 0.2709160300121185, "tokenization_path_logprob_std": 255.8280186555547, "trajectory_balance_rmse": 2261.5632881316765} {"canonical_length": 169, "canonical_nll": 0.800029506344767, "generation_rouge_l_f1": 0.25181598062953997, "generation_token_f1": 0.48426150121065387, "prompt_id": "58b4edff73083c67", "reading_canonical_log_prob": -135.20498657226562, "reading_distances": [54, 51, 52, 62], "reading_log_probs": [-161.34420776367188, -160.06097412109375, -155.84161376953125, -156.14614868164062], "reading_logprob_std": 2.77138125875445, "tokenization_path_logprob_std": 114.96996543284926, "trajectory_balance_rmse": 3642.692180756435} {"canonical_length": 6, "canonical_nll": 1.9212354024251301, "generation_rouge_l_f1": 0.10204081632653061, "generation_token_f1": 0.10204081632653061, "prompt_id": "dc845859825dc504", "reading_canonical_log_prob": -11.527412414550781, "reading_distances": [33, 29, 30, 35], "reading_log_probs": [-13.560018539428711, -14.970276832580566, -15.45458698272705, -13.6382417678833], "reading_logprob_std": 0.9527294381503372, "tokenization_path_logprob_std": 19.105551475344214, "trajectory_balance_rmse": 154.69280436032042} {"canonical_length": 55, "canonical_nll": 0.3118534781716087, "generation_rouge_l_f1": 0.9090909090909091, "generation_token_f1": 0.9292929292929293, "prompt_id": "f074eb196b10e4a0", "reading_canonical_log_prob": -17.151941299438477, "reading_distances": [30, 31, 27, 29], "reading_log_probs": [-19.76630210876465, -16.962398529052734, -19.530302047729492, -21.278888702392578], "reading_logprob_std": 1.7909322689358973, "tokenization_path_logprob_std": 53.10969026592298, "trajectory_balance_rmse": 523.5121858598793} {"canonical_length": 39, "canonical_nll": 2.176725338666867, "generation_rouge_l_f1": 0.16883116883116883, "generation_token_f1": 0.23376623376623376, "prompt_id": "b1076398ceba44c5", "reading_canonical_log_prob": -84.89228820800781, "reading_distances": [9, 12, 15, 13], "reading_log_probs": [-93.92228698730469, -96.08019256591797, -93.92964172363281, -96.45848083496094], "reading_logprob_std": 1.361735657227506, "tokenization_path_logprob_std": 65.01885722456403, "trajectory_balance_rmse": 1141.853031816657} {"canonical_length": 278, "canonical_nll": 1.8890508198909621, "generation_rouge_l_f1": 0.21374045801526714, "generation_token_f1": 0.37659033078880405, "prompt_id": "43d587270196a951", "reading_canonical_log_prob": -525.1561279296875, "reading_distances": [23, 24, 29, 22], "reading_log_probs": [-528.8345947265625, -523.9737548828125, -534.8594970703125, -528.682373046875], "reading_logprob_std": 4.460733430601818, "tokenization_path_logprob_std": 372.9029978433943, "trajectory_balance_rmse": 8060.208936673056} {"canonical_length": 77, "canonical_nll": 1.6656593223670861, "generation_rouge_l_f1": 0.21498371335504884, "generation_token_f1": 0.2996742671009772, "prompt_id": "9ec5781f60b3ed7d", "reading_canonical_log_prob": -128.25576782226562, "reading_distances": [14, 16, 9, 14], "reading_log_probs": [-127.25108337402344, -127.37725830078125, -126.58042907714844, -128.53506469726562], "reading_logprob_std": 0.8119091804352415, "tokenization_path_logprob_std": 290.33267086270183, "trajectory_balance_rmse": 2029.8722747206439} {"canonical_length": 50, "canonical_nll": 1.165788116455078, "generation_rouge_l_f1": 0.37593984962406013, "generation_token_f1": 0.481203007518797, "prompt_id": "9c285d8b80679a4e", "reading_canonical_log_prob": -58.289405822753906, "reading_distances": [21, 24, 26, 20], "reading_log_probs": [-58.8428955078125, -58.87839126586914, -59.67409896850586, -58.94581604003906], "reading_logprob_std": 0.39484645221970727, "tokenization_path_logprob_std": 23.83109087511987, "trajectory_balance_rmse": 1322.6526860392967} {"canonical_length": 260, "canonical_nll": 1.5013045090895432, "generation_rouge_l_f1": 0.30181086519114686, "generation_token_f1": 0.5030181086519115, "prompt_id": "46fcd75d760af9c5", "reading_canonical_log_prob": -390.33917236328125, "reading_distances": [24, 20, 34, 25], "reading_log_probs": [-393.94677734375, -392.5401916503906, -391.6795959472656, -387.8681945800781], "reading_logprob_std": 2.600684608316008, "tokenization_path_logprob_std": 300.8552497679351, "trajectory_balance_rmse": 6886.472830764982} {"canonical_length": 328, "canonical_nll": 1.0660173369616996, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.3428571428571428, "prompt_id": "5feba3faef17f1e8", "reading_canonical_log_prob": -349.6536865234375, "reading_distances": [27, 24, 28, 26], "reading_log_probs": [-348.9730529785156, -349.0184631347656, -348.940185546875, -350.0785217285156], "reading_logprob_std": 0.5515783695421428, "tokenization_path_logprob_std": 508.74579991747794, "trajectory_balance_rmse": 9193.032960122566} {"canonical_length": 232, "canonical_nll": 1.26138318818191, "generation_rouge_l_f1": 0.3122362869198312, "generation_token_f1": 0.5485232067510548, "prompt_id": "49ec1dbe6be4038c", "reading_canonical_log_prob": -292.6408996582031, "reading_distances": [17, 24, 23, 16], "reading_log_probs": [-297.590576171875, -299.23577880859375, -300.1452331542969, -293.82122802734375], "reading_logprob_std": 2.792526598855223, "tokenization_path_logprob_std": 281.15822627505173, "trajectory_balance_rmse": 6725.229186707102} {"canonical_length": 368, "canonical_nll": 1.3188113336977751, "generation_rouge_l_f1": 0.19106317411402157, "generation_token_f1": 0.38520801232665636, "prompt_id": "af5752cd92c29a4d", "reading_canonical_log_prob": -485.32257080078125, "reading_distances": [17, 20, 19, 15], "reading_log_probs": [-466.89990234375, -465.1998291015625, -466.87408447265625, -464.8680725097656], "reading_logprob_std": 1.0784450822301195, "tokenization_path_logprob_std": 490.8386309110614, "trajectory_balance_rmse": 9291.104894896162} {"canonical_length": 152, "canonical_nll": 1.5905350132992393, "generation_rouge_l_f1": 0.2115384615384615, "generation_token_f1": 0.37980769230769235, "prompt_id": "a82619f7741e0303", "reading_canonical_log_prob": -241.76132202148438, "reading_distances": [26, 16, 20, 19], "reading_log_probs": [-243.1777801513672, -244.34466552734375, -243.520263671875, -242.72451782226562], "reading_logprob_std": 0.6844761771000958, "tokenization_path_logprob_std": 206.05430609042529, "trajectory_balance_rmse": 4874.541534071556} {"canonical_length": 419, "canonical_nll": 1.5344685483946152, "generation_rouge_l_f1": 0.2731034482758621, "generation_token_f1": 0.5351724137931034, "prompt_id": "9294062be51005dc", "reading_canonical_log_prob": -642.9423217773438, "reading_distances": [19, 22, 20, 26], "reading_log_probs": [-649.2135009765625, -642.2823486328125, -651.7128295898438, -642.9382934570312], "reading_logprob_std": 4.654947768969411, "tokenization_path_logprob_std": 252.13603119554958, "trajectory_balance_rmse": 9230.973919659085} {"canonical_length": 9, "canonical_nll": 0.1125397417280409, "generation_rouge_l_f1": 0.9333333333333333, "generation_token_f1": 0.9333333333333333, "prompt_id": "3e7c9dafc0fa7462", "reading_canonical_log_prob": -1.0128576755523682, "reading_distances": [23, 32, 26, 29], "reading_log_probs": [-0.7618037462234497, -0.39648035168647766, -0.9940416812896729, -0.40565937757492065], "reading_logprob_std": 0.2912032487338779, "tokenization_path_logprob_std": 33.03750470307051, "trajectory_balance_rmse": 292.14223637988835} {"canonical_length": 12, "canonical_nll": 0.5428568919499716, "generation_rouge_l_f1": 0.75, "generation_token_f1": 0.75, "prompt_id": "c34fdb5fae4b8ff6", "reading_canonical_log_prob": -6.514282703399658, "reading_distances": [54, 44, 42, 59], "reading_log_probs": [-6.776762008666992, -6.296392440795898, -6.685921669006348, -7.080190181732178], "reading_logprob_std": 0.3230834675062158, "tokenization_path_logprob_std": 127.98376425734178, "trajectory_balance_rmse": 384.74310708224476} {"canonical_length": 122, "canonical_nll": 1.7499616028832607, "generation_rouge_l_f1": 0.19101123595505617, "generation_token_f1": 0.3314606741573034, "prompt_id": "d51b2c6c349d91e4", "reading_canonical_log_prob": -213.4953155517578, "reading_distances": [20, 25, 30, 31], "reading_log_probs": [-215.48245239257812, -209.07069396972656, -214.1727294921875, -213.9875030517578], "reading_logprob_std": 2.8181107817156983, "tokenization_path_logprob_std": 28.7167445538657, "trajectory_balance_rmse": 3177.39269081856} {"canonical_length": 311, "canonical_nll": 1.425866130080637, "generation_rouge_l_f1": 0.22727272727272727, "generation_token_f1": 0.46503496503496505, "prompt_id": "fc10a987783bea33", "reading_canonical_log_prob": -443.4443664550781, "reading_distances": [27, 25, 26, 25], "reading_log_probs": [-443.1077575683594, -446.613525390625, -439.2630310058594, -441.52813720703125], "reading_logprob_std": 3.0901615064432275, "tokenization_path_logprob_std": 491.29049988415784, "trajectory_balance_rmse": 9421.126217150882} {"canonical_length": 31, "canonical_nll": 1.6281494632844002, "generation_rouge_l_f1": 0.10606060606060605, "generation_token_f1": 0.11363636363636363, "prompt_id": "7386d29fb448f53c", "reading_canonical_log_prob": -50.472633361816406, "reading_distances": [15, 17, 12, 17], "reading_log_probs": [-45.6241455078125, -51.69001007080078, -44.723846435546875, -51.14491271972656], "reading_logprob_std": 3.63018343275396, "tokenization_path_logprob_std": 51.02765430425086, "trajectory_balance_rmse": 488.52029752494616} {"canonical_length": 253, "canonical_nll": 0.9459361200747283, "generation_rouge_l_f1": 0.1675977653631285, "generation_token_f1": 0.3575418994413408, "prompt_id": "d5d2ea5bf0257ac3", "reading_canonical_log_prob": -239.32183837890625, "reading_distances": [34, 33, 30, 34], "reading_log_probs": [-243.26693725585938, -243.49571228027344, -244.2134246826172, -240.6265411376953], "reading_logprob_std": 1.5687802008868372, "tokenization_path_logprob_std": 163.2171756026461, "trajectory_balance_rmse": 7699.604932967069} {"canonical_length": 379, "canonical_nll": 1.2211545274880442, "generation_rouge_l_f1": 0.14330218068535827, "generation_token_f1": 0.3052959501557633, "prompt_id": "0a54bd903d89ace0", "reading_canonical_log_prob": -462.81756591796875, "reading_distances": [21, 18, 22, 20], "reading_log_probs": [-460.97039794921875, -464.5991516113281, -462.70294189453125, -462.9152526855469], "reading_logprob_std": 1.4840323548368972, "tokenization_path_logprob_std": 376.2040605022848, "trajectory_balance_rmse": 9625.397883660755} {"canonical_length": 128, "canonical_nll": 1.121808648109436, "generation_rouge_l_f1": 0.3693931398416887, "generation_token_f1": 0.5435356200527705, "prompt_id": "49a85c182167a715", "reading_canonical_log_prob": -143.5915069580078, "reading_distances": [15, 10, 18, 18], "reading_log_probs": [-146.71234130859375, -144.2207794189453, -142.88217163085938, -145.36392211914062], "reading_logprob_std": 1.6318202068571601, "tokenization_path_logprob_std": 107.16361451069832, "trajectory_balance_rmse": 3670.6567528708483} {"canonical_length": 127, "canonical_nll": 1.2503669318251722, "generation_rouge_l_f1": 0.1616161616161616, "generation_token_f1": 0.31313131313131315, "prompt_id": "50a15e61cf0ab437", "reading_canonical_log_prob": -158.79660034179688, "reading_distances": [35, 28, 41, 28], "reading_log_probs": [-173.05531311035156, -177.156494140625, -176.90707397460938, -175.17498779296875], "reading_logprob_std": 1.89599289565034, "tokenization_path_logprob_std": 150.53544426093535, "trajectory_balance_rmse": 4050.948237625079} {"canonical_length": 14, "canonical_nll": 0.9352896554129464, "generation_rouge_l_f1": 0.7199999999999999, "generation_token_f1": 0.7199999999999999, "prompt_id": "394139a0ae5c1d75", "reading_canonical_log_prob": -13.09405517578125, "reading_distances": [20, 29, 26, 29], "reading_log_probs": [-11.597375869750977, -11.04491901397705, -11.931657791137695, -11.304391860961914], "reading_logprob_std": 0.3818689755952286, "tokenization_path_logprob_std": 66.44207396776902, "trajectory_balance_rmse": 311.2746888544801} {"canonical_length": 58, "canonical_nll": 0.4493723902209052, "generation_rouge_l_f1": 0.1527777777777778, "generation_token_f1": 0.1736111111111111, "prompt_id": "ee3925e13ca1cdfe", "reading_canonical_log_prob": -26.0635986328125, "reading_distances": [21, 20, 14, 22], "reading_log_probs": [-26.2611083984375, -27.356021881103516, -27.66820526123047, -26.798179626464844], "reading_logprob_std": 0.621342207698893, "tokenization_path_logprob_std": 39.62176419813533, "trajectory_balance_rmse": 1011.883638176189} {"canonical_length": 293, "canonical_nll": 1.3210045095189846, "generation_rouge_l_f1": 0.17863397548161122, "generation_token_f1": 0.30472854640980734, "prompt_id": "e5cf7a10fe204155", "reading_canonical_log_prob": -387.0543212890625, "reading_distances": [28, 24, 27, 30], "reading_log_probs": [-395.1435241699219, -391.9455871582031, -398.01953125, -392.747314453125], "reading_logprob_std": 2.7320912671035185, "tokenization_path_logprob_std": 219.96922092985272, "trajectory_balance_rmse": 7558.532956449909} {"canonical_length": 302, "canonical_nll": 1.3243183868610306, "generation_rouge_l_f1": 0.17863397548161122, "generation_token_f1": 0.3467600700525394, "prompt_id": "a6e94611dc8387e2", "reading_canonical_log_prob": -399.94415283203125, "reading_distances": [32, 34, 29, 28], "reading_log_probs": [-392.8511962890625, -399.0244140625, -395.7996520996094, -395.212158203125], "reading_logprob_std": 2.5438436147841577, "tokenization_path_logprob_std": 319.6036500483788, "trajectory_balance_rmse": 8860.294953889497} {"canonical_length": 195, "canonical_nll": 1.9164400540865385, "generation_rouge_l_f1": 0.2344497607655502, "generation_token_f1": 0.45454545454545453, "prompt_id": "747395c5d51f280d", "reading_canonical_log_prob": -373.705810546875, "reading_distances": [21, 17, 16, 15], "reading_log_probs": [-393.1099853515625, -394.14471435546875, -390.44439697265625, -394.51397705078125], "reading_logprob_std": 1.8379654488211892, "tokenization_path_logprob_std": 225.26790350183512, "trajectory_balance_rmse": 5046.23165343822} {"canonical_length": 35, "canonical_nll": 0.5696686880929129, "generation_rouge_l_f1": 0.5405405405405406, "generation_token_f1": 0.5405405405405406, "prompt_id": "c31ebbb64961b272", "reading_canonical_log_prob": -19.938404083251953, "reading_distances": [16, 17, 15, 20], "reading_log_probs": [-20.105308532714844, -21.499561309814453, -18.888818740844727, -20.946250915527344], "reading_logprob_std": 1.1360064780720731, "tokenization_path_logprob_std": 29.265717589101705, "trajectory_balance_rmse": 464.95357431890653} {"canonical_length": 111, "canonical_nll": 1.4196991791596283, "generation_rouge_l_f1": 0.2918918918918919, "generation_token_f1": 0.44324324324324327, "prompt_id": "0a9b1baeb01418e9", "reading_canonical_log_prob": -157.58660888671875, "reading_distances": [25, 28, 28, 22], "reading_log_probs": [-162.34454345703125, -163.75082397460938, -160.66534423828125, -160.16604614257812], "reading_logprob_std": 1.6371545782275416, "tokenization_path_logprob_std": 195.6989754458216, "trajectory_balance_rmse": 3258.858543959736} {"canonical_length": 86, "canonical_nll": 1.1858806166537972, "generation_rouge_l_f1": 0.3258426966292135, "generation_token_f1": 0.49438202247191015, "prompt_id": "ffff4b859d3e455e", "reading_canonical_log_prob": -101.98573303222656, "reading_distances": [18, 19, 17, 19], "reading_log_probs": [-102.01107025146484, -102.23104858398438, -101.98342895507812, -101.28817749023438], "reading_logprob_std": 0.4088015443239537, "tokenization_path_logprob_std": 105.68116280704335, "trajectory_balance_rmse": 2899.151933752042} {"canonical_length": 12, "canonical_nll": 0.5901928742726644, "generation_rouge_l_f1": 0.17307692307692307, "generation_token_f1": 0.17307692307692307, "prompt_id": "a9db5a2fccb43cd0", "reading_canonical_log_prob": -7.082314491271973, "reading_distances": [21, 28, 23, 23], "reading_log_probs": [-7.298975944519043, -8.452868461608887, -8.072940826416016, -8.816919326782227], "reading_logprob_std": 0.6496809307002287, "tokenization_path_logprob_std": 109.7026230084267, "trajectory_balance_rmse": 427.66229816600475} {"canonical_length": 115, "canonical_nll": 2.013282311480978, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "6d2c7ff3221f2284", "reading_canonical_log_prob": -231.5274658203125, "reading_distances": [11, 13, 15, 13], "reading_log_probs": [-230.5972900390625, -232.59317016601562, -229.335205078125, -231.11383056640625], "reading_logprob_std": 1.34813144879094, "tokenization_path_logprob_std": 207.1990255550992, "trajectory_balance_rmse": 2108.2966077456895} {"canonical_length": 16, "canonical_nll": 1.1701128482818604, "generation_rouge_l_f1": 0.5517241379310344, "generation_token_f1": 0.5517241379310344, "prompt_id": "686af343746a5e07", "reading_canonical_log_prob": -18.721805572509766, "reading_distances": [22, 25, 24, 26], "reading_log_probs": [-27.78120994567871, -27.02002716064453, -26.98904037475586, -25.552387237548828], "reading_logprob_std": 0.9306575581730187, "tokenization_path_logprob_std": 85.65200865458137, "trajectory_balance_rmse": 546.2669582441637} {"canonical_length": 8, "canonical_nll": 5.1013288497924805, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.5714285714285715, "prompt_id": "90063df346a3b4e5", "reading_canonical_log_prob": -40.810630798339844, "reading_distances": [19, 18, 22, 26], "reading_log_probs": [-35.076080322265625, -34.39630889892578, -36.441829681396484, -36.632015228271484], "reading_logprob_std": 1.0788497940136457, "tokenization_path_logprob_std": 25.276497614822638, "trajectory_balance_rmse": 295.5619688972774} {"canonical_length": 276, "canonical_nll": 0.8973565585371377, "generation_rouge_l_f1": 0.33712121212121215, "generation_token_f1": 0.6098484848484849, "prompt_id": "73f47d65b9b8ef25", "reading_canonical_log_prob": -247.67041015625, "reading_distances": [25, 20, 25, 29], "reading_log_probs": [-264.36297607421875, -262.9255065917969, -270.080078125, -260.07684326171875], "reading_logprob_std": 4.208027974203477, "tokenization_path_logprob_std": 457.7980410054989, "trajectory_balance_rmse": 6717.123046840332} {"canonical_length": 73, "canonical_nll": 0.9709608522180009, "generation_rouge_l_f1": 0.4761904761904762, "generation_token_f1": 0.6031746031746033, "prompt_id": "96bcd59d4201d57e", "reading_canonical_log_prob": -70.88014221191406, "reading_distances": [19, 20, 18, 20], "reading_log_probs": [-72.03573608398438, -75.29405212402344, -70.72805786132812, -74.13092041015625], "reading_logprob_std": 2.0513632789848297, "tokenization_path_logprob_std": 120.02610743813386, "trajectory_balance_rmse": 1707.2266316820721} {"canonical_length": 159, "canonical_nll": 1.5152085022356525, "generation_rouge_l_f1": 0.3085399449035813, "generation_token_f1": 0.4958677685950413, "prompt_id": "3d7e3fa66ce5f5da", "reading_canonical_log_prob": -240.91815185546875, "reading_distances": [24, 22, 25, 26], "reading_log_probs": [-230.85546875, -233.44976806640625, -231.7967071533203, -229.20843505859375], "reading_logprob_std": 1.773642841665395, "tokenization_path_logprob_std": 145.27846876832655, "trajectory_balance_rmse": 5144.052814457598} {"canonical_length": 40, "canonical_nll": 1.0416020393371581, "generation_rouge_l_f1": 0.2608695652173913, "generation_token_f1": 0.2608695652173913, "prompt_id": "d99d42ddbcbf4292", "reading_canonical_log_prob": -41.66408157348633, "reading_distances": [18, 15, 21, 17], "reading_log_probs": [-37.67708206176758, -38.327857971191406, -36.51911926269531, -41.054786682128906], "reading_logprob_std": 1.9246911111611027, "tokenization_path_logprob_std": 103.94932360725231, "trajectory_balance_rmse": 931.2956006166233} {"canonical_length": 15, "canonical_nll": 0.12205464045206706, "generation_rouge_l_f1": 0.11976047904191617, "generation_token_f1": 0.11976047904191617, "prompt_id": "4bb936b473e4256b", "reading_canonical_log_prob": -1.8308196067810059, "reading_distances": [17, 17, 23, 18], "reading_log_probs": [-16.926509857177734, -17.85553741455078, -5.222616672515869, -2.2553060054779053], "reading_logprob_std": 7.983583517779583, "tokenization_path_logprob_std": 66.49205780399492, "trajectory_balance_rmse": 321.8825653705532} {"canonical_length": 378, "canonical_nll": 1.439997435877563, "generation_rouge_l_f1": 0.275974025974026, "generation_token_f1": 0.5454545454545454, "prompt_id": "d413361106deda8e", "reading_canonical_log_prob": -544.3190307617188, "reading_distances": [24, 24, 35, 20], "reading_log_probs": [-549.3592529296875, -547.2599487304688, -551.0069580078125, -547.403076171875], "reading_logprob_std": 1.779446682013028, "tokenization_path_logprob_std": 178.6888526339603, "trajectory_balance_rmse": 10169.45019300753} {"canonical_length": 370, "canonical_nll": 0.6573937493401605, "generation_rouge_l_f1": 0.38888888888888884, "generation_token_f1": 0.5243055555555556, "prompt_id": "2807bad41faf5ef6", "reading_canonical_log_prob": -243.23568725585938, "reading_distances": [20, 12, 12, 15], "reading_log_probs": [-247.08538818359375, -246.52003479003906, -246.4070281982422, -247.41610717773438], "reading_logprob_std": 0.47636893817081516, "tokenization_path_logprob_std": 215.82188420368124, "trajectory_balance_rmse": 7437.138186336476} {"canonical_length": 76, "canonical_nll": 1.8105541028474506, "generation_rouge_l_f1": 0.19571865443425074, "generation_token_f1": 0.3180428134556575, "prompt_id": "f47b46ea69a09ffc", "reading_canonical_log_prob": -137.60211181640625, "reading_distances": [26, 30, 23, 20], "reading_log_probs": [-132.6056365966797, -136.8907470703125, -135.28240966796875, -136.68600463867188], "reading_logprob_std": 1.9743234433893662, "tokenization_path_logprob_std": 128.6287253070968, "trajectory_balance_rmse": 2370.6264973206676} {"canonical_length": 148, "canonical_nll": 1.5141620120486698, "generation_rouge_l_f1": 0.27213822894168466, "generation_token_f1": 0.3758099352051835, "prompt_id": "abdba20a44568a50", "reading_canonical_log_prob": -224.09597778320312, "reading_distances": [31, 31, 31, 29], "reading_log_probs": [-227.47178649902344, -225.79299926757812, -226.64840698242188, -229.95382690429688], "reading_logprob_std": 1.7941296592697726, "tokenization_path_logprob_std": 306.69131791901276, "trajectory_balance_rmse": 4517.466619607264} {"canonical_length": 19, "canonical_nll": 0.4245282725283974, "generation_rouge_l_f1": 0.21686746987951808, "generation_token_f1": 0.21686746987951808, "prompt_id": "f10790becafaebfa", "reading_canonical_log_prob": -8.06603717803955, "reading_distances": [31, 29, 23, 29], "reading_log_probs": [-11.468867301940918, -11.453824996948242, -10.508212089538574, -12.73996639251709], "reading_logprob_std": 0.9159625326688461, "tokenization_path_logprob_std": 113.98301008419223, "trajectory_balance_rmse": 362.45245761307285} {"canonical_length": 7, "canonical_nll": 0.0129493847489357, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "3eb075dbec50f70d", "reading_canonical_log_prob": -0.0906456932425499, "reading_distances": [65, 69, 68, 66], "reading_log_probs": [-0.06441041827201843, -0.03294297307729721, -0.049041975289583206, -0.02045549266040325], "reading_logprob_std": 0.019128352997604712, "tokenization_path_logprob_std": 74.72349172202678, "trajectory_balance_rmse": 214.39742631633496} {"canonical_length": 5, "canonical_nll": 2.2904150009155275, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "56ce6de59318b6ea", "reading_canonical_log_prob": -11.452075004577637, "reading_distances": [33, 24, 33, 24], "reading_log_probs": [-15.046140670776367, -14.238338470458984, -17.115581512451172, -13.822835922241211], "reading_logprob_std": 1.464153558429268, "tokenization_path_logprob_std": 56.179038363141636, "trajectory_balance_rmse": 117.36972325247899} {"canonical_length": 128, "canonical_nll": 1.4023208618164062, "generation_rouge_l_f1": 0.2005141388174807, "generation_token_f1": 0.40616966580976865, "prompt_id": "7b4da759d302cb91", "reading_canonical_log_prob": -179.4970703125, "reading_distances": [51, 42, 40, 46], "reading_log_probs": [-176.31723022460938, -177.71524047851562, -173.93141174316406, -177.84384155273438], "reading_logprob_std": 1.8170055446890594, "tokenization_path_logprob_std": 88.97148835867914, "trajectory_balance_rmse": 3319.119014473541} {"canonical_length": 179, "canonical_nll": 2.1361288571490924, "generation_rouge_l_f1": 0.34640522875817, "generation_token_f1": 0.5098039215686275, "prompt_id": "e1ce0d3f6081406f", "reading_canonical_log_prob": -382.3670654296875, "reading_distances": [15, 18, 20, 21], "reading_log_probs": [-380.696044921875, -379.03448486328125, -379.41912841796875, -379.34161376953125], "reading_logprob_std": 0.7345105333168471, "tokenization_path_logprob_std": 372.09838047586703, "trajectory_balance_rmse": 5882.030037258212} {"canonical_length": 193, "canonical_nll": 1.1014791696182804, "generation_rouge_l_f1": 0.15625, "generation_token_f1": 0.23958333333333337, "prompt_id": "50afdbd8ae3129b4", "reading_canonical_log_prob": -212.58547973632812, "reading_distances": [20, 20, 21, 18], "reading_log_probs": [-207.9947509765625, -204.15823364257812, -204.94760131835938, -204.45816040039062], "reading_logprob_std": 1.7669198481330917, "tokenization_path_logprob_std": 64.06518432374203, "trajectory_balance_rmse": 3339.9001193923687} {"canonical_length": 308, "canonical_nll": 0.9583339938869724, "generation_rouge_l_f1": 0.22586520947176686, "generation_token_f1": 0.4371584699453552, "prompt_id": "517e8812395e8d56", "reading_canonical_log_prob": -295.1668701171875, "reading_distances": [13, 15, 16, 16], "reading_log_probs": [-291.95001220703125, -292.6839904785156, -295.47576904296875, -297.1611022949219], "reading_logprob_std": 2.429061053386789, "tokenization_path_logprob_std": 400.4171517995713, "trajectory_balance_rmse": 6704.607350716221} {"canonical_length": 9, "canonical_nll": 2.6913935343424478, "generation_rouge_l_f1": 0.04790419161676646, "generation_token_f1": 0.04790419161676646, "prompt_id": "7febc4a2c9567c65", "reading_canonical_log_prob": -24.22254180908203, "reading_distances": [33, 34, 34, 39], "reading_log_probs": [-27.617624282836914, -26.093294143676758, -27.812484741210938, -32.03327178955078], "reading_logprob_std": 2.548095529451106, "tokenization_path_logprob_std": 60.28236411420153, "trajectory_balance_rmse": 151.3928083631667} {"canonical_length": 13, "canonical_nll": 1.035433108990009, "generation_rouge_l_f1": 0.4761904761904762, "generation_token_f1": 0.761904761904762, "prompt_id": "a3b4630e36098b06", "reading_canonical_log_prob": -13.460630416870117, "reading_distances": [39, 44, 46, 44], "reading_log_probs": [-15.69371509552002, -14.83864688873291, -15.018728256225586, -12.737902641296387], "reading_logprob_std": 1.2770831777203338, "tokenization_path_logprob_std": 93.53364248379575, "trajectory_balance_rmse": 489.64004216961325} {"canonical_length": 67, "canonical_nll": 1.144871725964902, "generation_rouge_l_f1": 0.20833333333333331, "generation_token_f1": 0.32142857142857145, "prompt_id": "0d478580ddc19eb6", "reading_canonical_log_prob": -76.70640563964844, "reading_distances": [26, 22, 27, 24], "reading_log_probs": [-73.23551177978516, -72.06657409667969, -75.14788818359375, -80.56761932373047], "reading_logprob_std": 3.7629681540280715, "tokenization_path_logprob_std": 173.4054691322093, "trajectory_balance_rmse": 2125.712887495611} {"canonical_length": 5, "canonical_nll": 2.9600343704223633, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "7c79a39b53212c05", "reading_canonical_log_prob": -14.800171852111816, "reading_distances": [33, 32, 33, 33], "reading_log_probs": [-15.041380882263184, -15.802328109741211, -15.481990814208984, -16.217626571655273], "reading_logprob_std": 0.49774328054730477, "tokenization_path_logprob_std": 26.766455551703327, "trajectory_balance_rmse": 127.79835810594803} {"canonical_length": 89, "canonical_nll": 1.036820143796085, "generation_rouge_l_f1": 0.3880597014925373, "generation_token_f1": 0.5472636815920398, "prompt_id": "1791e9720c6a20af", "reading_canonical_log_prob": -92.27699279785156, "reading_distances": [25, 25, 25, 20], "reading_log_probs": [-96.13014221191406, -94.1748046875, -93.91270446777344, -96.72496795654297], "reading_logprob_std": 1.4016373700134104, "tokenization_path_logprob_std": 173.7379973346251, "trajectory_balance_rmse": 3049.8745513016015} {"canonical_length": 14, "canonical_nll": 1.6526508331298828, "generation_rouge_l_f1": 0.17910447761194032, "generation_token_f1": 0.17910447761194032, "prompt_id": "a58d338f47d878b7", "reading_canonical_log_prob": -23.13711166381836, "reading_distances": [9, 9, 9, 8], "reading_log_probs": [-26.403963088989258, -22.48569107055664, -28.885608673095703, -25.18457794189453], "reading_logprob_std": 2.660496267762951, "tokenization_path_logprob_std": 77.5201974875506, "trajectory_balance_rmse": 345.73205467218787} {"canonical_length": 5, "canonical_nll": 0.9958343505859375, "generation_rouge_l_f1": 0.4285714285714285, "generation_token_f1": 0.4285714285714285, "prompt_id": "c979ade41611b06d", "reading_canonical_log_prob": -4.9791717529296875, "reading_distances": [30, 29, 36, 38], "reading_log_probs": [-17.0125675201416, -5.180399417877197, -11.707365989685059, -7.701155185699463], "reading_logprob_std": 5.1627914947364815, "tokenization_path_logprob_std": 22.125322124484516, "trajectory_balance_rmse": 111.62271409255817} {"canonical_length": 197, "canonical_nll": 1.173801247843631, "generation_rouge_l_f1": 0.23162583518930957, "generation_token_f1": 0.47216035634743875, "prompt_id": "58bae05f46b8fa96", "reading_canonical_log_prob": -231.2388458251953, "reading_distances": [21, 21, 21, 18], "reading_log_probs": [-228.75546264648438, -229.36264038085938, -226.519287109375, -230.05467224121094], "reading_logprob_std": 1.5307842586270899, "tokenization_path_logprob_std": 209.42110542812097, "trajectory_balance_rmse": 5441.361661648663} {"canonical_length": 91, "canonical_nll": 1.1071299458598043, "generation_rouge_l_f1": 0.3636363636363637, "generation_token_f1": 0.5340909090909092, "prompt_id": "005445d3abeec9e6", "reading_canonical_log_prob": -100.74882507324219, "reading_distances": [18, 23, 20, 21], "reading_log_probs": [-102.53158569335938, -100.5255126953125, -99.94911193847656, -101.60423278808594], "reading_logprob_std": 1.1470535760851304, "tokenization_path_logprob_std": 223.97489990484982, "trajectory_balance_rmse": 2844.877245489115} {"canonical_length": 48, "canonical_nll": 1.095739761988322, "generation_rouge_l_f1": 0.37398373983739835, "generation_token_f1": 0.45528455284552843, "prompt_id": "3ba7853ae081fbb0", "reading_canonical_log_prob": -52.59550857543945, "reading_distances": [23, 12, 18, 19], "reading_log_probs": [-61.999908447265625, -53.36565399169922, -58.74610900878906, -63.20207595825195], "reading_logprob_std": 4.398322618372863, "tokenization_path_logprob_std": 54.55065303672012, "trajectory_balance_rmse": 1169.7289365222011} {"canonical_length": 60, "canonical_nll": 1.1671206156412761, "generation_rouge_l_f1": 0.31016042780748665, "generation_token_f1": 0.39572192513368987, "prompt_id": "19c4e3cd0eb041a5", "reading_canonical_log_prob": -70.02723693847656, "reading_distances": [36, 31, 31, 26], "reading_log_probs": [-73.24917602539062, -68.61016845703125, -68.2070541381836, -67.6611557006836], "reading_logprob_std": 2.5744010138917455, "tokenization_path_logprob_std": 211.1335705294486, "trajectory_balance_rmse": 2097.2694234343767} {"canonical_length": 139, "canonical_nll": 1.5349409803212117, "generation_rouge_l_f1": 0.25680933852140075, "generation_token_f1": 0.4747081712062257, "prompt_id": "9a7d94fcb2412368", "reading_canonical_log_prob": -213.35679626464844, "reading_distances": [22, 32, 29, 23], "reading_log_probs": [-216.03012084960938, -221.79629516601562, -218.44752502441406, -215.91908264160156], "reading_logprob_std": 2.757623494919705, "tokenization_path_logprob_std": 220.6538713767568, "trajectory_balance_rmse": 4539.877492493965} {"canonical_length": 27, "canonical_nll": 1.3436026396574798, "generation_rouge_l_f1": 0.23008849557522124, "generation_token_f1": 0.3185840707964602, "prompt_id": "2a25293cb727ba1d", "reading_canonical_log_prob": -36.27727127075195, "reading_distances": [31, 27, 32, 29], "reading_log_probs": [-37.33515167236328, -37.95391845703125, -37.2263069152832, -39.15308380126953], "reading_logprob_std": 0.8840947341367527, "tokenization_path_logprob_std": 90.49313941208332, "trajectory_balance_rmse": 917.6094178496311} {"canonical_length": 33, "canonical_nll": 1.7501677310828008, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.32727272727272727, "prompt_id": "cc6712ec80ac0d9e", "reading_canonical_log_prob": -57.75553512573242, "reading_distances": [29, 23, 29, 25], "reading_log_probs": [-58.19784164428711, -58.54653549194336, -58.47463607788086, -58.02801513671875], "reading_logprob_std": 0.24161725071967047, "tokenization_path_logprob_std": 78.654829322537, "trajectory_balance_rmse": 970.8741240678394} {"canonical_length": 48, "canonical_nll": 0.6104582945505778, "generation_rouge_l_f1": 0.2464788732394366, "generation_token_f1": 0.2605633802816901, "prompt_id": "aa6e8f58031f9e55", "reading_canonical_log_prob": -29.301998138427734, "reading_distances": [30, 38, 36, 39], "reading_log_probs": [-34.8170166015625, -35.66346740722656, -33.57160186767578, -35.934478759765625], "reading_logprob_std": 1.0625752638732917, "tokenization_path_logprob_std": 174.13362487716813, "trajectory_balance_rmse": 1346.5658550850392} {"canonical_length": 259, "canonical_nll": 1.614010137020391, "generation_rouge_l_f1": 0.30000000000000004, "generation_token_f1": 0.6153846153846154, "prompt_id": "ef58deb0251c5bd3", "reading_canonical_log_prob": -418.02862548828125, "reading_distances": [24, 25, 18, 23], "reading_log_probs": [-415.29315185546875, -417.8916015625, -417.8982849121094, -416.17218017578125], "reading_logprob_std": 1.2989496411757124, "tokenization_path_logprob_std": 325.8634231232709, "trajectory_balance_rmse": 7018.674171392416} {"canonical_length": 136, "canonical_nll": 1.6000534506405102, "generation_rouge_l_f1": 0.16120906801007556, "generation_token_f1": 0.29219143576826195, "prompt_id": "209efb2fe475a408", "reading_canonical_log_prob": -217.60726928710938, "reading_distances": [25, 28, 23, 24], "reading_log_probs": [-219.93963623046875, -220.7753143310547, -221.54640197753906, -220.48690795898438], "reading_logprob_std": 0.6695671716336951, "tokenization_path_logprob_std": 190.76573054149074, "trajectory_balance_rmse": 4919.126603012132} {"canonical_length": 16, "canonical_nll": 0.8779265284538269, "generation_rouge_l_f1": 0.1, "generation_token_f1": 0.1, "prompt_id": "aa1f86a1c5954cf7", "reading_canonical_log_prob": -14.04682445526123, "reading_distances": [28, 29, 27, 28], "reading_log_probs": [-16.470352172851562, -18.418575286865234, -17.346296310424805, -14.21987247467041], "reading_logprob_std": 1.7837442785536288, "tokenization_path_logprob_std": 24.377880082464273, "trajectory_balance_rmse": 213.83448230202734} {"canonical_length": 8, "canonical_nll": 0.006488771643489599, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "878c5da5498115bc", "reading_canonical_log_prob": -0.051910173147916794, "reading_distances": [32, 33, 32, 30], "reading_log_probs": [-0.2296288013458252, -0.09735178202390671, -0.6023217439651489, -0.23634371161460876], "reading_logprob_std": 0.21692849598741032, "tokenization_path_logprob_std": 77.21218936555375, "trajectory_balance_rmse": 187.03243144905963} {"canonical_length": 30, "canonical_nll": 0.2741636594136556, "generation_rouge_l_f1": 0.2647058823529412, "generation_token_f1": 0.35294117647058826, "prompt_id": "3e49b0b70cea783a", "reading_canonical_log_prob": -8.224909782409668, "reading_distances": [41, 34, 38, 39], "reading_log_probs": [-9.937516212463379, -9.799339294433594, -9.710201263427734, -9.339005470275879], "reading_logprob_std": 0.25603046536399116, "tokenization_path_logprob_std": 130.8291425693324, "trajectory_balance_rmse": 784.3094405190363} {"canonical_length": 406, "canonical_nll": 1.2810272066463977, "generation_rouge_l_f1": 0.21461897356143078, "generation_token_f1": 0.39813374805598756, "prompt_id": "50289ab6cb9cf49e", "reading_canonical_log_prob": -520.0970458984375, "reading_distances": [25, 21, 20, 23], "reading_log_probs": [-517.0467529296875, -518.4366455078125, -518.591552734375, -518.229248046875], "reading_logprob_std": 0.7020670172005695, "tokenization_path_logprob_std": 290.0562778174867, "trajectory_balance_rmse": 8663.685787098379} {"canonical_length": 168, "canonical_nll": 0.37979920705159503, "generation_rouge_l_f1": 0.2640449438202247, "generation_token_f1": 0.5, "prompt_id": "97290896fed51229", "reading_canonical_log_prob": -63.80626678466797, "reading_distances": [30, 30, 34, 39], "reading_log_probs": [-70.9596939086914, -65.91094207763672, -69.20416259765625, -71.7733383178711], "reading_logprob_std": 2.5988371985134293, "tokenization_path_logprob_std": 81.14568504619679, "trajectory_balance_rmse": 2052.667071664291} {"canonical_length": 36, "canonical_nll": 1.2091058095296223, "generation_rouge_l_f1": 0.17543859649122806, "generation_token_f1": 0.31578947368421045, "prompt_id": "11c3ac95688ce9b6", "reading_canonical_log_prob": -43.527809143066406, "reading_distances": [23, 21, 17, 18], "reading_log_probs": [-45.617984771728516, -48.53107452392578, -47.458274841308594, -51.6070556640625], "reading_logprob_std": 2.5094267447417957, "tokenization_path_logprob_std": 80.66309570446457, "trajectory_balance_rmse": 915.9116063074321} {"canonical_length": 12, "canonical_nll": 0.6926022370656332, "generation_rouge_l_f1": 0.08148148148148149, "generation_token_f1": 0.08148148148148149, "prompt_id": "b6d0ecb07aebae2f", "reading_canonical_log_prob": -8.311226844787598, "reading_distances": [26, 27, 26, 27], "reading_log_probs": [-9.898724555969238, -10.050006866455078, -10.240078926086426, -9.160816192626953], "reading_logprob_std": 0.4721858202607984, "tokenization_path_logprob_std": 61.73724375580603, "trajectory_balance_rmse": 306.5868109286721} {"canonical_length": 27, "canonical_nll": 1.3328450520833333, "generation_rouge_l_f1": 0.1754385964912281, "generation_token_f1": 0.3157894736842105, "prompt_id": "a7179a571daab3f1", "reading_canonical_log_prob": -35.98681640625, "reading_distances": [28, 19, 18, 19], "reading_log_probs": [-42.228118896484375, -43.80186462402344, -43.56129455566406, -41.32488250732422], "reading_logprob_std": 1.16421368822042, "tokenization_path_logprob_std": 73.65143911324225, "trajectory_balance_rmse": 499.6298773497606} {"canonical_length": 226, "canonical_nll": 1.0600118721480918, "generation_rouge_l_f1": 0.22319474835886216, "generation_token_f1": 0.2975929978118162, "prompt_id": "a32be7c1cc7860a7", "reading_canonical_log_prob": -239.56268310546875, "reading_distances": [38, 29, 34, 38], "reading_log_probs": [-258.2060241699219, -256.7657470703125, -256.3676452636719, -257.53466796875], "reading_logprob_std": 0.8173341149803425, "tokenization_path_logprob_std": 257.1328552838783, "trajectory_balance_rmse": 4426.1615857007755} {"canonical_length": 317, "canonical_nll": 1.226394893995219, "generation_rouge_l_f1": 0.2391681109185442, "generation_token_f1": 0.4610051993067591, "prompt_id": "7ebfd1446f72fed6", "reading_canonical_log_prob": -388.7671813964844, "reading_distances": [25, 29, 19, 24], "reading_log_probs": [-391.24114990234375, -388.3402099609375, -391.0755310058594, -392.060302734375], "reading_logprob_std": 1.617733505654379, "tokenization_path_logprob_std": 431.99785019014405, "trajectory_balance_rmse": 10040.896814701444} {"canonical_length": 7, "canonical_nll": 1.6093185969761439, "generation_rouge_l_f1": 0.7142857142857143, "generation_token_f1": 0.7142857142857143, "prompt_id": "05c61d7a9c151e3a", "reading_canonical_log_prob": -11.265230178833008, "reading_distances": [48, 57, 44, 50], "reading_log_probs": [-9.402942657470703, -12.131343841552734, -8.383155822753906, -6.401218891143799], "reading_logprob_std": 2.3858227642915257, "tokenization_path_logprob_std": 75.30265177023306, "trajectory_balance_rmse": 265.43125004873383} {"canonical_length": 20, "canonical_nll": 2.178165817260742, "generation_rouge_l_f1": 0.6451612903225806, "generation_token_f1": 0.7741935483870968, "prompt_id": "3b1c1940eb0db061", "reading_canonical_log_prob": -43.563316345214844, "reading_distances": [19, 25, 18, 21], "reading_log_probs": [-42.399635314941406, -41.994258880615234, -41.06982421875, -42.140167236328125], "reading_logprob_std": 0.578903716554189, "tokenization_path_logprob_std": 66.46604933360183, "trajectory_balance_rmse": 704.0357564755135} {"canonical_length": 191, "canonical_nll": 1.532073974609375, "generation_rouge_l_f1": 0.2488262910798122, "generation_token_f1": 0.4507042253521127, "prompt_id": "13ae7fb963a43898", "reading_canonical_log_prob": -292.6261291503906, "reading_distances": [42, 40, 46, 36], "reading_log_probs": [-298.9359130859375, -301.5796813964844, -296.0997009277344, -295.468994140625], "reading_logprob_std": 2.8110142444152806, "tokenization_path_logprob_std": 99.72373252077769, "trajectory_balance_rmse": 4236.686186142209} {"canonical_length": 346, "canonical_nll": 1.9295223875541907, "generation_rouge_l_f1": 0.18900343642611683, "generation_token_f1": 0.3848797250859107, "prompt_id": "6aeff1f7dd4bd317", "reading_canonical_log_prob": -667.61474609375, "reading_distances": [22, 24, 20, 26], "reading_log_probs": [-666.49853515625, -668.3555908203125, -670.082763671875, -667.85009765625], "reading_logprob_std": 1.4817084148168154, "tokenization_path_logprob_std": 626.9254331527466, "trajectory_balance_rmse": 8617.7873706956} {"canonical_length": 35, "canonical_nll": 3.0340401785714284, "generation_rouge_l_f1": 0.2631578947368421, "generation_token_f1": 0.3157894736842105, "prompt_id": "a5fa29779141ee07", "reading_canonical_log_prob": -106.19140625, "reading_distances": [15, 16, 16, 13], "reading_log_probs": [-118.27164459228516, -115.77548217773438, -110.46598815917969, -108.96822357177734], "reading_logprob_std": 4.382593762612159, "tokenization_path_logprob_std": 91.65707673727063, "trajectory_balance_rmse": 948.848212172997} {"canonical_length": 44, "canonical_nll": 1.6286811828613281, "generation_rouge_l_f1": 0.21212121212121213, "generation_token_f1": 0.31313131313131315, "prompt_id": "635690e839a377b9", "reading_canonical_log_prob": -71.66197204589844, "reading_distances": [32, 30, 23, 35], "reading_log_probs": [-74.9152603149414, -77.47444152832031, -77.58963012695312, -77.42125701904297], "reading_logprob_std": 1.2918374628163176, "tokenization_path_logprob_std": 131.19575924616413, "trajectory_balance_rmse": 1817.200967771621} {"canonical_length": 124, "canonical_nll": 1.6995081747731855, "generation_rouge_l_f1": 0.14177215189873416, "generation_token_f1": 0.2632911392405063, "prompt_id": "908852670070fd4b", "reading_canonical_log_prob": -210.739013671875, "reading_distances": [21, 20, 23, 20], "reading_log_probs": [-221.77999877929688, -205.17697143554688, -211.15670776367188, -208.09423828125], "reading_logprob_std": 7.242589782688068, "tokenization_path_logprob_std": 153.99817300513692, "trajectory_balance_rmse": 3900.7975015530683} {"canonical_length": 14, "canonical_nll": 1.5734054020472936, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "410cbaa296c6e2c6", "reading_canonical_log_prob": -22.02767562866211, "reading_distances": [25, 24, 23, 21], "reading_log_probs": [-20.66840362548828, -21.272733688354492, -19.781719207763672, -20.596755981445312], "reading_logprob_std": 0.6124345169376274, "tokenization_path_logprob_std": 26.287235034688862, "trajectory_balance_rmse": 467.451437564601} {"canonical_length": 24, "canonical_nll": 1.5049540201822917, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.22222222222222224, "prompt_id": "b9eb5c3d26d069d5", "reading_canonical_log_prob": -36.118896484375, "reading_distances": [15, 12, 16, 17], "reading_log_probs": [-39.695152282714844, -38.19356918334961, -41.694786071777344, -38.42494201660156], "reading_logprob_std": 1.6039184945975888, "tokenization_path_logprob_std": 78.44568310251658, "trajectory_balance_rmse": 635.5555966574505} {"canonical_length": 120, "canonical_nll": 0.8776435852050781, "generation_rouge_l_f1": 0.20833333333333331, "generation_token_f1": 0.34375, "prompt_id": "fd96f68afab8f1c0", "reading_canonical_log_prob": -105.31723022460938, "reading_distances": [32, 25, 24, 30], "reading_log_probs": [-113.85520935058594, -108.69953918457031, -110.89983367919922, -112.96543884277344], "reading_logprob_std": 2.298782145417389, "tokenization_path_logprob_std": 101.17818498903537, "trajectory_balance_rmse": 1882.8518890926016} {"canonical_length": 92, "canonical_nll": 3.0081127000891645, "generation_rouge_l_f1": 0.24390243902439027, "generation_token_f1": 0.43089430894308944, "prompt_id": "930a5444ff77055e", "reading_canonical_log_prob": -276.7463684082031, "reading_distances": [46, 50, 45, 43], "reading_log_probs": [-283.95245361328125, -293.3092956542969, -286.8825378417969, -287.619873046875], "reading_logprob_std": 3.9136753895755825, "tokenization_path_logprob_std": 314.7386684369182, "trajectory_balance_rmse": 3058.027426733552} {"canonical_length": 88, "canonical_nll": 0.9118608994917436, "generation_rouge_l_f1": 0.22372881355932206, "generation_token_f1": 0.3932203389830509, "prompt_id": "6881f74e519464d9", "reading_canonical_log_prob": -80.24375915527344, "reading_distances": [14, 19, 18, 17], "reading_log_probs": [-89.36138916015625, -83.07535552978516, -85.54161834716797, -87.16993713378906], "reading_logprob_std": 2.6521501138525605, "tokenization_path_logprob_std": 284.07205475890254, "trajectory_balance_rmse": 2465.3070367438163} {"canonical_length": 242, "canonical_nll": 1.4810324424554493, "generation_rouge_l_f1": 0.17928286852589642, "generation_token_f1": 0.46613545816733065, "prompt_id": "1afba92bdf32d325", "reading_canonical_log_prob": -358.40985107421875, "reading_distances": [32, 28, 26, 33], "reading_log_probs": [-352.654296875, -355.064208984375, -357.10662841796875, -353.57537841796875], "reading_logprob_std": 1.9437323185243258, "tokenization_path_logprob_std": 444.91086130665684, "trajectory_balance_rmse": 6754.899517732978} {"canonical_length": 114, "canonical_nll": 1.402738470780222, "generation_rouge_l_f1": 0.3128491620111732, "generation_token_f1": 0.43575418994413406, "prompt_id": "19d150b4a3d84266", "reading_canonical_log_prob": -159.9121856689453, "reading_distances": [35, 36, 33, 35], "reading_log_probs": [-171.99884033203125, -164.54278564453125, -169.0330810546875, -162.85186767578125], "reading_logprob_std": 4.176171927848596, "tokenization_path_logprob_std": 65.24379778165115, "trajectory_balance_rmse": 3208.9365685156154} {"canonical_length": 6, "canonical_nll": 2.9399843215942383, "generation_rouge_l_f1": 0.8000000000000002, "generation_token_f1": 0.8000000000000002, "prompt_id": "dee72943ee9496d3", "reading_canonical_log_prob": -17.63990592956543, "reading_distances": [38, 45, 40, 41], "reading_log_probs": [-18.01272964477539, -18.6535587310791, -18.928977966308594, -19.034582138061523], "reading_logprob_std": 0.45885203274901704, "tokenization_path_logprob_std": 34.036691495757765, "trajectory_balance_rmse": 178.36604395651375} {"canonical_length": 40, "canonical_nll": 0.9041601181030273, "generation_rouge_l_f1": 0.17993079584775087, "generation_token_f1": 0.20761245674740483, "prompt_id": "f1a8da244ce8319b", "reading_canonical_log_prob": -36.166404724121094, "reading_distances": [21, 23, 19, 25], "reading_log_probs": [-36.90155029296875, -38.36174774169922, -40.58810043334961, -36.88700866699219], "reading_logprob_std": 1.7452941942738607, "tokenization_path_logprob_std": 22.174726187509272, "trajectory_balance_rmse": 884.6884278323759} {"canonical_length": 286, "canonical_nll": 1.6510787643752731, "generation_rouge_l_f1": 0.24091778202676864, "generation_token_f1": 0.39388145315487577, "prompt_id": "201696a6ff512973", "reading_canonical_log_prob": -472.2085266113281, "reading_distances": [23, 21, 20, 17], "reading_log_probs": [-462.8565673828125, -469.5976257324219, -466.9768371582031, -465.0242919921875], "reading_logprob_std": 2.8681275476138675, "tokenization_path_logprob_std": 195.6388485620203, "trajectory_balance_rmse": 5692.056714665686} {"canonical_length": 202, "canonical_nll": 2.2513542552985766, "generation_rouge_l_f1": 0.2608695652173913, "generation_token_f1": 0.3752860411899313, "prompt_id": "8d88867a9f02da1b", "reading_canonical_log_prob": -454.7735595703125, "reading_distances": [25, 29, 29, 32], "reading_log_probs": [-452.69915771484375, -453.1302185058594, -447.41302490234375, -450.71990966796875], "reading_logprob_std": 2.6057278988671446, "tokenization_path_logprob_std": 157.76993333105298, "trajectory_balance_rmse": 4288.238944422043} {"canonical_length": 71, "canonical_nll": 0.9211330145177706, "generation_rouge_l_f1": 0.25327510917030566, "generation_token_f1": 0.3580786026200873, "prompt_id": "de7efec5b0e0f3a3", "reading_canonical_log_prob": -65.40044403076172, "reading_distances": [21, 25, 25, 22], "reading_log_probs": [-70.5926513671875, -67.69317626953125, -65.60435485839844, -64.94645690917969], "reading_logprob_std": 2.5414897468952997, "tokenization_path_logprob_std": 31.50092620519188, "trajectory_balance_rmse": 1248.234965573479} {"canonical_length": 8, "canonical_nll": 3.942725896835327, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "5d9fef90efd251af", "reading_canonical_log_prob": -31.541807174682617, "reading_distances": [11, 13, 11, 15], "reading_log_probs": [-35.558326721191406, -37.95977783203125, -35.5141716003418, -38.5452766418457], "reading_logprob_std": 1.5864578839443628, "tokenization_path_logprob_std": 40.011556202206314, "trajectory_balance_rmse": 265.77541609947036} {"canonical_length": 35, "canonical_nll": 2.0288604736328124, "generation_rouge_l_f1": 0.17910447761194032, "generation_token_f1": 0.23880597014925375, "prompt_id": "207d314006c9124a", "reading_canonical_log_prob": -71.01011657714844, "reading_distances": [27, 21, 21, 30], "reading_log_probs": [-83.38737487792969, -81.34605407714844, -86.96305847167969, -83.23681640625], "reading_logprob_std": 2.3449550832802086, "tokenization_path_logprob_std": 102.857584093395, "trajectory_balance_rmse": 1164.644014640571} {"canonical_length": 243, "canonical_nll": 0.8395834086853781, "generation_rouge_l_f1": 0.34220532319391633, "generation_token_f1": 0.5361216730038022, "prompt_id": "182ccb8ddd147837", "reading_canonical_log_prob": -204.01876831054688, "reading_distances": [16, 6, 14, 12], "reading_log_probs": [-205.84176635742188, -203.06423950195312, -198.27926635742188, -207.2855682373047], "reading_logprob_std": 3.9667333683667856, "tokenization_path_logprob_std": 178.85746888011772, "trajectory_balance_rmse": 2298.296666175624} {"canonical_length": 8, "canonical_nll": 2.118380546569824, "generation_rouge_l_f1": 0.027972027972027972, "generation_token_f1": 0.04195804195804196, "prompt_id": "e6557b0f23801243", "reading_canonical_log_prob": -16.947044372558594, "reading_distances": [8, 6, 8, 9], "reading_log_probs": [-17.630550384521484, -16.706714630126953, -16.303646087646484, -17.55889129638672], "reading_logprob_std": 0.6508707240814421, "tokenization_path_logprob_std": 37.307842711485044, "trajectory_balance_rmse": 260.10682571592633} {"canonical_length": 56, "canonical_nll": 0.6725326265607562, "generation_rouge_l_f1": 0.19658119658119658, "generation_token_f1": 0.22222222222222224, "prompt_id": "f461f97cdf966620", "reading_canonical_log_prob": -37.661827087402344, "reading_distances": [24, 22, 20, 22], "reading_log_probs": [-43.403865814208984, -44.55900573730469, -46.549903869628906, -45.07575607299805], "reading_logprob_std": 1.3048288117142848, "tokenization_path_logprob_std": 34.173697977449926, "trajectory_balance_rmse": 1103.5355786293346} {"canonical_length": 319, "canonical_nll": 1.1374527025372256, "generation_rouge_l_f1": 0.2757417102966841, "generation_token_f1": 0.49214659685863876, "prompt_id": "e632ea14cc0c360e", "reading_canonical_log_prob": -362.847412109375, "reading_distances": [12, 15, 14, 15], "reading_log_probs": [-362.21514892578125, -363.65643310546875, -360.8562316894531, -364.0032958984375], "reading_logprob_std": 1.4429968999879872, "tokenization_path_logprob_std": 205.78372814122005, "trajectory_balance_rmse": 7725.182412788217} {"canonical_length": 123, "canonical_nll": 1.410386000222307, "generation_rouge_l_f1": 0.23157894736842105, "generation_token_f1": 0.41052631578947374, "prompt_id": "fbb4ad27a6d06f7c", "reading_canonical_log_prob": -173.47747802734375, "reading_distances": [43, 40, 44, 45], "reading_log_probs": [-171.94972229003906, -172.3875732421875, -174.579833984375, -171.25363159179688], "reading_logprob_std": 1.4361173473924636, "tokenization_path_logprob_std": 211.8943592642448, "trajectory_balance_rmse": 3360.063068146023} {"canonical_length": 285, "canonical_nll": 1.144853451377467, "generation_rouge_l_f1": 0.25341130604288503, "generation_token_f1": 0.4834307992202729, "prompt_id": "db0aefd45e9cb0fd", "reading_canonical_log_prob": -326.2832336425781, "reading_distances": [23, 26, 25, 26], "reading_log_probs": [-326.70269775390625, -328.4105529785156, -322.4159240722656, -325.755859375], "reading_logprob_std": 2.52203361496644, "tokenization_path_logprob_std": 192.46947924519208, "trajectory_balance_rmse": 6399.01286910776} {"canonical_length": 32, "canonical_nll": 0.1815548539161682, "generation_rouge_l_f1": 0.9583333333333334, "generation_token_f1": 0.9583333333333334, "prompt_id": "5f8ab685b6a4b0df", "reading_canonical_log_prob": -5.809755325317383, "reading_distances": [27, 24, 32, 28], "reading_log_probs": [-6.285677433013916, -7.550305366516113, -6.8119893074035645, -9.016206741333008], "reading_logprob_std": 1.186192446541284, "tokenization_path_logprob_std": 125.59791276079052, "trajectory_balance_rmse": 518.3460305354852} {"canonical_length": 12, "canonical_nll": 1.3975696563720703, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4210526315789474, "prompt_id": "6c7141219130f618", "reading_canonical_log_prob": -16.770835876464844, "reading_distances": [22, 20, 26, 20], "reading_log_probs": [-15.208690643310547, -15.189582824707031, -17.478626251220703, -18.399158477783203], "reading_logprob_std": 1.625846780322482, "tokenization_path_logprob_std": 23.205656563984878, "trajectory_balance_rmse": 428.3718430459476} {"canonical_length": 8, "canonical_nll": 0.02010912448167801, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1d180306e638dc8c", "reading_canonical_log_prob": -0.16087299585342407, "reading_distances": [24, 26, 26, 24], "reading_log_probs": [-0.28329625725746155, -0.7610350847244263, -0.44272905588150024, -0.25079694390296936], "reading_logprob_std": 0.23331113594481992, "tokenization_path_logprob_std": 25.27991748128206, "trajectory_balance_rmse": 181.64097231837172} {"canonical_length": 162, "canonical_nll": 1.3581989429615162, "generation_rouge_l_f1": 0.2770780856423174, "generation_token_f1": 0.4836272040302267, "prompt_id": "042d56a64e3fa50f", "reading_canonical_log_prob": -220.02822875976562, "reading_distances": [29, 27, 31, 21], "reading_log_probs": [-223.54721069335938, -225.41244506835938, -218.73141479492188, -224.11666870117188], "reading_logprob_std": 2.919923420441047, "tokenization_path_logprob_std": 161.1337293551167, "trajectory_balance_rmse": 4863.978078503723} {"canonical_length": 278, "canonical_nll": 1.5362598227082396, "generation_rouge_l_f1": 0.27067669172932335, "generation_token_f1": 0.5300751879699248, "prompt_id": "519601f060a2cff2", "reading_canonical_log_prob": -427.0802307128906, "reading_distances": [22, 21, 30, 27], "reading_log_probs": [-426.98004150390625, -426.1268005371094, -426.6287536621094, -427.101806640625], "reading_logprob_std": 0.4370948605859106, "tokenization_path_logprob_std": 302.1667393729713, "trajectory_balance_rmse": 7894.038964126352} {"canonical_length": 41, "canonical_nll": 1.5292490517220847, "generation_rouge_l_f1": 0.21621621621621623, "generation_token_f1": 0.3963963963963964, "prompt_id": "acf7f177082b10c2", "reading_canonical_log_prob": -62.69921112060547, "reading_distances": [10, 11, 15, 17], "reading_log_probs": [-63.2205810546875, -62.36932373046875, -63.206756591796875, -65.18199157714844], "reading_logprob_std": 1.1932416926788665, "tokenization_path_logprob_std": 228.0265224092791, "trajectory_balance_rmse": 1207.4519132630599} {"canonical_length": 10, "canonical_nll": 1.4504454612731934, "generation_rouge_l_f1": 0.5454545454545455, "generation_token_f1": 0.6363636363636364, "prompt_id": "78f8a9aafe333ebf", "reading_canonical_log_prob": -14.504454612731934, "reading_distances": [23, 24, 29, 25], "reading_log_probs": [-14.982461929321289, -15.487945556640625, -14.980913162231445, -14.253146171569824], "reading_logprob_std": 0.5081726850294274, "tokenization_path_logprob_std": 63.72539902488354, "trajectory_balance_rmse": 373.19209011943} {"canonical_length": 303, "canonical_nll": 1.1616928050226898, "generation_rouge_l_f1": 0.1837455830388693, "generation_token_f1": 0.38162544169611307, "prompt_id": "c2a3bc6c2ee0dd9e", "reading_canonical_log_prob": -351.992919921875, "reading_distances": [30, 28, 31, 32], "reading_log_probs": [-350.16619873046875, -350.1254577636719, -348.429931640625, -347.565673828125], "reading_logprob_std": 1.2894849488085511, "tokenization_path_logprob_std": 99.38221575692761, "trajectory_balance_rmse": 8797.336316707402} {"canonical_length": 335, "canonical_nll": 1.3192129562150186, "generation_rouge_l_f1": 0.24707846410684475, "generation_token_f1": 0.4941569282136895, "prompt_id": "994646dcabc2bac8", "reading_canonical_log_prob": -441.93634033203125, "reading_distances": [20, 18, 23, 20], "reading_log_probs": [-453.0657958984375, -452.45477294921875, -450.9022216796875, -448.329345703125], "reading_logprob_std": 2.1122187997320037, "tokenization_path_logprob_std": 482.8548792189574, "trajectory_balance_rmse": 9670.011180184767} {"canonical_length": 10, "canonical_nll": 1.2739569664001464, "generation_rouge_l_f1": 0.7000000000000001, "generation_token_f1": 0.7000000000000001, "prompt_id": "bb2e66b420114e36", "reading_canonical_log_prob": -12.739569664001465, "reading_distances": [25, 24, 29, 26], "reading_log_probs": [-10.607705116271973, -10.598665237426758, -9.917587280273438, -11.749005317687988], "reading_logprob_std": 0.7593940571528062, "tokenization_path_logprob_std": 27.138681661925048, "trajectory_balance_rmse": 290.8531351459116} {"canonical_length": 282, "canonical_nll": 1.0287910623753325, "generation_rouge_l_f1": 0.31739961759082214, "generation_token_f1": 0.5621414913957935, "prompt_id": "57dc4fcf5ed48b59", "reading_canonical_log_prob": -290.11907958984375, "reading_distances": [20, 17, 18, 18], "reading_log_probs": [-293.57269287109375, -293.8985595703125, -292.8546447753906, -293.6495056152344], "reading_logprob_std": 0.4482625809146325, "tokenization_path_logprob_std": 429.97549555055514, "trajectory_balance_rmse": 8337.28209297157} {"canonical_length": 15, "canonical_nll": 0.7832682927449545, "generation_rouge_l_f1": 0.04878048780487805, "generation_token_f1": 0.06829268292682926, "prompt_id": "3b7a47e13ee4a0c9", "reading_canonical_log_prob": -11.749024391174316, "reading_distances": [16, 17, 24, 21], "reading_log_probs": [-9.225898742675781, -8.849424362182617, -10.127660751342773, -9.017763137817383], "reading_logprob_std": 0.5695266332605654, "tokenization_path_logprob_std": 40.07163538288214, "trajectory_balance_rmse": 306.28620192600454} {"canonical_length": 38, "canonical_nll": 1.2960023378071033, "generation_rouge_l_f1": 0.3492063492063492, "generation_token_f1": 0.5396825396825397, "prompt_id": "4ae552c8baec5eb5", "reading_canonical_log_prob": -49.24808883666992, "reading_distances": [40, 48, 42, 38], "reading_log_probs": [-53.54033279418945, -53.2020149230957, -50.1575813293457, -48.015716552734375], "reading_logprob_std": 2.627303790400081, "tokenization_path_logprob_std": 142.47548884403696, "trajectory_balance_rmse": 1096.2151094355122} {"canonical_length": 27, "canonical_nll": 1.5389207910608362, "generation_rouge_l_f1": 0.48979591836734687, "generation_token_f1": 0.48979591836734687, "prompt_id": "17ccb345d2ac2714", "reading_canonical_log_prob": -41.55086135864258, "reading_distances": [33, 36, 31, 36], "reading_log_probs": [-38.65275573730469, -38.24931716918945, -38.88121795654297, -39.30458450317383], "reading_logprob_std": 0.440829274201406, "tokenization_path_logprob_std": 99.28044847642745, "trajectory_balance_rmse": 835.2164681563446} {"canonical_length": 3, "canonical_nll": 4.047277450561523, "generation_rouge_l_f1": 0.08163265306122448, "generation_token_f1": 0.08163265306122448, "prompt_id": "5e428837a52a1f42", "reading_canonical_log_prob": -12.14183235168457, "reading_distances": [38, 45, 47, 50], "reading_log_probs": [-14.681921005249023, -17.918466567993164, -15.374731063842773, -14.161865234375], "reading_logprob_std": 1.6653167460710088, "tokenization_path_logprob_std": 36.743171093672075, "trajectory_balance_rmse": 74.19448949535868} {"canonical_length": 22, "canonical_nll": 1.220290097323331, "generation_rouge_l_f1": 0.4878048780487805, "generation_token_f1": 0.4878048780487805, "prompt_id": "6a0a62922b0dc5d9", "reading_canonical_log_prob": -26.84638214111328, "reading_distances": [25, 21, 22, 14], "reading_log_probs": [-27.220340728759766, -25.693870544433594, -27.206815719604492, -24.191452026367188], "reading_logprob_std": 1.447501975530785, "tokenization_path_logprob_std": 109.72481028671524, "trajectory_balance_rmse": 869.3654243326117} {"canonical_length": 138, "canonical_nll": 1.9733243195906929, "generation_rouge_l_f1": 0.32707774798927614, "generation_token_f1": 0.4611260053619303, "prompt_id": "b8e84fe41fdcb18b", "reading_canonical_log_prob": -272.3187561035156, "reading_distances": [64, 67, 61, 50], "reading_log_probs": [-267.9781799316406, -266.3877258300781, -268.3294982910156, -265.4212341308594], "reading_logprob_std": 1.3648421028000255, "tokenization_path_logprob_std": 243.23607440493768, "trajectory_balance_rmse": 3271.274302818588} {"canonical_length": 16, "canonical_nll": 1.090004324913025, "generation_rouge_l_f1": 0.18, "generation_token_f1": 0.22, "prompt_id": "b63558947632bbd1", "reading_canonical_log_prob": -17.4400691986084, "reading_distances": [27, 27, 22, 28], "reading_log_probs": [-14.683798789978027, -16.46735382080078, -13.17072868347168, -15.412027359008789], "reading_logprob_std": 1.3846069699232884, "tokenization_path_logprob_std": 66.71974192784532, "trajectory_balance_rmse": 357.39988233920974} {"canonical_length": 145, "canonical_nll": 1.3348186624461207, "generation_rouge_l_f1": 0.3253588516746412, "generation_token_f1": 0.5502392344497606, "prompt_id": "f6960da157e7277f", "reading_canonical_log_prob": -193.5487060546875, "reading_distances": [18, 13, 15, 21], "reading_log_probs": [-194.43017578125, -193.79681396484375, -195.8943328857422, -194.47451782226562], "reading_logprob_std": 0.8860780811194628, "tokenization_path_logprob_std": 211.25813841896604, "trajectory_balance_rmse": 4234.047712899297} {"canonical_length": 147, "canonical_nll": 1.6750625298947703, "generation_rouge_l_f1": 0.33444816053511706, "generation_token_f1": 0.5551839464882944, "prompt_id": "92be2f9693f01c35", "reading_canonical_log_prob": -246.23419189453125, "reading_distances": [70, 65, 62, 72], "reading_log_probs": [-249.17581176757812, -254.37762451171875, -241.25265502929688, -253.77774047851562], "reading_logprob_std": 6.058846608218536, "tokenization_path_logprob_std": 176.69206955702668, "trajectory_balance_rmse": 3483.533033223327} {"canonical_length": 21, "canonical_nll": 0.09265352998461042, "generation_rouge_l_f1": 0.9743589743589743, "generation_token_f1": 0.9743589743589743, "prompt_id": "45f3d6045c73150a", "reading_canonical_log_prob": -1.9457241296768188, "reading_distances": [33, 34, 22, 17], "reading_log_probs": [-1.3310542106628418, -1.6033363342285156, -1.330772876739502, -1.568810224533081], "reading_logprob_std": 0.14798936494653442, "tokenization_path_logprob_std": 124.20370468109806, "trajectory_balance_rmse": 503.71690184469065} {"canonical_length": 98, "canonical_nll": 1.3044520786830358, "generation_rouge_l_f1": 0.2586206896551725, "generation_token_f1": 0.39080459770114945, "prompt_id": "949e9bf0a5c54d25", "reading_canonical_log_prob": -127.8363037109375, "reading_distances": [17, 15, 17, 14], "reading_log_probs": [-129.4319610595703, -129.3155517578125, -130.80906677246094, -130.92105102539062], "reading_logprob_std": 0.8635255094860118, "tokenization_path_logprob_std": 215.34444747818634, "trajectory_balance_rmse": 2753.5026711044793} {"canonical_length": 14, "canonical_nll": 0.7225604057312012, "generation_rouge_l_f1": 0.47058823529411764, "generation_token_f1": 0.47058823529411764, "prompt_id": "5602e1c51a5b2686", "reading_canonical_log_prob": -10.115845680236816, "reading_distances": [19, 17, 19, 20], "reading_log_probs": [-10.047080039978027, -9.95195198059082, -10.500202178955078, -9.673060417175293], "reading_logprob_std": 0.3436061186138359, "tokenization_path_logprob_std": 88.59751298615093, "trajectory_balance_rmse": 480.89384174691014} {"canonical_length": 179, "canonical_nll": 1.393174496442912, "generation_rouge_l_f1": 0.23608017817371935, "generation_token_f1": 0.4409799554565702, "prompt_id": "94a9111b40eefe1b", "reading_canonical_log_prob": -249.37823486328125, "reading_distances": [37, 36, 37, 38], "reading_log_probs": [-261.70721435546875, -254.41360473632812, -258.33984375, -257.23236083984375], "reading_logprob_std": 3.015895350985769, "tokenization_path_logprob_std": 132.9062236175493, "trajectory_balance_rmse": 5348.149147749799} {"canonical_length": 137, "canonical_nll": 1.700241868513344, "generation_rouge_l_f1": 0.26804123711340205, "generation_token_f1": 0.45360824742268047, "prompt_id": "726f150c14796002", "reading_canonical_log_prob": -232.93313598632812, "reading_distances": [17, 16, 13, 22], "reading_log_probs": [-241.09445190429688, -237.90740966796875, -230.5432891845703, -235.70703125], "reading_logprob_std": 4.437010461778579, "tokenization_path_logprob_std": 281.92695294287967, "trajectory_balance_rmse": 4784.719490541344} {"canonical_length": 57, "canonical_nll": 0.29916368451034814, "generation_rouge_l_f1": 0.4285714285714286, "generation_token_f1": 0.5974025974025974, "prompt_id": "39559806a2fa846b", "reading_canonical_log_prob": -17.052330017089844, "reading_distances": [27, 21, 21, 22], "reading_log_probs": [-20.42394256591797, -19.758140563964844, -21.226062774658203, -22.6057071685791], "reading_logprob_std": 1.2252086813592875, "tokenization_path_logprob_std": 33.30566918333874, "trajectory_balance_rmse": 838.4801866933641} {"canonical_length": 13, "canonical_nll": 0.9257575548612155, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.18181818181818182, "prompt_id": "3dc6a3c96d3d9987", "reading_canonical_log_prob": -12.0348482131958, "reading_distances": [24, 34, 30, 30], "reading_log_probs": [-21.412111282348633, -22.80109977722168, -15.912315368652344, -18.568607330322266], "reading_logprob_std": 3.0644165945960857, "tokenization_path_logprob_std": 43.076261894302384, "trajectory_balance_rmse": 302.1628047855901} {"canonical_length": 14, "canonical_nll": 0.7455158233642578, "generation_rouge_l_f1": 0.5185185185185186, "generation_token_f1": 0.5185185185185186, "prompt_id": "8e74067a85468916", "reading_canonical_log_prob": -10.43722152709961, "reading_distances": [17, 17, 17, 23], "reading_log_probs": [-9.13205623626709, -9.452692031860352, -8.873990058898926, -9.620872497558594], "reading_logprob_std": 0.33283639904492135, "tokenization_path_logprob_std": 56.33600582411497, "trajectory_balance_rmse": 437.85679565273136} {"canonical_length": 27, "canonical_nll": 1.7113575405544705, "generation_rouge_l_f1": 0.11904761904761904, "generation_token_f1": 0.14285714285714285, "prompt_id": "3090724d0b15fbad", "reading_canonical_log_prob": -46.2066535949707, "reading_distances": [25, 17, 19, 26], "reading_log_probs": [-52.997562408447266, -49.32810974121094, -47.31188201904297, -46.02623748779297], "reading_logprob_std": 3.0415406181618874, "tokenization_path_logprob_std": 58.964773168381555, "trajectory_balance_rmse": 684.9871901517458} {"canonical_length": 273, "canonical_nll": 1.6462637094350963, "generation_rouge_l_f1": 0.22932330827067668, "generation_token_f1": 0.4398496240601503, "prompt_id": "d410f93c21deda57", "reading_canonical_log_prob": -449.42999267578125, "reading_distances": [22, 17, 18, 13], "reading_log_probs": [-455.2220458984375, -454.47467041015625, -454.1609802246094, -451.9205322265625], "reading_logprob_std": 1.4208581220395704, "tokenization_path_logprob_std": 495.2405474978842, "trajectory_balance_rmse": 8583.936188934165} {"canonical_length": 354, "canonical_nll": 1.220027255473164, "generation_rouge_l_f1": 0.2070116861435726, "generation_token_f1": 0.4106844741235392, "prompt_id": "710012fa888db2a9", "reading_canonical_log_prob": -431.8896484375, "reading_distances": [33, 26, 36, 26], "reading_log_probs": [-433.3198547363281, -429.58203125, -431.55615234375, -433.09149169921875], "reading_logprob_std": 1.7249360832390361, "tokenization_path_logprob_std": 211.57796402671946, "trajectory_balance_rmse": 9251.117785957422} {"canonical_length": 191, "canonical_nll": 1.5032271939422448, "generation_rouge_l_f1": 0.28633405639913234, "generation_token_f1": 0.5466377440347072, "prompt_id": "b61ed23f42bef95d", "reading_canonical_log_prob": -287.11639404296875, "reading_distances": [20, 19, 23, 25], "reading_log_probs": [-283.89862060546875, -284.2058410644531, -285.89544677734375, -286.78985595703125], "reading_logprob_std": 1.3775827029025551, "tokenization_path_logprob_std": 230.63396793230814, "trajectory_balance_rmse": 6015.795380829467} {"canonical_length": 255, "canonical_nll": 1.0717264811197917, "generation_rouge_l_f1": 0.2426614481409002, "generation_token_f1": 0.5753424657534246, "prompt_id": "0cd98f075a94d815", "reading_canonical_log_prob": -273.2902526855469, "reading_distances": [17, 29, 27, 20], "reading_log_probs": [-279.29949951171875, -274.9647216796875, -277.7007751464844, -275.56451416015625], "reading_logprob_std": 1.9938590016171263, "tokenization_path_logprob_std": 445.1573812670095, "trajectory_balance_rmse": 6970.512746689186} {"canonical_length": 315, "canonical_nll": 1.3980643136160715, "generation_rouge_l_f1": 0.21441124780316348, "generation_token_f1": 0.4358523725834798, "prompt_id": "3f6acc98eba7e06b", "reading_canonical_log_prob": -440.3902587890625, "reading_distances": [22, 23, 25, 22], "reading_log_probs": [-436.43939208984375, -439.16827392578125, -436.86865234375, -437.3327941894531], "reading_logprob_std": 1.2007601997494302, "tokenization_path_logprob_std": 152.45757588146856, "trajectory_balance_rmse": 10638.790121019947} {"canonical_length": 44, "canonical_nll": 0.8025907169688832, "generation_rouge_l_f1": 0.3212435233160622, "generation_token_f1": 0.3730569948186529, "prompt_id": "43a708261a941980", "reading_canonical_log_prob": -35.31399154663086, "reading_distances": [15, 12, 13, 12], "reading_log_probs": [-32.86272048950195, -33.67919921875, -33.461002349853516, -33.183311462402344], "reading_logprob_std": 0.35331572767976854, "tokenization_path_logprob_std": 164.17224006627976, "trajectory_balance_rmse": 1421.5099724604745} {"canonical_length": 186, "canonical_nll": 1.8453290385584677, "generation_rouge_l_f1": 0.2535211267605634, "generation_token_f1": 0.5070422535211268, "prompt_id": "6a69846ab5b6aa8a", "reading_canonical_log_prob": -343.231201171875, "reading_distances": [19, 25, 20, 24], "reading_log_probs": [-336.2008972167969, -335.49212646484375, -337.153076171875, -334.9565124511719], "reading_logprob_std": 0.9499132708777287, "tokenization_path_logprob_std": 321.86357451925386, "trajectory_balance_rmse": 5609.858105986884} {"canonical_length": 99, "canonical_nll": 1.2094154743233112, "generation_rouge_l_f1": 0.21935483870967745, "generation_token_f1": 0.34838709677419355, "prompt_id": "692adc91823ed2ff", "reading_canonical_log_prob": -119.73213195800781, "reading_distances": [29, 20, 22, 31], "reading_log_probs": [-114.95635223388672, -116.18955993652344, -115.0845718383789, -114.66059875488281], "reading_logprob_std": 0.6685298418106038, "tokenization_path_logprob_std": 119.81162067296528, "trajectory_balance_rmse": 3209.0065432019874} {"canonical_length": 11, "canonical_nll": 2.4977248798717153, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "9477d8c348a14e9c", "reading_canonical_log_prob": -27.474973678588867, "reading_distances": [24, 27, 24, 25], "reading_log_probs": [-27.63546371459961, -29.867015838623047, -26.6219539642334, -28.398284912109375], "reading_logprob_std": 1.3672317474611506, "tokenization_path_logprob_std": 61.49187966153905, "trajectory_balance_rmse": 281.4915777495172} {"canonical_length": 53, "canonical_nll": 0.6569052642246462, "generation_rouge_l_f1": 0.588235294117647, "generation_token_f1": 0.6588235294117646, "prompt_id": "7e54ccf509187c23", "reading_canonical_log_prob": -34.81597900390625, "reading_distances": [14, 14, 14, 14], "reading_log_probs": [-34.80470657348633, -33.50788879394531, -31.83101463317871, -36.00800704956055], "reading_logprob_std": 1.7907697608040711, "tokenization_path_logprob_std": 106.15886237721189, "trajectory_balance_rmse": 1131.2302531810187} {"canonical_length": 65, "canonical_nll": 1.3155421330378605, "generation_rouge_l_f1": 0.15081967213114753, "generation_token_f1": 0.21639344262295085, "prompt_id": "ba0c8c0d4edf2258", "reading_canonical_log_prob": -85.51023864746094, "reading_distances": [25, 25, 22, 20], "reading_log_probs": [-85.47532653808594, -86.99462127685547, -85.63092041015625, -86.12350463867188], "reading_logprob_std": 0.6839690709196716, "tokenization_path_logprob_std": 166.2013628384546, "trajectory_balance_rmse": 2053.0374720335644} {"canonical_length": 181, "canonical_nll": 1.6889302796421788, "generation_rouge_l_f1": 0.2660098522167487, "generation_token_f1": 0.4975369458128079, "prompt_id": "1f441ed2a6f2f793", "reading_canonical_log_prob": -305.6963806152344, "reading_distances": [21, 19, 24, 17], "reading_log_probs": [-302.52862548828125, -304.19775390625, -309.4490966796875, -302.3909912109375], "reading_logprob_std": 3.308520559988009, "tokenization_path_logprob_std": 313.54532169699877, "trajectory_balance_rmse": 4600.521932243936} {"canonical_length": 234, "canonical_nll": 1.5363701714409723, "generation_rouge_l_f1": 0.25870646766169153, "generation_token_f1": 0.5174129353233831, "prompt_id": "a1958dd73d8cbad2", "reading_canonical_log_prob": -359.5106201171875, "reading_distances": [16, 12, 15, 15], "reading_log_probs": [-359.39837646484375, -360.2559814453125, -363.12127685546875, -361.6125793457031], "reading_logprob_std": 1.628511893453301, "tokenization_path_logprob_std": 223.74086360232693, "trajectory_balance_rmse": 7083.374417205813} {"canonical_length": 248, "canonical_nll": 0.9583382145051034, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.6169772256728779, "prompt_id": "3e4f398cc2a68a1a", "reading_canonical_log_prob": -237.66787719726562, "reading_distances": [14, 16, 12, 12], "reading_log_probs": [-240.4288330078125, -240.21229553222656, -238.8101806640625, -240.1337890625], "reading_logprob_std": 0.7347337034404599, "tokenization_path_logprob_std": 362.286427937218, "trajectory_balance_rmse": 6142.228020052841} {"canonical_length": 7, "canonical_nll": 0.7373800277709961, "generation_rouge_l_f1": 0.7272727272727273, "generation_token_f1": 0.7272727272727273, "prompt_id": "96fa3fcdd459d2ed", "reading_canonical_log_prob": -5.161660194396973, "reading_distances": [14, 15, 19, 19], "reading_log_probs": [-5.623548984527588, -6.2146992683410645, -5.453571319580078, -5.830036640167236], "reading_logprob_std": 0.32787134749126, "tokenization_path_logprob_std": 6.95853666616892, "trajectory_balance_rmse": 149.9701274436281} {"canonical_length": 17, "canonical_nll": 0.44383873659021716, "generation_rouge_l_f1": 0.8749999999999999, "generation_token_f1": 0.8749999999999999, "prompt_id": "27d776f87907a327", "reading_canonical_log_prob": -7.545258522033691, "reading_distances": [23, 20, 24, 22], "reading_log_probs": [-11.544598579406738, -10.366247177124023, -12.868040084838867, -10.516258239746094], "reading_logprob_std": 1.1550556342533438, "tokenization_path_logprob_std": 90.9852361448697, "trajectory_balance_rmse": 474.88261169792355} {"canonical_length": 37, "canonical_nll": 1.2011121801427893, "generation_rouge_l_f1": 0.14559386973180077, "generation_token_f1": 0.1992337164750958, "prompt_id": "38bff471c2090786", "reading_canonical_log_prob": -44.4411506652832, "reading_distances": [40, 34, 37, 46], "reading_log_probs": [-46.912506103515625, -50.697566986083984, -46.40960693359375, -48.788597106933594], "reading_logprob_std": 1.953408827517215, "tokenization_path_logprob_std": 52.35266416337949, "trajectory_balance_rmse": 729.6608587775258} {"canonical_length": 60, "canonical_nll": 1.4597681681315104, "generation_rouge_l_f1": 0.2183406113537118, "generation_token_f1": 0.29694323144104806, "prompt_id": "68df2af9448f2f1d", "reading_canonical_log_prob": -87.58609008789062, "reading_distances": [31, 30, 28, 24], "reading_log_probs": [-86.17538452148438, -82.45149993896484, -83.47867584228516, -87.39984130859375], "reading_logprob_std": 2.3013684293047465, "tokenization_path_logprob_std": 148.79102254798192, "trajectory_balance_rmse": 2357.3102431948555} {"canonical_length": 14, "canonical_nll": 0.43161732809884207, "generation_rouge_l_f1": 0.14285714285714288, "generation_token_f1": 0.14285714285714288, "prompt_id": "6da3ca8bcb458aa8", "reading_canonical_log_prob": -6.042642593383789, "reading_distances": [15, 21, 15, 19], "reading_log_probs": [-4.670011520385742, -5.668505668640137, -4.837993621826172, -4.734574317932129], "reading_logprob_std": 0.46565817984564967, "tokenization_path_logprob_std": 59.92284040065749, "trajectory_balance_rmse": 379.2267489357492} {"canonical_length": 29, "canonical_nll": 0.9805637885784281, "generation_rouge_l_f1": 0.17355371900826447, "generation_token_f1": 0.20661157024793386, "prompt_id": "eba9c966fdb6009a", "reading_canonical_log_prob": -28.436349868774414, "reading_distances": [35, 30, 32, 34], "reading_log_probs": [-28.92717742919922, -29.082199096679688, -30.057580947875977, -33.697933197021484], "reading_logprob_std": 2.2280501559135097, "tokenization_path_logprob_std": 123.93209513083393, "trajectory_balance_rmse": 716.0538415855493} {"canonical_length": 237, "canonical_nll": 1.4260681409876055, "generation_rouge_l_f1": 0.2563025210084034, "generation_token_f1": 0.47899159663865554, "prompt_id": "16c0d54d78758bbb", "reading_canonical_log_prob": -337.9781494140625, "reading_distances": [20, 34, 21, 33], "reading_log_probs": [-346.7633361816406, -342.76220703125, -346.78143310546875, -349.88360595703125], "reading_logprob_std": 2.9188671139208493, "tokenization_path_logprob_std": 166.72324407390218, "trajectory_balance_rmse": 6711.626748057741} {"canonical_length": 40, "canonical_nll": 1.2165765762329102, "generation_rouge_l_f1": 0.5454545454545454, "generation_token_f1": 0.5454545454545454, "prompt_id": "51ab3680c9c02305", "reading_canonical_log_prob": -48.663063049316406, "reading_distances": [23, 18, 18, 19], "reading_log_probs": [-49.062435150146484, -48.820899963378906, -49.20079803466797, -47.79584884643555], "reading_logprob_std": 0.6357842242353801, "tokenization_path_logprob_std": 93.2055762332142, "trajectory_balance_rmse": 882.1420648728806} {"canonical_length": 133, "canonical_nll": 1.2200785471980733, "generation_rouge_l_f1": 0.2805194805194805, "generation_token_f1": 0.43116883116883115, "prompt_id": "a142f5697eaf1704", "reading_canonical_log_prob": -162.27044677734375, "reading_distances": [23, 19, 18, 26], "reading_log_probs": [-162.7783203125, -163.58285522460938, -161.54681396484375, -169.2344970703125], "reading_logprob_std": 3.403834770408701, "tokenization_path_logprob_std": 324.83507903066123, "trajectory_balance_rmse": 4568.335407269603} {"canonical_length": 248, "canonical_nll": 1.2831235085764239, "generation_rouge_l_f1": 0.22264150943396227, "generation_token_f1": 0.41509433962264153, "prompt_id": "879233d0df49ade7", "reading_canonical_log_prob": -318.2146301269531, "reading_distances": [21, 27, 28, 22], "reading_log_probs": [-316.5818176269531, -317.0006103515625, -313.2514953613281, -315.2449951171875], "reading_logprob_std": 1.687345486740947, "tokenization_path_logprob_std": 132.43033908099113, "trajectory_balance_rmse": 6101.667387400307} {"canonical_length": 189, "canonical_nll": 1.3650698384279927, "generation_rouge_l_f1": 0.31100478468899523, "generation_token_f1": 0.5311004784688995, "prompt_id": "f6b26f3fe5bcfb2a", "reading_canonical_log_prob": -257.9981994628906, "reading_distances": [8, 13, 18, 16], "reading_log_probs": [-254.66265869140625, -253.36065673828125, -247.33338928222656, -249.3472442626953], "reading_logprob_std": 3.417578020026152, "tokenization_path_logprob_std": 133.34029860842904, "trajectory_balance_rmse": 3841.0649006539843} {"canonical_length": 12, "canonical_nll": 1.487505276997884, "generation_rouge_l_f1": 0.0625, "generation_token_f1": 0.078125, "prompt_id": "61b8f081f2257b08", "reading_canonical_log_prob": -17.85006332397461, "reading_distances": [43, 37, 38, 42], "reading_log_probs": [-24.111419677734375, -25.167068481445312, -24.248144149780273, -24.669780731201172], "reading_logprob_std": 0.4756004524740571, "tokenization_path_logprob_std": 36.75534480353266, "trajectory_balance_rmse": 361.07417607662575} {"canonical_length": 7, "canonical_nll": 0.265828047479902, "generation_rouge_l_f1": 0.7272727272727272, "generation_token_f1": 0.7272727272727272, "prompt_id": "756f69ca5eb428f4", "reading_canonical_log_prob": -1.860796332359314, "reading_distances": [39, 46, 37, 36], "reading_log_probs": [-4.823521137237549, -1.2292025089263916, -4.827121257781982, -2.815448522567749], "reading_logprob_std": 1.7430700387831841, "tokenization_path_logprob_std": 22.511087470931262, "trajectory_balance_rmse": 261.3547047633287} {"canonical_length": 59, "canonical_nll": 1.322351746639963, "generation_rouge_l_f1": 0.2556390977443609, "generation_token_f1": 0.4511278195488722, "prompt_id": "f3fd9530a70f7802", "reading_canonical_log_prob": -78.01875305175781, "reading_distances": [25, 30, 31, 37], "reading_log_probs": [-81.31431579589844, -86.46682739257812, -87.15290069580078, -87.68869018554688], "reading_logprob_std": 2.9371303155496697, "tokenization_path_logprob_std": 97.4965662368906, "trajectory_balance_rmse": 2166.6388811117886} {"canonical_length": 84, "canonical_nll": 0.9400483994256883, "generation_rouge_l_f1": 0.2992125984251968, "generation_token_f1": 0.3937007874015748, "prompt_id": "d431ad8fbed7b76c", "reading_canonical_log_prob": -78.96406555175781, "reading_distances": [27, 28, 21, 22], "reading_log_probs": [-80.28299713134766, -83.52249145507812, -82.25660705566406, -82.31957244873047], "reading_logprob_std": 1.341346982186544, "tokenization_path_logprob_std": 122.10889055751797, "trajectory_balance_rmse": 1794.8019057163488} {"canonical_length": 376, "canonical_nll": 1.486720308344415, "generation_rouge_l_f1": 0.19254658385093168, "generation_token_f1": 0.3260869565217392, "prompt_id": "455b2b67c74aee94", "reading_canonical_log_prob": -559.0068359375, "reading_distances": [27, 19, 18, 24], "reading_log_probs": [-555.750244140625, -554.081787109375, -558.7645263671875, -550.88623046875], "reading_logprob_std": 3.2880522434784383, "tokenization_path_logprob_std": 150.78916943244272, "trajectory_balance_rmse": 9727.152219020973} {"canonical_length": 307, "canonical_nll": 1.8158917256209284, "generation_rouge_l_f1": 0.19761499148211242, "generation_token_f1": 0.39522998296422485, "prompt_id": "b3e34a08eb2e7c34", "reading_canonical_log_prob": -557.478759765625, "reading_distances": [25, 26, 28, 27], "reading_log_probs": [-573.5586547851562, -579.7274169921875, -573.9324951171875, -563.9259033203125], "reading_logprob_std": 6.547162984792801, "tokenization_path_logprob_std": 244.89717725798516, "trajectory_balance_rmse": 8562.733264625072} {"canonical_length": 165, "canonical_nll": 1.4237380519057765, "generation_rouge_l_f1": 0.26760563380281693, "generation_token_f1": 0.4647887323943662, "prompt_id": "7461cdb891fcaab7", "reading_canonical_log_prob": -234.91677856445312, "reading_distances": [18, 18, 17, 23], "reading_log_probs": [-237.06515502929688, -234.95028686523438, -235.1611328125, -236.22891235351562], "reading_logprob_std": 0.9839021829426313, "tokenization_path_logprob_std": 228.19483699072316, "trajectory_balance_rmse": 4813.203421978814} {"canonical_length": 119, "canonical_nll": 1.3188526570296086, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.2842377260981912, "prompt_id": "eb07969998e95372", "reading_canonical_log_prob": -156.94346618652344, "reading_distances": [18, 19, 18, 17], "reading_log_probs": [-144.91836547851562, -191.802734375, -176.73245239257812, -179.24742126464844], "reading_logprob_std": 19.957952882081727, "tokenization_path_logprob_std": 121.58725882587638, "trajectory_balance_rmse": 2689.1731094056427} {"canonical_length": 134, "canonical_nll": 1.6614582574189598, "generation_rouge_l_f1": 0.2485875706214689, "generation_token_f1": 0.3954802259887006, "prompt_id": "b32326c583d0044a", "reading_canonical_log_prob": -222.63540649414062, "reading_distances": [23, 33, 27, 21], "reading_log_probs": [-216.6005859375, -217.70535278320312, -217.79840087890625, -217.75790405273438], "reading_logprob_std": 0.5779067985730814, "tokenization_path_logprob_std": 109.91864331548375, "trajectory_balance_rmse": 3445.2618169423226} {"canonical_length": 324, "canonical_nll": 1.2071189409420815, "generation_rouge_l_f1": 0.27450980392156865, "generation_token_f1": 0.5392156862745099, "prompt_id": "e86ca4a6dca6aa44", "reading_canonical_log_prob": -391.1065368652344, "reading_distances": [38, 31, 34, 32], "reading_log_probs": [-389.32464599609375, -389.33001708984375, -389.25225830078125, -388.15399169921875], "reading_logprob_std": 0.5752515382169366, "tokenization_path_logprob_std": 519.4686465689051, "trajectory_balance_rmse": 8497.321471340807} {"canonical_length": 17, "canonical_nll": 0.028142685399335975, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "39176769401db5cc", "reading_canonical_log_prob": -0.47842565178871155, "reading_distances": [14, 22, 19, 18], "reading_log_probs": [-0.36219608783721924, -1.0346249341964722, -0.3912959396839142, -0.7688072323799133], "reading_logprob_std": 0.3221523965910693, "tokenization_path_logprob_std": 111.67580663702485, "trajectory_balance_rmse": 467.8643438071167} {"canonical_length": 46, "canonical_nll": 1.7423968107804009, "generation_rouge_l_f1": 0.1111111111111111, "generation_token_f1": 0.19607843137254902, "prompt_id": "6614284ebae06e44", "reading_canonical_log_prob": -80.15025329589844, "reading_distances": [15, 22, 21, 18], "reading_log_probs": [-80.77100372314453, -82.91340637207031, -84.806640625, -79.17829895019531], "reading_logprob_std": 2.4601245938175604, "tokenization_path_logprob_std": 274.6366539528299, "trajectory_balance_rmse": 1330.7545244971284} {"canonical_length": 169, "canonical_nll": 2.1590756749260356, "generation_rouge_l_f1": 0.25925925925925924, "generation_token_f1": 0.47619047619047616, "prompt_id": "130451267d979653", "reading_canonical_log_prob": -364.8837890625, "reading_distances": [29, 24, 16, 21], "reading_log_probs": [-360.317626953125, -361.0224914550781, -359.8731689453125, -364.1512451171875], "reading_logprob_std": 1.9322478825760092, "tokenization_path_logprob_std": 102.4739424683277, "trajectory_balance_rmse": 4713.04598190056} {"canonical_length": 89, "canonical_nll": 1.9696022633756145, "generation_rouge_l_f1": 0.11764705882352941, "generation_token_f1": 0.2566844919786096, "prompt_id": "13cc5e7d950f14d1", "reading_canonical_log_prob": -175.2946014404297, "reading_distances": [24, 22, 20, 26], "reading_log_probs": [-174.52706909179688, -175.66244506835938, -174.36944580078125, -176.72940063476562], "reading_logprob_std": 1.1009017531199152, "tokenization_path_logprob_std": 198.17755281942067, "trajectory_balance_rmse": 2633.619349105529} {"canonical_length": 50, "canonical_nll": 1.415245361328125, "generation_rouge_l_f1": 0.3098591549295775, "generation_token_f1": 0.39436619718309857, "prompt_id": "aa5825f0f273154e", "reading_canonical_log_prob": -70.76226806640625, "reading_distances": [35, 32, 29, 29], "reading_log_probs": [-71.89633178710938, -70.37007141113281, -75.05786895751953, -74.13023376464844], "reading_logprob_std": 2.12700650116881, "tokenization_path_logprob_std": 23.025254516124853, "trajectory_balance_rmse": 1834.7766652575654} {"canonical_length": 217, "canonical_nll": 1.546357185609879, "generation_rouge_l_f1": 0.25820568927789933, "generation_token_f1": 0.49015317286652077, "prompt_id": "046e2cee68c73a19", "reading_canonical_log_prob": -335.55950927734375, "reading_distances": [23, 31, 24, 29], "reading_log_probs": [-336.2691650390625, -338.4152526855469, -337.3355712890625, -336.61279296875], "reading_logprob_std": 0.9485944022340328, "tokenization_path_logprob_std": 321.1188940929049, "trajectory_balance_rmse": 5633.254730892818} {"canonical_length": 114, "canonical_nll": 2.128209431966146, "generation_rouge_l_f1": 0.16045845272206305, "generation_token_f1": 0.28653295128939826, "prompt_id": "f13ae64aa84fa6f5", "reading_canonical_log_prob": -242.61587524414062, "reading_distances": [25, 29, 25, 26], "reading_log_probs": [-248.02191162109375, -248.0487060546875, -247.72239685058594, -245.7170867919922], "reading_logprob_std": 1.1167974691880314, "tokenization_path_logprob_std": 163.88812816862293, "trajectory_balance_rmse": 3055.1526504043227} {"canonical_length": 180, "canonical_nll": 1.52218017578125, "generation_rouge_l_f1": 0.22072072072072071, "generation_token_f1": 0.44594594594594594, "prompt_id": "55942a6d7a51de21", "reading_canonical_log_prob": -273.992431640625, "reading_distances": [17, 19, 19, 15], "reading_log_probs": [-271.9849548339844, -273.69549560546875, -272.29119873046875, -273.085205078125], "reading_logprob_std": 0.7748782687081939, "tokenization_path_logprob_std": 398.29096602726673, "trajectory_balance_rmse": 5783.424496803435} {"canonical_length": 158, "canonical_nll": 1.358128318303748, "generation_rouge_l_f1": 0.3205741626794259, "generation_token_f1": 0.5263157894736842, "prompt_id": "ab7cc228a99beb55", "reading_canonical_log_prob": -214.5842742919922, "reading_distances": [34, 32, 33, 32], "reading_log_probs": [-210.8031768798828, -214.4305419921875, -212.03872680664062, -212.37686157226562], "reading_logprob_std": 1.505920622609965, "tokenization_path_logprob_std": 126.59773939504791, "trajectory_balance_rmse": 4170.083802112481} {"canonical_length": 11, "canonical_nll": 0.43872837586836383, "generation_rouge_l_f1": 0.12244897959183675, "generation_token_f1": 0.12244897959183675, "prompt_id": "c2611a4dd26cfbb1", "reading_canonical_log_prob": -4.826012134552002, "reading_distances": [14, 11, 14, 16], "reading_log_probs": [-5.272530555725098, -4.881419658660889, -4.716747283935547, -5.51887845993042], "reading_logprob_std": 0.3650839180615889, "tokenization_path_logprob_std": 45.91367074126229, "trajectory_balance_rmse": 229.90428137798472} {"canonical_length": 138, "canonical_nll": 1.7435445370881453, "generation_rouge_l_f1": 0.29154518950437314, "generation_token_f1": 0.5364431486880467, "prompt_id": "6f0ecfb4c606f967", "reading_canonical_log_prob": -240.60914611816406, "reading_distances": [28, 22, 23, 24], "reading_log_probs": [-247.1893768310547, -247.83535766601562, -249.14419555664062, -261.88232421875], "reading_logprob_std": 6.960672609960471, "tokenization_path_logprob_std": 173.30236906220418, "trajectory_balance_rmse": 4178.670978914614} {"canonical_length": 202, "canonical_nll": 0.7453848961556312, "generation_rouge_l_f1": 0.4920634920634921, "generation_token_f1": 0.6150793650793651, "prompt_id": "5d860ca9ad882bc0", "reading_canonical_log_prob": -150.5677490234375, "reading_distances": [13, 10, 13, 6], "reading_log_probs": [-167.91090393066406, -161.6217041015625, -166.2477569580078, -156.22561645507812], "reading_logprob_std": 5.242673164440567, "tokenization_path_logprob_std": 271.3158311541978, "trajectory_balance_rmse": 4953.230907968339} {"canonical_length": 253, "canonical_nll": 1.7288813534461462, "generation_rouge_l_f1": 0.22727272727272727, "generation_token_f1": 0.42148760330578516, "prompt_id": "bef65ba5cd28353b", "reading_canonical_log_prob": -437.406982421875, "reading_distances": [38, 36, 31, 36], "reading_log_probs": [-432.05419921875, -434.7461853027344, -436.0799560546875, -437.5821838378906], "reading_logprob_std": 2.3468177310051113, "tokenization_path_logprob_std": 278.3902744754114, "trajectory_balance_rmse": 6617.272593863142} {"canonical_length": 49, "canonical_nll": 0.8226141637685348, "generation_rouge_l_f1": 0.24615384615384614, "generation_token_f1": 0.4461538461538461, "prompt_id": "ba40c7a321bb21e7", "reading_canonical_log_prob": -40.3080940246582, "reading_distances": [12, 16, 15, 19], "reading_log_probs": [-42.87849426269531, -41.6381721496582, -40.20037078857422, -44.111572265625], "reading_logprob_std": 1.6761489474815, "tokenization_path_logprob_std": 131.26490345468133, "trajectory_balance_rmse": 1347.0025454691095} {"canonical_length": 15, "canonical_nll": 0.9658104578653971, "generation_rouge_l_f1": 0.13114754098360656, "generation_token_f1": 0.1639344262295082, "prompt_id": "a165c584ac95ad42", "reading_canonical_log_prob": -14.487156867980957, "reading_distances": [26, 34, 24, 22], "reading_log_probs": [-17.292335510253906, -20.109159469604492, -18.4593505859375, -14.795648574829102], "reading_logprob_std": 2.2343501453832206, "tokenization_path_logprob_std": 39.320378168003586, "trajectory_balance_rmse": 456.801889020351} {"canonical_length": 242, "canonical_nll": 1.6554724322862862, "generation_rouge_l_f1": 0.21765913757700206, "generation_token_f1": 0.41478439425051333, "prompt_id": "9c4dbe029e6ba3bb", "reading_canonical_log_prob": -400.62432861328125, "reading_distances": [16, 16, 19, 19], "reading_log_probs": [-403.0313720703125, -406.7685852050781, -401.097412109375, -398.85894775390625], "reading_logprob_std": 3.3522552033721027, "tokenization_path_logprob_std": 329.319425641555, "trajectory_balance_rmse": 6713.5200209495315} {"canonical_length": 62, "canonical_nll": 0.7954822048064201, "generation_rouge_l_f1": 0.5137614678899082, "generation_token_f1": 0.6605504587155963, "prompt_id": "dcfe7ed6b29a3c6d", "reading_canonical_log_prob": -49.31989669799805, "reading_distances": [24, 17, 20, 19], "reading_log_probs": [-50.2242431640625, -50.6318359375, -50.87425231933594, -50.14698791503906], "reading_logprob_std": 0.34367729112538986, "tokenization_path_logprob_std": 114.03135552337832, "trajectory_balance_rmse": 1530.1911582989892} {"canonical_length": 148, "canonical_nll": 1.6629163381215688, "generation_rouge_l_f1": 0.23058823529411762, "generation_token_f1": 0.40470588235294114, "prompt_id": "c0cc49ea5faf89df", "reading_canonical_log_prob": -246.1116180419922, "reading_distances": [28, 23, 21, 26], "reading_log_probs": [-246.73643493652344, -252.64651489257812, -247.9322967529297, -247.03109741210938], "reading_logprob_std": 2.7540104536453685, "tokenization_path_logprob_std": 223.2001349487774, "trajectory_balance_rmse": 4909.59176233531} {"canonical_length": 14, "canonical_nll": 0.9798622131347656, "generation_rouge_l_f1": 0.043478260869565216, "generation_token_f1": 0.043478260869565216, "prompt_id": "adf9198c8ba81554", "reading_canonical_log_prob": -13.718070983886719, "reading_distances": [18, 15, 10, 13], "reading_log_probs": [-14.782997131347656, -16.10106658935547, -15.628997802734375, -14.35165786743164], "reading_logprob_std": 0.7934079929161953, "tokenization_path_logprob_std": 84.9884097680789, "trajectory_balance_rmse": 439.6602099280773} {"canonical_length": 280, "canonical_nll": 1.1188041687011718, "generation_rouge_l_f1": 0.2048417132216015, "generation_token_f1": 0.42830540037243947, "prompt_id": "7eceb236d5391917", "reading_canonical_log_prob": -313.2651672363281, "reading_distances": [21, 27, 22, 29], "reading_log_probs": [-311.1103820800781, -312.7447509765625, -310.09783935546875, -312.76910400390625], "reading_logprob_std": 1.3099030467098471, "tokenization_path_logprob_std": 207.72928588470538, "trajectory_balance_rmse": 7559.556367740374} {"canonical_length": 7, "canonical_nll": 2.1917310442243303, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.4444444444444444, "prompt_id": "8279f5e0aa9b5caf", "reading_canonical_log_prob": -15.342117309570312, "reading_distances": [22, 28, 24, 21], "reading_log_probs": [-14.85196590423584, -15.303973197937012, -15.205451011657715, -14.028518676757812], "reading_logprob_std": 0.579438537788091, "tokenization_path_logprob_std": 16.027031415283474, "trajectory_balance_rmse": 214.29506994765038} {"canonical_length": 323, "canonical_nll": 0.8092402880405863, "generation_rouge_l_f1": 0.1875, "generation_token_f1": 0.40808823529411764, "prompt_id": "f4edb7d5c1554a64", "reading_canonical_log_prob": -261.3846130371094, "reading_distances": [21, 21, 22, 13], "reading_log_probs": [-262.52813720703125, -261.26068115234375, -261.9860534667969, -264.41864013671875], "reading_logprob_std": 1.350637158510721, "tokenization_path_logprob_std": 162.464758142121, "trajectory_balance_rmse": 8897.108127133215} {"canonical_length": 18, "canonical_nll": 0.9960096147325304, "generation_rouge_l_f1": 0.21621621621621623, "generation_token_f1": 0.2702702702702703, "prompt_id": "39c97bd4effbfbe9", "reading_canonical_log_prob": -17.928173065185547, "reading_distances": [25, 27, 28, 26], "reading_log_probs": [-16.540983200073242, -17.18600845336914, -15.55964469909668, -17.655803680419922], "reading_logprob_std": 0.9074486498714297, "tokenization_path_logprob_std": 88.68422263884769, "trajectory_balance_rmse": 494.0129871753757} {"canonical_length": 32, "canonical_nll": 1.3331513404846191, "generation_rouge_l_f1": 0.1201716738197425, "generation_token_f1": 0.17167381974248927, "prompt_id": "e245da9a65e5ed0d", "reading_canonical_log_prob": -42.66084289550781, "reading_distances": [22, 20, 24, 17], "reading_log_probs": [-57.016326904296875, -54.075260162353516, -51.16529846191406, -51.936580657958984], "reading_logprob_std": 2.6192386407726618, "tokenization_path_logprob_std": 84.22954686644754, "trajectory_balance_rmse": 512.8394311010848} {"canonical_length": 244, "canonical_nll": 1.3283033527311732, "generation_rouge_l_f1": 0.2648221343873518, "generation_token_f1": 0.4980237154150198, "prompt_id": "1dc2c6d9af2ea2a5", "reading_canonical_log_prob": -324.10601806640625, "reading_distances": [15, 15, 18, 18], "reading_log_probs": [-316.4775085449219, -316.8563232421875, -317.2701416015625, -321.78515625], "reading_logprob_std": 2.479799884863495, "tokenization_path_logprob_std": 145.44967272637138, "trajectory_balance_rmse": 5856.756863263301} {"canonical_length": 143, "canonical_nll": 1.398834228515625, "generation_rouge_l_f1": 0.22055137844611528, "generation_token_f1": 0.41102756892230574, "prompt_id": "8214cf7581d9c6a8", "reading_canonical_log_prob": -200.03329467773438, "reading_distances": [13, 10, 10, 7], "reading_log_probs": [-197.88571166992188, -198.54608154296875, -199.1154022216797, -195.7593231201172], "reading_logprob_std": 1.466946118231609, "tokenization_path_logprob_std": 76.23182700169897, "trajectory_balance_rmse": 4535.060592768215} {"canonical_length": 79, "canonical_nll": 0.8380579888066159, "generation_rouge_l_f1": 0.3925233644859813, "generation_token_f1": 0.43925233644859807, "prompt_id": "1669f97fd38169a6", "reading_canonical_log_prob": -66.20658111572266, "reading_distances": [15, 27, 30, 22], "reading_log_probs": [-68.23078918457031, -66.96378326416016, -70.11161804199219, -66.92135620117188], "reading_logprob_std": 1.498495645318827, "tokenization_path_logprob_std": 141.85021331859588, "trajectory_balance_rmse": 1756.9956384254174} {"canonical_length": 354, "canonical_nll": 1.0906446209061618, "generation_rouge_l_f1": 0.28462998102466797, "generation_token_f1": 0.4895635673624289, "prompt_id": "85c649ae8170e371", "reading_canonical_log_prob": -386.08819580078125, "reading_distances": [22, 24, 24, 26], "reading_log_probs": [-384.9849548339844, -387.24493408203125, -387.7601623535156, -386.2564697265625], "reading_logprob_std": 1.2223490557881946, "tokenization_path_logprob_std": 404.6592335707731, "trajectory_balance_rmse": 7202.963355775903} {"canonical_length": 364, "canonical_nll": 1.0522196004678914, "generation_rouge_l_f1": 0.30132450331125826, "generation_token_f1": 0.4635761589403974, "prompt_id": "ff47696a7202ba2a", "reading_canonical_log_prob": -383.0079345703125, "reading_distances": [34, 28, 33, 29], "reading_log_probs": [-386.96197509765625, -385.04742431640625, -384.6692199707031, -387.99200439453125], "reading_logprob_std": 1.5768554628703766, "tokenization_path_logprob_std": 200.59176248474145, "trajectory_balance_rmse": 8978.309825707393} {"canonical_length": 41, "canonical_nll": 1.025944314351896, "generation_rouge_l_f1": 0.42592592592592593, "generation_token_f1": 0.4444444444444444, "prompt_id": "3a5ef38cf12c45a8", "reading_canonical_log_prob": -42.063716888427734, "reading_distances": [24, 17, 17, 21], "reading_log_probs": [-42.86988067626953, -45.13115692138672, -42.50847625732422, -44.26270294189453], "reading_logprob_std": 1.2211304433251908, "tokenization_path_logprob_std": 29.518210585317274, "trajectory_balance_rmse": 1034.3518540596604} {"canonical_length": 230, "canonical_nll": 1.1521368938943615, "generation_rouge_l_f1": 0.3044397463002114, "generation_token_f1": 0.5539112050739958, "prompt_id": "13b43b11a6bf37ad", "reading_canonical_log_prob": -264.9914855957031, "reading_distances": [20, 19, 22, 18], "reading_log_probs": [-262.2490539550781, -261.9552001953125, -262.5655517578125, -263.76959228515625], "reading_logprob_std": 0.796493303043662, "tokenization_path_logprob_std": 409.5400966734248, "trajectory_balance_rmse": 6172.832118121964} {"canonical_length": 111, "canonical_nll": 1.4257567809508727, "generation_rouge_l_f1": 0.38961038961038963, "generation_token_f1": 0.5887445887445888, "prompt_id": "d6b3da36a5892656", "reading_canonical_log_prob": -158.25900268554688, "reading_distances": [32, 35, 30, 40], "reading_log_probs": [-154.2913818359375, -154.14666748046875, -154.78488159179688, -155.07740783691406], "reading_logprob_std": 0.43219281914247704, "tokenization_path_logprob_std": 251.9615617477637, "trajectory_balance_rmse": 3158.851659857623} {"canonical_length": 184, "canonical_nll": 1.3717344532842222, "generation_rouge_l_f1": 0.2648401826484018, "generation_token_f1": 0.4429223744292237, "prompt_id": "f1036fae25a980bc", "reading_canonical_log_prob": -252.39913940429688, "reading_distances": [17, 8, 10, 14], "reading_log_probs": [-252.1661376953125, -254.39959716796875, -250.5860595703125, -252.63519287109375], "reading_logprob_std": 1.5695046902066496, "tokenization_path_logprob_std": 442.63457812259156, "trajectory_balance_rmse": 6521.474393831754} {"canonical_length": 199, "canonical_nll": 1.0733541364046797, "generation_rouge_l_f1": 0.31277533039647576, "generation_token_f1": 0.5242290748898678, "prompt_id": "272e476421ef865b", "reading_canonical_log_prob": -213.59747314453125, "reading_distances": [25, 28, 25, 30], "reading_log_probs": [-210.11883544921875, -211.78785705566406, -213.1602020263672, -208.11683654785156], "reading_logprob_std": 2.176365491243937, "tokenization_path_logprob_std": 206.2246194661163, "trajectory_balance_rmse": 4755.171638182177} {"canonical_length": 202, "canonical_nll": 1.1047804426438739, "generation_rouge_l_f1": 0.27292110874200426, "generation_token_f1": 0.4818763326226013, "prompt_id": "548a4dcfac376298", "reading_canonical_log_prob": -223.1656494140625, "reading_distances": [23, 23, 22, 25], "reading_log_probs": [-225.8123779296875, -225.468994140625, -223.1023712158203, -227.53826904296875], "reading_logprob_std": 1.825758773160728, "tokenization_path_logprob_std": 175.29621479255675, "trajectory_balance_rmse": 5637.026948119938} {"canonical_length": 415, "canonical_nll": 1.1682483351374247, "generation_rouge_l_f1": 0.20689655172413793, "generation_token_f1": 0.4545454545454546, "prompt_id": "8ea8da1e0ef16ea4", "reading_canonical_log_prob": -484.82305908203125, "reading_distances": [32, 28, 28, 26], "reading_log_probs": [-477.46026611328125, -470.33428955078125, -476.952880859375, -474.1943359375], "reading_logprob_std": 3.2662231091133753, "tokenization_path_logprob_std": 135.429581350368, "trajectory_balance_rmse": 5770.725642649257} {"canonical_length": 70, "canonical_nll": 1.0062554495675222, "generation_rouge_l_f1": 0.12103746397694526, "generation_token_f1": 0.19020172910662822, "prompt_id": "68d1893ed02225a9", "reading_canonical_log_prob": -70.43788146972656, "reading_distances": [12, 15, 21, 12], "reading_log_probs": [-67.65522766113281, -66.81246948242188, -65.68336486816406, -71.61847686767578], "reading_logprob_std": 2.5804387088090843, "tokenization_path_logprob_std": 99.44563685200526, "trajectory_balance_rmse": 2150.666813542099} {"canonical_length": 207, "canonical_nll": 1.2467276900286837, "generation_rouge_l_f1": 0.23427331887201736, "generation_token_f1": 0.49891540130151846, "prompt_id": "06716de5ff9ed39f", "reading_canonical_log_prob": -258.0726318359375, "reading_distances": [27, 29, 30, 30], "reading_log_probs": [-268.1848449707031, -258.9866943359375, -262.3989562988281, -269.4383850097656], "reading_logprob_std": 4.916712754082165, "tokenization_path_logprob_std": 369.2703634957022, "trajectory_balance_rmse": 6658.057303739223} {"canonical_length": 24, "canonical_nll": 0.590016762415568, "generation_rouge_l_f1": 0.15503875968992248, "generation_token_f1": 0.17829457364341086, "prompt_id": "8d76cf27c40ca6eb", "reading_canonical_log_prob": -14.160402297973633, "reading_distances": [33, 27, 25, 27], "reading_log_probs": [-15.409224510192871, -15.778456687927246, -13.528848648071289, -17.217023849487305], "reading_logprob_std": 1.5185830117448562, "tokenization_path_logprob_std": 104.35273818136999, "trajectory_balance_rmse": 606.0896698562325} {"canonical_length": 37, "canonical_nll": 1.7437709086650126, "generation_rouge_l_f1": 0.16, "generation_token_f1": 0.17333333333333334, "prompt_id": "f7669b85c591c347", "reading_canonical_log_prob": -64.51952362060547, "reading_distances": [13, 25, 26, 18], "reading_log_probs": [-67.68653869628906, -65.82244873046875, -65.556884765625, -65.38871002197266], "reading_logprob_std": 1.063688947039558, "tokenization_path_logprob_std": 125.23651999793532, "trajectory_balance_rmse": 1207.7248798626024} {"canonical_length": 292, "canonical_nll": 1.2404012810694027, "generation_rouge_l_f1": 0.2644927536231884, "generation_token_f1": 0.5217391304347826, "prompt_id": "9875c8d91c22a365", "reading_canonical_log_prob": -362.1971740722656, "reading_distances": [29, 21, 25, 22], "reading_log_probs": [-358.2465515136719, -362.8800048828125, -356.7835693359375, -357.92724609375], "reading_logprob_std": 2.6881763560351644, "tokenization_path_logprob_std": 342.22577711494984, "trajectory_balance_rmse": 8922.549947612017} {"canonical_length": 293, "canonical_nll": 1.4069942956084684, "generation_rouge_l_f1": 0.24778761061946902, "generation_token_f1": 0.46017699115044247, "prompt_id": "967c6a5c34b145df", "reading_canonical_log_prob": -412.24932861328125, "reading_distances": [37, 32, 28, 37], "reading_log_probs": [-407.0735168457031, -409.8276062011719, -416.44720458984375, -410.7802734375], "reading_logprob_std": 3.9373409058787425, "tokenization_path_logprob_std": 261.1534732439555, "trajectory_balance_rmse": 8838.515582833987} {"canonical_length": 9, "canonical_nll": 0.0035435499416457284, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "abcadc4b6374d8f6", "reading_canonical_log_prob": -0.031891949474811554, "reading_distances": [43, 51, 41, 50], "reading_log_probs": [-0.15598516166210175, -0.18436507880687714, -0.45503419637680054, -0.6538196206092834], "reading_logprob_std": 0.23650931558264376, "tokenization_path_logprob_std": 60.329628876810865, "trajectory_balance_rmse": 245.819596157829} {"canonical_length": 261, "canonical_nll": 1.344486865047294, "generation_rouge_l_f1": 0.13761467889908258, "generation_token_f1": 0.24311926605504589, "prompt_id": "ba38639bb894f700", "reading_canonical_log_prob": -350.91107177734375, "reading_distances": [21, 30, 29, 25], "reading_log_probs": [-351.19439697265625, -349.13653564453125, -347.001953125, -343.79949951171875], "reading_logprob_std": 3.1595357449074664, "tokenization_path_logprob_std": 382.6951983137156, "trajectory_balance_rmse": 5849.517276371519} {"canonical_length": 258, "canonical_nll": 1.2283583057019138, "generation_rouge_l_f1": 0.27176220806794055, "generation_token_f1": 0.4458598726114649, "prompt_id": "49e9fe2e1aacf453", "reading_canonical_log_prob": -316.91644287109375, "reading_distances": [15, 21, 17, 17], "reading_log_probs": [-317.15643310546875, -314.69415283203125, -315.3802490234375, -315.931396484375], "reading_logprob_std": 1.0417786122807136, "tokenization_path_logprob_std": 152.3216929541032, "trajectory_balance_rmse": 5400.534928404373} {"canonical_length": 255, "canonical_nll": 1.2114442114736519, "generation_rouge_l_f1": 0.2464065708418891, "generation_token_f1": 0.5051334702258727, "prompt_id": "6193ec67d5074a6a", "reading_canonical_log_prob": -308.91827392578125, "reading_distances": [21, 21, 17, 15], "reading_log_probs": [-308.48236083984375, -308.80810546875, -305.97576904296875, -306.74285888671875], "reading_logprob_std": 1.3629251308041914, "tokenization_path_logprob_std": 142.55736666909658, "trajectory_balance_rmse": 5909.121955288141} {"canonical_length": 289, "canonical_nll": 0.9864523072556228, "generation_rouge_l_f1": 0.26475849731663686, "generation_token_f1": 0.47584973166368516, "prompt_id": "821b32f3ecbd67a9", "reading_canonical_log_prob": -285.084716796875, "reading_distances": [17, 18, 17, 17], "reading_log_probs": [-284.1905822753906, -281.8750305175781, -282.58416748046875, -281.72802734375], "reading_logprob_std": 1.127829269243728, "tokenization_path_logprob_std": 316.87147199233, "trajectory_balance_rmse": 7333.279641586994} {"canonical_length": 22, "canonical_nll": 0.6009679274125532, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "ebadf5b1dd406fa9", "reading_canonical_log_prob": -13.221294403076172, "reading_distances": [30, 36, 37, 43], "reading_log_probs": [-13.495102882385254, -14.951982498168945, -14.181779861450195, -15.461559295654297], "reading_logprob_std": 0.8636979761518901, "tokenization_path_logprob_std": 78.60957092232023, "trajectory_balance_rmse": 650.7335469770229} {"canonical_length": 8, "canonical_nll": 0.5290312767028809, "generation_rouge_l_f1": 0.05818181818181818, "generation_token_f1": 0.05818181818181818, "prompt_id": "6edf6d71173f0e03", "reading_canonical_log_prob": -4.232250213623047, "reading_distances": [30, 32, 33, 27], "reading_log_probs": [-20.7014102935791, -22.488609313964844, -24.57305908203125, -21.66161346435547], "reading_logprob_std": 1.648512040303479, "tokenization_path_logprob_std": 30.85328100824256, "trajectory_balance_rmse": 198.1234296083126} {"canonical_length": 16, "canonical_nll": 1.4330600500106812, "generation_rouge_l_f1": 0.3, "generation_token_f1": 0.3, "prompt_id": "938d726d90109110", "reading_canonical_log_prob": -22.9289608001709, "reading_distances": [25, 24, 32, 23], "reading_log_probs": [-28.61180877685547, -30.16315269470215, -25.801467895507812, -30.01614761352539], "reading_logprob_std": 2.0225042500135957, "tokenization_path_logprob_std": 80.63918235288206, "trajectory_balance_rmse": 618.9169434664503} {"canonical_length": 6, "canonical_nll": 0.2561881939570109, "generation_rouge_l_f1": 0.8000000000000002, "generation_token_f1": 1.0, "prompt_id": "b4288f8c25b7da9a", "reading_canonical_log_prob": -1.5371291637420654, "reading_distances": [42, 33, 33, 37], "reading_log_probs": [-0.15423409640789032, -4.0070905685424805, -0.20809215307235718, -0.14623527228832245], "reading_logprob_std": 1.9189816422472405, "tokenization_path_logprob_std": 18.18552587218873, "trajectory_balance_rmse": 210.41053704790863} {"canonical_length": 266, "canonical_nll": 1.4422567267166941, "generation_rouge_l_f1": 0.24856596558317398, "generation_token_f1": 0.5353728489483748, "prompt_id": "3c7a013fe65f77bc", "reading_canonical_log_prob": -383.6402893066406, "reading_distances": [27, 25, 25, 26], "reading_log_probs": [-383.2251892089844, -378.7313537597656, -382.6119384765625, -384.9737243652344], "reading_logprob_std": 2.633631361863638, "tokenization_path_logprob_std": 83.10022672133287, "trajectory_balance_rmse": 7079.084620205322} {"canonical_length": 28, "canonical_nll": 2.0331663404192244, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "66c2a007594b76e9", "reading_canonical_log_prob": -56.92865753173828, "reading_distances": [61, 62, 51, 74], "reading_log_probs": [-64.21844482421875, -62.7441291809082, -58.416114807128906, -63.189537048339844], "reading_logprob_std": 2.559540126915601, "tokenization_path_logprob_std": 117.54200717929739, "trajectory_balance_rmse": 784.3581462878906} {"canonical_length": 321, "canonical_nll": 1.5837547801365361, "generation_rouge_l_f1": 0.17777777777777778, "generation_token_f1": 0.38974358974358975, "prompt_id": "dd27e1dd582c59a2", "reading_canonical_log_prob": -508.3852844238281, "reading_distances": [22, 25, 24, 24], "reading_log_probs": [-518.738525390625, -518.395263671875, -517.8321533203125, -515.88916015625], "reading_logprob_std": 1.2725005471814508, "tokenization_path_logprob_std": 301.4097166282274, "trajectory_balance_rmse": 8989.441525660226} {"canonical_length": 21, "canonical_nll": 0.19646694546654112, "generation_rouge_l_f1": 0.8400000000000001, "generation_token_f1": 0.8400000000000001, "prompt_id": "3c6903de8714c50f", "reading_canonical_log_prob": -4.125805854797363, "reading_distances": [18, 15, 16, 12], "reading_log_probs": [-3.5870752334594727, -2.919861316680908, -2.8039278984069824, -3.4038989543914795], "reading_logprob_std": 0.3763582035517345, "tokenization_path_logprob_std": 90.52627079095672, "trajectory_balance_rmse": 759.6277651690093} {"canonical_length": 45, "canonical_nll": 1.634641350640191, "generation_rouge_l_f1": 0.04166666666666667, "generation_token_f1": 0.04166666666666667, "prompt_id": "0692a96622cd2aed", "reading_canonical_log_prob": -73.5588607788086, "reading_distances": [21, 27, 23, 25], "reading_log_probs": [-70.4325180053711, -70.91218566894531, -69.72759246826172, -69.51921081542969], "reading_logprob_std": 0.6421444548292883, "tokenization_path_logprob_std": 89.93238787315589, "trajectory_balance_rmse": 1717.4434469980167} {"canonical_length": 322, "canonical_nll": 1.1676882157414596, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.4305555555555555, "prompt_id": "e5a0acf44a48d7ab", "reading_canonical_log_prob": -375.99560546875, "reading_distances": [25, 33, 23, 26], "reading_log_probs": [-381.7783203125, -380.3965759277344, -376.23992919921875, -376.05560302734375], "reading_logprob_std": 2.908152630674293, "tokenization_path_logprob_std": 271.9936994906593, "trajectory_balance_rmse": 6832.208117352132} {"canonical_length": 191, "canonical_nll": 0.9070000773325017, "generation_rouge_l_f1": 0.445, "generation_token_f1": 0.61, "prompt_id": "4e165cac00833611", "reading_canonical_log_prob": -173.2370147705078, "reading_distances": [18, 19, 18, 15], "reading_log_probs": [-172.3025665283203, -168.26016235351562, -166.48294067382812, -167.48995971679688], "reading_logprob_std": 2.5517327001199197, "tokenization_path_logprob_std": 33.436086213224016, "trajectory_balance_rmse": 3686.644367752883} {"canonical_length": 296, "canonical_nll": 1.3265867491026182, "generation_rouge_l_f1": 0.24604569420035147, "generation_token_f1": 0.49560632688927947, "prompt_id": "eadf46baa67f8e23", "reading_canonical_log_prob": -392.669677734375, "reading_distances": [25, 30, 24, 31], "reading_log_probs": [-399.840087890625, -402.15625, -396.53973388671875, -401.4811706542969], "reading_logprob_std": 2.5061404391171767, "tokenization_path_logprob_std": 430.28373520395456, "trajectory_balance_rmse": 9257.410476031817} {"canonical_length": 97, "canonical_nll": 1.3990366827581346, "generation_rouge_l_f1": 0.2413793103448276, "generation_token_f1": 0.33793103448275863, "prompt_id": "228080ce5449beea", "reading_canonical_log_prob": -135.70655822753906, "reading_distances": [17, 17, 23, 21], "reading_log_probs": [-130.64389038085938, -130.37942504882812, -130.04998779296875, -131.6073455810547], "reading_logprob_std": 0.6703599243227859, "tokenization_path_logprob_std": 117.04280391555001, "trajectory_balance_rmse": 1504.0902193653153} {"canonical_length": 345, "canonical_nll": 1.14148975317029, "generation_rouge_l_f1": 0.22666666666666666, "generation_token_f1": 0.44666666666666666, "prompt_id": "0584d61b9b2b3717", "reading_canonical_log_prob": -393.81396484375, "reading_distances": [39, 24, 40, 33], "reading_log_probs": [-398.1020812988281, -398.0255432128906, -399.1177062988281, -397.1690673828125], "reading_logprob_std": 0.7974663976988766, "tokenization_path_logprob_std": 271.673575602676, "trajectory_balance_rmse": 8112.987014725613} {"canonical_length": 16, "canonical_nll": 1.1578991413116455, "generation_rouge_l_f1": 0.64, "generation_token_f1": 0.64, "prompt_id": "f605a0e57df216ae", "reading_canonical_log_prob": -18.526386260986328, "reading_distances": [34, 36, 38, 30], "reading_log_probs": [-17.055191040039062, -17.689796447753906, -19.132461547851562, -16.192615509033203], "reading_logprob_std": 1.2391980410425039, "tokenization_path_logprob_std": 58.51004145866943, "trajectory_balance_rmse": 349.719204753025} {"canonical_length": 337, "canonical_nll": 1.195629991480434, "generation_rouge_l_f1": 0.22483221476510068, "generation_token_f1": 0.4228187919463087, "prompt_id": "ab315c4f854a9130", "reading_canonical_log_prob": -402.92730712890625, "reading_distances": [27, 25, 31, 21], "reading_log_probs": [-407.596435546875, -407.79718017578125, -407.1407165527344, -407.6544189453125], "reading_logprob_std": 0.28380681334299745, "tokenization_path_logprob_std": 180.11817371564734, "trajectory_balance_rmse": 10473.589689616816} {"canonical_length": 283, "canonical_nll": 1.3552322657285225, "generation_rouge_l_f1": 0.22627737226277372, "generation_token_f1": 0.4343065693430657, "prompt_id": "93f170181468154b", "reading_canonical_log_prob": -383.5307312011719, "reading_distances": [27, 25, 27, 26], "reading_log_probs": [-386.78155517578125, -385.8048095703125, -386.3958740234375, -385.1496276855469], "reading_logprob_std": 0.712841712621461, "tokenization_path_logprob_std": 66.11776706520122, "trajectory_balance_rmse": 7574.508321081849} {"canonical_length": 32, "canonical_nll": 0.9744933843612671, "generation_rouge_l_f1": 0.2835820895522388, "generation_token_f1": 0.34328358208955223, "prompt_id": "ce3af5b67c725855", "reading_canonical_log_prob": -31.183788299560547, "reading_distances": [15, 15, 15, 17], "reading_log_probs": [-33.940696716308594, -31.895187377929688, -35.3486328125, -37.604591369628906], "reading_logprob_std": 2.401448258455229, "tokenization_path_logprob_std": 114.10357569403254, "trajectory_balance_rmse": 1082.9942140795376} {"canonical_length": 64, "canonical_nll": 1.6122865676879883, "generation_rouge_l_f1": 0.1951219512195122, "generation_token_f1": 0.3252032520325203, "prompt_id": "5d5d479145243991", "reading_canonical_log_prob": -103.18634033203125, "reading_distances": [23, 19, 19, 21], "reading_log_probs": [-115.92854309082031, -109.74317932128906, -113.43143463134766, -113.3391342163086], "reading_logprob_std": 2.5452889840023816, "tokenization_path_logprob_std": 143.95646025924017, "trajectory_balance_rmse": 1704.6953787888738} {"canonical_length": 259, "canonical_nll": 1.8387203732052364, "generation_rouge_l_f1": 0.18285714285714283, "generation_token_f1": 0.4228571428571429, "prompt_id": "a2496cc91dc1f05e", "reading_canonical_log_prob": -476.22857666015625, "reading_distances": [14, 15, 15, 16], "reading_log_probs": [-473.26434326171875, -474.59326171875, -473.2658996582031, -474.2032470703125], "reading_logprob_std": 0.6733120058315831, "tokenization_path_logprob_std": 167.08157901778372, "trajectory_balance_rmse": 8467.047976401705} {"canonical_length": 201, "canonical_nll": 0.9784497313238495, "generation_rouge_l_f1": 0.2557077625570776, "generation_token_f1": 0.5022831050228311, "prompt_id": "419c50c5b36f610d", "reading_canonical_log_prob": -196.66839599609375, "reading_distances": [36, 29, 46, 37], "reading_log_probs": [-198.62648010253906, -196.3048095703125, -195.90676879882812, -200.33242797851562], "reading_logprob_std": 2.0749266546054095, "tokenization_path_logprob_std": 314.6777892223147, "trajectory_balance_rmse": 6548.0185030194225} {"canonical_length": 120, "canonical_nll": 0.7771199544270834, "generation_rouge_l_f1": 0.18230563002680963, "generation_token_f1": 0.2680965147453083, "prompt_id": "07d185247528da20", "reading_canonical_log_prob": -93.25439453125, "reading_distances": [20, 20, 15, 20], "reading_log_probs": [-115.3902816772461, -106.77815246582031, -112.42930603027344, -118.99315643310547], "reading_logprob_std": 5.165130229749365, "tokenization_path_logprob_std": 158.91688173974975, "trajectory_balance_rmse": 2495.6692336441406} {"canonical_length": 187, "canonical_nll": 1.4502922506893383, "generation_rouge_l_f1": 0.2471395881006865, "generation_token_f1": 0.576659038901602, "prompt_id": "f75ec5897fcc8e0d", "reading_canonical_log_prob": -271.20465087890625, "reading_distances": [29, 29, 22, 32], "reading_log_probs": [-267.1384582519531, -268.6759033203125, -267.26446533203125, -267.54791259765625], "reading_logprob_std": 0.7007216875728809, "tokenization_path_logprob_std": 216.08374480928143, "trajectory_balance_rmse": 5318.467245130037} {"canonical_length": 362, "canonical_nll": 1.6070330709383633, "generation_rouge_l_f1": 0.34210526315789475, "generation_token_f1": 0.5361842105263158, "prompt_id": "cd40c2a793f29f44", "reading_canonical_log_prob": -581.7459716796875, "reading_distances": [25, 21, 17, 17], "reading_log_probs": [-583.489501953125, -586.0699462890625, -587.072265625, -584.0948486328125], "reading_logprob_std": 1.6741166787859882, "tokenization_path_logprob_std": 477.7396123069714, "trajectory_balance_rmse": 10600.694660905781} {"canonical_length": 131, "canonical_nll": 1.0907102395559996, "generation_rouge_l_f1": 0.192, "generation_token_f1": 0.3466666666666666, "prompt_id": "79a9080b044df25c", "reading_canonical_log_prob": -142.88304138183594, "reading_distances": [32, 28, 28, 29], "reading_log_probs": [-143.17623901367188, -141.28091430664062, -142.3983154296875, -145.47735595703125], "reading_logprob_std": 1.775569354661641, "tokenization_path_logprob_std": 145.02451629480944, "trajectory_balance_rmse": 3404.147101518292} {"canonical_length": 59, "canonical_nll": 2.0433266849841103, "generation_rouge_l_f1": 0.15047021943573669, "generation_token_f1": 0.225705329153605, "prompt_id": "56223c1e8305751f", "reading_canonical_log_prob": -120.5562744140625, "reading_distances": [20, 19, 18, 21], "reading_log_probs": [-129.59799194335938, -123.80030822753906, -123.4066162109375, -125.33009338378906], "reading_logprob_std": 2.833670007049053, "tokenization_path_logprob_std": 137.92128357040733, "trajectory_balance_rmse": 1850.5341124421934} {"canonical_length": 24, "canonical_nll": 1.0027938683827717, "generation_rouge_l_f1": 0.15573770491803282, "generation_token_f1": 0.1885245901639344, "prompt_id": "5528bdba3c5f6117", "reading_canonical_log_prob": -24.067052841186523, "reading_distances": [47, 41, 48, 44], "reading_log_probs": [-25.003604888916016, -25.850440979003906, -24.811065673828125, -23.82163429260254], "reading_logprob_std": 0.8329960612812922, "tokenization_path_logprob_std": 92.35453580648526, "trajectory_balance_rmse": 854.3108252667722} {"canonical_length": 351, "canonical_nll": 1.4706325259303774, "generation_rouge_l_f1": 0.23300970873786406, "generation_token_f1": 0.48220064724919093, "prompt_id": "daf735e510f876d3", "reading_canonical_log_prob": -516.1920166015625, "reading_distances": [19, 21, 16, 18], "reading_log_probs": [-515.8193359375, -518.828857421875, -517.7061767578125, -513.8370361328125], "reading_logprob_std": 2.1927121940874796, "tokenization_path_logprob_std": 762.1852269917538, "trajectory_balance_rmse": 10585.7344932142} {"canonical_length": 73, "canonical_nll": 1.59321071677012, "generation_rouge_l_f1": 0.14534883720930233, "generation_token_f1": 0.25, "prompt_id": "d276e797ca2de2bf", "reading_canonical_log_prob": -116.30438232421875, "reading_distances": [53, 53, 48, 42], "reading_log_probs": [-113.34908294677734, -111.04116821289062, -111.586181640625, -112.9293441772461], "reading_logprob_std": 1.0907491988436029, "tokenization_path_logprob_std": 192.97611245974744, "trajectory_balance_rmse": 1632.784993330369} {"canonical_length": 107, "canonical_nll": 1.8293502665011683, "generation_rouge_l_f1": 0.2393162393162393, "generation_token_f1": 0.358974358974359, "prompt_id": "60ad70d4930701a2", "reading_canonical_log_prob": -195.740478515625, "reading_distances": [12, 13, 10, 12], "reading_log_probs": [-193.74911499023438, -200.1044921875, -199.4443359375, -198.82635498046875], "reading_logprob_std": 2.9019543462667414, "tokenization_path_logprob_std": 185.52020591092577, "trajectory_balance_rmse": 2955.8932907387834} {"canonical_length": 11, "canonical_nll": 0.015543126247145912, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "91d144e62a7c6f3d", "reading_canonical_log_prob": -0.17097438871860504, "reading_distances": [35, 35, 25, 26], "reading_log_probs": [-0.43351903557777405, -0.24071937799453735, -0.14063365757465363, -0.11365607380867004], "reading_logprob_std": 0.14495858364040726, "tokenization_path_logprob_std": 64.2124517719905, "trajectory_balance_rmse": 412.98215281575847} {"canonical_length": 31, "canonical_nll": 0.9979871934460055, "generation_rouge_l_f1": 0.14746543778801843, "generation_token_f1": 0.1751152073732719, "prompt_id": "663423620e0283d5", "reading_canonical_log_prob": -30.937602996826172, "reading_distances": [15, 15, 18, 18], "reading_log_probs": [-33.33076095581055, -32.841270446777344, -32.45473098754883, -34.16018295288086], "reading_logprob_std": 0.7355541942022431, "tokenization_path_logprob_std": 47.62017609440368, "trajectory_balance_rmse": 582.3164460668369} {"canonical_length": 51, "canonical_nll": 0.053121178757910635, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "eece35ce9b12a6f2", "reading_canonical_log_prob": -2.7091801166534424, "reading_distances": [24, 22, 23, 23], "reading_log_probs": [-3.1379234790802, -5.650351047515869, -3.856102466583252, -3.469287872314453], "reading_logprob_std": 1.120411597753541, "tokenization_path_logprob_std": 116.38435092368965, "trajectory_balance_rmse": 1469.8086903572396} {"canonical_length": 156, "canonical_nll": 1.558509533221905, "generation_rouge_l_f1": 0.18840579710144928, "generation_token_f1": 0.33816425120772947, "prompt_id": "25055edbaab1f2e6", "reading_canonical_log_prob": -243.1274871826172, "reading_distances": [9, 5, 8, 8], "reading_log_probs": [-246.85107421875, -242.74179077148438, -247.15420532226562, -244.9303741455078], "reading_logprob_std": 2.0386251932564825, "tokenization_path_logprob_std": 223.97216957863023, "trajectory_balance_rmse": 4167.116225791973} {"canonical_length": 297, "canonical_nll": 0.9917935458096591, "generation_rouge_l_f1": 0.2950257289879932, "generation_token_f1": 0.5454545454545455, "prompt_id": "51338c6674b54120", "reading_canonical_log_prob": -294.56268310546875, "reading_distances": [28, 22, 22, 29], "reading_log_probs": [-292.6104736328125, -289.8512878417969, -291.7911376953125, -293.941162109375], "reading_logprob_std": 1.7119151894229663, "tokenization_path_logprob_std": 231.91606465467532, "trajectory_balance_rmse": 6836.283301940973} {"canonical_length": 130, "canonical_nll": 2.25653076171875, "generation_rouge_l_f1": 0.1377410468319559, "generation_token_f1": 0.2920110192837465, "prompt_id": "156614ecc1fc9430", "reading_canonical_log_prob": -293.3489990234375, "reading_distances": [67, 59, 68, 64], "reading_log_probs": [-298.6119689941406, -296.54547119140625, -301.4550476074219, -302.394775390625], "reading_logprob_std": 2.6749536118953716, "tokenization_path_logprob_std": 85.64914100713199, "trajectory_balance_rmse": 4359.456369089373} {"canonical_length": 234, "canonical_nll": 1.6724571814903846, "generation_rouge_l_f1": 0.255863539445629, "generation_token_f1": 0.4946695095948827, "prompt_id": "39459679cbe7ca40", "reading_canonical_log_prob": -391.35498046875, "reading_distances": [25, 26, 27, 29], "reading_log_probs": [-393.426025390625, -394.27935791015625, -392.24481201171875, -393.074462890625], "reading_logprob_std": 0.8429366933827682, "tokenization_path_logprob_std": 38.91500701197136, "trajectory_balance_rmse": 5198.774229442756} {"canonical_length": 81, "canonical_nll": 1.2766975120261863, "generation_rouge_l_f1": 0.3044982698961938, "generation_token_f1": 0.43598615916955014, "prompt_id": "0fc229d90ecbfd68", "reading_canonical_log_prob": -103.4124984741211, "reading_distances": [24, 26, 32, 30], "reading_log_probs": [-105.60910034179688, -107.06410217285156, -106.13249969482422, -105.75637817382812], "reading_logprob_std": 0.6539725149789499, "tokenization_path_logprob_std": 110.42244731832939, "trajectory_balance_rmse": 2293.31011500321} {"canonical_length": 7, "canonical_nll": 3.7816260201590404, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "fc7a567baaf9e2df", "reading_canonical_log_prob": -26.47138214111328, "reading_distances": [19, 24, 24, 21], "reading_log_probs": [-30.52596664428711, -29.490692138671875, -28.303600311279297, -31.968246459960938], "reading_logprob_std": 1.5563839046384391, "tokenization_path_logprob_std": 53.961162216184384, "trajectory_balance_rmse": 148.065051878989} {"canonical_length": 3, "canonical_nll": 1.3390828768412273, "generation_rouge_l_f1": 0.125, "generation_token_f1": 0.125, "prompt_id": "14871a1dde668d87", "reading_canonical_log_prob": -4.017248630523682, "reading_distances": [30, 26, 20, 26], "reading_log_probs": [-4.235827922821045, -4.022003650665283, -4.091213226318359, -4.099737644195557], "reading_logprob_std": 0.0894706649114002, "tokenization_path_logprob_std": 30.317325913973054, "trajectory_balance_rmse": 47.63830765056466} {"canonical_length": 3, "canonical_nll": 2.5755236943562827, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "408cd1b200daeec5", "reading_canonical_log_prob": -7.726571083068848, "reading_distances": [31, 29, 23, 30], "reading_log_probs": [-4.035592555999756, -6.293456077575684, -6.656405448913574, -2.551363945007324], "reading_logprob_std": 1.939843077738346, "tokenization_path_logprob_std": 24.81337682845002, "trajectory_balance_rmse": 74.34758868887134} {"canonical_length": 129, "canonical_nll": 1.6526753033778465, "generation_rouge_l_f1": 0.16560509554140126, "generation_token_f1": 0.19108280254777069, "prompt_id": "6db695863766c033", "reading_canonical_log_prob": -213.1951141357422, "reading_distances": [9, 13, 11, 12], "reading_log_probs": [-217.0444793701172, -216.55416870117188, -216.5780029296875, -216.61688232421875], "reading_logprob_std": 0.23217391676326377, "tokenization_path_logprob_std": 188.78880028175325, "trajectory_balance_rmse": 3299.2232543213045} {"canonical_length": 6, "canonical_nll": 0.526294986406962, "generation_rouge_l_f1": 0.888888888888889, "generation_token_f1": 0.888888888888889, "prompt_id": "66a509bfd96df65b", "reading_canonical_log_prob": -3.1577699184417725, "reading_distances": [16, 17, 13, 19], "reading_log_probs": [-3.733999729156494, -4.158585548400879, -3.948112964630127, -3.839993953704834], "reading_logprob_std": 0.1813930824663171, "tokenization_path_logprob_std": 42.69372300772576, "trajectory_balance_rmse": 131.21672915988387} {"canonical_length": 3, "canonical_nll": 0.31315626700719196, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "2c889c2667571076", "reading_canonical_log_prob": -0.9394688010215759, "reading_distances": [13, 18, 20, 19], "reading_log_probs": [-1.9085921049118042, -2.985124111175537, -4.826411247253418, -2.3758251667022705], "reading_logprob_std": 1.2799040361608047, "tokenization_path_logprob_std": 63.41183673965995, "trajectory_balance_rmse": 113.50179998913805} {"canonical_length": 15, "canonical_nll": 0.4852407455444336, "generation_rouge_l_f1": 0.2608695652173913, "generation_token_f1": 0.2608695652173913, "prompt_id": "79afdb8eb4cf0c62", "reading_canonical_log_prob": -7.278611183166504, "reading_distances": [23, 24, 26, 24], "reading_log_probs": [-8.195491790771484, -8.298343658447266, -6.177175521850586, -8.683954238891602], "reading_logprob_std": 1.1274885051510177, "tokenization_path_logprob_std": 94.90587987883733, "trajectory_balance_rmse": 371.8422806473523} {"canonical_length": 145, "canonical_nll": 1.4821442702720906, "generation_rouge_l_f1": 0.33163265306122447, "generation_token_f1": 0.48469387755102045, "prompt_id": "51fd58caa03ae57f", "reading_canonical_log_prob": -214.91091918945312, "reading_distances": [29, 30, 29, 22], "reading_log_probs": [-224.75770568847656, -227.29531860351562, -226.97364807128906, -223.57571411132812], "reading_logprob_std": 1.7849339097201846, "tokenization_path_logprob_std": 196.26222622983724, "trajectory_balance_rmse": 4050.9577045322662} {"canonical_length": 19, "canonical_nll": 1.1002928081311678, "generation_rouge_l_f1": 0.2962962962962963, "generation_token_f1": 0.345679012345679, "prompt_id": "8671c943f4179a67", "reading_canonical_log_prob": -20.905563354492188, "reading_distances": [16, 13, 15, 16], "reading_log_probs": [-21.143259048461914, -22.479684829711914, -21.519046783447266, -21.644060134887695], "reading_logprob_std": 0.5638243257204865, "tokenization_path_logprob_std": 97.13938287280665, "trajectory_balance_rmse": 619.5014271061997} {"canonical_length": 289, "canonical_nll": 1.2700250223021194, "generation_rouge_l_f1": 0.2314647377938517, "generation_token_f1": 0.49547920433996384, "prompt_id": "9cb312e17f963300", "reading_canonical_log_prob": -367.0372314453125, "reading_distances": [21, 13, 21, 18], "reading_log_probs": [-379.31658935546875, -378.45245361328125, -380.0731201171875, -376.9024963378906], "reading_logprob_std": 1.3610258720947406, "tokenization_path_logprob_std": 236.48622485385533, "trajectory_balance_rmse": 8805.59578676468} {"canonical_length": 28, "canonical_nll": 2.000167574201311, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "9227ddd5a86a7e6a", "reading_canonical_log_prob": -56.00469207763672, "reading_distances": [25, 25, 27, 24], "reading_log_probs": [-59.64359664916992, -59.80759811401367, -59.16737747192383, -59.84409713745117], "reading_logprob_std": 0.3113205456907351, "tokenization_path_logprob_std": 9.622435891227639, "trajectory_balance_rmse": 870.2567765606705} {"canonical_length": 223, "canonical_nll": 1.6807561626348795, "generation_rouge_l_f1": 0.29831932773109243, "generation_token_f1": 0.5672268907563025, "prompt_id": "3216cdb4c067a3d8", "reading_canonical_log_prob": -374.8086242675781, "reading_distances": [46, 47, 49, 44], "reading_log_probs": [-382.857666015625, -383.01611328125, -386.40380859375, -382.135498046875], "reading_logprob_std": 1.9059598837457146, "tokenization_path_logprob_std": 255.39261837489911, "trajectory_balance_rmse": 7040.364283819195} {"canonical_length": 108, "canonical_nll": 0.999661057083695, "generation_rouge_l_f1": 0.34599156118143454, "generation_token_f1": 0.4725738396624472, "prompt_id": "1ef389c9aee61602", "reading_canonical_log_prob": -107.96339416503906, "reading_distances": [14, 9, 14, 15], "reading_log_probs": [-111.48809051513672, -110.37100982666016, -111.56272888183594, -111.21331024169922], "reading_logprob_std": 0.5462525385707766, "tokenization_path_logprob_std": 91.62426644359478, "trajectory_balance_rmse": 2117.253760775031} {"canonical_length": 6, "canonical_nll": 3.970017115275065, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "a730d1195f266d12", "reading_canonical_log_prob": -23.82010269165039, "reading_distances": [32, 30, 25, 29], "reading_log_probs": [-28.991390228271484, -24.983165740966797, -22.069704055786133, -29.59998321533203], "reading_logprob_std": 3.5455882826198524, "tokenization_path_logprob_std": 53.19269351882341, "trajectory_balance_rmse": 164.67191623794605} {"canonical_length": 26, "canonical_nll": 1.6013411008394682, "generation_rouge_l_f1": 0.22641509433962262, "generation_token_f1": 0.33962264150943394, "prompt_id": "7f409abdb6fbebf8", "reading_canonical_log_prob": -41.63486862182617, "reading_distances": [15, 11, 11, 11], "reading_log_probs": [-50.05390930175781, -47.016990661621094, -47.33257293701172, -52.46959686279297], "reading_logprob_std": 2.5606576987349534, "tokenization_path_logprob_std": 93.92074645509149, "trajectory_balance_rmse": 633.2830695131169} {"canonical_length": 168, "canonical_nll": 1.504530770438058, "generation_rouge_l_f1": 0.27791563275434245, "generation_token_f1": 0.531017369727047, "prompt_id": "1c78ad010cb27d62", "reading_canonical_log_prob": -252.76116943359375, "reading_distances": [39, 33, 35, 41], "reading_log_probs": [-253.80712890625, -245.9904022216797, -248.9807891845703, -250.67401123046875], "reading_logprob_std": 3.2654352182159734, "tokenization_path_logprob_std": 66.27807626420433, "trajectory_balance_rmse": 4897.241488236464} {"canonical_length": 15, "canonical_nll": 0.6527185440063477, "generation_rouge_l_f1": 0.40909090909090906, "generation_token_f1": 0.40909090909090906, "prompt_id": "779d0398df55911b", "reading_canonical_log_prob": -9.790778160095215, "reading_distances": [6, 5, 8, 7], "reading_log_probs": [-19.61983299255371, -9.849069595336914, -9.072845458984375, -10.111064910888672], "reading_logprob_std": 4.990595312803291, "tokenization_path_logprob_std": 73.20252795833866, "trajectory_balance_rmse": 308.9994978912819} {"canonical_length": 11, "canonical_nll": 0.5713568167252974, "generation_rouge_l_f1": 0.23880597014925373, "generation_token_f1": 0.23880597014925373, "prompt_id": "3e344435d6b1501d", "reading_canonical_log_prob": -6.2849249839782715, "reading_distances": [41, 47, 43, 38], "reading_log_probs": [-5.806331157684326, -6.577402591705322, -4.879091262817383, -4.179652690887451], "reading_logprob_std": 1.0497256818625402, "tokenization_path_logprob_std": 44.61334520227334, "trajectory_balance_rmse": 201.568265822137} {"canonical_length": 376, "canonical_nll": 0.551007534595246, "generation_rouge_l_f1": 0.39285714285714285, "generation_token_f1": 0.6666666666666667, "prompt_id": "e2124bce16d0a6b8", "reading_canonical_log_prob": -207.1788330078125, "reading_distances": [13, 17, 9, 12], "reading_log_probs": [-204.78094482421875, -204.64154052734375, -206.47576904296875, -205.681396484375], "reading_logprob_std": 0.8553458512729883, "tokenization_path_logprob_std": 229.64078320099821, "trajectory_balance_rmse": 6246.968979680632} {"canonical_length": 240, "canonical_nll": 2.239446258544922, "generation_rouge_l_f1": 0.29045643153526973, "generation_token_f1": 0.5228215767634854, "prompt_id": "6a9d3b5c30c1cdf4", "reading_canonical_log_prob": -537.4671020507812, "reading_distances": [11, 18, 14, 13], "reading_log_probs": [-537.7205810546875, -542.4285278320312, -536.7572021484375, -542.0823974609375], "reading_logprob_std": 2.9263138145767975, "tokenization_path_logprob_std": 239.61407873497205, "trajectory_balance_rmse": 5423.683459350112} {"canonical_length": 271, "canonical_nll": 1.1402449379108048, "generation_rouge_l_f1": 0.2723735408560311, "generation_token_f1": 0.556420233463035, "prompt_id": "63c15bf241bee263", "reading_canonical_log_prob": -309.0063781738281, "reading_distances": [33, 24, 32, 30], "reading_log_probs": [-306.2494201660156, -308.3268127441406, -306.82208251953125, -307.80462646484375], "reading_logprob_std": 0.9382809143280204, "tokenization_path_logprob_std": 99.24749068019545, "trajectory_balance_rmse": 8286.204177402276} {"canonical_length": 25, "canonical_nll": 1.4976075744628907, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "9932e0e2da85e27d", "reading_canonical_log_prob": -37.440189361572266, "reading_distances": [24, 18, 24, 22], "reading_log_probs": [-39.448265075683594, -38.33454132080078, -39.50776672363281, -39.99073791503906], "reading_logprob_std": 0.700648747357438, "tokenization_path_logprob_std": 50.62779400552438, "trajectory_balance_rmse": 679.9152613675699} {"canonical_length": 114, "canonical_nll": 0.9967810647529468, "generation_rouge_l_f1": 0.27586206896551724, "generation_token_f1": 0.5039787798408488, "prompt_id": "f5ce2237f02c3721", "reading_canonical_log_prob": -113.63304138183594, "reading_distances": [24, 22, 21, 23], "reading_log_probs": [-113.26689910888672, -114.40803527832031, -117.5634765625, -111.93428039550781], "reading_logprob_std": 2.4031709020229743, "tokenization_path_logprob_std": 210.99711128132404, "trajectory_balance_rmse": 3208.2136866312985} {"canonical_length": 297, "canonical_nll": 1.5796050511626682, "generation_rouge_l_f1": 0.2932330827067669, "generation_token_f1": 0.5601503759398495, "prompt_id": "3c5a7e9edbe46ab0", "reading_canonical_log_prob": -469.1427001953125, "reading_distances": [13, 21, 18, 16], "reading_log_probs": [-470.1946716308594, -471.1255187988281, -471.91607666015625, -473.549072265625], "reading_logprob_std": 1.421471746979492, "tokenization_path_logprob_std": 423.90653726483896, "trajectory_balance_rmse": 7796.220884866433} {"canonical_length": 129, "canonical_nll": 1.3037302179854044, "generation_rouge_l_f1": 0.2448979591836735, "generation_token_f1": 0.4336734693877551, "prompt_id": "6097f6fb796f153f", "reading_canonical_log_prob": -168.1811981201172, "reading_distances": [26, 25, 32, 21], "reading_log_probs": [-167.493896484375, -168.42755126953125, -166.964599609375, -165.68392944335938], "reading_logprob_std": 1.145121954510228, "tokenization_path_logprob_std": 288.1169051413054, "trajectory_balance_rmse": 3413.071356514855} {"canonical_length": 10, "canonical_nll": 0.5878170013427735, "generation_rouge_l_f1": 0.10294117647058823, "generation_token_f1": 0.10294117647058823, "prompt_id": "b2680f937aebb340", "reading_canonical_log_prob": -5.878170013427734, "reading_distances": [7, 10, 5, 6], "reading_log_probs": [-15.221783638000488, -20.1915283203125, -4.132822036743164, -10.175614356994629], "reading_logprob_std": 6.878972640483948, "tokenization_path_logprob_std": 31.172457466113602, "trajectory_balance_rmse": 76.2984009942617} {"canonical_length": 8, "canonical_nll": 0.9891908764839172, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "a6d7ae5badbbf976", "reading_canonical_log_prob": -7.913527011871338, "reading_distances": [29, 35, 25, 32], "reading_log_probs": [-8.734855651855469, -9.333364486694336, -9.537989616394043, -10.096299171447754], "reading_logprob_std": 0.5621695906325886, "tokenization_path_logprob_std": 24.739307374576665, "trajectory_balance_rmse": 109.47596777600708} {"canonical_length": 31, "canonical_nll": 1.8310984949911795, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.18181818181818182, "prompt_id": "5eee9bc9190ed48e", "reading_canonical_log_prob": -56.76405334472656, "reading_distances": [26, 31, 23, 22], "reading_log_probs": [-69.18014526367188, -70.57032012939453, -67.93341064453125, -66.33455657958984], "reading_logprob_std": 1.803598882073978, "tokenization_path_logprob_std": 186.19761820079762, "trajectory_balance_rmse": 960.7034019810442} {"canonical_length": 14, "canonical_nll": 0.670114244733538, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "89b79b80175b4a32", "reading_canonical_log_prob": -9.381599426269531, "reading_distances": [28, 41, 37, 37], "reading_log_probs": [-9.40075397491455, -8.381987571716309, -8.962419509887695, -9.391345977783203], "reading_logprob_std": 0.48043332346387196, "tokenization_path_logprob_std": 72.01674574987933, "trajectory_balance_rmse": 328.48336140394224} {"canonical_length": 10, "canonical_nll": 0.2786454916000366, "generation_rouge_l_f1": 0.6666666666666667, "generation_token_f1": 0.6666666666666667, "prompt_id": "5a773e9fe3bdab08", "reading_canonical_log_prob": -2.786454916000366, "reading_distances": [19, 16, 20, 21], "reading_log_probs": [-0.9439915418624878, -0.6485942602157593, -0.29238995909690857, -0.31381985545158386], "reading_logprob_std": 0.3093508226511604, "tokenization_path_logprob_std": 50.828436046268756, "trajectory_balance_rmse": 291.08894497326486} {"canonical_length": 37, "canonical_nll": 1.2140222497888513, "generation_rouge_l_f1": 0.3516483516483516, "generation_token_f1": 0.3956043956043956, "prompt_id": "bc2e832a6508e078", "reading_canonical_log_prob": -44.9188232421875, "reading_distances": [22, 12, 14, 17], "reading_log_probs": [-49.267723083496094, -45.150306701660156, -45.78334045410156, -47.89017868041992], "reading_logprob_std": 1.9003961441298467, "tokenization_path_logprob_std": 52.76643318290274, "trajectory_balance_rmse": 1161.6353503161042} {"canonical_length": 245, "canonical_nll": 0.7414211973852041, "generation_rouge_l_f1": 0.22921348314606743, "generation_token_f1": 0.3460674157303371, "prompt_id": "66a81182d7c90c5b", "reading_canonical_log_prob": -181.648193359375, "reading_distances": [16, 10, 19, 16], "reading_log_probs": [-181.14260864257812, -179.3778076171875, -180.28457641601562, -178.4797821044922], "reading_logprob_std": 1.1484536384439434, "tokenization_path_logprob_std": 204.39541715057848, "trajectory_balance_rmse": 4820.753677041484} {"canonical_length": 92, "canonical_nll": 1.3673439025878906, "generation_rouge_l_f1": 0.23717948717948723, "generation_token_f1": 0.3205128205128205, "prompt_id": "aed52e93b5146883", "reading_canonical_log_prob": -125.79563903808594, "reading_distances": [24, 29, 23, 28], "reading_log_probs": [-133.28536987304688, -134.71893310546875, -128.16209411621094, -130.90089416503906], "reading_logprob_std": 2.873142400072956, "tokenization_path_logprob_std": 235.09999513902778, "trajectory_balance_rmse": 1958.708977026517} {"canonical_length": 131, "canonical_nll": 0.8743420084014194, "generation_rouge_l_f1": 0.362116991643454, "generation_token_f1": 0.5236768802228413, "prompt_id": "b1543c03f1b0c7a8", "reading_canonical_log_prob": -114.53880310058594, "reading_distances": [17, 20, 19, 19], "reading_log_probs": [-119.90132141113281, -119.62541961669922, -117.11094665527344, -120.26669311523438], "reading_logprob_std": 1.4343509349088839, "tokenization_path_logprob_std": 218.33808059728935, "trajectory_balance_rmse": 3580.4353845517544} {"canonical_length": 501, "canonical_nll": 0.3434403166323603, "generation_rouge_l_f1": 0.12085769980506822, "generation_token_f1": 0.17543859649122806, "prompt_id": "c52f77a538102872", "reading_canonical_log_prob": -172.0635986328125, "reading_distances": [32, 29, 23, 31], "reading_log_probs": [-174.384033203125, -176.1771240234375, -173.87533569335938, -179.90843200683594], "reading_logprob_std": 2.7327042451863623, "tokenization_path_logprob_std": 70.28220225675376, "trajectory_balance_rmse": 3709.990014901977} {"canonical_length": 67, "canonical_nll": 0.7460614104769123, "generation_rouge_l_f1": 0.1833810888252149, "generation_token_f1": 0.21203438395415472, "prompt_id": "fdb23a8f949c30eb", "reading_canonical_log_prob": -49.986114501953125, "reading_distances": [19, 20, 23, 21], "reading_log_probs": [-54.87068176269531, -51.40739059448242, -50.55787658691406, -53.1062126159668], "reading_logprob_std": 1.9107091265750695, "tokenization_path_logprob_std": 27.768902617550356, "trajectory_balance_rmse": 1116.9722898360367} {"canonical_length": 268, "canonical_nll": 1.3207982760756762, "generation_rouge_l_f1": 0.2567049808429118, "generation_token_f1": 0.49042145593869735, "prompt_id": "8aa2f18f3606aa8c", "reading_canonical_log_prob": -353.97393798828125, "reading_distances": [22, 27, 24, 22], "reading_log_probs": [-355.12957763671875, -358.70220947265625, -362.8825988769531, -354.40045166015625], "reading_logprob_std": 3.8872897427894744, "tokenization_path_logprob_std": 569.7985124001789, "trajectory_balance_rmse": 7831.158769668171} {"canonical_length": 123, "canonical_nll": 1.104145081062627, "generation_rouge_l_f1": 0.14285714285714285, "generation_token_f1": 0.20833333333333334, "prompt_id": "75e528b4a6af3ee2", "reading_canonical_log_prob": -135.80984497070312, "reading_distances": [21, 18, 15, 19], "reading_log_probs": [-130.60179138183594, -133.39727783203125, -134.26748657226562, -133.660400390625], "reading_logprob_std": 1.6279385009721845, "tokenization_path_logprob_std": 170.4584315893789, "trajectory_balance_rmse": 1849.737107269339} {"canonical_length": 323, "canonical_nll": 1.632276222063661, "generation_rouge_l_f1": 0.16433566433566432, "generation_token_f1": 0.37412587412587417, "prompt_id": "af23751b53e953f7", "reading_canonical_log_prob": -527.2252197265625, "reading_distances": [33, 36, 31, 28], "reading_log_probs": [-523.1520385742188, -523.5457763671875, -526.200927734375, -524.57666015625], "reading_logprob_std": 1.361099327354454, "tokenization_path_logprob_std": 263.60724102379993, "trajectory_balance_rmse": 8401.696700411347} {"canonical_length": 250, "canonical_nll": 1.6856932373046876, "generation_rouge_l_f1": 0.30142566191446024, "generation_token_f1": 0.5336048879837068, "prompt_id": "b8fd0dbaa9eea228", "reading_canonical_log_prob": -421.4233093261719, "reading_distances": [15, 20, 19, 21], "reading_log_probs": [-423.2566833496094, -424.4896240234375, -427.5362854003906, -425.7755126953125], "reading_logprob_std": 1.8306559259872894, "tokenization_path_logprob_std": 185.07399130901945, "trajectory_balance_rmse": 6887.195612290222} {"canonical_length": 13, "canonical_nll": 1.951868203970102, "generation_rouge_l_f1": 0.1818181818181818, "generation_token_f1": 0.1818181818181818, "prompt_id": "44c43dead4708e04", "reading_canonical_log_prob": -25.374286651611328, "reading_distances": [34, 24, 29, 22], "reading_log_probs": [-28.862747192382812, -26.279041290283203, -29.352201461791992, -26.69866943359375], "reading_logprob_std": 1.534600047231438, "tokenization_path_logprob_std": 48.487153169540775, "trajectory_balance_rmse": 514.2528369708349} {"canonical_length": 138, "canonical_nll": 1.9253265823143115, "generation_rouge_l_f1": 0.25855513307984795, "generation_token_f1": 0.44106463878326996, "prompt_id": "36e5fedd319115db", "reading_canonical_log_prob": -265.695068359375, "reading_distances": [30, 27, 35, 26], "reading_log_probs": [-257.47015380859375, -259.855224609375, -259.823974609375, -255.187744140625], "reading_logprob_std": 2.230833409800688, "tokenization_path_logprob_std": 122.34499798223874, "trajectory_balance_rmse": 4086.529910156771} {"canonical_length": 13, "canonical_nll": 0.6665157171396109, "generation_rouge_l_f1": 0.8695652173913043, "generation_token_f1": 0.8695652173913043, "prompt_id": "15b19e1f65328c81", "reading_canonical_log_prob": -8.664704322814941, "reading_distances": [24, 21, 16, 24], "reading_log_probs": [-7.754962921142578, -7.081103324890137, -7.693391799926758, -7.0048346519470215], "reading_logprob_std": 0.39532636431231954, "tokenization_path_logprob_std": 45.060908323090935, "trajectory_balance_rmse": 296.1106180094085} {"canonical_length": 20, "canonical_nll": 1.2180566787719727, "generation_rouge_l_f1": 0.22448979591836735, "generation_token_f1": 0.30612244897959184, "prompt_id": "7e1ececb041b4ecb", "reading_canonical_log_prob": -24.361133575439453, "reading_distances": [21, 24, 26, 23], "reading_log_probs": [-26.308624267578125, -25.683692932128906, -26.89679527282715, -23.891277313232422], "reading_logprob_std": 1.3005626938766577, "tokenization_path_logprob_std": 104.36098059325877, "trajectory_balance_rmse": 633.1995594116579} {"canonical_length": 359, "canonical_nll": 1.4020389992546571, "generation_rouge_l_f1": 0.22186495176848875, "generation_token_f1": 0.38263665594855306, "prompt_id": "4db416527bef041c", "reading_canonical_log_prob": -503.3320007324219, "reading_distances": [18, 21, 22, 15], "reading_log_probs": [-500.4862060546875, -501.59002685546875, -502.0563049316406, -499.9340515136719], "reading_logprob_std": 0.9769056005166203, "tokenization_path_logprob_std": 469.2655824757705, "trajectory_balance_rmse": 10971.568148377875} {"canonical_length": 78, "canonical_nll": 0.7841573372865335, "generation_rouge_l_f1": 0.4748201438848921, "generation_token_f1": 0.7338129496402876, "prompt_id": "ba11bfd32f28c7ee", "reading_canonical_log_prob": -61.16427230834961, "reading_distances": [16, 14, 14, 18], "reading_log_probs": [-62.82697296142578, -64.84217834472656, -65.58905029296875, -63.02451705932617], "reading_logprob_std": 1.3591568634909597, "tokenization_path_logprob_std": 163.64698026681836, "trajectory_balance_rmse": 2237.1871943088336} {"canonical_length": 292, "canonical_nll": 1.6292974393661708, "generation_rouge_l_f1": 0.2003514938488577, "generation_token_f1": 0.37258347978910367, "prompt_id": "2bb0a1d6597bcbfc", "reading_canonical_log_prob": -475.7548522949219, "reading_distances": [50, 64, 52, 54], "reading_log_probs": [-491.5003662109375, -484.0654602050781, -486.97943115234375, -492.541259765625], "reading_logprob_std": 3.9587927758741954, "tokenization_path_logprob_std": 320.5255356050811, "trajectory_balance_rmse": 8325.515175233064} {"canonical_length": 2, "canonical_nll": 5.06764554977417, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "bad903d11c6658df", "reading_canonical_log_prob": -10.13529109954834, "reading_distances": [28, 34, 34, 30], "reading_log_probs": [-15.394673347473145, -12.885958671569824, -13.295641899108887, -14.84607219696045], "reading_logprob_std": 1.2046529602787712, "tokenization_path_logprob_std": 26.496294960420354, "trajectory_balance_rmse": 49.921250071147064} {"canonical_length": 99, "canonical_nll": 1.1074955487492109, "generation_rouge_l_f1": 0.17796610169491528, "generation_token_f1": 0.2711864406779661, "prompt_id": "7c8a24a51c291f25", "reading_canonical_log_prob": -109.64205932617188, "reading_distances": [10, 13, 11, 10], "reading_log_probs": [-113.81546020507812, -113.09782409667969, -114.82230377197266, -115.53549194335938], "reading_logprob_std": 1.0767209783260716, "tokenization_path_logprob_std": 177.7442682076206, "trajectory_balance_rmse": 2781.9582850404445} {"canonical_length": 108, "canonical_nll": 1.929830621790003, "generation_rouge_l_f1": 0.2293333333333333, "generation_token_f1": 0.38933333333333336, "prompt_id": "cef16e435e5a8a53", "reading_canonical_log_prob": -208.4217071533203, "reading_distances": [20, 21, 18, 23], "reading_log_probs": [-215.33267211914062, -216.0108642578125, -211.4710693359375, -224.84011840820312], "reading_logprob_std": 5.64992916956432, "tokenization_path_logprob_std": 195.92808450767927, "trajectory_balance_rmse": 3690.3475754150463} {"canonical_length": 38, "canonical_nll": 3.2035660994680306, "generation_rouge_l_f1": 0.2247191011235955, "generation_token_f1": 0.3146067415730337, "prompt_id": "289239d155fcb98c", "reading_canonical_log_prob": -121.73551177978516, "reading_distances": [28, 17, 22, 19], "reading_log_probs": [-124.56327819824219, -125.38156127929688, -123.76495361328125, -123.39842224121094], "reading_logprob_std": 0.8824400254048074, "tokenization_path_logprob_std": 81.83174551166073, "trajectory_balance_rmse": 1022.4873211667679} {"canonical_length": 8, "canonical_nll": 1.1440492868423462, "generation_rouge_l_f1": 0.7272727272727273, "generation_token_f1": 0.7272727272727273, "prompt_id": "c036d970930b8362", "reading_canonical_log_prob": -9.15239429473877, "reading_distances": [24, 36, 31, 26], "reading_log_probs": [-12.896659851074219, -22.819835662841797, -11.087900161743164, -14.399388313293457], "reading_logprob_std": 5.192201047263691, "tokenization_path_logprob_std": 37.73624222391526, "trajectory_balance_rmse": 142.83149695410546} {"canonical_length": 99, "canonical_nll": 1.3537499013573233, "generation_rouge_l_f1": 0.1741573033707865, "generation_token_f1": 0.34269662921348315, "prompt_id": "1d5889fbec769861", "reading_canonical_log_prob": -134.021240234375, "reading_distances": [22, 16, 23, 24], "reading_log_probs": [-134.675537109375, -135.47410583496094, -137.51510620117188, -137.9301300048828], "reading_logprob_std": 1.5722388900286515, "tokenization_path_logprob_std": 232.86571565044673, "trajectory_balance_rmse": 2568.1470341970307} {"canonical_length": 185, "canonical_nll": 1.017669017894848, "generation_rouge_l_f1": 0.17972350230414746, "generation_token_f1": 0.28571428571428575, "prompt_id": "c20bcd26ca03079c", "reading_canonical_log_prob": -188.26876831054688, "reading_distances": [38, 36, 38, 37], "reading_log_probs": [-182.4345245361328, -185.04806518554688, -186.195556640625, -186.5855255126953], "reading_logprob_std": 1.8717051704884846, "tokenization_path_logprob_std": 217.30011406899126, "trajectory_balance_rmse": 4603.1264724477605} {"canonical_length": 16, "canonical_nll": 0.9744229316711426, "generation_rouge_l_f1": 0.12154696132596686, "generation_token_f1": 0.15469613259668508, "prompt_id": "93426790937338d2", "reading_canonical_log_prob": -15.590766906738281, "reading_distances": [15, 17, 15, 16], "reading_log_probs": [-16.224523544311523, -13.960103988647461, -13.904256820678711, -14.386168479919434], "reading_logprob_std": 1.0919276395383206, "tokenization_path_logprob_std": 49.610813548210345, "trajectory_balance_rmse": 294.93778996841183} {"canonical_length": 203, "canonical_nll": 1.6486847976158405, "generation_rouge_l_f1": 0.18390804597701152, "generation_token_f1": 0.3218390804597701, "prompt_id": "f380c029d22e1217", "reading_canonical_log_prob": -334.6830139160156, "reading_distances": [19, 12, 12, 11], "reading_log_probs": [-332.914306640625, -333.40753173828125, -331.15625, -330.77423095703125], "reading_logprob_std": 1.2930080915883126, "tokenization_path_logprob_std": 176.42067113422857, "trajectory_balance_rmse": 5528.710387600094} {"canonical_length": 191, "canonical_nll": 0.8165661077848904, "generation_rouge_l_f1": 0.4316939890710382, "generation_token_f1": 0.7103825136612022, "prompt_id": "b8396f86fe7c1b16", "reading_canonical_log_prob": -155.96412658691406, "reading_distances": [41, 36, 35, 34], "reading_log_probs": [-155.67433166503906, -154.9482421875, -157.44915771484375, -158.2892608642578], "reading_logprob_std": 1.5448253873387032, "tokenization_path_logprob_std": 267.7323550654303, "trajectory_balance_rmse": 3825.8372591881866} {"canonical_length": 149, "canonical_nll": 1.9667759837720218, "generation_rouge_l_f1": 0.16585365853658537, "generation_token_f1": 0.33658536585365856, "prompt_id": "6cb0fdf8b6463b4b", "reading_canonical_log_prob": -293.04962158203125, "reading_distances": [45, 51, 38, 49], "reading_log_probs": [-288.8053894042969, -289.30731201171875, -285.991455078125, -291.47186279296875], "reading_logprob_std": 2.2545378136463547, "tokenization_path_logprob_std": 177.24259198843328, "trajectory_balance_rmse": 4592.215967103724} {"canonical_length": 118, "canonical_nll": 1.5669691441422802, "generation_rouge_l_f1": 0.2464985994397759, "generation_token_f1": 0.3585434173669468, "prompt_id": "7c85ca0add25adb8", "reading_canonical_log_prob": -184.90235900878906, "reading_distances": [33, 33, 38, 35], "reading_log_probs": [-179.08218383789062, -184.0982666015625, -184.20281982421875, -185.92922973632812], "reading_logprob_std": 2.9525114351128057, "tokenization_path_logprob_std": 216.09635172568593, "trajectory_balance_rmse": 3745.0055748696104} {"canonical_length": 114, "canonical_nll": 1.607045759234512, "generation_rouge_l_f1": 0.20578778135048229, "generation_token_f1": 0.37299035369774924, "prompt_id": "9a1872631342b4a7", "reading_canonical_log_prob": -183.20321655273438, "reading_distances": [30, 29, 23, 28], "reading_log_probs": [-185.03402709960938, -190.94679260253906, -184.2318878173828, -186.82362365722656], "reading_logprob_std": 2.994640243398694, "tokenization_path_logprob_std": 304.41773044848054, "trajectory_balance_rmse": 3673.533845922867} {"canonical_length": 13, "canonical_nll": 2.4568088238055887, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "d49306af1b19f211", "reading_canonical_log_prob": -31.938514709472656, "reading_distances": [22, 22, 20, 22], "reading_log_probs": [-32.22325134277344, -32.70178985595703, -31.75176239013672, -32.79942321777344], "reading_logprob_std": 0.4824376159598637, "tokenization_path_logprob_std": 99.72972760187328, "trajectory_balance_rmse": 310.76227624094514} {"canonical_length": 253, "canonical_nll": 1.693931006631361, "generation_rouge_l_f1": 0.11584327086882451, "generation_token_f1": 0.21124361158432708, "prompt_id": "807f35218626237b", "reading_canonical_log_prob": -428.5645446777344, "reading_distances": [17, 15, 21, 21], "reading_log_probs": [-439.8203430175781, -439.16619873046875, -438.97821044921875, -438.9295959472656], "reading_logprob_std": 0.41071139451633554, "tokenization_path_logprob_std": 243.50764240846942, "trajectory_balance_rmse": 7227.838205998426} {"canonical_length": 4, "canonical_nll": 4.403063774108887, "generation_rouge_l_f1": 0.05940594059405941, "generation_token_f1": 0.05940594059405941, "prompt_id": "1a44d44c0ed30f8a", "reading_canonical_log_prob": -17.612255096435547, "reading_distances": [20, 20, 28, 22], "reading_log_probs": [-17.81511688232422, -17.758623123168945, -17.292842864990234, -17.436241149902344], "reading_logprob_std": 0.25181887808765685, "tokenization_path_logprob_std": 45.18549117223305, "trajectory_balance_rmse": 176.5371520096534} {"canonical_length": 119, "canonical_nll": 0.18759721066771434, "generation_rouge_l_f1": 0.3198090692124105, "generation_token_f1": 0.3770883054892602, "prompt_id": "c78f29d81d80f8fc", "reading_canonical_log_prob": -22.324068069458008, "reading_distances": [21, 27, 20, 22], "reading_log_probs": [-28.56121826171875, -27.60655403137207, -27.14554786682129, -26.826873779296875], "reading_logprob_std": 0.7552835227410263, "tokenization_path_logprob_std": 79.58879480562373, "trajectory_balance_rmse": 2069.483801853007} {"canonical_length": 279, "canonical_nll": 1.0107510474420363, "generation_rouge_l_f1": 0.17557251908396945, "generation_token_f1": 0.34732824427480913, "prompt_id": "dbd0e34044b57777", "reading_canonical_log_prob": -281.9995422363281, "reading_distances": [25, 18, 19, 23], "reading_log_probs": [-279.82720947265625, -281.6820068359375, -298.0347900390625, -290.9985046386719], "reading_logprob_std": 8.482698376989509, "tokenization_path_logprob_std": 355.5645001684247, "trajectory_balance_rmse": 6354.087921392081} {"canonical_length": 11, "canonical_nll": 0.06050971963188865, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1c94de058e5cb066", "reading_canonical_log_prob": -0.6656069159507751, "reading_distances": [20, 22, 23, 12], "reading_log_probs": [-1.094534158706665, -0.7639855146408081, -0.8293938636779785, -0.7815672159194946], "reading_logprob_std": 0.15394394163641162, "tokenization_path_logprob_std": 38.176164447245775, "trajectory_balance_rmse": 312.994706968105} {"canonical_length": 40, "canonical_nll": 0.38380892276763917, "generation_rouge_l_f1": 0.26455026455026454, "generation_token_f1": 0.2751322751322751, "prompt_id": "91069c7b26de61af", "reading_canonical_log_prob": -15.352356910705566, "reading_distances": [45, 43, 46, 36], "reading_log_probs": [-16.17087173461914, -15.517782211303711, -17.34064483642578, -16.796367645263672], "reading_logprob_std": 0.7874000831972565, "tokenization_path_logprob_std": 114.17414048972822, "trajectory_balance_rmse": 898.3273457529298} {"canonical_length": 10, "canonical_nll": 0.25897960662841796, "generation_rouge_l_f1": 0.875, "generation_token_f1": 0.875, "prompt_id": "c9ec0273bafaec5d", "reading_canonical_log_prob": -2.5897960662841797, "reading_distances": [41, 38, 42, 49], "reading_log_probs": [-2.613532066345215, -2.153672218322754, -1.59260892868042, -3.701496124267578], "reading_logprob_std": 0.894211803148612, "tokenization_path_logprob_std": 75.60743086699331, "trajectory_balance_rmse": 234.59275551657478} {"canonical_length": 12, "canonical_nll": 0.07092547913392384, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1cb768d05dd2b1b9", "reading_canonical_log_prob": -0.8511057496070862, "reading_distances": [41, 43, 37, 40], "reading_log_probs": [-0.9445269107818604, -1.0688207149505615, -0.8442401885986328, -0.942312479019165], "reading_logprob_std": 0.09200099674837196, "tokenization_path_logprob_std": 46.0361343384974, "trajectory_balance_rmse": 233.63805882577702} {"canonical_length": 117, "canonical_nll": 0.9462400257077992, "generation_rouge_l_f1": 0.380952380952381, "generation_token_f1": 0.5238095238095238, "prompt_id": "31716783a47d0f33", "reading_canonical_log_prob": -110.7100830078125, "reading_distances": [29, 30, 29, 26], "reading_log_probs": [-108.95602416992188, -108.61299133300781, -110.78338623046875, -109.421142578125], "reading_logprob_std": 0.9527455490771838, "tokenization_path_logprob_std": 183.06475368908218, "trajectory_balance_rmse": 3115.019520860265} {"canonical_length": 38, "canonical_nll": 0.5978695718865645, "generation_rouge_l_f1": 0.27624309392265195, "generation_token_f1": 0.30939226519337015, "prompt_id": "64444cea846a2353", "reading_canonical_log_prob": -22.719043731689453, "reading_distances": [621, 595, 616, 591], "reading_log_probs": [-53.107330322265625, -42.5085334777832, -46.65373992919922, -49.277252197265625], "reading_logprob_std": 4.4584557136292595, "tokenization_path_logprob_std": 50.572125733372026, "trajectory_balance_rmse": 1163.4691708376915} {"canonical_length": 17, "canonical_nll": 1.5189639820772058, "generation_rouge_l_f1": 0.45161290322580644, "generation_token_f1": 0.45161290322580644, "prompt_id": "bcb6a50b95a69177", "reading_canonical_log_prob": -25.8223876953125, "reading_distances": [31, 30, 30, 36], "reading_log_probs": [-25.63232421875, -27.39872932434082, -27.092281341552734, -27.735225677490234], "reading_logprob_std": 0.9262097460260381, "tokenization_path_logprob_std": 62.50843796906355, "trajectory_balance_rmse": 554.2051627335707} {"canonical_length": 47, "canonical_nll": 0.796327956179355, "generation_rouge_l_f1": 0.4307692307692308, "generation_token_f1": 0.5076923076923077, "prompt_id": "5a3dfb14391da9f8", "reading_canonical_log_prob": -37.42741394042969, "reading_distances": [25, 28, 24, 23], "reading_log_probs": [-39.66561508178711, -40.5410270690918, -37.64649963378906, -40.71063232421875], "reading_logprob_std": 1.406271623172639, "tokenization_path_logprob_std": 99.4060360257328, "trajectory_balance_rmse": 987.3881470884457} {"canonical_length": 255, "canonical_nll": 0.15314492618336398, "generation_rouge_l_f1": 0.9528795811518325, "generation_token_f1": 0.9528795811518325, "prompt_id": "1b64247b2a4cbf46", "reading_canonical_log_prob": -39.05195617675781, "reading_distances": [40, 50, 41, 46], "reading_log_probs": [-47.4119987487793, -44.52996063232422, -43.35826873779297, -44.65601348876953], "reading_logprob_std": 1.7177315010665888, "tokenization_path_logprob_std": 139.32146352834272, "trajectory_balance_rmse": 3791.4789185165146} {"canonical_length": 8, "canonical_nll": 0.0077181365340948105, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "ffe7a66de245047f", "reading_canonical_log_prob": -0.061745092272758484, "reading_distances": [23, 21, 28, 26], "reading_log_probs": [-0.21222653985023499, -0.1808602213859558, -0.17769528925418854, -0.14419874548912048], "reading_logprob_std": 0.027809070855126974, "tokenization_path_logprob_std": 40.21486376591478, "trajectory_balance_rmse": 216.81024349071353} {"canonical_length": 195, "canonical_nll": 1.3778514372996795, "generation_rouge_l_f1": 0.35204081632653067, "generation_token_f1": 0.5918367346938777, "prompt_id": "9b0fbce06c332eb0", "reading_canonical_log_prob": -268.6810302734375, "reading_distances": [22, 24, 21, 17], "reading_log_probs": [-265.491455078125, -266.1229248046875, -265.12982177734375, -266.9853820800781], "reading_logprob_std": 0.8131492666807214, "tokenization_path_logprob_std": 376.7182380288842, "trajectory_balance_rmse": 6277.398875069962} {"canonical_length": 103, "canonical_nll": 1.587513155150182, "generation_rouge_l_f1": 0.27004219409282704, "generation_token_f1": 0.5569620253164557, "prompt_id": "f63c128779f1525d", "reading_canonical_log_prob": -163.51385498046875, "reading_distances": [33, 28, 31, 28], "reading_log_probs": [-162.7452392578125, -159.55393981933594, -158.8134765625, -161.72857666015625], "reading_logprob_std": 1.8360258857851446, "tokenization_path_logprob_std": 185.76685379192347, "trajectory_balance_rmse": 2949.3652758967014} {"canonical_length": 45, "canonical_nll": 0.6608673943413629, "generation_rouge_l_f1": 0.46376811594202905, "generation_token_f1": 0.5507246376811594, "prompt_id": "e0f5d30e662ee161", "reading_canonical_log_prob": -29.739032745361328, "reading_distances": [36, 27, 29, 30], "reading_log_probs": [-33.7054443359375, -35.4450798034668, -34.504905700683594, -32.64603042602539], "reading_logprob_std": 1.1889018011649013, "tokenization_path_logprob_std": 39.26046350995382, "trajectory_balance_rmse": 1462.269624512863} {"canonical_length": 10, "canonical_nll": 1.8034976959228515, "generation_rouge_l_f1": 0.5714285714285714, "generation_token_f1": 0.5714285714285714, "prompt_id": "10787d0a825f897d", "reading_canonical_log_prob": -18.034976959228516, "reading_distances": [35, 37, 43, 32], "reading_log_probs": [-19.33698272705078, -21.532821655273438, -21.115568161010742, -19.77117919921875], "reading_logprob_std": 1.0511290605452408, "tokenization_path_logprob_std": 60.22260846649291, "trajectory_balance_rmse": 255.89456499268292} {"canonical_length": 138, "canonical_nll": 1.0299148559570312, "generation_rouge_l_f1": 0.241025641025641, "generation_token_f1": 0.4256410256410257, "prompt_id": "55c070ce04e00179", "reading_canonical_log_prob": -142.1282501220703, "reading_distances": [20, 14, 18, 20], "reading_log_probs": [-143.26165771484375, -142.8596954345703, -151.91427612304688, -146.77117919921875], "reading_logprob_std": 4.1940702455486605, "tokenization_path_logprob_std": 195.68519461389712, "trajectory_balance_rmse": 4174.27394293266} {"canonical_length": 57, "canonical_nll": 1.8706362540261787, "generation_rouge_l_f1": 0.34234234234234234, "generation_token_f1": 0.5045045045045045, "prompt_id": "9718a6939e1171c6", "reading_canonical_log_prob": -106.62626647949219, "reading_distances": [19, 17, 25, 27], "reading_log_probs": [-104.66923522949219, -106.93611145019531, -106.97705078125, -106.09024810791016], "reading_logprob_std": 1.0796449194698974, "tokenization_path_logprob_std": 138.4520447074364, "trajectory_balance_rmse": 1913.8778659364864} {"canonical_length": 180, "canonical_nll": 1.5796661376953125, "generation_rouge_l_f1": 0.22421524663677128, "generation_token_f1": 0.43946188340807174, "prompt_id": "abf26ab3207cd57f", "reading_canonical_log_prob": -284.33990478515625, "reading_distances": [23, 22, 24, 26], "reading_log_probs": [-286.87548828125, -285.7469482421875, -287.3980712890625, -285.34381103515625], "reading_logprob_std": 0.9574899007481573, "tokenization_path_logprob_std": 356.7919488868472, "trajectory_balance_rmse": 4207.4696467665} {"canonical_length": 8, "canonical_nll": 0.022078506648540497, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "9706792c2f395e2c", "reading_canonical_log_prob": -0.17662805318832397, "reading_distances": [26, 27, 19, 24], "reading_log_probs": [-0.32977667450904846, -0.13709206879138947, -0.09125220030546188, -0.11807189881801605], "reading_logprob_std": 0.10878976313743285, "tokenization_path_logprob_std": 49.79090722310486, "trajectory_balance_rmse": 212.46740789631917} {"canonical_length": 198, "canonical_nll": 1.4147890649660668, "generation_rouge_l_f1": 0.2554112554112554, "generation_token_f1": 0.42857142857142855, "prompt_id": "4d13949d8c9ff709", "reading_canonical_log_prob": -280.12823486328125, "reading_distances": [31, 26, 23, 26], "reading_log_probs": [-281.18475341796875, -277.9440612792969, -279.70355224609375, -276.8740234375], "reading_logprob_std": 1.9045003311784248, "tokenization_path_logprob_std": 435.86111787102374, "trajectory_balance_rmse": 6350.435970258623} {"canonical_length": 259, "canonical_nll": 1.5537982484088442, "generation_rouge_l_f1": 0.2980935875216638, "generation_token_f1": 0.6239168110918544, "prompt_id": "cf2b1f1029bc9513", "reading_canonical_log_prob": -402.4337463378906, "reading_distances": [29, 30, 35, 24], "reading_log_probs": [-403.11065673828125, -405.2013854980469, -401.653076171875, -404.3244934082031], "reading_logprob_std": 1.5401568150057303, "tokenization_path_logprob_std": 66.31489685034177, "trajectory_balance_rmse": 5781.133974581756} {"canonical_length": 159, "canonical_nll": 1.7911910530906054, "generation_rouge_l_f1": 0.2142857142857143, "generation_token_f1": 0.4081632653061224, "prompt_id": "0c1e711645beb310", "reading_canonical_log_prob": -284.79937744140625, "reading_distances": [20, 22, 27, 25], "reading_log_probs": [-290.3857421875, -292.1253662109375, -298.99267578125, -295.4012451171875], "reading_logprob_std": 3.797483430132105, "tokenization_path_logprob_std": 90.69626543730274, "trajectory_balance_rmse": 3785.518555213731} {"canonical_length": 36, "canonical_nll": 0.9426498413085938, "generation_rouge_l_f1": 0.3063063063063063, "generation_token_f1": 0.36036036036036034, "prompt_id": "602cd10603b86b5c", "reading_canonical_log_prob": -33.935394287109375, "reading_distances": [11, 13, 14, 14], "reading_log_probs": [-33.43626403808594, -33.32017517089844, -34.173492431640625, -32.884395599365234], "reading_logprob_std": 0.535518901660871, "tokenization_path_logprob_std": 99.11756605874943, "trajectory_balance_rmse": 940.1691587579687} {"canonical_length": 8, "canonical_nll": 2.89385986328125, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "b36657ba534be993", "reading_canonical_log_prob": -23.15087890625, "reading_distances": [31, 35, 29, 34], "reading_log_probs": [-20.763944625854492, -19.345109939575195, -22.62203025817871, -21.534515380859375], "reading_logprob_std": 1.3776110040483123, "tokenization_path_logprob_std": 10.528515929226389, "trajectory_balance_rmse": 145.83370766318407} {"canonical_length": 33, "canonical_nll": 1.8287471424449573, "generation_rouge_l_f1": 0.28, "generation_token_f1": 0.36000000000000004, "prompt_id": "f11461f699664b03", "reading_canonical_log_prob": -60.348655700683594, "reading_distances": [21, 20, 20, 18], "reading_log_probs": [-68.51408386230469, -71.3163070678711, -65.52867126464844, -70.31700897216797], "reading_logprob_std": 2.540327818410928, "tokenization_path_logprob_std": 82.07505007360555, "trajectory_balance_rmse": 965.5403326056584} {"canonical_length": 323, "canonical_nll": 1.4049527667255224, "generation_rouge_l_f1": 0.27303754266211605, "generation_token_f1": 0.5255972696245734, "prompt_id": "167bb9079aab9f14", "reading_canonical_log_prob": -453.79974365234375, "reading_distances": [19, 26, 28, 23], "reading_log_probs": [-451.65966796875, -451.107421875, -453.12799072265625, -456.43756103515625], "reading_logprob_std": 2.3933161652096397, "tokenization_path_logprob_std": 413.0293426785192, "trajectory_balance_rmse": 10715.740292189654} {"canonical_length": 330, "canonical_nll": 1.2146417791193183, "generation_rouge_l_f1": 0.19366197183098594, "generation_token_f1": 0.39788732394366194, "prompt_id": "e95dc10521ada79e", "reading_canonical_log_prob": -400.831787109375, "reading_distances": [39, 43, 31, 31], "reading_log_probs": [-409.947509765625, -424.11114501953125, -422.91046142578125, -420.753173828125], "reading_logprob_std": 6.472895775040616, "tokenization_path_logprob_std": 231.02270690758414, "trajectory_balance_rmse": 8080.741292331165} {"canonical_length": 126, "canonical_nll": 1.4199715266152033, "generation_rouge_l_f1": 0.20555555555555555, "generation_token_f1": 0.36111111111111116, "prompt_id": "78584ee870db48f1", "reading_canonical_log_prob": -178.91641235351562, "reading_distances": [32, 35, 34, 38], "reading_log_probs": [-184.73806762695312, -182.74163818359375, -185.2165069580078, -185.17410278320312], "reading_logprob_std": 1.1707696530581093, "tokenization_path_logprob_std": 160.40486885087824, "trajectory_balance_rmse": 3617.5911185646846} {"canonical_length": 318, "canonical_nll": 1.44480617691136, "generation_rouge_l_f1": 0.2150170648464164, "generation_token_f1": 0.4436860068259385, "prompt_id": "16eff2d1d4178803", "reading_canonical_log_prob": -459.4483642578125, "reading_distances": [17, 25, 25, 22], "reading_log_probs": [-455.52838134765625, -456.49859619140625, -456.2811584472656, -458.1939392089844], "reading_logprob_std": 1.1252179892153689, "tokenization_path_logprob_std": 336.19826995622236, "trajectory_balance_rmse": 10411.139212314369} {"canonical_length": 104, "canonical_nll": 1.288488241342398, "generation_rouge_l_f1": 0.375, "generation_token_f1": 0.6145833333333334, "prompt_id": "22619d053543f675", "reading_canonical_log_prob": -134.00277709960938, "reading_distances": [43, 39, 43, 38], "reading_log_probs": [-129.09703063964844, -128.42356872558594, -133.6046142578125, -130.2307586669922], "reading_logprob_std": 2.3012545510266915, "tokenization_path_logprob_std": 174.96040948764136, "trajectory_balance_rmse": 2761.6446457962543} {"canonical_length": 180, "canonical_nll": 0.997607930501302, "generation_rouge_l_f1": 0.2536873156342183, "generation_token_f1": 0.5191740412979351, "prompt_id": "3c84043f06521168", "reading_canonical_log_prob": -179.56942749023438, "reading_distances": [20, 28, 25, 24], "reading_log_probs": [-179.4725341796875, -181.65213012695312, -179.923583984375, -180.23556518554688], "reading_logprob_std": 0.941105777161269, "tokenization_path_logprob_std": 116.87306802482195, "trajectory_balance_rmse": 5015.230914560907} {"canonical_length": 238, "canonical_nll": 1.7337084858357406, "generation_rouge_l_f1": 0.25346534653465347, "generation_token_f1": 0.4712871287128713, "prompt_id": "4c550a924c0f2e5b", "reading_canonical_log_prob": -412.62261962890625, "reading_distances": [31, 22, 30, 26], "reading_log_probs": [-409.7347412109375, -411.009765625, -411.7393798828125, -410.9844055175781], "reading_logprob_std": 0.8321103058357389, "tokenization_path_logprob_std": 251.99088027745918, "trajectory_balance_rmse": 6098.852404939762} {"canonical_length": 72, "canonical_nll": 1.4520770178900824, "generation_rouge_l_f1": 0.12844036697247704, "generation_token_f1": 0.18348623853211007, "prompt_id": "9e6e21f46355104c", "reading_canonical_log_prob": -104.54954528808594, "reading_distances": [19, 13, 20, 21], "reading_log_probs": [-105.37559509277344, -103.70320892333984, -101.016845703125, -102.86561584472656], "reading_logprob_std": 1.8127241957856495, "tokenization_path_logprob_std": 155.04386789474162, "trajectory_balance_rmse": 1733.7536434706644} {"canonical_length": 27, "canonical_nll": 1.6199381792986836, "generation_rouge_l_f1": 0.27450980392156865, "generation_token_f1": 0.47058823529411764, "prompt_id": "4a624dc7ffd3988d", "reading_canonical_log_prob": -43.73833084106445, "reading_distances": [30, 26, 25, 36], "reading_log_probs": [-52.76067352294922, -54.008995056152344, -51.473976135253906, -53.97550964355469], "reading_logprob_std": 1.2032884010428053, "tokenization_path_logprob_std": 116.56510259915501, "trajectory_balance_rmse": 886.5908488098348} {"canonical_length": 7, "canonical_nll": 0.11274910824639457, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "0cd953773dcd678b", "reading_canonical_log_prob": -0.789243757724762, "reading_distances": [27, 29, 36, 31], "reading_log_probs": [-0.5340098142623901, -0.6841273307800293, -0.671411395072937, -0.4645426869392395], "reading_logprob_std": 0.10701009227704104, "tokenization_path_logprob_std": 14.271306694383835, "trajectory_balance_rmse": 221.33206694032438} {"canonical_length": 165, "canonical_nll": 0.6414577599727747, "generation_rouge_l_f1": 0.21465968586387435, "generation_token_f1": 0.5706806282722513, "prompt_id": "a86300073620913b", "reading_canonical_log_prob": -105.84053039550781, "reading_distances": [135, 135, 154, 137], "reading_log_probs": [-98.7673110961914, -99.27931213378906, -101.24531555175781, -94.84789276123047], "reading_logprob_std": 2.680083280491595, "tokenization_path_logprob_std": 110.295457361066, "trajectory_balance_rmse": 3556.5668089001406} {"canonical_length": 294, "canonical_nll": 1.4817338800754676, "generation_rouge_l_f1": 0.19310344827586204, "generation_token_f1": 0.3275862068965517, "prompt_id": "2837da3fee02c3ea", "reading_canonical_log_prob": -435.6297607421875, "reading_distances": [32, 24, 25, 27], "reading_log_probs": [-431.1231689453125, -431.57025146484375, -425.8236083984375, -430.5067138671875], "reading_logprob_std": 2.657563690336668, "tokenization_path_logprob_std": 315.30888394215793, "trajectory_balance_rmse": 7975.144131772458} {"canonical_length": 23, "canonical_nll": 1.2109247290569802, "generation_rouge_l_f1": 0.13846153846153847, "generation_token_f1": 0.14615384615384616, "prompt_id": "8041588a81157d9c", "reading_canonical_log_prob": -27.851268768310547, "reading_distances": [54, 53, 49, 37], "reading_log_probs": [-37.778411865234375, -32.743858337402344, -33.16156768798828, -33.46345901489258], "reading_logprob_std": 2.3463489739756898, "tokenization_path_logprob_std": 131.86787239782524, "trajectory_balance_rmse": 740.644387771819} {"canonical_length": 308, "canonical_nll": 0.9832287082424411, "generation_rouge_l_f1": 0.20502901353965183, "generation_token_f1": 0.39071566731141194, "prompt_id": "bc95295408e3badf", "reading_canonical_log_prob": -302.8344421386719, "reading_distances": [30, 32, 28, 32], "reading_log_probs": [-312.84124755859375, -322.87066650390625, -311.53240966796875, -316.0137634277344], "reading_logprob_std": 5.066416199990732, "tokenization_path_logprob_std": 195.30341404011472, "trajectory_balance_rmse": 6659.283422911646} {"canonical_length": 126, "canonical_nll": 1.4855092366536458, "generation_rouge_l_f1": 0.3421052631578947, "generation_token_f1": 0.4842105263157895, "prompt_id": "2a5a99cace11afef", "reading_canonical_log_prob": -187.17416381835938, "reading_distances": [27, 28, 22, 22], "reading_log_probs": [-190.0923309326172, -186.54852294921875, -188.6173553466797, -186.6965789794922], "reading_logprob_std": 1.6895914017668279, "tokenization_path_logprob_std": 293.26091153428695, "trajectory_balance_rmse": 4130.375541709546} {"canonical_length": 14, "canonical_nll": 0.9816591399056571, "generation_rouge_l_f1": 0.11111111111111112, "generation_token_f1": 0.13131313131313133, "prompt_id": "90db691d0627dd6d", "reading_canonical_log_prob": -13.7432279586792, "reading_distances": [16, 21, 16, 22], "reading_log_probs": [-12.262479782104492, -12.563587188720703, -13.167642593383789, -13.18375015258789], "reading_logprob_std": 0.4572079184108993, "tokenization_path_logprob_std": 19.506647394261382, "trajectory_balance_rmse": 393.58669766490556} {"canonical_length": 14, "canonical_nll": 1.4145984649658203, "generation_rouge_l_f1": 0.45454545454545453, "generation_token_f1": 0.45454545454545453, "prompt_id": "07c0c6eb902aa919", "reading_canonical_log_prob": -19.804378509521484, "reading_distances": [26, 26, 23, 27], "reading_log_probs": [-22.645904541015625, -19.99702262878418, -20.757862091064453, -18.904932022094727], "reading_logprob_std": 1.5753582525402579, "tokenization_path_logprob_std": 50.777664136974934, "trajectory_balance_rmse": 388.55863288800384} {"canonical_length": 6, "canonical_nll": 0.9905261198679606, "generation_rouge_l_f1": 0.8, "generation_token_f1": 0.8, "prompt_id": "285daef948799d4d", "reading_canonical_log_prob": -5.943156719207764, "reading_distances": [43, 43, 52, 50], "reading_log_probs": [-5.835887432098389, -5.6763482093811035, -5.509216785430908, -5.455158710479736], "reading_logprob_std": 0.1724579217297278, "tokenization_path_logprob_std": 31.41072136789563, "trajectory_balance_rmse": 168.22453575229358} {"canonical_length": 68, "canonical_nll": 0.7946654488058651, "generation_rouge_l_f1": 0.29577464788732394, "generation_token_f1": 0.3661971830985916, "prompt_id": "924060b32d38a1e4", "reading_canonical_log_prob": -54.03725051879883, "reading_distances": [34, 30, 33, 36], "reading_log_probs": [-57.20393371582031, -57.820533752441406, -56.99606704711914, -54.55213165283203], "reading_logprob_std": 1.4373145150694568, "tokenization_path_logprob_std": 93.70508241260184, "trajectory_balance_rmse": 1672.7333912807424} {"canonical_length": 147, "canonical_nll": 1.1790484992825254, "generation_rouge_l_f1": 0.3793103448275862, "generation_token_f1": 0.5310344827586206, "prompt_id": "ac4b8108300a4ae5", "reading_canonical_log_prob": -173.32012939453125, "reading_distances": [32, 32, 30, 29], "reading_log_probs": [-168.85882568359375, -167.26626586914062, -168.9225311279297, -168.4394073486328], "reading_logprob_std": 0.7675231272546575, "tokenization_path_logprob_std": 88.48342786469861, "trajectory_balance_rmse": 3041.825940166243} {"canonical_length": 53, "canonical_nll": 1.8800873666439417, "generation_rouge_l_f1": 0.15286624203821655, "generation_token_f1": 0.25477707006369427, "prompt_id": "10e2361fe555104f", "reading_canonical_log_prob": -99.6446304321289, "reading_distances": [25, 24, 18, 28], "reading_log_probs": [-102.51472473144531, -99.96141052246094, -101.91752624511719, -100.81765747070312], "reading_logprob_std": 1.1374443446855336, "tokenization_path_logprob_std": 81.85500764297471, "trajectory_balance_rmse": 1393.7777401582064} {"canonical_length": 27, "canonical_nll": 2.07154874448423, "generation_rouge_l_f1": 0.10408921933085502, "generation_token_f1": 0.1338289962825279, "prompt_id": "e53f2608a52ca6d9", "reading_canonical_log_prob": -55.93181610107422, "reading_distances": [24, 33, 32, 36], "reading_log_probs": [-61.035282135009766, -59.532344818115234, -58.642520904541016, -61.543365478515625], "reading_logprob_std": 1.3383187682244044, "tokenization_path_logprob_std": 86.55545199019853, "trajectory_balance_rmse": 595.4345592303782} {"canonical_length": 18, "canonical_nll": 1.1246038013034396, "generation_rouge_l_f1": 0.163265306122449, "generation_token_f1": 0.22448979591836732, "prompt_id": "3f4c8b5b47d9a555", "reading_canonical_log_prob": -20.242868423461914, "reading_distances": [17, 19, 15, 16], "reading_log_probs": [-21.022891998291016, -20.746505737304688, -19.603294372558594, -19.71002197265625], "reading_logprob_std": 0.7192528581767096, "tokenization_path_logprob_std": 84.31987930861082, "trajectory_balance_rmse": 538.5779618584666} {"canonical_length": 11, "canonical_nll": 0.8739865909923207, "generation_rouge_l_f1": 0.6153846153846153, "generation_token_f1": 0.6153846153846153, "prompt_id": "94a2d875be1c7a06", "reading_canonical_log_prob": -9.613852500915527, "reading_distances": [17, 16, 13, 17], "reading_log_probs": [-11.653623580932617, -8.95141887664795, -9.730624198913574, -8.071630477905273], "reading_logprob_std": 1.5265418490931695, "tokenization_path_logprob_std": 47.62395301638431, "trajectory_balance_rmse": 289.042881794629} {"canonical_length": 18, "canonical_nll": 0.4125303692287869, "generation_rouge_l_f1": 0.18987341772151897, "generation_token_f1": 0.18987341772151897, "prompt_id": "a86c032862f0a2ee", "reading_canonical_log_prob": -7.425546646118164, "reading_distances": [17, 17, 15, 14], "reading_log_probs": [-7.374669075012207, -7.995539665222168, -9.758840560913086, -8.654863357543945], "reading_logprob_std": 1.0194506204551788, "tokenization_path_logprob_std": 40.18934512715934, "trajectory_balance_rmse": 357.20716325051427} {"canonical_length": 80, "canonical_nll": 1.619894027709961, "generation_rouge_l_f1": 0.1927710843373494, "generation_token_f1": 0.29518072289156627, "prompt_id": "26b8508f97bd840b", "reading_canonical_log_prob": -129.59152221679688, "reading_distances": [33, 29, 38, 28], "reading_log_probs": [-129.3614959716797, -130.5476837158203, -128.3179931640625, -128.04656982421875], "reading_logprob_std": 1.1374581139505442, "tokenization_path_logprob_std": 169.54320258415422, "trajectory_balance_rmse": 2591.822705972661} {"canonical_length": 282, "canonical_nll": 0.9733194121232269, "generation_rouge_l_f1": 0.3097514340344169, "generation_token_f1": 0.44359464627151046, "prompt_id": "d2afb79833a3e8c4", "reading_canonical_log_prob": -274.47607421875, "reading_distances": [21, 26, 22, 22], "reading_log_probs": [-276.27044677734375, -279.00225830078125, -277.64190673828125, -277.68109130859375], "reading_logprob_std": 1.1154665930668755, "tokenization_path_logprob_std": 244.60945399675097, "trajectory_balance_rmse": 7981.150163907787} {"canonical_length": 206, "canonical_nll": 1.1926085092489003, "generation_rouge_l_f1": 0.3010752688172043, "generation_token_f1": 0.5548387096774194, "prompt_id": "8d6230df8e48bdfc", "reading_canonical_log_prob": -245.67735290527344, "reading_distances": [14, 14, 12, 17], "reading_log_probs": [-237.7877197265625, -242.6355743408203, -233.75067138671875, -231.92388916015625], "reading_logprob_std": 4.753970233635265, "tokenization_path_logprob_std": 137.03940399047923, "trajectory_balance_rmse": 5126.853667670152} {"canonical_length": 14, "canonical_nll": 3.297197069440569, "generation_rouge_l_f1": 0.1904761904761905, "generation_token_f1": 0.1904761904761905, "prompt_id": "74ebf94d0f403056", "reading_canonical_log_prob": -46.16075897216797, "reading_distances": [22, 24, 30, 26], "reading_log_probs": [-52.72685241699219, -56.00913619995117, -47.14952087402344, -46.68138885498047], "reading_logprob_std": 4.510602745143528, "tokenization_path_logprob_std": 15.315384875829157, "trajectory_balance_rmse": 334.59302779098726} {"canonical_length": 129, "canonical_nll": 1.3029621922692587, "generation_rouge_l_f1": 0.39111111111111113, "generation_token_f1": 0.6133333333333334, "prompt_id": "c307d4b735027f71", "reading_canonical_log_prob": -168.08212280273438, "reading_distances": [18, 17, 16, 15], "reading_log_probs": [-167.5872802734375, -170.2922821044922, -168.61119079589844, -167.13772583007812], "reading_logprob_std": 1.3998785406638568, "tokenization_path_logprob_std": 280.8141308808215, "trajectory_balance_rmse": 3527.4686645887728} {"canonical_length": 73, "canonical_nll": 0.9509530263404323, "generation_rouge_l_f1": 0.28169014084507044, "generation_token_f1": 0.46478873239436613, "prompt_id": "1044800b3405ca7b", "reading_canonical_log_prob": -69.41957092285156, "reading_distances": [12, 13, 13, 12], "reading_log_probs": [-77.6229476928711, -77.75663757324219, -76.21682739257812, -79.24030303955078], "reading_logprob_std": 1.2357376068794268, "tokenization_path_logprob_std": 124.59312397155577, "trajectory_balance_rmse": 1868.055423959214} {"canonical_length": 205, "canonical_nll": 1.4524772830125763, "generation_rouge_l_f1": 0.24411134903640258, "generation_token_f1": 0.44539614561027835, "prompt_id": "c87b3b9596ed8821", "reading_canonical_log_prob": -297.7578430175781, "reading_distances": [30, 29, 32, 27], "reading_log_probs": [-300.435791015625, -298.856689453125, -297.01812744140625, -299.63751220703125], "reading_logprob_std": 1.4623726754212043, "tokenization_path_logprob_std": 95.1897533108923, "trajectory_balance_rmse": 6143.660616454802} {"canonical_length": 11, "canonical_nll": 1.5577952645041726, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.18181818181818182, "prompt_id": "f2f04e5d3550740c", "reading_canonical_log_prob": -17.1357479095459, "reading_distances": [19, 11, 13, 17], "reading_log_probs": [-19.178974151611328, -17.029359817504883, -19.999752044677734, -20.042451858520508], "reading_logprob_std": 1.412559402599125, "tokenization_path_logprob_std": 75.82146232836982, "trajectory_balance_rmse": 280.01779431143126} {"canonical_length": 149, "canonical_nll": 1.4767234853450084, "generation_rouge_l_f1": 0.20297029702970298, "generation_token_f1": 0.3415841584158416, "prompt_id": "e64143ec52b6e7c2", "reading_canonical_log_prob": -220.03179931640625, "reading_distances": [20, 18, 16, 15], "reading_log_probs": [-219.10791015625, -218.5699462890625, -221.34815979003906, -229.46377563476562], "reading_logprob_std": 5.039918004976957, "tokenization_path_logprob_std": 160.93236516544684, "trajectory_balance_rmse": 3681.6219718609777} {"canonical_length": 48, "canonical_nll": 1.2765041987101238, "generation_rouge_l_f1": 0.29906542056074764, "generation_token_f1": 0.3738317757009346, "prompt_id": "c95d33d4bb11096d", "reading_canonical_log_prob": -61.27220153808594, "reading_distances": [18, 26, 25, 24], "reading_log_probs": [-63.05720520019531, -62.529136657714844, -63.730552673339844, -63.18639373779297], "reading_logprob_std": 0.49332537795748743, "tokenization_path_logprob_std": 172.5909560123881, "trajectory_balance_rmse": 1514.929771310302} {"canonical_length": 211, "canonical_nll": 1.3919226479191351, "generation_rouge_l_f1": 0.3273542600896861, "generation_token_f1": 0.5291479820627802, "prompt_id": "0accd9d43b5400d9", "reading_canonical_log_prob": -293.6956787109375, "reading_distances": [23, 20, 20, 30], "reading_log_probs": [-296.62677001953125, -294.726806640625, -293.0676574707031, -296.3832702636719], "reading_logprob_std": 1.6539440489060868, "tokenization_path_logprob_std": 198.0241387036867, "trajectory_balance_rmse": 5987.994138868196} {"canonical_length": 28, "canonical_nll": 1.1517058781215124, "generation_rouge_l_f1": 0.48648648648648646, "generation_token_f1": 0.5405405405405406, "prompt_id": "c5287ae056861247", "reading_canonical_log_prob": -32.247764587402344, "reading_distances": [19, 16, 23, 19], "reading_log_probs": [-29.942045211791992, -30.014026641845703, -30.57020378112793, -32.947776794433594], "reading_logprob_std": 1.414309703915098, "tokenization_path_logprob_std": 58.27630474818635, "trajectory_balance_rmse": 843.7772485665885} {"canonical_length": 9, "canonical_nll": 0.9489611519707574, "generation_rouge_l_f1": 0.05172413793103449, "generation_token_f1": 0.05172413793103449, "prompt_id": "9ba45d22569ee03c", "reading_canonical_log_prob": -8.540650367736816, "reading_distances": [41, 44, 43, 46], "reading_log_probs": [-12.874401092529297, -10.702020645141602, -9.139006614685059, -11.227527618408203], "reading_logprob_std": 1.5401756224105807, "tokenization_path_logprob_std": 63.657936244658515, "trajectory_balance_rmse": 304.1232826083328} {"canonical_length": 9, "canonical_nll": 1.144646750556098, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "61752b6441afd01d", "reading_canonical_log_prob": -10.301820755004883, "reading_distances": [20, 20, 16, 23], "reading_log_probs": [-17.230791091918945, -16.79564094543457, -16.268224716186523, -16.396434783935547], "reading_logprob_std": 0.43455066697997874, "tokenization_path_logprob_std": 54.49871960355543, "trajectory_balance_rmse": 342.9490194283296} {"canonical_length": 147, "canonical_nll": 1.383944764429209, "generation_rouge_l_f1": 0.27669902912621364, "generation_token_f1": 0.44660194174757284, "prompt_id": "1c1ecaad77fec061", "reading_canonical_log_prob": -203.43988037109375, "reading_distances": [14, 14, 17, 16], "reading_log_probs": [-201.8427734375, -205.07142639160156, -209.99264526367188, -204.88577270507812], "reading_logprob_std": 3.3719117355910337, "tokenization_path_logprob_std": 151.4109322087364, "trajectory_balance_rmse": 3852.5694055879967} {"canonical_length": 58, "canonical_nll": 1.1798535051016972, "generation_rouge_l_f1": 0.43410852713178294, "generation_token_f1": 0.6201550387596899, "prompt_id": "697abdbcde745481", "reading_canonical_log_prob": -68.43150329589844, "reading_distances": [21, 25, 17, 22], "reading_log_probs": [-70.40357971191406, -74.72714233398438, -68.87301635742188, -71.58958435058594], "reading_logprob_std": 2.4822239296338617, "tokenization_path_logprob_std": 59.99507343974391, "trajectory_balance_rmse": 1425.044662974733} {"canonical_length": 338, "canonical_nll": 1.3763424122827292, "generation_rouge_l_f1": 0.16176470588235295, "generation_token_f1": 0.2450980392156863, "prompt_id": "4034832cbc1502c1", "reading_canonical_log_prob": -465.2037353515625, "reading_distances": [29, 20, 22, 17], "reading_log_probs": [-466.60125732421875, -468.1158142089844, -463.5596618652344, -467.28680419921875], "reading_logprob_std": 1.9864652812776162, "tokenization_path_logprob_std": 254.81778214604486, "trajectory_balance_rmse": 9667.790045549307} {"canonical_length": 13, "canonical_nll": 1.4517857478215144, "generation_rouge_l_f1": 0.16, "generation_token_f1": 0.19999999999999996, "prompt_id": "228b1bd7f7bc8ed3", "reading_canonical_log_prob": -18.873214721679688, "reading_distances": [14, 16, 14, 15], "reading_log_probs": [-18.41461944580078, -19.864566802978516, -18.638988494873047, -16.565406799316406], "reading_logprob_std": 1.3619384527177927, "tokenization_path_logprob_std": 50.943270185593335, "trajectory_balance_rmse": 318.21361432891536} {"canonical_length": 115, "canonical_nll": 1.4500062362007473, "generation_rouge_l_f1": 0.30225080385852093, "generation_token_f1": 0.4887459807073955, "prompt_id": "28e1d60f144a6d02", "reading_canonical_log_prob": -166.75071716308594, "reading_distances": [20, 24, 27, 23], "reading_log_probs": [-167.06463623046875, -165.0445556640625, -167.48028564453125, -166.72682189941406], "reading_logprob_std": 1.0684145688828244, "tokenization_path_logprob_std": 71.75822846346261, "trajectory_balance_rmse": 2896.383805564476} {"canonical_length": 131, "canonical_nll": 1.475250244140625, "generation_rouge_l_f1": 0.2784090909090909, "generation_token_f1": 0.4602272727272727, "prompt_id": "f38548affd15660d", "reading_canonical_log_prob": -193.25778198242188, "reading_distances": [27, 26, 30, 26], "reading_log_probs": [-194.7229766845703, -189.2088165283203, -192.92465209960938, -190.48678588867188], "reading_logprob_std": 2.465918985932524, "tokenization_path_logprob_std": 185.20159388221475, "trajectory_balance_rmse": 3617.2563937339114} {"canonical_length": 39, "canonical_nll": 1.4547285422300682, "generation_rouge_l_f1": 0.2517482517482518, "generation_token_f1": 0.3496503496503497, "prompt_id": "b15afd8e011cefe5", "reading_canonical_log_prob": -56.734413146972656, "reading_distances": [20, 19, 30, 25], "reading_log_probs": [-58.84701919555664, -59.63373565673828, -59.660728454589844, -58.524009704589844], "reading_logprob_std": 0.5707984536590194, "tokenization_path_logprob_std": 146.04146964591288, "trajectory_balance_rmse": 1020.8472503479776} {"canonical_length": 286, "canonical_nll": 0.9862580199341674, "generation_rouge_l_f1": 0.15417558886509636, "generation_token_f1": 0.2698072805139186, "prompt_id": "60d1476fd22fd625", "reading_canonical_log_prob": -282.0697937011719, "reading_distances": [17, 24, 20, 22], "reading_log_probs": [-278.75140380859375, -280.8139343261719, -276.35986328125, -277.6197509765625], "reading_logprob_std": 1.8903925851315937, "tokenization_path_logprob_std": 154.31223332960278, "trajectory_balance_rmse": 6727.793996066484} {"canonical_length": 58, "canonical_nll": 1.3158556181808998, "generation_rouge_l_f1": 0.3278688524590164, "generation_token_f1": 0.39344262295081966, "prompt_id": "f1aff688c2307612", "reading_canonical_log_prob": -76.31962585449219, "reading_distances": [35, 51, 43, 44], "reading_log_probs": [-75.48596954345703, -76.02229309082031, -76.63279724121094, -77.16586303710938], "reading_logprob_std": 0.7296988847156932, "tokenization_path_logprob_std": 122.56417568663868, "trajectory_balance_rmse": 1619.1504197355098} {"canonical_length": 123, "canonical_nll": 1.622092766490409, "generation_rouge_l_f1": 0.18681318681318682, "generation_token_f1": 0.3846153846153846, "prompt_id": "a3f93481aca6425b", "reading_canonical_log_prob": -199.5174102783203, "reading_distances": [24, 22, 23, 28], "reading_log_probs": [-197.4419403076172, -196.10397338867188, -198.5654754638672, -199.07847595214844], "reading_logprob_std": 1.3197399714413225, "tokenization_path_logprob_std": 223.20252912972794, "trajectory_balance_rmse": 3298.835525343576} {"canonical_length": 163, "canonical_nll": 1.3225970121980444, "generation_rouge_l_f1": 0.18981481481481483, "generation_token_f1": 0.4351851851851852, "prompt_id": "39b8bde022351616", "reading_canonical_log_prob": -215.58331298828125, "reading_distances": [17, 22, 26, 16], "reading_log_probs": [-211.67453002929688, -211.32742309570312, -213.77178955078125, -211.34207153320312], "reading_logprob_std": 1.1728946018223039, "tokenization_path_logprob_std": 199.7623136629598, "trajectory_balance_rmse": 4756.910890033489} {"canonical_length": 56, "canonical_nll": 1.5880770002092635, "generation_rouge_l_f1": 0.2268041237113402, "generation_token_f1": 0.3917525773195877, "prompt_id": "969cc5c017329874", "reading_canonical_log_prob": -88.93231201171875, "reading_distances": [32, 31, 33, 33], "reading_log_probs": [-96.55075073242188, -98.11241912841797, -92.94498443603516, -95.42535400390625], "reading_logprob_std": 2.1752741213541102, "tokenization_path_logprob_std": 129.94115289020937, "trajectory_balance_rmse": 1237.0495160206644} {"canonical_length": 262, "canonical_nll": 1.734935964336832, "generation_rouge_l_f1": 0.2978723404255319, "generation_token_f1": 0.49129593810444877, "prompt_id": "a4cec9116b07e32b", "reading_canonical_log_prob": -454.55322265625, "reading_distances": [12, 20, 13, 16], "reading_log_probs": [-450.17041015625, -451.6413269042969, -451.72418212890625, -453.1202392578125], "reading_logprob_std": 1.2049314432751652, "tokenization_path_logprob_std": 215.56038377215617, "trajectory_balance_rmse": 6752.59900637679} {"canonical_length": 412, "canonical_nll": 0.9745933754930218, "generation_rouge_l_f1": 0.15536723163841806, "generation_token_f1": 0.46892655367231645, "prompt_id": "63a1199c25b071e9", "reading_canonical_log_prob": -401.532470703125, "reading_distances": [209, 221, 217, 245], "reading_log_probs": [-402.69439697265625, -405.24749755859375, -401.646484375, -405.62994384765625], "reading_logprob_std": 1.9411205816978463, "tokenization_path_logprob_std": 152.6165381261626, "trajectory_balance_rmse": 9135.486268410375} {"canonical_length": 34, "canonical_nll": 1.6462281171013327, "generation_rouge_l_f1": 0.0955631399317406, "generation_token_f1": 0.136518771331058, "prompt_id": "85280303ae668b99", "reading_canonical_log_prob": -55.97175598144531, "reading_distances": [10, 14, 10, 15], "reading_log_probs": [-55.59783935546875, -57.18743133544922, -57.94399642944336, -60.65019989013672], "reading_logprob_std": 2.110377071881969, "tokenization_path_logprob_std": 75.3242612631727, "trajectory_balance_rmse": 970.446062696084} {"canonical_length": 18, "canonical_nll": 1.167848163180881, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.25, "prompt_id": "963484b861b43113", "reading_canonical_log_prob": -21.02126693725586, "reading_distances": [15, 15, 17, 12], "reading_log_probs": [-21.10708999633789, -20.519550323486328, -21.350502014160156, -21.020727157592773], "reading_logprob_std": 0.34908466079324674, "tokenization_path_logprob_std": 24.156190359734715, "trajectory_balance_rmse": 405.0225639898692} {"canonical_length": 242, "canonical_nll": 1.2256552956321023, "generation_rouge_l_f1": 0.20077220077220076, "generation_token_f1": 0.5057915057915058, "prompt_id": "edfd3e88db49ba9c", "reading_canonical_log_prob": -296.60858154296875, "reading_distances": [24, 28, 23, 31], "reading_log_probs": [-304.06903076171875, -303.51861572265625, -302.01470947265625, -307.729248046875], "reading_logprob_std": 2.4249998787501186, "tokenization_path_logprob_std": 171.8742950246398, "trajectory_balance_rmse": 6961.507818515875} {"canonical_length": 3, "canonical_nll": 5.587650299072266, "generation_rouge_l_f1": 0.01680672268907563, "generation_token_f1": 0.01680672268907563, "prompt_id": "315bed08881fe880", "reading_canonical_log_prob": -16.762950897216797, "reading_distances": [31, 32, 31, 29], "reading_log_probs": [-23.50668716430664, -23.6466007232666, -24.835105895996094, -22.945404052734375], "reading_logprob_std": 0.7944864816773936, "tokenization_path_logprob_std": 25.639811833615823, "trajectory_balance_rmse": 109.64460907919018} {"canonical_length": 40, "canonical_nll": 1.4702775001525878, "generation_rouge_l_f1": 0.3225806451612903, "generation_token_f1": 0.45161290322580644, "prompt_id": "ec3686b4d6045b07", "reading_canonical_log_prob": -58.811100006103516, "reading_distances": [20, 30, 28, 25], "reading_log_probs": [-60.93951416015625, -62.95969772338867, -62.45181655883789, -62.06937789916992], "reading_logprob_std": 0.8583736675666392, "tokenization_path_logprob_std": 190.50318309426336, "trajectory_balance_rmse": 1423.0808927338674} {"canonical_length": 266, "canonical_nll": 1.5499854984140038, "generation_rouge_l_f1": 0.19921874999999997, "generation_token_f1": 0.41406249999999994, "prompt_id": "a1b8dba044d5aaa9", "reading_canonical_log_prob": -412.296142578125, "reading_distances": [26, 27, 19, 23], "reading_log_probs": [-415.1412353515625, -416.6497802734375, -416.11395263671875, -415.8592529296875], "reading_logprob_std": 0.6267876181306633, "tokenization_path_logprob_std": 81.79734131814631, "trajectory_balance_rmse": 8225.310696935883} {"canonical_length": 154, "canonical_nll": 1.4844468352082487, "generation_rouge_l_f1": 0.21395348837209302, "generation_token_f1": 0.37209302325581395, "prompt_id": "e45ec8d7efaf0407", "reading_canonical_log_prob": -228.6048126220703, "reading_distances": [22, 15, 16, 15], "reading_log_probs": [-246.84555053710938, -243.85194396972656, -246.83721923828125, -238.21829223632812], "reading_logprob_std": 4.065373480449252, "tokenization_path_logprob_std": 118.0893046462779, "trajectory_balance_rmse": 4906.46941189572} {"canonical_length": 3, "canonical_nll": 5.324610710144043, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "3521244aca15fc91", "reading_canonical_log_prob": -15.973832130432129, "reading_distances": [49, 49, 46, 48], "reading_log_probs": [-12.24083423614502, -11.313753128051758, -11.5802640914917, -11.952545166015625], "reading_logprob_std": 0.40790311807209145, "tokenization_path_logprob_std": 23.554191363103055, "trajectory_balance_rmse": 90.02466590970538} {"canonical_length": 23, "canonical_nll": 1.0984482972518257, "generation_rouge_l_f1": 0.27586206896551724, "generation_token_f1": 0.4137931034482759, "prompt_id": "99de541f6c007900", "reading_canonical_log_prob": -25.264310836791992, "reading_distances": [26, 23, 26, 26], "reading_log_probs": [-27.199737548828125, -25.399873733520508, -24.648786544799805, -27.416086196899414], "reading_logprob_std": 1.3564923990198792, "tokenization_path_logprob_std": 87.74773330008934, "trajectory_balance_rmse": 390.7227993456295} {"canonical_length": 176, "canonical_nll": 1.387504144148393, "generation_rouge_l_f1": 0.28506787330316746, "generation_token_f1": 0.48416289592760176, "prompt_id": "124ab50eaad8157c", "reading_canonical_log_prob": -244.2007293701172, "reading_distances": [18, 19, 23, 25], "reading_log_probs": [-245.44186401367188, -242.6542510986328, -248.34573364257812, -250.463134765625], "reading_logprob_std": 3.4067510150034974, "tokenization_path_logprob_std": 341.51967743303896, "trajectory_balance_rmse": 5430.220058075198} {"canonical_length": 15, "canonical_nll": 0.13964349428812664, "generation_rouge_l_f1": 0.8421052631578948, "generation_token_f1": 0.8421052631578948, "prompt_id": "7efa869db692612b", "reading_canonical_log_prob": -2.0946524143218994, "reading_distances": [48, 52, 54, 51], "reading_log_probs": [-24.446794509887695, -26.741456985473633, -31.6627197265625, -30.78610610961914], "reading_logprob_std": 3.401818007890367, "tokenization_path_logprob_std": 56.17158663337281, "trajectory_balance_rmse": 322.61866805510056} {"canonical_length": 248, "canonical_nll": 1.5392270242014239, "generation_rouge_l_f1": 0.20502901353965183, "generation_token_f1": 0.4796905222437137, "prompt_id": "b9153e23659ffc43", "reading_canonical_log_prob": -381.7283020019531, "reading_distances": [30, 27, 29, 36], "reading_log_probs": [-381.3863830566406, -382.92938232421875, -382.983154296875, -379.2807312011719], "reading_logprob_std": 1.741354312314023, "tokenization_path_logprob_std": 257.59164141217286, "trajectory_balance_rmse": 7867.419796621147} {"canonical_length": 48, "canonical_nll": 0.7913288275400797, "generation_rouge_l_f1": 0.693069306930693, "generation_token_f1": 0.693069306930693, "prompt_id": "2f3de773aa866413", "reading_canonical_log_prob": -37.98378372192383, "reading_distances": [23, 33, 25, 29], "reading_log_probs": [-42.08176040649414, -44.64399719238281, -41.42713165283203, -42.208404541015625], "reading_logprob_std": 1.4112747688303218, "tokenization_path_logprob_std": 89.0382308155189, "trajectory_balance_rmse": 1085.052559795704} {"canonical_length": 301, "canonical_nll": 1.413327340667826, "generation_rouge_l_f1": 0.2010968921389397, "generation_token_f1": 0.3692870201096893, "prompt_id": "54066b815b8806cf", "reading_canonical_log_prob": -425.4115295410156, "reading_distances": [32, 37, 45, 40], "reading_log_probs": [-429.8294372558594, -428.3064880371094, -430.5697937011719, -431.4682922363281], "reading_logprob_std": 1.3379152183175032, "tokenization_path_logprob_std": 504.88196904420033, "trajectory_balance_rmse": 8212.62145012609} {"canonical_length": 394, "canonical_nll": 1.393175773814245, "generation_rouge_l_f1": 0.20121951219512196, "generation_token_f1": 0.43902439024390244, "prompt_id": "38de2822dc803e85", "reading_canonical_log_prob": -548.9112548828125, "reading_distances": [26, 31, 38, 36], "reading_log_probs": [-539.5607299804688, -546.5276489257812, -543.0926513671875, -543.0669555664062], "reading_logprob_std": 2.844326416472849, "tokenization_path_logprob_std": 386.7159108818464, "trajectory_balance_rmse": 10471.373161466727} {"canonical_length": 329, "canonical_nll": 0.996651971231478, "generation_rouge_l_f1": 0.1821366024518389, "generation_token_f1": 0.39579684763572676, "prompt_id": "93d5783f4b9e5595", "reading_canonical_log_prob": -327.89849853515625, "reading_distances": [32, 31, 30, 32], "reading_log_probs": [-326.69488525390625, -327.8449401855469, -327.4240417480469, -327.3025817871094], "reading_logprob_std": 0.47518866993034437, "tokenization_path_logprob_std": 313.1769747375388, "trajectory_balance_rmse": 10375.010214367416} {"canonical_length": 25, "canonical_nll": 1.177150115966797, "generation_rouge_l_f1": 0.6857142857142857, "generation_token_f1": 0.6857142857142857, "prompt_id": "fffceba44eca0753", "reading_canonical_log_prob": -29.428752899169922, "reading_distances": [27, 31, 30, 28], "reading_log_probs": [-26.001453399658203, -24.218685150146484, -24.28890609741211, -24.497373580932617], "reading_logprob_std": 0.8415954050477252, "tokenization_path_logprob_std": 84.97871459842959, "trajectory_balance_rmse": 488.55071422919656} {"canonical_length": 49, "canonical_nll": 2.2157868365852202, "generation_rouge_l_f1": 0.21333333333333337, "generation_token_f1": 0.4, "prompt_id": "4d1f490ef0029856", "reading_canonical_log_prob": -108.57355499267578, "reading_distances": [21, 21, 25, 25], "reading_log_probs": [-110.43925476074219, -111.36091613769531, -110.53472900390625, -111.17473602294922], "reading_logprob_std": 0.45883592851205635, "tokenization_path_logprob_std": 125.61755315487129, "trajectory_balance_rmse": 1625.8057339222019} {"canonical_length": 58, "canonical_nll": 1.8494002243568157, "generation_rouge_l_f1": 0.3636363636363637, "generation_token_f1": 0.49090909090909085, "prompt_id": "04c2c0f3f6e1a037", "reading_canonical_log_prob": -107.26521301269531, "reading_distances": [11, 14, 10, 10], "reading_log_probs": [-114.41441345214844, -109.13412475585938, -119.09066009521484, -111.15536499023438], "reading_logprob_std": 4.3450838571891595, "tokenization_path_logprob_std": 127.63642008471406, "trajectory_balance_rmse": 1725.5643674279897} {"canonical_length": 68, "canonical_nll": 0.4400007023530848, "generation_rouge_l_f1": 0.30837004405286345, "generation_token_f1": 0.36123348017621143, "prompt_id": "42b3ed92d904a820", "reading_canonical_log_prob": -29.920047760009766, "reading_distances": [10, 11, 11, 14], "reading_log_probs": [-31.037084579467773, -43.530174255371094, -35.72032165527344, -33.39004898071289], "reading_logprob_std": 5.422119431786703, "tokenization_path_logprob_std": 65.66804763766024, "trajectory_balance_rmse": 1628.0185052602994} {"canonical_length": 4, "canonical_nll": 0.36969494819641113, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "a0febc02441ecdc9", "reading_canonical_log_prob": -1.4787797927856445, "reading_distances": [48, 47, 48, 41], "reading_log_probs": [-5.619123458862305, -4.997758388519287, -4.958317756652832, -3.5039782524108887], "reading_logprob_std": 0.8965041375235482, "tokenization_path_logprob_std": 33.514331198867374, "trajectory_balance_rmse": 138.34867090052407} {"canonical_length": 84, "canonical_nll": 2.334564208984375, "generation_rouge_l_f1": 0.1408450704225352, "generation_token_f1": 0.21971830985915491, "prompt_id": "95d007f697f6007d", "reading_canonical_log_prob": -196.1033935546875, "reading_distances": [28, 23, 21, 24], "reading_log_probs": [-213.8029022216797, -212.62680053710938, -224.14942932128906, -214.28164672851562], "reading_logprob_std": 5.334991860441201, "tokenization_path_logprob_std": 201.1428242353179, "trajectory_balance_rmse": 2354.999203031452} {"canonical_length": 178, "canonical_nll": 1.209368073538448, "generation_rouge_l_f1": 0.27505827505827507, "generation_token_f1": 0.4801864801864802, "prompt_id": "42731119470564d3", "reading_canonical_log_prob": -215.26751708984375, "reading_distances": [57, 55, 51, 60], "reading_log_probs": [-235.86602783203125, -237.28094482421875, -236.22239685058594, -223.54949951171875], "reading_logprob_std": 6.481391954288278, "tokenization_path_logprob_std": 192.3365254240227, "trajectory_balance_rmse": 3379.9316210233005} {"canonical_length": 302, "canonical_nll": 1.3018300643819847, "generation_rouge_l_f1": 0.25615763546798026, "generation_token_f1": 0.46962233169129725, "prompt_id": "f73070b2f6b0ff25", "reading_canonical_log_prob": -393.1526794433594, "reading_distances": [25, 27, 22, 27], "reading_log_probs": [-387.8410339355469, -385.9972229003906, -385.8952941894531, -385.8882751464844], "reading_logprob_std": 0.9583456893397326, "tokenization_path_logprob_std": 148.34189978742793, "trajectory_balance_rmse": 8441.466231567514} {"canonical_length": 143, "canonical_nll": 1.4062453049879808, "generation_rouge_l_f1": 0.28482972136222906, "generation_token_f1": 0.45820433436532504, "prompt_id": "08c2f6d9de2bf434", "reading_canonical_log_prob": -201.09307861328125, "reading_distances": [21, 22, 20, 18], "reading_log_probs": [-206.76168823242188, -203.77630615234375, -205.59149169921875, -207.6619415283203], "reading_logprob_std": 1.6776011004456957, "tokenization_path_logprob_std": 250.87967906586874, "trajectory_balance_rmse": 4025.6083405581976} {"canonical_length": 118, "canonical_nll": 1.1415053545418432, "generation_rouge_l_f1": 0.3510971786833856, "generation_token_f1": 0.4827586206896552, "prompt_id": "5f825922a6155ea5", "reading_canonical_log_prob": -134.6976318359375, "reading_distances": [37, 29, 34, 36], "reading_log_probs": [-134.1151580810547, -135.3520965576172, -135.3396453857422, -136.94659423828125], "reading_logprob_std": 1.1608645903877093, "tokenization_path_logprob_std": 258.758067117317, "trajectory_balance_rmse": 3239.1397831400004} {"canonical_length": 55, "canonical_nll": 1.4887989390980114, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.42391304347826086, "prompt_id": "3828b93ba3b0437a", "reading_canonical_log_prob": -81.88394165039062, "reading_distances": [22, 30, 26, 26], "reading_log_probs": [-76.15713500976562, -78.03461456298828, -75.9384994506836, -77.74967956542969], "reading_logprob_std": 1.07487166160884, "tokenization_path_logprob_std": 135.43972401919294, "trajectory_balance_rmse": 1417.2179173771085} {"canonical_length": 109, "canonical_nll": 2.162608155416786, "generation_rouge_l_f1": 0.2906574394463668, "generation_token_f1": 0.47058823529411764, "prompt_id": "bf57ac1eadcdbbef", "reading_canonical_log_prob": -235.7242889404297, "reading_distances": [36, 45, 38, 37], "reading_log_probs": [-233.0312042236328, -234.930908203125, -237.1479034423828, -233.26832580566406], "reading_logprob_std": 1.900497414507303, "tokenization_path_logprob_std": 162.63830996746196, "trajectory_balance_rmse": 3181.3088943961848} {"canonical_length": 15, "canonical_nll": 3.0285873413085938, "generation_rouge_l_f1": 0.1904761904761905, "generation_token_f1": 0.1904761904761905, "prompt_id": "fd15860ce78ef664", "reading_canonical_log_prob": -45.428810119628906, "reading_distances": [34, 33, 22, 31], "reading_log_probs": [-45.680747985839844, -45.665977478027344, -44.557003021240234, -46.47654724121094], "reading_logprob_std": 0.7888712964012734, "tokenization_path_logprob_std": 67.12976123644829, "trajectory_balance_rmse": 338.48442991682253} {"canonical_length": 25, "canonical_nll": 1.8851051330566406, "generation_rouge_l_f1": 0.13953488372093023, "generation_token_f1": 0.13953488372093023, "prompt_id": "b2df929cec29e4f9", "reading_canonical_log_prob": -47.127628326416016, "reading_distances": [31, 39, 30, 29], "reading_log_probs": [-64.44588470458984, -67.75164794921875, -59.88422393798828, -60.13263702392578], "reading_logprob_std": 3.7677172824683653, "tokenization_path_logprob_std": 32.289127495903365, "trajectory_balance_rmse": 794.539698952545} {"canonical_length": 280, "canonical_nll": 0.9155455453055246, "generation_rouge_l_f1": 0.4115384615384616, "generation_token_f1": 0.5846153846153846, "prompt_id": "190309fcf6be3c92", "reading_canonical_log_prob": -256.3527526855469, "reading_distances": [24, 29, 28, 27], "reading_log_probs": [-260.7547302246094, -260.2995300292969, -262.48870849609375, -260.8603515625], "reading_logprob_std": 0.9567136974031009, "tokenization_path_logprob_std": 266.04997779696384, "trajectory_balance_rmse": 5314.4527662786895} {"canonical_length": 20, "canonical_nll": 1.0578950881958007, "generation_rouge_l_f1": 0.12962962962962965, "generation_token_f1": 0.1388888888888889, "prompt_id": "48be81c0aa95d468", "reading_canonical_log_prob": -21.157901763916016, "reading_distances": [14, 23, 18, 23], "reading_log_probs": [-20.482437133789062, -20.93424415588379, -19.547203063964844, -20.102428436279297], "reading_logprob_std": 0.5878828225698063, "tokenization_path_logprob_std": 78.55502678774805, "trajectory_balance_rmse": 676.5064409225668} {"canonical_length": 86, "canonical_nll": 2.4418528357217477, "generation_rouge_l_f1": 0.24000000000000002, "generation_token_f1": 0.3644444444444444, "prompt_id": "87fbf3fe9c551bf6", "reading_canonical_log_prob": -209.9993438720703, "reading_distances": [32, 30, 29, 27], "reading_log_probs": [-212.14886474609375, -208.710693359375, -209.18658447265625, -211.97943115234375], "reading_logprob_std": 1.8105235507228128, "tokenization_path_logprob_std": 112.52370355901675, "trajectory_balance_rmse": 2281.6837720455096} {"canonical_length": 122, "canonical_nll": 1.3570350271756533, "generation_rouge_l_f1": 0.1568627450980392, "generation_token_f1": 0.2969187675070028, "prompt_id": "31fa13f44ebbb510", "reading_canonical_log_prob": -165.5582733154297, "reading_distances": [22, 22, 21, 22], "reading_log_probs": [-166.50567626953125, -157.0506591796875, -163.28836059570312, -164.30569458007812], "reading_logprob_std": 4.053473960693832, "tokenization_path_logprob_std": 123.59414611508582, "trajectory_balance_rmse": 2763.029585290881} {"canonical_length": 113, "canonical_nll": 1.5083331892975664, "generation_rouge_l_f1": 0.2689075630252101, "generation_token_f1": 0.453781512605042, "prompt_id": "1e48963a206d1f06", "reading_canonical_log_prob": -170.441650390625, "reading_distances": [13, 13, 15, 13], "reading_log_probs": [-170.8324432373047, -166.81976318359375, -162.94302368164062, -161.83847045898438], "reading_logprob_std": 4.085528450843714, "tokenization_path_logprob_std": 295.261846399541, "trajectory_balance_rmse": 2981.0223915930674} {"canonical_length": 7, "canonical_nll": 2.753434589930943, "generation_rouge_l_f1": 0.7692307692307692, "generation_token_f1": 0.7692307692307692, "prompt_id": "b92fcbf5058fa547", "reading_canonical_log_prob": -19.2740421295166, "reading_distances": [21, 17, 20, 17], "reading_log_probs": [-23.659875869750977, -21.258583068847656, -23.76401710510254, -22.52039909362793], "reading_logprob_std": 1.1722992004594797, "tokenization_path_logprob_std": 36.155642747152825, "trajectory_balance_rmse": 161.773855173417} {"canonical_length": 14, "canonical_nll": 1.0026392255510603, "generation_rouge_l_f1": 0.380952380952381, "generation_token_f1": 0.4761904761904762, "prompt_id": "5f82c19500742c50", "reading_canonical_log_prob": -14.036949157714844, "reading_distances": [28, 27, 25, 23], "reading_log_probs": [-13.282378196716309, -13.330039978027344, -14.77069091796875, -13.967105865478516], "reading_logprob_std": 0.6960169654482721, "tokenization_path_logprob_std": 57.32795712501046, "trajectory_balance_rmse": 519.0116252940211} {"canonical_length": 107, "canonical_nll": 1.4607934684396904, "generation_rouge_l_f1": 0.3004291845493562, "generation_token_f1": 0.5150214592274679, "prompt_id": "cb0faecefbcafa87", "reading_canonical_log_prob": -156.30490112304688, "reading_distances": [26, 27, 26, 28], "reading_log_probs": [-157.45639038085938, -158.04608154296875, -159.98260498046875, -157.05496215820312], "reading_logprob_std": 1.297246238560194, "tokenization_path_logprob_std": 210.52004442078157, "trajectory_balance_rmse": 3155.2710254895746} {"canonical_length": 111, "canonical_nll": 1.8849226118208051, "generation_rouge_l_f1": 0.14516129032258066, "generation_token_f1": 0.19354838709677416, "prompt_id": "935eaf327e8c6a5b", "reading_canonical_log_prob": -209.22640991210938, "reading_distances": [36, 33, 34, 33], "reading_log_probs": [-200.48452758789062, -201.63877868652344, -208.5606689453125, -207.7994384765625], "reading_logprob_std": 4.148393682728712, "tokenization_path_logprob_std": 120.59072299303638, "trajectory_balance_rmse": 2979.2996004978936} {"canonical_length": 72, "canonical_nll": 1.2709159851074219, "generation_rouge_l_f1": 0.11875000000000001, "generation_token_f1": 0.2, "prompt_id": "116cdf21f0043438", "reading_canonical_log_prob": -91.50595092773438, "reading_distances": [12, 18, 18, 16], "reading_log_probs": [-93.74942016601562, -90.1851806640625, -89.56988525390625, -89.71281433105469], "reading_logprob_std": 1.9809225353160997, "tokenization_path_logprob_std": 319.8386670437834, "trajectory_balance_rmse": 2278.4150229998863} {"canonical_length": 31, "canonical_nll": 1.532438216670867, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.3125, "prompt_id": "da79267701ae23c7", "reading_canonical_log_prob": -47.505584716796875, "reading_distances": [33, 35, 40, 31], "reading_log_probs": [-52.04666519165039, -49.710693359375, -48.0345344543457, -50.83568572998047], "reading_logprob_std": 1.706409331725912, "tokenization_path_logprob_std": 60.10444957809057, "trajectory_balance_rmse": 1175.5530343160192} {"canonical_length": 269, "canonical_nll": 0.8741105657527881, "generation_rouge_l_f1": 0.3208191126279863, "generation_token_f1": 0.5392491467576791, "prompt_id": "3348c8386d3ef99b", "reading_canonical_log_prob": -235.1357421875, "reading_distances": [15, 17, 11, 12], "reading_log_probs": [-230.80093383789062, -234.5707244873047, -235.15335083007812, -233.0579071044922], "reading_logprob_std": 1.9422439307104082, "tokenization_path_logprob_std": 222.42613083475237, "trajectory_balance_rmse": 6167.849330758589} {"canonical_length": 10, "canonical_nll": 0.05975120067596436, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "d04f01367f709567", "reading_canonical_log_prob": -0.5975120067596436, "reading_distances": [20, 27, 14, 25], "reading_log_probs": [-0.6308605670928955, -0.5174828171730042, -0.770123302936554, -0.528937816619873], "reading_logprob_std": 0.11717712436895411, "tokenization_path_logprob_std": 50.10952349522876, "trajectory_balance_rmse": 220.14375690276566} {"canonical_length": 281, "canonical_nll": 1.0106876685526023, "generation_rouge_l_f1": 0.2705882352941177, "generation_token_f1": 0.3803921568627451, "prompt_id": "f04ba139f8e40a51", "reading_canonical_log_prob": -284.00323486328125, "reading_distances": [14, 12, 15, 9], "reading_log_probs": [-286.27618408203125, -298.1314697265625, -285.54620361328125, -287.0662536621094], "reading_logprob_std": 5.950093563589492, "tokenization_path_logprob_std": 214.08230924641347, "trajectory_balance_rmse": 5343.697538065344} {"canonical_length": 501, "canonical_nll": 0.8672809410475923, "generation_rouge_l_f1": 0.3617647058823529, "generation_token_f1": 0.46470588235294125, "prompt_id": "16085c364de7d67f", "reading_canonical_log_prob": -434.50775146484375, "reading_distances": [14, 13, 17, 15], "reading_log_probs": [-432.1387634277344, -435.1512451171875, -429.55438232421875, -430.38897705078125], "reading_logprob_std": 2.4751505442381068, "tokenization_path_logprob_std": 143.25484406434705, "trajectory_balance_rmse": 8654.527722559294} {"canonical_length": 405, "canonical_nll": 1.2162778501157407, "generation_rouge_l_f1": 0.20933977455716588, "generation_token_f1": 0.3768115942028986, "prompt_id": "122921a791ee9365", "reading_canonical_log_prob": -492.592529296875, "reading_distances": [24, 22, 23, 21], "reading_log_probs": [-490.5439758300781, -490.1861877441406, -490.30267333984375, -490.9731140136719], "reading_logprob_std": 0.3479357377996289, "tokenization_path_logprob_std": 153.48188369677527, "trajectory_balance_rmse": 9277.518912228828} {"canonical_length": 67, "canonical_nll": 1.7560881429643773, "generation_rouge_l_f1": 0.3111111111111111, "generation_token_f1": 0.3555555555555555, "prompt_id": "5090b423053cf048", "reading_canonical_log_prob": -117.65790557861328, "reading_distances": [33, 38, 36, 33], "reading_log_probs": [-118.4838638305664, -114.53350830078125, -113.76005554199219, -116.35078430175781], "reading_logprob_std": 2.1032177446238864, "tokenization_path_logprob_std": 154.48468056632444, "trajectory_balance_rmse": 1941.4657425717307} {"canonical_length": 349, "canonical_nll": 1.1239824267718033, "generation_rouge_l_f1": 0.24761904761904763, "generation_token_f1": 0.4666666666666667, "prompt_id": "4811a788bad2e4f1", "reading_canonical_log_prob": -392.2698669433594, "reading_distances": [31, 24, 29, 22], "reading_log_probs": [-393.11102294921875, -394.7322692871094, -395.2825927734375, -396.27490234375], "reading_logprob_std": 1.3235538958645634, "tokenization_path_logprob_std": 155.22730014538786, "trajectory_balance_rmse": 10461.433296522702} {"canonical_length": 14, "canonical_nll": 0.6735469954354423, "generation_rouge_l_f1": 0.02158273381294964, "generation_token_f1": 0.04316546762589928, "prompt_id": "b2fd6c1ae0104b61", "reading_canonical_log_prob": -9.429657936096191, "reading_distances": [24, 18, 27, 20], "reading_log_probs": [-13.413192749023438, -11.708380699157715, -13.455894470214844, -11.837981224060059], "reading_logprob_std": 0.9608044745121105, "tokenization_path_logprob_std": 29.882612655729517, "trajectory_balance_rmse": 393.41025872301344} {"canonical_length": 12, "canonical_nll": 0.3251016139984131, "generation_rouge_l_f1": 0.05405405405405405, "generation_token_f1": 0.06177606177606177, "prompt_id": "3b02f9d0f5b11977", "reading_canonical_log_prob": -3.901219367980957, "reading_distances": [26, 20, 18, 19], "reading_log_probs": [-5.671591281890869, -4.611889362335205, -5.600390434265137, -4.098993301391602], "reading_logprob_std": 0.7689544967260155, "tokenization_path_logprob_std": 29.739587511897437, "trajectory_balance_rmse": 294.31934210134307} {"canonical_length": 6, "canonical_nll": 0.6972141265869141, "generation_rouge_l_f1": 0.13114754098360656, "generation_token_f1": 0.13114754098360656, "prompt_id": "842d7e43c22f686e", "reading_canonical_log_prob": -4.183284759521484, "reading_distances": [23, 27, 22, 26], "reading_log_probs": [-3.7523033618927, -5.224377155303955, -4.433011054992676, -4.480452060699463], "reading_logprob_std": 0.6015604877036009, "tokenization_path_logprob_std": 26.790577887241763, "trajectory_balance_rmse": 201.43025691505838} {"canonical_length": 74, "canonical_nll": 2.045186017010663, "generation_rouge_l_f1": 0.19642857142857142, "generation_token_f1": 0.3571428571428571, "prompt_id": "f726d0133e785ebc", "reading_canonical_log_prob": -151.34376525878906, "reading_distances": [22, 22, 20, 22], "reading_log_probs": [-153.86978149414062, -151.11807250976562, -151.9213409423828, -150.9156036376953], "reading_logprob_std": 1.347628505652707, "tokenization_path_logprob_std": 129.73338269416885, "trajectory_balance_rmse": 1760.9884075081968} {"canonical_length": 48, "canonical_nll": 1.340798060099284, "generation_rouge_l_f1": 0.32653061224489793, "generation_token_f1": 0.47619047619047605, "prompt_id": "cb06f10a99b5f8fb", "reading_canonical_log_prob": -64.35830688476562, "reading_distances": [21, 19, 24, 17], "reading_log_probs": [-63.80813217163086, -63.50506591796875, -64.05479431152344, -66.09803009033203], "reading_logprob_std": 1.1760385763613548, "tokenization_path_logprob_std": 36.662117733337226, "trajectory_balance_rmse": 1331.899047190987} {"canonical_length": 171, "canonical_nll": 1.0712578310603984, "generation_rouge_l_f1": 0.2776203966005666, "generation_token_f1": 0.45892351274787535, "prompt_id": "443ce871037b41c0", "reading_canonical_log_prob": -183.18508911132812, "reading_distances": [30, 36, 30, 28], "reading_log_probs": [-194.58749389648438, -191.11660766601562, -192.53402709960938, -193.83172607421875], "reading_logprob_std": 1.5247931707254783, "tokenization_path_logprob_std": 43.521377371551615, "trajectory_balance_rmse": 3272.820573430198} {"canonical_length": 307, "canonical_nll": 1.4214934805705415, "generation_rouge_l_f1": 0.28716216216216217, "generation_token_f1": 0.5675675675675675, "prompt_id": "d1c3cdcce82ccde0", "reading_canonical_log_prob": -436.39849853515625, "reading_distances": [44, 41, 34, 36], "reading_log_probs": [-440.6087341308594, -442.419677734375, -442.55938720703125, -441.65130615234375], "reading_logprob_std": 0.8947087703726293, "tokenization_path_logprob_std": 376.68925051970456, "trajectory_balance_rmse": 9023.695796913973} {"canonical_length": 61, "canonical_nll": 0.5509449693023182, "generation_rouge_l_f1": 0.4264705882352941, "generation_token_f1": 0.5147058823529411, "prompt_id": "66bf9508c6f9f94b", "reading_canonical_log_prob": -33.607643127441406, "reading_distances": [41, 46, 45, 47], "reading_log_probs": [-37.22161865234375, -33.990909576416016, -39.09477615356445, -38.73661804199219], "reading_logprob_std": 2.3263151015206365, "tokenization_path_logprob_std": 205.7859168234189, "trajectory_balance_rmse": 1713.2545969579844} {"canonical_length": 265, "canonical_nll": 0.9857843363060141, "generation_rouge_l_f1": 0.23461538461538461, "generation_token_f1": 0.4769230769230769, "prompt_id": "c9e81615c39cdb45", "reading_canonical_log_prob": -261.23284912109375, "reading_distances": [19, 27, 23, 24], "reading_log_probs": [-255.609375, -257.7939147949219, -257.44317626953125, -257.17266845703125], "reading_logprob_std": 0.9644111608769228, "tokenization_path_logprob_std": 510.7972178457482, "trajectory_balance_rmse": 7569.279110057874} {"canonical_length": 41, "canonical_nll": 1.7239331501286204, "generation_rouge_l_f1": 0.05921052631578947, "generation_token_f1": 0.1118421052631579, "prompt_id": "768e651293911174", "reading_canonical_log_prob": -70.68125915527344, "reading_distances": [55, 54, 52, 61], "reading_log_probs": [-79.85760498046875, -76.70103454589844, -78.1954345703125, -78.67060852050781], "reading_logprob_std": 1.3062015966832272, "tokenization_path_logprob_std": 101.47340031620251, "trajectory_balance_rmse": 1098.892728725572} {"canonical_length": 367, "canonical_nll": 1.1766630167532357, "generation_rouge_l_f1": 0.1605136436597111, "generation_token_f1": 0.27929373996789725, "prompt_id": "bd6b470fc22dea74", "reading_canonical_log_prob": -431.8353271484375, "reading_distances": [48, 35, 46, 46], "reading_log_probs": [-430.767822265625, -429.8377380371094, -436.77001953125, -434.927734375], "reading_logprob_std": 3.311028307993582, "tokenization_path_logprob_std": 370.09067870157355, "trajectory_balance_rmse": 10213.193018823704} {"canonical_length": 303, "canonical_nll": 1.3233760418278155, "generation_rouge_l_f1": 0.2371134020618557, "generation_token_f1": 0.43986254295532645, "prompt_id": "ca8e3b0a9ea7294b", "reading_canonical_log_prob": -400.9829406738281, "reading_distances": [24, 17, 25, 21], "reading_log_probs": [-409.6418151855469, -406.0638427734375, -409.7685546875, -412.168701171875], "reading_logprob_std": 2.5159158678418305, "tokenization_path_logprob_std": 220.87991559165147, "trajectory_balance_rmse": 8644.266852038007} {"canonical_length": 246, "canonical_nll": 1.545191198829713, "generation_rouge_l_f1": 0.2992424242424242, "generation_token_f1": 0.6287878787878788, "prompt_id": "5cd1f02fb9a1877d", "reading_canonical_log_prob": -380.1170349121094, "reading_distances": [25, 17, 30, 29], "reading_log_probs": [-382.53094482421875, -381.871337890625, -388.20037841796875, -383.3780517578125], "reading_logprob_std": 2.870495367082673, "tokenization_path_logprob_std": 256.73642536470044, "trajectory_balance_rmse": 6551.892259240629} {"canonical_length": 210, "canonical_nll": 0.9385657900855655, "generation_rouge_l_f1": 0.23043478260869565, "generation_token_f1": 0.4565217391304348, "prompt_id": "754987e5b5772585", "reading_canonical_log_prob": -197.09881591796875, "reading_distances": [27, 24, 27, 33], "reading_log_probs": [-192.27603149414062, -195.72402954101562, -194.349609375, -192.18292236328125], "reading_logprob_std": 1.715616714022383, "tokenization_path_logprob_std": 347.54662125714015, "trajectory_balance_rmse": 6616.9996185546115} {"canonical_length": 221, "canonical_nll": 1.2569948774657098, "generation_rouge_l_f1": 0.1735357917570499, "generation_token_f1": 0.38177874186550975, "prompt_id": "5aac80ddf699bdd3", "reading_canonical_log_prob": -277.7958679199219, "reading_distances": [20, 20, 20, 20], "reading_log_probs": [-279.0480041503906, -276.8165283203125, -279.623291015625, -280.75048828125], "reading_logprob_std": 1.6541216967284027, "tokenization_path_logprob_std": 298.21086053363496, "trajectory_balance_rmse": 5616.550277625051} {"canonical_length": 25, "canonical_nll": 1.487811279296875, "generation_rouge_l_f1": 0.11494252873563218, "generation_token_f1": 0.13793103448275862, "prompt_id": "f552b8f246a9b742", "reading_canonical_log_prob": -37.195281982421875, "reading_distances": [23, 22, 27, 26], "reading_log_probs": [-41.13569259643555, -39.08750915527344, -44.36176300048828, -42.729705810546875], "reading_logprob_std": 2.252598953764392, "tokenization_path_logprob_std": 69.9878650586512, "trajectory_balance_rmse": 660.6801965335376} {"canonical_length": 212, "canonical_nll": 1.3743620098761793, "generation_rouge_l_f1": 0.23093681917211328, "generation_token_f1": 0.44008714596949894, "prompt_id": "1dcab7a046130541", "reading_canonical_log_prob": -291.36474609375, "reading_distances": [20, 18, 19, 26], "reading_log_probs": [-291.74285888671875, -292.80816650390625, -291.2510986328125, -292.9619140625], "reading_logprob_std": 0.8285424998947192, "tokenization_path_logprob_std": 148.4583570023922, "trajectory_balance_rmse": 5517.341857187225} {"canonical_length": 21, "canonical_nll": 2.0142760503859747, "generation_rouge_l_f1": 0.2790697674418604, "generation_token_f1": 0.3255813953488372, "prompt_id": "721929550e55e7e5", "reading_canonical_log_prob": -42.29979705810547, "reading_distances": [24, 25, 26, 27], "reading_log_probs": [-45.999427795410156, -50.92213821411133, -45.21250915527344, -44.9144287109375], "reading_logprob_std": 2.8108544258706116, "tokenization_path_logprob_std": 26.34273936556568, "trajectory_balance_rmse": 434.97203157223703} {"canonical_length": 106, "canonical_nll": 2.421192097214033, "generation_rouge_l_f1": 0.28205128205128205, "generation_token_f1": 0.4615384615384615, "prompt_id": "a21c76c0694212ca", "reading_canonical_log_prob": -256.6463623046875, "reading_distances": [34, 36, 35, 33], "reading_log_probs": [-256.47344970703125, -258.0689697265625, -258.90032958984375, -255.84024047851562], "reading_logprob_std": 1.4100509056427102, "tokenization_path_logprob_std": 104.21373001840523, "trajectory_balance_rmse": 2344.8903292753503} {"canonical_length": 10, "canonical_nll": 1.2367225646972657, "generation_rouge_l_f1": 0.09815950920245399, "generation_token_f1": 0.09815950920245399, "prompt_id": "7cb87ccdae4ac547", "reading_canonical_log_prob": -12.367225646972656, "reading_distances": [19, 12, 18, 15], "reading_log_probs": [-12.559120178222656, -15.980419158935547, -12.422323226928711, -12.309158325195312], "reading_logprob_std": 1.778048792801213, "tokenization_path_logprob_std": 24.73557319300319, "trajectory_balance_rmse": 319.87210328695795} {"canonical_length": 263, "canonical_nll": 1.0173935110578065, "generation_rouge_l_f1": 0.3648068669527897, "generation_token_f1": 0.575107296137339, "prompt_id": "02274883d489f1d7", "reading_canonical_log_prob": -267.5744934082031, "reading_distances": [52, 51, 65, 54], "reading_log_probs": [-262.90557861328125, -262.6676940917969, -259.9671630859375, -260.28887939453125], "reading_logprob_std": 1.5436190732348645, "tokenization_path_logprob_std": 325.82710625441325, "trajectory_balance_rmse": 5990.902942036937} {"canonical_length": 219, "canonical_nll": 1.2259557715289668, "generation_rouge_l_f1": 0.21338912133891214, "generation_token_f1": 0.4058577405857741, "prompt_id": "dd480c5be4f865e4", "reading_canonical_log_prob": -268.48431396484375, "reading_distances": [25, 18, 27, 22], "reading_log_probs": [-268.256591796875, -270.73553466796875, -267.1639099121094, -266.6883544921875], "reading_logprob_std": 1.80648838525135, "tokenization_path_logprob_std": 400.0272831525805, "trajectory_balance_rmse": 6605.3807684956955} {"canonical_length": 17, "canonical_nll": 0.6051837135763729, "generation_rouge_l_f1": 0.1176470588235294, "generation_token_f1": 0.1176470588235294, "prompt_id": "48b6b6146fd695d3", "reading_canonical_log_prob": -10.28812313079834, "reading_distances": [17, 19, 18, 17], "reading_log_probs": [-15.552363395690918, -13.188682556152344, -18.513221740722656, -11.810644149780273], "reading_logprob_std": 2.937237731637047, "tokenization_path_logprob_std": 24.417586236227066, "trajectory_balance_rmse": 291.4404213610809} {"canonical_length": 230, "canonical_nll": 2.2554740574048915, "generation_rouge_l_f1": 0.1991701244813278, "generation_token_f1": 0.4149377593360996, "prompt_id": "619b20aa15b49441", "reading_canonical_log_prob": -518.759033203125, "reading_distances": [26, 22, 19, 24], "reading_log_probs": [-520.6077880859375, -520.47119140625, -520.1713256835938, -518.0299682617188], "reading_logprob_std": 1.2072425408389311, "tokenization_path_logprob_std": 151.933591394849, "trajectory_balance_rmse": 8023.764813832432} {"canonical_length": 264, "canonical_nll": 1.3180477257930872, "generation_rouge_l_f1": 0.25757575757575757, "generation_token_f1": 0.4848484848484848, "prompt_id": "49784b3e9fd7ae19", "reading_canonical_log_prob": -347.964599609375, "reading_distances": [38, 31, 38, 35], "reading_log_probs": [-354.1139221191406, -352.3510437011719, -353.270263671875, -357.10284423828125], "reading_logprob_std": 2.058850869285218, "tokenization_path_logprob_std": 360.2175539182797, "trajectory_balance_rmse": 8368.94594393542} {"canonical_length": 128, "canonical_nll": 1.3980557918548584, "generation_rouge_l_f1": 0.13471502590673573, "generation_token_f1": 0.2590673575129534, "prompt_id": "211ba51e9f80e5d0", "reading_canonical_log_prob": -178.95114135742188, "reading_distances": [11, 9, 14, 12], "reading_log_probs": [-176.9449462890625, -177.79653930664062, -179.48092651367188, -177.6060333251953], "reading_logprob_std": 1.0794334816974323, "tokenization_path_logprob_std": 150.47233110941696, "trajectory_balance_rmse": 3615.6671759543833} {"canonical_length": 19, "canonical_nll": 0.4740691937898335, "generation_rouge_l_f1": 0.7142857142857143, "generation_token_f1": 0.7142857142857143, "prompt_id": "c295bb5aa04e8e4c", "reading_canonical_log_prob": -9.007314682006836, "reading_distances": [23, 23, 25, 18], "reading_log_probs": [-11.53062629699707, -22.89398956298828, -14.64625358581543, -10.388028144836426], "reading_logprob_std": 5.647236730640069, "tokenization_path_logprob_std": 112.3570121468692, "trajectory_balance_rmse": 798.8100868982754} {"canonical_length": 163, "canonical_nll": 0.6318337844193347, "generation_rouge_l_f1": 0.4189944134078212, "generation_token_f1": 0.5027932960893855, "prompt_id": "55b91ee2bfd58fff", "reading_canonical_log_prob": -102.98890686035156, "reading_distances": [39, 24, 36, 32], "reading_log_probs": [-106.60490417480469, -105.76242065429688, -105.87300109863281, -108.05647277832031], "reading_logprob_std": 1.0565259014159223, "tokenization_path_logprob_std": 183.46040794279506, "trajectory_balance_rmse": 2766.3530900636506} {"canonical_length": 79, "canonical_nll": 1.2292289251013646, "generation_rouge_l_f1": 0.4367816091954023, "generation_token_f1": 0.5632183908045978, "prompt_id": "aacdb99cf2030827", "reading_canonical_log_prob": -97.10908508300781, "reading_distances": [13, 29, 28, 24], "reading_log_probs": [-98.16339111328125, -100.0236587524414, -101.61431121826172, -98.5158462524414], "reading_logprob_std": 1.5784440870674323, "tokenization_path_logprob_std": 141.8015904661688, "trajectory_balance_rmse": 2538.0437160124156} {"canonical_length": 7, "canonical_nll": 0.04430235283715384, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "732d6c9c54c4fcf0", "reading_canonical_log_prob": -0.3101164698600769, "reading_distances": [19, 17, 23, 18], "reading_log_probs": [-0.2283443808555603, -0.15412890911102295, -0.17832720279693604, -0.3325609862804413], "reading_logprob_std": 0.07910034699746762, "tokenization_path_logprob_std": 71.73780017379775, "trajectory_balance_rmse": 256.8711066587533} {"canonical_length": 234, "canonical_nll": 0.6825397198016827, "generation_rouge_l_f1": 0.31672597864768687, "generation_token_f1": 0.49466192170818507, "prompt_id": "1e288eaee9c72d47", "reading_canonical_log_prob": -159.71429443359375, "reading_distances": [34, 34, 31, 32], "reading_log_probs": [-164.46755981445312, -170.37171936035156, -163.65219116210938, -167.26324462890625], "reading_logprob_std": 3.0440400931914473, "tokenization_path_logprob_std": 490.48085615185926, "trajectory_balance_rmse": 5318.817535870204} {"canonical_length": 85, "canonical_nll": 1.2537782556870405, "generation_rouge_l_f1": 0.4088397790055249, "generation_token_f1": 0.6187845303867404, "prompt_id": "34251c7523ed14e6", "reading_canonical_log_prob": -106.57115173339844, "reading_distances": [26, 25, 27, 25], "reading_log_probs": [-108.00682067871094, -108.4400405883789, -106.36723327636719, -110.97650909423828], "reading_logprob_std": 1.9076731425249307, "tokenization_path_logprob_std": 110.27267369811962, "trajectory_balance_rmse": 2288.20469776299} {"canonical_length": 62, "canonical_nll": 2.9586759997952368, "generation_rouge_l_f1": 0.13333333333333333, "generation_token_f1": 0.19682539682539682, "prompt_id": "915708e38c0c1bff", "reading_canonical_log_prob": -183.4379119873047, "reading_distances": [13, 14, 11, 14], "reading_log_probs": [-187.85635375976562, -193.26974487304688, -187.63131713867188, -184.56039428710938], "reading_logprob_std": 3.620474367818463, "tokenization_path_logprob_std": 100.16439113970158, "trajectory_balance_rmse": 1340.0733327040393} {"canonical_length": 145, "canonical_nll": 1.8699902764682113, "generation_rouge_l_f1": 0.2544529262086514, "generation_token_f1": 0.4681933842239185, "prompt_id": "fc76d6d236be83db", "reading_canonical_log_prob": -271.1485900878906, "reading_distances": [23, 27, 22, 25], "reading_log_probs": [-271.9643859863281, -269.39739990234375, -268.87109375, -269.2682189941406], "reading_logprob_std": 1.410631808712899, "tokenization_path_logprob_std": 323.40131768605414, "trajectory_balance_rmse": 3314.2321982509516} {"canonical_length": 391, "canonical_nll": 0.8043648410026375, "generation_rouge_l_f1": 0.3115264797507788, "generation_token_f1": 0.42990654205607476, "prompt_id": "cf72c2330315d4cd", "reading_canonical_log_prob": -314.50665283203125, "reading_distances": [23, 19, 28, 22], "reading_log_probs": [-307.79278564453125, -312.6431884765625, -309.7310791015625, -310.3440246582031], "reading_logprob_std": 1.9986338499595502, "tokenization_path_logprob_std": 168.2072651455893, "trajectory_balance_rmse": 5121.028048549924} {"canonical_length": 247, "canonical_nll": 1.27479015767333, "generation_rouge_l_f1": 0.297029702970297, "generation_token_f1": 0.5544554455445545, "prompt_id": "12725fe24fa84c62", "reading_canonical_log_prob": -314.8731689453125, "reading_distances": [34, 37, 32, 31], "reading_log_probs": [-302.25164794921875, -305.09228515625, -309.2184753417969, -308.76776123046875], "reading_logprob_std": 3.288912548037193, "tokenization_path_logprob_std": 413.52800914911137, "trajectory_balance_rmse": 7945.989555018052} {"canonical_length": 164, "canonical_nll": 1.7876482707698171, "generation_rouge_l_f1": 0.3050847457627119, "generation_token_f1": 0.5181598062953995, "prompt_id": "767d0dc7cec8c49a", "reading_canonical_log_prob": -293.17431640625, "reading_distances": [13, 12, 17, 13], "reading_log_probs": [-299.0050048828125, -302.05596923828125, -298.5015869140625, -300.5299072265625], "reading_logprob_std": 1.6063344029187832, "tokenization_path_logprob_std": 261.48651517218826, "trajectory_balance_rmse": 4815.980240012374} {"canonical_length": 211, "canonical_nll": 1.386150342028288, "generation_rouge_l_f1": 0.30626450116009285, "generation_token_f1": 0.4361948955916473, "prompt_id": "0305770999fedf01", "reading_canonical_log_prob": -292.47772216796875, "reading_distances": [33, 26, 28, 37], "reading_log_probs": [-287.11572265625, -285.40692138671875, -286.3768615722656, -290.1524658203125], "reading_logprob_std": 2.0494692381459694, "tokenization_path_logprob_std": 153.6136924538643, "trajectory_balance_rmse": 3721.8775132548617} {"canonical_length": 28, "canonical_nll": 0.45791629382542204, "generation_rouge_l_f1": 0.8135593220338982, "generation_token_f1": 0.8135593220338982, "prompt_id": "47deaa7fb6aebf76", "reading_canonical_log_prob": -12.821656227111816, "reading_distances": [17, 18, 14, 15], "reading_log_probs": [-13.27426528930664, -13.531761169433594, -13.320001602172852, -12.751127243041992], "reading_logprob_std": 0.3316522817504194, "tokenization_path_logprob_std": 51.35167956874512, "trajectory_balance_rmse": 1089.7689443469646} {"canonical_length": 55, "canonical_nll": 1.3459625244140625, "generation_rouge_l_f1": 0.23076923076923078, "generation_token_f1": 0.3384615384615385, "prompt_id": "83b035c8880eabe5", "reading_canonical_log_prob": -74.02793884277344, "reading_distances": [38, 38, 29, 28], "reading_log_probs": [-71.27692413330078, -68.24555969238281, -71.78904724121094, -72.55181121826172], "reading_logprob_std": 1.8876504935200609, "tokenization_path_logprob_std": 172.0095878745867, "trajectory_balance_rmse": 2148.405896291448} {"canonical_length": 235, "canonical_nll": 1.5528637175864362, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.4579256360078278, "prompt_id": "3f8f300bcd0a09f9", "reading_canonical_log_prob": -364.9229736328125, "reading_distances": [30, 24, 30, 33], "reading_log_probs": [-359.82763671875, -365.12420654296875, -354.9493713378906, -350.568603515625], "reading_logprob_std": 6.272724013126199, "tokenization_path_logprob_std": 561.3595490843221, "trajectory_balance_rmse": 6779.825197503703} {"canonical_length": 32, "canonical_nll": 1.1589419841766357, "generation_rouge_l_f1": 0.18543046357615892, "generation_token_f1": 0.2781456953642384, "prompt_id": "de8521c6b6a04f6f", "reading_canonical_log_prob": -37.086143493652344, "reading_distances": [20, 21, 24, 20], "reading_log_probs": [-38.001129150390625, -35.29716110229492, -35.17577362060547, -34.7840690612793], "reading_logprob_std": 1.4740811690910045, "tokenization_path_logprob_std": 106.28496261404322, "trajectory_balance_rmse": 799.4053564322109} {"canonical_length": 12, "canonical_nll": 0.6843885580698649, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "89d6bb48c1695f7e", "reading_canonical_log_prob": -8.212662696838379, "reading_distances": [23, 26, 34, 26], "reading_log_probs": [-9.377965927124023, -12.47899055480957, -10.955065727233887, -13.53903865814209], "reading_logprob_std": 1.815238477514451, "tokenization_path_logprob_std": 76.84908953241978, "trajectory_balance_rmse": 344.5623455676155} {"canonical_length": 325, "canonical_nll": 1.7180658428485578, "generation_rouge_l_f1": 0.21, "generation_token_f1": 0.2766666666666667, "prompt_id": "b1d79661f636c046", "reading_canonical_log_prob": -558.3713989257812, "reading_distances": [16, 19, 22, 14], "reading_log_probs": [-554.648193359375, -558.89111328125, -557.3584594726562, -559.6231689453125], "reading_logprob_std": 2.2006085547854797, "tokenization_path_logprob_std": 534.104727623852, "trajectory_balance_rmse": 9130.355745016463} {"canonical_length": 15, "canonical_nll": 1.2491703033447266, "generation_rouge_l_f1": 0.06872852233676975, "generation_token_f1": 0.08934707903780069, "prompt_id": "db698be75c6d5259", "reading_canonical_log_prob": -18.7375545501709, "reading_distances": [22, 19, 26, 22], "reading_log_probs": [-20.700626373291016, -21.075654983520508, -20.01498031616211, -17.82830238342285], "reading_logprob_std": 1.4523800041594068, "tokenization_path_logprob_std": 59.61553633134879, "trajectory_balance_rmse": 473.3907665939593} {"canonical_length": 33, "canonical_nll": 0.7940758213852391, "generation_rouge_l_f1": 0.24242424242424243, "generation_token_f1": 0.3151515151515151, "prompt_id": "4c3924ee604d1034", "reading_canonical_log_prob": -26.20450210571289, "reading_distances": [31, 23, 26, 19], "reading_log_probs": [-26.099979400634766, -25.199087142944336, -25.575109481811523, -25.556934356689453], "reading_logprob_std": 0.3710088694235331, "tokenization_path_logprob_std": 74.05157125945438, "trajectory_balance_rmse": 1057.5868254300722} {"canonical_length": 63, "canonical_nll": 1.4982047913566467, "generation_rouge_l_f1": 0.16909620991253643, "generation_token_f1": 0.2740524781341108, "prompt_id": "1fd45a1cb9cbeccc", "reading_canonical_log_prob": -94.38690185546875, "reading_distances": [16, 16, 20, 19], "reading_log_probs": [-98.50933837890625, -99.22831726074219, -103.11021423339844, -100.95414733886719], "reading_logprob_std": 2.0485425333281237, "tokenization_path_logprob_std": 146.7837809614035, "trajectory_balance_rmse": 1721.3290002322428} {"canonical_length": 205, "canonical_nll": 1.2124465570217227, "generation_rouge_l_f1": 0.3866348448687351, "generation_token_f1": 0.5966587112171838, "prompt_id": "326b7dac8c34a0d5", "reading_canonical_log_prob": -248.55154418945312, "reading_distances": [21, 21, 21, 12], "reading_log_probs": [-249.4193572998047, -249.67637634277344, -246.82467651367188, -249.60464477539062], "reading_logprob_std": 1.3753278241994105, "tokenization_path_logprob_std": 58.06186026253939, "trajectory_balance_rmse": 5456.386566590818} {"canonical_length": 322, "canonical_nll": 1.4703971910180513, "generation_rouge_l_f1": 0.22064056939501778, "generation_token_f1": 0.44128113879003555, "prompt_id": "1e2cec78b7ee8d07", "reading_canonical_log_prob": -473.4678955078125, "reading_distances": [15, 19, 18, 17], "reading_log_probs": [-475.1269226074219, -478.16851806640625, -471.09210205078125, -478.17681884765625], "reading_logprob_std": 3.3553674855581024, "tokenization_path_logprob_std": 369.3434383620071, "trajectory_balance_rmse": 8099.1046325039815} {"canonical_length": 360, "canonical_nll": 1.2487518310546875, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.47863247863247865, "prompt_id": "e3633ae650b9ffe5", "reading_canonical_log_prob": -449.5506591796875, "reading_distances": [22, 15, 20, 21], "reading_log_probs": [-446.2684631347656, -449.9292297363281, -443.522216796875, -447.2843017578125], "reading_logprob_std": 2.6484862223342716, "tokenization_path_logprob_std": 329.4704237371404, "trajectory_balance_rmse": 7787.001002576934} {"canonical_length": 199, "canonical_nll": 1.1366617020650125, "generation_rouge_l_f1": 0.26352941176470585, "generation_token_f1": 0.5505882352941175, "prompt_id": "e7d8d0860929d3e3", "reading_canonical_log_prob": -226.1956787109375, "reading_distances": [24, 30, 28, 25], "reading_log_probs": [-224.6353302001953, -224.90377807617188, -219.5519256591797, -227.01748657226562], "reading_logprob_std": 3.1679702047400737, "tokenization_path_logprob_std": 282.7297609003143, "trajectory_balance_rmse": 4753.221610263994} {"canonical_length": 58, "canonical_nll": 2.244786492709456, "generation_rouge_l_f1": 0.34108527131782945, "generation_token_f1": 0.4961240310077519, "prompt_id": "8c4c85aa12a6cb37", "reading_canonical_log_prob": -130.19761657714844, "reading_distances": [30, 24, 30, 29], "reading_log_probs": [-121.73562622070312, -120.370849609375, -120.50062561035156, -121.20297241210938], "reading_logprob_std": 0.6373201043827876, "tokenization_path_logprob_std": 130.01910620892062, "trajectory_balance_rmse": 1840.8580878026642} {"canonical_length": 35, "canonical_nll": 1.9345103672572546, "generation_rouge_l_f1": 0.3888888888888889, "generation_token_f1": 0.5277777777777778, "prompt_id": "5f2bc0710dc7dd88", "reading_canonical_log_prob": -67.7078628540039, "reading_distances": [13, 16, 14, 25], "reading_log_probs": [-72.04884338378906, -71.44895935058594, -71.91936492919922, -72.39537048339844], "reading_logprob_std": 0.39160601827486735, "tokenization_path_logprob_std": 58.73791683527287, "trajectory_balance_rmse": 922.4904339749427} {"canonical_length": 24, "canonical_nll": 1.2314472993214924, "generation_rouge_l_f1": 0.07719298245614034, "generation_token_f1": 0.08421052631578949, "prompt_id": "e06153205819c799", "reading_canonical_log_prob": -29.55473518371582, "reading_distances": [70, 71, 64, 77], "reading_log_probs": [-38.22832489013672, -24.671255111694336, -37.82597732543945, -29.948612213134766], "reading_logprob_std": 6.554004896267361, "tokenization_path_logprob_std": 100.912516745204, "trajectory_balance_rmse": 652.9382985675762} {"canonical_length": 81, "canonical_nll": 0.9117052054699556, "generation_rouge_l_f1": 0.24126984126984127, "generation_token_f1": 0.26031746031746034, "prompt_id": "788acaefe94c61a6", "reading_canonical_log_prob": -73.8481216430664, "reading_distances": [19, 19, 26, 26], "reading_log_probs": [-79.6629638671875, -78.97257995605469, -80.53522491455078, -82.2794418334961], "reading_logprob_std": 1.4289539772971558, "tokenization_path_logprob_std": 27.307298917961624, "trajectory_balance_rmse": 1292.7496102647924} {"canonical_length": 21, "canonical_nll": 2.11338624500093, "generation_rouge_l_f1": 0.21917808219178084, "generation_token_f1": 0.30136986301369867, "prompt_id": "24304bef6f934812", "reading_canonical_log_prob": -44.38111114501953, "reading_distances": [17, 17, 14, 14], "reading_log_probs": [-50.80864334106445, -49.53163528442383, -48.20603561401367, -45.56137466430664], "reading_logprob_std": 2.2444851031744326, "tokenization_path_logprob_std": 73.27469490451561, "trajectory_balance_rmse": 461.6408937758811} {"canonical_length": 315, "canonical_nll": 1.510045902312748, "generation_rouge_l_f1": 0.2381786339754816, "generation_token_f1": 0.49036777583187396, "prompt_id": "969a799cc7257588", "reading_canonical_log_prob": -475.6644592285156, "reading_distances": [20, 15, 16, 18], "reading_log_probs": [-468.5528259277344, -468.57781982421875, -467.06732177734375, -468.93975830078125], "reading_logprob_std": 0.8304459786631325, "tokenization_path_logprob_std": 345.07225788515836, "trajectory_balance_rmse": 8638.342764659641} {"canonical_length": 152, "canonical_nll": 1.4508396951775802, "generation_rouge_l_f1": 0.1309823677581864, "generation_token_f1": 0.2770780856423174, "prompt_id": "7f301b40ae5af8a5", "reading_canonical_log_prob": -220.5276336669922, "reading_distances": [23, 32, 31, 24], "reading_log_probs": [-228.56683349609375, -223.94398498535156, -221.73255920410156, -226.57528686523438], "reading_logprob_std": 2.990407667349286, "tokenization_path_logprob_std": 312.91326167747883, "trajectory_balance_rmse": 3756.9452238934505} {"canonical_length": 354, "canonical_nll": 1.463351147323005, "generation_rouge_l_f1": 0.19182389937106917, "generation_token_f1": 0.39937106918238996, "prompt_id": "d838a6ece3beb4b3", "reading_canonical_log_prob": -518.0263061523438, "reading_distances": [15, 19, 22, 17], "reading_log_probs": [-511.30230712890625, -511.04315185546875, -513.031494140625, -514.7492065429688], "reading_logprob_std": 1.7218509187005304, "tokenization_path_logprob_std": 254.88029031780607, "trajectory_balance_rmse": 10383.332343003429} {"canonical_length": 15, "canonical_nll": 2.9645350138346354, "generation_rouge_l_f1": 0.35714285714285715, "generation_token_f1": 0.42857142857142855, "prompt_id": "3ad7d2823c1e6bd8", "reading_canonical_log_prob": -44.46802520751953, "reading_distances": [40, 44, 31, 41], "reading_log_probs": [-48.84687805175781, -48.978919982910156, -50.20793151855469, -49.366905212402344], "reading_logprob_std": 0.612961186887197, "tokenization_path_logprob_std": 63.6987856558276, "trajectory_balance_rmse": 629.4018630535846} {"canonical_length": 8, "canonical_nll": 0.08127240836620331, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "7a4144ac9a54046a", "reading_canonical_log_prob": -0.6501792669296265, "reading_distances": [32, 28, 27, 30], "reading_log_probs": [-0.2928937077522278, -0.16035988926887512, -0.2020237147808075, -0.27668020129203796], "reading_logprob_std": 0.06253354218590419, "tokenization_path_logprob_std": 41.29522674562039, "trajectory_balance_rmse": 162.1953871029578} {"canonical_length": 34, "canonical_nll": 0.2893378874834846, "generation_rouge_l_f1": 0.17627118644067796, "generation_token_f1": 0.1830508474576271, "prompt_id": "c898f115e1329204", "reading_canonical_log_prob": -9.837488174438477, "reading_distances": [10, 7, 6, 7], "reading_log_probs": [-11.05525016784668, -10.009218215942383, -10.569112777709961, -10.447016716003418], "reading_logprob_std": 0.43016636642812117, "tokenization_path_logprob_std": 93.20044934489606, "trajectory_balance_rmse": 735.5933691504737} {"canonical_length": 104, "canonical_nll": 1.8650385049673228, "generation_rouge_l_f1": 0.19886363636363638, "generation_token_f1": 0.34659090909090906, "prompt_id": "5328de30202d5ac5", "reading_canonical_log_prob": -193.96400451660156, "reading_distances": [14, 12, 11, 14], "reading_log_probs": [-196.05020141601562, -196.555908203125, -197.44552612304688, -198.17196655273438], "reading_logprob_std": 0.9414234797479261, "tokenization_path_logprob_std": 72.01632428034937, "trajectory_balance_rmse": 3487.0662452726087} {"canonical_length": 35, "canonical_nll": 1.455383518763951, "generation_rouge_l_f1": 0.4285714285714286, "generation_token_f1": 0.4285714285714286, "prompt_id": "f9e71d1877230558", "reading_canonical_log_prob": -50.93842315673828, "reading_distances": [29, 22, 28, 26], "reading_log_probs": [-43.963523864746094, -44.68568420410156, -43.416839599609375, -43.335533142089844], "reading_logprob_std": 0.6227785086384432, "tokenization_path_logprob_std": 181.17914550648845, "trajectory_balance_rmse": 1076.5209306161648} {"canonical_length": 211, "canonical_nll": 1.6488974331679502, "generation_rouge_l_f1": 0.23340040241448695, "generation_token_f1": 0.45875251509054327, "prompt_id": "4df4b024265f29c3", "reading_canonical_log_prob": -347.9173583984375, "reading_distances": [27, 28, 31, 30], "reading_log_probs": [-350.94097900390625, -350.81951904296875, -351.4041748046875, -354.5592956542969], "reading_logprob_std": 1.7702179077836233, "tokenization_path_logprob_std": 155.3296158043707, "trajectory_balance_rmse": 6287.716517226756} {"canonical_length": 21, "canonical_nll": 0.3939139048258464, "generation_rouge_l_f1": 0.26804123711340205, "generation_token_f1": 0.26804123711340205, "prompt_id": "6e5cc5a69c7269eb", "reading_canonical_log_prob": -8.272192001342773, "reading_distances": [19, 26, 29, 24], "reading_log_probs": [-7.744494438171387, -11.495570182800293, -11.054658889770508, -15.04930305480957], "reading_logprob_std": 2.9884303148906115, "tokenization_path_logprob_std": 99.1593728876065, "trajectory_balance_rmse": 663.6872191935448} {"canonical_length": 81, "canonical_nll": 1.5297752662941262, "generation_rouge_l_f1": 0.30841121495327095, "generation_token_f1": 0.4766355140186916, "prompt_id": "15d75a153e79b8a7", "reading_canonical_log_prob": -123.91179656982422, "reading_distances": [37, 41, 41, 42], "reading_log_probs": [-123.60221862792969, -122.39775085449219, -122.81292724609375, -122.30148315429688], "reading_logprob_std": 0.5922309684809683, "tokenization_path_logprob_std": 177.08680667130406, "trajectory_balance_rmse": 2844.087503921434} {"canonical_length": 4, "canonical_nll": 2.8556697368621826, "generation_rouge_l_f1": 0.015037593984962407, "generation_token_f1": 0.015037593984962407, "prompt_id": "9acb0ff71abeda7d", "reading_canonical_log_prob": -11.42267894744873, "reading_distances": [21, 16, 14, 15], "reading_log_probs": [-17.144197463989258, -18.00872039794922, -15.517714500427246, -14.167996406555176], "reading_logprob_std": 1.70852347464257, "tokenization_path_logprob_std": 22.80947128244256, "trajectory_balance_rmse": 76.89183232542746} {"canonical_length": 114, "canonical_nll": 0.7073662741142407, "generation_rouge_l_f1": 0.2870813397129187, "generation_token_f1": 0.5550239234449761, "prompt_id": "1b111d554cbb9115", "reading_canonical_log_prob": -80.63975524902344, "reading_distances": [9, 12, 11, 10], "reading_log_probs": [-83.00433349609375, -81.51206970214844, -80.75267028808594, -80.53282165527344], "reading_logprob_std": 1.1176298150202955, "tokenization_path_logprob_std": 42.17475502400535, "trajectory_balance_rmse": 2105.095161607536} {"canonical_length": 135, "canonical_nll": 1.513404224537037, "generation_rouge_l_f1": 0.24937655860349128, "generation_token_f1": 0.40399002493765584, "prompt_id": "4c468581b163e3ed", "reading_canonical_log_prob": -204.3095703125, "reading_distances": [19, 23, 22, 28], "reading_log_probs": [-203.23770141601562, -205.33853149414062, -205.23687744140625, -205.80059814453125], "reading_logprob_std": 1.1372573589437194, "tokenization_path_logprob_std": 412.3169905568248, "trajectory_balance_rmse": 4083.1896480298187} {"canonical_length": 38, "canonical_nll": 0.8690820995130037, "generation_rouge_l_f1": 0.4807692307692308, "generation_token_f1": 0.5192307692307692, "prompt_id": "5a9339c4c26875b0", "reading_canonical_log_prob": -33.02511978149414, "reading_distances": [20, 17, 15, 19], "reading_log_probs": [-44.219688415527344, -45.93971252441406, -47.11412048339844, -48.59315872192383], "reading_logprob_std": 1.8500231381397416, "tokenization_path_logprob_std": 66.94653252553361, "trajectory_balance_rmse": 647.3626015494837} {"canonical_length": 245, "canonical_nll": 1.3615934410873725, "generation_rouge_l_f1": 0.27145708582834327, "generation_token_f1": 0.5269461077844311, "prompt_id": "dadb86ce514e9684", "reading_canonical_log_prob": -333.59039306640625, "reading_distances": [14, 16, 17, 17], "reading_log_probs": [-335.1749267578125, -333.9745788574219, -336.01983642578125, -339.1368408203125], "reading_logprob_std": 2.206037434662739, "tokenization_path_logprob_std": 130.16213844560428, "trajectory_balance_rmse": 7354.4600542752505} {"canonical_length": 153, "canonical_nll": 1.776415856055964, "generation_rouge_l_f1": 0.2478386167146974, "generation_token_f1": 0.4668587896253602, "prompt_id": "8be6dabc44180a81", "reading_canonical_log_prob": -271.7916259765625, "reading_distances": [21, 22, 22, 24], "reading_log_probs": [-268.5735168457031, -271.8257141113281, -281.3406677246094, -270.2955322265625], "reading_logprob_std": 5.711195749175368, "tokenization_path_logprob_std": 229.99366309309622, "trajectory_balance_rmse": 4717.441839238738} {"canonical_length": 8, "canonical_nll": 1.7314844131469727, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "5786bd96f280e551", "reading_canonical_log_prob": -13.851875305175781, "reading_distances": [14, 23, 19, 22], "reading_log_probs": [-16.76325225830078, -15.917428970336914, -15.07463264465332, -17.595247268676758], "reading_logprob_std": 1.0854318227867508, "tokenization_path_logprob_std": 38.99678453241652, "trajectory_balance_rmse": 130.33894945710813} {"canonical_length": 103, "canonical_nll": 1.3518553798638502, "generation_rouge_l_f1": 0.199288256227758, "generation_token_f1": 0.4056939501779359, "prompt_id": "edcbad1e6d6e267f", "reading_canonical_log_prob": -139.24110412597656, "reading_distances": [20, 28, 20, 24], "reading_log_probs": [-141.2638702392578, -143.23989868164062, -141.69566345214844, -143.68423461914062], "reading_logprob_std": 1.1720996001037778, "tokenization_path_logprob_std": 276.1551462121959, "trajectory_balance_rmse": 2900.344214784709} {"canonical_length": 15, "canonical_nll": 2.63779296875, "generation_rouge_l_f1": 0.22222222222222227, "generation_token_f1": 0.22222222222222227, "prompt_id": "f125519b3e50ff0b", "reading_canonical_log_prob": -39.56689453125, "reading_distances": [14, 14, 15, 14], "reading_log_probs": [-32.7816047668457, -36.12154006958008, -34.981651306152344, -31.38815689086914], "reading_logprob_std": 2.1321846075249153, "tokenization_path_logprob_std": 58.76239929245162, "trajectory_balance_rmse": 411.272271073111} {"canonical_length": 335, "canonical_nll": 0.7913842044659515, "generation_rouge_l_f1": 0.25597269624573377, "generation_token_f1": 0.48805460750853247, "prompt_id": "805fd7652d88dc11", "reading_canonical_log_prob": -265.11370849609375, "reading_distances": [24, 26, 23, 25], "reading_log_probs": [-274.7189025878906, -275.8734130859375, -267.07843017578125, -275.740966796875], "reading_logprob_std": 4.2146879633294425, "tokenization_path_logprob_std": 187.24300677045895, "trajectory_balance_rmse": 6528.457028383359} {"canonical_length": 58, "canonical_nll": 1.7356635784280712, "generation_rouge_l_f1": 0.37383177570093457, "generation_token_f1": 0.5420560747663552, "prompt_id": "7efa21552d311157", "reading_canonical_log_prob": -100.66848754882812, "reading_distances": [38, 36, 33, 36], "reading_log_probs": [-102.205322265625, -101.36039733886719, -101.87371063232422, -98.93699645996094], "reading_logprob_std": 1.479483875840079, "tokenization_path_logprob_std": 131.95466000665593, "trajectory_balance_rmse": 2011.6893847187303} {"canonical_length": 337, "canonical_nll": 1.2115039316058513, "generation_rouge_l_f1": 0.25944170771756975, "generation_token_f1": 0.4729064039408866, "prompt_id": "961872748d95ea69", "reading_canonical_log_prob": -408.2768249511719, "reading_distances": [27, 32, 27, 29], "reading_log_probs": [-416.88714599609375, -414.9058837890625, -415.2463073730469, -416.21917724609375], "reading_logprob_std": 0.9060462314163559, "tokenization_path_logprob_std": 496.1702139737182, "trajectory_balance_rmse": 10760.476649844653} {"canonical_length": 62, "canonical_nll": 0.884141368250693, "generation_rouge_l_f1": 0.27983539094650206, "generation_token_f1": 0.34567901234567905, "prompt_id": "e754a1aad6ea85da", "reading_canonical_log_prob": -54.81676483154297, "reading_distances": [6, 6, 9, 7], "reading_log_probs": [-54.07026290893555, -53.06831359863281, -52.653228759765625, -52.65709686279297], "reading_logprob_std": 0.6677284917173361, "tokenization_path_logprob_std": 157.9033070070835, "trajectory_balance_rmse": 1345.197513670558} {"canonical_length": 200, "canonical_nll": 2.0644380187988283, "generation_rouge_l_f1": 0.16913319238900634, "generation_token_f1": 0.3382663847780127, "prompt_id": "4b3ba24fb950898f", "reading_canonical_log_prob": -412.8876037597656, "reading_distances": [42, 39, 42, 44], "reading_log_probs": [-409.59619140625, -411.1595458984375, -411.03387451171875, -408.35333251953125], "reading_logprob_std": 1.3270154605216442, "tokenization_path_logprob_std": 292.33113566028226, "trajectory_balance_rmse": 6644.9602605697455} {"canonical_length": 33, "canonical_nll": 1.4260080510919744, "generation_rouge_l_f1": 0.09929078014184396, "generation_token_f1": 0.1347517730496454, "prompt_id": "7b02e9bcfb31548c", "reading_canonical_log_prob": -47.058265686035156, "reading_distances": [11, 16, 15, 12], "reading_log_probs": [-50.833274841308594, -49.105224609375, -50.76537322998047, -52.35090637207031], "reading_logprob_std": 1.3259723660998541, "tokenization_path_logprob_std": 90.80543912687341, "trajectory_balance_rmse": 971.3837113567665} {"canonical_length": 64, "canonical_nll": 0.9101220369338989, "generation_rouge_l_f1": 0.19933554817275745, "generation_token_f1": 0.29900332225913623, "prompt_id": "2e9002691e2c9828", "reading_canonical_log_prob": -58.24781036376953, "reading_distances": [33, 45, 41, 34], "reading_log_probs": [-64.91325378417969, -67.95118713378906, -64.35307312011719, -67.97807312011719], "reading_logprob_std": 1.9370022217901017, "tokenization_path_logprob_std": 191.73101292391502, "trajectory_balance_rmse": 1711.2302996427202} {"canonical_length": 110, "canonical_nll": 1.3367731267755683, "generation_rouge_l_f1": 0.2831325301204819, "generation_token_f1": 0.46987951807228917, "prompt_id": "660c03c6bae9ac45", "reading_canonical_log_prob": -147.0450439453125, "reading_distances": [31, 30, 29, 28], "reading_log_probs": [-143.87173461914062, -144.48597717285156, -145.64407348632812, -145.48460388183594], "reading_logprob_std": 0.8408180125836777, "tokenization_path_logprob_std": 199.35894043203822, "trajectory_balance_rmse": 3262.087979936736} {"canonical_length": 129, "canonical_nll": 1.1970664327458818, "generation_rouge_l_f1": 0.24403183023872685, "generation_token_f1": 0.45623342175066317, "prompt_id": "dc5df7204109382e", "reading_canonical_log_prob": -154.42156982421875, "reading_distances": [34, 21, 31, 29], "reading_log_probs": [-158.499267578125, -154.99412536621094, -156.0923614501953, -157.7398681640625], "reading_logprob_std": 1.5841772545194734, "tokenization_path_logprob_std": 248.18003866143596, "trajectory_balance_rmse": 3533.917628652026} {"canonical_length": 330, "canonical_nll": 1.4195547392874053, "generation_rouge_l_f1": 0.35175879396984927, "generation_token_f1": 0.5661641541038527, "prompt_id": "e7afe4dd1ceeee4d", "reading_canonical_log_prob": -468.45306396484375, "reading_distances": [24, 17, 18, 16], "reading_log_probs": [-474.18939208984375, -477.44482421875, -478.3687744140625, -477.8642883300781], "reading_logprob_std": 1.8897547823161285, "tokenization_path_logprob_std": 352.51109807228625, "trajectory_balance_rmse": 7649.403500807979} {"canonical_length": 142, "canonical_nll": 0.9172865101988886, "generation_rouge_l_f1": 0.19075144508670522, "generation_token_f1": 0.2601156069364162, "prompt_id": "64cb8edf4fa19a8d", "reading_canonical_log_prob": -130.2546844482422, "reading_distances": [14, 17, 19, 20], "reading_log_probs": [-136.50003051757812, -126.12014770507812, -122.25364685058594, -127.95416259765625], "reading_logprob_std": 6.017630988462904, "tokenization_path_logprob_std": 187.67700517740784, "trajectory_balance_rmse": 2260.5375790978983} {"canonical_length": 67, "canonical_nll": 1.7151582632491837, "generation_rouge_l_f1": 0.19827586206896552, "generation_token_f1": 0.2758620689655173, "prompt_id": "d242cdc5b38c5c82", "reading_canonical_log_prob": -114.91560363769531, "reading_distances": [24, 28, 31, 27], "reading_log_probs": [-110.67294311523438, -110.4783935546875, -108.92559051513672, -109.79019165039062], "reading_logprob_std": 0.7907114440520818, "tokenization_path_logprob_std": 161.53342279721343, "trajectory_balance_rmse": 2004.0665147397233} {"canonical_length": 71, "canonical_nll": 3.808630714953785, "generation_rouge_l_f1": 0.19771863117870722, "generation_token_f1": 0.2813688212927757, "prompt_id": "406fdefc259c7cdd", "reading_canonical_log_prob": -270.41278076171875, "reading_distances": [32, 31, 32, 35], "reading_log_probs": [-268.57373046875, -267.369140625, -266.1402587890625, -263.2023620605469], "reading_logprob_std": 2.3044856809321415, "tokenization_path_logprob_std": 160.321645647252, "trajectory_balance_rmse": 1590.2529123313243} {"canonical_length": 112, "canonical_nll": 1.2261483328683036, "generation_rouge_l_f1": 0.21164021164021163, "generation_token_f1": 0.39682539682539686, "prompt_id": "8b97cb84ee36bc33", "reading_canonical_log_prob": -137.32861328125, "reading_distances": [27, 21, 23, 24], "reading_log_probs": [-144.04110717773438, -141.53079223632812, -140.3355255126953, -141.76416015625], "reading_logprob_std": 1.5476224881896778, "tokenization_path_logprob_std": 169.57792635256098, "trajectory_balance_rmse": 3359.0517377044994} {"canonical_length": 3, "canonical_nll": 3.9190149307250977, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "3c3cb4aabb9bd15f", "reading_canonical_log_prob": -11.757044792175293, "reading_distances": [24, 22, 21, 22], "reading_log_probs": [-11.11372184753418, -10.287054061889648, -10.491006851196289, -10.451871871948242], "reading_logprob_std": 0.3628007008042722, "tokenization_path_logprob_std": 17.41727454661859, "trajectory_balance_rmse": 54.234684442730035} {"canonical_length": 76, "canonical_nll": 2.3653046457391036, "generation_rouge_l_f1": 0.21917808219178084, "generation_token_f1": 0.3835616438356165, "prompt_id": "4d20c1324c854a9e", "reading_canonical_log_prob": -179.76315307617188, "reading_distances": [11, 17, 12, 15], "reading_log_probs": [-172.88966369628906, -174.55575561523438, -175.3211212158203, -172.38092041015625], "reading_logprob_std": 1.3816397604621904, "tokenization_path_logprob_std": 224.33901202224564, "trajectory_balance_rmse": 1720.8897839019803} {"canonical_length": 194, "canonical_nll": 0.9518245421734053, "generation_rouge_l_f1": 0.38571428571428573, "generation_token_f1": 0.5190476190476191, "prompt_id": "45005e42ada68a08", "reading_canonical_log_prob": -184.65396118164062, "reading_distances": [25, 26, 20, 23], "reading_log_probs": [-186.01617431640625, -180.4222412109375, -181.52980041503906, -180.60079956054688], "reading_logprob_std": 2.627856147448261, "tokenization_path_logprob_std": 90.80430893175136, "trajectory_balance_rmse": 3512.8170035775756} {"canonical_length": 3, "canonical_nll": 2.4980700810750327, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "1b11c060139de682", "reading_canonical_log_prob": -7.494210243225098, "reading_distances": [17, 18, 19, 19], "reading_log_probs": [-13.537188529968262, -14.872187614440918, -15.57654857635498, -12.899950981140137], "reading_logprob_std": 1.22124594974046, "tokenization_path_logprob_std": 27.438087681250575, "trajectory_balance_rmse": 48.90706848580138} {"canonical_length": 180, "canonical_nll": 1.531303236219618, "generation_rouge_l_f1": 0.2073732718894009, "generation_token_f1": 0.4285714285714285, "prompt_id": "fb27819d14f6e258", "reading_canonical_log_prob": -275.63458251953125, "reading_distances": [18, 20, 19, 19], "reading_log_probs": [-274.25201416015625, -277.1975402832031, -276.71539306640625, -279.44219970703125], "reading_logprob_std": 2.1289438326570354, "tokenization_path_logprob_std": 281.7456828313804, "trajectory_balance_rmse": 5447.668935507665} {"canonical_length": 92, "canonical_nll": 2.043757894764776, "generation_rouge_l_f1": 0.1807909604519774, "generation_token_f1": 0.3050847457627119, "prompt_id": "b1a0ce36dec4c1a2", "reading_canonical_log_prob": -188.02572631835938, "reading_distances": [31, 34, 28, 28], "reading_log_probs": [-200.991455078125, -198.8856201171875, -198.4253692626953, -196.5819091796875], "reading_logprob_std": 1.8115530246987932, "tokenization_path_logprob_std": 298.48658044097266, "trajectory_balance_rmse": 2455.5651510598304} {"canonical_length": 205, "canonical_nll": 1.363515565453506, "generation_rouge_l_f1": 0.19393939393939394, "generation_token_f1": 0.3313131313131313, "prompt_id": "fe99f791079b2601", "reading_canonical_log_prob": -279.52069091796875, "reading_distances": [21, 31, 27, 25], "reading_log_probs": [-278.3788757324219, -277.974365234375, -273.79034423828125, -277.8692626953125], "reading_logprob_std": 2.1531490403964777, "tokenization_path_logprob_std": 266.87481875804315, "trajectory_balance_rmse": 5188.490498220524} {"canonical_length": 265, "canonical_nll": 1.9228810436320756, "generation_rouge_l_f1": 0.2015209125475285, "generation_token_f1": 0.4106463878326996, "prompt_id": "7fb35b5d3ec740ba", "reading_canonical_log_prob": -509.5634765625, "reading_distances": [22, 26, 27, 27], "reading_log_probs": [-513.4209594726562, -514.1891479492188, -513.1404418945312, -510.21697998046875], "reading_logprob_std": 1.7406612861590443, "tokenization_path_logprob_std": 249.40144128038833, "trajectory_balance_rmse": 8967.779767378493} {"canonical_length": 11, "canonical_nll": 0.30604869669133966, "generation_rouge_l_f1": 0.1652892561983471, "generation_token_f1": 0.18181818181818182, "prompt_id": "cb7972227cbb9dc7", "reading_canonical_log_prob": -3.3665356636047363, "reading_distances": [29, 19, 21, 21], "reading_log_probs": [-4.809101104736328, -4.596527576446533, -5.293715000152588, -3.617661952972412], "reading_logprob_std": 0.7043299069843638, "tokenization_path_logprob_std": 103.95648579351511, "trajectory_balance_rmse": 337.22074230978126} {"canonical_length": 109, "canonical_nll": 1.771903501738102, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.4215686274509804, "prompt_id": "ce54506448843c44", "reading_canonical_log_prob": -193.13748168945312, "reading_distances": [22, 21, 20, 20], "reading_log_probs": [-192.52252197265625, -193.04196166992188, -196.93740844726562, -193.91555786132812], "reading_logprob_std": 1.9742265581403469, "tokenization_path_logprob_std": 69.82470651547808, "trajectory_balance_rmse": 2504.192522544211} {"canonical_length": 90, "canonical_nll": 0.94666748046875, "generation_rouge_l_f1": 0.248995983935743, "generation_token_f1": 0.37751004016064255, "prompt_id": "110898cad92a62bf", "reading_canonical_log_prob": -85.2000732421875, "reading_distances": [12, 16, 16, 17], "reading_log_probs": [-83.99111938476562, -88.33790588378906, -87.41026306152344, -86.75524139404297], "reading_logprob_std": 1.8712711435066867, "tokenization_path_logprob_std": 157.07583449419627, "trajectory_balance_rmse": 1551.6205172408781} {"canonical_length": 6, "canonical_nll": 1.4449013074239094, "generation_rouge_l_f1": 0.03571428571428572, "generation_token_f1": 0.03571428571428572, "prompt_id": "bb43b5a4398228cc", "reading_canonical_log_prob": -8.669407844543457, "reading_distances": [34, 27, 22, 30], "reading_log_probs": [-9.475004196166992, -9.4750337600708, -10.201282501220703, -11.67701530456543], "reading_logprob_std": 1.0380382283406957, "tokenization_path_logprob_std": 29.370462213122963, "trajectory_balance_rmse": 156.5832801885293} {"canonical_length": 344, "canonical_nll": 1.4556627495344294, "generation_rouge_l_f1": 0.20714285714285713, "generation_token_f1": 0.39999999999999997, "prompt_id": "6dc0804e4c40522b", "reading_canonical_log_prob": -500.74798583984375, "reading_distances": [16, 21, 19, 19], "reading_log_probs": [-512.4367065429688, -503.0844421386719, -510.80999755859375, -511.7156982421875], "reading_logprob_std": 4.336223075985813, "tokenization_path_logprob_std": 220.34188627039953, "trajectory_balance_rmse": 7160.332474358751} {"canonical_length": 325, "canonical_nll": 1.2542364971454327, "generation_rouge_l_f1": 0.22410865874363328, "generation_token_f1": 0.44482173174872663, "prompt_id": "8ed2c6a41c30f4da", "reading_canonical_log_prob": -407.6268615722656, "reading_distances": [30, 35, 29, 23], "reading_log_probs": [-405.1260986328125, -403.46697998046875, -403.7618103027344, -405.48406982421875], "reading_logprob_std": 0.9943111074193257, "tokenization_path_logprob_std": 69.39322971505236, "trajectory_balance_rmse": 10091.521099472007} {"canonical_length": 116, "canonical_nll": 0.9676187449488146, "generation_rouge_l_f1": 0.3024691358024691, "generation_token_f1": 0.4444444444444445, "prompt_id": "b2cdbde556d30889", "reading_canonical_log_prob": -112.2437744140625, "reading_distances": [15, 15, 13, 12], "reading_log_probs": [-117.39642333984375, -117.49557495117188, -117.79417419433594, -117.76307678222656], "reading_logprob_std": 0.19667174804842058, "tokenization_path_logprob_std": 181.02935861024977, "trajectory_balance_rmse": 1913.855552141314} {"canonical_length": 15, "canonical_nll": 1.4273568471272786, "generation_rouge_l_f1": 0.48000000000000004, "generation_token_f1": 0.5599999999999999, "prompt_id": "eb54d1c2e52f2aa1", "reading_canonical_log_prob": -21.41035270690918, "reading_distances": [21, 23, 21, 19], "reading_log_probs": [-22.483226776123047, -23.39234733581543, -22.78385353088379, -23.62727165222168], "reading_logprob_std": 0.5293489388501351, "tokenization_path_logprob_std": 45.81204569970604, "trajectory_balance_rmse": 373.59886253366096} {"canonical_length": 61, "canonical_nll": 1.6751767768234502, "generation_rouge_l_f1": 0.14563106796116504, "generation_token_f1": 0.22330097087378642, "prompt_id": "2d9f537706729121", "reading_canonical_log_prob": -102.18578338623047, "reading_distances": [11, 15, 10, 10], "reading_log_probs": [-103.90161895751953, -101.86575317382812, -103.5658950805664, -101.84745025634766], "reading_logprob_std": 1.0924337957574437, "tokenization_path_logprob_std": 98.09393344346196, "trajectory_balance_rmse": 1696.135542841959} {"canonical_length": 2, "canonical_nll": 2.5827925205230713, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "e6a02b338700cb67", "reading_canonical_log_prob": -5.165585041046143, "reading_distances": [22, 19, 25, 21], "reading_log_probs": [-4.270348072052002, -4.49778413772583, -5.475268840789795, -4.302481651306152], "reading_logprob_std": 0.5681580610155391, "tokenization_path_logprob_std": 32.06284484287843, "trajectory_balance_rmse": 57.829394506766455} {"canonical_length": 169, "canonical_nll": 1.0531701082308618, "generation_rouge_l_f1": 0.2640586797066015, "generation_token_f1": 0.4938875305623473, "prompt_id": "56062aaca5480bb3", "reading_canonical_log_prob": -177.98574829101562, "reading_distances": [27, 24, 27, 23], "reading_log_probs": [-173.6464080810547, -176.9484405517578, -177.73947143554688, -175.40780639648438], "reading_logprob_std": 1.8072782344581197, "tokenization_path_logprob_std": 149.70765777826728, "trajectory_balance_rmse": 2962.6542578136923} {"canonical_length": 3, "canonical_nll": 0.5229362646738688, "generation_rouge_l_f1": 0.05769230769230769, "generation_token_f1": 0.05769230769230769, "prompt_id": "a6378b47a0666677", "reading_canonical_log_prob": -1.5688087940216064, "reading_distances": [29, 31, 27, 30], "reading_log_probs": [-4.106221675872803, -6.255492210388184, -6.649192810058594, -5.021649360656738], "reading_logprob_std": 1.163698840107562, "tokenization_path_logprob_std": 29.2533349311519, "trajectory_balance_rmse": 75.68208212572135} {"canonical_length": 199, "canonical_nll": 1.1374688076613537, "generation_rouge_l_f1": 0.26280623608017817, "generation_token_f1": 0.5345211581291759, "prompt_id": "41d1448457af2a4f", "reading_canonical_log_prob": -226.35629272460938, "reading_distances": [13, 18, 21, 16], "reading_log_probs": [-224.0036163330078, -217.8275909423828, -222.7141571044922, -221.82110595703125], "reading_logprob_std": 2.6644820892900603, "tokenization_path_logprob_std": 151.6720583539859, "trajectory_balance_rmse": 5868.087607583633} {"canonical_length": 94, "canonical_nll": 2.1352855601209275, "generation_rouge_l_f1": 0.1763085399449036, "generation_token_f1": 0.30303030303030304, "prompt_id": "5749ffa6095e5a40", "reading_canonical_log_prob": -200.7168426513672, "reading_distances": [31, 23, 22, 19], "reading_log_probs": [-192.5596923828125, -193.1227569580078, -194.19061279296875, -197.8304443359375], "reading_logprob_std": 2.368346459825997, "tokenization_path_logprob_std": 116.93150178240023, "trajectory_balance_rmse": 3061.402226852518} {"canonical_length": 58, "canonical_nll": 1.3916660177296605, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.5, "prompt_id": "e503fe8212044208", "reading_canonical_log_prob": -80.71662902832031, "reading_distances": [31, 31, 34, 34], "reading_log_probs": [-82.82653045654297, -81.05723571777344, -84.7785873413086, -81.20755004882812], "reading_logprob_std": 1.7365048028284993, "tokenization_path_logprob_std": 121.83173613115183, "trajectory_balance_rmse": 1893.6387010861017} {"canonical_length": 161, "canonical_nll": 0.9808390362662558, "generation_rouge_l_f1": 0.2569832402234637, "generation_token_f1": 0.5363128491620112, "prompt_id": "cf2e46f2d41b486e", "reading_canonical_log_prob": -157.9150848388672, "reading_distances": [25, 21, 22, 22], "reading_log_probs": [-154.66851806640625, -159.7145233154297, -155.01519775390625, -158.52235412597656], "reading_logprob_std": 2.5205733818677696, "tokenization_path_logprob_std": 240.85575797178709, "trajectory_balance_rmse": 3936.0161048122236} {"canonical_length": 62, "canonical_nll": 0.7204335735690209, "generation_rouge_l_f1": 0.16326530612244897, "generation_token_f1": 0.21224489795918366, "prompt_id": "45e020ec1f1a052d", "reading_canonical_log_prob": -44.6668815612793, "reading_distances": [12, 12, 12, 18], "reading_log_probs": [-46.971675872802734, -46.14058303833008, -47.63194274902344, -47.639976501464844], "reading_logprob_std": 0.7097941287658196, "tokenization_path_logprob_std": 42.84029571641404, "trajectory_balance_rmse": 1284.3160426365134} {"canonical_length": 284, "canonical_nll": 1.5995106226961378, "generation_rouge_l_f1": 0.22826086956521738, "generation_token_f1": 0.44565217391304346, "prompt_id": "238ad0e8d97d1bdf", "reading_canonical_log_prob": -454.2610168457031, "reading_distances": [13, 19, 17, 26], "reading_log_probs": [-459.372802734375, -452.60107421875, -459.91729736328125, -455.2835998535156], "reading_logprob_std": 3.476930868654408, "tokenization_path_logprob_std": 494.6137130845724, "trajectory_balance_rmse": 7085.342353235328} {"canonical_length": 30, "canonical_nll": 0.7787069320678711, "generation_rouge_l_f1": 0.04098360655737705, "generation_token_f1": 0.05737704918032787, "prompt_id": "aca807bc21ebbe28", "reading_canonical_log_prob": -23.361207962036133, "reading_distances": [9, 11, 9, 11], "reading_log_probs": [-27.33200454711914, -26.450927734375, -23.947795867919922, -25.90730857849121], "reading_logprob_std": 1.433521431293829, "tokenization_path_logprob_std": 15.231492695789026, "trajectory_balance_rmse": 490.87605210462937} {"canonical_length": 39, "canonical_nll": 1.2692010341546474, "generation_rouge_l_f1": 0.14332247557003258, "generation_token_f1": 0.20846905537459284, "prompt_id": "1c53f5181f02d61c", "reading_canonical_log_prob": -49.49884033203125, "reading_distances": [21, 21, 21, 22], "reading_log_probs": [-55.383705139160156, -50.84385681152344, -53.80711364746094, -53.115211486816406], "reading_logprob_std": 1.8854839952688163, "tokenization_path_logprob_std": 41.87219235248549, "trajectory_balance_rmse": 636.8882820613885} {"canonical_length": 10, "canonical_nll": 4.261152267456055, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "1a83cc65795284e0", "reading_canonical_log_prob": -42.61152267456055, "reading_distances": [21, 14, 22, 17], "reading_log_probs": [-28.017169952392578, -28.469234466552734, -27.03034019470215, -27.25111961364746], "reading_logprob_std": 0.6688293560007423, "tokenization_path_logprob_std": 65.88798752914327, "trajectory_balance_rmse": 151.7310678117878} {"canonical_length": 30, "canonical_nll": 1.9058268229166666, "generation_rouge_l_f1": 0.4657534246575342, "generation_token_f1": 0.4931506849315069, "prompt_id": "a5ddf34551bf57f2", "reading_canonical_log_prob": -57.1748046875, "reading_distances": [24, 20, 18, 17], "reading_log_probs": [-57.56973648071289, -58.92197799682617, -57.697723388671875, -58.445960998535156], "reading_logprob_std": 0.6388765022290461, "tokenization_path_logprob_std": 120.42555158349181, "trajectory_balance_rmse": 1057.9296280945466} {"canonical_length": 33, "canonical_nll": 1.0170696142948035, "generation_rouge_l_f1": 0.14592274678111586, "generation_token_f1": 0.1716738197424893, "prompt_id": "626308b96b881737", "reading_canonical_log_prob": -33.563297271728516, "reading_distances": [22, 18, 14, 14], "reading_log_probs": [-33.283233642578125, -33.57728576660156, -34.85144805908203, -32.316932678222656], "reading_logprob_std": 1.0454364452333518, "tokenization_path_logprob_std": 45.06075728464541, "trajectory_balance_rmse": 928.6032457442643} {"canonical_length": 79, "canonical_nll": 1.1057351027862936, "generation_rouge_l_f1": 0.16181229773462785, "generation_token_f1": 0.2783171521035599, "prompt_id": "b884d1bbe7793313", "reading_canonical_log_prob": -87.35307312011719, "reading_distances": [19, 23, 24, 21], "reading_log_probs": [-90.17314147949219, -94.2047119140625, -93.2576904296875, -97.31248474121094], "reading_logprob_std": 2.940162738544334, "tokenization_path_logprob_std": 61.51267384741294, "trajectory_balance_rmse": 1173.0044686607396} {"canonical_length": 50, "canonical_nll": 1.4483078002929688, "generation_rouge_l_f1": 0.2608695652173913, "generation_token_f1": 0.3850931677018634, "prompt_id": "d8395ff4ce8fbfae", "reading_canonical_log_prob": -72.41539001464844, "reading_distances": [38, 31, 29, 29], "reading_log_probs": [-82.70549011230469, -81.7010498046875, -79.5194320678711, -77.22586059570312], "reading_logprob_std": 2.4364155765013704, "tokenization_path_logprob_std": 185.90702188800893, "trajectory_balance_rmse": 1441.786696605393} {"canonical_length": 228, "canonical_nll": 1.5040055659779332, "generation_rouge_l_f1": 0.2494802494802495, "generation_token_f1": 0.5197505197505198, "prompt_id": "cabaeed82d3c2f7b", "reading_canonical_log_prob": -342.91326904296875, "reading_distances": [24, 21, 27, 23], "reading_log_probs": [-346.45062255859375, -348.11956787109375, -347.683349609375, -348.1184997558594], "reading_logprob_std": 0.788799295644687, "tokenization_path_logprob_std": 351.6689253960613, "trajectory_balance_rmse": 7108.951008791292} {"canonical_length": 60, "canonical_nll": 0.026119466622670492, "generation_rouge_l_f1": 0.6486486486486486, "generation_token_f1": 0.954954954954955, "prompt_id": "ef7598410b16ef36", "reading_canonical_log_prob": -1.5671679973602295, "reading_distances": [32, 36, 31, 32], "reading_log_probs": [-0.9937041401863098, -2.238217353820801, -2.154501438140869, -1.5204596519470215], "reading_logprob_std": 0.5843749035505722, "tokenization_path_logprob_std": 175.1435501160081, "trajectory_balance_rmse": 1443.5434480594622} {"canonical_length": 239, "canonical_nll": 1.2774886765739408, "generation_rouge_l_f1": 0.311578947368421, "generation_token_f1": 0.5178947368421053, "prompt_id": "39157986b3a30c4b", "reading_canonical_log_prob": -305.3197937011719, "reading_distances": [42, 44, 40, 40], "reading_log_probs": [-296.89752197265625, -298.29119873046875, -308.1842041015625, -304.0927734375], "reading_logprob_std": 5.239058449594101, "tokenization_path_logprob_std": 165.89527253947153, "trajectory_balance_rmse": 5595.3062653154375} {"canonical_length": 113, "canonical_nll": 1.9211709351666206, "generation_rouge_l_f1": 0.3243243243243243, "generation_token_f1": 0.4648648648648649, "prompt_id": "952e6be856268bb7", "reading_canonical_log_prob": -217.09231567382812, "reading_distances": [25, 27, 29, 33], "reading_log_probs": [-216.7091064453125, -218.57693481445312, -216.0467071533203, -216.89779663085938], "reading_logprob_std": 1.0766100338664044, "tokenization_path_logprob_std": 79.28251904479492, "trajectory_balance_rmse": 2989.0227461737923} {"canonical_length": 324, "canonical_nll": 1.0511572567033178, "generation_rouge_l_f1": 0.2381852551984877, "generation_token_f1": 0.45746691871455575, "prompt_id": "b30cec3b9127ceb8", "reading_canonical_log_prob": -340.574951171875, "reading_distances": [24, 19, 21, 20], "reading_log_probs": [-343.7160339355469, -343.3336486816406, -342.9854736328125, -346.1484680175781], "reading_logprob_std": 1.4331096619593953, "tokenization_path_logprob_std": 319.57877456533623, "trajectory_balance_rmse": 8971.085444844332} {"canonical_length": 211, "canonical_nll": 1.324955800133294, "generation_rouge_l_f1": 0.23476297968397294, "generation_token_f1": 0.41986455981941306, "prompt_id": "ce18c132bb5bd7f2", "reading_canonical_log_prob": -279.565673828125, "reading_distances": [25, 28, 28, 24], "reading_log_probs": [-277.93963623046875, -287.5829162597656, -276.8502197265625, -284.06060791015625], "reading_logprob_std": 5.09274893915868, "tokenization_path_logprob_std": 543.2379343919066, "trajectory_balance_rmse": 5623.276381303485} {"canonical_length": 53, "canonical_nll": 1.656855025381412, "generation_rouge_l_f1": 0.07294832826747721, "generation_token_f1": 0.1519756838905775, "prompt_id": "f1a703b446eae2b2", "reading_canonical_log_prob": -87.81331634521484, "reading_distances": [54, 59, 45, 56], "reading_log_probs": [-87.55804443359375, -84.78266906738281, -87.312744140625, -86.8996810913086], "reading_logprob_std": 1.2665558224296147, "tokenization_path_logprob_std": 118.84131057641183, "trajectory_balance_rmse": 1331.6338227361682} {"canonical_length": 56, "canonical_nll": 1.7900853838239397, "generation_rouge_l_f1": 0.20560747663551404, "generation_token_f1": 0.4485981308411215, "prompt_id": "7188e613a859c6bb", "reading_canonical_log_prob": -100.24478149414062, "reading_distances": [26, 25, 28, 24], "reading_log_probs": [-104.3916015625, -98.57054901123047, -102.21087646484375, -100.43687438964844], "reading_logprob_std": 2.4859994555901572, "tokenization_path_logprob_std": 108.71895630976611, "trajectory_balance_rmse": 1291.9550570951956} {"canonical_length": 198, "canonical_nll": 1.7216198853772096, "generation_rouge_l_f1": 0.3056179775280899, "generation_token_f1": 0.5033707865168539, "prompt_id": "85df6700a3ed9560", "reading_canonical_log_prob": -340.8807373046875, "reading_distances": [40, 30, 32, 41], "reading_log_probs": [-342.67218017578125, -340.3863525390625, -339.7376708984375, -343.9316101074219], "reading_logprob_std": 1.9579172952797301, "tokenization_path_logprob_std": 186.9117363997934, "trajectory_balance_rmse": 5434.99363288842} {"canonical_length": 94, "canonical_nll": 1.3331043973882148, "generation_rouge_l_f1": 0.21487603305785122, "generation_token_f1": 0.35261707988980723, "prompt_id": "0baa5b0629968fb8", "reading_canonical_log_prob": -125.31181335449219, "reading_distances": [26, 24, 29, 21], "reading_log_probs": [-129.98257446289062, -128.37969970703125, -126.98411560058594, -129.45291137695312], "reading_logprob_std": 1.3239743479557216, "tokenization_path_logprob_std": 118.04589634305985, "trajectory_balance_rmse": 3106.297821019979} {"canonical_length": 331, "canonical_nll": 1.2456837449549187, "generation_rouge_l_f1": 0.19649122807017544, "generation_token_f1": 0.47719298245614034, "prompt_id": "3d704d3c66ac7a28", "reading_canonical_log_prob": -412.3213195800781, "reading_distances": [25, 32, 31, 32], "reading_log_probs": [-400.7656555175781, -398.68963623046875, -398.7269287109375, -400.6505126953125], "reading_logprob_std": 1.1556428054877124, "tokenization_path_logprob_std": 188.76227648055146, "trajectory_balance_rmse": 7825.301444879182} {"canonical_length": 14, "canonical_nll": 0.7464733804975238, "generation_rouge_l_f1": 0.761904761904762, "generation_token_f1": 0.761904761904762, "prompt_id": "d990b55cc5500b31", "reading_canonical_log_prob": -10.450627326965332, "reading_distances": [59, 60, 60, 60], "reading_log_probs": [-16.08380126953125, -17.921592712402344, -17.962940216064453, -18.318517684936523], "reading_logprob_std": 1.0078149801840177, "tokenization_path_logprob_std": 60.02216573660233, "trajectory_balance_rmse": 426.88915262615893} {"canonical_length": 125, "canonical_nll": 1.72955712890625, "generation_rouge_l_f1": 0.18528610354223432, "generation_token_f1": 0.33242506811989103, "prompt_id": "96948513433b0e29", "reading_canonical_log_prob": -216.19464111328125, "reading_distances": [14, 19, 14, 18], "reading_log_probs": [-220.0175018310547, -221.04446411132812, -218.98165893554688, -218.54539489746094], "reading_logprob_std": 1.1174956857884264, "tokenization_path_logprob_std": 173.80053829676294, "trajectory_balance_rmse": 3797.892003604955} {"canonical_length": 88, "canonical_nll": 1.9869445453990588, "generation_rouge_l_f1": 0.28691983122362874, "generation_token_f1": 0.4388185654008439, "prompt_id": "5b5a0a9c17c96287", "reading_canonical_log_prob": -174.8511199951172, "reading_distances": [43, 45, 35, 38], "reading_log_probs": [-179.07107543945312, -178.65176391601562, -177.53640747070312, -174.88796997070312], "reading_logprob_std": 1.8809081438000081, "tokenization_path_logprob_std": 178.2285497521701, "trajectory_balance_rmse": 3043.2968137893745} {"canonical_length": 310, "canonical_nll": 1.3772998440650202, "generation_rouge_l_f1": 0.23941068139963168, "generation_token_f1": 0.5672191528545121, "prompt_id": "9ae64366f18d52ad", "reading_canonical_log_prob": -426.96295166015625, "reading_distances": [20, 27, 15, 21], "reading_log_probs": [-436.3290710449219, -436.5566711425781, -438.7095947265625, -436.564208984375], "reading_logprob_std": 1.1184737150164252, "tokenization_path_logprob_std": 401.7605748449449, "trajectory_balance_rmse": 9413.490327448659} {"canonical_length": 192, "canonical_nll": 1.4766182899475098, "generation_rouge_l_f1": 0.19815668202764977, "generation_token_f1": 0.4470046082949309, "prompt_id": "eaabe2e9ef0c1d39", "reading_canonical_log_prob": -283.5107116699219, "reading_distances": [28, 29, 31, 35], "reading_log_probs": [-285.8160705566406, -281.83154296875, -282.84979248046875, -284.695556640625], "reading_logprob_std": 1.792973993015138, "tokenization_path_logprob_std": 312.8786491923445, "trajectory_balance_rmse": 5381.640292956192} {"canonical_length": 208, "canonical_nll": 1.5223102569580078, "generation_rouge_l_f1": 0.2898550724637681, "generation_token_f1": 0.5590062111801243, "prompt_id": "74426ee6c2f5003e", "reading_canonical_log_prob": -316.6405334472656, "reading_distances": [20, 21, 28, 22], "reading_log_probs": [-316.01904296875, -317.1894836425781, -314.4510192871094, -315.8199768066406], "reading_logprob_std": 1.1223872498744059, "tokenization_path_logprob_std": 175.84678883873212, "trajectory_balance_rmse": 5941.148319642517} {"canonical_length": 341, "canonical_nll": 1.3548369197901393, "generation_rouge_l_f1": 0.2957317073170732, "generation_token_f1": 0.5670731707317074, "prompt_id": "6909a2a1a6b2d043", "reading_canonical_log_prob": -461.9993896484375, "reading_distances": [36, 44, 34, 33], "reading_log_probs": [-476.76904296875, -477.2103271484375, -476.4200439453125, -480.87127685546875], "reading_logprob_std": 2.0612583055413465, "tokenization_path_logprob_std": 383.2471885108086, "trajectory_balance_rmse": 6736.118493319458} {"canonical_length": 19, "canonical_nll": 1.8603601957622327, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "fcaeb274bce92488", "reading_canonical_log_prob": -35.34684371948242, "reading_distances": [24, 30, 32, 34], "reading_log_probs": [-37.50080108642578, -45.171566009521484, -33.86253356933594, -45.92300796508789], "reading_logprob_std": 5.894383740297714, "tokenization_path_logprob_std": 30.740856725163468, "trajectory_balance_rmse": 393.89658395225723} {"canonical_length": 20, "canonical_nll": 0.69988112449646, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.8000000000000002, "prompt_id": "59d62ad713859ec4", "reading_canonical_log_prob": -13.9976224899292, "reading_distances": [13, 15, 14, 16], "reading_log_probs": [-15.288256645202637, -16.254087448120117, -14.371963500976562, -13.851146697998047], "reading_logprob_std": 1.0577283917925056, "tokenization_path_logprob_std": 51.12411620548866, "trajectory_balance_rmse": 217.26662038028422} {"canonical_length": 227, "canonical_nll": 1.3770421233996422, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.3423076923076923, "prompt_id": "0205b5c76a68ce56", "reading_canonical_log_prob": -312.58856201171875, "reading_distances": [32, 25, 25, 26], "reading_log_probs": [-317.78216552734375, -312.20965576171875, -320.81402587890625, -318.3642578125], "reading_logprob_std": 3.6343219083713416, "tokenization_path_logprob_std": 254.23529949093995, "trajectory_balance_rmse": 5160.145246726837} {"canonical_length": 251, "canonical_nll": 1.689862376665214, "generation_rouge_l_f1": 0.22007722007722008, "generation_token_f1": 0.4362934362934363, "prompt_id": "28e138f0fd885003", "reading_canonical_log_prob": -424.15545654296875, "reading_distances": [23, 25, 29, 28], "reading_log_probs": [-432.51165771484375, -432.0673522949219, -433.57861328125, -423.8673095703125], "reading_logprob_std": 4.471153715308492, "tokenization_path_logprob_std": 229.69734086301605, "trajectory_balance_rmse": 5982.679498744484} {"canonical_length": 25, "canonical_nll": 0.32258785247802735, "generation_rouge_l_f1": 0.6829268292682927, "generation_token_f1": 0.6829268292682927, "prompt_id": "3cdea1f3e20549fa", "reading_canonical_log_prob": -8.064696311950684, "reading_distances": [31, 36, 33, 37], "reading_log_probs": [-9.767156600952148, -10.116240501403809, -7.660866737365723, -10.683610916137695], "reading_logprob_std": 1.3192787470737368, "tokenization_path_logprob_std": 56.508782483240076, "trajectory_balance_rmse": 554.4659178715201} {"canonical_length": 143, "canonical_nll": 2.49180506993007, "generation_rouge_l_f1": 0.13574660633484162, "generation_token_f1": 0.248868778280543, "prompt_id": "b7ff5b60544bcb0d", "reading_canonical_log_prob": -356.328125, "reading_distances": [17, 16, 16, 18], "reading_log_probs": [-375.7494201660156, -388.42352294921875, -389.38482666015625, -376.4790954589844], "reading_logprob_std": 7.400681096943312, "tokenization_path_logprob_std": 168.56347342171952, "trajectory_balance_rmse": 3709.3229252464816} {"canonical_length": 297, "canonical_nll": 1.4091542048085017, "generation_rouge_l_f1": 0.21673003802281368, "generation_token_f1": 0.4714828897338403, "prompt_id": "2c96188d7d0e9844", "reading_canonical_log_prob": -418.518798828125, "reading_distances": [20, 19, 18, 15], "reading_log_probs": [-416.44439697265625, -416.27618408203125, -414.10968017578125, -418.6334228515625], "reading_logprob_std": 1.848097985898073, "tokenization_path_logprob_std": 335.1911209564547, "trajectory_balance_rmse": 6834.369753715711} {"canonical_length": 101, "canonical_nll": 1.2061062047977258, "generation_rouge_l_f1": 0.381294964028777, "generation_token_f1": 0.5683453237410072, "prompt_id": "f320a0730139386c", "reading_canonical_log_prob": -121.81672668457031, "reading_distances": [35, 41, 38, 37], "reading_log_probs": [-119.68212127685547, -118.25840759277344, -119.12146759033203, -119.0013656616211], "reading_logprob_std": 0.5856623844602259, "tokenization_path_logprob_std": 311.0911082396128, "trajectory_balance_rmse": 2211.431201861955} {"canonical_length": 109, "canonical_nll": 2.111291167932913, "generation_rouge_l_f1": 0.18994413407821228, "generation_token_f1": 0.4245810055865922, "prompt_id": "2a4224ce9ad0c001", "reading_canonical_log_prob": -230.1307373046875, "reading_distances": [17, 13, 19, 14], "reading_log_probs": [-235.94017028808594, -239.77252197265625, -247.1388397216797, -233.38934326171875], "reading_logprob_std": 5.990687881853266, "tokenization_path_logprob_std": 243.45026727718513, "trajectory_balance_rmse": 2777.2217823329297} {"canonical_length": 20, "canonical_nll": 0.8552451133728027, "generation_rouge_l_f1": 0.3243243243243243, "generation_token_f1": 0.3783783783783784, "prompt_id": "af538126b4c7deeb", "reading_canonical_log_prob": -17.104902267456055, "reading_distances": [27, 37, 35, 32], "reading_log_probs": [-18.123916625976562, -19.658935546875, -19.32957649230957, -19.96564483642578], "reading_logprob_std": 0.8066903844037959, "tokenization_path_logprob_std": 59.352580714022594, "trajectory_balance_rmse": 470.1609408518066} {"canonical_length": 48, "canonical_nll": 1.7421050071716309, "generation_rouge_l_f1": 0.3529411764705882, "generation_token_f1": 0.4999999999999999, "prompt_id": "6c3b3c80c0d299e0", "reading_canonical_log_prob": -83.62104034423828, "reading_distances": [29, 23, 19, 32], "reading_log_probs": [-79.86764526367188, -81.359619140625, -79.3731689453125, -80.79631805419922], "reading_logprob_std": 0.8954315490935651, "tokenization_path_logprob_std": 37.042271145737054, "trajectory_balance_rmse": 1438.4654747274012} {"canonical_length": 131, "canonical_nll": 1.3854634525211713, "generation_rouge_l_f1": 0.25641025641025644, "generation_token_f1": 0.46153846153846156, "prompt_id": "783057aa1290e5ce", "reading_canonical_log_prob": -181.49571228027344, "reading_distances": [35, 27, 24, 27], "reading_log_probs": [-187.82635498046875, -186.15725708007812, -190.18557739257812, -184.01986694335938], "reading_logprob_std": 2.6085269765359964, "tokenization_path_logprob_std": 331.7909703535823, "trajectory_balance_rmse": 3676.6570725999236} {"canonical_length": 358, "canonical_nll": 1.6376351297900664, "generation_rouge_l_f1": 0.21016949152542372, "generation_token_f1": 0.4067796610169492, "prompt_id": "609ea42a956a4ffa", "reading_canonical_log_prob": -586.2733764648438, "reading_distances": [13, 18, 13, 14], "reading_log_probs": [-581.6015625, -583.6124267578125, -583.53662109375, -584.7730712890625], "reading_logprob_std": 1.3142851712308303, "tokenization_path_logprob_std": 296.43982217345496, "trajectory_balance_rmse": 8370.122968969903} {"canonical_length": 185, "canonical_nll": 1.0382923847920187, "generation_rouge_l_f1": 0.3776223776223776, "generation_token_f1": 0.634032634032634, "prompt_id": "0fd242df3fc4b686", "reading_canonical_log_prob": -192.08409118652344, "reading_distances": [21, 23, 25, 23], "reading_log_probs": [-189.97923278808594, -188.28346252441406, -186.9634552001953, -191.9931182861328], "reading_logprob_std": 2.1761541384306717, "tokenization_path_logprob_std": 216.99720436703728, "trajectory_balance_rmse": 5012.6097407099805} {"canonical_length": 78, "canonical_nll": 0.6265907776661408, "generation_rouge_l_f1": 0.2845849802371542, "generation_token_f1": 0.4110671936758893, "prompt_id": "c7ad7826ab1191ed", "reading_canonical_log_prob": -48.874080657958984, "reading_distances": [33, 30, 29, 24], "reading_log_probs": [-49.959327697753906, -52.22422790527344, -50.93304443359375, -54.62255859375], "reading_logprob_std": 2.0177396430963266, "tokenization_path_logprob_std": 159.52515265928258, "trajectory_balance_rmse": 1887.455891966354} {"canonical_length": 9, "canonical_nll": 1.8062491946750217, "generation_rouge_l_f1": 0.06896551724137931, "generation_token_f1": 0.10344827586206896, "prompt_id": "f4e51cd8f0bc2528", "reading_canonical_log_prob": -16.256242752075195, "reading_distances": [30, 27, 27, 24], "reading_log_probs": [-20.936939239501953, -21.66376304626465, -21.186134338378906, -19.789108276367188], "reading_logprob_std": 0.795928622749766, "tokenization_path_logprob_std": 55.0999069993853, "trajectory_balance_rmse": 225.5131020812216} {"canonical_length": 110, "canonical_nll": 1.394601162997159, "generation_rouge_l_f1": 0.3249097472924188, "generation_token_f1": 0.4404332129963899, "prompt_id": "67c67cc47e6c2592", "reading_canonical_log_prob": -153.4061279296875, "reading_distances": [20, 31, 25, 28], "reading_log_probs": [-145.58180236816406, -141.419677734375, -143.4654998779297, -140.79428100585938], "reading_logprob_std": 2.1686074269273017, "tokenization_path_logprob_std": 95.35034924341745, "trajectory_balance_rmse": 2480.598472829687} {"canonical_length": 18, "canonical_nll": 1.3220475514729817, "generation_rouge_l_f1": 0.6666666666666667, "generation_token_f1": 0.6666666666666667, "prompt_id": "82aeadf42ab79a24", "reading_canonical_log_prob": -23.796855926513672, "reading_distances": [41, 46, 37, 40], "reading_log_probs": [-26.82012939453125, -25.222993850708008, -25.097562789916992, -26.898046493530273], "reading_logprob_std": 0.9826590033003265, "tokenization_path_logprob_std": 31.615361041979682, "trajectory_balance_rmse": 540.8860546066614} {"canonical_length": 12, "canonical_nll": 0.010252140462398529, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "736b52db67801c89", "reading_canonical_log_prob": -0.12302568554878235, "reading_distances": [24, 21, 27, 28], "reading_log_probs": [-0.412161648273468, -0.2487332969903946, -0.2692430913448334, -0.367511510848999], "reading_logprob_std": 0.0781631694562937, "tokenization_path_logprob_std": 39.347698247177746, "trajectory_balance_rmse": 298.7683260390721} {"canonical_length": 29, "canonical_nll": 1.1530015879663928, "generation_rouge_l_f1": 0.5074626865671643, "generation_token_f1": 0.537313432835821, "prompt_id": "3c9993234c0b5914", "reading_canonical_log_prob": -33.43704605102539, "reading_distances": [21, 21, 19, 29], "reading_log_probs": [-33.8502197265625, -31.927085876464844, -32.234771728515625, -32.445068359375], "reading_logprob_std": 0.8509684440682266, "tokenization_path_logprob_std": 64.04046054781487, "trajectory_balance_rmse": 1022.3279354807094} {"canonical_length": 221, "canonical_nll": 1.1341974595013786, "generation_rouge_l_f1": 0.1955193482688391, "generation_token_f1": 0.4725050916496945, "prompt_id": "1560e8f9b6710dda", "reading_canonical_log_prob": -250.6576385498047, "reading_distances": [14, 14, 16, 10], "reading_log_probs": [-249.7227325439453, -252.13034057617188, -249.1243896484375, -244.5242156982422], "reading_logprob_std": 3.1784383621411907, "tokenization_path_logprob_std": 266.5533630070748, "trajectory_balance_rmse": 6024.615979103001} {"canonical_length": 184, "canonical_nll": 2.009441375732422, "generation_rouge_l_f1": 0.2058165548098434, "generation_token_f1": 0.3266219239373602, "prompt_id": "1073dfb1d10b700d", "reading_canonical_log_prob": -369.7372131347656, "reading_distances": [15, 18, 21, 20], "reading_log_probs": [-369.2649230957031, -371.8089904785156, -377.80615234375, -369.5355224609375], "reading_logprob_std": 3.9690051845066736, "tokenization_path_logprob_std": 182.14915122570588, "trajectory_balance_rmse": 5384.862034007533} {"canonical_length": 67, "canonical_nll": 1.5340396254809934, "generation_rouge_l_f1": 0.25142857142857145, "generation_token_f1": 0.44571428571428573, "prompt_id": "2045b4674c0f9cd0", "reading_canonical_log_prob": -102.78065490722656, "reading_distances": [29, 22, 24, 21], "reading_log_probs": [-99.27041625976562, -99.19124603271484, -102.2420654296875, -100.717529296875], "reading_logprob_std": 1.4402648156899087, "tokenization_path_logprob_std": 236.78604340140217, "trajectory_balance_rmse": 2070.6069864889037} {"canonical_length": 18, "canonical_nll": 0.48998408847384983, "generation_rouge_l_f1": 0.7200000000000001, "generation_token_f1": 0.7200000000000001, "prompt_id": "e0eb2633a5eb27ba", "reading_canonical_log_prob": -8.819713592529297, "reading_distances": [17, 23, 24, 27], "reading_log_probs": [-9.124402046203613, -8.511990547180176, -9.498918533325195, -9.15183162689209], "reading_logprob_std": 0.410279945372592, "tokenization_path_logprob_std": 56.42587438513848, "trajectory_balance_rmse": 626.5100170468234} {"canonical_length": 10, "canonical_nll": 2.6236030578613283, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.4, "prompt_id": "d46e9e081fc93a1e", "reading_canonical_log_prob": -26.23603057861328, "reading_distances": [15, 16, 10, 18], "reading_log_probs": [-24.737457275390625, -25.21078872680664, -25.02464485168457, -24.194129943847656], "reading_logprob_std": 0.44344518558585194, "tokenization_path_logprob_std": 53.661570101544946, "trajectory_balance_rmse": 381.36343177460543} {"canonical_length": 104, "canonical_nll": 1.9411778083214393, "generation_rouge_l_f1": 0.1759530791788856, "generation_token_f1": 0.34604105571847504, "prompt_id": "d0a49199c6983c98", "reading_canonical_log_prob": -201.8824920654297, "reading_distances": [26, 23, 34, 26], "reading_log_probs": [-200.0170135498047, -199.0071563720703, -199.39913940429688, -197.13662719726562], "reading_logprob_std": 1.240621386554482, "tokenization_path_logprob_std": 179.01231586986762, "trajectory_balance_rmse": 2342.036513166303} {"canonical_length": 2, "canonical_nll": 2.420947790145874, "generation_rouge_l_f1": 0.025641025641025647, "generation_token_f1": 0.025641025641025647, "prompt_id": "22a76db8343e67a9", "reading_canonical_log_prob": -4.841895580291748, "reading_distances": [28, 20, 26, 22], "reading_log_probs": [-5.321838855743408, -5.945690631866455, -4.646981239318848, -5.008481025695801], "reading_logprob_std": 0.5506440924356052, "tokenization_path_logprob_std": 17.055484134909687, "trajectory_balance_rmse": 40.06668733173946} {"canonical_length": 193, "canonical_nll": 1.3713941821162565, "generation_rouge_l_f1": 0.2368421052631579, "generation_token_f1": 0.5087719298245614, "prompt_id": "ab11d19cfada82a0", "reading_canonical_log_prob": -264.6790771484375, "reading_distances": [23, 24, 17, 23], "reading_log_probs": [-266.6654052734375, -269.65093994140625, -264.73114013671875, -267.48883056640625], "reading_logprob_std": 2.037499113691538, "tokenization_path_logprob_std": 264.1657077257584, "trajectory_balance_rmse": 5954.582040837999} {"canonical_length": 246, "canonical_nll": 1.521346549677655, "generation_rouge_l_f1": 0.29333333333333333, "generation_token_f1": 0.5257142857142857, "prompt_id": "65ef693d02b228c6", "reading_canonical_log_prob": -374.2512512207031, "reading_distances": [76, 81, 86, 85], "reading_log_probs": [-386.6204833984375, -382.3462829589844, -380.7850036621094, -373.72235107421875], "reading_logprob_std": 5.364810390995037, "tokenization_path_logprob_std": 279.86794620856864, "trajectory_balance_rmse": 7113.025587310011} {"canonical_length": 24, "canonical_nll": 0.7058135668436686, "generation_rouge_l_f1": 0.5714285714285714, "generation_token_f1": 0.6428571428571429, "prompt_id": "4fb4dab7cec52a74", "reading_canonical_log_prob": -16.939525604248047, "reading_distances": [19, 27, 21, 27], "reading_log_probs": [-23.199167251586914, -23.444042205810547, -22.819684982299805, -24.470542907714844], "reading_logprob_std": 0.7064715209151361, "tokenization_path_logprob_std": 120.34453250738842, "trajectory_balance_rmse": 623.6114499065902} {"canonical_length": 36, "canonical_nll": 1.3923091888427734, "generation_rouge_l_f1": 0.14332247557003255, "generation_token_f1": 0.17589576547231273, "prompt_id": "d445e9b3b0bb59b8", "reading_canonical_log_prob": -50.123130798339844, "reading_distances": [49, 43, 38, 42], "reading_log_probs": [-59.87841796875, -55.51862716674805, -56.60491180419922, -55.682159423828125], "reading_logprob_std": 2.028760567656833, "tokenization_path_logprob_std": 127.81054976921645, "trajectory_balance_rmse": 1141.9010299839113} {"canonical_length": 17, "canonical_nll": 1.1546789057114546, "generation_rouge_l_f1": 0.14388489208633093, "generation_token_f1": 0.15827338129496402, "prompt_id": "9384aa5dbac32997", "reading_canonical_log_prob": -19.629541397094727, "reading_distances": [27, 24, 26, 25], "reading_log_probs": [-16.892009735107422, -18.16619300842285, -17.137006759643555, -17.958412170410156], "reading_logprob_std": 0.6189968208188614, "tokenization_path_logprob_std": 74.18327397694387, "trajectory_balance_rmse": 408.1055460245265} {"canonical_length": 227, "canonical_nll": 1.5706182135359306, "generation_rouge_l_f1": 0.17787418655097614, "generation_token_f1": 0.32537960954446854, "prompt_id": "7e391a560a4818b8", "reading_canonical_log_prob": -356.53033447265625, "reading_distances": [12, 17, 13, 11], "reading_log_probs": [-358.189208984375, -356.10650634765625, -354.22149658203125, -358.6169128417969], "reading_logprob_std": 2.0297456583132587, "tokenization_path_logprob_std": 203.17925787152143, "trajectory_balance_rmse": 6082.691521309454} {"canonical_length": 30, "canonical_nll": 7.021948750813802, "generation_rouge_l_f1": 0.05333333333333333, "generation_token_f1": 0.05333333333333333, "prompt_id": "62ca6c153160a733", "reading_canonical_log_prob": -210.65846252441406, "reading_distances": [47, 42, 48, 45], "reading_log_probs": [-230.00201416015625, -228.31813049316406, -226.14364624023438, -218.60464477539062], "reading_logprob_std": 5.029407070294851, "tokenization_path_logprob_std": 0.0, "trajectory_balance_rmse": 210.85601806640625} {"canonical_length": 85, "canonical_nll": 0.45335814532111673, "generation_rouge_l_f1": 0.1389728096676737, "generation_token_f1": 0.25981873111782483, "prompt_id": "5fd334d7d6294d39", "reading_canonical_log_prob": -38.53544235229492, "reading_distances": [25, 29, 20, 25], "reading_log_probs": [-42.03627395629883, -39.846336364746094, -39.55478286743164, -42.60771942138672], "reading_logprob_std": 1.535980577903626, "tokenization_path_logprob_std": 140.9866288834231, "trajectory_balance_rmse": 1921.8095989417268} {"canonical_length": 161, "canonical_nll": 1.7858462126358696, "generation_rouge_l_f1": 0.24213075060532688, "generation_token_f1": 0.4067796610169492, "prompt_id": "ec23e8fead4fc5cf", "reading_canonical_log_prob": -287.521240234375, "reading_distances": [25, 31, 28, 26], "reading_log_probs": [-290.5517272949219, -293.4093933105469, -291.77154541015625, -294.3336181640625], "reading_logprob_std": 1.6846825605580207, "tokenization_path_logprob_std": 306.81693604122626, "trajectory_balance_rmse": 3825.6021340983957} {"canonical_length": 296, "canonical_nll": 1.2625280844198692, "generation_rouge_l_f1": 0.28102189781021897, "generation_token_f1": 0.5328467153284672, "prompt_id": "6d37aa3b89a989f6", "reading_canonical_log_prob": -373.70831298828125, "reading_distances": [16, 20, 24, 16], "reading_log_probs": [-386.60662841796875, -381.8048095703125, -385.1957702636719, -383.61505126953125], "reading_logprob_std": 2.0670385576197705, "tokenization_path_logprob_std": 93.0251882498399, "trajectory_balance_rmse": 7940.210557748819} {"canonical_length": 19, "canonical_nll": 3.560460140830592, "generation_rouge_l_f1": 0.20000000000000004, "generation_token_f1": 0.33333333333333337, "prompt_id": "b2ecf966e39675a0", "reading_canonical_log_prob": -67.64874267578125, "reading_distances": [23, 29, 29, 24], "reading_log_probs": [-68.76990509033203, -67.44590759277344, -68.97792053222656, -69.56480407714844], "reading_logprob_std": 0.8948613696311825, "tokenization_path_logprob_std": 72.73198635748014, "trajectory_balance_rmse": 517.9328980600836} {"canonical_length": 17, "canonical_nll": 1.4377831851734835, "generation_rouge_l_f1": 0.6000000000000001, "generation_token_f1": 0.6000000000000001, "prompt_id": "efd422672564486e", "reading_canonical_log_prob": -24.44231414794922, "reading_distances": [14, 19, 18, 17], "reading_log_probs": [-33.06229019165039, -37.6436767578125, -26.200345993041992, -39.04987335205078], "reading_logprob_std": 5.787659964381419, "tokenization_path_logprob_std": 44.671289387187585, "trajectory_balance_rmse": 307.76324700153486} {"canonical_length": 196, "canonical_nll": 1.796255617725606, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.47685185185185186, "prompt_id": "743badce8c66694d", "reading_canonical_log_prob": -352.06610107421875, "reading_distances": [18, 18, 22, 15], "reading_log_probs": [-351.9268798828125, -355.00213623046875, -355.1374206542969, -361.92071533203125], "reading_logprob_std": 4.218408526724586, "tokenization_path_logprob_std": 458.5331965437675, "trajectory_balance_rmse": 5987.168715621845} {"canonical_length": 14, "canonical_nll": 0.045280473572867255, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "685c7dde1bc34a6f", "reading_canonical_log_prob": -0.6339266300201416, "reading_distances": [33, 28, 26, 35], "reading_log_probs": [-0.6546462774276733, -0.651309609413147, -0.5702091455459595, -0.6317965984344482], "reading_logprob_std": 0.03917261757716539, "tokenization_path_logprob_std": 31.24544693323673, "trajectory_balance_rmse": 265.6994673290595} {"canonical_length": 214, "canonical_nll": 1.473424180645809, "generation_rouge_l_f1": 0.24545454545454545, "generation_token_f1": 0.4954545454545454, "prompt_id": "d0bf7e97f3a575eb", "reading_canonical_log_prob": -315.3127746582031, "reading_distances": [28, 25, 35, 29], "reading_log_probs": [-328.7340087890625, -330.89813232421875, -329.7182922363281, -330.8505859375], "reading_logprob_std": 1.033139056583679, "tokenization_path_logprob_std": 397.04093128147184, "trajectory_balance_rmse": 6036.2127469893285} {"canonical_length": 120, "canonical_nll": 0.8885658899943034, "generation_rouge_l_f1": 0.15466666666666665, "generation_token_f1": 0.3093333333333333, "prompt_id": "181490950025f621", "reading_canonical_log_prob": -106.6279067993164, "reading_distances": [31, 28, 22, 22], "reading_log_probs": [-102.7503890991211, -114.3336181640625, -102.59782409667969, -105.73685455322266], "reading_logprob_std": 5.5121091451699735, "tokenization_path_logprob_std": 245.60981050364506, "trajectory_balance_rmse": 3008.1315899917413} {"canonical_length": 38, "canonical_nll": 1.8441471300627057, "generation_rouge_l_f1": 0.11803278688524589, "generation_token_f1": 0.17704918032786887, "prompt_id": "70f652d8f02698c0", "reading_canonical_log_prob": -70.07759094238281, "reading_distances": [28, 29, 27, 30], "reading_log_probs": [-69.43666076660156, -68.51900482177734, -68.01025390625, -68.33050537109375], "reading_logprob_std": 0.6121838817594467, "tokenization_path_logprob_std": 42.31485402228177, "trajectory_balance_rmse": 1368.916736693079} {"canonical_length": 278, "canonical_nll": 1.2219259138587568, "generation_rouge_l_f1": 0.2448210922787194, "generation_token_f1": 0.5084745762711865, "prompt_id": "fe3deab131dcdb79", "reading_canonical_log_prob": -339.6954040527344, "reading_distances": [25, 32, 23, 30], "reading_log_probs": [-340.50445556640625, -340.04046630859375, -341.4206237792969, -340.1966552734375], "reading_logprob_std": 0.6175704943594486, "tokenization_path_logprob_std": 228.0115063805787, "trajectory_balance_rmse": 8645.211184882843} {"canonical_length": 16, "canonical_nll": 1.0673850774765015, "generation_rouge_l_f1": 0.20689655172413793, "generation_token_f1": 0.20689655172413793, "prompt_id": "16c98792a8b466f0", "reading_canonical_log_prob": -17.078161239624023, "reading_distances": [33, 33, 34, 31], "reading_log_probs": [-25.455760955810547, -27.507434844970703, -26.539976119995117, -25.456512451171875], "reading_logprob_std": 0.9874648266322973, "tokenization_path_logprob_std": 58.156717859754124, "trajectory_balance_rmse": 402.9057670565711} {"canonical_length": 53, "canonical_nll": 1.4665993744472288, "generation_rouge_l_f1": 0.28888888888888886, "generation_token_f1": 0.37777777777777777, "prompt_id": "52f4380d6ba7596b", "reading_canonical_log_prob": -77.72976684570312, "reading_distances": [13, 13, 14, 16], "reading_log_probs": [-80.21247863769531, -80.31552124023438, -82.24114990234375, -81.35053253173828], "reading_logprob_std": 0.9571565947463265, "tokenization_path_logprob_std": 153.17369075363237, "trajectory_balance_rmse": 1320.9745648229746} {"canonical_length": 78, "canonical_nll": 0.699178451146835, "generation_rouge_l_f1": 0.21052631578947367, "generation_token_f1": 0.2763157894736842, "prompt_id": "9b002f163aa12acd", "reading_canonical_log_prob": -54.535919189453125, "reading_distances": [18, 19, 21, 20], "reading_log_probs": [-56.425079345703125, -55.589805603027344, -55.8062629699707, -56.41999816894531], "reading_logprob_std": 0.4275303237885957, "tokenization_path_logprob_std": 113.25975212141337, "trajectory_balance_rmse": 1567.3774046700487} {"canonical_length": 22, "canonical_nll": 2.9385195645419033, "generation_rouge_l_f1": 0.2777777777777778, "generation_token_f1": 0.4444444444444445, "prompt_id": "6e455bb516dd4dc1", "reading_canonical_log_prob": -64.64743041992188, "reading_distances": [12, 14, 16, 15], "reading_log_probs": [-63.393310546875, -63.92219543457031, -64.91241455078125, -63.76835632324219], "reading_logprob_std": 0.6481464039712895, "tokenization_path_logprob_std": 47.799960205850944, "trajectory_balance_rmse": 567.2316004501992} {"canonical_length": 130, "canonical_nll": 1.6628040020282453, "generation_rouge_l_f1": 0.20472440944881892, "generation_token_f1": 0.4146981627296588, "prompt_id": "7c910d294599c2d9", "reading_canonical_log_prob": -216.16452026367188, "reading_distances": [14, 14, 13, 6], "reading_log_probs": [-224.3050079345703, -223.17396545410156, -222.98971557617188, -223.70838928222656], "reading_logprob_std": 0.591696264872581, "tokenization_path_logprob_std": 197.19232764436592, "trajectory_balance_rmse": 3292.187174291622} {"canonical_length": 22, "canonical_nll": 1.9134795448996804, "generation_rouge_l_f1": 0.09599999999999999, "generation_token_f1": 0.128, "prompt_id": "e55841df337a493f", "reading_canonical_log_prob": -42.09654998779297, "reading_distances": [19, 21, 21, 24], "reading_log_probs": [-40.411354064941406, -41.075904846191406, -42.612945556640625, -43.267059326171875], "reading_logprob_std": 1.323983959507845, "tokenization_path_logprob_std": 38.44548108049331, "trajectory_balance_rmse": 676.7474380437891} {"canonical_length": 440, "canonical_nll": 2.135097295587713, "generation_rouge_l_f1": 0.22889842632331903, "generation_token_f1": 0.4206008583690987, "prompt_id": "58ceb01ab8ce3495", "reading_canonical_log_prob": -939.4428100585938, "reading_distances": [80, 81, 74, 88], "reading_log_probs": [-957.9198608398438, -957.2946166992188, -958.4404296875, -975.2461547851562], "reading_logprob_std": 8.693222394260564, "tokenization_path_logprob_std": 140.59348767339412, "trajectory_balance_rmse": 9279.09852951269} {"canonical_length": 32, "canonical_nll": 1.382392406463623, "generation_rouge_l_f1": 0.21719457013574658, "generation_token_f1": 0.25339366515837103, "prompt_id": "a8059315f9c843b5", "reading_canonical_log_prob": -44.23655700683594, "reading_distances": [23, 31, 21, 20], "reading_log_probs": [-43.2631950378418, -43.13835906982422, -42.35060501098633, -41.784706115722656], "reading_logprob_std": 0.6956716422446292, "tokenization_path_logprob_std": 158.93586557312014, "trajectory_balance_rmse": 1141.1702661926267} {"canonical_length": 12, "canonical_nll": 1.5149356524149578, "generation_rouge_l_f1": 0.11965811965811965, "generation_token_f1": 0.15384615384615385, "prompt_id": "3715d1d4c18fe319", "reading_canonical_log_prob": -18.179227828979492, "reading_distances": [13, 9, 11, 19], "reading_log_probs": [-21.960710525512695, -19.873764038085938, -18.87398910522461, -20.097110748291016], "reading_logprob_std": 1.2878165992803363, "tokenization_path_logprob_std": 82.12022671071094, "trajectory_balance_rmse": 319.19958736847707} {"canonical_length": 264, "canonical_nll": 2.071935017903646, "generation_rouge_l_f1": 0.21194605009633913, "generation_token_f1": 0.43930635838150284, "prompt_id": "996c63537529bc50", "reading_canonical_log_prob": -546.9908447265625, "reading_distances": [31, 19, 25, 22], "reading_log_probs": [-536.2017822265625, -548.5693359375, -544.3031005859375, -539.829345703125], "reading_logprob_std": 5.37238026741671, "tokenization_path_logprob_std": 123.0519802310832, "trajectory_balance_rmse": 7901.313098374913} {"canonical_length": 15, "canonical_nll": 0.9697075525919596, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "a9d8f6ac4429b707", "reading_canonical_log_prob": -14.545613288879395, "reading_distances": [23, 23, 20, 23], "reading_log_probs": [-15.445356369018555, -16.72439193725586, -16.318647384643555, -15.166690826416016], "reading_logprob_std": 0.729970831161031, "tokenization_path_logprob_std": 103.10706635203208, "trajectory_balance_rmse": 310.99199328669044} {"canonical_length": 17, "canonical_nll": 0.8009519016041475, "generation_rouge_l_f1": 0.5925925925925927, "generation_token_f1": 0.5925925925925927, "prompt_id": "b1822d4ae9e3de29", "reading_canonical_log_prob": -13.616182327270508, "reading_distances": [21, 22, 25, 16], "reading_log_probs": [-13.958012580871582, -13.207449913024902, -13.583782196044922, -13.654038429260254], "reading_logprob_std": 0.30846342469872906, "tokenization_path_logprob_std": 73.1584680321179, "trajectory_balance_rmse": 569.0864520765477} {"canonical_length": 14, "canonical_nll": 0.8565895898001534, "generation_rouge_l_f1": 0.27848101265822783, "generation_token_f1": 0.3037974683544304, "prompt_id": "92e5aa90e47873e8", "reading_canonical_log_prob": -11.992254257202148, "reading_distances": [27, 17, 21, 21], "reading_log_probs": [-11.276143074035645, -12.138198852539062, -11.96121597290039, -12.64774227142334], "reading_logprob_std": 0.566864563061712, "tokenization_path_logprob_std": 68.32906550866774, "trajectory_balance_rmse": 402.3872375951377} {"canonical_length": 176, "canonical_nll": 1.995274457064542, "generation_rouge_l_f1": 0.24068767908309452, "generation_token_f1": 0.4469914040114613, "prompt_id": "d3105cc3dd2ecc61", "reading_canonical_log_prob": -351.1683044433594, "reading_distances": [31, 31, 33, 35], "reading_log_probs": [-339.019287109375, -341.5979919433594, -339.5771179199219, -346.8775634765625], "reading_logprob_std": 3.5819886703640207, "tokenization_path_logprob_std": 156.00397513613703, "trajectory_balance_rmse": 4798.02043840591} {"canonical_length": 46, "canonical_nll": 1.0185164575991423, "generation_rouge_l_f1": 0.12318840579710146, "generation_token_f1": 0.15217391304347827, "prompt_id": "f2effbcb126018c5", "reading_canonical_log_prob": -46.85175704956055, "reading_distances": [23, 26, 29, 23], "reading_log_probs": [-45.19939422607422, -47.25115966796875, -42.39845275878906, -46.95377731323242], "reading_logprob_std": 2.227125519510151, "tokenization_path_logprob_std": 33.87812117786692, "trajectory_balance_rmse": 590.3752825579356} {"canonical_length": 226, "canonical_nll": 1.4572834926368916, "generation_rouge_l_f1": 0.22970297029702968, "generation_token_f1": 0.4396039603960396, "prompt_id": "e73472036e75a1c6", "reading_canonical_log_prob": -329.3460693359375, "reading_distances": [41, 37, 54, 44], "reading_log_probs": [-329.4167785644531, -330.3470458984375, -327.8944091796875, -330.667724609375], "reading_logprob_std": 1.2435638318293052, "tokenization_path_logprob_std": 225.9355129147528, "trajectory_balance_rmse": 5388.275945149303} {"canonical_length": 8, "canonical_nll": 0.28323695063591003, "generation_rouge_l_f1": 0.7999999999999999, "generation_token_f1": 0.7999999999999999, "prompt_id": "dea67577bfdb2da1", "reading_canonical_log_prob": -2.2658956050872803, "reading_distances": [22, 26, 30, 27], "reading_log_probs": [-1.7038331031799316, -1.2059993743896484, -1.1253283023834229, -1.857305884361267], "reading_logprob_std": 0.36200360183953456, "tokenization_path_logprob_std": 24.523951740325128, "trajectory_balance_rmse": 172.69139646240183} {"canonical_length": 107, "canonical_nll": 1.5296789151485835, "generation_rouge_l_f1": 0.14749262536873156, "generation_token_f1": 0.26548672566371684, "prompt_id": "90fd3570044a7e38", "reading_canonical_log_prob": -163.67564392089844, "reading_distances": [24, 25, 24, 18], "reading_log_probs": [-164.70733642578125, -185.4206085205078, -176.9949951171875, -164.00289916992188], "reading_logprob_std": 10.324026269603086, "tokenization_path_logprob_std": 89.09517585222676, "trajectory_balance_rmse": 2716.650199351797} {"canonical_length": 222, "canonical_nll": 0.8063212901622325, "generation_rouge_l_f1": 0.24821002386634844, "generation_token_f1": 0.4391408114558472, "prompt_id": "1c7402f9ee727b52", "reading_canonical_log_prob": -179.00332641601562, "reading_distances": [78, 90, 86, 86], "reading_log_probs": [-176.89512634277344, -188.02850341796875, -184.40533447265625, -184.9373321533203], "reading_logprob_std": 4.725798874949905, "tokenization_path_logprob_std": 131.37571607367576, "trajectory_balance_rmse": 3461.11054874006} {"canonical_length": 19, "canonical_nll": 1.4254429465846012, "generation_rouge_l_f1": 0.3846153846153846, "generation_token_f1": 0.5384615384615385, "prompt_id": "219202257fbe0a6c", "reading_canonical_log_prob": -27.083415985107422, "reading_distances": [21, 22, 20, 19], "reading_log_probs": [-26.441390991210938, -25.60305404663086, -25.59428596496582, -27.57071304321289], "reading_logprob_std": 0.9342459908481476, "tokenization_path_logprob_std": 54.00009962858534, "trajectory_balance_rmse": 505.6916242228778} {"canonical_length": 47, "canonical_nll": 1.7025928903133312, "generation_rouge_l_f1": 0.3010752688172043, "generation_token_f1": 0.45161290322580644, "prompt_id": "1eb21676f35636bf", "reading_canonical_log_prob": -80.02186584472656, "reading_distances": [20, 29, 28, 23], "reading_log_probs": [-81.41698455810547, -84.60432434082031, -80.95270538330078, -84.32014465332031], "reading_logprob_std": 1.9052067893900473, "tokenization_path_logprob_std": 32.61978105356766, "trajectory_balance_rmse": 1376.9956599447225} {"canonical_length": 15, "canonical_nll": 0.1171119769414266, "generation_rouge_l_f1": 0.962962962962963, "generation_token_f1": 0.962962962962963, "prompt_id": "5f8f43706c38c575", "reading_canonical_log_prob": -1.756679654121399, "reading_distances": [29, 31, 29, 33], "reading_log_probs": [-2.7984743118286133, -3.4597623348236084, -3.4108564853668213, -3.593794584274292], "reading_logprob_std": 0.35339671824816293, "tokenization_path_logprob_std": 58.1799586205189, "trajectory_balance_rmse": 401.711959858074} {"canonical_length": 18, "canonical_nll": 0.9156099955240885, "generation_rouge_l_f1": 0.8387096774193549, "generation_token_f1": 0.8387096774193549, "prompt_id": "46d2c294deb9298b", "reading_canonical_log_prob": -16.480979919433594, "reading_distances": [73, 64, 59, 69], "reading_log_probs": [-17.232444763183594, -16.027023315429688, -13.095845222473145, -19.75585174560547], "reading_logprob_std": 2.7656182670475804, "tokenization_path_logprob_std": 44.26354279900337, "trajectory_balance_rmse": 545.5654516495725} {"canonical_length": 228, "canonical_nll": 1.3276084766053318, "generation_rouge_l_f1": 0.26423690205011385, "generation_token_f1": 0.4738041002277904, "prompt_id": "7cb804cc64427944", "reading_canonical_log_prob": -302.6947326660156, "reading_distances": [24, 19, 19, 19], "reading_log_probs": [-299.09423828125, -299.43438720703125, -298.4573059082031, -297.9095153808594], "reading_logprob_std": 0.6773077158177555, "tokenization_path_logprob_std": 262.76898161344985, "trajectory_balance_rmse": 6281.831443605048} {"canonical_length": 171, "canonical_nll": 1.6722237213313231, "generation_rouge_l_f1": 0.3176178660049628, "generation_token_f1": 0.47146401985111663, "prompt_id": "d5a4beae5409050e", "reading_canonical_log_prob": -285.95025634765625, "reading_distances": [29, 18, 28, 22], "reading_log_probs": [-289.2406005859375, -294.7537841796875, -292.447509765625, -300.8365478515625], "reading_logprob_std": 4.897621262583036, "tokenization_path_logprob_std": 473.53972287381873, "trajectory_balance_rmse": 5233.527997434481} {"canonical_length": 168, "canonical_nll": 1.9349619547526042, "generation_rouge_l_f1": 0.21395348837209302, "generation_token_f1": 0.37209302325581395, "prompt_id": "ad2f9057c92483ba", "reading_canonical_log_prob": -325.0736083984375, "reading_distances": [13, 14, 14, 11], "reading_log_probs": [-319.5107421875, -318.4609069824219, -324.65789794921875, -320.32403564453125], "reading_logprob_std": 2.722026214828955, "tokenization_path_logprob_std": 211.96069543955042, "trajectory_balance_rmse": 4678.9466216278615} {"canonical_length": 159, "canonical_nll": 1.3313802083333333, "generation_rouge_l_f1": 0.2923076923076923, "generation_token_f1": 0.5128205128205129, "prompt_id": "240077b10fcaa781", "reading_canonical_log_prob": -211.689453125, "reading_distances": [27, 25, 26, 24], "reading_log_probs": [-209.43780517578125, -211.5785369873047, -210.15737915039062, -208.37933349609375], "reading_logprob_std": 1.3427869220042117, "tokenization_path_logprob_std": 185.64561053009655, "trajectory_balance_rmse": 4796.292577318499} {"canonical_length": 327, "canonical_nll": 0.9454440895570527, "generation_rouge_l_f1": 0.1793570219966159, "generation_token_f1": 0.4230118443316413, "prompt_id": "dcea5e8dbdd12a1a", "reading_canonical_log_prob": -309.16021728515625, "reading_distances": [68, 68, 70, 69], "reading_log_probs": [-318.1174011230469, -317.7635192871094, -308.48687744140625, -317.019775390625], "reading_logprob_std": 4.596163139997967, "tokenization_path_logprob_std": 195.07313356034228, "trajectory_balance_rmse": 7090.356102693105} {"canonical_length": 4, "canonical_nll": 2.935407876968384, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "09f4cd01fd363213", "reading_canonical_log_prob": -11.741631507873535, "reading_distances": [30, 33, 33, 28], "reading_log_probs": [-16.39084243774414, -15.343008041381836, -16.29135513305664, -15.06021499633789], "reading_logprob_std": 0.6691701442004139, "tokenization_path_logprob_std": 15.327045301286033, "trajectory_balance_rmse": 98.32622080850595} {"canonical_length": 139, "canonical_nll": 1.2299899094396358, "generation_rouge_l_f1": 0.24128686327077745, "generation_token_f1": 0.42359249329758714, "prompt_id": "0eb3ed1212b155f2", "reading_canonical_log_prob": -170.96859741210938, "reading_distances": [26, 19, 22, 22], "reading_log_probs": [-174.56134033203125, -175.5375213623047, -175.56051635742188, -176.18348693847656], "reading_logprob_std": 0.6701078511853071, "tokenization_path_logprob_std": 207.03453614224043, "trajectory_balance_rmse": 3303.506198016072} {"canonical_length": 3, "canonical_nll": 7.0837656656901045, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "423d33be2ee3658e", "reading_canonical_log_prob": -21.251296997070312, "reading_distances": [22, 21, 21, 23], "reading_log_probs": [-18.31716537475586, -17.88665771484375, -18.545997619628906, -17.61270523071289], "reading_logprob_std": 0.4197994454591591, "tokenization_path_logprob_std": 14.208333923972715, "trajectory_balance_rmse": 51.85261458974941} {"canonical_length": 33, "canonical_nll": 1.4102268796978574, "generation_rouge_l_f1": 0.09688581314878893, "generation_token_f1": 0.1245674740484429, "prompt_id": "82dd428e952ac035", "reading_canonical_log_prob": -46.5374870300293, "reading_distances": [36, 31, 30, 32], "reading_log_probs": [-46.21718215942383, -48.33936309814453, -48.12522888183594, -46.49138641357422], "reading_logprob_std": 1.09353382454074, "tokenization_path_logprob_std": 59.76032413606881, "trajectory_balance_rmse": 935.2280631021367} {"canonical_length": 293, "canonical_nll": 1.2463222672914889, "generation_rouge_l_f1": 0.2898032200357782, "generation_token_f1": 0.4865831842576029, "prompt_id": "a3746b2776781f66", "reading_canonical_log_prob": -365.17242431640625, "reading_distances": [21, 25, 23, 27], "reading_log_probs": [-372.54632568359375, -374.4720458984375, -374.1075439453125, -370.6362609863281], "reading_logprob_std": 1.7485722962237789, "tokenization_path_logprob_std": 380.8106201744669, "trajectory_balance_rmse": 8716.943242384454} {"canonical_length": 64, "canonical_nll": 2.072859287261963, "generation_rouge_l_f1": 0.15547703180212014, "generation_token_f1": 0.24734982332155478, "prompt_id": "d4e9c302f1022491", "reading_canonical_log_prob": -132.66299438476562, "reading_distances": [30, 25, 32, 28], "reading_log_probs": [-138.54867553710938, -145.53341674804688, -146.31130981445312, -147.29705810546875], "reading_logprob_std": 3.981902250846394, "tokenization_path_logprob_std": 181.10158053670418, "trajectory_balance_rmse": 1769.1245091556827} {"canonical_length": 99, "canonical_nll": 1.177187447596078, "generation_rouge_l_f1": 0.26775956284153, "generation_token_f1": 0.4262295081967213, "prompt_id": "ed00a510b9a4ea44", "reading_canonical_log_prob": -116.54155731201172, "reading_distances": [30, 33, 35, 28], "reading_log_probs": [-115.0386962890625, -117.265625, -115.696533203125, -117.67210388183594], "reading_logprob_std": 1.2535599338736336, "tokenization_path_logprob_std": 345.19484862297526, "trajectory_balance_rmse": 3131.143069088714} {"canonical_length": 275, "canonical_nll": 1.430707674893466, "generation_rouge_l_f1": 0.21284403669724772, "generation_token_f1": 0.436697247706422, "prompt_id": "00a5d180fc30eed4", "reading_canonical_log_prob": -393.4446105957031, "reading_distances": [30, 25, 32, 26], "reading_log_probs": [-399.1133728027344, -396.16168212890625, -399.41644287109375, -399.691162109375], "reading_logprob_std": 1.6397239824755054, "tokenization_path_logprob_std": 234.54753453247082, "trajectory_balance_rmse": 7711.5640097930445} {"canonical_length": 115, "canonical_nll": 1.2408709982167119, "generation_rouge_l_f1": 0.32628398791540786, "generation_token_f1": 0.45317220543806647, "prompt_id": "300b5dc6095f893e", "reading_canonical_log_prob": -142.70016479492188, "reading_distances": [6, 10, 5, 6], "reading_log_probs": [-143.19204711914062, -140.52691650390625, -142.98387145996094, -142.67799377441406], "reading_logprob_std": 1.230441363253161, "tokenization_path_logprob_std": 177.43303625507735, "trajectory_balance_rmse": 3150.2984234289083} {"canonical_length": 6, "canonical_nll": 3.6754283905029297, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "2c288dbc9a356333", "reading_canonical_log_prob": -22.052570343017578, "reading_distances": [19, 21, 19, 22], "reading_log_probs": [-24.18385124206543, -24.98461151123047, -26.192235946655273, -23.970462799072266], "reading_logprob_std": 1.0059610274357416, "tokenization_path_logprob_std": 39.78363235896558, "trajectory_balance_rmse": 170.98590241097114} {"canonical_length": 233, "canonical_nll": 1.165672858897197, "generation_rouge_l_f1": 0.29166666666666663, "generation_token_f1": 0.5185185185185185, "prompt_id": "810a6641955f3c0f", "reading_canonical_log_prob": -271.6017761230469, "reading_distances": [42, 39, 33, 35], "reading_log_probs": [-276.7427062988281, -267.00152587890625, -278.146240234375, -269.3876953125], "reading_logprob_std": 5.458687738138315, "tokenization_path_logprob_std": 130.62791895426514, "trajectory_balance_rmse": 3931.392098885061} {"canonical_length": 167, "canonical_nll": 2.1765684938716316, "generation_rouge_l_f1": 0.2934782608695652, "generation_token_f1": 0.532608695652174, "prompt_id": "eb867c21f2710dc4", "reading_canonical_log_prob": -363.4869384765625, "reading_distances": [23, 31, 25, 26], "reading_log_probs": [-364.6002197265625, -360.4056396484375, -362.5015869140625, -365.98577880859375], "reading_logprob_std": 2.4424897957516554, "tokenization_path_logprob_std": 219.2212433137804, "trajectory_balance_rmse": 4998.29497119602} {"canonical_length": 48, "canonical_nll": 2.381009896596273, "generation_rouge_l_f1": 0.1732283464566929, "generation_token_f1": 0.25196850393700787, "prompt_id": "ee8008c4016a6c6e", "reading_canonical_log_prob": -114.2884750366211, "reading_distances": [24, 23, 32, 28], "reading_log_probs": [-113.26298522949219, -118.09152221679688, -115.35740661621094, -115.03607940673828], "reading_logprob_std": 1.9949854511322458, "tokenization_path_logprob_std": 83.75856241481522, "trajectory_balance_rmse": 1583.6055775823493} {"canonical_length": 232, "canonical_nll": 1.4203766132223195, "generation_rouge_l_f1": 0.24532224532224534, "generation_token_f1": 0.5530145530145529, "prompt_id": "5adec8000ecf22fe", "reading_canonical_log_prob": -329.5273742675781, "reading_distances": [22, 18, 20, 23], "reading_log_probs": [-333.56622314453125, -335.01605224609375, -332.50750732421875, -332.68194580078125], "reading_logprob_std": 1.1465931179588331, "tokenization_path_logprob_std": 286.5282178834775, "trajectory_balance_rmse": 6496.901160921815} {"canonical_length": 283, "canonical_nll": 1.5185502662254307, "generation_rouge_l_f1": 0.21153846153846154, "generation_token_f1": 0.4461538461538462, "prompt_id": "893df9640ff6dc30", "reading_canonical_log_prob": -429.7497253417969, "reading_distances": [22, 23, 24, 19], "reading_log_probs": [-433.05841064453125, -431.7928161621094, -435.52606201171875, -436.6072692871094], "reading_logprob_std": 2.2092706746746007, "tokenization_path_logprob_std": 97.08153618280208, "trajectory_balance_rmse": 7630.533122115735} {"canonical_length": 13, "canonical_nll": 0.9926600089439979, "generation_rouge_l_f1": 0.14285714285714285, "generation_token_f1": 0.14285714285714285, "prompt_id": "5f407731bdd7e46c", "reading_canonical_log_prob": -12.904580116271973, "reading_distances": [25, 31, 31, 29], "reading_log_probs": [-14.411996841430664, -15.006020545959473, -15.119232177734375, -14.28305435180664], "reading_logprob_std": 0.4187643028521924, "tokenization_path_logprob_std": 58.60204489657951, "trajectory_balance_rmse": 186.15767525538234} {"canonical_length": 226, "canonical_nll": 1.260302383287818, "generation_rouge_l_f1": 0.28750000000000003, "generation_token_f1": 0.5416666666666667, "prompt_id": "c371f7349e2c9b37", "reading_canonical_log_prob": -284.8283386230469, "reading_distances": [24, 30, 21, 19], "reading_log_probs": [-287.15325927734375, -283.12823486328125, -281.9685363769531, -285.54736328125], "reading_logprob_std": 2.339269115390513, "tokenization_path_logprob_std": 310.88265117963533, "trajectory_balance_rmse": 6762.808264827273} {"canonical_length": 9, "canonical_nll": 1.3371645609537761, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4210526315789474, "prompt_id": "59b66b10bf9d5bcd", "reading_canonical_log_prob": -12.034481048583984, "reading_distances": [64, 50, 67, 58], "reading_log_probs": [-12.221231460571289, -21.864683151245117, -28.563751220703125, -15.757144927978516], "reading_logprob_std": 7.1807932335408875, "tokenization_path_logprob_std": 20.65213379148838, "trajectory_balance_rmse": 231.5924881650986} {"canonical_length": 387, "canonical_nll": 1.414013293362403, "generation_rouge_l_f1": 0.18313953488372092, "generation_token_f1": 0.3313953488372093, "prompt_id": "d9970f787452110a", "reading_canonical_log_prob": -547.22314453125, "reading_distances": [32, 28, 31, 31], "reading_log_probs": [-543.5475463867188, -547.6737060546875, -547.4156494140625, -548.5814208984375], "reading_logprob_std": 2.2281694627460116, "tokenization_path_logprob_std": 362.2302839671818, "trajectory_balance_rmse": 9869.209681290618} {"canonical_length": 199, "canonical_nll": 1.2093666114998822, "generation_rouge_l_f1": 0.28635346756152125, "generation_token_f1": 0.5100671140939597, "prompt_id": "4f2d283e14485973", "reading_canonical_log_prob": -240.66395568847656, "reading_distances": [20, 21, 24, 20], "reading_log_probs": [-242.48809814453125, -241.2786102294922, -241.06924438476562, -237.76075744628906], "reading_logprob_std": 2.024618588952447, "tokenization_path_logprob_std": 217.3510848033781, "trajectory_balance_rmse": 5570.694850288925} {"canonical_length": 366, "canonical_nll": 1.4978841145833333, "generation_rouge_l_f1": 0.18213058419243985, "generation_token_f1": 0.37800687285223367, "prompt_id": "74c803f89db0a539", "reading_canonical_log_prob": -548.2255859375, "reading_distances": [12, 15, 17, 12], "reading_log_probs": [-539.401611328125, -537.6448974609375, -543.1738891601562, -538.1349487304688], "reading_logprob_std": 2.5020221471881965, "tokenization_path_logprob_std": 105.0806036288825, "trajectory_balance_rmse": 9175.37217685931} {"canonical_length": 132, "canonical_nll": 1.3556540518095999, "generation_rouge_l_f1": 0.3416149068322981, "generation_token_f1": 0.5838509316770186, "prompt_id": "456727f2988746cd", "reading_canonical_log_prob": -178.9463348388672, "reading_distances": [26, 29, 30, 31], "reading_log_probs": [-183.1552734375, -184.25735473632812, -184.47572326660156, -186.75802612304688], "reading_logprob_std": 1.5123917697233287, "tokenization_path_logprob_std": 239.57896469645934, "trajectory_balance_rmse": 4764.140560236306} {"canonical_length": 12, "canonical_nll": 0.10645994544029236, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "4105efabcbebe823", "reading_canonical_log_prob": -1.2775193452835083, "reading_distances": [29, 18, 24, 23], "reading_log_probs": [-5.959370136260986, -2.8726248741149902, -2.293191432952881, -3.1284356117248535], "reading_logprob_std": 1.6350807310488578, "tokenization_path_logprob_std": 32.214700195565946, "trajectory_balance_rmse": 367.89700849378795} {"canonical_length": 57, "canonical_nll": 2.0341250770970394, "generation_rouge_l_f1": 0.0948905109489051, "generation_token_f1": 0.1386861313868613, "prompt_id": "61331d091d23a9af", "reading_canonical_log_prob": -115.94512939453125, "reading_distances": [16, 16, 15, 16], "reading_log_probs": [-120.2199478149414, -121.52790069580078, -119.4844970703125, -122.36909484863281], "reading_logprob_std": 1.2933957834210448, "tokenization_path_logprob_std": 211.70787807052196, "trajectory_balance_rmse": 1719.9470567679928} {"canonical_length": 32, "canonical_nll": 3.077443838119507, "generation_rouge_l_f1": 0.1206896551724138, "generation_token_f1": 0.22413793103448282, "prompt_id": "1a1747f5e15bef6b", "reading_canonical_log_prob": -98.47820281982422, "reading_distances": [21, 38, 30, 33], "reading_log_probs": [-103.30342102050781, -105.36074829101562, -101.45887756347656, -102.26424407958984], "reading_logprob_std": 1.6876110706278418, "tokenization_path_logprob_std": 40.54039698928188, "trajectory_balance_rmse": 1260.6621791320376} {"canonical_length": 9, "canonical_nll": 0.36050301127963597, "generation_rouge_l_f1": 0.875, "generation_token_f1": 0.875, "prompt_id": "8d4efdacc1d27322", "reading_canonical_log_prob": -3.2445271015167236, "reading_distances": [33, 36, 36, 33], "reading_log_probs": [-1.875776767730713, -3.1869702339172363, -1.6548646688461304, -1.715721845626831], "reading_logprob_std": 0.7251017697438631, "tokenization_path_logprob_std": 30.330154172063946, "trajectory_balance_rmse": 349.03737604936254} {"canonical_length": 263, "canonical_nll": 1.8059140049459363, "generation_rouge_l_f1": 0.19723865877712032, "generation_token_f1": 0.37080867850098614, "prompt_id": "123b7e61f56166d9", "reading_canonical_log_prob": -474.95538330078125, "reading_distances": [34, 36, 32, 39], "reading_log_probs": [-484.71392822265625, -475.8785705566406, -478.5941162109375, -482.9557189941406], "reading_logprob_std": 4.032066683495577, "tokenization_path_logprob_std": 231.81162059936307, "trajectory_balance_rmse": 6070.654841497014} {"canonical_length": 63, "canonical_nll": 1.2202396695576017, "generation_rouge_l_f1": 0.5846153846153845, "generation_token_f1": 0.5846153846153845, "prompt_id": "0407e3355bf3ebea", "reading_canonical_log_prob": -76.8750991821289, "reading_distances": [18, 15, 20, 19], "reading_log_probs": [-83.49775695800781, -87.14634704589844, -85.6638412475586, -86.579833984375], "reading_logprob_std": 1.6036711206957086, "tokenization_path_logprob_std": 21.029892407406372, "trajectory_balance_rmse": 1511.817688403382} {"canonical_length": 29, "canonical_nll": 0.7826458503460062, "generation_rouge_l_f1": 0.36363636363636365, "generation_token_f1": 0.5, "prompt_id": "f217efc1d8ea8fee", "reading_canonical_log_prob": -22.69672966003418, "reading_distances": [10, 17, 13, 15], "reading_log_probs": [-23.66295623779297, -23.21240234375, -23.07852554321289, -23.438461303710938], "reading_logprob_std": 0.25715355762966347, "tokenization_path_logprob_std": 46.348543185376634, "trajectory_balance_rmse": 833.2460276486082} {"canonical_length": 125, "canonical_nll": 1.754979248046875, "generation_rouge_l_f1": 0.29721362229102166, "generation_token_f1": 0.5139318885448916, "prompt_id": "65c5ecab43401060", "reading_canonical_log_prob": -219.37240600585938, "reading_distances": [31, 17, 21, 32], "reading_log_probs": [-219.34820556640625, -219.16390991210938, -221.1428680419922, -221.86073303222656], "reading_logprob_std": 1.3314163173478861, "tokenization_path_logprob_std": 335.16689281293816, "trajectory_balance_rmse": 3844.995579851472} {"canonical_length": 41, "canonical_nll": 1.1165038318168827, "generation_rouge_l_f1": 0.5185185185185185, "generation_token_f1": 0.5185185185185185, "prompt_id": "5cd588db0ae30114", "reading_canonical_log_prob": -45.77665710449219, "reading_distances": [26, 32, 31, 25], "reading_log_probs": [-52.00745391845703, -50.24766540527344, -51.472206115722656, -52.1119499206543], "reading_logprob_std": 0.8553066739346015, "tokenization_path_logprob_std": 32.00014963297812, "trajectory_balance_rmse": 1081.9425831044043} {"canonical_length": 145, "canonical_nll": 2.005979761584052, "generation_rouge_l_f1": 0.23300970873786406, "generation_token_f1": 0.4983818770226537, "prompt_id": "f88ab53f7308d0a9", "reading_canonical_log_prob": -290.8670654296875, "reading_distances": [26, 29, 27, 18], "reading_log_probs": [-309.4172668457031, -308.0010986328125, -310.1339111328125, -301.864013671875], "reading_logprob_std": 3.7657961745795423, "tokenization_path_logprob_std": 262.98466668497997, "trajectory_balance_rmse": 4466.337697890425} {"canonical_length": 55, "canonical_nll": 1.4249624078924006, "generation_rouge_l_f1": 0.216, "generation_token_f1": 0.328, "prompt_id": "8c737ded95e1b468", "reading_canonical_log_prob": -78.37293243408203, "reading_distances": [12, 20, 18, 16], "reading_log_probs": [-95.01317596435547, -91.67655944824219, -93.13113403320312, -96.26190948486328], "reading_logprob_std": 2.024380648530323, "tokenization_path_logprob_std": 115.40873077287688, "trajectory_balance_rmse": 1388.5644202892113} {"canonical_length": 22, "canonical_nll": 0.6296345103870739, "generation_rouge_l_f1": 0.1069182389937107, "generation_token_f1": 0.11320754716981132, "prompt_id": "cd208c8b818d2286", "reading_canonical_log_prob": -13.851959228515625, "reading_distances": [24, 27, 28, 30], "reading_log_probs": [-9.897745132446289, -19.234535217285156, -18.19893455505371, -20.19504165649414], "reading_logprob_std": 4.726690134529656, "tokenization_path_logprob_std": 61.38840356614284, "trajectory_balance_rmse": 417.615291756824} {"canonical_length": 165, "canonical_nll": 1.0812738591974431, "generation_rouge_l_f1": 0.3020833333333333, "generation_token_f1": 0.515625, "prompt_id": "a88030476f29d6bf", "reading_canonical_log_prob": -178.41018676757812, "reading_distances": [14, 12, 16, 17], "reading_log_probs": [-181.21875, -178.9310302734375, -181.5421142578125, -181.3018798828125], "reading_logprob_std": 1.2193419183660472, "tokenization_path_logprob_std": 263.13820721193724, "trajectory_balance_rmse": 5378.295465858275} {"canonical_length": 377, "canonical_nll": 1.3593851995088693, "generation_rouge_l_f1": 0.25391849529780564, "generation_token_f1": 0.48902821316614425, "prompt_id": "842add8204336566", "reading_canonical_log_prob": -512.4882202148438, "reading_distances": [35, 31, 24, 24], "reading_log_probs": [-519.7718505859375, -518.852294921875, -514.6595458984375, -523.8914794921875], "reading_logprob_std": 3.7876301688190197, "tokenization_path_logprob_std": 124.76369891000702, "trajectory_balance_rmse": 9931.994805864768} {"canonical_length": 112, "canonical_nll": 1.2509448187691825, "generation_rouge_l_f1": 0.27893175074183973, "generation_token_f1": 0.43323442136498513, "prompt_id": "95ed700077e4b7ce", "reading_canonical_log_prob": -140.10581970214844, "reading_distances": [12, 13, 12, 14], "reading_log_probs": [-135.76141357421875, -133.63555908203125, -133.84945678710938, -135.69866943359375], "reading_logprob_std": 1.151105876530258, "tokenization_path_logprob_std": 257.99464981346597, "trajectory_balance_rmse": 2139.005008282121} {"canonical_length": 11, "canonical_nll": 0.6299553350968794, "generation_rouge_l_f1": 0.7272727272727272, "generation_token_f1": 0.7272727272727272, "prompt_id": "e5cef34a838082a8", "reading_canonical_log_prob": -6.929508686065674, "reading_distances": [35, 36, 30, 38], "reading_log_probs": [-6.315258502960205, -7.042255401611328, -6.325786113739014, -6.356428146362305], "reading_logprob_std": 0.35531165421416483, "tokenization_path_logprob_std": 36.117343662915225, "trajectory_balance_rmse": 473.2954727830365} {"canonical_length": 10, "canonical_nll": 0.3206679105758667, "generation_rouge_l_f1": 0.8888888888888888, "generation_token_f1": 0.8888888888888888, "prompt_id": "47116aef05367776", "reading_canonical_log_prob": -3.206679105758667, "reading_distances": [32, 37, 32, 38], "reading_log_probs": [-12.569576263427734, -11.412885665893555, -14.797836303710938, -9.696256637573242], "reading_logprob_std": 2.140674791375166, "tokenization_path_logprob_std": 39.478847605766916, "trajectory_balance_rmse": 199.7924904512462} {"canonical_length": 375, "canonical_nll": 1.50410498046875, "generation_rouge_l_f1": 0.17896389324960751, "generation_token_f1": 0.3704866562009419, "prompt_id": "7dea199ced8c5f4e", "reading_canonical_log_prob": -564.0393676757812, "reading_distances": [32, 26, 28, 27], "reading_log_probs": [-560.05029296875, -561.84619140625, -567.6849365234375, -568.3648681640625], "reading_logprob_std": 4.1602443070860335, "tokenization_path_logprob_std": 103.34471397500157, "trajectory_balance_rmse": 9292.4030605408} {"canonical_length": 377, "canonical_nll": 1.3072685423833306, "generation_rouge_l_f1": 0.2694136291600634, "generation_token_f1": 0.49762282091917587, "prompt_id": "c009e5e48dec12c7", "reading_canonical_log_prob": -492.8402404785156, "reading_distances": [27, 29, 32, 26], "reading_log_probs": [-484.7891540527344, -487.6924133300781, -485.26922607421875, -485.3719177246094], "reading_logprob_std": 1.299553736915449, "tokenization_path_logprob_std": 403.2254684512533, "trajectory_balance_rmse": 8775.643225879066} {"canonical_length": 206, "canonical_nll": 1.1296502270744841, "generation_rouge_l_f1": 0.2602739726027397, "generation_token_f1": 0.5068493150684931, "prompt_id": "12ce96bbad2517ca", "reading_canonical_log_prob": -232.70794677734375, "reading_distances": [21, 23, 23, 20], "reading_log_probs": [-236.86843872070312, -237.28228759765625, -239.62496948242188, -237.86312866210938], "reading_logprob_std": 1.2141100528190518, "tokenization_path_logprob_std": 155.65065173593354, "trajectory_balance_rmse": 6024.422056954285} {"canonical_length": 193, "canonical_nll": 1.40306636830068, "generation_rouge_l_f1": 0.3141592920353983, "generation_token_f1": 0.6150442477876105, "prompt_id": "cda8b27fc8b243fa", "reading_canonical_log_prob": -270.79180908203125, "reading_distances": [22, 12, 18, 13], "reading_log_probs": [-270.8585510253906, -274.0378112792969, -270.32391357421875, -267.97735595703125], "reading_logprob_std": 2.4953841640774965, "tokenization_path_logprob_std": 265.50185369914124, "trajectory_balance_rmse": 5280.082889658781} {"canonical_length": 247, "canonical_nll": 1.0051842199163399, "generation_rouge_l_f1": 0.3295019157088123, "generation_token_f1": 0.5402298850574714, "prompt_id": "1ba77d2f5f7df1e4", "reading_canonical_log_prob": -248.28050231933594, "reading_distances": [17, 17, 17, 19], "reading_log_probs": [-247.96536254882812, -249.7113037109375, -250.01593017578125, -250.31988525390625], "reading_logprob_std": 1.0548488347478917, "tokenization_path_logprob_std": 169.44697030549037, "trajectory_balance_rmse": 7577.01815660589} {"canonical_length": 10, "canonical_nll": 0.9275984764099121, "generation_rouge_l_f1": 0.39999999999999997, "generation_token_f1": 0.5333333333333333, "prompt_id": "78c9a1c087a98da8", "reading_canonical_log_prob": -9.275984764099121, "reading_distances": [16, 21, 21, 21], "reading_log_probs": [-10.102370262145996, -9.658797264099121, -10.411859512329102, -10.659429550170898], "reading_logprob_std": 0.4313268575208189, "tokenization_path_logprob_std": 69.57584610649266, "trajectory_balance_rmse": 246.66135891154502} {"canonical_length": 128, "canonical_nll": 1.1855747699737549, "generation_rouge_l_f1": 0.22012578616352202, "generation_token_f1": 0.42138364779874216, "prompt_id": "6de4ae8779beb5f7", "reading_canonical_log_prob": -151.75357055664062, "reading_distances": [38, 40, 33, 35], "reading_log_probs": [-151.04798889160156, -150.93838500976562, -150.23779296875, -149.6863555908203], "reading_logprob_std": 0.6380285877694005, "tokenization_path_logprob_std": 273.21302818778196, "trajectory_balance_rmse": 3528.074763352785} {"canonical_length": 292, "canonical_nll": 0.9923994508508134, "generation_rouge_l_f1": 0.3672014260249554, "generation_token_f1": 0.5525846702317292, "prompt_id": "bcb2851b6b713edf", "reading_canonical_log_prob": -289.7806396484375, "reading_distances": [46, 47, 43, 50], "reading_log_probs": [-294.9302978515625, -296.17919921875, -297.88250732421875, -297.0089111328125], "reading_logprob_std": 1.2566085056122722, "tokenization_path_logprob_std": 397.28465518128723, "trajectory_balance_rmse": 6891.539734353869} {"canonical_length": 207, "canonical_nll": 1.6418530745206823, "generation_rouge_l_f1": 0.31304347826086953, "generation_token_f1": 0.5260869565217391, "prompt_id": "790529293961a686", "reading_canonical_log_prob": -339.86358642578125, "reading_distances": [30, 26, 26, 35], "reading_log_probs": [-340.9595947265625, -340.85565185546875, -339.768310546875, -337.212646484375], "reading_logprob_std": 1.7429572882126518, "tokenization_path_logprob_std": 163.29257185741378, "trajectory_balance_rmse": 6266.171908288409} {"canonical_length": 130, "canonical_nll": 1.8612726064828726, "generation_rouge_l_f1": 0.22282608695652173, "generation_token_f1": 0.3532608695652174, "prompt_id": "ecb22237f8778fd5", "reading_canonical_log_prob": -241.96543884277344, "reading_distances": [19, 21, 20, 14], "reading_log_probs": [-249.3092498779297, -246.49105834960938, -247.80368041992188, -244.20489501953125], "reading_logprob_std": 2.1634108338641274, "tokenization_path_logprob_std": 229.02245308724494, "trajectory_balance_rmse": 3379.3177890382226} {"canonical_length": 37, "canonical_nll": 1.266657442659945, "generation_rouge_l_f1": 0.12371134020618557, "generation_token_f1": 0.1649484536082474, "prompt_id": "db3b67bebed71ebb", "reading_canonical_log_prob": -46.86632537841797, "reading_distances": [29, 33, 32, 29], "reading_log_probs": [-42.125816345214844, -46.19251251220703, -44.214439392089844, -43.07139205932617], "reading_logprob_std": 1.750123822584729, "tokenization_path_logprob_std": 51.30997114704466, "trajectory_balance_rmse": 1023.2421217926131} {"canonical_length": 10, "canonical_nll": 0.13049669265747071, "generation_rouge_l_f1": 0.9333333333333333, "generation_token_f1": 0.9333333333333333, "prompt_id": "a75ceffa04e21449", "reading_canonical_log_prob": -1.304966926574707, "reading_distances": [38, 37, 35, 32], "reading_log_probs": [-1.8451735973358154, -1.3764138221740723, -1.691650629043579, -2.1927266120910645], "reading_logprob_std": 0.33922903802135895, "tokenization_path_logprob_std": 33.908502446298435, "trajectory_balance_rmse": 167.0209803237609} {"canonical_length": 7, "canonical_nll": 0.10825840064457484, "generation_rouge_l_f1": 0.8, "generation_token_f1": 0.8, "prompt_id": "7af908712adcd78e", "reading_canonical_log_prob": -0.7578088045120239, "reading_distances": [31, 36, 38, 38], "reading_log_probs": [-0.290282279253006, -9.784805297851562, -10.908563613891602, -0.40080907940864563], "reading_logprob_std": 5.792532586322902, "tokenization_path_logprob_std": 86.64851761912901, "trajectory_balance_rmse": 248.96836146012097} {"canonical_length": 45, "canonical_nll": 0.7659735785590278, "generation_rouge_l_f1": 0.15384615384615385, "generation_token_f1": 0.15384615384615385, "prompt_id": "3c24680236bfa43e", "reading_canonical_log_prob": -34.46881103515625, "reading_distances": [26, 20, 23, 28], "reading_log_probs": [-35.60588455200195, -33.45554733276367, -35.462799072265625, -37.253379821777344], "reading_logprob_std": 1.5550304084847888, "tokenization_path_logprob_std": 166.07196610507046, "trajectory_balance_rmse": 958.613402290784} {"canonical_length": 22, "canonical_nll": 0.39250789989124646, "generation_rouge_l_f1": 0.1875, "generation_token_f1": 0.1875, "prompt_id": "8be7f076246178b6", "reading_canonical_log_prob": -8.635173797607422, "reading_distances": [28, 32, 38, 35], "reading_log_probs": [-8.64622688293457, -8.468868255615234, -8.019993782043457, -8.159950256347656], "reading_logprob_std": 0.28527696701283894, "tokenization_path_logprob_std": 109.24910016230561, "trajectory_balance_rmse": 652.3225056268435} {"canonical_length": 20, "canonical_nll": 1.5561473846435547, "generation_rouge_l_f1": 0.2758620689655172, "generation_token_f1": 0.3448275862068965, "prompt_id": "ba65e8996b196d2c", "reading_canonical_log_prob": -31.122947692871094, "reading_distances": [39, 40, 26, 40], "reading_log_probs": [-34.12565994262695, -32.232295989990234, -31.796123504638672, -31.746578216552734], "reading_logprob_std": 1.1217627445372869, "tokenization_path_logprob_std": 88.48499471299824, "trajectory_balance_rmse": 414.5237340807689} {"canonical_length": 8, "canonical_nll": 1.0732849836349487, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "e026b81849f6617b", "reading_canonical_log_prob": -8.58627986907959, "reading_distances": [29, 25, 43, 28], "reading_log_probs": [-11.11316967010498, -11.09714412689209, -9.823235511779785, -12.019441604614258], "reading_logprob_std": 0.9028802416619212, "tokenization_path_logprob_std": 84.17809630738773, "trajectory_balance_rmse": 291.1351949327413} {"canonical_length": 84, "canonical_nll": 2.9691994076683406, "generation_rouge_l_f1": 0.46846846846846846, "generation_token_f1": 0.48648648648648646, "prompt_id": "1c6873b5ff06ab16", "reading_canonical_log_prob": -249.41275024414062, "reading_distances": [22, 21, 23, 21], "reading_log_probs": [-251.952392578125, -263.1724853515625, -248.57142639160156, -250.7526092529297], "reading_logprob_std": 6.525352467865712, "tokenization_path_logprob_std": 85.52573956648052, "trajectory_balance_rmse": 2629.631272525637} {"canonical_length": 212, "canonical_nll": 1.014828088148585, "generation_rouge_l_f1": 0.31162790697674414, "generation_token_f1": 0.5023255813953488, "prompt_id": "9cabb6227a1b5412", "reading_canonical_log_prob": -215.1435546875, "reading_distances": [16, 14, 17, 20], "reading_log_probs": [-215.06814575195312, -213.8406982421875, -218.55470275878906, -213.43080139160156], "reading_logprob_std": 2.32715532601586, "tokenization_path_logprob_std": 197.48665048615447, "trajectory_balance_rmse": 4667.9140420650765} {"canonical_length": 123, "canonical_nll": 1.4470574603817328, "generation_rouge_l_f1": 0.21409921671018275, "generation_token_f1": 0.42297650130548303, "prompt_id": "3aaeb505baa59e0c", "reading_canonical_log_prob": -177.98806762695312, "reading_distances": [16, 18, 18, 19], "reading_log_probs": [-180.08575439453125, -177.3660888671875, -180.16311645507812, -179.89581298828125], "reading_logprob_std": 1.3457637796207236, "tokenization_path_logprob_std": 292.55179384417704, "trajectory_balance_rmse": 3651.581916145256} {"canonical_length": 305, "canonical_nll": 1.1663656266009221, "generation_rouge_l_f1": 0.2024432809773124, "generation_token_f1": 0.4223385689354276, "prompt_id": "5964386ba6fbd7d7", "reading_canonical_log_prob": -355.74151611328125, "reading_distances": [30, 32, 37, 40], "reading_log_probs": [-353.42364501953125, -346.7601318359375, -345.26019287109375, -345.5675354003906], "reading_logprob_std": 3.8354636819029744, "tokenization_path_logprob_std": 216.8628401036133, "trajectory_balance_rmse": 8007.3736133025595} {"canonical_length": 125, "canonical_nll": 1.049992431640625, "generation_rouge_l_f1": 0.2185792349726776, "generation_token_f1": 0.4043715846994535, "prompt_id": "a562f2a21ae863a6", "reading_canonical_log_prob": -131.24905395507812, "reading_distances": [10, 8, 8, 7], "reading_log_probs": [-135.53729248046875, -132.9274139404297, -132.8172607421875, -131.8525390625], "reading_logprob_std": 1.5781218809023314, "tokenization_path_logprob_std": 135.39428383037585, "trajectory_balance_rmse": 3272.2505124347676} {"canonical_length": 166, "canonical_nll": 1.4614805658179593, "generation_rouge_l_f1": 0.17391304347826086, "generation_token_f1": 0.30917874396135264, "prompt_id": "f84f62888219bcaf", "reading_canonical_log_prob": -242.60577392578125, "reading_distances": [87, 91, 90, 94], "reading_log_probs": [-240.87362670898438, -241.25111389160156, -245.01602172851562, -241.73211669921875], "reading_logprob_std": 1.8980003753302237, "tokenization_path_logprob_std": 133.9461699194771, "trajectory_balance_rmse": 3549.7978566250717} {"canonical_length": 86, "canonical_nll": 0.6332320279853288, "generation_rouge_l_f1": 0.4124513618677043, "generation_token_f1": 0.5058365758754864, "prompt_id": "6b3af0ffe0ba6c9d", "reading_canonical_log_prob": -54.45795440673828, "reading_distances": [32, 26, 25, 21], "reading_log_probs": [-54.07251739501953, -54.27970504760742, -56.415199279785156, -54.31601333618164], "reading_logprob_std": 1.101461564742595, "tokenization_path_logprob_std": 214.30158878370653, "trajectory_balance_rmse": 2582.859039968606} {"canonical_length": 61, "canonical_nll": 1.383272765112705, "generation_rouge_l_f1": 0.4324324324324324, "generation_token_f1": 0.4324324324324324, "prompt_id": "53e2d36ebe126c1f", "reading_canonical_log_prob": -84.379638671875, "reading_distances": [22, 21, 23, 24], "reading_log_probs": [-90.73609161376953, -90.44648742675781, -91.47529602050781, -91.34825134277344], "reading_logprob_std": 0.49098531095071707, "tokenization_path_logprob_std": 90.67263036399353, "trajectory_balance_rmse": 1784.2249404051279} {"canonical_length": 133, "canonical_nll": 0.8803323731386572, "generation_rouge_l_f1": 0.3885350318471337, "generation_token_f1": 0.573248407643312, "prompt_id": "bce0ef5bdff3466e", "reading_canonical_log_prob": -117.0842056274414, "reading_distances": [23, 21, 18, 19], "reading_log_probs": [-125.09833526611328, -127.75942993164062, -123.17415618896484, -121.84188079833984], "reading_logprob_std": 2.569135172213944, "tokenization_path_logprob_std": 108.1064032470185, "trajectory_balance_rmse": 3268.5901140434407} {"canonical_length": 74, "canonical_nll": 2.66918759732633, "generation_rouge_l_f1": 0.17094017094017094, "generation_token_f1": 0.24786324786324784, "prompt_id": "2c030e502077ba6b", "reading_canonical_log_prob": -197.51988220214844, "reading_distances": [32, 25, 29, 23], "reading_log_probs": [-199.66351318359375, -203.47703552246094, -200.3975067138672, -204.53872680664062], "reading_logprob_std": 2.356017064958652, "tokenization_path_logprob_std": 59.43481600549604, "trajectory_balance_rmse": 1594.4146174676002} {"canonical_length": 152, "canonical_nll": 1.5352287292480469, "generation_rouge_l_f1": 0.18004866180048662, "generation_token_f1": 0.3260340632603407, "prompt_id": "b73bca8ae43941f1", "reading_canonical_log_prob": -233.35476684570312, "reading_distances": [18, 23, 21, 12], "reading_log_probs": [-236.27297973632812, -235.74383544921875, -238.20831298828125, -237.90151977539062], "reading_logprob_std": 1.2076500734537, "tokenization_path_logprob_std": 100.10763374544108, "trajectory_balance_rmse": 4194.834370368963} {"canonical_length": 139, "canonical_nll": 1.5478928380732915, "generation_rouge_l_f1": 0.19950124688279303, "generation_token_f1": 0.4189526184538653, "prompt_id": "98eb458a641b4b4a", "reading_canonical_log_prob": -215.1571044921875, "reading_distances": [37, 38, 30, 23], "reading_log_probs": [-218.62399291992188, -220.988037109375, -220.09854125976562, -221.4541473388672], "reading_logprob_std": 1.2456223754113642, "tokenization_path_logprob_std": 253.1609604108509, "trajectory_balance_rmse": 4676.171720626664} {"canonical_length": 79, "canonical_nll": 2.073082259938687, "generation_rouge_l_f1": 0.20216606498194947, "generation_token_f1": 0.27436823104693137, "prompt_id": "3cfac8c1932346e3", "reading_canonical_log_prob": -163.77349853515625, "reading_distances": [22, 28, 26, 21], "reading_log_probs": [-162.50119018554688, -164.54605102539062, -165.5926971435547, -163.78863525390625], "reading_logprob_std": 1.301286926438053, "tokenization_path_logprob_std": 123.12961226058673, "trajectory_balance_rmse": 2275.591390930745} {"canonical_length": 22, "canonical_nll": 1.4249724474820225, "generation_rouge_l_f1": 0.07692307692307691, "generation_token_f1": 0.07692307692307691, "prompt_id": "7f70df4099ae17f5", "reading_canonical_log_prob": -31.349393844604492, "reading_distances": [28, 29, 27, 31], "reading_log_probs": [-36.915985107421875, -38.771018981933594, -38.497859954833984, -39.58245849609375], "reading_logprob_std": 1.116652847273404, "tokenization_path_logprob_std": 84.92154888653853, "trajectory_balance_rmse": 562.6045670329962} {"canonical_length": 87, "canonical_nll": 0.683611902697333, "generation_rouge_l_f1": 0.3473053892215569, "generation_token_f1": 0.4730538922155689, "prompt_id": "0f424e2f0c95df2d", "reading_canonical_log_prob": -59.47423553466797, "reading_distances": [21, 22, 21, 18], "reading_log_probs": [-58.61553955078125, -56.75695037841797, -60.29203796386719, -60.695213317871094], "reading_logprob_std": 1.7972050801492365, "tokenization_path_logprob_std": 110.7525717672239, "trajectory_balance_rmse": 2099.3370380606225} {"canonical_length": 222, "canonical_nll": 1.2131058976456925, "generation_rouge_l_f1": 0.2971887550200803, "generation_token_f1": 0.5261044176706827, "prompt_id": "7d4605b87d9b1016", "reading_canonical_log_prob": -269.30950927734375, "reading_distances": [26, 19, 24, 24], "reading_log_probs": [-268.2469482421875, -267.7900390625, -268.0038757324219, -269.136962890625], "reading_logprob_std": 0.5918750393297281, "tokenization_path_logprob_std": 368.2413022452973, "trajectory_balance_rmse": 6981.530074986701} {"canonical_length": 68, "canonical_nll": 2.399687823127298, "generation_rouge_l_f1": 0.13043478260869565, "generation_token_f1": 0.2318840579710145, "prompt_id": "7bba6cec3efa18f3", "reading_canonical_log_prob": -163.17877197265625, "reading_distances": [16, 18, 20, 17], "reading_log_probs": [-168.2213134765625, -167.54351806640625, -170.43618774414062, -167.0116729736328], "reading_logprob_std": 1.505711439435244, "tokenization_path_logprob_std": 169.27116702014726, "trajectory_balance_rmse": 2008.2792555455626} {"canonical_length": 10, "canonical_nll": 0.7362863063812256, "generation_rouge_l_f1": 0.6250000000000001, "generation_token_f1": 0.75, "prompt_id": "23ac4432bd5b4331", "reading_canonical_log_prob": -7.362863063812256, "reading_distances": [23, 18, 24, 23], "reading_log_probs": [-8.42819595336914, -8.459797859191895, -7.772502899169922, -7.0465617179870605], "reading_logprob_std": 0.6668612157946145, "tokenization_path_logprob_std": 74.64852302467665, "trajectory_balance_rmse": 294.3105582875805} {"canonical_length": 35, "canonical_nll": 0.912080819266183, "generation_rouge_l_f1": 0.1422924901185771, "generation_token_f1": 0.20553359683794464, "prompt_id": "f25ec758946024eb", "reading_canonical_log_prob": -31.922828674316406, "reading_distances": [37, 39, 42, 43], "reading_log_probs": [-37.25108337402344, -38.705474853515625, -36.6727409362793, -36.797271728515625], "reading_logprob_std": 0.9329359491801289, "tokenization_path_logprob_std": 49.54336573825564, "trajectory_balance_rmse": 538.2766025491999} {"canonical_length": 225, "canonical_nll": 1.5711433919270834, "generation_rouge_l_f1": 0.24737945492662475, "generation_token_f1": 0.5618448637316563, "prompt_id": "e70426e74522ced9", "reading_canonical_log_prob": -353.50726318359375, "reading_distances": [29, 30, 30, 31], "reading_log_probs": [-360.77197265625, -369.8587646484375, -364.7940673828125, -366.32977294921875], "reading_logprob_std": 3.764964709745228, "tokenization_path_logprob_std": 372.3666285909537, "trajectory_balance_rmse": 5711.252959710832} {"canonical_length": 17, "canonical_nll": 0.7942539663875804, "generation_rouge_l_f1": 0.5185185185185185, "generation_token_f1": 0.5185185185185185, "prompt_id": "fa1698c1de8ba5af", "reading_canonical_log_prob": -13.502317428588867, "reading_distances": [22, 23, 21, 22], "reading_log_probs": [-19.93519401550293, -21.71040153503418, -21.190269470214844, -20.427141189575195], "reading_logprob_std": 0.788894131484651, "tokenization_path_logprob_std": 54.24507482531951, "trajectory_balance_rmse": 447.64380590083545} {"canonical_length": 41, "canonical_nll": 1.5821873734637004, "generation_rouge_l_f1": 0.49230769230769234, "generation_token_f1": 0.5846153846153848, "prompt_id": "2ce41f448f87f273", "reading_canonical_log_prob": -64.86968231201172, "reading_distances": [38, 37, 28, 41], "reading_log_probs": [-58.31409454345703, -60.75841522216797, -59.56476974487305, -58.618865966796875], "reading_logprob_std": 1.100339802034652, "tokenization_path_logprob_std": 102.07758551064835, "trajectory_balance_rmse": 1067.2840399822157} {"canonical_length": 17, "canonical_nll": 0.1054539540234734, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "f7f7943a60f6d582", "reading_canonical_log_prob": -1.7927172183990479, "reading_distances": [33, 25, 33, 33], "reading_log_probs": [-1.30794095993042, -2.5045528411865234, -3.11608624458313, -1.858139157295227], "reading_logprob_std": 0.784125949160723, "tokenization_path_logprob_std": 18.68752814444111, "trajectory_balance_rmse": 453.5039678913013} {"canonical_length": 45, "canonical_nll": 2.4744216918945314, "generation_rouge_l_f1": 0.1035856573705179, "generation_token_f1": 0.13545816733067728, "prompt_id": "d6cc1411da921bc9", "reading_canonical_log_prob": -111.3489761352539, "reading_distances": [17, 21, 19, 19], "reading_log_probs": [-114.87519073486328, -113.25053405761719, -115.140869140625, -114.36308288574219], "reading_logprob_std": 0.8360837210190604, "tokenization_path_logprob_std": 19.703676374229463, "trajectory_balance_rmse": 1389.2613201733516} {"canonical_length": 19, "canonical_nll": 0.19394771676314504, "generation_rouge_l_f1": 0.9375, "generation_token_f1": 0.9375, "prompt_id": "9f836d8d503579f9", "reading_canonical_log_prob": -3.685006618499756, "reading_distances": [34, 42, 46, 38], "reading_log_probs": [-3.747847318649292, -5.4356279373168945, -4.994923114776611, -4.042693138122559], "reading_logprob_std": 0.7922523555846358, "tokenization_path_logprob_std": 83.1398332951237, "trajectory_balance_rmse": 429.04887842359796} {"canonical_length": 193, "canonical_nll": 1.5287024937763116, "generation_rouge_l_f1": 0.24945295404814002, "generation_token_f1": 0.4551422319474836, "prompt_id": "d3c6858742c763af", "reading_canonical_log_prob": -295.0395812988281, "reading_distances": [35, 34, 29, 42], "reading_log_probs": [-300.1994323730469, -300.27593994140625, -295.9664306640625, -297.1372375488281], "reading_logprob_std": 2.181270763031172, "tokenization_path_logprob_std": 108.6032730939995, "trajectory_balance_rmse": 6431.484323513426} {"canonical_length": 19, "canonical_nll": 0.6437279550652755, "generation_rouge_l_f1": 0.6976744186046512, "generation_token_f1": 0.6976744186046512, "prompt_id": "923c3d05ff40a1cb", "reading_canonical_log_prob": -12.230831146240234, "reading_distances": [31, 34, 27, 31], "reading_log_probs": [-14.108118057250977, -13.136428833007812, -14.235498428344727, -15.36581039428711], "reading_logprob_std": 0.9127749350017036, "tokenization_path_logprob_std": 17.06704206027932, "trajectory_balance_rmse": 732.4062618866632} {"canonical_length": 200, "canonical_nll": 1.536587677001953, "generation_rouge_l_f1": 0.29321663019693656, "generation_token_f1": 0.5733041575492341, "prompt_id": "180d045694c4d4f7", "reading_canonical_log_prob": -307.3175354003906, "reading_distances": [12, 11, 15, 8], "reading_log_probs": [-307.9620666503906, -307.0052490234375, -306.4001159667969, -307.4765625], "reading_logprob_std": 0.6669560041429143, "tokenization_path_logprob_std": 388.41331521901054, "trajectory_balance_rmse": 5416.312898013742} {"canonical_length": 314, "canonical_nll": 1.014468830861863, "generation_rouge_l_f1": 0.2547332185886403, "generation_token_f1": 0.523235800344234, "prompt_id": "1350704a8f43bb33", "reading_canonical_log_prob": -318.543212890625, "reading_distances": [33, 45, 36, 44], "reading_log_probs": [-318.7496337890625, -321.668701171875, -320.000732421875, -321.32391357421875], "reading_logprob_std": 1.334320213538249, "tokenization_path_logprob_std": 563.6357835817586, "trajectory_balance_rmse": 9054.957710607669} {"canonical_length": 79, "canonical_nll": 0.7142282316956339, "generation_rouge_l_f1": 0.2545454545454546, "generation_token_f1": 0.4303030303030303, "prompt_id": "3269989441292641", "reading_canonical_log_prob": -56.42403030395508, "reading_distances": [29, 31, 32, 34], "reading_log_probs": [-59.39371871948242, -59.64634323120117, -60.04729461669922, -60.43230056762695], "reading_logprob_std": 0.45610245405165173, "tokenization_path_logprob_std": 72.60091033120484, "trajectory_balance_rmse": 2435.109122444765} {"canonical_length": 204, "canonical_nll": 1.237129435819738, "generation_rouge_l_f1": 0.22174840085287847, "generation_token_f1": 0.5415778251599147, "prompt_id": "0f5981f2ed369cdc", "reading_canonical_log_prob": -252.37440490722656, "reading_distances": [28, 42, 38, 33], "reading_log_probs": [-263.25738525390625, -262.88726806640625, -258.41302490234375, -264.02288818359375], "reading_logprob_std": 2.532618204870074, "tokenization_path_logprob_std": 399.50574454989845, "trajectory_balance_rmse": 6853.914861125287} {"canonical_length": 53, "canonical_nll": 1.1081688359098614, "generation_rouge_l_f1": 0.2824858757062147, "generation_token_f1": 0.33898305084745767, "prompt_id": "801510c180dde0a4", "reading_canonical_log_prob": -58.732948303222656, "reading_distances": [34, 36, 39, 36], "reading_log_probs": [-56.00984191894531, -55.685791015625, -55.17462158203125, -55.601715087890625], "reading_logprob_std": 0.3439890167307302, "tokenization_path_logprob_std": 125.00029297240017, "trajectory_balance_rmse": 1681.9299768723288} {"canonical_length": 18, "canonical_nll": 2.5391693115234375, "generation_rouge_l_f1": 0.1572052401746725, "generation_token_f1": 0.17467248908296942, "prompt_id": "a62ba8086bae83ba", "reading_canonical_log_prob": -45.705047607421875, "reading_distances": [21, 20, 17, 18], "reading_log_probs": [-48.53256607055664, -50.095306396484375, -50.97052001953125, -51.75661087036133], "reading_logprob_std": 1.3821514606281529, "tokenization_path_logprob_std": 42.619751874544505, "trajectory_balance_rmse": 456.22934794396934} {"canonical_length": 278, "canonical_nll": 1.1641935719002923, "generation_rouge_l_f1": 0.17174515235457063, "generation_token_f1": 0.31578947368421056, "prompt_id": "6197952fa284f196", "reading_canonical_log_prob": -323.64581298828125, "reading_distances": [20, 24, 23, 23], "reading_log_probs": [-321.52252197265625, -323.63763427734375, -329.35711669921875, -324.6270751953125], "reading_logprob_std": 3.311062146252951, "tokenization_path_logprob_std": 149.94505402867972, "trajectory_balance_rmse": 7549.031435365946} {"canonical_length": 13, "canonical_nll": 1.3003375713641827, "generation_rouge_l_f1": 0.3636363636363636, "generation_token_f1": 0.5454545454545454, "prompt_id": "d2ae9ca225f37e77", "reading_canonical_log_prob": -16.904388427734375, "reading_distances": [24, 28, 33, 31], "reading_log_probs": [-21.306852340698242, -17.565654754638672, -23.412059783935547, -21.436389923095703], "reading_logprob_std": 2.4411669281343347, "tokenization_path_logprob_std": 26.674235417687953, "trajectory_balance_rmse": 331.3471417851219} {"canonical_length": 14, "canonical_nll": 2.1659882409232005, "generation_rouge_l_f1": 0.4166666666666667, "generation_token_f1": 0.4166666666666667, "prompt_id": "3d0956e900302c45", "reading_canonical_log_prob": -30.323835372924805, "reading_distances": [19, 18, 20, 17], "reading_log_probs": [-32.709068298339844, -37.68927001953125, -39.096675872802734, -35.48826217651367], "reading_logprob_std": 2.7864812710356084, "tokenization_path_logprob_std": 34.78513036039553, "trajectory_balance_rmse": 325.3007341276934} {"canonical_length": 13, "canonical_nll": 0.3568740624647874, "generation_rouge_l_f1": 0.09230769230769231, "generation_token_f1": 0.09230769230769231, "prompt_id": "360b4e609ec3ff2b", "reading_canonical_log_prob": -4.639362812042236, "reading_distances": [22, 16, 20, 23], "reading_log_probs": [-5.299400329589844, -5.251311779022217, -5.159611701965332, -5.671791076660156], "reading_logprob_std": 0.2251051038495758, "tokenization_path_logprob_std": 23.436107427588524, "trajectory_balance_rmse": 150.62148638738222} {"canonical_length": 86, "canonical_nll": 2.03274944216706, "generation_rouge_l_f1": 0.32499999999999996, "generation_token_f1": 0.5, "prompt_id": "1ad837078d8b4fb2", "reading_canonical_log_prob": -174.8164520263672, "reading_distances": [31, 36, 32, 30], "reading_log_probs": [-172.3393096923828, -170.4192352294922, -173.61936950683594, -172.094482421875], "reading_logprob_std": 1.3152240821277175, "tokenization_path_logprob_std": 254.3868003927755, "trajectory_balance_rmse": 2365.0196540522893} {"canonical_length": 20, "canonical_nll": 0.6682137489318848, "generation_rouge_l_f1": 0.19540229885057472, "generation_token_f1": 0.19540229885057472, "prompt_id": "24f992c3a95d7b8a", "reading_canonical_log_prob": -13.364274978637695, "reading_distances": [38, 41, 32, 31], "reading_log_probs": [-14.592750549316406, -14.264164924621582, -15.225738525390625, -13.92412281036377], "reading_logprob_std": 0.5545389408236117, "tokenization_path_logprob_std": 52.741640908180884, "trajectory_balance_rmse": 734.2323434266594} {"canonical_length": 29, "canonical_nll": 0.8147546176252694, "generation_rouge_l_f1": 0.27397260273972607, "generation_token_f1": 0.3561643835616438, "prompt_id": "429af0c846ac90d6", "reading_canonical_log_prob": -23.627883911132812, "reading_distances": [20, 20, 14, 22], "reading_log_probs": [-25.313331604003906, -25.917495727539062, -25.46491050720215, -26.231937408447266], "reading_logprob_std": 0.42070023924526645, "tokenization_path_logprob_std": 90.75006196706657, "trajectory_balance_rmse": 842.7727248516558} {"canonical_length": 132, "canonical_nll": 1.301470323042436, "generation_rouge_l_f1": 0.22871046228710465, "generation_token_f1": 0.437956204379562, "prompt_id": "a4d2b99ec03649f1", "reading_canonical_log_prob": -171.79408264160156, "reading_distances": [19, 21, 19, 21], "reading_log_probs": [-174.85055541992188, -175.72970581054688, -178.69277954101562, -173.44903564453125], "reading_logprob_std": 2.2169416184826676, "tokenization_path_logprob_std": 190.08656678058958, "trajectory_balance_rmse": 3708.596690913146} {"canonical_length": 27, "canonical_nll": 0.5196691795631692, "generation_rouge_l_f1": 0.24365482233502536, "generation_token_f1": 0.24365482233502536, "prompt_id": "363a6f01446a3b83", "reading_canonical_log_prob": -14.031067848205566, "reading_distances": [45, 40, 44, 33], "reading_log_probs": [-14.130695343017578, -12.489630699157715, -13.439142227172852, -13.110097885131836], "reading_logprob_std": 0.6836044115652572, "tokenization_path_logprob_std": 64.53745775212496, "trajectory_balance_rmse": 864.8682902909622} {"canonical_length": 56, "canonical_nll": 2.194054194859096, "generation_rouge_l_f1": 0.09316770186335405, "generation_token_f1": 0.1925465838509317, "prompt_id": "15d4cbd9fbd2e327", "reading_canonical_log_prob": -122.86703491210938, "reading_distances": [28, 29, 30, 28], "reading_log_probs": [-126.74039459228516, -125.98127746582031, -124.49215698242188, -127.35345458984375], "reading_logprob_std": 1.234708048984007, "tokenization_path_logprob_std": 94.85110353434686, "trajectory_balance_rmse": 1733.1659080374588} {"canonical_length": 56, "canonical_nll": 1.150582722255162, "generation_rouge_l_f1": 0.28571428571428564, "generation_token_f1": 0.4523809523809524, "prompt_id": "c741581c2c318b9d", "reading_canonical_log_prob": -64.43263244628906, "reading_distances": [19, 28, 30, 29], "reading_log_probs": [-62.59345626831055, -64.14321899414062, -62.99983596801758, -64.7508544921875], "reading_logprob_std": 0.9984941820057869, "tokenization_path_logprob_std": 134.28664887491985, "trajectory_balance_rmse": 1501.0952269273384} {"canonical_length": 10, "canonical_nll": 0.9613726615905762, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "1551065a718dd716", "reading_canonical_log_prob": -9.613726615905762, "reading_distances": [19, 24, 17, 29], "reading_log_probs": [-12.024121284484863, -12.014558792114258, -8.955917358398438, -9.918285369873047], "reading_logprob_std": 1.5417607235162747, "tokenization_path_logprob_std": 71.48147848175036, "trajectory_balance_rmse": 234.8940043065903} {"canonical_length": 22, "canonical_nll": 2.0184844623912466, "generation_rouge_l_f1": 0.5333333333333333, "generation_token_f1": 0.5333333333333333, "prompt_id": "0281e958164e52e0", "reading_canonical_log_prob": -44.40665817260742, "reading_distances": [20, 23, 25, 22], "reading_log_probs": [-44.65061950683594, -45.695518493652344, -46.193668365478516, -45.98213577270508], "reading_logprob_std": 0.6843974257415358, "tokenization_path_logprob_std": 68.42573389104398, "trajectory_balance_rmse": 558.1049124196591} {"canonical_length": 338, "canonical_nll": 1.3501465927214311, "generation_rouge_l_f1": 0.22617124394184168, "generation_token_f1": 0.38126009693053314, "prompt_id": "059afc2f6690f050", "reading_canonical_log_prob": -456.34954833984375, "reading_distances": [28, 20, 24, 34], "reading_log_probs": [-454.2857666015625, -455.0997009277344, -456.27911376953125, -453.55572509765625], "reading_logprob_std": 1.1676402303418596, "tokenization_path_logprob_std": 450.23818586127743, "trajectory_balance_rmse": 8528.443875940342} {"canonical_length": 11, "canonical_nll": 1.351039019497958, "generation_rouge_l_f1": 0.12000000000000001, "generation_token_f1": 0.14, "prompt_id": "3463b16713cbef14", "reading_canonical_log_prob": -14.861429214477539, "reading_distances": [37, 40, 33, 39], "reading_log_probs": [-14.711685180664062, -10.860197067260742, -11.065831184387207, -12.577818870544434], "reading_logprob_std": 1.7785343753507532, "tokenization_path_logprob_std": 52.81532117143432, "trajectory_balance_rmse": 380.7419716351259} {"canonical_length": 184, "canonical_nll": 2.0024530161981997, "generation_rouge_l_f1": 0.26987951807228916, "generation_token_f1": 0.5012048192771084, "prompt_id": "75052e7415fb5476", "reading_canonical_log_prob": -368.45135498046875, "reading_distances": [25, 20, 30, 28], "reading_log_probs": [-370.586181640625, -370.3179931640625, -369.83526611328125, -370.2462158203125], "reading_logprob_std": 0.3107025642915916, "tokenization_path_logprob_std": 298.0616661690898, "trajectory_balance_rmse": 4868.581232260451} {"canonical_length": 155, "canonical_nll": 2.6150947816910284, "generation_rouge_l_f1": 0.24277456647398843, "generation_token_f1": 0.43352601156069365, "prompt_id": "3e8a52f189b6e093", "reading_canonical_log_prob": -405.3396911621094, "reading_distances": [27, 25, 28, 31], "reading_log_probs": [-410.09979248046875, -415.5330810546875, -414.0540771484375, -409.16009521484375], "reading_logprob_std": 3.0658533770249785, "tokenization_path_logprob_std": 181.35316663119784, "trajectory_balance_rmse": 4555.796636289765} {"canonical_length": 278, "canonical_nll": 1.5795647051694581, "generation_rouge_l_f1": 0.21132075471698114, "generation_token_f1": 0.4679245283018868, "prompt_id": "39b34b632ecf238e", "reading_canonical_log_prob": -439.1189880371094, "reading_distances": [27, 21, 17, 23], "reading_log_probs": [-442.0977478027344, -440.1891174316406, -441.70672607421875, -441.31573486328125], "reading_logprob_std": 0.8232349340376476, "tokenization_path_logprob_std": 121.04272980915248, "trajectory_balance_rmse": 8091.033506204649} {"canonical_length": 144, "canonical_nll": 2.175011952718099, "generation_rouge_l_f1": 0.19464720194647203, "generation_token_f1": 0.38929440389294406, "prompt_id": "76cab7f49b66aca5", "reading_canonical_log_prob": -313.20172119140625, "reading_distances": [17, 14, 15, 11], "reading_log_probs": [-306.9272766113281, -311.69451904296875, -306.19659423828125, -308.2947998046875], "reading_logprob_std": 2.437867091549632, "tokenization_path_logprob_std": 254.3811047221776, "trajectory_balance_rmse": 3759.797503799846} {"canonical_length": 40, "canonical_nll": 1.4021584510803222, "generation_rouge_l_f1": 0.38333333333333336, "generation_token_f1": 0.41666666666666663, "prompt_id": "b6ca714d2fc1e411", "reading_canonical_log_prob": -56.08633804321289, "reading_distances": [14, 16, 13, 18], "reading_log_probs": [-57.284759521484375, -58.40534973144531, -58.02772521972656, -58.088436126708984], "reading_logprob_std": 0.4743712375480637, "tokenization_path_logprob_std": 93.29600464268826, "trajectory_balance_rmse": 1058.8717291272515} {"canonical_length": 39, "canonical_nll": 1.415660369090545, "generation_rouge_l_f1": 0.3364485981308411, "generation_token_f1": 0.4672897196261682, "prompt_id": "56f7c951f93e4273", "reading_canonical_log_prob": -55.21075439453125, "reading_distances": [34, 27, 29, 26], "reading_log_probs": [-55.74704360961914, -53.38975143432617, -58.0975456237793, -56.45302200317383], "reading_logprob_std": 1.9543008608660335, "tokenization_path_logprob_std": 124.705870807758, "trajectory_balance_rmse": 976.688222700796} {"canonical_length": 17, "canonical_nll": 0.6970804999856388, "generation_rouge_l_f1": 0.30769230769230765, "generation_token_f1": 0.30769230769230765, "prompt_id": "3f9be20b25197a77", "reading_canonical_log_prob": -11.85036849975586, "reading_distances": [19, 14, 11, 10], "reading_log_probs": [-24.300168991088867, -12.371601104736328, -15.770011901855469, -17.86931037902832], "reading_logprob_std": 5.0215478322716525, "tokenization_path_logprob_std": 81.54122025783332, "trajectory_balance_rmse": 327.989219914358} {"canonical_length": 231, "canonical_nll": 1.4642266607903816, "generation_rouge_l_f1": 0.2309278350515464, "generation_token_f1": 0.49484536082474223, "prompt_id": "053c343603667759", "reading_canonical_log_prob": -338.2363586425781, "reading_distances": [22, 20, 18, 20], "reading_log_probs": [-338.6156311035156, -334.95330810546875, -336.6397705078125, -336.9059143066406], "reading_logprob_std": 1.4990948739773067, "tokenization_path_logprob_std": 272.5348598500465, "trajectory_balance_rmse": 6513.28550178827} {"canonical_length": 53, "canonical_nll": 1.6327170605929393, "generation_rouge_l_f1": 0.3384615384615384, "generation_token_f1": 0.4615384615384615, "prompt_id": "e23c57e239d810b9", "reading_canonical_log_prob": -86.53400421142578, "reading_distances": [29, 21, 23, 29], "reading_log_probs": [-83.30052947998047, -83.27243041992188, -82.85786437988281, -86.02751922607422], "reading_logprob_std": 1.4560879723493525, "tokenization_path_logprob_std": 128.87488353281836, "trajectory_balance_rmse": 1451.4050259656617} {"canonical_length": 66, "canonical_nll": 1.305787519975142, "generation_rouge_l_f1": 0.22545454545454544, "generation_token_f1": 0.2836363636363637, "prompt_id": "e9b590e25ca62203", "reading_canonical_log_prob": -86.18197631835938, "reading_distances": [21, 22, 30, 20], "reading_log_probs": [-86.17727661132812, -84.37306213378906, -85.26319122314453, -82.95924377441406], "reading_logprob_std": 1.3707012594788759, "tokenization_path_logprob_std": 182.6156069779296, "trajectory_balance_rmse": 1726.7206854187289} {"canonical_length": 16, "canonical_nll": 1.9194512367248535, "generation_rouge_l_f1": 0.608695652173913, "generation_token_f1": 0.608695652173913, "prompt_id": "d046e49475ca84ec", "reading_canonical_log_prob": -30.711219787597656, "reading_distances": [37, 44, 42, 37], "reading_log_probs": [-33.707923889160156, -37.44549560546875, -33.06025695800781, -36.50312042236328], "reading_logprob_std": 2.1247301056565764, "tokenization_path_logprob_std": 143.74407005290615, "trajectory_balance_rmse": 646.235423883019} {"canonical_length": 56, "canonical_nll": 1.0185319355555944, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.5052631578947367, "prompt_id": "750881d0bf75959b", "reading_canonical_log_prob": -57.03778839111328, "reading_distances": [48, 44, 52, 48], "reading_log_probs": [-64.40941619873047, -59.6645622253418, -61.54143142700195, -60.02506637573242], "reading_logprob_std": 2.1585771444281856, "tokenization_path_logprob_std": 137.3722776997835, "trajectory_balance_rmse": 1852.352981995451} {"canonical_length": 2, "canonical_nll": 0.009165593422949314, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "73f8da9f02677521", "reading_canonical_log_prob": -0.018331186845898628, "reading_distances": [18, 14, 15, 17], "reading_log_probs": [-0.026315761730074883, -0.04113525152206421, -0.03964863717556, -0.018631432205438614], "reading_logprob_std": 0.010827382633819183, "tokenization_path_logprob_std": 11.377149526140629, "trajectory_balance_rmse": 61.76137161152225} {"canonical_length": 156, "canonical_nll": 2.4884379460261417, "generation_rouge_l_f1": 0.2506393861892583, "generation_token_f1": 0.3938618925831202, "prompt_id": "9c25897ff633e0cd", "reading_canonical_log_prob": -388.1963195800781, "reading_distances": [29, 34, 28, 35], "reading_log_probs": [-393.6549072265625, -392.80120849609375, -390.95556640625, -394.018310546875], "reading_logprob_std": 1.3667308366004554, "tokenization_path_logprob_std": 248.3498421662053, "trajectory_balance_rmse": 3526.772650047109} {"canonical_length": 263, "canonical_nll": 1.236812577048182, "generation_rouge_l_f1": 0.3946360153256705, "generation_token_f1": 0.6130268199233716, "prompt_id": "5d0b3ece32e43007", "reading_canonical_log_prob": -325.2817077636719, "reading_distances": [13, 21, 23, 25], "reading_log_probs": [-322.03521728515625, -319.3595275878906, -316.55267333984375, -322.8049621582031], "reading_logprob_std": 2.8379963039408493, "tokenization_path_logprob_std": 189.43088530349786, "trajectory_balance_rmse": 8577.742610462616} {"canonical_length": 21, "canonical_nll": 0.5857676551455543, "generation_rouge_l_f1": 0.6, "generation_token_f1": 0.6, "prompt_id": "16cea9158da9fd5a", "reading_canonical_log_prob": -12.30112075805664, "reading_distances": [31, 29, 34, 38], "reading_log_probs": [-19.039260864257812, -13.331960678100586, -14.269460678100586, -12.943382263183594], "reading_logprob_std": 2.817688008321399, "tokenization_path_logprob_std": 31.30573112457584, "trajectory_balance_rmse": 598.1107898703023} {"canonical_length": 18, "canonical_nll": 1.4033554924858942, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.48, "prompt_id": "d643de98c824ec5b", "reading_canonical_log_prob": -25.260398864746094, "reading_distances": [27, 23, 18, 22], "reading_log_probs": [-38.34441375732422, -38.41977310180664, -33.951805114746094, -37.99640655517578], "reading_logprob_std": 2.1587530488017452, "tokenization_path_logprob_std": 75.37790674516549, "trajectory_balance_rmse": 440.74701455512485} {"canonical_length": 97, "canonical_nll": 1.8564472001852448, "generation_rouge_l_f1": 0.2885572139303483, "generation_token_f1": 0.46766169154228854, "prompt_id": "cdeaf78291feca5e", "reading_canonical_log_prob": -180.07537841796875, "reading_distances": [37, 29, 32, 39], "reading_log_probs": [-192.97186279296875, -190.64498901367188, -194.38369750976562, -192.4162139892578], "reading_logprob_std": 1.5465697252783455, "tokenization_path_logprob_std": 84.28044708844511, "trajectory_balance_rmse": 2327.03124048603} {"canonical_length": 308, "canonical_nll": 2.0229452554281657, "generation_rouge_l_f1": 0.22700587084148727, "generation_token_f1": 0.3953033268101761, "prompt_id": "26a8450ba4613a47", "reading_canonical_log_prob": -623.067138671875, "reading_distances": [22, 17, 20, 17], "reading_log_probs": [-632.5413208007812, -633.1278686523438, -626.4239501953125, -634.426513671875], "reading_logprob_std": 3.5589117886152835, "tokenization_path_logprob_std": 260.49725745081315, "trajectory_balance_rmse": 6900.427473954585} {"canonical_length": 104, "canonical_nll": 0.7454131933359, "generation_rouge_l_f1": 0.5549738219895288, "generation_token_f1": 0.6910994764397905, "prompt_id": "c15330337653f483", "reading_canonical_log_prob": -77.5229721069336, "reading_distances": [281, 304, 302, 287], "reading_log_probs": [-103.51140594482422, -100.87041473388672, -110.0531997680664, -104.69676971435547], "reading_logprob_std": 3.860384961470664, "tokenization_path_logprob_std": 214.7373753292863, "trajectory_balance_rmse": 2502.075366784408} {"canonical_length": 11, "canonical_nll": 1.1841351769187234, "generation_rouge_l_f1": 0.5714285714285714, "generation_token_f1": 0.5714285714285714, "prompt_id": "8609ed9bb93a15ed", "reading_canonical_log_prob": -13.025486946105957, "reading_distances": [17, 14, 15, 19], "reading_log_probs": [-18.686508178710938, -20.42215347290039, -20.002653121948242, -26.15380859375], "reading_logprob_std": 3.308698664056874, "tokenization_path_logprob_std": 45.00744983347245, "trajectory_balance_rmse": 204.24534147962174} {"canonical_length": 367, "canonical_nll": 1.432784826294278, "generation_rouge_l_f1": 0.27140549273021003, "generation_token_f1": 0.507269789983845, "prompt_id": "4cb8d591dfd52931", "reading_canonical_log_prob": -525.83203125, "reading_distances": [22, 25, 30, 31], "reading_log_probs": [-527.1785888671875, -524.761474609375, -530.2610473632812, -528.8759155273438], "reading_logprob_std": 2.3684985443072946, "tokenization_path_logprob_std": 243.59174246491966, "trajectory_balance_rmse": 10412.97365459294} {"canonical_length": 277, "canonical_nll": 0.5261178688021774, "generation_rouge_l_f1": 0.5086419753086419, "generation_token_f1": 0.6419753086419754, "prompt_id": "7e37b68aba5b5c9c", "reading_canonical_log_prob": -145.73464965820312, "reading_distances": [35, 28, 36, 36], "reading_log_probs": [-144.30027770996094, -146.418701171875, -144.64012145996094, -148.19393920898438], "reading_logprob_std": 1.7960198607078777, "tokenization_path_logprob_std": 125.40615954127702, "trajectory_balance_rmse": 5843.835691033986} {"canonical_length": 87, "canonical_nll": 1.5325598004220546, "generation_rouge_l_f1": 0.15909090909090912, "generation_token_f1": 0.3125, "prompt_id": "fef695db887a3a23", "reading_canonical_log_prob": -133.33270263671875, "reading_distances": [17, 19, 19, 15], "reading_log_probs": [-144.76956176757812, -137.48455810546875, -146.06417846679688, -139.23678588867188], "reading_logprob_std": 4.169856081105911, "tokenization_path_logprob_std": 92.99620343037324, "trajectory_balance_rmse": 2317.1269279468793} {"canonical_length": 174, "canonical_nll": 0.6559466657967403, "generation_rouge_l_f1": 0.39069767441860465, "generation_token_f1": 0.6511627906976745, "prompt_id": "471b9b9cabf2decb", "reading_canonical_log_prob": -114.13471984863281, "reading_distances": [22, 24, 22, 22], "reading_log_probs": [-112.36996459960938, -109.50652313232422, -111.22506713867188, -109.07296752929688], "reading_logprob_std": 1.5316976857292584, "tokenization_path_logprob_std": 163.08722691549696, "trajectory_balance_rmse": 3555.195421563502} {"canonical_length": 10, "canonical_nll": 1.6584566116333008, "generation_rouge_l_f1": 0.05228758169934641, "generation_token_f1": 0.05228758169934641, "prompt_id": "71f95c366e84f3b3", "reading_canonical_log_prob": -16.584566116333008, "reading_distances": [29, 41, 33, 31], "reading_log_probs": [-13.50218391418457, -10.735614776611328, -10.604007720947266, -11.391242980957031], "reading_logprob_std": 1.3409048492725772, "tokenization_path_logprob_std": 63.24477977685359, "trajectory_balance_rmse": 374.1994280091138} {"canonical_length": 7, "canonical_nll": 0.7956836564200265, "generation_rouge_l_f1": 0.631578947368421, "generation_token_f1": 0.631578947368421, "prompt_id": "78998e3018151a20", "reading_canonical_log_prob": -5.5697855949401855, "reading_distances": [15, 19, 16, 24], "reading_log_probs": [-7.045882701873779, -8.193305969238281, -7.235249042510986, -6.491592884063721], "reading_logprob_std": 0.7086608232366101, "tokenization_path_logprob_std": 26.909228791381736, "trajectory_balance_rmse": 130.74769662119792} {"canonical_length": 179, "canonical_nll": 1.4967971887002445, "generation_rouge_l_f1": 0.184549356223176, "generation_token_f1": 0.3776824034334764, "prompt_id": "3025be82a097d5b7", "reading_canonical_log_prob": -267.92669677734375, "reading_distances": [24, 21, 30, 26], "reading_log_probs": [-258.8413391113281, -261.0240478515625, -264.64166259765625, -263.53802490234375], "reading_logprob_std": 2.599552215218961, "tokenization_path_logprob_std": 99.30007453363751, "trajectory_balance_rmse": 4815.01888692511} {"canonical_length": 259, "canonical_nll": 1.9040932673745175, "generation_rouge_l_f1": 0.1410488245931284, "generation_token_f1": 0.2531645569620253, "prompt_id": "317ccf6fb834b830", "reading_canonical_log_prob": -493.16015625, "reading_distances": [20, 24, 29, 20], "reading_log_probs": [-488.0413818359375, -486.3052673339844, -488.5327453613281, -488.897705078125], "reading_logprob_std": 1.1476210389146815, "tokenization_path_logprob_std": 290.7016928342517, "trajectory_balance_rmse": 7248.516007137403} {"canonical_length": 35, "canonical_nll": 1.5550824846540179, "generation_rouge_l_f1": 0.5416666666666667, "generation_token_f1": 0.5416666666666667, "prompt_id": "43805108665c2c05", "reading_canonical_log_prob": -54.427886962890625, "reading_distances": [31, 17, 23, 25], "reading_log_probs": [-61.44076919555664, -57.09679412841797, -57.59574508666992, -60.31417465209961], "reading_logprob_std": 2.099878842582321, "tokenization_path_logprob_std": 98.16656241807284, "trajectory_balance_rmse": 1171.998036483513} {"canonical_length": 118, "canonical_nll": 1.5264053344726562, "generation_rouge_l_f1": 0.26558265582655827, "generation_token_f1": 0.45528455284552855, "prompt_id": "5056b26f2553e0fc", "reading_canonical_log_prob": -180.11582946777344, "reading_distances": [26, 26, 33, 14], "reading_log_probs": [-175.6908416748047, -180.3030548095703, -178.19940185546875, -177.96376037597656], "reading_logprob_std": 1.8860170710577469, "tokenization_path_logprob_std": 199.211426258728, "trajectory_balance_rmse": 3344.0145467178872} {"canonical_length": 372, "canonical_nll": 1.055459955687164, "generation_rouge_l_f1": 0.26810477657935283, "generation_token_f1": 0.4622496147919876, "prompt_id": "1ea4408820560cde", "reading_canonical_log_prob": -392.631103515625, "reading_distances": [12, 13, 13, 14], "reading_log_probs": [-390.1463928222656, -382.42706298828125, -383.1549072265625, -380.63348388671875], "reading_logprob_std": 4.174013049255305, "tokenization_path_logprob_std": 496.99571687923867, "trajectory_balance_rmse": 8490.146380126209} {"canonical_length": 9, "canonical_nll": 0.7848557896084256, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.8571428571428571, "prompt_id": "2b79f5bffeeb5daf", "reading_canonical_log_prob": -7.06370210647583, "reading_distances": [35, 37, 32, 32], "reading_log_probs": [-7.380903720855713, -6.829647064208984, -7.1427507400512695, -6.230660438537598], "reading_logprob_std": 0.49769367753968974, "tokenization_path_logprob_std": 64.5276101804866, "trajectory_balance_rmse": 226.65661183069986} {"canonical_length": 36, "canonical_nll": 0.9239728715684679, "generation_rouge_l_f1": 0.11764705882352942, "generation_token_f1": 0.18685121107266436, "prompt_id": "91748c9b4543a8c6", "reading_canonical_log_prob": -33.263023376464844, "reading_distances": [10, 13, 13, 11], "reading_log_probs": [-30.947994232177734, -31.733261108398438, -32.09965133666992, -31.710113525390625], "reading_logprob_std": 0.48393364616548634, "tokenization_path_logprob_std": 104.3131069006856, "trajectory_balance_rmse": 629.5897797900936} {"canonical_length": 187, "canonical_nll": 1.2672856784759359, "generation_rouge_l_f1": 0.31186440677966104, "generation_token_f1": 0.5288135593220339, "prompt_id": "9a3c9e43bea33f6c", "reading_canonical_log_prob": -236.982421875, "reading_distances": [21, 17, 25, 16], "reading_log_probs": [-237.7435302734375, -235.48004150390625, -238.79733276367188, -234.75076293945312], "reading_logprob_std": 1.895203007651779, "tokenization_path_logprob_std": 87.4146104376613, "trajectory_balance_rmse": 5621.815637872553} {"canonical_length": 376, "canonical_nll": 1.4059169038813164, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.43606557377049177, "prompt_id": "7473e8ab94e776ba", "reading_canonical_log_prob": -528.624755859375, "reading_distances": [22, 19, 22, 14], "reading_log_probs": [-536.9044799804688, -534.0601806640625, -536.225830078125, -532.7721557617188], "reading_logprob_std": 1.9127558556007414, "tokenization_path_logprob_std": 193.48735771867297, "trajectory_balance_rmse": 9953.620925837367} {"canonical_length": 87, "canonical_nll": 0.5547630923917923, "generation_rouge_l_f1": 0.22895622895622897, "generation_token_f1": 0.3771043771043771, "prompt_id": "81563152acabd240", "reading_canonical_log_prob": -48.26438903808594, "reading_distances": [40, 35, 40, 34], "reading_log_probs": [-45.407562255859375, -43.78177261352539, -46.88130569458008, -48.09819793701172], "reading_logprob_std": 1.865790531215023, "tokenization_path_logprob_std": 149.51074832053578, "trajectory_balance_rmse": 1473.1546596252083} {"canonical_length": 19, "canonical_nll": 3.0313216761538855, "generation_rouge_l_f1": 0.06060606060606061, "generation_token_f1": 0.06060606060606061, "prompt_id": "49b74500a1071228", "reading_canonical_log_prob": -57.59511184692383, "reading_distances": [22, 20, 18, 28], "reading_log_probs": [-55.943023681640625, -57.65809631347656, -58.4447135925293, -56.25432205200195], "reading_logprob_std": 1.179123871065614, "tokenization_path_logprob_std": 86.72279772919235, "trajectory_balance_rmse": 692.9757015348841} {"canonical_length": 83, "canonical_nll": 1.2208479915756778, "generation_rouge_l_f1": 0.2165605095541401, "generation_token_f1": 0.3121019108280255, "prompt_id": "c1a613d92b944071", "reading_canonical_log_prob": -101.33038330078125, "reading_distances": [36, 36, 31, 34], "reading_log_probs": [-100.2488021850586, -101.84738159179688, -101.88800048828125, -103.12448120117188], "reading_logprob_std": 1.1787519834206677, "tokenization_path_logprob_std": 82.96297914714755, "trajectory_balance_rmse": 2155.2609064012945} {"canonical_length": 190, "canonical_nll": 1.298240179764597, "generation_rouge_l_f1": 0.24892703862660945, "generation_token_f1": 0.4935622317596566, "prompt_id": "00050e95a0ce36d0", "reading_canonical_log_prob": -246.66563415527344, "reading_distances": [34, 29, 22, 38], "reading_log_probs": [-245.44410705566406, -244.74758911132812, -248.2015380859375, -248.3475341796875], "reading_logprob_std": 1.858071021960952, "tokenization_path_logprob_std": 237.21961655699488, "trajectory_balance_rmse": 4944.619743137892} {"canonical_length": 309, "canonical_nll": 1.4607761222567759, "generation_rouge_l_f1": 0.3071428571428571, "generation_token_f1": 0.5892857142857143, "prompt_id": "3e9a72eac16389eb", "reading_canonical_log_prob": -451.37982177734375, "reading_distances": [18, 19, 20, 22], "reading_log_probs": [-450.7088317871094, -452.996826171875, -449.5238342285156, -452.0753173828125], "reading_logprob_std": 1.525541930361022, "tokenization_path_logprob_std": 522.998798430979, "trajectory_balance_rmse": 9280.043371130963} {"canonical_length": 335, "canonical_nll": 1.256955912575793, "generation_rouge_l_f1": 0.32116788321167883, "generation_token_f1": 0.5145985401459854, "prompt_id": "bef02b456531b6b6", "reading_canonical_log_prob": -421.0802307128906, "reading_distances": [18, 20, 20, 21], "reading_log_probs": [-422.99542236328125, -421.49346923828125, -422.064208984375, -420.8206787109375], "reading_logprob_std": 0.9209275212419848, "tokenization_path_logprob_std": 121.01080712483673, "trajectory_balance_rmse": 7853.900345235269} {"canonical_length": 110, "canonical_nll": 1.726150235262784, "generation_rouge_l_f1": 0.22038567493112948, "generation_token_f1": 0.3856749311294766, "prompt_id": "9d2a25bad82737d8", "reading_canonical_log_prob": -189.87652587890625, "reading_distances": [16, 19, 17, 18], "reading_log_probs": [-187.40499877929688, -192.77439880371094, -194.54547119140625, -189.83761596679688], "reading_logprob_std": 3.157790906640337, "tokenization_path_logprob_std": 192.41291080275616, "trajectory_balance_rmse": 3874.2408890662036} {"canonical_length": 10, "canonical_nll": 2.092181396484375, "generation_rouge_l_f1": 0.5217391304347826, "generation_token_f1": 0.5217391304347826, "prompt_id": "77bd9a6bbbf6a3af", "reading_canonical_log_prob": -20.92181396484375, "reading_distances": [21, 22, 17, 19], "reading_log_probs": [-22.02654266357422, -19.965660095214844, -21.399051666259766, -19.6713924407959], "reading_logprob_std": 1.1296653535391672, "tokenization_path_logprob_std": 59.06065813601638, "trajectory_balance_rmse": 212.88481450071455} {"canonical_length": 18, "canonical_nll": 1.1233880784776475, "generation_rouge_l_f1": 0.09090909090909091, "generation_token_f1": 0.09090909090909091, "prompt_id": "45f03bee685e0c93", "reading_canonical_log_prob": -20.220985412597656, "reading_distances": [23, 23, 27, 23], "reading_log_probs": [-14.933863639831543, -14.020051956176758, -15.5057373046875, -14.55993938446045], "reading_logprob_std": 0.6255120386796912, "tokenization_path_logprob_std": 96.66932426781189, "trajectory_balance_rmse": 585.7142670254716} {"canonical_length": 300, "canonical_nll": 1.0387705485026042, "generation_rouge_l_f1": 0.24738675958188153, "generation_token_f1": 0.4982578397212543, "prompt_id": "1c2335f122ec1aa9", "reading_canonical_log_prob": -311.63116455078125, "reading_distances": [29, 22, 25, 22], "reading_log_probs": [-313.5610046386719, -314.3601989746094, -314.3709411621094, -314.37713623046875], "reading_logprob_std": 0.40427096707342786, "tokenization_path_logprob_std": 200.6753990504788, "trajectory_balance_rmse": 7169.381558254658} {"canonical_length": 24, "canonical_nll": 0.5529553890228271, "generation_rouge_l_f1": 0.4999999999999999, "generation_token_f1": 0.4999999999999999, "prompt_id": "0b6723f81e5854b3", "reading_canonical_log_prob": -13.270929336547852, "reading_distances": [20, 28, 25, 33], "reading_log_probs": [-16.236127853393555, -14.55894660949707, -14.555459976196289, -13.733379364013672], "reading_logprob_std": 1.0511391204215697, "tokenization_path_logprob_std": 114.00642852461742, "trajectory_balance_rmse": 822.8478273011724} {"canonical_length": 43, "canonical_nll": 0.61565283841865, "generation_rouge_l_f1": 0.7536231884057971, "generation_token_f1": 0.7826086956521738, "prompt_id": "6d26cacb241c8ba0", "reading_canonical_log_prob": -26.473072052001953, "reading_distances": [32, 37, 42, 38], "reading_log_probs": [-20.94927978515625, -18.96894073486328, -15.493404388427734, -15.636924743652344], "reading_logprob_std": 2.663201656202213, "tokenization_path_logprob_std": 57.36350953602525, "trajectory_balance_rmse": 1167.8893393535006} {"canonical_length": 10, "canonical_nll": 2.7239246368408203, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.7368421052631577, "prompt_id": "d7683fe01587262b", "reading_canonical_log_prob": -27.239246368408203, "reading_distances": [40, 33, 45, 37], "reading_log_probs": [-27.607664108276367, -27.962343215942383, -28.158842086791992, -27.862722396850586], "reading_logprob_std": 0.22928732858548678, "tokenization_path_logprob_std": 25.457830994523935, "trajectory_balance_rmse": 304.4570114514076} {"canonical_length": 71, "canonical_nll": 0.8071421770982339, "generation_rouge_l_f1": 0.6056338028169015, "generation_token_f1": 0.676056338028169, "prompt_id": "e455ade71af9293b", "reading_canonical_log_prob": -57.30709457397461, "reading_distances": [52, 34, 39, 53], "reading_log_probs": [-49.02117919921875, -52.81646728515625, -51.80527877807617, -52.618717193603516], "reading_logprob_std": 1.751689448615982, "tokenization_path_logprob_std": 176.9136334894641, "trajectory_balance_rmse": 2020.2106330163617} {"canonical_length": 334, "canonical_nll": 1.8862454534290793, "generation_rouge_l_f1": 0.20309810671256456, "generation_token_f1": 0.3476764199655766, "prompt_id": "83deb2f6c2a5fe75", "reading_canonical_log_prob": -630.0059814453125, "reading_distances": [43, 35, 31, 37], "reading_log_probs": [-627.3423461914062, -620.6343994140625, -626.4844360351562, -625.3822021484375], "reading_logprob_std": 2.9938019764001176, "tokenization_path_logprob_std": 208.8689317601759, "trajectory_balance_rmse": 7918.735719761686} {"canonical_length": 251, "canonical_nll": 1.36563949281001, "generation_rouge_l_f1": 0.14849187935034802, "generation_token_f1": 0.2830626450116009, "prompt_id": "cc738e3b9180c6b5", "reading_canonical_log_prob": -342.7755126953125, "reading_distances": [11, 11, 12, 15], "reading_log_probs": [-335.6231689453125, -342.70477294921875, -338.65087890625, -347.3341064453125], "reading_logprob_std": 5.080398819887847, "tokenization_path_logprob_std": 167.7667087042545, "trajectory_balance_rmse": 5989.0869942462705} {"canonical_length": 15, "canonical_nll": 0.15862933794657388, "generation_rouge_l_f1": 0.888888888888889, "generation_token_f1": 0.888888888888889, "prompt_id": "d829b51f56a6d0c5", "reading_canonical_log_prob": -2.3794400691986084, "reading_distances": [18, 16, 15, 19], "reading_log_probs": [-2.4890825748443604, -1.8126399517059326, -2.216338872909546, -2.163311004638672], "reading_logprob_std": 0.27791572860267283, "tokenization_path_logprob_std": 76.53743468606714, "trajectory_balance_rmse": 554.4471053156627} {"canonical_length": 158, "canonical_nll": 1.4884286228614518, "generation_rouge_l_f1": 0.24175824175824176, "generation_token_f1": 0.4065934065934066, "prompt_id": "f53fbba7ee710f29", "reading_canonical_log_prob": -235.17172241210938, "reading_distances": [22, 23, 22, 29], "reading_log_probs": [-238.86428833007812, -236.52577209472656, -236.45594787597656, -238.63104248046875], "reading_logprob_std": 1.3067530023345109, "tokenization_path_logprob_std": 260.99038926863346, "trajectory_balance_rmse": 5203.1887223307} {"canonical_length": 268, "canonical_nll": 1.3494652136048275, "generation_rouge_l_f1": 0.21818181818181817, "generation_token_f1": 0.3959595959595959, "prompt_id": "2ce5516156798d8a", "reading_canonical_log_prob": -361.65667724609375, "reading_distances": [31, 29, 26, 30], "reading_log_probs": [-363.2513732910156, -361.145751953125, -362.90826416015625, -359.08099365234375], "reading_logprob_std": 1.914007729988286, "tokenization_path_logprob_std": 446.07644443744795, "trajectory_balance_rmse": 6685.695765478276} {"canonical_length": 28, "canonical_nll": 0.1554532561983381, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4210526315789474, "prompt_id": "83deea31656f880e", "reading_canonical_log_prob": -4.352691173553467, "reading_distances": [33, 26, 28, 23], "reading_log_probs": [-4.763876438140869, -4.719607353210449, -4.050641059875488, -4.227502346038818], "reading_logprob_std": 0.35582354915665854, "tokenization_path_logprob_std": 78.95544563666184, "trajectory_balance_rmse": 705.5211805523061} {"canonical_length": 16, "canonical_nll": 0.1654813587665558, "generation_rouge_l_f1": 0.9444444444444444, "generation_token_f1": 0.9444444444444444, "prompt_id": "2a011ac439c0c89f", "reading_canonical_log_prob": -2.6477017402648926, "reading_distances": [50, 50, 40, 47], "reading_log_probs": [-6.512495040893555, -5.989223480224609, -5.269588470458984, -5.493678092956543], "reading_logprob_std": 0.5530426279133881, "tokenization_path_logprob_std": 54.66827967143787, "trajectory_balance_rmse": 529.199372134464} {"canonical_length": 27, "canonical_nll": 1.3140785782425493, "generation_rouge_l_f1": 0.1111111111111111, "generation_token_f1": 0.1574074074074074, "prompt_id": "a62ae1af2ae45a72", "reading_canonical_log_prob": -35.48012161254883, "reading_distances": [27, 23, 29, 22], "reading_log_probs": [-33.260414123535156, -36.597877502441406, -39.72035598754883, -37.037620544433594], "reading_logprob_std": 2.6501118581539647, "tokenization_path_logprob_std": 50.23704705220078, "trajectory_balance_rmse": 564.2364985055793} {"canonical_length": 99, "canonical_nll": 1.764631637419113, "generation_rouge_l_f1": 0.21022727272727273, "generation_token_f1": 0.3295454545454545, "prompt_id": "53aea8a2764798f8", "reading_canonical_log_prob": -174.6985321044922, "reading_distances": [34, 31, 28, 33], "reading_log_probs": [-179.635498046875, -177.93653869628906, -176.36019897460938, -176.55496215820312], "reading_logprob_std": 1.5147959597346536, "tokenization_path_logprob_std": 196.2181589002936, "trajectory_balance_rmse": 2660.6548161435144} {"canonical_length": 159, "canonical_nll": 0.5578023562641263, "generation_rouge_l_f1": 0.36129032258064514, "generation_token_f1": 0.5483870967741935, "prompt_id": "7dff62aaf925a1c6", "reading_canonical_log_prob": -88.6905746459961, "reading_distances": [30, 32, 34, 32], "reading_log_probs": [-90.8215560913086, -87.94508361816406, -90.66354370117188, -89.75540161132812], "reading_logprob_std": 1.3205978978605213, "tokenization_path_logprob_std": 109.73058178777377, "trajectory_balance_rmse": 3040.0098214510394} {"canonical_length": 217, "canonical_nll": 1.4050817533572149, "generation_rouge_l_f1": 0.1551362683438155, "generation_token_f1": 0.3731656184486373, "prompt_id": "5470eb105173aff8", "reading_canonical_log_prob": -304.9027404785156, "reading_distances": [25, 38, 33, 35], "reading_log_probs": [-313.25811767578125, -319.836669921875, -307.4599609375, -314.785888671875], "reading_logprob_std": 5.095689265785607, "tokenization_path_logprob_std": 211.95344034446376, "trajectory_balance_rmse": 6349.150996123346} {"canonical_length": 15, "canonical_nll": 0.5370452880859375, "generation_rouge_l_f1": 0.2622950819672131, "generation_token_f1": 0.36065573770491804, "prompt_id": "861fa387c936a923", "reading_canonical_log_prob": -8.055679321289062, "reading_distances": [9, 9, 11, 10], "reading_log_probs": [-4.822939395904541, -7.926171779632568, -4.586926460266113, -4.456701755523682], "reading_logprob_std": 1.658930965191763, "tokenization_path_logprob_std": 87.55686248695073, "trajectory_balance_rmse": 241.01652908013375} {"canonical_length": 11, "canonical_nll": 1.814980766989968, "generation_rouge_l_f1": 0.15730337078651682, "generation_token_f1": 0.15730337078651682, "prompt_id": "6c2751afcc2de989", "reading_canonical_log_prob": -19.96478843688965, "reading_distances": [11, 9, 9, 14], "reading_log_probs": [-15.78017520904541, -15.071673393249512, -17.484691619873047, -16.586889266967773], "reading_logprob_std": 1.0401411666966096, "tokenization_path_logprob_std": 49.26729134677173, "trajectory_balance_rmse": 255.0260372039014} {"canonical_length": 69, "canonical_nll": 0.855737326801687, "generation_rouge_l_f1": 0.2711864406779661, "generation_token_f1": 0.3644067796610169, "prompt_id": "2dc95060fe386398", "reading_canonical_log_prob": -59.045875549316406, "reading_distances": [21, 18, 24, 20], "reading_log_probs": [-59.31017303466797, -57.5552978515625, -58.34907531738281, -59.03253173828125], "reading_logprob_std": 0.783145139831005, "tokenization_path_logprob_std": 178.6814919376104, "trajectory_balance_rmse": 2116.3081291747826} {"canonical_length": 367, "canonical_nll": 1.3993265843196525, "generation_rouge_l_f1": 0.21487603305785122, "generation_token_f1": 0.36363636363636365, "prompt_id": "346df7037bb677e7", "reading_canonical_log_prob": -513.5528564453125, "reading_distances": [27, 25, 27, 28], "reading_log_probs": [-514.1088256835938, -512.830810546875, -513.2125244140625, -515.6163940429688], "reading_logprob_std": 1.2380499147357102, "tokenization_path_logprob_std": 334.5917347516317, "trajectory_balance_rmse": 8714.871027842068} {"canonical_length": 168, "canonical_nll": 1.8624605451311385, "generation_rouge_l_f1": 0.21363636363636362, "generation_token_f1": 0.44545454545454544, "prompt_id": "cccaf997f15db0c8", "reading_canonical_log_prob": -312.89337158203125, "reading_distances": [19, 12, 16, 12], "reading_log_probs": [-313.1019287109375, -313.052490234375, -313.4236145019531, -313.1831970214844], "reading_logprob_std": 0.16460774093333458, "tokenization_path_logprob_std": 186.41077318158253, "trajectory_balance_rmse": 5024.964593643851} {"canonical_length": 15, "canonical_nll": 0.1792755126953125, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "c7d4acf2c03f4f98", "reading_canonical_log_prob": -2.6891326904296875, "reading_distances": [15, 14, 19, 20], "reading_log_probs": [-5.712167739868164, -1.4493249654769897, -2.6062121391296387, -2.3629398345947266], "reading_logprob_std": 1.8544625850047303, "tokenization_path_logprob_std": 31.75620057837534, "trajectory_balance_rmse": 301.97602341189605} {"canonical_length": 31, "canonical_nll": 2.22289792952999, "generation_rouge_l_f1": 0.24489795918367346, "generation_token_f1": 0.3061224489795918, "prompt_id": "f4a99dd0e66e3bbe", "reading_canonical_log_prob": -68.90983581542969, "reading_distances": [22, 18, 23, 19], "reading_log_probs": [-69.07760620117188, -68.73091125488281, -68.5179443359375, -68.66389465332031], "reading_logprob_std": 0.2372979250154963, "tokenization_path_logprob_std": 52.71118888838146, "trajectory_balance_rmse": 952.0818759160825} {"canonical_length": 51, "canonical_nll": 1.5008682550168504, "generation_rouge_l_f1": 0.1923076923076923, "generation_token_f1": 0.25384615384615383, "prompt_id": "b701439d534ac688", "reading_canonical_log_prob": -76.54428100585938, "reading_distances": [21, 24, 23, 21], "reading_log_probs": [-80.73097229003906, -72.61384582519531, -75.939453125, -73.62449645996094], "reading_logprob_std": 3.614651892618928, "tokenization_path_logprob_std": 66.94012484480902, "trajectory_balance_rmse": 1379.3441849530916} {"canonical_length": 448, "canonical_nll": 1.3381919860839844, "generation_rouge_l_f1": 0.17440225035161744, "generation_token_f1": 0.360056258790436, "prompt_id": "7fdb8e03f79c4b21", "reading_canonical_log_prob": -599.510009765625, "reading_distances": [15, 14, 13, 13], "reading_log_probs": [-601.40869140625, -598.8356323242188, -598.3743286132812, -596.8886108398438], "reading_logprob_std": 1.8812710685958813, "tokenization_path_logprob_std": 403.6203338869796, "trajectory_balance_rmse": 9643.207831310348} {"canonical_length": 392, "canonical_nll": 1.3651870416135203, "generation_rouge_l_f1": 0.22585924713584288, "generation_token_f1": 0.45499181669394434, "prompt_id": "79079ac51debbea1", "reading_canonical_log_prob": -535.1533203125, "reading_distances": [26, 23, 23, 25], "reading_log_probs": [-524.8365478515625, -522.2517700195312, -527.9221801757812, -528.367919921875], "reading_logprob_std": 2.864034404901259, "tokenization_path_logprob_std": 393.4098163251884, "trajectory_balance_rmse": 9085.462698564372} {"canonical_length": 5, "canonical_nll": 2.927642250061035, "generation_rouge_l_f1": 0.007874015748031496, "generation_token_f1": 0.007874015748031496, "prompt_id": "e0c4de6979cf5d26", "reading_canonical_log_prob": -14.638211250305176, "reading_distances": [19, 19, 26, 24], "reading_log_probs": [-14.160826683044434, -14.000885963439941, -14.439661979675293, -14.86092472076416], "reading_logprob_std": 0.3767316432606783, "tokenization_path_logprob_std": 21.35779110928222, "trajectory_balance_rmse": 152.4402617689092} {"canonical_length": 199, "canonical_nll": 1.6745228216276695, "generation_rouge_l_f1": 0.22641509433962262, "generation_token_f1": 0.41509433962264153, "prompt_id": "a5453cc0769ffb94", "reading_canonical_log_prob": -333.23004150390625, "reading_distances": [22, 13, 22, 20], "reading_log_probs": [-356.0133361816406, -336.22271728515625, -354.6113586425781, -354.8201904296875], "reading_logprob_std": 9.4829209773856, "tokenization_path_logprob_std": 142.53148926290592, "trajectory_balance_rmse": 4156.489749469364} {"canonical_length": 16, "canonical_nll": 1.0790603160858154, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.37500000000000006, "prompt_id": "c1820a57ac4bcbc9", "reading_canonical_log_prob": -17.264965057373047, "reading_distances": [17, 13, 13, 14], "reading_log_probs": [-26.248958587646484, -23.622737884521484, -25.25788116455078, -29.574539184570312], "reading_logprob_std": 2.5111405886239466, "tokenization_path_logprob_std": 64.26981582201402, "trajectory_balance_rmse": 594.2204317907857} {"canonical_length": 87, "canonical_nll": 0.4749004320166577, "generation_rouge_l_f1": 0.30985915492957744, "generation_token_f1": 0.3661971830985915, "prompt_id": "326b224fa8e4dc3a", "reading_canonical_log_prob": -41.31633758544922, "reading_distances": [25, 28, 21, 30], "reading_log_probs": [-41.311397552490234, -43.26959991455078, -40.07071304321289, -41.84630584716797], "reading_logprob_std": 1.325121203556408, "tokenization_path_logprob_std": 182.3153863441616, "trajectory_balance_rmse": 1484.574664115573} {"canonical_length": 210, "canonical_nll": 1.0163026355561755, "generation_rouge_l_f1": 0.3258145363408521, "generation_token_f1": 0.49122807017543857, "prompt_id": "a8f7e780841cbd23", "reading_canonical_log_prob": -213.42355346679688, "reading_distances": [15, 21, 12, 16], "reading_log_probs": [-214.60890197753906, -214.5292205810547, -215.45870971679688, -216.7985076904297], "reading_logprob_std": 1.0540227758829632, "tokenization_path_logprob_std": 172.86144774360096, "trajectory_balance_rmse": 3806.060135291495} {"canonical_length": 221, "canonical_nll": 1.57182802226209, "generation_rouge_l_f1": 0.20524017467248906, "generation_token_f1": 0.46724890829694327, "prompt_id": "675093cc227e317b", "reading_canonical_log_prob": -347.3739929199219, "reading_distances": [15, 17, 14, 19], "reading_log_probs": [-340.8197937011719, -344.03521728515625, -343.30126953125, -343.160400390625], "reading_logprob_std": 1.3934052656956182, "tokenization_path_logprob_std": 203.93897070331508, "trajectory_balance_rmse": 6108.841268337494} {"canonical_length": 327, "canonical_nll": 1.2150153763797304, "generation_rouge_l_f1": 0.24067796610169492, "generation_token_f1": 0.471186440677966, "prompt_id": "df931a7b15ed70cf", "reading_canonical_log_prob": -397.3100280761719, "reading_distances": [34, 35, 42, 34], "reading_log_probs": [-397.2620544433594, -399.2982177734375, -396.17352294921875, -399.847412109375], "reading_logprob_std": 1.721862603547766, "tokenization_path_logprob_std": 243.4026566210518, "trajectory_balance_rmse": 9218.988299463901} {"canonical_length": 267, "canonical_nll": 1.4080887126565427, "generation_rouge_l_f1": 0.22975517890772126, "generation_token_f1": 0.4858757062146893, "prompt_id": "64f98636b7dde269", "reading_canonical_log_prob": -375.9596862792969, "reading_distances": [27, 26, 32, 24], "reading_log_probs": [-377.4718017578125, -378.06463623046875, -376.8666076660156, -377.09722900390625], "reading_logprob_std": 0.5230003771051697, "tokenization_path_logprob_std": 121.47883130801787, "trajectory_balance_rmse": 6802.513032677453} {"canonical_length": 110, "canonical_nll": 1.3780520352450285, "generation_rouge_l_f1": 0.20963172804532576, "generation_token_f1": 0.311614730878187, "prompt_id": "3485bbb40ac08228", "reading_canonical_log_prob": -151.58572387695312, "reading_distances": [16, 15, 10, 14], "reading_log_probs": [-152.6039581298828, -156.56161499023438, -151.70120239257812, -156.944580078125], "reading_logprob_std": 2.686110835923313, "tokenization_path_logprob_std": 64.7293757075367, "trajectory_balance_rmse": 3236.681028358489} {"canonical_length": 20, "canonical_nll": 1.1019770622253418, "generation_rouge_l_f1": 0.09558823529411764, "generation_token_f1": 0.11029411764705882, "prompt_id": "5de8b78dfeffe43c", "reading_canonical_log_prob": -22.039541244506836, "reading_distances": [13, 19, 12, 11], "reading_log_probs": [-21.342082977294922, -21.756664276123047, -20.755741119384766, -24.756025314331055], "reading_logprob_std": 1.783511614573795, "tokenization_path_logprob_std": 46.667654105213586, "trajectory_balance_rmse": 338.872269981673} {"canonical_length": 14, "canonical_nll": 3.1210081917898997, "generation_rouge_l_f1": 0.16470588235294115, "generation_token_f1": 0.18823529411764706, "prompt_id": "75d59497e8843334", "reading_canonical_log_prob": -43.694114685058594, "reading_distances": [7, 7, 11, 10], "reading_log_probs": [-42.56341552734375, -41.11561584472656, -52.21974182128906, -38.91389846801758], "reading_logprob_std": 5.872636947704371, "tokenization_path_logprob_std": 55.90383505526009, "trajectory_balance_rmse": 347.4731337560302} {"canonical_length": 53, "canonical_nll": 2.7793230740529187, "generation_rouge_l_f1": 0.18556701030927833, "generation_token_f1": 0.24054982817869416, "prompt_id": "9fe6000f50208baa", "reading_canonical_log_prob": -147.3041229248047, "reading_distances": [24, 22, 20, 23], "reading_log_probs": [-142.6141357421875, -151.71063232421875, -146.02674865722656, -154.16363525390625], "reading_logprob_std": 5.2624122822948385, "tokenization_path_logprob_std": 106.48416961331523, "trajectory_balance_rmse": 1849.6518867612988} {"canonical_length": 13, "canonical_nll": 0.03858781777895414, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "2321fcc6e5fd7b96", "reading_canonical_log_prob": -0.5016416311264038, "reading_distances": [31, 35, 36, 45], "reading_log_probs": [-0.7272003293037415, -0.7044439315795898, -0.7646722197532654, -0.384860634803772], "reading_logprob_std": 0.17538914660877475, "tokenization_path_logprob_std": 82.80595101731939, "trajectory_balance_rmse": 268.47578104725994} {"canonical_length": 6, "canonical_nll": 1.6695165634155273, "generation_rouge_l_f1": 0.47058823529411764, "generation_token_f1": 0.47058823529411764, "prompt_id": "9ecc7ea82de4a5d6", "reading_canonical_log_prob": -10.017099380493164, "reading_distances": [47, 46, 47, 40], "reading_log_probs": [-7.003138542175293, -6.4261064529418945, -6.5405988693237305, -7.445913314819336], "reading_logprob_std": 0.4668750077573366, "tokenization_path_logprob_std": 17.52453734316237, "trajectory_balance_rmse": 89.6175212171466} {"canonical_length": 268, "canonical_nll": 1.20481360136573, "generation_rouge_l_f1": 0.212, "generation_token_f1": 0.37200000000000005, "prompt_id": "aceb2f4e30eb8855", "reading_canonical_log_prob": -322.8900451660156, "reading_distances": [25, 27, 32, 27], "reading_log_probs": [-329.02520751953125, -324.9996337890625, -332.9351806640625, -323.88665771484375], "reading_logprob_std": 4.122956215460917, "tokenization_path_logprob_std": 203.69549132865126, "trajectory_balance_rmse": 5930.069029393113} {"canonical_length": 118, "canonical_nll": 1.5777726254220736, "generation_rouge_l_f1": 0.31316725978647686, "generation_token_f1": 0.4697508896797153, "prompt_id": "c575c5fadea559fc", "reading_canonical_log_prob": -186.1771697998047, "reading_distances": [53, 46, 38, 46], "reading_log_probs": [-170.29660034179688, -174.35586547851562, -172.5255889892578, -174.7538604736328], "reading_logprob_std": 2.0368827812552843, "tokenization_path_logprob_std": 147.97881647904833, "trajectory_balance_rmse": 3461.5949841627526} {"canonical_length": 61, "canonical_nll": 2.8949394851434427, "generation_rouge_l_f1": 0.3047619047619047, "generation_token_f1": 0.419047619047619, "prompt_id": "3adbad55ce76ffec", "reading_canonical_log_prob": -176.59130859375, "reading_distances": [25, 21, 24, 19], "reading_log_probs": [-172.91659545898438, -172.89686584472656, -172.56753540039062, -172.61605834960938], "reading_logprob_std": 0.18308026558887036, "tokenization_path_logprob_std": 108.1069894294587, "trajectory_balance_rmse": 1469.6756468683056} {"canonical_length": 12, "canonical_nll": 2.178410053253174, "generation_rouge_l_f1": 0.7, "generation_token_f1": 0.7, "prompt_id": "70c54675265e6b63", "reading_canonical_log_prob": -26.140920639038086, "reading_distances": [27, 34, 22, 28], "reading_log_probs": [-25.7175235748291, -25.665977478027344, -26.55718421936035, -25.790782928466797], "reading_logprob_std": 0.41934976709043686, "tokenization_path_logprob_std": 60.02511562585757, "trajectory_balance_rmse": 428.74105943516844} {"canonical_length": 11, "canonical_nll": 1.797500263560902, "generation_rouge_l_f1": 0.7142857142857143, "generation_token_f1": 0.7142857142857143, "prompt_id": "5e3e21467d43fd88", "reading_canonical_log_prob": -19.772502899169922, "reading_distances": [16, 20, 12, 17], "reading_log_probs": [-21.461414337158203, -21.890336990356445, -21.931745529174805, -21.98917579650879], "reading_logprob_std": 0.2412640743530109, "tokenization_path_logprob_std": 26.488474656868668, "trajectory_balance_rmse": 314.99138082530965} {"canonical_length": 17, "canonical_nll": 0.1541150036980124, "generation_rouge_l_f1": 0.14814814814814814, "generation_token_f1": 0.14814814814814814, "prompt_id": "46d8b242e85d5c4e", "reading_canonical_log_prob": -2.619955062866211, "reading_distances": [12, 8, 11, 10], "reading_log_probs": [-3.7646782398223877, -3.002365827560425, -3.82733154296875, -3.761561870574951], "reading_logprob_std": 0.39225077811437176, "tokenization_path_logprob_std": 25.866412914816532, "trajectory_balance_rmse": 173.5005442103066} {"canonical_length": 48, "canonical_nll": 1.451042652130127, "generation_rouge_l_f1": 0.087248322147651, "generation_token_f1": 0.15436241610738255, "prompt_id": "6509666af0922805", "reading_canonical_log_prob": -69.6500473022461, "reading_distances": [30, 28, 32, 23], "reading_log_probs": [-76.75125122070312, -78.17430877685547, -75.86054229736328, -72.65340423583984], "reading_logprob_std": 2.340413815330769, "tokenization_path_logprob_std": 114.16159431156548, "trajectory_balance_rmse": 1245.7035964690976} {"canonical_length": 175, "canonical_nll": 1.1814927455357143, "generation_rouge_l_f1": 0.3186274509803922, "generation_token_f1": 0.5294117647058824, "prompt_id": "42f6dd40e9d218d5", "reading_canonical_log_prob": -206.76123046875, "reading_distances": [8, 15, 12, 13], "reading_log_probs": [-209.134521484375, -206.2650909423828, -208.20608520507812, -207.85964965820312], "reading_logprob_std": 1.1955130539978864, "tokenization_path_logprob_std": 216.77094933670023, "trajectory_balance_rmse": 4075.3172223134675} {"canonical_length": 85, "canonical_nll": 1.111701067756204, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.2946708463949843, "prompt_id": "68880c3a241cb55b", "reading_canonical_log_prob": -94.49459075927734, "reading_distances": [35, 41, 42, 37], "reading_log_probs": [-98.02836608886719, -97.0603256225586, -100.45513916015625, -95.86836242675781], "reading_logprob_std": 1.9467066574413094, "tokenization_path_logprob_std": 175.25543902666516, "trajectory_balance_rmse": 2256.5446009044504} {"canonical_length": 14, "canonical_nll": 1.475299562726702, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.4666666666666667, "prompt_id": "a0d2e6c5b9eebb8b", "reading_canonical_log_prob": -20.654193878173828, "reading_distances": [26, 28, 25, 32], "reading_log_probs": [-23.136455535888672, -22.07075309753418, -22.121835708618164, -21.843307495117188], "reading_logprob_std": 0.5751322918313596, "tokenization_path_logprob_std": 54.46033985035012, "trajectory_balance_rmse": 441.2381350201268} {"canonical_length": 256, "canonical_nll": 0.5593691468238831, "generation_rouge_l_f1": 0.21611721611721613, "generation_token_f1": 0.43223443223443225, "prompt_id": "c51ea1a2a74eb3ff", "reading_canonical_log_prob": -143.19850158691406, "reading_distances": [28, 24, 27, 20], "reading_log_probs": [-144.65774536132812, -143.6055450439453, -144.4455108642578, -149.43399047851562], "reading_logprob_std": 2.638274910398004, "tokenization_path_logprob_std": 267.06291437195017, "trajectory_balance_rmse": 4289.758792569278} {"canonical_length": 61, "canonical_nll": 0.9046790951588115, "generation_rouge_l_f1": 0.16149068322981366, "generation_token_f1": 0.2670807453416149, "prompt_id": "9ce480763a8f4a02", "reading_canonical_log_prob": -55.1854248046875, "reading_distances": [30, 32, 30, 29], "reading_log_probs": [-61.410186767578125, -62.3670539855957, -61.48347473144531, -61.87757110595703], "reading_logprob_std": 0.4392258206733367, "tokenization_path_logprob_std": 94.78152679769676, "trajectory_balance_rmse": 1252.7151148740688} {"canonical_length": 17, "canonical_nll": 0.7097245384665096, "generation_rouge_l_f1": 0.13333333333333333, "generation_token_f1": 0.13333333333333333, "prompt_id": "690ac1ef381d5161", "reading_canonical_log_prob": -12.065317153930664, "reading_distances": [32, 37, 39, 41], "reading_log_probs": [-8.563095092773438, -11.29543399810791, -8.369405746459961, -12.081559181213379], "reading_logprob_std": 1.8894994543138357, "tokenization_path_logprob_std": 30.348099434091445, "trajectory_balance_rmse": 387.8673014039741} {"canonical_length": 49, "canonical_nll": 1.2636121633101483, "generation_rouge_l_f1": 0.3111111111111111, "generation_token_f1": 0.41481481481481486, "prompt_id": "f544199aa0f474b9", "reading_canonical_log_prob": -61.916996002197266, "reading_distances": [20, 18, 19, 19], "reading_log_probs": [-63.8045768737793, -61.48359680175781, -62.81761169433594, -62.104488372802734], "reading_logprob_std": 0.9968699605781363, "tokenization_path_logprob_std": 149.80505901380127, "trajectory_balance_rmse": 1430.4220088526813} {"canonical_length": 236, "canonical_nll": 0.5699441554182667, "generation_rouge_l_f1": 0.7301587301587301, "generation_token_f1": 0.8072562358276643, "prompt_id": "672d37d3ef4c74d7", "reading_canonical_log_prob": -134.50682067871094, "reading_distances": [20, 24, 25, 26], "reading_log_probs": [-141.71983337402344, -141.5166015625, -141.8927001953125, -144.05661010742188], "reading_logprob_std": 1.1834733945212124, "tokenization_path_logprob_std": 223.14492120189246, "trajectory_balance_rmse": 4904.608720572234} {"canonical_length": 76, "canonical_nll": 1.1369179173519737, "generation_rouge_l_f1": 0.22325581395348837, "generation_token_f1": 0.4837209302325581, "prompt_id": "e905bcd075fa835d", "reading_canonical_log_prob": -86.40576171875, "reading_distances": [19, 17, 18, 19], "reading_log_probs": [-88.87676239013672, -90.17523193359375, -87.55258178710938, -87.2060546875], "reading_logprob_std": 1.3553944688019823, "tokenization_path_logprob_std": 101.24699456192549, "trajectory_balance_rmse": 2152.2622818343857} {"canonical_length": 10, "canonical_nll": 0.9058901786804199, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.6666666666666666, "prompt_id": "1d805df4e7b3cce0", "reading_canonical_log_prob": -9.0589017868042, "reading_distances": [34, 41, 36, 34], "reading_log_probs": [-7.458908557891846, -9.408050537109375, -9.510082244873047, -8.820121765136719], "reading_logprob_std": 0.9439052529602034, "tokenization_path_logprob_std": 43.93954014343489, "trajectory_balance_rmse": 304.16632323127163} {"canonical_length": 140, "canonical_nll": 1.7177957807268416, "generation_rouge_l_f1": 0.16176470588235292, "generation_token_f1": 0.34803921568627455, "prompt_id": "6a438344f95ac9cf", "reading_canonical_log_prob": -240.4914093017578, "reading_distances": [25, 26, 18, 25], "reading_log_probs": [-241.9031982421875, -240.3523406982422, -244.13922119140625, -240.62521362304688], "reading_logprob_std": 1.7272660268685196, "tokenization_path_logprob_std": 345.8835527091611, "trajectory_balance_rmse": 4468.6355920578835} {"canonical_length": 103, "canonical_nll": 1.9623484194857403, "generation_rouge_l_f1": 0.25773195876288657, "generation_token_f1": 0.43298969072164945, "prompt_id": "208ec2e52d9d1ce6", "reading_canonical_log_prob": -202.12188720703125, "reading_distances": [32, 29, 36, 30], "reading_log_probs": [-202.30343627929688, -206.24710083007812, -201.576904296875, -197.66464233398438], "reading_logprob_std": 3.5163175724024316, "tokenization_path_logprob_std": 161.54247314287318, "trajectory_balance_rmse": 2782.4860262659745} {"canonical_length": 14, "canonical_nll": 0.21990656852722168, "generation_rouge_l_f1": 0.8, "generation_token_f1": 0.8, "prompt_id": "6a81e01870d01f92", "reading_canonical_log_prob": -3.0786919593811035, "reading_distances": [25, 28, 31, 24], "reading_log_probs": [-3.3973228931427, -3.460034132003784, -3.8218727111816406, -3.119300603866577], "reading_logprob_std": 0.2889789065782791, "tokenization_path_logprob_std": 51.15731491206429, "trajectory_balance_rmse": 452.73372851534083} {"canonical_length": 274, "canonical_nll": 1.3088773184449134, "generation_rouge_l_f1": 0.19521178637200734, "generation_token_f1": 0.41252302025782683, "prompt_id": "7857d9054d4c3033", "reading_canonical_log_prob": -358.63238525390625, "reading_distances": [29, 31, 23, 30], "reading_log_probs": [-365.96417236328125, -366.4139404296875, -366.02166748046875, -366.7718505859375], "reading_logprob_std": 0.376684389529085, "tokenization_path_logprob_std": 209.4672083658752, "trajectory_balance_rmse": 5710.861239699183} {"canonical_length": 158, "canonical_nll": 1.5481905393962618, "generation_rouge_l_f1": 0.24358974358974358, "generation_token_f1": 0.40384615384615385, "prompt_id": "77515334ad84a428", "reading_canonical_log_prob": -244.61410522460938, "reading_distances": [19, 18, 20, 16], "reading_log_probs": [-245.45501708984375, -246.65597534179688, -243.82310485839844, -241.8485107421875], "reading_logprob_std": 2.084630648055394, "tokenization_path_logprob_std": 220.91420678494902, "trajectory_balance_rmse": 3647.980657477055} {"canonical_length": 43, "canonical_nll": 0.7659441925758539, "generation_rouge_l_f1": 0.22535211267605634, "generation_token_f1": 0.2535211267605633, "prompt_id": "78695f86d2d683db", "reading_canonical_log_prob": -32.93560028076172, "reading_distances": [40, 37, 35, 31], "reading_log_probs": [-34.94581985473633, -38.60866165161133, -35.07939529418945, -35.70450210571289], "reading_logprob_std": 1.7148973098600375, "tokenization_path_logprob_std": 178.33903471937597, "trajectory_balance_rmse": 1647.6771320443306} {"canonical_length": 8, "canonical_nll": 0.2760107219219208, "generation_rouge_l_f1": 0.14953271028037382, "generation_token_f1": 0.14953271028037382, "prompt_id": "73a081a89d8a40bd", "reading_canonical_log_prob": -2.208085775375366, "reading_distances": [28, 28, 32, 25], "reading_log_probs": [-1.612440824508667, -1.5431451797485352, -2.266007423400879, -1.125016689300537], "reading_logprob_std": 0.47158903244588113, "tokenization_path_logprob_std": 42.77121388050049, "trajectory_balance_rmse": 149.86086435051334} {"canonical_length": 84, "canonical_nll": 1.234274637131464, "generation_rouge_l_f1": 0.17252396166134185, "generation_token_f1": 0.30031948881789133, "prompt_id": "7879ca5705298aac", "reading_canonical_log_prob": -103.67906951904297, "reading_distances": [46, 46, 41, 35], "reading_log_probs": [-112.26181030273438, -111.98670959472656, -113.31034088134766, -108.50196838378906], "reading_logprob_std": 2.088208466499017, "tokenization_path_logprob_std": 160.5214830160196, "trajectory_balance_rmse": 1749.3488233806258} {"canonical_length": 355, "canonical_nll": 1.362522694762324, "generation_rouge_l_f1": 0.17610062893081763, "generation_token_f1": 0.3742138364779874, "prompt_id": "2e88e8dc9b895643", "reading_canonical_log_prob": -483.695556640625, "reading_distances": [23, 20, 20, 21], "reading_log_probs": [-481.7052001953125, -483.96954345703125, -481.36895751953125, -480.80072021484375], "reading_logprob_std": 1.3900172412589398, "tokenization_path_logprob_std": 90.1029673732847, "trajectory_balance_rmse": 10419.953555598622} {"canonical_length": 343, "canonical_nll": 1.4502320581552934, "generation_rouge_l_f1": 0.1986754966887417, "generation_token_f1": 0.423841059602649, "prompt_id": "a81b52bde10fe226", "reading_canonical_log_prob": -497.4295959472656, "reading_distances": [19, 20, 20, 19], "reading_log_probs": [-496.8741455078125, -500.23663330078125, -496.7639465332031, -496.4499816894531], "reading_logprob_std": 1.7794016000487138, "tokenization_path_logprob_std": 162.77765952447032, "trajectory_balance_rmse": 9017.578358244584} {"canonical_length": 269, "canonical_nll": 1.2760799365415891, "generation_rouge_l_f1": 0.34317343173431736, "generation_token_f1": 0.5498154981549815, "prompt_id": "ff000a10ec04c16c", "reading_canonical_log_prob": -343.2655029296875, "reading_distances": [33, 25, 32, 26], "reading_log_probs": [-343.2071533203125, -344.9923095703125, -342.6618347167969, -347.2706298828125], "reading_logprob_std": 2.078838745585078, "tokenization_path_logprob_std": 527.6127686722552, "trajectory_balance_rmse": 7654.669920846681} {"canonical_length": 405, "canonical_nll": 1.3014484170042437, "generation_rouge_l_f1": 0.21909233176838813, "generation_token_f1": 0.4131455399061033, "prompt_id": "e7f10884569482cf", "reading_canonical_log_prob": -527.0866088867188, "reading_distances": [31, 38, 41, 31], "reading_log_probs": [-525.8197021484375, -523.5132446289062, -523.7188110351562, -527.078857421875], "reading_logprob_std": 1.7167004596889377, "tokenization_path_logprob_std": 236.97675520344188, "trajectory_balance_rmse": 8587.690706623227} {"canonical_length": 10, "canonical_nll": 1.1778715133666993, "generation_rouge_l_f1": 0.7000000000000001, "generation_token_f1": 0.7000000000000001, "prompt_id": "7d1f1626ed9a7e46", "reading_canonical_log_prob": -11.778715133666992, "reading_distances": [34, 38, 35, 40], "reading_log_probs": [-16.136371612548828, -12.801141738891602, -14.866290092468262, -13.622274398803711], "reading_logprob_std": 1.4590014219631275, "tokenization_path_logprob_std": 39.06285741713642, "trajectory_balance_rmse": 352.76388641602114} {"canonical_length": 41, "canonical_nll": 1.6735045270221989, "generation_rouge_l_f1": 0.17582417582417584, "generation_token_f1": 0.21978021978021978, "prompt_id": "8de4f5d47e6be004", "reading_canonical_log_prob": -68.61368560791016, "reading_distances": [13, 16, 15, 15], "reading_log_probs": [-74.20664978027344, -71.94999694824219, -72.18341064453125, -68.3458023071289], "reading_logprob_std": 2.437680769777616, "tokenization_path_logprob_std": 130.4614411057854, "trajectory_balance_rmse": 1178.887176095629} {"canonical_length": 187, "canonical_nll": 1.303601106857871, "generation_rouge_l_f1": 0.351219512195122, "generation_token_f1": 0.5463414634146342, "prompt_id": "37dc889754565fdd", "reading_canonical_log_prob": -243.77340698242188, "reading_distances": [32, 40, 40, 36], "reading_log_probs": [-242.259033203125, -240.5120391845703, -242.26528930664062, -240.580322265625], "reading_logprob_std": 0.9911172210518298, "tokenization_path_logprob_std": 203.27281168303736, "trajectory_balance_rmse": 5276.644295448492} {"canonical_length": 81, "canonical_nll": 1.476715276270737, "generation_rouge_l_f1": 0.2758620689655173, "generation_token_f1": 0.4224137931034483, "prompt_id": "fccb0a4a57193f97", "reading_canonical_log_prob": -119.61393737792969, "reading_distances": [24, 26, 24, 21], "reading_log_probs": [-117.69125366210938, -117.31365203857422, -117.2457046508789, -117.71676635742188], "reading_logprob_std": 0.24677333358987955, "tokenization_path_logprob_std": 204.65916035692797, "trajectory_balance_rmse": 2408.4453429018204} {"canonical_length": 8, "canonical_nll": 0.022073661908507347, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "a29eafa220e503a1", "reading_canonical_log_prob": -0.17658929526805878, "reading_distances": [35, 27, 28, 37], "reading_log_probs": [-0.1149541512131691, -0.2255191206932068, -0.1759863644838333, -0.14030945301055908], "reading_logprob_std": 0.047940465057046334, "tokenization_path_logprob_std": 56.0878526857905, "trajectory_balance_rmse": 215.82421049584983} {"canonical_length": 20, "canonical_nll": 1.6929689407348634, "generation_rouge_l_f1": 0.2926829268292683, "generation_token_f1": 0.4878048780487805, "prompt_id": "276e3276fb9b91f2", "reading_canonical_log_prob": -33.859378814697266, "reading_distances": [40, 49, 46, 48], "reading_log_probs": [-30.961273193359375, -34.13962936401367, -32.39274597167969, -37.7224235534668], "reading_logprob_std": 2.9177287880984943, "tokenization_path_logprob_std": 45.08353054042505, "trajectory_balance_rmse": 558.0482009974564} {"canonical_length": 229, "canonical_nll": 1.3138936804892194, "generation_rouge_l_f1": 0.2115768463073852, "generation_token_f1": 0.4231536926147704, "prompt_id": "14d153fd0e1b36b1", "reading_canonical_log_prob": -300.88165283203125, "reading_distances": [15, 21, 16, 22], "reading_log_probs": [-302.2176513671875, -301.339111328125, -299.685302734375, -298.07177734375], "reading_logprob_std": 1.8345506269238852, "tokenization_path_logprob_std": 173.1307946376693, "trajectory_balance_rmse": 5745.603410632138} {"canonical_length": 315, "canonical_nll": 0.5777182927207342, "generation_rouge_l_f1": 0.36996336996337004, "generation_token_f1": 0.4981684981684982, "prompt_id": "0ee97cbd046603bc", "reading_canonical_log_prob": -181.98126220703125, "reading_distances": [25, 28, 24, 29], "reading_log_probs": [-183.89041137695312, -183.37509155273438, -183.17080688476562, -185.40365600585938], "reading_logprob_std": 1.0089481163131173, "tokenization_path_logprob_std": 279.0397837982936, "trajectory_balance_rmse": 6360.656358401767} {"canonical_length": 282, "canonical_nll": 0.9723348414644282, "generation_rouge_l_f1": 0.32667876588021777, "generation_token_f1": 0.5807622504537204, "prompt_id": "740c83c01085ccfc", "reading_canonical_log_prob": -274.19842529296875, "reading_distances": [16, 17, 28, 24], "reading_log_probs": [-281.2405700683594, -277.3754577636719, -281.6573181152344, -280.9577941894531], "reading_logprob_std": 1.9758865166597703, "tokenization_path_logprob_std": 482.0544574870683, "trajectory_balance_rmse": 9155.341861334662} {"canonical_length": 128, "canonical_nll": 1.4359228610992432, "generation_rouge_l_f1": 0.23711340206185566, "generation_token_f1": 0.3917525773195876, "prompt_id": "055891d31fa97386", "reading_canonical_log_prob": -183.79812622070312, "reading_distances": [26, 24, 21, 22], "reading_log_probs": [-181.45372009277344, -184.3058624267578, -182.844970703125, -189.31260681152344], "reading_logprob_std": 3.4261786950560875, "tokenization_path_logprob_std": 476.26977533880813, "trajectory_balance_rmse": 3433.9528207715416} {"canonical_length": 197, "canonical_nll": 1.3848275896256346, "generation_rouge_l_f1": 0.1782608695652174, "generation_token_f1": 0.4956521739130434, "prompt_id": "54415ac83a3ddec9", "reading_canonical_log_prob": -272.81103515625, "reading_distances": [20, 19, 19, 19], "reading_log_probs": [-270.70623779296875, -269.6048889160156, -268.7908935546875, -271.6512451171875], "reading_logprob_std": 1.2518760905851352, "tokenization_path_logprob_std": 158.3569820565142, "trajectory_balance_rmse": 4511.5993476838} {"canonical_length": 71, "canonical_nll": 1.8406028210277288, "generation_rouge_l_f1": 0.22666666666666663, "generation_token_f1": 0.45333333333333325, "prompt_id": "6bb4c6466fd9ddad", "reading_canonical_log_prob": -130.68280029296875, "reading_distances": [11, 17, 13, 18], "reading_log_probs": [-128.35830688476562, -128.6893310546875, -130.2526092529297, -132.62237548828125], "reading_logprob_std": 1.9452658081122352, "tokenization_path_logprob_std": 140.05973920017283, "trajectory_balance_rmse": 1993.336245067784} {"canonical_length": 96, "canonical_nll": 0.9821522235870361, "generation_rouge_l_f1": 0.2689655172413793, "generation_token_f1": 0.39999999999999997, "prompt_id": "5ca3d0261fdb6645", "reading_canonical_log_prob": -94.28661346435547, "reading_distances": [20, 27, 25, 23], "reading_log_probs": [-101.722900390625, -101.65205383300781, -101.32291412353516, -102.78863525390625], "reading_logprob_std": 0.6356943548920444, "tokenization_path_logprob_std": 257.0071804496394, "trajectory_balance_rmse": 2537.3489938748394} {"canonical_length": 120, "canonical_nll": 1.4356981913248699, "generation_rouge_l_f1": 0.2888283378746594, "generation_token_f1": 0.44141689373297005, "prompt_id": "2749992f70103af2", "reading_canonical_log_prob": -172.28378295898438, "reading_distances": [38, 44, 37, 44], "reading_log_probs": [-182.0706329345703, -186.07623291015625, -182.8633270263672, -183.58255004882812], "reading_logprob_std": 1.732473872698019, "tokenization_path_logprob_std": 228.60155263318663, "trajectory_balance_rmse": 3684.3107269374646} {"canonical_length": 18, "canonical_nll": 0.11960886584387885, "generation_rouge_l_f1": 0.5185185185185185, "generation_token_f1": 0.5185185185185185, "prompt_id": "67a8cb2f3ba290b7", "reading_canonical_log_prob": -2.1529595851898193, "reading_distances": [39, 50, 47, 39], "reading_log_probs": [-2.810774803161621, -1.9943900108337402, -4.9586005210876465, -2.364337682723999], "reading_logprob_std": 1.3270440349948414, "tokenization_path_logprob_std": 76.22492815944733, "trajectory_balance_rmse": 520.3965595434746} {"canonical_length": 144, "canonical_nll": 1.1113583246866863, "generation_rouge_l_f1": 0.26392961876832843, "generation_token_f1": 0.4281524926686217, "prompt_id": "14aa2b70fa0eb52a", "reading_canonical_log_prob": -160.0355987548828, "reading_distances": [32, 26, 28, 30], "reading_log_probs": [-168.12188720703125, -171.231689453125, -163.55372619628906, -167.9735107421875], "reading_logprob_std": 3.1578252580307815, "tokenization_path_logprob_std": 284.97734584404907, "trajectory_balance_rmse": 3283.391870187473} {"canonical_length": 46, "canonical_nll": 1.8018916586171025, "generation_rouge_l_f1": 0.11074918566775244, "generation_token_f1": 0.1498371335504886, "prompt_id": "632380e80642e576", "reading_canonical_log_prob": -82.88701629638672, "reading_distances": [26, 32, 28, 25], "reading_log_probs": [-87.22359466552734, -93.1947021484375, -88.24136352539062, -86.5260009765625], "reading_logprob_std": 3.015601411655085, "tokenization_path_logprob_std": 111.77447105152142, "trajectory_balance_rmse": 1285.225057138638} {"canonical_length": 186, "canonical_nll": 1.512322046423471, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.5152224824355972, "prompt_id": "e0e28697fbbba739", "reading_canonical_log_prob": -281.2919006347656, "reading_distances": [36, 28, 34, 28], "reading_log_probs": [-280.3227233886719, -280.1826171875, -284.8193359375, -276.08795166015625], "reading_logprob_std": 3.566919348922179, "tokenization_path_logprob_std": 423.5002022428041, "trajectory_balance_rmse": 5573.634747027098} {"canonical_length": 41, "canonical_nll": 1.208273445687643, "generation_rouge_l_f1": 0.20809248554913293, "generation_token_f1": 0.28901734104046245, "prompt_id": "dd2bdffacb538a3a", "reading_canonical_log_prob": -49.53921127319336, "reading_distances": [145, 138, 159, 126], "reading_log_probs": [-66.19252014160156, -57.61344909667969, -66.48567962646484, -62.74932098388672], "reading_logprob_std": 4.129124740168853, "tokenization_path_logprob_std": 109.57624241710077, "trajectory_balance_rmse": 1282.3638599094636} {"canonical_length": 11, "canonical_nll": 1.022945924238725, "generation_rouge_l_f1": 0.0622568093385214, "generation_token_f1": 0.0622568093385214, "prompt_id": "fbdd1075d5b5f58e", "reading_canonical_log_prob": -11.252405166625977, "reading_distances": [25, 19, 24, 15], "reading_log_probs": [-13.918963432312012, -13.119829177856445, -13.503959655761719, -13.316648483276367], "reading_logprob_std": 0.3409553401115047, "tokenization_path_logprob_std": 49.53077968538298, "trajectory_balance_rmse": 238.49236007608138} {"canonical_length": 25, "canonical_nll": 2.6603546142578125, "generation_rouge_l_f1": 0.22727272727272727, "generation_token_f1": 0.29545454545454547, "prompt_id": "71ff4b9aa3b9c919", "reading_canonical_log_prob": -66.50886535644531, "reading_distances": [30, 22, 22, 23], "reading_log_probs": [-61.371456146240234, -63.8125, -61.080169677734375, -61.530548095703125], "reading_logprob_std": 1.2564708298101335, "tokenization_path_logprob_std": 152.00216842844745, "trajectory_balance_rmse": 697.4027234547247} {"canonical_length": 227, "canonical_nll": 0.6497582927149298, "generation_rouge_l_f1": 0.2592592592592593, "generation_token_f1": 0.4773662551440329, "prompt_id": "84c9e255e3abf202", "reading_canonical_log_prob": -147.49513244628906, "reading_distances": [29, 23, 23, 24], "reading_log_probs": [-153.61541748046875, -152.5644989013672, -154.0198974609375, -150.76698303222656], "reading_logprob_std": 1.4523633977062222, "tokenization_path_logprob_std": 146.9735407844363, "trajectory_balance_rmse": 5737.843908851935} {"canonical_length": 20, "canonical_nll": 0.44307408332824705, "generation_rouge_l_f1": 0.30952380952380953, "generation_token_f1": 0.33333333333333337, "prompt_id": "7aac721bb785059a", "reading_canonical_log_prob": -8.861481666564941, "reading_distances": [19, 22, 28, 23], "reading_log_probs": [-7.863408088684082, -9.53589916229248, -8.72900390625, -8.272751808166504], "reading_logprob_std": 0.7169862830412678, "tokenization_path_logprob_std": 81.55216171136588, "trajectory_balance_rmse": 532.2806060444354} {"canonical_length": 16, "canonical_nll": 0.6599107384681702, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "e8e5a8f77a9d33c4", "reading_canonical_log_prob": -10.558571815490723, "reading_distances": [43, 41, 42, 43], "reading_log_probs": [-9.86374282836914, -9.830755233764648, -9.113779067993164, -10.414325714111328], "reading_logprob_std": 0.5332843397204486, "tokenization_path_logprob_std": 50.24174930590651, "trajectory_balance_rmse": 332.33526476851506} {"canonical_length": 139, "canonical_nll": 1.6968556136536084, "generation_rouge_l_f1": 0.11990407673860912, "generation_token_f1": 0.1870503597122302, "prompt_id": "b0bb5fcadef34afc", "reading_canonical_log_prob": -235.86293029785156, "reading_distances": [16, 10, 16, 17], "reading_log_probs": [-245.3663787841797, -241.2401885986328, -237.60885620117188, -244.64584350585938], "reading_logprob_std": 3.559354152731252, "tokenization_path_logprob_std": 132.68921010978303, "trajectory_balance_rmse": 3442.843364152893} {"canonical_length": 214, "canonical_nll": 1.612067819755768, "generation_rouge_l_f1": 0.21390374331550804, "generation_token_f1": 0.41711229946524064, "prompt_id": "6db457998b4e4d8a", "reading_canonical_log_prob": -344.9825134277344, "reading_distances": [39, 38, 39, 35], "reading_log_probs": [-334.0236511230469, -334.2428283691406, -329.1206359863281, -330.78662109375], "reading_logprob_std": 2.5087129738477643, "tokenization_path_logprob_std": 319.42953423883034, "trajectory_balance_rmse": 4910.26284396942} {"canonical_length": 124, "canonical_nll": 1.6917859969600555, "generation_rouge_l_f1": 0.1657142857142857, "generation_token_f1": 0.2571428571428571, "prompt_id": "b7fe2d1d82e7517d", "reading_canonical_log_prob": -209.78146362304688, "reading_distances": [21, 28, 32, 26], "reading_log_probs": [-205.15521240234375, -207.64610290527344, -207.99520874023438, -208.99041748046875], "reading_logprob_std": 1.6303989717708791, "tokenization_path_logprob_std": 91.7827707761474, "trajectory_balance_rmse": 2888.2289770029834} {"canonical_length": 3, "canonical_nll": 0.020743281890948612, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "fe8dff1c5ccd0457", "reading_canonical_log_prob": -0.06222984567284584, "reading_distances": [57, 77, 54, 55], "reading_log_probs": [-0.2828231155872345, -1.4944555759429932, -0.2467963695526123, -0.23957650363445282], "reading_logprob_std": 0.6193174013067049, "tokenization_path_logprob_std": 17.63307972999859, "trajectory_balance_rmse": 93.89769988380563} {"canonical_length": 36, "canonical_nll": 1.6857824325561523, "generation_rouge_l_f1": 0.11347517730496452, "generation_token_f1": 0.15602836879432624, "prompt_id": "007cda46c19b41d2", "reading_canonical_log_prob": -60.688167572021484, "reading_distances": [22, 16, 22, 16], "reading_log_probs": [-63.71944046020508, -66.76129150390625, -62.191673278808594, -63.87947463989258], "reading_logprob_std": 1.907168360568042, "tokenization_path_logprob_std": 167.8101010542213, "trajectory_balance_rmse": 1121.8336202578669} {"canonical_length": 121, "canonical_nll": 1.7521073522646564, "generation_rouge_l_f1": 0.19186046511627905, "generation_token_f1": 0.4011627906976744, "prompt_id": "0f24b7ddfdbc9dca", "reading_canonical_log_prob": -212.00498962402344, "reading_distances": [20, 21, 20, 22], "reading_log_probs": [-221.19351196289062, -222.31362915039062, -223.46051025390625, -222.13519287109375], "reading_logprob_std": 0.9302485615249311, "tokenization_path_logprob_std": 100.2266348757195, "trajectory_balance_rmse": 2605.430568443319} {"canonical_length": 4, "canonical_nll": 0.3952106237411499, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "0131c4e6a39cb194", "reading_canonical_log_prob": -1.5808424949645996, "reading_distances": [17, 27, 22, 21], "reading_log_probs": [-1.4810030460357666, -1.3867759704589844, -1.4037529230117798, -1.3196628093719482], "reading_logprob_std": 0.06629513290754253, "tokenization_path_logprob_std": 4.870989203684813, "trajectory_balance_rmse": 61.125724527803136} {"canonical_length": 182, "canonical_nll": 1.1335741818606198, "generation_rouge_l_f1": 0.18527315914489315, "generation_token_f1": 0.31353919239904987, "prompt_id": "21b0719467934ebb", "reading_canonical_log_prob": -206.3105010986328, "reading_distances": [46, 42, 48, 46], "reading_log_probs": [-203.36911010742188, -199.85580444335938, -195.03152465820312, -188.98228454589844], "reading_logprob_std": 6.237920419312595, "tokenization_path_logprob_std": 264.9606778402592, "trajectory_balance_rmse": 4066.7881395199233} {"canonical_length": 246, "canonical_nll": 1.4687829986820375, "generation_rouge_l_f1": 0.23366336633663365, "generation_token_f1": 0.47920792079207924, "prompt_id": "5d16c846337cb10b", "reading_canonical_log_prob": -361.32061767578125, "reading_distances": [17, 19, 21, 15], "reading_log_probs": [-362.8782958984375, -359.806396484375, -359.42657470703125, -361.2255859375], "reading_logprob_std": 1.5673012926168406, "tokenization_path_logprob_std": 290.9482334363843, "trajectory_balance_rmse": 7025.800110385049} {"canonical_length": 116, "canonical_nll": 0.6149775406409954, "generation_rouge_l_f1": 0.3384615384615384, "generation_token_f1": 0.4861538461538461, "prompt_id": "d66394bc1b63c0cb", "reading_canonical_log_prob": -71.33739471435547, "reading_distances": [22, 14, 17, 19], "reading_log_probs": [-70.97148895263672, -71.95623779296875, -71.317138671875, -70.3832015991211], "reading_logprob_std": 0.6576736162954021, "tokenization_path_logprob_std": 62.44485482282373, "trajectory_balance_rmse": 1566.852456819989} {"canonical_length": 75, "canonical_nll": 1.0141935221354166, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.34285714285714286, "prompt_id": "4b9d4e7ec5d5b8b5", "reading_canonical_log_prob": -76.06451416015625, "reading_distances": [39, 32, 36, 40], "reading_log_probs": [-73.06172943115234, -73.15713500976562, -73.89205169677734, -72.65139770507812], "reading_logprob_std": 0.5165594915694834, "tokenization_path_logprob_std": 179.43028179617826, "trajectory_balance_rmse": 1984.0715453015275} {"canonical_length": 363, "canonical_nll": 1.3262544321948175, "generation_rouge_l_f1": 0.30990415335463256, "generation_token_f1": 0.4856230031948882, "prompt_id": "cddcde885956c90e", "reading_canonical_log_prob": -481.43035888671875, "reading_distances": [21, 21, 25, 24], "reading_log_probs": [-481.9728698730469, -481.8173828125, -480.00787353515625, -481.31396484375], "reading_logprob_std": 0.8922429907257795, "tokenization_path_logprob_std": 245.87039623595192, "trajectory_balance_rmse": 10356.718631326403} {"canonical_length": 204, "canonical_nll": 1.7727743410596661, "generation_rouge_l_f1": 0.15111111111111108, "generation_token_f1": 0.32, "prompt_id": "e708ffc93eb420d5", "reading_canonical_log_prob": -361.6459655761719, "reading_distances": [40, 42, 33, 39], "reading_log_probs": [-358.84130859375, -359.318115234375, -355.9862365722656, -356.744873046875], "reading_logprob_std": 1.6091494362386334, "tokenization_path_logprob_std": 75.60321896726335, "trajectory_balance_rmse": 5441.77302027256} {"canonical_length": 73, "canonical_nll": 2.1776438673881637, "generation_rouge_l_f1": 0.18404907975460125, "generation_token_f1": 0.25766871165644173, "prompt_id": "276028c9e0375416", "reading_canonical_log_prob": -158.96800231933594, "reading_distances": [30, 25, 29, 25], "reading_log_probs": [-168.7662353515625, -162.99227905273438, -162.9929962158203, -167.09188842773438], "reading_logprob_std": 2.93087037680185, "tokenization_path_logprob_std": 260.9247571950499, "trajectory_balance_rmse": 2282.8114558358584} {"canonical_length": 100, "canonical_nll": 1.1513768005371094, "generation_rouge_l_f1": 0.4421052631578947, "generation_token_f1": 0.5684210526315788, "prompt_id": "328ee04612e4bcf9", "reading_canonical_log_prob": -115.13768005371094, "reading_distances": [28, 27, 25, 29], "reading_log_probs": [-127.67777252197266, -118.54916381835938, -118.78245544433594, -120.88165283203125], "reading_logprob_std": 4.267580943086556, "tokenization_path_logprob_std": 139.69848607040075, "trajectory_balance_rmse": 2081.8336777211607} {"canonical_length": 90, "canonical_nll": 1.4493204752604167, "generation_rouge_l_f1": 0.14245014245014245, "generation_token_f1": 0.2336182336182336, "prompt_id": "bdf95ab3a5ffa3ff", "reading_canonical_log_prob": -130.4388427734375, "reading_distances": [25, 34, 30, 30], "reading_log_probs": [-119.4862060546875, -114.62875366210938, -111.03861999511719, -111.1322021484375], "reading_logprob_std": 3.977746693582748, "tokenization_path_logprob_std": 108.32002352565252, "trajectory_balance_rmse": 1428.320293922467} {"canonical_length": 132, "canonical_nll": 1.9643977772105823, "generation_rouge_l_f1": 0.24675324675324675, "generation_token_f1": 0.4480519480519481, "prompt_id": "80019145cee29a1a", "reading_canonical_log_prob": -259.3005065917969, "reading_distances": [18, 18, 19, 24], "reading_log_probs": [-265.005615234375, -261.1194152832031, -262.9706115722656, -260.543212890625], "reading_logprob_std": 2.016763618198609, "tokenization_path_logprob_std": 141.06792651579698, "trajectory_balance_rmse": 3461.3542708106597} {"canonical_length": 250, "canonical_nll": 1.47932470703125, "generation_rouge_l_f1": 0.2235528942115769, "generation_token_f1": 0.4870259481037924, "prompt_id": "77663077eb493ddd", "reading_canonical_log_prob": -369.8311767578125, "reading_distances": [31, 30, 29, 30], "reading_log_probs": [-370.5291748046875, -368.9122619628906, -370.3162536621094, -371.83905029296875], "reading_logprob_std": 1.1983220402309007, "tokenization_path_logprob_std": 120.90362394203427, "trajectory_balance_rmse": 6123.455656677111} {"canonical_length": 369, "canonical_nll": 1.3057753813622122, "generation_rouge_l_f1": 0.27786752827140543, "generation_token_f1": 0.5589660743134087, "prompt_id": "4825c22b71594c0c", "reading_canonical_log_prob": -481.83111572265625, "reading_distances": [32, 32, 36, 32], "reading_log_probs": [-480.59075927734375, -483.3409423828125, -482.6903381347656, -477.48748779296875], "reading_logprob_std": 2.635619797440902, "tokenization_path_logprob_std": 183.08696908347517, "trajectory_balance_rmse": 9447.898131133947} {"canonical_length": 198, "canonical_nll": 1.157041761610243, "generation_rouge_l_f1": 0.32786885245901637, "generation_token_f1": 0.5620608899297423, "prompt_id": "c520430df28e5bdb", "reading_canonical_log_prob": -229.09426879882812, "reading_distances": [17, 21, 22, 21], "reading_log_probs": [-228.62213134765625, -228.729736328125, -229.42318725585938, -228.83358764648438], "reading_logprob_std": 0.3579187700068621, "tokenization_path_logprob_std": 317.4491994863267, "trajectory_balance_rmse": 5102.69820183111} {"canonical_length": 290, "canonical_nll": 1.4652554216056035, "generation_rouge_l_f1": 0.21206581352833637, "generation_token_f1": 0.4643510054844607, "prompt_id": "c3e62d8cbe21b9a4", "reading_canonical_log_prob": -424.924072265625, "reading_distances": [25, 23, 28, 26], "reading_log_probs": [-425.884765625, -426.7178649902344, -426.2341003417969, -424.8399658203125], "reading_logprob_std": 0.7964421154973171, "tokenization_path_logprob_std": 242.49455650682677, "trajectory_balance_rmse": 8161.872517517709} {"canonical_length": 38, "canonical_nll": 0.8079804872211657, "generation_rouge_l_f1": 0.19594594594594594, "generation_token_f1": 0.22972972972972977, "prompt_id": "ef7fdf70b61d5a5a", "reading_canonical_log_prob": -30.703258514404297, "reading_distances": [34, 27, 27, 31], "reading_log_probs": [-30.861486434936523, -31.04925537109375, -35.43348693847656, -31.699480056762695], "reading_logprob_std": 2.1453003880372457, "tokenization_path_logprob_std": 139.7851402228828, "trajectory_balance_rmse": 901.3216675140221} {"canonical_length": 262, "canonical_nll": 1.00257908478948, "generation_rouge_l_f1": 0.27680311890838205, "generation_token_f1": 0.5886939571150097, "prompt_id": "f0bfbb195bb03bbd", "reading_canonical_log_prob": -262.67572021484375, "reading_distances": [38, 26, 27, 32], "reading_log_probs": [-271.5803527832031, -268.0647888183594, -267.94024658203125, -269.7115478515625], "reading_logprob_std": 1.7070156533345675, "tokenization_path_logprob_std": 219.29676624128243, "trajectory_balance_rmse": 7529.841916354509} {"canonical_length": 12, "canonical_nll": 0.4370686213175456, "generation_rouge_l_f1": 0.7000000000000001, "generation_token_f1": 0.7000000000000001, "prompt_id": "064f3342c7039feb", "reading_canonical_log_prob": -5.244823455810547, "reading_distances": [32, 31, 33, 30], "reading_log_probs": [-6.712794303894043, -7.31895637512207, -8.719356536865234, -8.026961326599121], "reading_logprob_std": 0.8690302145913892, "tokenization_path_logprob_std": 73.42504967091851, "trajectory_balance_rmse": 391.6611452664219} {"canonical_length": 174, "canonical_nll": 1.5810676662401222, "generation_rouge_l_f1": 0.20754716981132076, "generation_token_f1": 0.41037735849056606, "prompt_id": "5319439232a42bef", "reading_canonical_log_prob": -275.10577392578125, "reading_distances": [27, 27, 25, 31], "reading_log_probs": [-268.132568359375, -267.1974792480469, -266.75152587890625, -266.4429931640625], "reading_logprob_std": 0.7359602805236718, "tokenization_path_logprob_std": 298.29271240870776, "trajectory_balance_rmse": 5380.453666894242} {"canonical_length": 274, "canonical_nll": 1.253873170727361, "generation_rouge_l_f1": 0.23783783783783782, "generation_token_f1": 0.5225225225225225, "prompt_id": "7f189f92ecaccbe7", "reading_canonical_log_prob": -343.5612487792969, "reading_distances": [21, 13, 20, 19], "reading_log_probs": [-341.63616943359375, -340.213134765625, -342.1609802246094, -343.1676025390625], "reading_logprob_std": 1.2309214580093644, "tokenization_path_logprob_std": 123.37290564152072, "trajectory_balance_rmse": 7582.04401701349} {"canonical_length": 225, "canonical_nll": 1.1644333224826389, "generation_rouge_l_f1": 0.47844827586206895, "generation_token_f1": 0.685344827586207, "prompt_id": "89ab6e8e7235950e", "reading_canonical_log_prob": -261.99749755859375, "reading_distances": [30, 18, 26, 22], "reading_log_probs": [-264.5504150390625, -261.3412170410156, -262.76239013671875, -262.6505126953125], "reading_logprob_std": 1.3182095937945175, "tokenization_path_logprob_std": 305.07901780115475, "trajectory_balance_rmse": 6436.297764616131} {"canonical_length": 188, "canonical_nll": 1.4606128449135638, "generation_rouge_l_f1": 0.3438395415472779, "generation_token_f1": 0.5616045845272206, "prompt_id": "3d38b4ece7a08884", "reading_canonical_log_prob": -274.59521484375, "reading_distances": [26, 21, 21, 29], "reading_log_probs": [-281.8724670410156, -278.42755126953125, -283.1600036621094, -285.78076171875], "reading_logprob_std": 3.0568794663465106, "tokenization_path_logprob_std": 120.81657719363889, "trajectory_balance_rmse": 5055.724764798094} {"canonical_length": 169, "canonical_nll": 1.9032412196052144, "generation_rouge_l_f1": 0.2631578947368421, "generation_token_f1": 0.3909774436090226, "prompt_id": "49135037e7b65f9f", "reading_canonical_log_prob": -321.64776611328125, "reading_distances": [10, 15, 17, 16], "reading_log_probs": [-316.96307373046875, -322.03924560546875, -327.497802734375, -320.06585693359375], "reading_logprob_std": 4.428122503400351, "tokenization_path_logprob_std": 273.37691526478994, "trajectory_balance_rmse": 5898.3508853260455} {"canonical_length": 250, "canonical_nll": 1.4732435302734375, "generation_rouge_l_f1": 0.2666666666666667, "generation_token_f1": 0.5058823529411766, "prompt_id": "daaf9a26e9fd85ac", "reading_canonical_log_prob": -368.3108825683594, "reading_distances": [37, 30, 33, 34], "reading_log_probs": [-364.5060119628906, -361.47088623046875, -364.00048828125, -364.1839599609375], "reading_logprob_std": 1.3953666589763725, "tokenization_path_logprob_std": 157.07988191383066, "trajectory_balance_rmse": 8758.332308299628} {"canonical_length": 32, "canonical_nll": 1.475199818611145, "generation_rouge_l_f1": 0.13620071684587812, "generation_token_f1": 0.17921146953405015, "prompt_id": "06a9af54aedbd3a8", "reading_canonical_log_prob": -47.20639419555664, "reading_distances": [43, 44, 45, 39], "reading_log_probs": [-50.48387145996094, -50.05448913574219, -49.048179626464844, -47.90077209472656], "reading_logprob_std": 1.150568066829959, "tokenization_path_logprob_std": 114.77237172840775, "trajectory_balance_rmse": 972.2984516753306} {"canonical_length": 48, "canonical_nll": 0.9145568211873373, "generation_rouge_l_f1": 0.16501650165016502, "generation_token_f1": 0.22442244224422445, "prompt_id": "4148986bb9e9364b", "reading_canonical_log_prob": -43.89872741699219, "reading_distances": [11, 9, 6, 10], "reading_log_probs": [-47.788246154785156, -47.36351013183594, -48.099395751953125, -49.937339782714844], "reading_logprob_std": 1.134312683245814, "tokenization_path_logprob_std": 102.07624102249875, "trajectory_balance_rmse": 1209.9393489766246} {"canonical_length": 19, "canonical_nll": 0.6616859937969007, "generation_rouge_l_f1": 0.1954022988505747, "generation_token_f1": 0.21839080459770116, "prompt_id": "1c5e2235e187365f", "reading_canonical_log_prob": -12.572033882141113, "reading_distances": [23, 24, 19, 15], "reading_log_probs": [-13.573274612426758, -14.981101036071777, -12.517932891845703, -14.418777465820312], "reading_logprob_std": 1.0726606973707926, "tokenization_path_logprob_std": 92.33136707644569, "trajectory_balance_rmse": 446.63921414413505} {"canonical_length": 369, "canonical_nll": 1.1210067459561315, "generation_rouge_l_f1": 0.2695507487520799, "generation_token_f1": 0.5424292845257904, "prompt_id": "6125f7b585f4ca3f", "reading_canonical_log_prob": -413.6514892578125, "reading_distances": [32, 19, 29, 26], "reading_log_probs": [-416.4185791015625, -416.0469665527344, -419.61053466796875, -417.9033203125], "reading_logprob_std": 1.6225181184567237, "tokenization_path_logprob_std": 508.9889747424669, "trajectory_balance_rmse": 10021.631496603692} {"canonical_length": 21, "canonical_nll": 0.9661415645054409, "generation_rouge_l_f1": 0.3684210526315789, "generation_token_f1": 0.7368421052631579, "prompt_id": "bac6bf1b8815878e", "reading_canonical_log_prob": -20.288972854614258, "reading_distances": [19, 20, 17, 16], "reading_log_probs": [-19.34343147277832, -19.98617935180664, -20.521564483642578, -20.962522506713867], "reading_logprob_std": 0.6986239556615373, "tokenization_path_logprob_std": 89.69019536075015, "trajectory_balance_rmse": 538.0426028933949} {"canonical_length": 81, "canonical_nll": 2.3987545437282987, "generation_rouge_l_f1": 0.24324324324324323, "generation_token_f1": 0.29729729729729726, "prompt_id": "3e2b40881e16b7da", "reading_canonical_log_prob": -194.2991180419922, "reading_distances": [22, 14, 19, 18], "reading_log_probs": [-198.89645385742188, -199.76527404785156, -197.0284423828125, -198.50363159179688], "reading_logprob_std": 1.1422491401078498, "tokenization_path_logprob_std": 102.52163920583087, "trajectory_balance_rmse": 1965.3051729814506} {"canonical_length": 32, "canonical_nll": 1.3058414459228516, "generation_rouge_l_f1": 0.2173913043478261, "generation_token_f1": 0.2236024844720497, "prompt_id": "e01d08fb030556a3", "reading_canonical_log_prob": -41.78692626953125, "reading_distances": [32, 27, 24, 30], "reading_log_probs": [-46.48081588745117, -47.124977111816406, -46.24629211425781, -47.007606506347656], "reading_logprob_std": 0.4196145281576577, "tokenization_path_logprob_std": 155.19104708197483, "trajectory_balance_rmse": 984.6330335472775} {"canonical_length": 64, "canonical_nll": 1.5917003154754639, "generation_rouge_l_f1": 0.16901408450704225, "generation_token_f1": 0.16901408450704225, "prompt_id": "68a9acf034dc69e5", "reading_canonical_log_prob": -101.86882019042969, "reading_distances": [28, 30, 23, 21], "reading_log_probs": [-99.25564575195312, -98.96092224121094, -99.04302215576172, -98.56379699707031], "reading_logprob_std": 0.2893715963515114, "tokenization_path_logprob_std": 51.11365366804793, "trajectory_balance_rmse": 2015.1545677256624} {"canonical_length": 196, "canonical_nll": 1.1774122277084662, "generation_rouge_l_f1": 0.18705035971223022, "generation_token_f1": 0.32374100719424465, "prompt_id": "12fd41da356aa8b2", "reading_canonical_log_prob": -230.77279663085938, "reading_distances": [32, 31, 33, 31], "reading_log_probs": [-230.71078491210938, -232.56141662597656, -229.55760192871094, -231.804931640625], "reading_logprob_std": 1.3101364082833526, "tokenization_path_logprob_std": 257.2040892790095, "trajectory_balance_rmse": 5141.87888601028} {"canonical_length": 404, "canonical_nll": 0.9789664391243812, "generation_rouge_l_f1": 0.18960244648318045, "generation_token_f1": 0.3333333333333333, "prompt_id": "7a350261a78f28df", "reading_canonical_log_prob": -395.50244140625, "reading_distances": [20, 20, 20, 25], "reading_log_probs": [-395.554443359375, -398.74139404296875, -398.205322265625, -400.8502502441406], "reading_logprob_std": 2.1786784150644163, "tokenization_path_logprob_std": 257.9024022062341, "trajectory_balance_rmse": 8858.221572043809} {"canonical_length": 316, "canonical_nll": 1.428980187524723, "generation_rouge_l_f1": 0.20309810671256456, "generation_token_f1": 0.42340791738382105, "prompt_id": "732a9ef0a365d23a", "reading_canonical_log_prob": -451.5577392578125, "reading_distances": [20, 24, 26, 24], "reading_log_probs": [-447.2325744628906, -448.00030517578125, -447.6842346191406, -449.32501220703125], "reading_logprob_std": 0.8999353433120492, "tokenization_path_logprob_std": 230.64503794082063, "trajectory_balance_rmse": 10853.361049832876} {"canonical_length": 244, "canonical_nll": 1.6255260530065319, "generation_rouge_l_f1": 0.310838445807771, "generation_token_f1": 0.5725971370143148, "prompt_id": "26e0cf383cca3d0a", "reading_canonical_log_prob": -396.62835693359375, "reading_distances": [27, 33, 27, 30], "reading_log_probs": [-400.4534606933594, -398.14276123046875, -399.9403991699219, -403.49395751953125], "reading_logprob_std": 2.2237686839033777, "tokenization_path_logprob_std": 76.0287181323196, "trajectory_balance_rmse": 7514.386179789575} {"canonical_length": 123, "canonical_nll": 1.8548772548272359, "generation_rouge_l_f1": 0.2532981530343008, "generation_token_f1": 0.42216358839050133, "prompt_id": "92f09a8894ee3f5f", "reading_canonical_log_prob": -228.14990234375, "reading_distances": [14, 20, 14, 18], "reading_log_probs": [-250.45516967773438, -249.915283203125, -245.7395477294922, -243.76708984375], "reading_logprob_std": 3.2453379134665736, "tokenization_path_logprob_std": 177.92799450864146, "trajectory_balance_rmse": 3826.954726242733} {"canonical_length": 114, "canonical_nll": 1.3902143511855811, "generation_rouge_l_f1": 0.287292817679558, "generation_token_f1": 0.4033149171270718, "prompt_id": "eb8e055ee1d3ca60", "reading_canonical_log_prob": -158.48443603515625, "reading_distances": [21, 21, 17, 21], "reading_log_probs": [-160.05126953125, -160.80545043945312, -158.36044311523438, -159.05612182617188], "reading_logprob_std": 1.077813452005807, "tokenization_path_logprob_std": 233.5789532362759, "trajectory_balance_rmse": 3487.751257951601} {"canonical_length": 221, "canonical_nll": 1.6837159584010888, "generation_rouge_l_f1": 0.28095238095238095, "generation_token_f1": 0.4714285714285714, "prompt_id": "5a05098dc8854389", "reading_canonical_log_prob": -372.1012268066406, "reading_distances": [22, 20, 19, 25], "reading_log_probs": [-376.3916931152344, -374.60784912109375, -366.873291015625, -377.48321533203125], "reading_logprob_std": 4.792660296474034, "tokenization_path_logprob_std": 184.84894201941376, "trajectory_balance_rmse": 5903.098428225272} {"canonical_length": 6, "canonical_nll": 1.23324982325236, "generation_rouge_l_f1": 0.3478260869565218, "generation_token_f1": 0.3478260869565218, "prompt_id": "0744be99f0e84455", "reading_canonical_log_prob": -7.39949893951416, "reading_distances": [20, 24, 17, 15], "reading_log_probs": [-8.771235466003418, -7.775871276855469, -8.957867622375488, -9.6129789352417], "reading_logprob_std": 0.7602282683732154, "tokenization_path_logprob_std": 58.575819848009694, "trajectory_balance_rmse": 204.05402180654505} {"canonical_length": 3, "canonical_nll": 3.7670911153157554, "generation_rouge_l_f1": 0.021052631578947368, "generation_token_f1": 0.021052631578947368, "prompt_id": "1e2019bd6514f4f1", "reading_canonical_log_prob": -11.301273345947266, "reading_distances": [45, 55, 50, 54], "reading_log_probs": [-11.423116683959961, -12.620678901672363, -12.429720878601074, -12.213555335998535], "reading_logprob_std": 0.5260816348810013, "tokenization_path_logprob_std": 19.3954779051508, "trajectory_balance_rmse": 42.19860553618197} {"canonical_length": 17, "canonical_nll": 0.8667134677662569, "generation_rouge_l_f1": 0.5384615384615385, "generation_token_f1": 0.5384615384615385, "prompt_id": "982c9409d78c0796", "reading_canonical_log_prob": -14.734128952026367, "reading_distances": [16, 16, 19, 17], "reading_log_probs": [-19.740774154663086, -17.033187866210938, -15.220016479492188, -16.127479553222656], "reading_logprob_std": 1.9526810330040516, "tokenization_path_logprob_std": 68.83298126641972, "trajectory_balance_rmse": 541.9654096968236} {"canonical_length": 185, "canonical_nll": 1.5995882601351352, "generation_rouge_l_f1": 0.2460136674259681, "generation_token_f1": 0.4510250569476082, "prompt_id": "f2e5fb29aaf06273", "reading_canonical_log_prob": -295.923828125, "reading_distances": [35, 39, 36, 37], "reading_log_probs": [-301.1756896972656, -308.6205749511719, -303.81195068359375, -307.70159912109375], "reading_logprob_std": 3.4648284124209097, "tokenization_path_logprob_std": 240.15577829956212, "trajectory_balance_rmse": 5043.104495526706} {"canonical_length": 309, "canonical_nll": 1.0988105847997574, "generation_rouge_l_f1": 0.2195540308747856, "generation_token_f1": 0.4391080617495712, "prompt_id": "3aa4d1f8b5499bdf", "reading_canonical_log_prob": -339.532470703125, "reading_distances": [65, 56, 54, 65], "reading_log_probs": [-355.24652099609375, -351.69610595703125, -348.72503662109375, -359.09027099609375], "reading_logprob_std": 4.4800356466029205, "tokenization_path_logprob_std": 284.9323821510479, "trajectory_balance_rmse": 6499.09706642483} {"canonical_length": 253, "canonical_nll": 1.623820308640069, "generation_rouge_l_f1": 0.2119460500963391, "generation_token_f1": 0.4778420038535645, "prompt_id": "aa727754a09ad816", "reading_canonical_log_prob": -410.8265380859375, "reading_distances": [24, 22, 21, 24], "reading_log_probs": [-418.891357421875, -424.8817138671875, -420.895751953125, -426.94580078125], "reading_logprob_std": 3.6688720292032477, "tokenization_path_logprob_std": 79.51455543572315, "trajectory_balance_rmse": 7188.115096367929} {"canonical_length": 218, "canonical_nll": 1.3619821530963303, "generation_rouge_l_f1": 0.25672877846790887, "generation_token_f1": 0.4761904761904762, "prompt_id": "4d9af47029652710", "reading_canonical_log_prob": -296.912109375, "reading_distances": [22, 24, 22, 19], "reading_log_probs": [-315.4004211425781, -303.36358642578125, -307.9521484375, -314.1275634765625], "reading_logprob_std": 5.605326165922062, "tokenization_path_logprob_std": 306.5969114612681, "trajectory_balance_rmse": 6558.45477608115} {"canonical_length": 286, "canonical_nll": 1.2874416538051792, "generation_rouge_l_f1": 0.16380952380952382, "generation_token_f1": 0.3466666666666667, "prompt_id": "4d68d14019e3f5eb", "reading_canonical_log_prob": -368.20831298828125, "reading_distances": [25, 24, 24, 22], "reading_log_probs": [-361.45068359375, -363.58538818359375, -362.5157775878906, -361.90960693359375], "reading_logprob_std": 0.9229369796348684, "tokenization_path_logprob_std": 483.87012177818696, "trajectory_balance_rmse": 6622.22634088843} {"canonical_length": 104, "canonical_nll": 1.4227218627929688, "generation_rouge_l_f1": 0.2109090909090909, "generation_token_f1": 0.43636363636363634, "prompt_id": "9752f3aa2c81e0d1", "reading_canonical_log_prob": -147.96307373046875, "reading_distances": [23, 32, 17, 28], "reading_log_probs": [-140.1127471923828, -142.64328002929688, -146.14889526367188, -143.86862182617188], "reading_logprob_std": 2.515547022723324, "tokenization_path_logprob_std": 125.20309620313044, "trajectory_balance_rmse": 3073.2497883156543} {"canonical_length": 129, "canonical_nll": 0.4464690925539002, "generation_rouge_l_f1": 0.48395061728395056, "generation_token_f1": 0.5432098765432098, "prompt_id": "46cccd71028358c4", "reading_canonical_log_prob": -57.594512939453125, "reading_distances": [20, 25, 24, 25], "reading_log_probs": [-58.973323822021484, -54.43925857543945, -55.524940490722656, -59.82418441772461], "reading_logprob_std": 2.6113963203796664, "tokenization_path_logprob_std": 120.20781751882508, "trajectory_balance_rmse": 1672.3653235362656} {"canonical_length": 3, "canonical_nll": 0.46290381749471027, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "f385579e0d438649", "reading_canonical_log_prob": -1.3887114524841309, "reading_distances": [15, 11, 14, 12], "reading_log_probs": [-2.153836965560913, -1.5109679698944092, -1.8086601495742798, -1.9172953367233276], "reading_logprob_std": 0.2667557698252359, "tokenization_path_logprob_std": 18.675916499559204, "trajectory_balance_rmse": 40.684198712319436} {"canonical_length": 3, "canonical_nll": 3.909118970235189, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.25, "prompt_id": "f015c3c34e66dddf", "reading_canonical_log_prob": -11.727356910705566, "reading_distances": [26, 19, 21, 22], "reading_log_probs": [-13.355910301208496, -13.811944961547852, -13.754863739013672, -12.403685569763184], "reading_logprob_std": 0.6510247449033919, "tokenization_path_logprob_std": 20.302021271330077, "trajectory_balance_rmse": 39.82615346575179} {"canonical_length": 47, "canonical_nll": 0.9245071411132812, "generation_rouge_l_f1": 0.26315789473684215, "generation_token_f1": 0.3289473684210526, "prompt_id": "f91ba1a01be59af5", "reading_canonical_log_prob": -43.45183563232422, "reading_distances": [25, 28, 29, 28], "reading_log_probs": [-43.2816162109375, -44.18501281738281, -44.28546142578125, -43.79742431640625], "reading_logprob_std": 0.45537380567273245, "tokenization_path_logprob_std": 126.25483609222375, "trajectory_balance_rmse": 971.2299378440649} {"canonical_length": 179, "canonical_nll": 1.5801753038800628, "generation_rouge_l_f1": 0.24036281179138322, "generation_token_f1": 0.4716553287981859, "prompt_id": "0764d55b71d1b589", "reading_canonical_log_prob": -282.85137939453125, "reading_distances": [14, 15, 18, 17], "reading_log_probs": [-287.1983642578125, -278.8380126953125, -280.17779541015625, -280.349609375], "reading_logprob_std": 3.7660630518008826, "tokenization_path_logprob_std": 347.6872689649914, "trajectory_balance_rmse": 5040.733769325434} {"canonical_length": 75, "canonical_nll": 1.7683638509114583, "generation_rouge_l_f1": 0.13994169096209913, "generation_token_f1": 0.23323615160349853, "prompt_id": "a40186552cfde2fc", "reading_canonical_log_prob": -132.62728881835938, "reading_distances": [12, 12, 6, 5], "reading_log_probs": [-142.638671875, -137.05311584472656, -134.3473663330078, -133.2249755859375], "reading_logprob_std": 4.20116413262384, "tokenization_path_logprob_std": 151.55640606278033, "trajectory_balance_rmse": 1999.0630209818328} {"canonical_length": 263, "canonical_nll": 1.1696929351458532, "generation_rouge_l_f1": 0.26565464895635676, "generation_token_f1": 0.6110056925996206, "prompt_id": "a3ec61d54b8539e4", "reading_canonical_log_prob": -307.6292419433594, "reading_distances": [22, 25, 26, 23], "reading_log_probs": [-306.0880432128906, -309.4056396484375, -307.7529296875, -306.07794189453125], "reading_logprob_std": 1.5913569100522416, "tokenization_path_logprob_std": 275.4362767008222, "trajectory_balance_rmse": 8038.666330176756} {"canonical_length": 210, "canonical_nll": 1.6234531947544644, "generation_rouge_l_f1": 0.23275862068965517, "generation_token_f1": 0.47413793103448276, "prompt_id": "ca23abf600fc132e", "reading_canonical_log_prob": -340.9251708984375, "reading_distances": [43, 45, 50, 45], "reading_log_probs": [-341.5489501953125, -341.0155029296875, -341.51885986328125, -341.25872802734375], "reading_logprob_std": 0.24998258987044664, "tokenization_path_logprob_std": 241.50561702835492, "trajectory_balance_rmse": 6629.507239820201} {"canonical_length": 132, "canonical_nll": 1.8530439897017046, "generation_rouge_l_f1": 0.24552429667519182, "generation_token_f1": 0.34782608695652173, "prompt_id": "a83cfaa0169d4436", "reading_canonical_log_prob": -244.601806640625, "reading_distances": [40, 56, 39, 47], "reading_log_probs": [-254.91380310058594, -259.97442626953125, -257.2561340332031, -260.86199951171875], "reading_logprob_std": 2.702724341487899, "tokenization_path_logprob_std": 125.48268035415923, "trajectory_balance_rmse": 1337.2523809671818} {"canonical_length": 115, "canonical_nll": 1.6156232750934103, "generation_rouge_l_f1": 0.2546419098143236, "generation_token_f1": 0.40848806366047746, "prompt_id": "bd9008593d2eaa20", "reading_canonical_log_prob": -185.7966766357422, "reading_distances": [41, 42, 41, 41], "reading_log_probs": [-191.8316650390625, -191.12069702148438, -191.76406860351562, -193.41363525390625], "reading_logprob_std": 0.9749033039708214, "tokenization_path_logprob_std": 117.36894695758672, "trajectory_balance_rmse": 3715.341384800438} {"canonical_length": 7, "canonical_nll": 0.7858751160757882, "generation_rouge_l_f1": 0.12903225806451613, "generation_token_f1": 0.12903225806451613, "prompt_id": "48bde41f90ff68c1", "reading_canonical_log_prob": -5.501125812530518, "reading_distances": [31, 34, 27, 35], "reading_log_probs": [-6.644209861755371, -6.238243103027344, -5.673730373382568, -6.28234338760376], "reading_logprob_std": 0.4008965005427956, "tokenization_path_logprob_std": 53.64897244436506, "trajectory_balance_rmse": 202.5553767635184} {"canonical_length": 130, "canonical_nll": 0.7899605384239784, "generation_rouge_l_f1": 0.34502923976608185, "generation_token_f1": 0.4970760233918129, "prompt_id": "8dbb7f1da28c2195", "reading_canonical_log_prob": -102.69486999511719, "reading_distances": [11, 9, 18, 15], "reading_log_probs": [-103.9903793334961, -101.71113586425781, -107.65623474121094, -103.78829956054688], "reading_logprob_std": 2.4714033654437784, "tokenization_path_logprob_std": 219.65199172790614, "trajectory_balance_rmse": 2432.391115172838} {"canonical_length": 260, "canonical_nll": 2.2563061053936297, "generation_rouge_l_f1": 0.2918287937743191, "generation_token_f1": 0.5291828793774319, "prompt_id": "d859a6a8a580026f", "reading_canonical_log_prob": -586.6395874023438, "reading_distances": [38, 35, 33, 31], "reading_log_probs": [-598.8822021484375, -598.6319580078125, -608.7421875, -601.5729370117188], "reading_logprob_std": 4.7151019591066605, "tokenization_path_logprob_std": 295.3437260356061, "trajectory_balance_rmse": 7923.456143656326} {"canonical_length": 242, "canonical_nll": 1.4131616009168388, "generation_rouge_l_f1": 0.2454728370221328, "generation_token_f1": 0.44668008048289737, "prompt_id": "b4d85be43e32fd7f", "reading_canonical_log_prob": -341.985107421875, "reading_distances": [37, 43, 34, 38], "reading_log_probs": [-339.95166015625, -337.95758056640625, -341.7193298339844, -347.08770751953125], "reading_logprob_std": 3.919541174021393, "tokenization_path_logprob_std": 125.54392268007754, "trajectory_balance_rmse": 7471.839595263438} {"canonical_length": 19, "canonical_nll": 1.4557866548237048, "generation_rouge_l_f1": 0.35294117647058826, "generation_token_f1": 0.47058823529411764, "prompt_id": "64c4e6c526576026", "reading_canonical_log_prob": -27.65994644165039, "reading_distances": [27, 28, 30, 29], "reading_log_probs": [-28.118654251098633, -25.81131362915039, -29.659191131591797, -28.13388442993164], "reading_logprob_std": 1.5870403144628775, "tokenization_path_logprob_std": 34.96994319312636, "trajectory_balance_rmse": 587.723956534817} {"canonical_length": 56, "canonical_nll": 1.762063435145787, "generation_rouge_l_f1": 0.061971830985915494, "generation_token_f1": 0.1295774647887324, "prompt_id": "505a139a155b8b27", "reading_canonical_log_prob": -98.67555236816406, "reading_distances": [26, 29, 18, 32], "reading_log_probs": [-99.28633880615234, -97.70951843261719, -99.91729736328125, -99.39445495605469], "reading_logprob_std": 0.9523139962993523, "tokenization_path_logprob_std": 100.26580607140318, "trajectory_balance_rmse": 1723.615002805816} {"canonical_length": 171, "canonical_nll": 1.3298181009571455, "generation_rouge_l_f1": 0.38929440389294406, "generation_token_f1": 0.583941605839416, "prompt_id": "ca5c2935eb3033fb", "reading_canonical_log_prob": -227.39889526367188, "reading_distances": [18, 24, 21, 20], "reading_log_probs": [-222.73501586914062, -223.9147186279297, -224.53257751464844, -222.24407958984375], "reading_logprob_std": 1.0517426828100802, "tokenization_path_logprob_std": 310.32584021079197, "trajectory_balance_rmse": 4734.1475221072815} {"canonical_length": 78, "canonical_nll": 0.8035146762163211, "generation_rouge_l_f1": 0.43037974683544306, "generation_token_f1": 0.5569620253164557, "prompt_id": "f8aa35c9d5781c5a", "reading_canonical_log_prob": -62.67414474487305, "reading_distances": [22, 28, 23, 24], "reading_log_probs": [-63.23880386352539, -63.887062072753906, -62.43943786621094, -63.12495040893555], "reading_logprob_std": 0.5929126404417041, "tokenization_path_logprob_std": 100.3382657864175, "trajectory_balance_rmse": 2028.129453506291} {"canonical_length": 339, "canonical_nll": 1.3154888321868086, "generation_rouge_l_f1": 0.2487394957983193, "generation_token_f1": 0.5176470588235293, "prompt_id": "de613d5bcfdf1d4f", "reading_canonical_log_prob": -445.9507141113281, "reading_distances": [30, 37, 37, 36], "reading_log_probs": [-447.80615234375, -448.9051818847656, -448.21356201171875, -449.0841369628906], "reading_logprob_std": 0.5968917385595317, "tokenization_path_logprob_std": 446.05112897165753, "trajectory_balance_rmse": 10062.673690374068} {"canonical_length": 19, "canonical_nll": 1.9521147075452303, "generation_rouge_l_f1": 0.29411764705882354, "generation_token_f1": 0.29411764705882354, "prompt_id": "cb6d19eef2166daa", "reading_canonical_log_prob": -37.090179443359375, "reading_distances": [23, 30, 27, 26], "reading_log_probs": [-37.58990478515625, -37.823516845703125, -39.0438346862793, -37.91653823852539], "reading_logprob_std": 0.648318469693081, "tokenization_path_logprob_std": 126.18215372020467, "trajectory_balance_rmse": 653.911648885213} {"canonical_length": 389, "canonical_nll": 1.2308413939488272, "generation_rouge_l_f1": 0.1987179487179487, "generation_token_f1": 0.3814102564102564, "prompt_id": "980fa61309c5ee0c", "reading_canonical_log_prob": -478.79730224609375, "reading_distances": [20, 23, 21, 26], "reading_log_probs": [-478.28076171875, -478.05120849609375, -478.8700866699219, -477.0813903808594], "reading_logprob_std": 0.7443696013503613, "tokenization_path_logprob_std": 438.782317390445, "trajectory_balance_rmse": 9733.01860700048} {"canonical_length": 14, "canonical_nll": 2.232830592564174, "generation_rouge_l_f1": 0.12500000000000003, "generation_token_f1": 0.12500000000000003, "prompt_id": "0f4edb0d2ff1676b", "reading_canonical_log_prob": -31.259628295898438, "reading_distances": [14, 13, 14, 17], "reading_log_probs": [-34.279823303222656, -34.530513763427734, -30.991371154785156, -33.160400390625], "reading_logprob_std": 1.6134151093570617, "tokenization_path_logprob_std": 138.35539516287815, "trajectory_balance_rmse": 418.8713454951279} {"canonical_length": 80, "canonical_nll": 1.2553171157836913, "generation_rouge_l_f1": 0.35789473684210527, "generation_token_f1": 0.4631578947368421, "prompt_id": "7bc8145a69124b53", "reading_canonical_log_prob": -100.42536926269531, "reading_distances": [25, 28, 26, 23], "reading_log_probs": [-104.56924438476562, -104.38938903808594, -104.5416259765625, -106.02579498291016], "reading_logprob_std": 0.7669424417914231, "tokenization_path_logprob_std": 94.46540411076518, "trajectory_balance_rmse": 1779.7793719067888} {"canonical_length": 12, "canonical_nll": 1.133975346883138, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "5167b2d9f21714f8", "reading_canonical_log_prob": -13.607704162597656, "reading_distances": [27, 20, 22, 22], "reading_log_probs": [-12.347274780273438, -11.788385391235352, -13.415422439575195, -12.152026176452637], "reading_logprob_std": 0.6992301173171986, "tokenization_path_logprob_std": 21.374338315029433, "trajectory_balance_rmse": 332.31640014875677} {"canonical_length": 130, "canonical_nll": 1.0142481877253606, "generation_rouge_l_f1": 0.25888324873096447, "generation_token_f1": 0.4771573604060914, "prompt_id": "786583b4f8ed0931", "reading_canonical_log_prob": -131.85226440429688, "reading_distances": [23, 23, 21, 20], "reading_log_probs": [-137.05235290527344, -139.04443359375, -133.3184356689453, -134.8655242919922], "reading_logprob_std": 2.5056030740416193, "tokenization_path_logprob_std": 253.85075659910572, "trajectory_balance_rmse": 4617.539942506072} {"canonical_length": 351, "canonical_nll": 1.1850418134292646, "generation_rouge_l_f1": 0.24742268041237112, "generation_token_f1": 0.4020618556701031, "prompt_id": "99a05025a306c903", "reading_canonical_log_prob": -415.9496765136719, "reading_distances": [18, 15, 13, 19], "reading_log_probs": [-422.9200744628906, -420.47442626953125, -420.6921691894531, -419.81048583984375], "reading_logprob_std": 1.3503052988121886, "tokenization_path_logprob_std": 423.38232452706274, "trajectory_balance_rmse": 8156.6165932092845} {"canonical_length": 32, "canonical_nll": 1.3640458583831787, "generation_rouge_l_f1": 0.1764705882352941, "generation_token_f1": 0.1764705882352941, "prompt_id": "4d4e1a944b4c608a", "reading_canonical_log_prob": -43.64946746826172, "reading_distances": [12, 11, 14, 14], "reading_log_probs": [-42.832340240478516, -42.14048767089844, -43.11235427856445, -39.42743682861328], "reading_logprob_std": 1.6840977127373566, "tokenization_path_logprob_std": 36.502664517009194, "trajectory_balance_rmse": 819.9167093910853} {"canonical_length": 375, "canonical_nll": 1.3622792154947916, "generation_rouge_l_f1": 0.2712369597615499, "generation_token_f1": 0.5126676602086437, "prompt_id": "8de18ae965c9405f", "reading_canonical_log_prob": -510.8547058105469, "reading_distances": [29, 25, 28, 29], "reading_log_probs": [-506.4842224121094, -508.8252258300781, -508.18865966796875, -508.2745361328125], "reading_logprob_std": 1.0126893637435543, "tokenization_path_logprob_std": 567.9351649157472, "trajectory_balance_rmse": 9498.977553380657} {"canonical_length": 154, "canonical_nll": 1.2315772911170861, "generation_rouge_l_f1": 0.4202898550724637, "generation_token_f1": 0.5555555555555555, "prompt_id": "577845af12167b4c", "reading_canonical_log_prob": -189.66290283203125, "reading_distances": [9, 11, 10, 8], "reading_log_probs": [-194.86334228515625, -191.37161254882812, -190.69833374023438, -191.6422576904297], "reading_logprob_std": 1.8559005960211825, "tokenization_path_logprob_std": 98.8962944448817, "trajectory_balance_rmse": 3704.8213505567} {"canonical_length": 94, "canonical_nll": 1.1195671406198056, "generation_rouge_l_f1": 0.1956521739130435, "generation_token_f1": 0.3423913043478261, "prompt_id": "aea07a40db5e5411", "reading_canonical_log_prob": -105.23931121826172, "reading_distances": [18, 24, 30, 29], "reading_log_probs": [-114.67285919189453, -118.08644104003906, -113.68511962890625, -115.34708404541016], "reading_logprob_std": 1.8868090846103087, "tokenization_path_logprob_std": 122.1150273393599, "trajectory_balance_rmse": 2837.3031125300586} {"canonical_length": 111, "canonical_nll": 1.500672348984727, "generation_rouge_l_f1": 0.20320855614973263, "generation_token_f1": 0.34759358288770054, "prompt_id": "fdb4e9b8596975db", "reading_canonical_log_prob": -166.5746307373047, "reading_distances": [24, 23, 23, 20], "reading_log_probs": [-165.3470916748047, -165.441162109375, -165.44644165039062, -165.02813720703125], "reading_logprob_std": 0.19707182862800032, "tokenization_path_logprob_std": 183.82952289646937, "trajectory_balance_rmse": 3585.252822322553} {"canonical_length": 364, "canonical_nll": 1.6771696321256868, "generation_rouge_l_f1": 0.2666666666666667, "generation_token_f1": 0.5466666666666666, "prompt_id": "219a3ab27e90cb77", "reading_canonical_log_prob": -610.48974609375, "reading_distances": [13, 15, 18, 13], "reading_log_probs": [-618.8710327148438, -616.8851928710938, -615.067626953125, -613.469482421875], "reading_logprob_std": 2.329357753448348, "tokenization_path_logprob_std": 517.6331672082479, "trajectory_balance_rmse": 10049.362549884261} {"canonical_length": 42, "canonical_nll": 1.4088572547549294, "generation_rouge_l_f1": 0.16406249999999997, "generation_token_f1": 0.234375, "prompt_id": "c1bdd5a5ad1d792e", "reading_canonical_log_prob": -59.17200469970703, "reading_distances": [14, 21, 19, 18], "reading_log_probs": [-64.66127014160156, -65.02381896972656, -60.514915466308594, -60.98842239379883], "reading_logprob_std": 2.374383331575514, "tokenization_path_logprob_std": 32.44994597498157, "trajectory_balance_rmse": 775.139063158554} {"canonical_length": 262, "canonical_nll": 1.824057775599356, "generation_rouge_l_f1": 0.19805825242718447, "generation_token_f1": 0.4271844660194175, "prompt_id": "52d1f167e2d1bb05", "reading_canonical_log_prob": -477.90313720703125, "reading_distances": [30, 24, 24, 28], "reading_log_probs": [-478.939453125, -477.5557861328125, -475.288818359375, -482.47137451171875], "reading_logprob_std": 3.008424312069828, "tokenization_path_logprob_std": 326.9758346964288, "trajectory_balance_rmse": 7425.629524828317} {"canonical_length": 17, "canonical_nll": 2.4084387386546418, "generation_rouge_l_f1": 0.4615384615384615, "generation_token_f1": 0.4615384615384615, "prompt_id": "c86b63feb26be380", "reading_canonical_log_prob": -40.943458557128906, "reading_distances": [37, 37, 36, 36], "reading_log_probs": [-33.13646697998047, -34.54829788208008, -34.097145080566406, -32.19496536254883], "reading_logprob_std": 1.047321770124857, "tokenization_path_logprob_std": 60.57913518200982, "trajectory_balance_rmse": 542.0554468062359} {"canonical_length": 10, "canonical_nll": 2.764059638977051, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.2105263157894737, "prompt_id": "197710f492b62fe4", "reading_canonical_log_prob": -27.640596389770508, "reading_distances": [14, 7, 9, 16], "reading_log_probs": [-26.07790184020996, -26.08492660522461, -25.58625030517578, -25.498668670654297], "reading_logprob_std": 0.31322630326982565, "tokenization_path_logprob_std": 57.66415382218633, "trajectory_balance_rmse": 283.4684199928364} {"canonical_length": 25, "canonical_nll": 1.1079010009765624, "generation_rouge_l_f1": 0.4225352112676057, "generation_token_f1": 0.4507042253521127, "prompt_id": "23c24264db31cb91", "reading_canonical_log_prob": -27.697525024414062, "reading_distances": [31, 23, 27, 30], "reading_log_probs": [-29.56105613708496, -30.652488708496094, -31.18987274169922, -31.59348487854004], "reading_logprob_std": 0.8809177660580826, "tokenization_path_logprob_std": 101.7983550044223, "trajectory_balance_rmse": 544.1594395105062} {"canonical_length": 169, "canonical_nll": 0.9490379920372596, "generation_rouge_l_f1": 0.3980343980343981, "generation_token_f1": 0.7076167076167077, "prompt_id": "742a95dae13136e2", "reading_canonical_log_prob": -160.38742065429688, "reading_distances": [19, 18, 18, 16], "reading_log_probs": [-162.76116943359375, -165.18478393554688, -159.21493530273438, -161.56155395507812], "reading_logprob_std": 2.485998597995585, "tokenization_path_logprob_std": 199.63752276592416, "trajectory_balance_rmse": 3521.0661591038015} {"canonical_length": 19, "canonical_nll": 2.932228690699527, "generation_rouge_l_f1": 0.42424242424242425, "generation_token_f1": 0.48484848484848486, "prompt_id": "87a7422c0b830ad1", "reading_canonical_log_prob": -55.712345123291016, "reading_distances": [48, 51, 43, 43], "reading_log_probs": [-62.624603271484375, -59.251224517822266, -55.277198791503906, -60.117210388183594], "reading_logprob_std": 3.0498576488149807, "tokenization_path_logprob_std": 77.54018358077549, "trajectory_balance_rmse": 506.047787831422} {"canonical_length": 44, "canonical_nll": 0.9786148938265714, "generation_rouge_l_f1": 0.5060240963855421, "generation_token_f1": 0.5783132530120483, "prompt_id": "73b66ff67ed60622", "reading_canonical_log_prob": -43.05905532836914, "reading_distances": [29, 26, 25, 22], "reading_log_probs": [-50.50556182861328, -48.15530014038086, -50.08739471435547, -46.65491485595703], "reading_logprob_std": 1.7863402827535226, "tokenization_path_logprob_std": 78.41944977770692, "trajectory_balance_rmse": 1619.4979542074832} {"canonical_length": 126, "canonical_nll": 1.256945049951947, "generation_rouge_l_f1": 0.4086021505376344, "generation_token_f1": 0.6021505376344086, "prompt_id": "f610e308eb24edb1", "reading_canonical_log_prob": -158.3750762939453, "reading_distances": [23, 21, 24, 25], "reading_log_probs": [-155.47634887695312, -157.038330078125, -157.37765502929688, -156.92422485351562], "reading_logprob_std": 0.8408753152886436, "tokenization_path_logprob_std": 165.66880361668308, "trajectory_balance_rmse": 2720.658728178308} {"canonical_length": 251, "canonical_nll": 1.2738499128486056, "generation_rouge_l_f1": 0.26156941649899396, "generation_token_f1": 0.4708249496981891, "prompt_id": "da58850d3c7fdcfd", "reading_canonical_log_prob": -319.736328125, "reading_distances": [18, 23, 25, 22], "reading_log_probs": [-314.8025207519531, -318.20831298828125, -315.1402893066406, -313.95208740234375], "reading_logprob_std": 1.8568969310511885, "tokenization_path_logprob_std": 251.82471094959078, "trajectory_balance_rmse": 7505.93029148772} {"canonical_length": 296, "canonical_nll": 1.35428186365076, "generation_rouge_l_f1": 0.19962335216572505, "generation_token_f1": 0.4180790960451977, "prompt_id": "0a2b0e3667ca975a", "reading_canonical_log_prob": -400.867431640625, "reading_distances": [12, 16, 15, 19], "reading_log_probs": [-403.633544921875, -402.87847900390625, -407.7467956542969, -407.56854248046875], "reading_logprob_std": 2.560959106321656, "tokenization_path_logprob_std": 399.851267140962, "trajectory_balance_rmse": 8004.297112232311} {"canonical_length": 111, "canonical_nll": 1.351389841990428, "generation_rouge_l_f1": 0.24277456647398846, "generation_token_f1": 0.38150289017341044, "prompt_id": "b3bdd02a596bc2d5", "reading_canonical_log_prob": -150.0042724609375, "reading_distances": [39, 36, 44, 46], "reading_log_probs": [-153.21676635742188, -148.73944091796875, -152.7650146484375, -150.70001220703125], "reading_logprob_std": 2.0594880608378325, "tokenization_path_logprob_std": 88.03511551384666, "trajectory_balance_rmse": 2100.10366359304} {"canonical_length": 14, "canonical_nll": 0.016479490058762685, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "3a6cc5e39c6939e7", "reading_canonical_log_prob": -0.2307128608226776, "reading_distances": [22, 27, 23, 27], "reading_log_probs": [-1.6822640895843506, -1.6399176120758057, -2.190547227859497, -2.0308072566986084], "reading_logprob_std": 0.26819331625281184, "tokenization_path_logprob_std": 93.46528460784945, "trajectory_balance_rmse": 411.19957367818154} {"canonical_length": 79, "canonical_nll": 0.8385256996637658, "generation_rouge_l_f1": 0.3659305993690852, "generation_token_f1": 0.4794952681388013, "prompt_id": "9431e7948140e38d", "reading_canonical_log_prob": -66.2435302734375, "reading_distances": [36, 34, 37, 37], "reading_log_probs": [-64.07019805908203, -67.08131408691406, -65.32150268554688, -67.45106506347656], "reading_logprob_std": 1.5766930735251796, "tokenization_path_logprob_std": 177.79000173568846, "trajectory_balance_rmse": 1921.720762137274} {"canonical_length": 56, "canonical_nll": 1.0569403512137276, "generation_rouge_l_f1": 0.42580645161290326, "generation_token_f1": 0.4903225806451613, "prompt_id": "8230c7c1426f816d", "reading_canonical_log_prob": -59.18865966796875, "reading_distances": [24, 23, 25, 21], "reading_log_probs": [-58.132240295410156, -56.74845886230469, -58.577369689941406, -56.81157302856445], "reading_logprob_std": 0.9275455660397665, "tokenization_path_logprob_std": 150.3697988173149, "trajectory_balance_rmse": 1803.4145927063735} {"canonical_length": 101, "canonical_nll": 1.2062877994952816, "generation_rouge_l_f1": 0.23577235772357724, "generation_token_f1": 0.44715447154471544, "prompt_id": "211cd27653ee878e", "reading_canonical_log_prob": -121.83506774902344, "reading_distances": [27, 24, 31, 27], "reading_log_probs": [-126.61083984375, -125.14881896972656, -125.59075164794922, -125.55362701416016], "reading_logprob_std": 0.6229189496298111, "tokenization_path_logprob_std": 131.28979104747998, "trajectory_balance_rmse": 2794.0201323299425} {"canonical_length": 19, "canonical_nll": 2.0047364485891244, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4210526315789474, "prompt_id": "bd8cff3238489dd7", "reading_canonical_log_prob": -38.08999252319336, "reading_distances": [31, 29, 37, 36], "reading_log_probs": [-44.63774871826172, -46.43077087402344, -45.011940002441406, -38.4373779296875], "reading_logprob_std": 3.546498551825693, "tokenization_path_logprob_std": 51.13192325215373, "trajectory_balance_rmse": 591.4946515357218} {"canonical_length": 21, "canonical_nll": 1.290829068138486, "generation_rouge_l_f1": 0.3243243243243243, "generation_token_f1": 0.3243243243243243, "prompt_id": "9444b92ac8847c28", "reading_canonical_log_prob": -27.107410430908203, "reading_distances": [28, 21, 24, 23], "reading_log_probs": [-28.787019729614258, -28.442121505737305, -27.69506072998047, -26.983837127685547], "reading_logprob_std": 0.8038101392751623, "tokenization_path_logprob_std": 90.55190950577756, "trajectory_balance_rmse": 737.8488354300326} {"canonical_length": 13, "canonical_nll": 2.323496598463792, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4666666666666666, "prompt_id": "d790284919634e0e", "reading_canonical_log_prob": -30.205455780029297, "reading_distances": [32, 30, 32, 24], "reading_log_probs": [-23.43587303161621, -24.706111907958984, -22.891162872314453, -22.033781051635742], "reading_logprob_std": 1.1197683419283397, "tokenization_path_logprob_std": 24.797048063159775, "trajectory_balance_rmse": 502.6764978767841} {"canonical_length": 86, "canonical_nll": 1.2770602204078851, "generation_rouge_l_f1": 0.1075268817204301, "generation_token_f1": 0.1075268817204301, "prompt_id": "58457960c61ffd91", "reading_canonical_log_prob": -109.82717895507812, "reading_distances": [32, 30, 29, 29], "reading_log_probs": [-111.74832153320312, -109.70280456542969, -108.18609619140625, -110.16746520996094], "reading_logprob_std": 1.4667092633971774, "tokenization_path_logprob_std": 175.0194037022449, "trajectory_balance_rmse": 2611.2501654372686} {"canonical_length": 6, "canonical_nll": 1.8904215494791667, "generation_rouge_l_f1": 0.125, "generation_token_f1": 0.125, "prompt_id": "02cbb58a2f84ec5b", "reading_canonical_log_prob": -11.342529296875, "reading_distances": [19, 16, 17, 20], "reading_log_probs": [-9.931002616882324, -11.852660179138184, -11.670324325561523, -11.305889129638672], "reading_logprob_std": 0.8695467531955382, "tokenization_path_logprob_std": 19.44513139549386, "trajectory_balance_rmse": 206.34078842467437} {"canonical_length": 50, "canonical_nll": 2.80357666015625, "generation_rouge_l_f1": 0.23452768729641696, "generation_token_f1": 0.31921824104234525, "prompt_id": "d0f9c73543784d9e", "reading_canonical_log_prob": -140.1788330078125, "reading_distances": [40, 32, 29, 40], "reading_log_probs": [-142.93905639648438, -147.53921508789062, -141.7637939453125, -149.599609375], "reading_logprob_std": 3.718270041693767, "tokenization_path_logprob_std": 57.280412177379965, "trajectory_balance_rmse": 661.0602122791819} {"canonical_length": 76, "canonical_nll": 1.6626644134521484, "generation_rouge_l_f1": 0.1584158415841584, "generation_token_f1": 0.29702970297029707, "prompt_id": "e67d8b619feb7d26", "reading_canonical_log_prob": -126.36249542236328, "reading_distances": [6, 10, 9, 8], "reading_log_probs": [-127.96365356445312, -129.64187622070312, -132.14141845703125, -129.7691192626953], "reading_logprob_std": 1.7180794223858618, "tokenization_path_logprob_std": 138.60625791699954, "trajectory_balance_rmse": 2555.8080226582456} {"canonical_length": 260, "canonical_nll": 1.3209754356971153, "generation_rouge_l_f1": 0.2836363636363637, "generation_token_f1": 0.48000000000000004, "prompt_id": "aa422d282fda03cd", "reading_canonical_log_prob": -343.45361328125, "reading_distances": [28, 30, 37, 23], "reading_log_probs": [-343.41619873046875, -347.33001708984375, -344.59112548828125, -346.4390869140625], "reading_logprob_std": 1.768863223026727, "tokenization_path_logprob_std": 234.17980483318632, "trajectory_balance_rmse": 4851.584224799797} {"canonical_length": 281, "canonical_nll": 1.4635559299238212, "generation_rouge_l_f1": 0.18786692759295504, "generation_token_f1": 0.3131115459882583, "prompt_id": "050b55490a11894a", "reading_canonical_log_prob": -411.25921630859375, "reading_distances": [19, 17, 26, 18], "reading_log_probs": [-410.89678955078125, -413.90863037109375, -410.46246337890625, -411.84796142578125], "reading_logprob_std": 1.5331498231956897, "tokenization_path_logprob_std": 232.80335670813767, "trajectory_balance_rmse": 7846.097548478457} {"canonical_length": 251, "canonical_nll": 0.9679008575074701, "generation_rouge_l_f1": 0.2653061224489796, "generation_token_f1": 0.46530612244897956, "prompt_id": "2d149cddaf627a2b", "reading_canonical_log_prob": -242.943115234375, "reading_distances": [26, 29, 33, 35], "reading_log_probs": [-237.10107421875, -241.76809692382812, -241.76365661621094, -241.97996520996094], "reading_logprob_std": 2.3702329375977174, "tokenization_path_logprob_std": 208.00359306471148, "trajectory_balance_rmse": 7539.187915647862} {"canonical_length": 157, "canonical_nll": 1.7447128781847134, "generation_rouge_l_f1": 0.23728813559322032, "generation_token_f1": 0.41646489104116224, "prompt_id": "f7af4006bdd897fe", "reading_canonical_log_prob": -273.919921875, "reading_distances": [29, 22, 22, 32], "reading_log_probs": [-267.8726501464844, -266.205322265625, -268.96038818359375, -272.2963562011719], "reading_logprob_std": 2.5715121489859034, "tokenization_path_logprob_std": 169.0905993742715, "trajectory_balance_rmse": 5038.4479924341795} {"canonical_length": 21, "canonical_nll": 0.2649850164140974, "generation_rouge_l_f1": 0.2035928143712575, "generation_token_f1": 0.2035928143712575, "prompt_id": "2d0d3b6d72c67323", "reading_canonical_log_prob": -5.564685344696045, "reading_distances": [16, 31, 24, 23], "reading_log_probs": [-8.074777603149414, -6.479693412780762, -7.89390230178833, -5.848203182220459], "reading_logprob_std": 1.0846771670941038, "tokenization_path_logprob_std": 41.658306168575194, "trajectory_balance_rmse": 512.0522536123958} {"canonical_length": 7, "canonical_nll": 0.3111274923597063, "generation_rouge_l_f1": 0.7692307692307693, "generation_token_f1": 0.7692307692307693, "prompt_id": "974b052b2f839275", "reading_canonical_log_prob": -2.1778924465179443, "reading_distances": [15, 25, 19, 23], "reading_log_probs": [-1.9726566076278687, -1.875311255455017, -1.757969856262207, -2.2623867988586426], "reading_logprob_std": 0.21555027154477074, "tokenization_path_logprob_std": 34.811918010562586, "trajectory_balance_rmse": 121.25672216843857} {"canonical_length": 71, "canonical_nll": 1.567260849643761, "generation_rouge_l_f1": 0.18666666666666668, "generation_token_f1": 0.2866666666666667, "prompt_id": "7f8aa3c30a8d7e14", "reading_canonical_log_prob": -111.27552032470703, "reading_distances": [32, 30, 32, 37], "reading_log_probs": [-121.04087829589844, -116.981689453125, -118.8568344116211, -125.68965148925781], "reading_logprob_std": 3.7515587272446944, "tokenization_path_logprob_std": 108.69307059346731, "trajectory_balance_rmse": 2198.818169360048} {"canonical_length": 12, "canonical_nll": 0.3726062774658203, "generation_rouge_l_f1": 0.6363636363636365, "generation_token_f1": 0.6363636363636365, "prompt_id": "537e5de8a8d8abfa", "reading_canonical_log_prob": -4.471275329589844, "reading_distances": [57, 49, 64, 54], "reading_log_probs": [-7.590386390686035, -7.28987979888916, -5.141229629516602, -10.775209426879883], "reading_logprob_std": 2.322673017372416, "tokenization_path_logprob_std": 42.88275983031805, "trajectory_balance_rmse": 381.38537548655603} {"canonical_length": 235, "canonical_nll": 1.5531417522024602, "generation_rouge_l_f1": 0.2950819672131148, "generation_token_f1": 0.5163934426229508, "prompt_id": "219d66b292bdf703", "reading_canonical_log_prob": -364.9883117675781, "reading_distances": [38, 33, 45, 36], "reading_log_probs": [-369.05450439453125, -371.2886962890625, -369.68310546875, -368.8962097167969], "reading_logprob_std": 1.092890995718927, "tokenization_path_logprob_std": 483.67112265844486, "trajectory_balance_rmse": 7205.363127110297} {"canonical_length": 233, "canonical_nll": 1.7180495364471566, "generation_rouge_l_f1": 0.20469083155650317, "generation_token_f1": 0.33262260127931775, "prompt_id": "b3b45f7b904e5cc3", "reading_canonical_log_prob": -400.3055419921875, "reading_distances": [26, 14, 21, 19], "reading_log_probs": [-389.11859130859375, -391.48968505859375, -384.103515625, -382.29229736328125], "reading_logprob_std": 4.279790781709782, "tokenization_path_logprob_std": 210.08129282310568, "trajectory_balance_rmse": 6972.432239473595} {"canonical_length": 121, "canonical_nll": 1.6102294921875, "generation_rouge_l_f1": 0.272463768115942, "generation_token_f1": 0.42318840579710143, "prompt_id": "dd6de628b051978a", "reading_canonical_log_prob": -194.8377685546875, "reading_distances": [38, 39, 33, 35], "reading_log_probs": [-196.31150817871094, -195.80789184570312, -195.2406005859375, -195.07852172851562], "reading_logprob_std": 0.5627905003028089, "tokenization_path_logprob_std": 142.0822599359511, "trajectory_balance_rmse": 3298.6212726803487} {"canonical_length": 105, "canonical_nll": 0.9945435296921503, "generation_rouge_l_f1": 0.21764705882352944, "generation_token_f1": 0.2823529411764706, "prompt_id": "d2e124ed83b7b0d6", "reading_canonical_log_prob": -104.42707061767578, "reading_distances": [29, 23, 31, 31], "reading_log_probs": [-105.6799545288086, -104.84074401855469, -106.01658630371094, -106.82325744628906], "reading_logprob_std": 0.8209963305079763, "tokenization_path_logprob_std": 150.94038701878435, "trajectory_balance_rmse": 2153.9787913776754} {"canonical_length": 102, "canonical_nll": 1.0105596056171493, "generation_rouge_l_f1": 0.2724458204334365, "generation_token_f1": 0.40866873065015485, "prompt_id": "40e867680ebfa198", "reading_canonical_log_prob": -103.07707977294922, "reading_distances": [36, 38, 40, 29], "reading_log_probs": [-111.08113861083984, -119.54147338867188, -117.56761169433594, -115.9928970336914], "reading_logprob_std": 3.6141529530710788, "tokenization_path_logprob_std": 102.47804363008407, "trajectory_balance_rmse": 1967.0099129850048} {"canonical_length": 217, "canonical_nll": 1.6536080496651786, "generation_rouge_l_f1": 0.20374220374220373, "generation_token_f1": 0.45738045738045735, "prompt_id": "7763d96632870ad4", "reading_canonical_log_prob": -358.83294677734375, "reading_distances": [13, 12, 12, 12], "reading_log_probs": [-349.732666015625, -348.7489929199219, -351.63525390625, -349.014892578125], "reading_logprob_std": 1.3028862430423775, "tokenization_path_logprob_std": 230.75773314553547, "trajectory_balance_rmse": 6667.484038780943} {"canonical_length": 11, "canonical_nll": 0.9779719439419833, "generation_rouge_l_f1": 0.031007751937984496, "generation_token_f1": 0.031007751937984496, "prompt_id": "151a13661f1584a8", "reading_canonical_log_prob": -10.757691383361816, "reading_distances": [15, 12, 13, 15], "reading_log_probs": [-10.806578636169434, -10.804913520812988, -10.111254692077637, -11.100086212158203], "reading_logprob_std": 0.4198910692571196, "tokenization_path_logprob_std": 45.85240191796482, "trajectory_balance_rmse": 273.0806586529095} {"canonical_length": 166, "canonical_nll": 0.9636952963220068, "generation_rouge_l_f1": 0.38255033557046986, "generation_token_f1": 0.5704697986577181, "prompt_id": "f8789af8051fda52", "reading_canonical_log_prob": -159.97341918945312, "reading_distances": [19, 21, 25, 28], "reading_log_probs": [-161.00469970703125, -161.27175903320312, -161.5763397216797, -162.1083984375], "reading_logprob_std": 0.473644138865212, "tokenization_path_logprob_std": 252.36062595161383, "trajectory_balance_rmse": 4740.201555429014} {"canonical_length": 199, "canonical_nll": 0.7539398346714039, "generation_rouge_l_f1": 0.252212389380531, "generation_token_f1": 0.44690265486725667, "prompt_id": "b3a3308a3d93f328", "reading_canonical_log_prob": -150.03402709960938, "reading_distances": [7, 10, 10, 6], "reading_log_probs": [-151.60560607910156, -148.95001220703125, -149.99951171875, -150.3650360107422], "reading_logprob_std": 1.0957524659810463, "tokenization_path_logprob_std": 158.34922555756768, "trajectory_balance_rmse": 5013.245227164507} {"canonical_length": 33, "canonical_nll": 0.42831865946451825, "generation_rouge_l_f1": 0.8461538461538461, "generation_token_f1": 0.9615384615384615, "prompt_id": "06bad79ff1fc809f", "reading_canonical_log_prob": -14.134515762329102, "reading_distances": [21, 25, 19, 26], "reading_log_probs": [-16.117149353027344, -17.07196044921875, -19.943769454956055, -16.212974548339844], "reading_logprob_std": 1.790433897144486, "tokenization_path_logprob_std": 67.13995691337982, "trajectory_balance_rmse": 772.7090800251057} {"canonical_length": 153, "canonical_nll": 1.8517388237847223, "generation_rouge_l_f1": 0.2810457516339869, "generation_token_f1": 0.4509803921568628, "prompt_id": "df98426df5120860", "reading_canonical_log_prob": -283.3160400390625, "reading_distances": [21, 18, 21, 20], "reading_log_probs": [-288.57696533203125, -286.5716552734375, -285.76263427734375, -287.09515380859375], "reading_logprob_std": 1.1846823448032342, "tokenization_path_logprob_std": 209.6992419599406, "trajectory_balance_rmse": 4420.45349417782} {"canonical_length": 25, "canonical_nll": 0.9502638244628906, "generation_rouge_l_f1": 0.1212121212121212, "generation_token_f1": 0.1575757575757576, "prompt_id": "e01eb25b0bbbab91", "reading_canonical_log_prob": -23.756595611572266, "reading_distances": [28, 30, 28, 22], "reading_log_probs": [-24.73423194885254, -25.40186309814453, -23.198657989501953, -23.449298858642578], "reading_logprob_std": 1.0481814604300497, "tokenization_path_logprob_std": 42.12413447032489, "trajectory_balance_rmse": 258.42370204932416} {"canonical_length": 42, "canonical_nll": 2.3169056120372953, "generation_rouge_l_f1": 0.27941176470588236, "generation_token_f1": 0.39705882352941185, "prompt_id": "050fda8d8b928503", "reading_canonical_log_prob": -97.3100357055664, "reading_distances": [29, 30, 28, 33], "reading_log_probs": [-101.67472076416016, -96.63467407226562, -97.32793426513672, -95.82654571533203], "reading_logprob_std": 2.6122407947981134, "tokenization_path_logprob_std": 104.20750818668812, "trajectory_balance_rmse": 1372.6382836184957} {"canonical_length": 191, "canonical_nll": 1.2240809894981184, "generation_rouge_l_f1": 0.35185185185185186, "generation_token_f1": 0.6018518518518517, "prompt_id": "c86e9e8e522d0430", "reading_canonical_log_prob": -233.79946899414062, "reading_distances": [30, 21, 19, 24], "reading_log_probs": [-234.8538055419922, -237.79124450683594, -228.40806579589844, -236.5758514404297], "reading_logprob_std": 4.177075292445963, "tokenization_path_logprob_std": 302.6522845080511, "trajectory_balance_rmse": 4438.170892445512} {"canonical_length": 71, "canonical_nll": 1.297855216012874, "generation_rouge_l_f1": 0.23904382470119523, "generation_token_f1": 0.398406374501992, "prompt_id": "0d01c13272391fbb", "reading_canonical_log_prob": -92.14772033691406, "reading_distances": [27, 26, 27, 32], "reading_log_probs": [-93.04542541503906, -94.51577758789062, -93.10493469238281, -94.27839660644531], "reading_logprob_std": 0.7697150498208774, "tokenization_path_logprob_std": 56.17885629739002, "trajectory_balance_rmse": 1967.1836374522738} {"canonical_length": 96, "canonical_nll": 1.7355200449625652, "generation_rouge_l_f1": 0.1762917933130699, "generation_token_f1": 0.3100303951367781, "prompt_id": "f4314cbcdd2435c8", "reading_canonical_log_prob": -166.60992431640625, "reading_distances": [23, 30, 30, 30], "reading_log_probs": [-166.88914489746094, -162.6883544921875, -166.95156860351562, -166.6385955810547], "reading_logprob_std": 2.073456515098577, "tokenization_path_logprob_std": 111.77997677141633, "trajectory_balance_rmse": 2915.4201524267605} {"canonical_length": 28, "canonical_nll": 1.0323497227260046, "generation_rouge_l_f1": 0.1477272727272727, "generation_token_f1": 0.17045454545454547, "prompt_id": "6039b41734030698", "reading_canonical_log_prob": -28.905792236328125, "reading_distances": [13, 18, 17, 15], "reading_log_probs": [-27.360382080078125, -29.92364501953125, -31.291765213012695, -30.741559982299805], "reading_logprob_std": 1.7392986562406108, "tokenization_path_logprob_std": 63.34917299254993, "trajectory_balance_rmse": 405.4377159695148} {"canonical_length": 180, "canonical_nll": 1.059262424045139, "generation_rouge_l_f1": 0.33604336043360433, "generation_token_f1": 0.5311653116531165, "prompt_id": "bc991d771f2ffcd2", "reading_canonical_log_prob": -190.667236328125, "reading_distances": [28, 34, 27, 26], "reading_log_probs": [-191.4764404296875, -190.90948486328125, -190.5722198486328, -191.31661987304688], "reading_logprob_std": 0.40806801786356145, "tokenization_path_logprob_std": 146.41286358648532, "trajectory_balance_rmse": 4421.172126496352} {"canonical_length": 322, "canonical_nll": 1.0335357855565799, "generation_rouge_l_f1": 0.34210526315789475, "generation_token_f1": 0.5888157894736842, "prompt_id": "ee15b1eaf2aa94fb", "reading_canonical_log_prob": -332.79852294921875, "reading_distances": [31, 29, 29, 32], "reading_log_probs": [-336.0684814453125, -336.2201232910156, -339.22601318359375, -337.1443786621094], "reading_logprob_std": 1.454114308524476, "tokenization_path_logprob_std": 154.7013462173597, "trajectory_balance_rmse": 7985.730321153056} {"canonical_length": 42, "canonical_nll": 1.59163084484282, "generation_rouge_l_f1": 0.1437908496732026, "generation_token_f1": 0.1895424836601307, "prompt_id": "e36603aac9ffa742", "reading_canonical_log_prob": -66.84849548339844, "reading_distances": [24, 27, 27, 29], "reading_log_probs": [-70.4825439453125, -72.33804321289062, -69.03913879394531, -70.61717987060547], "reading_logprob_std": 1.3502708815622972, "tokenization_path_logprob_std": 161.71488121282485, "trajectory_balance_rmse": 1329.2669055368651} {"canonical_length": 248, "canonical_nll": 1.5855856864683089, "generation_rouge_l_f1": 0.20990099009900992, "generation_token_f1": 0.44356435643564357, "prompt_id": "46c0bc7a74b4f8f8", "reading_canonical_log_prob": -393.2252502441406, "reading_distances": [19, 11, 10, 18], "reading_log_probs": [-393.74371337890625, -391.96734619140625, -385.14599609375, -388.894775390625], "reading_logprob_std": 3.770646244476152, "tokenization_path_logprob_std": 203.13423621301536, "trajectory_balance_rmse": 6890.766866612442} {"canonical_length": 193, "canonical_nll": 1.578518249827963, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.47807017543859653, "prompt_id": "17af0bbd6fc62e57", "reading_canonical_log_prob": -304.6540222167969, "reading_distances": [22, 30, 30, 29], "reading_log_probs": [-300.06689453125, -302.04486083984375, -301.60174560546875, -300.3536376953125], "reading_logprob_std": 0.9558896541541373, "tokenization_path_logprob_std": 80.97003587930828, "trajectory_balance_rmse": 4536.966740596175} {"canonical_length": 27, "canonical_nll": 2.1057347898130065, "generation_rouge_l_f1": 0.0879120879120879, "generation_token_f1": 0.10989010989010989, "prompt_id": "917c56964f6f4a40", "reading_canonical_log_prob": -56.85483932495117, "reading_distances": [36, 35, 28, 33], "reading_log_probs": [-61.281700134277344, -62.14257049560547, -59.264930725097656, -58.832679748535156], "reading_logprob_std": 1.5871669149695151, "tokenization_path_logprob_std": 73.1633904840927, "trajectory_balance_rmse": 627.7571476466987} {"canonical_length": 350, "canonical_nll": 1.5133224051339285, "generation_rouge_l_f1": 0.18685121107266434, "generation_token_f1": 0.4117647058823529, "prompt_id": "74f8c2e7c18cb610", "reading_canonical_log_prob": -529.662841796875, "reading_distances": [43, 41, 32, 39], "reading_log_probs": [-540.330322265625, -540.5850219726562, -539.3504638671875, -547.7323608398438], "reading_logprob_std": 3.8587557681026357, "tokenization_path_logprob_std": 220.55969014030487, "trajectory_balance_rmse": 8021.227898379087} {"canonical_length": 26, "canonical_nll": 1.4001312255859375, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "c533f2b5441efb8c", "reading_canonical_log_prob": -36.403411865234375, "reading_distances": [25, 25, 23, 27], "reading_log_probs": [-35.95316696166992, -35.76081848144531, -37.57973861694336, -36.009979248046875], "reading_logprob_std": 0.8426474622006853, "tokenization_path_logprob_std": 56.925401270908054, "trajectory_balance_rmse": 328.2280827213326} {"canonical_length": 91, "canonical_nll": 1.2223036210615557, "generation_rouge_l_f1": 0.21256038647342995, "generation_token_f1": 0.41545893719806765, "prompt_id": "ce28dbb2751f3b46", "reading_canonical_log_prob": -111.22962951660156, "reading_distances": [21, 16, 15, 17], "reading_log_probs": [-111.6229248046875, -112.1698226928711, -112.08378601074219, -111.73072814941406], "reading_logprob_std": 0.26582702626130905, "tokenization_path_logprob_std": 125.98243463522863, "trajectory_balance_rmse": 2789.7391259928013} {"canonical_length": 13, "canonical_nll": 1.002213478088379, "generation_rouge_l_f1": 0.09448818897637797, "generation_token_f1": 0.11023622047244096, "prompt_id": "a24d7fb2090fb702", "reading_canonical_log_prob": -13.028775215148926, "reading_distances": [25, 25, 28, 31], "reading_log_probs": [-13.11198616027832, -12.897298812866211, -12.799163818359375, -13.741325378417969], "reading_logprob_std": 0.42325091572918466, "tokenization_path_logprob_std": 50.77327930763897, "trajectory_balance_rmse": 249.7040511539} {"canonical_length": 133, "canonical_nll": 1.2189803732965225, "generation_rouge_l_f1": 0.49056603773584906, "generation_token_f1": 0.6188679245283019, "prompt_id": "4973bf9719e2751b", "reading_canonical_log_prob": -162.1243896484375, "reading_distances": [18, 18, 23, 24], "reading_log_probs": [-168.91844177246094, -171.3017120361328, -168.1915283203125, -172.42184448242188], "reading_logprob_std": 1.9854832564467726, "tokenization_path_logprob_std": 227.0563697746386, "trajectory_balance_rmse": 3738.7829482610136} {"canonical_length": 15, "canonical_nll": 3.5739710489908854, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "c4461f4a56e11ca5", "reading_canonical_log_prob": -53.60956573486328, "reading_distances": [18, 23, 21, 22], "reading_log_probs": [-66.55831909179688, -62.96710205078125, -56.45787811279297, -55.23651123046875], "reading_logprob_std": 5.375274014891952, "tokenization_path_logprob_std": 11.274453775620378, "trajectory_balance_rmse": 365.78552809817586} {"canonical_length": 51, "canonical_nll": 0.9077625648648131, "generation_rouge_l_f1": 0.5862068965517242, "generation_token_f1": 0.6206896551724138, "prompt_id": "cdeadd1b86cf8871", "reading_canonical_log_prob": -46.29589080810547, "reading_distances": [13, 10, 11, 10], "reading_log_probs": [-47.90821838378906, -47.380428314208984, -46.53881072998047, -47.830177307128906], "reading_logprob_std": 0.6283677294383689, "tokenization_path_logprob_std": 167.37227375931744, "trajectory_balance_rmse": 1061.4937145433344} {"canonical_length": 11, "canonical_nll": 1.0386805100874468, "generation_rouge_l_f1": 0.3, "generation_token_f1": 0.3, "prompt_id": "55ee0fbefdef0ed3", "reading_canonical_log_prob": -11.425485610961914, "reading_distances": [15, 14, 16, 12], "reading_log_probs": [-11.367488861083984, -10.847179412841797, -10.487194061279297, -19.163856506347656], "reading_logprob_std": 4.147390422836708, "tokenization_path_logprob_std": 64.53631619203645, "trajectory_balance_rmse": 312.46407209729074} {"canonical_length": 282, "canonical_nll": 1.6567953123268506, "generation_rouge_l_f1": 0.1895910780669145, "generation_token_f1": 0.42750929368029744, "prompt_id": "0b1cafe0e71072e8", "reading_canonical_log_prob": -467.2162780761719, "reading_distances": [30, 22, 21, 21], "reading_log_probs": [-464.04779052734375, -457.07958984375, -461.1708984375, -456.6903991699219], "reading_logprob_std": 3.5110383735485287, "tokenization_path_logprob_std": 409.2127515404675, "trajectory_balance_rmse": 7741.311667139057} {"canonical_length": 98, "canonical_nll": 1.3510726617307078, "generation_rouge_l_f1": 0.3411764705882353, "generation_token_f1": 0.5529411764705884, "prompt_id": "3a37b4e0a87fe5c3", "reading_canonical_log_prob": -132.40512084960938, "reading_distances": [15, 15, 20, 16], "reading_log_probs": [-140.25009155273438, -138.38754272460938, -138.5015869140625, -141.9261932373047], "reading_logprob_std": 1.6732865729596227, "tokenization_path_logprob_std": 184.9012630646642, "trajectory_balance_rmse": 2906.8459970387617} {"canonical_length": 261, "canonical_nll": 0.9457853470725575, "generation_rouge_l_f1": 0.20233463035019456, "generation_token_f1": 0.46303501945525294, "prompt_id": "7236a1de81b123d6", "reading_canonical_log_prob": -246.8499755859375, "reading_distances": [24, 22, 23, 24], "reading_log_probs": [-245.10903930664062, -245.21315002441406, -245.24607849121094, -245.4002685546875], "reading_logprob_std": 0.12052159507502358, "tokenization_path_logprob_std": 450.911995710412, "trajectory_balance_rmse": 7303.506135338232} {"canonical_length": 181, "canonical_nll": 1.7413517230123445, "generation_rouge_l_f1": 0.2366589327146172, "generation_token_f1": 0.4686774941995359, "prompt_id": "564e5eeb9b5a33fe", "reading_canonical_log_prob": -315.1846618652344, "reading_distances": [42, 46, 36, 36], "reading_log_probs": [-314.27490234375, -309.26397705078125, -310.913330078125, -312.2733459472656], "reading_logprob_std": 2.1221466750535605, "tokenization_path_logprob_std": 211.16685181816675, "trajectory_balance_rmse": 4767.713434262819} {"canonical_length": 141, "canonical_nll": 1.225975956477172, "generation_rouge_l_f1": 0.29508196721311475, "generation_token_f1": 0.48633879781420764, "prompt_id": "8686f6b687c7ab8c", "reading_canonical_log_prob": -172.86260986328125, "reading_distances": [22, 20, 25, 24], "reading_log_probs": [-170.51226806640625, -168.18634033203125, -170.42129516601562, -176.86520385742188], "reading_logprob_std": 3.737419603197668, "tokenization_path_logprob_std": 264.1282517685386, "trajectory_balance_rmse": 3770.3049728036735} {"canonical_length": 101, "canonical_nll": 1.2993541755298577, "generation_rouge_l_f1": 0.31223628691983124, "generation_token_f1": 0.5232067510548523, "prompt_id": "79b5d532fcbe3323", "reading_canonical_log_prob": -131.23477172851562, "reading_distances": [28, 26, 24, 28], "reading_log_probs": [-133.37893676757812, -132.90341186523438, -133.3142852783203, -131.10000610351562], "reading_logprob_std": 1.0703562602537076, "tokenization_path_logprob_std": 137.6478322350297, "trajectory_balance_rmse": 3649.5099107542937} {"canonical_length": 333, "canonical_nll": 1.1775778464010886, "generation_rouge_l_f1": 0.25723472668810293, "generation_token_f1": 0.48553054662379425, "prompt_id": "f5411f1c63da4002", "reading_canonical_log_prob": -392.1334228515625, "reading_distances": [45, 44, 44, 54], "reading_log_probs": [-399.9033203125, -398.24053955078125, -396.824462890625, -396.6275329589844], "reading_logprob_std": 1.5171472803100203, "tokenization_path_logprob_std": 247.12217153195346, "trajectory_balance_rmse": 10338.93353971688} {"canonical_length": 105, "canonical_nll": 1.3445790608723958, "generation_rouge_l_f1": 0.18528610354223432, "generation_token_f1": 0.34332425068119893, "prompt_id": "0709a316fd9c3cd1", "reading_canonical_log_prob": -141.18080139160156, "reading_distances": [30, 26, 31, 25], "reading_log_probs": [-142.525390625, -149.8031005859375, -142.45547485351562, -139.59974670410156], "reading_logprob_std": 4.356800818333731, "tokenization_path_logprob_std": 69.59901403560825, "trajectory_balance_rmse": 2248.2225530376772} {"canonical_length": 68, "canonical_nll": 0.702184284434599, "generation_rouge_l_f1": 0.5748502994011976, "generation_token_f1": 0.6826347305389221, "prompt_id": "28c4e87a8705b13a", "reading_canonical_log_prob": -47.748531341552734, "reading_distances": [29, 29, 25, 27], "reading_log_probs": [-53.80752182006836, -53.85226058959961, -55.17607879638672, -55.94922637939453], "reading_logprob_std": 1.0491808522076256, "tokenization_path_logprob_std": 218.10842008772164, "trajectory_balance_rmse": 1831.743545883179} {"canonical_length": 10, "canonical_nll": 1.1430986404418946, "generation_rouge_l_f1": 0.7000000000000001, "generation_token_f1": 0.7999999999999999, "prompt_id": "d86571d08d8aa4d7", "reading_canonical_log_prob": -11.430986404418945, "reading_distances": [36, 42, 39, 47], "reading_log_probs": [-11.251605033874512, -10.874823570251465, -11.007807731628418, -10.698806762695312], "reading_logprob_std": 0.23294055519454226, "tokenization_path_logprob_std": 41.39664613049451, "trajectory_balance_rmse": 365.9102899398063} {"canonical_length": 30, "canonical_nll": 0.5446627934773763, "generation_rouge_l_f1": 0.291970802919708, "generation_token_f1": 0.32116788321167883, "prompt_id": "f8bb815af5d18371", "reading_canonical_log_prob": -16.33988380432129, "reading_distances": [26, 20, 27, 18], "reading_log_probs": [-17.10895538330078, -17.454442977905273, -15.624621391296387, -15.856348991394043], "reading_logprob_std": 0.9058826708465597, "tokenization_path_logprob_std": 181.7684624999807, "trajectory_balance_rmse": 752.2447830333323} {"canonical_length": 39, "canonical_nll": 1.7812349368364384, "generation_rouge_l_f1": 0.39473684210526316, "generation_token_f1": 0.42105263157894735, "prompt_id": "e8ce7622bc908c26", "reading_canonical_log_prob": -69.4681625366211, "reading_distances": [19, 20, 22, 15], "reading_log_probs": [-79.97084045410156, -78.2378158569336, -77.98799133300781, -67.65497589111328], "reading_logprob_std": 5.608369541606479, "tokenization_path_logprob_std": 107.77792685044501, "trajectory_balance_rmse": 1075.935814683866} {"canonical_length": 35, "canonical_nll": 0.9661170959472656, "generation_rouge_l_f1": 0.10158730158730159, "generation_token_f1": 0.13968253968253969, "prompt_id": "300a5e4b8236fdd5", "reading_canonical_log_prob": -33.8140983581543, "reading_distances": [33, 33, 27, 25], "reading_log_probs": [-43.69884490966797, -41.35809326171875, -39.735740661621094, -31.753517150878906], "reading_logprob_std": 5.183885055523552, "tokenization_path_logprob_std": 153.62195545796436, "trajectory_balance_rmse": 1009.5299246253137} {"canonical_length": 64, "canonical_nll": 1.0313912630081177, "generation_rouge_l_f1": 0.23241590214067276, "generation_token_f1": 0.3241590214067278, "prompt_id": "20d20afcda52df5b", "reading_canonical_log_prob": -66.00904083251953, "reading_distances": [28, 29, 26, 27], "reading_log_probs": [-66.64213562011719, -69.12718200683594, -67.51876068115234, -66.21328735351562], "reading_logprob_std": 1.2880831914864599, "tokenization_path_logprob_std": 237.7889861089945, "trajectory_balance_rmse": 1857.5152750503014} {"canonical_length": 92, "canonical_nll": 2.2155383566151494, "generation_rouge_l_f1": 0.303030303030303, "generation_token_f1": 0.509090909090909, "prompt_id": "26bff204aae37f53", "reading_canonical_log_prob": -203.82952880859375, "reading_distances": [15, 14, 19, 14], "reading_log_probs": [-195.48159790039062, -195.89266967773438, -193.53045654296875, -194.35794067382812], "reading_logprob_std": 1.0746593321409335, "tokenization_path_logprob_std": 200.54619798420927, "trajectory_balance_rmse": 2790.5593051149003} {"canonical_length": 259, "canonical_nll": 1.6090990456835184, "generation_rouge_l_f1": 0.2434456928838951, "generation_token_f1": 0.4681647940074906, "prompt_id": "d7bb721e18ff9d29", "reading_canonical_log_prob": -416.75665283203125, "reading_distances": [21, 22, 26, 25], "reading_log_probs": [-440.97528076171875, -440.2879333496094, -430.48944091796875, -439.657470703125], "reading_logprob_std": 4.938139108571857, "tokenization_path_logprob_std": 215.85214328477178, "trajectory_balance_rmse": 7917.745054212072} {"canonical_length": 110, "canonical_nll": 0.8433538263494318, "generation_rouge_l_f1": 0.4010989010989011, "generation_token_f1": 0.46153846153846156, "prompt_id": "6df1e56bb497e6a2", "reading_canonical_log_prob": -92.7689208984375, "reading_distances": [41, 43, 35, 42], "reading_log_probs": [-101.58406066894531, -100.86077880859375, -98.3013916015625, -102.010986328125], "reading_logprob_std": 1.6612207264231258, "tokenization_path_logprob_std": 191.465796317009, "trajectory_balance_rmse": 2661.918881261583} {"canonical_length": 9, "canonical_nll": 0.4277775552537706, "generation_rouge_l_f1": 0.5833333333333334, "generation_token_f1": 0.5833333333333334, "prompt_id": "196c78145c93b0b5", "reading_canonical_log_prob": -3.8499979972839355, "reading_distances": [36, 41, 38, 35], "reading_log_probs": [-5.070362091064453, -5.745269775390625, -6.646976470947266, -5.953277111053467], "reading_logprob_std": 0.6492504034240999, "tokenization_path_logprob_std": 65.30805451794943, "trajectory_balance_rmse": 265.5911610712737} {"canonical_length": 61, "canonical_nll": 0.551062349413262, "generation_rouge_l_f1": 0.4635761589403974, "generation_token_f1": 0.7019867549668874, "prompt_id": "31d1718162ea0ff4", "reading_canonical_log_prob": -33.614803314208984, "reading_distances": [25, 22, 22, 19], "reading_log_probs": [-35.87263870239258, -35.52953338623047, -36.710693359375, -37.46167755126953], "reading_logprob_std": 0.8678228766304402, "tokenization_path_logprob_std": 103.50810315693116, "trajectory_balance_rmse": 1239.4060720249993} {"canonical_length": 117, "canonical_nll": 0.9967470739641761, "generation_rouge_l_f1": 0.19498607242339835, "generation_token_f1": 0.3231197771587744, "prompt_id": "823548660cad1f9e", "reading_canonical_log_prob": -116.6194076538086, "reading_distances": [23, 23, 25, 25], "reading_log_probs": [-114.20648193359375, -113.4980239868164, -117.10868072509766, -112.0015640258789], "reading_logprob_std": 2.143696358771926, "tokenization_path_logprob_std": 164.398061380837, "trajectory_balance_rmse": 2313.9707621813823} {"canonical_length": 14, "canonical_nll": 1.4356209891183036, "generation_rouge_l_f1": 0.2456140350877193, "generation_token_f1": 0.2456140350877193, "prompt_id": "912aaeeaa0cce8d9", "reading_canonical_log_prob": -20.09869384765625, "reading_distances": [32, 32, 28, 26], "reading_log_probs": [-19.178260803222656, -20.224693298339844, -17.071155548095703, -15.641669273376465], "reading_logprob_std": 2.0622547073776984, "tokenization_path_logprob_std": 61.8062174717195, "trajectory_balance_rmse": 457.2841196265051} {"canonical_length": 368, "canonical_nll": 1.5837929767111074, "generation_rouge_l_f1": 0.21874999999999997, "generation_token_f1": 0.415625, "prompt_id": "4547f9ca1d45d608", "reading_canonical_log_prob": -582.8358154296875, "reading_distances": [25, 26, 27, 25], "reading_log_probs": [-584.4132690429688, -583.8143310546875, -584.27880859375, -581.7919921875], "reading_logprob_std": 1.2157901544418064, "tokenization_path_logprob_std": 304.8677901311815, "trajectory_balance_rmse": 9694.347706229431} {"canonical_length": 228, "canonical_nll": 1.35111945972108, "generation_rouge_l_f1": 0.3292682926829268, "generation_token_f1": 0.5487804878048781, "prompt_id": "99e3f7e86895d3aa", "reading_canonical_log_prob": -308.05523681640625, "reading_distances": [20, 23, 20, 19], "reading_log_probs": [-312.08221435546875, -313.9240417480469, -313.4397888183594, -311.27850341796875], "reading_logprob_std": 1.2174363903700192, "tokenization_path_logprob_std": 58.3207722668965, "trajectory_balance_rmse": 6894.7251665980875} {"canonical_length": 318, "canonical_nll": 0.8806804920892296, "generation_rouge_l_f1": 0.3449612403100775, "generation_token_f1": 0.5658914728682171, "prompt_id": "d3e72bee48c9454b", "reading_canonical_log_prob": -280.056396484375, "reading_distances": [17, 25, 23, 29], "reading_log_probs": [-283.02593994140625, -281.03912353515625, -286.245361328125, -285.81207275390625], "reading_logprob_std": 2.452011839567431, "tokenization_path_logprob_std": 90.42359415582932, "trajectory_balance_rmse": 6727.332442433204} {"canonical_length": 281, "canonical_nll": 1.5024172962772464, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.44067796610169485, "prompt_id": "3126c4e8c381cc6a", "reading_canonical_log_prob": -422.17926025390625, "reading_distances": [25, 25, 24, 26], "reading_log_probs": [-427.25140380859375, -424.594970703125, -425.1885681152344, -425.3345642089844], "reading_logprob_std": 1.151335213756221, "tokenization_path_logprob_std": 354.4727507225938, "trajectory_balance_rmse": 8345.777192030497} {"canonical_length": 23, "canonical_nll": 0.3762157274329144, "generation_rouge_l_f1": 0.23952095808383236, "generation_token_f1": 0.23952095808383236, "prompt_id": "963018756c1975a3", "reading_canonical_log_prob": -8.652961730957031, "reading_distances": [44, 40, 38, 46], "reading_log_probs": [-6.024344444274902, -6.251129150390625, -6.805119037628174, -7.189414024353027], "reading_logprob_std": 0.5286297800026143, "tokenization_path_logprob_std": 100.56678302924031, "trajectory_balance_rmse": 534.7561435586232} {"canonical_length": 50, "canonical_nll": 2.257957000732422, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.419047619047619, "prompt_id": "b9207d2d35a02140", "reading_canonical_log_prob": -112.8978500366211, "reading_distances": [35, 37, 39, 39], "reading_log_probs": [-120.34673309326172, -122.93025207519531, -120.36663055419922, -123.14484405517578], "reading_logprob_std": 1.5502974572626727, "tokenization_path_logprob_std": 39.716366433683696, "trajectory_balance_rmse": 1629.651461271403} {"canonical_length": 59, "canonical_nll": 0.6852596088991327, "generation_rouge_l_f1": 0.3424657534246575, "generation_token_f1": 0.4383561643835616, "prompt_id": "cc824ceca446b0f2", "reading_canonical_log_prob": -40.43031692504883, "reading_distances": [12, 11, 16, 14], "reading_log_probs": [-48.965309143066406, -47.07881546020508, -47.253273010253906, -51.41826629638672], "reading_logprob_std": 2.0148482780243975, "tokenization_path_logprob_std": 129.10453712339464, "trajectory_balance_rmse": 1536.6194387294931} {"canonical_length": 27, "canonical_nll": 1.2607177451804832, "generation_rouge_l_f1": 0.6415094339622641, "generation_token_f1": 0.7924528301886793, "prompt_id": "9cebf3474b92652d", "reading_canonical_log_prob": -34.03937911987305, "reading_distances": [21, 26, 27, 25], "reading_log_probs": [-32.99272155761719, -33.374298095703125, -34.166038513183594, -33.54916000366211], "reading_logprob_std": 0.4890352533228738, "tokenization_path_logprob_std": 98.06275049516279, "trajectory_balance_rmse": 827.8855482179366} {"canonical_length": 5, "canonical_nll": 2.926459312438965, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "c004648702d6143d", "reading_canonical_log_prob": -14.632296562194824, "reading_distances": [22, 19, 24, 21], "reading_log_probs": [-16.61676025390625, -18.23118019104004, -20.769573211669922, -18.063398361206055], "reading_logprob_std": 1.7257834259635283, "tokenization_path_logprob_std": 3.1611522216486305, "trajectory_balance_rmse": 68.06089586509204} {"canonical_length": 120, "canonical_nll": 1.2832024892171223, "generation_rouge_l_f1": 0.2476190476190476, "generation_token_f1": 0.43809523809523804, "prompt_id": "ebe4917bfacc9af5", "reading_canonical_log_prob": -153.9842987060547, "reading_distances": [28, 27, 18, 32], "reading_log_probs": [-158.4168701171875, -162.99835205078125, -158.1737060546875, -162.72735595703125], "reading_logprob_std": 2.6412712997752514, "tokenization_path_logprob_std": 236.37811108839344, "trajectory_balance_rmse": 3751.207640791597} {"canonical_length": 127, "canonical_nll": 1.604205033910556, "generation_rouge_l_f1": 0.2506393861892583, "generation_token_f1": 0.4757033248081841, "prompt_id": "bd51aa5053398c97", "reading_canonical_log_prob": -203.73403930664062, "reading_distances": [31, 21, 23, 18], "reading_log_probs": [-203.67335510253906, -204.08779907226562, -203.29934692382812, -203.97244262695312], "reading_logprob_std": 0.35226895435894107, "tokenization_path_logprob_std": 139.05940138432825, "trajectory_balance_rmse": 3972.2992428226326} {"canonical_length": 168, "canonical_nll": 1.418056851341611, "generation_rouge_l_f1": 0.15420560747663553, "generation_token_f1": 0.31308411214953263, "prompt_id": "0d7273ba3d245c40", "reading_canonical_log_prob": -238.23355102539062, "reading_distances": [28, 22, 20, 15], "reading_log_probs": [-244.7667236328125, -240.98492431640625, -243.20315551757812, -244.34677124023438], "reading_logprob_std": 1.6944402383184463, "tokenization_path_logprob_std": 334.4611597513298, "trajectory_balance_rmse": 5020.699146886599} {"canonical_length": 211, "canonical_nll": 1.839324083373445, "generation_rouge_l_f1": 0.18969072164948453, "generation_token_f1": 0.3463917525773195, "prompt_id": "16d4d8a6fd8e0475", "reading_canonical_log_prob": -388.0973815917969, "reading_distances": [32, 31, 29, 33], "reading_log_probs": [-387.89398193359375, -379.20233154296875, -382.3970947265625, -396.96844482421875], "reading_logprob_std": 7.779631358141412, "tokenization_path_logprob_std": 407.30118784984046, "trajectory_balance_rmse": 6049.272622142141} {"canonical_length": 131, "canonical_nll": 1.8848329500387644, "generation_rouge_l_f1": 0.20540540540540542, "generation_token_f1": 0.37297297297297294, "prompt_id": "5669cff2cb390e91", "reading_canonical_log_prob": -246.91311645507812, "reading_distances": [26, 24, 21, 20], "reading_log_probs": [-256.01416015625, -253.77444458007812, -255.04505920410156, -254.41641235351562], "reading_logprob_std": 0.9543785747110781, "tokenization_path_logprob_std": 168.04273063506653, "trajectory_balance_rmse": 3800.8601859796704} {"canonical_length": 314, "canonical_nll": 1.0796415486912818, "generation_rouge_l_f1": 0.24, "generation_token_f1": 0.46956521739130436, "prompt_id": "ea8af27e7f1b048e", "reading_canonical_log_prob": -339.0074462890625, "reading_distances": [27, 27, 28, 30], "reading_log_probs": [-340.1058654785156, -337.5634460449219, -339.884521484375, -338.6779479980469], "reading_logprob_std": 1.177467743653992, "tokenization_path_logprob_std": 433.4254106381774, "trajectory_balance_rmse": 8786.012580401732} {"canonical_length": 8, "canonical_nll": 1.0972609519958496, "generation_rouge_l_f1": 0.12903225806451613, "generation_token_f1": 0.12903225806451613, "prompt_id": "b75f2d55eed057da", "reading_canonical_log_prob": -8.778087615966797, "reading_distances": [14, 17, 18, 17], "reading_log_probs": [-8.906538963317871, -9.538366317749023, -8.635348320007324, -8.1361722946167], "reading_logprob_std": 0.5843073721759965, "tokenization_path_logprob_std": 50.583252450684306, "trajectory_balance_rmse": 191.17697243947865} {"canonical_length": 22, "canonical_nll": 0.852214899930087, "generation_rouge_l_f1": 0.1094890510948905, "generation_token_f1": 0.1240875912408759, "prompt_id": "55725f89c2c9fced", "reading_canonical_log_prob": -18.748727798461914, "reading_distances": [35, 29, 30, 32], "reading_log_probs": [-20.547773361206055, -19.80510139465332, -19.619338989257812, -19.74418067932129], "reading_logprob_std": 0.41963453183393, "tokenization_path_logprob_std": 32.94847311739017, "trajectory_balance_rmse": 582.027856762107} {"canonical_length": 6, "canonical_nll": 1.7917974789937336, "generation_rouge_l_f1": 0.6666666666666665, "generation_token_f1": 0.6666666666666665, "prompt_id": "e8e15c8f276bd3e5", "reading_canonical_log_prob": -10.750784873962402, "reading_distances": [21, 24, 21, 25], "reading_log_probs": [-11.64720630645752, -12.034913063049316, -10.934405326843262, -11.959108352661133], "reading_logprob_std": 0.501876858631568, "tokenization_path_logprob_std": 21.657528908871377, "trajectory_balance_rmse": 114.75140004005029} {"canonical_length": 37, "canonical_nll": 1.6622295895138302, "generation_rouge_l_f1": 0.15209125475285168, "generation_token_f1": 0.20532319391634984, "prompt_id": "6e9b2ca73b16e2a8", "reading_canonical_log_prob": -61.50249481201172, "reading_distances": [36, 33, 30, 32], "reading_log_probs": [-66.97525024414062, -64.99003601074219, -70.38614654541016, -76.63005065917969], "reading_logprob_std": 5.102162520971563, "tokenization_path_logprob_std": 133.94073533995856, "trajectory_balance_rmse": 1185.9594295683567} {"canonical_length": 266, "canonical_nll": 1.5751683514817316, "generation_rouge_l_f1": 0.25296442687747034, "generation_token_f1": 0.5059288537549407, "prompt_id": "fb1406f3dcf9e9ba", "reading_canonical_log_prob": -418.9947814941406, "reading_distances": [24, 25, 22, 33], "reading_log_probs": [-418.3111267089844, -413.6390380859375, -411.9322814941406, -411.79583740234375], "reading_logprob_std": 3.045435066614642, "tokenization_path_logprob_std": 133.52228782841402, "trajectory_balance_rmse": 6806.198435199039} {"canonical_length": 3, "canonical_nll": 0.28390417496363324, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "f6b05fa9b7a6a33d", "reading_canonical_log_prob": -0.8517125248908997, "reading_distances": [34, 24, 24, 32], "reading_log_probs": [-5.085577011108398, -5.193701267242432, -4.608722686767578, -3.0447731018066406], "reading_logprob_std": 0.9920514100314859, "tokenization_path_logprob_std": 37.14626768223163, "trajectory_balance_rmse": 138.84904822420725} {"canonical_length": 274, "canonical_nll": 1.3735970824304289, "generation_rouge_l_f1": 0.22889305816135086, "generation_token_f1": 0.48405253283302063, "prompt_id": "b4b2d4b733fbf2fd", "reading_canonical_log_prob": -376.3656005859375, "reading_distances": [27, 40, 31, 33], "reading_log_probs": [-380.726806640625, -380.33056640625, -381.22271728515625, -387.36505126953125], "reading_logprob_std": 3.322616974388987, "tokenization_path_logprob_std": 456.36565735195364, "trajectory_balance_rmse": 6957.655700979236} {"canonical_length": 309, "canonical_nll": 0.8473311242162217, "generation_rouge_l_f1": 0.2840466926070038, "generation_token_f1": 0.49416342412451364, "prompt_id": "8c9f3f4649945cbf", "reading_canonical_log_prob": -261.8253173828125, "reading_distances": [52, 46, 43, 42], "reading_log_probs": [-265.7213439941406, -261.56134033203125, -266.3604736328125, -263.4570617675781], "reading_logprob_std": 2.1965186816179956, "tokenization_path_logprob_std": 168.92569671914254, "trajectory_balance_rmse": 3790.486482747116} {"canonical_length": 8, "canonical_nll": 0.2929045557975769, "generation_rouge_l_f1": 0.8750000000000001, "generation_token_f1": 0.8750000000000001, "prompt_id": "eccb7866ba7ed3da", "reading_canonical_log_prob": -2.3432364463806152, "reading_distances": [20, 26, 26, 26], "reading_log_probs": [-3.7184898853302, -1.759159803390503, -5.745469093322754, -3.2017722129821777], "reading_logprob_std": 1.6496666928733197, "tokenization_path_logprob_std": 51.65944986152843, "trajectory_balance_rmse": 261.24876491440205} {"canonical_length": 71, "canonical_nll": 2.3539321254676495, "generation_rouge_l_f1": 0.14836795252225518, "generation_token_f1": 0.2789317507418398, "prompt_id": "f78297865f8b46fc", "reading_canonical_log_prob": -167.12918090820312, "reading_distances": [26, 17, 21, 20], "reading_log_probs": [-163.83688354492188, -165.21463012695312, -165.74920654296875, -169.9681396484375], "reading_logprob_std": 2.6430488639050655, "tokenization_path_logprob_std": 106.52943607651784, "trajectory_balance_rmse": 2503.143337805848} {"canonical_length": 147, "canonical_nll": 1.0867305418261055, "generation_rouge_l_f1": 0.2978723404255319, "generation_token_f1": 0.4340425531914894, "prompt_id": "65091372e7851e26", "reading_canonical_log_prob": -159.7493896484375, "reading_distances": [27, 26, 19, 24], "reading_log_probs": [-170.37767028808594, -163.60122680664062, -164.13412475585938, -165.50672912597656], "reading_logprob_std": 3.087975023445534, "tokenization_path_logprob_std": 210.06814780086978, "trajectory_balance_rmse": 3913.1282647734647} {"canonical_length": 16, "canonical_nll": 1.225841760635376, "generation_rouge_l_f1": 0.24999999999999994, "generation_token_f1": 0.33333333333333337, "prompt_id": "aa225c057d9ef9cd", "reading_canonical_log_prob": -19.613468170166016, "reading_distances": [28, 29, 24, 25], "reading_log_probs": [-21.371721267700195, -23.002147674560547, -22.239185333251953, -22.535131454467773], "reading_logprob_std": 0.6863009074398558, "tokenization_path_logprob_std": 27.764402221505584, "trajectory_balance_rmse": 273.3331776515941} {"canonical_length": 352, "canonical_nll": 1.076249902898615, "generation_rouge_l_f1": 0.2828685258964143, "generation_token_f1": 0.5219123505976095, "prompt_id": "1b2704eec1f82afa", "reading_canonical_log_prob": -378.8399658203125, "reading_distances": [31, 31, 32, 34], "reading_log_probs": [-377.2015380859375, -378.2789306640625, -383.23516845703125, -386.1222229003906], "reading_logprob_std": 4.1988162747078315, "tokenization_path_logprob_std": 542.4462478075317, "trajectory_balance_rmse": 8832.134038731334} {"canonical_length": 284, "canonical_nll": 1.481966475365867, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.48161764705882354, "prompt_id": "aa3d50bf234f7b48", "reading_canonical_log_prob": -420.87847900390625, "reading_distances": [36, 33, 34, 30], "reading_log_probs": [-420.38287353515625, -420.7754821777344, -423.709228515625, -421.45635986328125], "reading_logprob_std": 1.4865236588952275, "tokenization_path_logprob_std": 306.1440492987962, "trajectory_balance_rmse": 9462.131955112589} {"canonical_length": 11, "canonical_nll": 1.261374907060103, "generation_rouge_l_f1": 0.07, "generation_token_f1": 0.07, "prompt_id": "38ebe3c79fd5a7f3", "reading_canonical_log_prob": -13.875123977661133, "reading_distances": [23, 13, 19, 23], "reading_log_probs": [-16.870861053466797, -14.599067687988281, -13.406620979309082, -15.340424537658691], "reading_logprob_std": 1.4495796991320737, "tokenization_path_logprob_std": 44.52999330525013, "trajectory_balance_rmse": 152.8439743156313} {"canonical_length": 266, "canonical_nll": 1.5031529476768093, "generation_rouge_l_f1": 0.20149253731343283, "generation_token_f1": 0.3656716417910448, "prompt_id": "df8d0cc99eb20244", "reading_canonical_log_prob": -399.83868408203125, "reading_distances": [24, 18, 23, 28], "reading_log_probs": [-387.10675048828125, -400.66424560546875, -397.64471435546875, -389.3175048828125], "reading_logprob_std": 6.4996814211035225, "tokenization_path_logprob_std": 411.45854829587836, "trajectory_balance_rmse": 7847.487667562398} {"canonical_length": 131, "canonical_nll": 1.802703857421875, "generation_rouge_l_f1": 0.24671916010498685, "generation_token_f1": 0.4146981627296588, "prompt_id": "0f4d389ccaad372e", "reading_canonical_log_prob": -236.15420532226562, "reading_distances": [32, 29, 29, 26], "reading_log_probs": [-235.76815795898438, -231.6179656982422, -240.68438720703125, -236.43006896972656], "reading_logprob_std": 3.7113238040415406, "tokenization_path_logprob_std": 141.14819085853787, "trajectory_balance_rmse": 3012.66375137929} {"canonical_length": 30, "canonical_nll": 1.6609695434570313, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.35555555555555557, "prompt_id": "936a9a79e3069ca4", "reading_canonical_log_prob": -49.82908630371094, "reading_distances": [27, 27, 24, 26], "reading_log_probs": [-47.20335006713867, -43.4227409362793, -43.59859848022461, -44.96632385253906], "reading_logprob_std": 1.7458467973476075, "tokenization_path_logprob_std": 146.94525862537276, "trajectory_balance_rmse": 959.1713311163023} {"canonical_length": 427, "canonical_nll": 1.6440801330137589, "generation_rouge_l_f1": 0.19384615384615386, "generation_token_f1": 0.3323076923076923, "prompt_id": "82391a0f7d399b89", "reading_canonical_log_prob": -702.022216796875, "reading_distances": [24, 29, 30, 30], "reading_log_probs": [-701.858154296875, -699.267578125, -700.599365234375, -702.6959228515625], "reading_logprob_std": 1.4977830636313956, "tokenization_path_logprob_std": 302.8537133276816, "trajectory_balance_rmse": 9548.20729447782} {"canonical_length": 189, "canonical_nll": 1.0692805537471064, "generation_rouge_l_f1": 0.3469387755102041, "generation_token_f1": 0.489795918367347, "prompt_id": "0bf403e155ca208a", "reading_canonical_log_prob": -202.09402465820312, "reading_distances": [23, 21, 22, 24], "reading_log_probs": [-201.24053955078125, -200.8041534423828, -201.00363159179688, -200.48265075683594], "reading_logprob_std": 0.3208751288696356, "tokenization_path_logprob_std": 110.90348561635047, "trajectory_balance_rmse": 4565.964494835522} {"canonical_length": 175, "canonical_nll": 0.7929244559151786, "generation_rouge_l_f1": 0.2900763358778626, "generation_token_f1": 0.5038167938931298, "prompt_id": "3b3aada8838b1247", "reading_canonical_log_prob": -138.76177978515625, "reading_distances": [23, 18, 21, 16], "reading_log_probs": [-129.0572052001953, -130.34829711914062, -130.50698852539062, -129.57489013671875], "reading_logprob_std": 0.6787817798407159, "tokenization_path_logprob_std": 78.85129212819058, "trajectory_balance_rmse": 3585.991381779247} {"canonical_length": 62, "canonical_nll": 2.10351316390499, "generation_rouge_l_f1": 0.22429906542056074, "generation_token_f1": 0.2990654205607477, "prompt_id": "0221b725811d035d", "reading_canonical_log_prob": -130.41781616210938, "reading_distances": [14, 20, 17, 12], "reading_log_probs": [-124.43058776855469, -127.51539611816406, -124.22693634033203, -130.731201171875], "reading_logprob_std": 3.064801197588448, "tokenization_path_logprob_std": 119.49203422520345, "trajectory_balance_rmse": 1932.4308277634818} {"canonical_length": 109, "canonical_nll": 0.8967450343140768, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.4103896103896104, "prompt_id": "7ac2b23c73ad8de9", "reading_canonical_log_prob": -97.74520874023438, "reading_distances": [39, 38, 40, 33], "reading_log_probs": [-101.16143035888672, -101.98516845703125, -100.08238220214844, -98.6692886352539], "reading_logprob_std": 1.4337057956744588, "tokenization_path_logprob_std": 181.32198586366277, "trajectory_balance_rmse": 2867.1323723093615} {"canonical_length": 98, "canonical_nll": 1.4863633136360013, "generation_rouge_l_f1": 0.2524271844660194, "generation_token_f1": 0.42071197411003236, "prompt_id": "a2a34e9410c7790e", "reading_canonical_log_prob": -145.66360473632812, "reading_distances": [21, 22, 25, 22], "reading_log_probs": [-148.80499267578125, -149.40858459472656, -150.09454345703125, -151.2069091796875], "reading_logprob_std": 1.0303047966647874, "tokenization_path_logprob_std": 146.51493308507148, "trajectory_balance_rmse": 2891.7781585518387} {"canonical_length": 139, "canonical_nll": 1.1581912692502248, "generation_rouge_l_f1": 0.27091633466135456, "generation_token_f1": 0.41434262948207173, "prompt_id": "1c2356a54b0c56f0", "reading_canonical_log_prob": -160.98858642578125, "reading_distances": [31, 27, 25, 29], "reading_log_probs": [-158.28103637695312, -157.84609985351562, -159.00222778320312, -158.67874145507812], "reading_logprob_std": 0.5001685869097062, "tokenization_path_logprob_std": 262.4542409281478, "trajectory_balance_rmse": 4590.168516305769} {"canonical_length": 150, "canonical_nll": 1.6577828979492188, "generation_rouge_l_f1": 0.18686868686868688, "generation_token_f1": 0.4343434343434343, "prompt_id": "55d3c9fe027b6cf4", "reading_canonical_log_prob": -248.6674346923828, "reading_distances": [20, 18, 19, 24], "reading_log_probs": [-252.0584716796875, -253.84542846679688, -257.0670471191406, -253.69732666015625], "reading_logprob_std": 2.0960412089749454, "tokenization_path_logprob_std": 90.78079618733399, "trajectory_balance_rmse": 4126.303826306269} {"canonical_length": 22, "canonical_nll": 2.569326574152166, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.2222222222222222, "prompt_id": "99817bb6bf526c93", "reading_canonical_log_prob": -56.525184631347656, "reading_distances": [17, 20, 20, 23], "reading_log_probs": [-64.95523071289062, -65.88523864746094, -60.65425491333008, -64.9351806640625], "reading_logprob_std": 2.3444230966464494, "tokenization_path_logprob_std": 54.178823128139356, "trajectory_balance_rmse": 574.5100141501852} {"canonical_length": 177, "canonical_nll": 1.6496769964358227, "generation_rouge_l_f1": 0.23234624145785876, "generation_token_f1": 0.5239179954441913, "prompt_id": "a99bf80957c76958", "reading_canonical_log_prob": -291.9928283691406, "reading_distances": [22, 28, 30, 26], "reading_log_probs": [-298.3709716796875, -297.54962158203125, -302.07421875, -296.9402160644531], "reading_logprob_std": 2.30284343565498, "tokenization_path_logprob_std": 343.292965097021, "trajectory_balance_rmse": 5340.639550972378} {"canonical_length": 333, "canonical_nll": 1.7643188843140014, "generation_rouge_l_f1": 0.27814569536423844, "generation_token_f1": 0.5033112582781456, "prompt_id": "6b033b3db0a5c2fe", "reading_canonical_log_prob": -587.5181884765625, "reading_distances": [23, 27, 26, 29], "reading_log_probs": [-594.033935546875, -594.3909912109375, -600.82275390625, -595.9840698242188], "reading_logprob_std": 3.126991387448264, "tokenization_path_logprob_std": 229.9228256330368, "trajectory_balance_rmse": 10300.476973022222} {"canonical_length": 190, "canonical_nll": 1.2300680060135691, "generation_rouge_l_f1": 0.2669584245076586, "generation_token_f1": 0.4638949671772428, "prompt_id": "f5608ead26273538", "reading_canonical_log_prob": -233.71292114257812, "reading_distances": [34, 36, 38, 46], "reading_log_probs": [-242.13442993164062, -244.64633178710938, -245.61215209960938, -247.363037109375], "reading_logprob_std": 2.1817697409371974, "tokenization_path_logprob_std": 161.16984960300718, "trajectory_balance_rmse": 4952.852374014579} {"canonical_length": 305, "canonical_nll": 1.2902443807633197, "generation_rouge_l_f1": 0.20848056537102472, "generation_token_f1": 0.42402826855123676, "prompt_id": "eb21c6c6320f2679", "reading_canonical_log_prob": -393.5245361328125, "reading_distances": [32, 25, 34, 31], "reading_log_probs": [-388.9815368652344, -382.4517822265625, -388.87774658203125, -390.55804443359375], "reading_logprob_std": 3.5935329634648574, "tokenization_path_logprob_std": 415.4382893754868, "trajectory_balance_rmse": 8815.878426246212} {"canonical_length": 222, "canonical_nll": 1.0914068823462133, "generation_rouge_l_f1": 0.33406593406593404, "generation_token_f1": 0.5978021978021978, "prompt_id": "bd9aa33662e77c04", "reading_canonical_log_prob": -242.29232788085938, "reading_distances": [18, 17, 19, 19], "reading_log_probs": [-247.66754150390625, -246.6545867919922, -241.0154266357422, -241.91891479492188], "reading_logprob_std": 3.3337469646572626, "tokenization_path_logprob_std": 161.987760128344, "trajectory_balance_rmse": 2970.18421025266} {"canonical_length": 85, "canonical_nll": 1.0944494808421414, "generation_rouge_l_f1": 0.3081967213114754, "generation_token_f1": 0.3868852459016393, "prompt_id": "4fcfacdaa4123cbd", "reading_canonical_log_prob": -93.02820587158203, "reading_distances": [28, 26, 23, 26], "reading_log_probs": [-94.15066528320312, -88.39376831054688, -92.17230224609375, -85.87712860107422], "reading_logprob_std": 3.716485768855151, "tokenization_path_logprob_std": 231.2765514984685, "trajectory_balance_rmse": 1754.1864532429518} {"canonical_length": 214, "canonical_nll": 1.7781414851964077, "generation_rouge_l_f1": 0.24430641821946167, "generation_token_f1": 0.5134575569358177, "prompt_id": "42e23f667d7b63a7", "reading_canonical_log_prob": -380.52227783203125, "reading_distances": [17, 18, 22, 22], "reading_log_probs": [-382.2546081542969, -375.2679138183594, -389.27435302734375, -377.3776550292969], "reading_logprob_std": 6.2185083833047115, "tokenization_path_logprob_std": 340.56339516650115, "trajectory_balance_rmse": 5891.476660175094} {"canonical_length": 13, "canonical_nll": 1.3026373936579778, "generation_rouge_l_f1": 0.056338028169014086, "generation_token_f1": 0.07042253521126761, "prompt_id": "f4ba682f79ee92c5", "reading_canonical_log_prob": -16.93428611755371, "reading_distances": [12, 12, 19, 18], "reading_log_probs": [-17.3255558013916, -16.776918411254883, -16.645931243896484, -18.090396881103516], "reading_logprob_std": 0.6568065867045481, "tokenization_path_logprob_std": 68.98535054305393, "trajectory_balance_rmse": 384.3186878601919} {"canonical_length": 112, "canonical_nll": 1.141462734767369, "generation_rouge_l_f1": 0.15633423180592992, "generation_token_f1": 0.3018867924528302, "prompt_id": "8c5355410cf1d176", "reading_canonical_log_prob": -127.84382629394531, "reading_distances": [22, 20, 30, 24], "reading_log_probs": [-129.73484802246094, -128.05528259277344, -125.79510498046875, -130.7135467529297], "reading_logprob_std": 2.1537991071587044, "tokenization_path_logprob_std": 110.90606470854813, "trajectory_balance_rmse": 2929.9537672455535} {"canonical_length": 255, "canonical_nll": 1.3903526156556372, "generation_rouge_l_f1": 0.2533081285444234, "generation_token_f1": 0.46124763705103966, "prompt_id": "131fb5156c48d9ba", "reading_canonical_log_prob": -354.5399169921875, "reading_distances": [22, 25, 26, 26], "reading_log_probs": [-364.62969970703125, -359.7781677246094, -366.827392578125, -362.06011962890625], "reading_logprob_std": 3.0631645566216164, "tokenization_path_logprob_std": 255.3627956912632, "trajectory_balance_rmse": 7764.652419944434} {"canonical_length": 17, "canonical_nll": 1.452403236837948, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.2857142857142857, "prompt_id": "20a85de31d07820a", "reading_canonical_log_prob": -24.690855026245117, "reading_distances": [17, 18, 17, 23], "reading_log_probs": [-24.44076156616211, -24.385093688964844, -24.888622283935547, -24.504924774169922], "reading_logprob_std": 0.22783754560961875, "tokenization_path_logprob_std": 94.22679552085424, "trajectory_balance_rmse": 343.46356170182173} {"canonical_length": 236, "canonical_nll": 1.2349897481627383, "generation_rouge_l_f1": 0.22821576763485477, "generation_token_f1": 0.4854771784232365, "prompt_id": "8bdd0bdfad8b9b1b", "reading_canonical_log_prob": -291.45758056640625, "reading_distances": [20, 18, 17, 16], "reading_log_probs": [-293.7846374511719, -292.8379211425781, -296.43798828125, -294.3924865722656], "reading_logprob_std": 1.523900075191685, "tokenization_path_logprob_std": 169.74818703408255, "trajectory_balance_rmse": 7074.789345878389} {"canonical_length": 314, "canonical_nll": 1.39722068446457, "generation_rouge_l_f1": 0.2383419689119171, "generation_token_f1": 0.5043177892918825, "prompt_id": "f543bc422c7c86fd", "reading_canonical_log_prob": -438.727294921875, "reading_distances": [23, 23, 27, 26], "reading_log_probs": [-435.1246032714844, -435.1764221191406, -437.47552490234375, -436.83795166015625], "reading_logprob_std": 1.187368732075707, "tokenization_path_logprob_std": 280.1217889427247, "trajectory_balance_rmse": 10590.937627734847} {"canonical_length": 17, "canonical_nll": 1.1513155768899357, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.7857142857142856, "prompt_id": "6b9813758e34b6e5", "reading_canonical_log_prob": -19.572364807128906, "reading_distances": [28, 34, 32, 32], "reading_log_probs": [-17.836034774780273, -19.882307052612305, -23.108386993408203, -18.303421020507812], "reading_logprob_std": 2.3838265621772377, "tokenization_path_logprob_std": 97.1906810480238, "trajectory_balance_rmse": 633.1100482299347} {"canonical_length": 133, "canonical_nll": 1.3438174599095394, "generation_rouge_l_f1": 0.20151133501259444, "generation_token_f1": 0.35264483627204035, "prompt_id": "89680ea889169bf7", "reading_canonical_log_prob": -178.72772216796875, "reading_distances": [24, 22, 25, 25], "reading_log_probs": [-172.67544555664062, -175.50343322753906, -174.09164428710938, -173.06983947753906], "reading_logprob_std": 1.2622166776811135, "tokenization_path_logprob_std": 200.63768315476318, "trajectory_balance_rmse": 4153.566965538062} {"canonical_length": 15, "canonical_nll": 1.109875996907552, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.25, "prompt_id": "5d244886eb8450ed", "reading_canonical_log_prob": -16.64813995361328, "reading_distances": [33, 28, 32, 31], "reading_log_probs": [-15.299854278564453, -12.637275695800781, -14.477880477905273, -14.715807914733887], "reading_logprob_std": 1.1500292535304752, "tokenization_path_logprob_std": 83.39625910028253, "trajectory_balance_rmse": 592.1932033470309} {"canonical_length": 245, "canonical_nll": 0.9794073610889669, "generation_rouge_l_f1": 0.326923076923077, "generation_token_f1": 0.6, "prompt_id": "8dfd747741dad44f", "reading_canonical_log_prob": -239.95480346679688, "reading_distances": [21, 20, 17, 20], "reading_log_probs": [-237.27020263671875, -235.56536865234375, -237.58163452148438, -235.13694763183594], "reading_logprob_std": 1.2172233519804916, "tokenization_path_logprob_std": 172.77013572546818, "trajectory_balance_rmse": 6572.580226834908} {"canonical_length": 18, "canonical_nll": 0.5139378971523709, "generation_rouge_l_f1": 0.19512195121951217, "generation_token_f1": 0.21138211382113822, "prompt_id": "73b880653e3d865c", "reading_canonical_log_prob": -9.250882148742676, "reading_distances": [18, 22, 27, 17], "reading_log_probs": [-10.73807144165039, -11.357889175415039, -8.92208480834961, -9.946895599365234], "reading_logprob_std": 1.05207032529844, "tokenization_path_logprob_std": 42.79668247831252, "trajectory_balance_rmse": 422.50268874653085} {"canonical_length": 18, "canonical_nll": 2.3399204678005643, "generation_rouge_l_f1": 0.22857142857142856, "generation_token_f1": 0.2857142857142857, "prompt_id": "50d05075f2282d29", "reading_canonical_log_prob": -42.118568420410156, "reading_distances": [22, 24, 29, 30], "reading_log_probs": [-37.92011260986328, -39.65223693847656, -39.9987678527832, -39.85725402832031], "reading_logprob_std": 0.9684919991067952, "tokenization_path_logprob_std": 35.604995449969984, "trajectory_balance_rmse": 523.2428936181831} {"canonical_length": 12, "canonical_nll": 1.4696178436279297, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "4907b79891256514", "reading_canonical_log_prob": -17.635414123535156, "reading_distances": [36, 32, 29, 40], "reading_log_probs": [-14.468664169311523, -13.913551330566406, -13.804490089416504, -14.763277053833008], "reading_logprob_std": 0.4554553423138987, "tokenization_path_logprob_std": 59.96029538759544, "trajectory_balance_rmse": 324.5131207480627} {"canonical_length": 58, "canonical_nll": 2.786145440463362, "generation_rouge_l_f1": 0.21978021978021978, "generation_token_f1": 0.2637362637362637, "prompt_id": "d9bc5b8ac130964c", "reading_canonical_log_prob": -161.596435546875, "reading_distances": [13, 12, 10, 14], "reading_log_probs": [-167.6801300048828, -166.41958618164062, -173.86329650878906, -166.8908233642578], "reading_logprob_std": 3.472388627758801, "tokenization_path_logprob_std": 62.39591267800753, "trajectory_balance_rmse": 1722.7737190653652} {"canonical_length": 14, "canonical_nll": 1.1358306748526437, "generation_rouge_l_f1": 0.8461538461538461, "generation_token_f1": 0.8461538461538461, "prompt_id": "0946ba5ceddc203c", "reading_canonical_log_prob": -15.901629447937012, "reading_distances": [17, 14, 12, 15], "reading_log_probs": [-13.928694725036621, -13.886106491088867, -12.506608963012695, -13.186399459838867], "reading_logprob_std": 0.6726726799418058, "tokenization_path_logprob_std": 34.12829636428918, "trajectory_balance_rmse": 322.55662583462873} {"canonical_length": 17, "canonical_nll": 1.3534095988554113, "generation_rouge_l_f1": 0.09929078014184398, "generation_token_f1": 0.12056737588652483, "prompt_id": "d08a5a0e92cb2564", "reading_canonical_log_prob": -23.007963180541992, "reading_distances": [27, 16, 21, 20], "reading_log_probs": [-29.115638732910156, -30.077688217163086, -29.187814712524414, -32.963775634765625], "reading_logprob_std": 1.8055046044481284, "tokenization_path_logprob_std": 89.41067718643215, "trajectory_balance_rmse": 397.9571247425718} {"canonical_length": 65, "canonical_nll": 0.6311901972844051, "generation_rouge_l_f1": 0.4545454545454546, "generation_token_f1": 0.5844155844155844, "prompt_id": "02aff4ae4d6fb42e", "reading_canonical_log_prob": -41.02736282348633, "reading_distances": [18, 26, 25, 20], "reading_log_probs": [-42.730438232421875, -42.311527252197266, -42.16472625732422, -41.81922912597656], "reading_logprob_std": 0.3773917867786838, "tokenization_path_logprob_std": 83.81523536031634, "trajectory_balance_rmse": 1481.087813121873} {"canonical_length": 281, "canonical_nll": 1.2894725867437722, "generation_rouge_l_f1": 0.20143884892086333, "generation_token_f1": 0.40287769784172667, "prompt_id": "a79f040945146e56", "reading_canonical_log_prob": -362.341796875, "reading_distances": [35, 35, 41, 35], "reading_log_probs": [-363.0069580078125, -360.9144287109375, -365.3960266113281, -364.224365234375], "reading_logprob_std": 1.9144537497261405, "tokenization_path_logprob_std": 377.27859609748924, "trajectory_balance_rmse": 8150.0339949361005} {"canonical_length": 341, "canonical_nll": 1.4719042288592834, "generation_rouge_l_f1": 0.2530329289428076, "generation_token_f1": 0.464471403812825, "prompt_id": "79e68a828bcf94f4", "reading_canonical_log_prob": -501.9193420410156, "reading_distances": [30, 25, 29, 23], "reading_log_probs": [-502.35455322265625, -501.4935607910156, -501.0308837890625, -500.6542053222656], "reading_logprob_std": 0.7328634478048504, "tokenization_path_logprob_std": 204.12426226997098, "trajectory_balance_rmse": 9250.454075915117} {"canonical_length": 50, "canonical_nll": 0.7713029479980469, "generation_rouge_l_f1": 0.625, "generation_token_f1": 0.6666666666666666, "prompt_id": "5debb498a4c42600", "reading_canonical_log_prob": -38.565147399902344, "reading_distances": [33, 32, 30, 30], "reading_log_probs": [-38.50274658203125, -38.55611801147461, -39.02342987060547, -37.669097900390625], "reading_logprob_std": 0.5633478689628821, "tokenization_path_logprob_std": 71.02400763787789, "trajectory_balance_rmse": 1250.7241110490027} {"canonical_length": 34, "canonical_nll": 1.8472907122443705, "generation_rouge_l_f1": 0.4571428571428572, "generation_token_f1": 0.5428571428571428, "prompt_id": "fa77d10032456ab0", "reading_canonical_log_prob": -62.807884216308594, "reading_distances": [27, 28, 32, 27], "reading_log_probs": [-63.607906341552734, -62.17234802246094, -64.6597900390625, -62.94163131713867], "reading_logprob_std": 1.0544525979509638, "tokenization_path_logprob_std": 127.55306015205427, "trajectory_balance_rmse": 1374.9095904330836} {"canonical_length": 8, "canonical_nll": 0.0046138581819832325, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "253388f751e77ae9", "reading_canonical_log_prob": -0.03691086545586586, "reading_distances": [48, 51, 45, 55], "reading_log_probs": [-0.04247984662652016, -0.025185709819197655, -0.02099042572081089, -0.04504208639264107], "reading_logprob_std": 0.01210305034470527, "tokenization_path_logprob_std": 39.8401257529966, "trajectory_balance_rmse": 193.46143755149808} {"canonical_length": 51, "canonical_nll": 1.4856588326248468, "generation_rouge_l_f1": 0.22140221402214022, "generation_token_f1": 0.3025830258302583, "prompt_id": "0182330e9a378205", "reading_canonical_log_prob": -75.76860046386719, "reading_distances": [25, 22, 24, 26], "reading_log_probs": [-76.7530746459961, -75.47671508789062, -73.87421417236328, -77.4413070678711], "reading_logprob_std": 1.5690330826123542, "tokenization_path_logprob_std": 152.35996558834742, "trajectory_balance_rmse": 1135.7234077377952} {"canonical_length": 111, "canonical_nll": 1.9694355457752675, "generation_rouge_l_f1": 0.22627737226277372, "generation_token_f1": 0.48175182481751827, "prompt_id": "8709e23cb34a4482", "reading_canonical_log_prob": -218.6073455810547, "reading_distances": [22, 24, 26, 28], "reading_log_probs": [-219.1006317138672, -216.76087951660156, -214.93441772460938, -215.01058959960938], "reading_logprob_std": 1.9571557481664599, "tokenization_path_logprob_std": 145.75535672039226, "trajectory_balance_rmse": 4023.463278515049} {"canonical_length": 6, "canonical_nll": 1.3792778650919597, "generation_rouge_l_f1": 0.888888888888889, "generation_token_f1": 0.888888888888889, "prompt_id": "0852cc501290063e", "reading_canonical_log_prob": -8.275667190551758, "reading_distances": [25, 34, 28, 28], "reading_log_probs": [-10.74315071105957, -9.152681350708008, -8.133549690246582, -8.982686042785645], "reading_logprob_std": 1.088860505817892, "tokenization_path_logprob_std": 27.08550353930776, "trajectory_balance_rmse": 124.95815552718163} {"canonical_length": 12, "canonical_nll": 1.386412779490153, "generation_rouge_l_f1": 0.5882352941176471, "generation_token_f1": 0.5882352941176471, "prompt_id": "0108b707d4dec974", "reading_canonical_log_prob": -16.636953353881836, "reading_distances": [35, 40, 31, 43], "reading_log_probs": [-19.225269317626953, -19.033246994018555, -18.0828800201416, -19.852210998535156], "reading_logprob_std": 0.7325419946930258, "tokenization_path_logprob_std": 105.07992997618294, "trajectory_balance_rmse": 430.9947664485574} {"canonical_length": 104, "canonical_nll": 1.8022318619948168, "generation_rouge_l_f1": 0.16216216216216217, "generation_token_f1": 0.30810810810810807, "prompt_id": "5ffba76c8fed8b0d", "reading_canonical_log_prob": -187.43211364746094, "reading_distances": [18, 19, 19, 19], "reading_log_probs": [-187.17620849609375, -187.83502197265625, -186.72891235351562, -185.51388549804688], "reading_logprob_std": 0.978300825782852, "tokenization_path_logprob_std": 67.26160905138859, "trajectory_balance_rmse": 3211.247305773662} {"canonical_length": 245, "canonical_nll": 1.2138082698899872, "generation_rouge_l_f1": 0.33061224489795915, "generation_token_f1": 0.5510204081632654, "prompt_id": "66d201ed6635605c", "reading_canonical_log_prob": -297.3830261230469, "reading_distances": [43, 40, 50, 36], "reading_log_probs": [-297.6175842285156, -300.8301696777344, -299.8088684082031, -304.35546875], "reading_logprob_std": 2.8086753097861803, "tokenization_path_logprob_std": 339.52393043470755, "trajectory_balance_rmse": 7079.7616987650135} {"canonical_length": 62, "canonical_nll": 1.0272659793976815, "generation_rouge_l_f1": 0.13013698630136988, "generation_token_f1": 0.15753424657534246, "prompt_id": "d9f443de104a8944", "reading_canonical_log_prob": -63.69049072265625, "reading_distances": [22, 17, 18, 17], "reading_log_probs": [-66.14605712890625, -66.4613265991211, -65.87046813964844, -64.44776916503906], "reading_logprob_std": 0.8891534255448563, "tokenization_path_logprob_std": 35.989793760462966, "trajectory_balance_rmse": 709.0075873900996} {"canonical_length": 19, "canonical_nll": 2.0962038542094983, "generation_rouge_l_f1": 0.08695652173913045, "generation_token_f1": 0.08695652173913045, "prompt_id": "d416f7f8cc539e7b", "reading_canonical_log_prob": -39.82787322998047, "reading_distances": [23, 30, 33, 28], "reading_log_probs": [-34.89556884765625, -39.56525421142578, -39.67356491088867, -39.55779266357422], "reading_logprob_std": 2.3522458910252726, "tokenization_path_logprob_std": 129.23514344021484, "trajectory_balance_rmse": 720.9677189853631} {"canonical_length": 12, "canonical_nll": 1.4251023928324382, "generation_rouge_l_f1": 0.3703703703703704, "generation_token_f1": 0.5185185185185186, "prompt_id": "111f88e5ddbba7a9", "reading_canonical_log_prob": -17.101228713989258, "reading_distances": [24, 22, 18, 19], "reading_log_probs": [-27.444278717041016, -25.42494773864746, -33.461753845214844, -21.50856590270996], "reading_logprob_std": 4.986042119431741, "tokenization_path_logprob_std": 36.740400194068435, "trajectory_balance_rmse": 307.9773684915716} {"canonical_length": 11, "canonical_nll": 1.5106433521617542, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.18181818181818182, "prompt_id": "47566df03369ba07", "reading_canonical_log_prob": -16.617076873779297, "reading_distances": [50, 40, 54, 48], "reading_log_probs": [-15.426626205444336, -26.25701904296875, -11.92772102355957, -20.380250930786133], "reading_logprob_std": 6.227550290258653, "tokenization_path_logprob_std": 40.29066834722799, "trajectory_balance_rmse": 270.80826469864854} {"canonical_length": 26, "canonical_nll": 0.7615588995126578, "generation_rouge_l_f1": 0.20444444444444448, "generation_token_f1": 0.21333333333333335, "prompt_id": "0ad3db1386b28ee8", "reading_canonical_log_prob": -19.8005313873291, "reading_distances": [30, 31, 33, 37], "reading_log_probs": [-21.987844467163086, -22.535648345947266, -21.553293228149414, -21.302141189575195], "reading_logprob_std": 0.5407365102801004, "tokenization_path_logprob_std": 56.596030362520764, "trajectory_balance_rmse": 812.7949935265227} {"canonical_length": 104, "canonical_nll": 1.9527549743652344, "generation_rouge_l_f1": 0.1883656509695291, "generation_token_f1": 0.38227146814404434, "prompt_id": "136b8765ce15fa6a", "reading_canonical_log_prob": -203.08651733398438, "reading_distances": [18, 21, 18, 21], "reading_log_probs": [-197.95407104492188, -197.65911865234375, -205.5750274658203, -197.76123046875], "reading_logprob_std": 3.893698154804009, "tokenization_path_logprob_std": 70.25929551245703, "trajectory_balance_rmse": 2782.015916441818} {"canonical_length": 35, "canonical_nll": 0.7639004298618861, "generation_rouge_l_f1": 0.3516483516483517, "generation_token_f1": 0.5274725274725275, "prompt_id": "92e8091e1c9b04af", "reading_canonical_log_prob": -26.736515045166016, "reading_distances": [28, 31, 24, 38], "reading_log_probs": [-29.509912490844727, -28.3997802734375, -29.220787048339844, -26.84207534790039], "reading_logprob_std": 1.1969443107411821, "tokenization_path_logprob_std": 162.45148582716334, "trajectory_balance_rmse": 949.0908555985623} {"canonical_length": 20, "canonical_nll": 2.4943958282470704, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.25, "prompt_id": "3bfad973aaa202e7", "reading_canonical_log_prob": -49.887916564941406, "reading_distances": [21, 15, 21, 17], "reading_log_probs": [-42.93124771118164, -44.705169677734375, -44.13075256347656, -46.899566650390625], "reading_logprob_std": 1.661947421451489, "tokenization_path_logprob_std": 151.4460064154039, "trajectory_balance_rmse": 670.0787761568223} {"canonical_length": 292, "canonical_nll": 1.0692981040641054, "generation_rouge_l_f1": 0.31401869158878504, "generation_token_f1": 0.508411214953271, "prompt_id": "be5dbfe122219a1b", "reading_canonical_log_prob": -312.23504638671875, "reading_distances": [18, 28, 24, 21], "reading_log_probs": [-318.62762451171875, -319.07342529296875, -318.7686462402344, -320.19537353515625], "reading_logprob_std": 0.7108487513865763, "tokenization_path_logprob_std": 270.43726325592263, "trajectory_balance_rmse": 7780.708621250366} {"canonical_length": 39, "canonical_nll": 1.5683980110364082, "generation_rouge_l_f1": 0.12179487179487179, "generation_token_f1": 0.16666666666666669, "prompt_id": "6acd370678d1b813", "reading_canonical_log_prob": -61.16752243041992, "reading_distances": [16, 16, 14, 17], "reading_log_probs": [-60.5869255065918, -68.12373352050781, -71.3701171875, -72.06822204589844], "reading_logprob_std": 5.255839729314611, "tokenization_path_logprob_std": 52.9575956344679, "trajectory_balance_rmse": 480.2368714402327} {"canonical_length": 197, "canonical_nll": 1.2065871960015466, "generation_rouge_l_f1": 0.3103448275862069, "generation_token_f1": 0.522167487684729, "prompt_id": "2bb939d56ab15ee3", "reading_canonical_log_prob": -237.6976776123047, "reading_distances": [22, 20, 21, 20], "reading_log_probs": [-236.82748413085938, -235.87103271484375, -235.4061279296875, -234.79193115234375], "reading_logprob_std": 0.8581163000482723, "tokenization_path_logprob_std": 510.0651503385081, "trajectory_balance_rmse": 5866.795168839857} {"canonical_length": 14, "canonical_nll": 1.5424033573695592, "generation_rouge_l_f1": 0.4666666666666667, "generation_token_f1": 0.4666666666666667, "prompt_id": "9d34cf3c82e91821", "reading_canonical_log_prob": -21.593647003173828, "reading_distances": [48, 30, 40, 42], "reading_log_probs": [-21.74169921875, -19.5990047454834, -18.275463104248047, -19.828428268432617], "reading_logprob_std": 1.4283625988846582, "tokenization_path_logprob_std": 70.20056558143465, "trajectory_balance_rmse": 284.0363913942351} {"canonical_length": 13, "canonical_nll": 0.09912356046529916, "generation_rouge_l_f1": 0.1904761904761905, "generation_token_f1": 0.1904761904761905, "prompt_id": "b8a06c7365e05ce4", "reading_canonical_log_prob": -1.2886062860488892, "reading_distances": [30, 24, 29, 22], "reading_log_probs": [-3.332195997238159, -3.5314042568206787, -2.1523842811584473, -4.416821479797363], "reading_logprob_std": 0.9319060947014505, "tokenization_path_logprob_std": 58.944598028327526, "trajectory_balance_rmse": 400.020241063129} {"canonical_length": 44, "canonical_nll": 1.6663605083118787, "generation_rouge_l_f1": 0.18965517241379312, "generation_token_f1": 0.22413793103448276, "prompt_id": "3c94fb62a787deaa", "reading_canonical_log_prob": -73.31986236572266, "reading_distances": [26, 19, 21, 24], "reading_log_probs": [-81.54287719726562, -78.14395904541016, -85.01234436035156, -90.44462585449219], "reading_logprob_std": 5.250569645042629, "tokenization_path_logprob_std": 58.13551353329354, "trajectory_balance_rmse": 1062.6052689500548} {"canonical_length": 230, "canonical_nll": 1.1188747240149457, "generation_rouge_l_f1": 0.3045267489711934, "generation_token_f1": 0.5308641975308642, "prompt_id": "84f970de38e4a1fd", "reading_canonical_log_prob": -257.3411865234375, "reading_distances": [25, 25, 21, 29], "reading_log_probs": [-257.09332275390625, -252.6204376220703, -257.0837097167969, -256.273681640625], "reading_logprob_std": 2.1331068988394906, "tokenization_path_logprob_std": 223.3631358371433, "trajectory_balance_rmse": 5220.4078382593525} {"canonical_length": 43, "canonical_nll": 1.7105141573174054, "generation_rouge_l_f1": 0.11464968152866242, "generation_token_f1": 0.18471337579617836, "prompt_id": "b9d9b86bb406a853", "reading_canonical_log_prob": -73.55210876464844, "reading_distances": [16, 18, 17, 12], "reading_log_probs": [-80.37482452392578, -82.57743072509766, -83.22664642333984, -78.0355224609375], "reading_logprob_std": 2.3532756613785155, "tokenization_path_logprob_std": 42.114318510577554, "trajectory_balance_rmse": 1231.1012218602857} {"canonical_length": 217, "canonical_nll": 1.0588331090689804, "generation_rouge_l_f1": 0.30148619957537154, "generation_token_f1": 0.5690021231422504, "prompt_id": "2dfd04cd2fcb5555", "reading_canonical_log_prob": -229.76678466796875, "reading_distances": [42, 36, 40, 40], "reading_log_probs": [-222.4669189453125, -219.09384155273438, -219.21022033691406, -218.82569885253906], "reading_logprob_std": 1.7193875359970503, "tokenization_path_logprob_std": 274.3717975824578, "trajectory_balance_rmse": 6861.257443004472} {"canonical_length": 193, "canonical_nll": 0.9596598728950777, "generation_rouge_l_f1": 0.3665158371040724, "generation_token_f1": 0.5520361990950227, "prompt_id": "daff264becf42549", "reading_canonical_log_prob": -185.21435546875, "reading_distances": [16, 11, 14, 14], "reading_log_probs": [-185.35411071777344, -183.1112060546875, -182.49862670898438, -185.96707153320312], "reading_logprob_std": 1.6862547701490254, "tokenization_path_logprob_std": 290.93499822920404, "trajectory_balance_rmse": 5526.681376621404} {"canonical_length": 254, "canonical_nll": 1.2958568663109007, "generation_rouge_l_f1": 0.22868217054263568, "generation_token_f1": 0.39534883720930236, "prompt_id": "f80fc3edc248f324", "reading_canonical_log_prob": -329.14764404296875, "reading_distances": [30, 28, 30, 27], "reading_log_probs": [-329.83477783203125, -328.76422119140625, -327.2769470214844, -326.84716796875], "reading_logprob_std": 1.3749603091400904, "tokenization_path_logprob_std": 267.692265838723, "trajectory_balance_rmse": 8011.0636683212715} {"canonical_length": 72, "canonical_nll": 2.159581078423394, "generation_rouge_l_f1": 0.1567398119122257, "generation_token_f1": 0.27586206896551724, "prompt_id": "7da5b774aaa5c557", "reading_canonical_log_prob": -155.48983764648438, "reading_distances": [25, 18, 24, 29], "reading_log_probs": [-154.71026611328125, -151.55841064453125, -155.17115783691406, -158.1085205078125], "reading_logprob_std": 2.6813977692655198, "tokenization_path_logprob_std": 66.9892255527187, "trajectory_balance_rmse": 2262.814105830938} {"canonical_length": 138, "canonical_nll": 1.7467967600062273, "generation_rouge_l_f1": 0.23737373737373738, "generation_token_f1": 0.40404040404040403, "prompt_id": "0e6878e78767cf2b", "reading_canonical_log_prob": -241.05795288085938, "reading_distances": [16, 20, 24, 22], "reading_log_probs": [-243.26255798339844, -239.6376953125, -244.64859008789062, -244.12539672851562], "reading_logprob_std": 2.2606685371491184, "tokenization_path_logprob_std": 165.13718696057134, "trajectory_balance_rmse": 3571.4311697431394} {"canonical_length": 152, "canonical_nll": 1.3099967554995888, "generation_rouge_l_f1": 0.22541966426858515, "generation_token_f1": 0.4364508393285372, "prompt_id": "0ac01e4320c0e1b4", "reading_canonical_log_prob": -199.1195068359375, "reading_distances": [29, 28, 26, 29], "reading_log_probs": [-195.97372436523438, -203.8350830078125, -199.51116943359375, -199.72943115234375], "reading_logprob_std": 3.214810190394037, "tokenization_path_logprob_std": 125.36628292758438, "trajectory_balance_rmse": 4246.949076284625} {"canonical_length": 259, "canonical_nll": 1.2521608566225266, "generation_rouge_l_f1": 0.22396856581532418, "generation_token_f1": 0.5304518664047151, "prompt_id": "cc034e8024f676e6", "reading_canonical_log_prob": -324.3096618652344, "reading_distances": [23, 15, 30, 24], "reading_log_probs": [-323.29718017578125, -322.78204345703125, -322.950439453125, -323.27703857421875], "reading_logprob_std": 0.25266001782785774, "tokenization_path_logprob_std": 408.3422057803604, "trajectory_balance_rmse": 7300.305017418723} {"canonical_length": 253, "canonical_nll": 1.387076999830163, "generation_rouge_l_f1": 0.23046875000000003, "generation_token_f1": 0.47265625000000006, "prompt_id": "e0ae35419a54967e", "reading_canonical_log_prob": -350.93048095703125, "reading_distances": [23, 23, 27, 30], "reading_log_probs": [-345.7344970703125, -346.9996337890625, -350.89849853515625, -349.89117431640625], "reading_logprob_std": 2.417338460399436, "tokenization_path_logprob_std": 198.0755212625271, "trajectory_balance_rmse": 6632.103767277323} {"canonical_length": 6, "canonical_nll": 1.0672895908355713, "generation_rouge_l_f1": 0.7272727272727273, "generation_token_f1": 0.7272727272727273, "prompt_id": "9ad445292250739f", "reading_canonical_log_prob": -6.403737545013428, "reading_distances": [27, 19, 18, 27], "reading_log_probs": [-4.468608379364014, -4.882070541381836, -5.311599254608154, -19.03813362121582], "reading_logprob_std": 7.083719673198469, "tokenization_path_logprob_std": 37.775134532903294, "trajectory_balance_rmse": 194.06288422152764} {"canonical_length": 21, "canonical_nll": 1.4564156305222284, "generation_rouge_l_f1": 0.11940298507462686, "generation_token_f1": 0.208955223880597, "prompt_id": "1a95335c2e86bbe5", "reading_canonical_log_prob": -30.584728240966797, "reading_distances": [17, 23, 23, 26], "reading_log_probs": [-33.61387634277344, -34.4788703918457, -33.61784362792969, -30.95771026611328], "reading_logprob_std": 1.5280623321971756, "tokenization_path_logprob_std": 24.42254720946648, "trajectory_balance_rmse": 568.2017343491075} {"canonical_length": 25, "canonical_nll": 0.35330326080322266, "generation_rouge_l_f1": 0.14728682170542634, "generation_token_f1": 0.14728682170542634, "prompt_id": "8fe26f83cb600778", "reading_canonical_log_prob": -8.832581520080566, "reading_distances": [22, 23, 25, 23], "reading_log_probs": [-8.416481018066406, -8.34697151184082, -10.192699432373047, -9.060188293457031], "reading_logprob_std": 0.8549909208654692, "tokenization_path_logprob_std": 59.87329858817305, "trajectory_balance_rmse": 267.9545639507925} {"canonical_length": 17, "canonical_nll": 1.0187575396369486, "generation_rouge_l_f1": 0.061224489795918366, "generation_token_f1": 0.08163265306122451, "prompt_id": "8bd44913279a5599", "reading_canonical_log_prob": -17.318878173828125, "reading_distances": [14, 12, 11, 13], "reading_log_probs": [-16.824697494506836, -20.718547821044922, -18.815814971923828, -22.867218017578125], "reading_logprob_std": 2.5866599440369895, "tokenization_path_logprob_std": 9.257315594525963, "trajectory_balance_rmse": 274.8046469537517} {"canonical_length": 355, "canonical_nll": 1.3813854808538732, "generation_rouge_l_f1": 0.24957841483979762, "generation_token_f1": 0.5193929173693086, "prompt_id": "e131aed6216cf197", "reading_canonical_log_prob": -490.391845703125, "reading_distances": [45, 39, 43, 45], "reading_log_probs": [-493.97454833984375, -490.7816162109375, -491.7174072265625, -493.49566650390625], "reading_logprob_std": 1.497854278431999, "tokenization_path_logprob_std": 125.74355287090061, "trajectory_balance_rmse": 10166.966256958784} {"canonical_length": 46, "canonical_nll": 1.7586306696352751, "generation_rouge_l_f1": 0.09395973154362415, "generation_token_f1": 0.14093959731543626, "prompt_id": "87d0cec84789988b", "reading_canonical_log_prob": -80.89701080322266, "reading_distances": [15, 21, 25, 22], "reading_log_probs": [-81.07556915283203, -81.02670288085938, -79.07324981689453, -81.67933654785156], "reading_logprob_std": 1.1332035463149732, "tokenization_path_logprob_std": 224.47469073336208, "trajectory_balance_rmse": 1461.311537580184} {"canonical_length": 138, "canonical_nll": 1.0818386354308198, "generation_rouge_l_f1": 0.3175965665236052, "generation_token_f1": 0.5236051502145923, "prompt_id": "d60af44d5acef8d0", "reading_canonical_log_prob": -149.29373168945312, "reading_distances": [28, 34, 31, 35], "reading_log_probs": [-142.64488220214844, -146.43746948242188, -142.04888916015625, -140.13624572753906], "reading_logprob_std": 2.640319518239312, "tokenization_path_logprob_std": 182.94474843344108, "trajectory_balance_rmse": 4160.262165512347} {"canonical_length": 35, "canonical_nll": 1.8370380946568081, "generation_rouge_l_f1": 0.2121212121212121, "generation_token_f1": 0.24242424242424243, "prompt_id": "a9f740ec35ab9678", "reading_canonical_log_prob": -64.29633331298828, "reading_distances": [14, 15, 17, 11], "reading_log_probs": [-68.80897521972656, -65.9163589477539, -66.52842712402344, -70.86760711669922], "reading_logprob_std": 2.2642891212006875, "tokenization_path_logprob_std": 153.8783003104535, "trajectory_balance_rmse": 983.95228025559} {"canonical_length": 257, "canonical_nll": 1.340679955389713, "generation_rouge_l_f1": 0.11071428571428572, "generation_token_f1": 0.23214285714285715, "prompt_id": "a25427203b062343", "reading_canonical_log_prob": -344.55474853515625, "reading_distances": [32, 33, 30, 31], "reading_log_probs": [-345.39312744140625, -347.26025390625, -346.00823974609375, -345.7732238769531], "reading_logprob_std": 0.8084464477947646, "tokenization_path_logprob_std": 363.7682182849103, "trajectory_balance_rmse": 7026.929081759083} {"canonical_length": 28, "canonical_nll": 1.230607169015067, "generation_rouge_l_f1": 0.32, "generation_token_f1": 0.37333333333333335, "prompt_id": "b0bf00b191d5ab96", "reading_canonical_log_prob": -34.457000732421875, "reading_distances": [26, 29, 27, 31], "reading_log_probs": [-31.922306060791016, -31.486427307128906, -32.7894401550293, -32.14438247680664], "reading_logprob_std": 0.5429914559769772, "tokenization_path_logprob_std": 109.29511428536848, "trajectory_balance_rmse": 1123.353100752561} {"canonical_length": 31, "canonical_nll": 1.0015925745810232, "generation_rouge_l_f1": 0.14814814814814817, "generation_token_f1": 0.20987654320987656, "prompt_id": "77df38b1cb264cc8", "reading_canonical_log_prob": -31.04936981201172, "reading_distances": [8, 10, 12, 10], "reading_log_probs": [-34.29690170288086, -34.63984680175781, -31.18789291381836, -32.108856201171875], "reading_logprob_std": 1.6768314005445952, "tokenization_path_logprob_std": 88.86738674681374, "trajectory_balance_rmse": 878.682237943508} {"canonical_length": 54, "canonical_nll": 2.076250429506655, "generation_rouge_l_f1": 0.2914572864321608, "generation_token_f1": 0.38190954773869346, "prompt_id": "93f9306a97ca7347", "reading_canonical_log_prob": -112.11752319335938, "reading_distances": [13, 14, 13, 13], "reading_log_probs": [-113.6092758178711, -116.79353332519531, -115.75582885742188, -114.56165313720703], "reading_logprob_std": 1.3885959266366588, "tokenization_path_logprob_std": 167.31712279758466, "trajectory_balance_rmse": 1558.4407417926056} {"canonical_length": 31, "canonical_nll": 2.0600435810704387, "generation_rouge_l_f1": 0.2564102564102564, "generation_token_f1": 0.2564102564102564, "prompt_id": "1cc77327b9384870", "reading_canonical_log_prob": -63.861351013183594, "reading_distances": [26, 23, 26, 19], "reading_log_probs": [-57.85228729248047, -57.04092025756836, -60.67094802856445, -59.922218322753906], "reading_logprob_std": 1.706051865457701, "tokenization_path_logprob_std": 101.89751305633234, "trajectory_balance_rmse": 849.4752055055125} {"canonical_length": 182, "canonical_nll": 1.5572875305846496, "generation_rouge_l_f1": 0.2296650717703349, "generation_token_f1": 0.44019138755980863, "prompt_id": "0d40240cbf3216d7", "reading_canonical_log_prob": -283.42633056640625, "reading_distances": [27, 22, 27, 25], "reading_log_probs": [-286.034423828125, -283.51171875, -283.99627685546875, -296.46893310546875], "reading_logprob_std": 6.076515052465813, "tokenization_path_logprob_std": 236.30130288810287, "trajectory_balance_rmse": 4671.780715924986} {"canonical_length": 136, "canonical_nll": 1.9310163610121782, "generation_rouge_l_f1": 0.21276595744680848, "generation_token_f1": 0.34042553191489355, "prompt_id": "c887cead4589c177", "reading_canonical_log_prob": -262.61822509765625, "reading_distances": [12, 12, 19, 17], "reading_log_probs": [-273.32244873046875, -272.77532958984375, -271.19091796875, -272.21868896484375], "reading_logprob_std": 0.9100169665667852, "tokenization_path_logprob_std": 328.5595844789316, "trajectory_balance_rmse": 4157.684769787655} {"canonical_length": 169, "canonical_nll": 2.1035809940134986, "generation_rouge_l_f1": 0.24438902743142144, "generation_token_f1": 0.35910224438902744, "prompt_id": "5686c4dfdf197ade", "reading_canonical_log_prob": -355.50518798828125, "reading_distances": [13, 11, 11, 10], "reading_log_probs": [-351.9205017089844, -351.5334777832031, -355.2510681152344, -355.38848876953125], "reading_logprob_std": 2.0810627568512294, "tokenization_path_logprob_std": 221.4843152242913, "trajectory_balance_rmse": 4276.850023414904} {"canonical_length": 348, "canonical_nll": 1.1974256668967762, "generation_rouge_l_f1": 0.2532467532467532, "generation_token_f1": 0.5, "prompt_id": "497e17449fccf12b", "reading_canonical_log_prob": -416.7041320800781, "reading_distances": [35, 24, 33, 27], "reading_log_probs": [-418.5816345214844, -424.5047607421875, -420.71563720703125, -421.82977294921875], "reading_logprob_std": 2.4654632803872345, "tokenization_path_logprob_std": 492.9975940120599, "trajectory_balance_rmse": 10125.188646119885} {"canonical_length": 38, "canonical_nll": 1.424698879844264, "generation_rouge_l_f1": 0.34545454545454546, "generation_token_f1": 0.39999999999999997, "prompt_id": "faaabffbce1c5a62", "reading_canonical_log_prob": -54.13855743408203, "reading_distances": [35, 38, 39, 31], "reading_log_probs": [-58.188053131103516, -63.68614959716797, -57.159339904785156, -58.13440704345703], "reading_logprob_std": 2.9673506649178005, "tokenization_path_logprob_std": 99.1886373828814, "trajectory_balance_rmse": 1200.2730895642571} {"canonical_length": 76, "canonical_nll": 0.35121581428929377, "generation_rouge_l_f1": 0.6845637583892618, "generation_token_f1": 0.7516778523489933, "prompt_id": "bf9932d84bc4ffec", "reading_canonical_log_prob": -26.692401885986328, "reading_distances": [21, 18, 17, 16], "reading_log_probs": [-40.91644287109375, -41.244415283203125, -36.332679748535156, -41.20350646972656], "reading_logprob_std": 2.3988300658499013, "tokenization_path_logprob_std": 25.52243005087057, "trajectory_balance_rmse": 478.2567481958361} {"canonical_length": 165, "canonical_nll": 1.9391832756273675, "generation_rouge_l_f1": 0.2412060301507538, "generation_token_f1": 0.4623115577889447, "prompt_id": "d69cbb797502e917", "reading_canonical_log_prob": -319.9652404785156, "reading_distances": [18, 20, 13, 19], "reading_log_probs": [-321.48968505859375, -320.541259765625, -320.65203857421875, -320.90216064453125], "reading_logprob_std": 0.4234215339680527, "tokenization_path_logprob_std": 226.4481026708388, "trajectory_balance_rmse": 5670.836690949617} {"canonical_length": 119, "canonical_nll": 1.4296323631991859, "generation_rouge_l_f1": 0.2159383033419023, "generation_token_f1": 0.3804627249357327, "prompt_id": "98ec80f0ace1a8dc", "reading_canonical_log_prob": -170.12625122070312, "reading_distances": [34, 25, 35, 36], "reading_log_probs": [-171.77835083007812, -167.78829956054688, -170.77691650390625, -169.95166015625], "reading_logprob_std": 1.6968889916917702, "tokenization_path_logprob_std": 238.50663614683234, "trajectory_balance_rmse": 3807.7021663531154} {"canonical_length": 14, "canonical_nll": 1.6120529174804688, "generation_rouge_l_f1": 0.09937888198757765, "generation_token_f1": 0.13664596273291926, "prompt_id": "e2aae622b59df148", "reading_canonical_log_prob": -22.568740844726562, "reading_distances": [33, 23, 26, 32], "reading_log_probs": [-27.323928833007812, -27.151607513427734, -29.245601654052734, -27.009614944458008], "reading_logprob_std": 1.0498382496428744, "tokenization_path_logprob_std": 57.235773373237734, "trajectory_balance_rmse": 308.34257750001973} {"canonical_length": 10, "canonical_nll": 0.6702260494232177, "generation_rouge_l_f1": 0.7272727272727273, "generation_token_f1": 0.7272727272727273, "prompt_id": "aefa9cb1b8d05c05", "reading_canonical_log_prob": -6.702260494232178, "reading_distances": [33, 28, 33, 26], "reading_log_probs": [-7.745118141174316, -10.343437194824219, -6.498170852661133, -12.81273078918457], "reading_logprob_std": 2.809863411336813, "tokenization_path_logprob_std": 51.47067595294978, "trajectory_balance_rmse": 398.5204779619797} {"canonical_length": 121, "canonical_nll": 1.2047882080078125, "generation_rouge_l_f1": 0.24289405684754528, "generation_token_f1": 0.4444444444444444, "prompt_id": "7467b21b2ac04b77", "reading_canonical_log_prob": -145.7793731689453, "reading_distances": [11, 5, 12, 9], "reading_log_probs": [-144.28167724609375, -144.00057983398438, -141.65643310546875, -143.23379516601562], "reading_logprob_std": 1.1775692863063472, "tokenization_path_logprob_std": 106.12670739903866, "trajectory_balance_rmse": 3555.280822658444} {"canonical_length": 92, "canonical_nll": 1.424108422320822, "generation_rouge_l_f1": 0.38181818181818183, "generation_token_f1": 0.509090909090909, "prompt_id": "80fb331e54c070fd", "reading_canonical_log_prob": -131.01797485351562, "reading_distances": [18, 21, 19, 23], "reading_log_probs": [-133.0511474609375, -133.82090759277344, -131.643310546875, -132.92398071289062], "reading_logprob_std": 0.9026456306904317, "tokenization_path_logprob_std": 105.14282951312882, "trajectory_balance_rmse": 2222.8424251181923} {"canonical_length": 288, "canonical_nll": 1.3666884104410808, "generation_rouge_l_f1": 0.2422018348623853, "generation_token_f1": 0.4770642201834862, "prompt_id": "1baa92d9506c1060", "reading_canonical_log_prob": -393.60626220703125, "reading_distances": [17, 27, 17, 16], "reading_log_probs": [-403.01904296875, -400.37261962890625, -401.7355041503906, -398.981201171875], "reading_logprob_std": 1.7400881013956548, "tokenization_path_logprob_std": 215.14385086416405, "trajectory_balance_rmse": 8627.295899972458} {"canonical_length": 232, "canonical_nll": 1.6137362513048896, "generation_rouge_l_f1": 0.2663934426229508, "generation_token_f1": 0.4713114754098361, "prompt_id": "23fd1d0b807b2a3d", "reading_canonical_log_prob": -374.3868103027344, "reading_distances": [27, 26, 23, 21], "reading_log_probs": [-386.163818359375, -379.35565185546875, -382.203369140625, -380.0229187011719], "reading_logprob_std": 3.069419606779532, "tokenization_path_logprob_std": 94.63399326945463, "trajectory_balance_rmse": 6696.0655423754715} {"canonical_length": 23, "canonical_nll": 1.4880105723505435, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "e7e3fd3a1ae85dcc", "reading_canonical_log_prob": -34.2242431640625, "reading_distances": [20, 26, 14, 28], "reading_log_probs": [-33.16746520996094, -33.13331985473633, -32.81943893432617, -31.11404800415039], "reading_logprob_std": 0.9756685177536184, "tokenization_path_logprob_std": 81.40434291141666, "trajectory_balance_rmse": 706.2870967639119} {"canonical_length": 11, "canonical_nll": 0.5947885513305664, "generation_rouge_l_f1": 0.4000000000000001, "generation_token_f1": 0.4000000000000001, "prompt_id": "7a61f927f0ede558", "reading_canonical_log_prob": -6.5426740646362305, "reading_distances": [24, 18, 17, 20], "reading_log_probs": [-9.908349990844727, -11.390374183654785, -11.633511543273926, -6.389918327331543], "reading_logprob_std": 2.4171416284153917, "tokenization_path_logprob_std": 41.958547359326076, "trajectory_balance_rmse": 338.88491773198905} {"canonical_length": 244, "canonical_nll": 1.735441114081711, "generation_rouge_l_f1": 0.29514563106796116, "generation_token_f1": 0.5320388349514563, "prompt_id": "d3428656e283f9ef", "reading_canonical_log_prob": -423.4476318359375, "reading_distances": [27, 34, 40, 40], "reading_log_probs": [-423.4302978515625, -428.9549560546875, -426.9656066894531, -429.5035095214844], "reading_logprob_std": 2.7477798525358716, "tokenization_path_logprob_std": 164.70045272974923, "trajectory_balance_rmse": 8390.885504244283} {"canonical_length": 18, "canonical_nll": 0.5126095347934299, "generation_rouge_l_f1": 0.5833333333333334, "generation_token_f1": 0.5833333333333334, "prompt_id": "f6a9e161c915b813", "reading_canonical_log_prob": -9.226971626281738, "reading_distances": [12, 15, 12, 19], "reading_log_probs": [-9.998424530029297, -10.113264083862305, -10.838650703430176, -9.893899917602539], "reading_logprob_std": 0.4278779004969656, "tokenization_path_logprob_std": 29.20325694910424, "trajectory_balance_rmse": 476.50161630288994} {"canonical_length": 4, "canonical_nll": 1.5063883066177368, "generation_rouge_l_f1": 0.2666666666666667, "generation_token_f1": 0.2666666666666667, "prompt_id": "60d17bcb4782ac31", "reading_canonical_log_prob": -6.025553226470947, "reading_distances": [14, 12, 12, 19], "reading_log_probs": [-5.306275367736816, -5.306537628173828, -5.202054977416992, -5.974785804748535], "reading_logprob_std": 0.3550062610064832, "tokenization_path_logprob_std": 30.432048804550377, "trajectory_balance_rmse": 72.77580663410788} {"canonical_length": 12, "canonical_nll": 1.8846673965454102, "generation_rouge_l_f1": 0.17204301075268816, "generation_token_f1": 0.23655913978494622, "prompt_id": "469967c1d14dd489", "reading_canonical_log_prob": -22.616008758544922, "reading_distances": [26, 23, 30, 26], "reading_log_probs": [-24.394126892089844, -24.691999435424805, -29.209779739379883, -30.309669494628906], "reading_logprob_std": 3.0475563697647514, "tokenization_path_logprob_std": 40.09091521875457, "trajectory_balance_rmse": 274.7100987256357} {"canonical_length": 269, "canonical_nll": 1.4763816634961664, "generation_rouge_l_f1": 0.1871750433275563, "generation_token_f1": 0.31542461005199307, "prompt_id": "812c384ebd7a8b2e", "reading_canonical_log_prob": -397.14666748046875, "reading_distances": [30, 30, 26, 37], "reading_log_probs": [-395.87255859375, -393.0088195800781, -393.33673095703125, -395.21875], "reading_logprob_std": 1.4021472411080411, "tokenization_path_logprob_std": 205.70929328714627, "trajectory_balance_rmse": 5940.560508926606} {"canonical_length": 99, "canonical_nll": 1.0884327743992661, "generation_rouge_l_f1": 0.1907356948228883, "generation_token_f1": 0.37057220708446864, "prompt_id": "40633ecaff73b957", "reading_canonical_log_prob": -107.75484466552734, "reading_distances": [19, 26, 23, 15], "reading_log_probs": [-107.68319702148438, -110.41767120361328, -112.34938049316406, -115.07573699951172], "reading_logprob_std": 3.1193263443249877, "tokenization_path_logprob_std": 269.0439133449895, "trajectory_balance_rmse": 3393.617640367435} {"canonical_length": 210, "canonical_nll": 1.401382300967262, "generation_rouge_l_f1": 0.25820568927789933, "generation_token_f1": 0.47264770240700216, "prompt_id": "cd8cb6a38db69f38", "reading_canonical_log_prob": -294.290283203125, "reading_distances": [16, 23, 20, 20], "reading_log_probs": [-298.5019836425781, -300.0389404296875, -297.4910888671875, -300.75775146484375], "reading_logprob_std": 1.4762552369662967, "tokenization_path_logprob_std": 193.54584110926533, "trajectory_balance_rmse": 5955.013227237892} {"canonical_length": 105, "canonical_nll": 1.2221099853515625, "generation_rouge_l_f1": 0.1939393939393939, "generation_token_f1": 0.2606060606060606, "prompt_id": "2de9ac9d5457eaa3", "reading_canonical_log_prob": -128.32154846191406, "reading_distances": [30, 28, 31, 30], "reading_log_probs": [-127.0050277709961, -126.51934051513672, -127.45323181152344, -126.5814437866211], "reading_logprob_std": 0.43322689344300336, "tokenization_path_logprob_std": 82.20197569096776, "trajectory_balance_rmse": 1431.7240123174486} {"canonical_length": 99, "canonical_nll": 1.3716365929805872, "generation_rouge_l_f1": 0.2696629213483146, "generation_token_f1": 0.40449438202247195, "prompt_id": "0b44770e70f57065", "reading_canonical_log_prob": -135.79202270507812, "reading_distances": [28, 23, 29, 32], "reading_log_probs": [-137.46917724609375, -139.27590942382812, -135.6884002685547, -134.241943359375], "reading_logprob_std": 2.1823870767413114, "tokenization_path_logprob_std": 90.03357860563767, "trajectory_balance_rmse": 2938.1058878150034} {"canonical_length": 111, "canonical_nll": 1.7548756642384573, "generation_rouge_l_f1": 0.2674418604651163, "generation_token_f1": 0.41279069767441856, "prompt_id": "f5df13177d4b7c6c", "reading_canonical_log_prob": -194.79119873046875, "reading_distances": [16, 19, 15, 16], "reading_log_probs": [-194.64935302734375, -193.1903839111328, -196.32891845703125, -195.79873657226562], "reading_logprob_std": 1.3906116453910375, "tokenization_path_logprob_std": 124.8610827985738, "trajectory_balance_rmse": 3091.225738388048} {"canonical_length": 129, "canonical_nll": 1.9770071340161701, "generation_rouge_l_f1": 0.23737373737373738, "generation_token_f1": 0.42929292929292934, "prompt_id": "b85464f813712359", "reading_canonical_log_prob": -255.03392028808594, "reading_distances": [41, 46, 44, 43], "reading_log_probs": [-252.3194580078125, -258.67303466796875, -254.28677368164062, -256.94659423828125], "reading_logprob_std": 2.812815677443458, "tokenization_path_logprob_std": 266.78555405595097, "trajectory_balance_rmse": 4470.391145414591} {"canonical_length": 52, "canonical_nll": 1.2977183415339544, "generation_rouge_l_f1": 0.14339622641509434, "generation_token_f1": 0.24150943396226413, "prompt_id": "fa52ad4bc1072fe5", "reading_canonical_log_prob": -67.48135375976562, "reading_distances": [47, 48, 45, 59], "reading_log_probs": [-62.948646545410156, -73.37863159179688, -67.22860717773438, -64.52144622802734], "reading_logprob_std": 4.5932704687863115, "tokenization_path_logprob_std": 20.238195527824185, "trajectory_balance_rmse": 1106.8053821251965} {"canonical_length": 192, "canonical_nll": 1.4199660619099934, "generation_rouge_l_f1": 0.3506493506493506, "generation_token_f1": 0.6277056277056278, "prompt_id": "63e9efbaa63103ce", "reading_canonical_log_prob": -272.63348388671875, "reading_distances": [16, 24, 25, 23], "reading_log_probs": [-276.17901611328125, -277.0229187011719, -275.6929931640625, -276.1348876953125], "reading_logprob_std": 0.5554959481295476, "tokenization_path_logprob_std": 186.94440047931326, "trajectory_balance_rmse": 5861.696700136095} {"canonical_length": 265, "canonical_nll": 1.3688832408977005, "generation_rouge_l_f1": 0.22676579925650558, "generation_token_f1": 0.44981412639405205, "prompt_id": "867f08ace4ea14ea", "reading_canonical_log_prob": -362.7540588378906, "reading_distances": [22, 25, 33, 21], "reading_log_probs": [-358.4359436035156, -352.11859130859375, -359.11944580078125, -351.5337219238281], "reading_logprob_std": 4.030240087376563, "tokenization_path_logprob_std": 70.9858210396289, "trajectory_balance_rmse": 7756.007651841746} {"canonical_length": 6, "canonical_nll": 0.0017995266243815422, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "126ead873a35c47e", "reading_canonical_log_prob": -0.010797159746289253, "reading_distances": [31, 37, 32, 27], "reading_log_probs": [-3.739593982696533, -5.126742362976074, -5.1210222244262695, -4.39491605758667], "reading_logprob_std": 0.6661333865779943, "tokenization_path_logprob_std": 48.628509501377195, "trajectory_balance_rmse": 148.14183022505466} {"canonical_length": 51, "canonical_nll": 2.982708500880821, "generation_rouge_l_f1": 0.13333333333333333, "generation_token_f1": 0.26666666666666666, "prompt_id": "9800df60020cf919", "reading_canonical_log_prob": -152.11813354492188, "reading_distances": [51, 49, 47, 41], "reading_log_probs": [-156.44882202148438, -156.00546264648438, -154.43743896484375, -158.98374938964844], "reading_logprob_std": 1.8856014683028186, "tokenization_path_logprob_std": 134.13457006865997, "trajectory_balance_rmse": 1448.850364015133} {"canonical_length": 11, "canonical_nll": 0.995550502430309, "generation_rouge_l_f1": 0.1818181818181818, "generation_token_f1": 0.19999999999999998, "prompt_id": "0aba661f60975419", "reading_canonical_log_prob": -10.951055526733398, "reading_distances": [12, 13, 20, 19], "reading_log_probs": [-7.372899055480957, -8.194276809692383, -7.507536888122559, -9.386259078979492], "reading_logprob_std": 0.9205269263626815, "tokenization_path_logprob_std": 53.29576643059389, "trajectory_balance_rmse": 306.5097588053844} {"canonical_length": 245, "canonical_nll": 1.2094037737165177, "generation_rouge_l_f1": 0.34467120181405897, "generation_token_f1": 0.5850340136054423, "prompt_id": "a78682b2f45a9311", "reading_canonical_log_prob": -296.3039245605469, "reading_distances": [15, 17, 17, 15], "reading_log_probs": [-296.0792541503906, -296.94482421875, -297.0755920410156, -296.0091857910156], "reading_logprob_std": 0.5609925795225699, "tokenization_path_logprob_std": 244.75991139752995, "trajectory_balance_rmse": 6455.449260870061} {"canonical_length": 41, "canonical_nll": 1.1384499712688168, "generation_rouge_l_f1": 0.11940298507462686, "generation_token_f1": 0.13432835820895522, "prompt_id": "e970a5e256be139f", "reading_canonical_log_prob": -46.676448822021484, "reading_distances": [14, 16, 12, 15], "reading_log_probs": [-42.17972183227539, -42.189308166503906, -38.732818603515625, -40.75924301147461], "reading_logprob_std": 1.6329369291359108, "tokenization_path_logprob_std": 64.65609286538317, "trajectory_balance_rmse": 447.3514422440459} {"canonical_length": 112, "canonical_nll": 1.763247353690011, "generation_rouge_l_f1": 0.24242424242424243, "generation_token_f1": 0.44696969696969696, "prompt_id": "0b4d03b967102211", "reading_canonical_log_prob": -197.48370361328125, "reading_distances": [23, 28, 26, 24], "reading_log_probs": [-192.6880340576172, -192.21392822265625, -193.5604248046875, -195.82406616210938], "reading_logprob_std": 1.6018416614700886, "tokenization_path_logprob_std": 333.3782316580653, "trajectory_balance_rmse": 3008.947855037819} {"canonical_length": 354, "canonical_nll": 1.2480715304444738, "generation_rouge_l_f1": 0.19834710743801653, "generation_token_f1": 0.509090909090909, "prompt_id": "e03c0c9ce1e8af04", "reading_canonical_log_prob": -441.81732177734375, "reading_distances": [11, 15, 15, 18], "reading_log_probs": [-442.85540771484375, -440.95257568359375, -443.08892822265625, -442.1701354980469], "reading_logprob_std": 0.9589722676542298, "tokenization_path_logprob_std": 304.0849552118258, "trajectory_balance_rmse": 10018.827432852906} {"canonical_length": 114, "canonical_nll": 2.124293277138158, "generation_rouge_l_f1": 0.14537444933920707, "generation_token_f1": 0.24229074889867844, "prompt_id": "6aab22322acba4ab", "reading_canonical_log_prob": -242.16943359375, "reading_distances": [49, 44, 49, 43], "reading_log_probs": [-263.2599182128906, -248.7513427734375, -255.71803283691406, -269.6463317871094], "reading_logprob_std": 9.07054722079315, "tokenization_path_logprob_std": 125.32204288350013, "trajectory_balance_rmse": 2864.336782705457} {"canonical_length": 72, "canonical_nll": 1.0971724192301433, "generation_rouge_l_f1": 0.384, "generation_token_f1": 0.528, "prompt_id": "1498d74fffe04fb0", "reading_canonical_log_prob": -78.99641418457031, "reading_distances": [27, 29, 20, 31], "reading_log_probs": [-88.02192687988281, -79.38339233398438, -79.766357421875, -86.12966918945312], "reading_logprob_std": 4.401798833661047, "tokenization_path_logprob_std": 85.85625092619031, "trajectory_balance_rmse": 1835.3119187981001} {"canonical_length": 203, "canonical_nll": 1.7632432759101755, "generation_rouge_l_f1": 0.21524663677130043, "generation_token_f1": 0.46636771300448426, "prompt_id": "363570662fa8a4e6", "reading_canonical_log_prob": -357.9383850097656, "reading_distances": [31, 30, 29, 32], "reading_log_probs": [-367.6279296875, -368.04736328125, -368.0283203125, -367.9965515136719], "reading_logprob_std": 0.19918018849072888, "tokenization_path_logprob_std": 252.9763301974234, "trajectory_balance_rmse": 6371.627935067935} {"canonical_length": 267, "canonical_nll": 1.7525717060217696, "generation_rouge_l_f1": 0.21301775147928995, "generation_token_f1": 0.4536489151873767, "prompt_id": "2b4383e7d7d3624e", "reading_canonical_log_prob": -467.9366455078125, "reading_distances": [25, 34, 34, 33], "reading_log_probs": [-464.78570556640625, -463.3837585449219, -464.05859375, -466.3780517578125], "reading_logprob_std": 1.2855217256907059, "tokenization_path_logprob_std": 376.019531332322, "trajectory_balance_rmse": 7657.349331810766} {"canonical_length": 72, "canonical_nll": 1.47495608859592, "generation_rouge_l_f1": 0.29059829059829057, "generation_token_f1": 0.37037037037037035, "prompt_id": "fadbe59576c37aff", "reading_canonical_log_prob": -106.19683837890625, "reading_distances": [19, 22, 16, 20], "reading_log_probs": [-110.7093734741211, -110.34587097167969, -105.9189453125, -106.91261291503906], "reading_logprob_std": 2.412951749822568, "tokenization_path_logprob_std": 106.59776054515706, "trajectory_balance_rmse": 1508.1199360000808} {"canonical_length": 119, "canonical_nll": 1.6984158043100053, "generation_rouge_l_f1": 0.16230366492146597, "generation_token_f1": 0.3769633507853403, "prompt_id": "6caa7e0c0b6b5ae9", "reading_canonical_log_prob": -202.11148071289062, "reading_distances": [26, 30, 24, 30], "reading_log_probs": [-202.5353240966797, -201.86172485351562, -200.1983642578125, -203.45339965820312], "reading_logprob_std": 1.373967266062858, "tokenization_path_logprob_std": 48.31218884270413, "trajectory_balance_rmse": 4008.6527033229045} {"canonical_length": 246, "canonical_nll": 1.23643741762735, "generation_rouge_l_f1": 0.23856858846918488, "generation_token_f1": 0.47713717693836977, "prompt_id": "09ec255dc57dfab9", "reading_canonical_log_prob": -304.1636047363281, "reading_distances": [26, 27, 21, 29], "reading_log_probs": [-321.5634765625, -311.6923522949219, -317.40484619140625, -319.99285888671875], "reading_logprob_std": 4.334252594755744, "tokenization_path_logprob_std": 173.03708339156836, "trajectory_balance_rmse": 7089.893972336543} {"canonical_length": 83, "canonical_nll": 1.7458288353609752, "generation_rouge_l_f1": 0.3549295774647887, "generation_token_f1": 0.43943661971830983, "prompt_id": "a8f654f305639b3f", "reading_canonical_log_prob": -144.90379333496094, "reading_distances": [17, 20, 13, 15], "reading_log_probs": [-159.84674072265625, -156.27731323242188, -161.9899444580078, -156.2120361328125], "reading_logprob_std": 2.8367800635628417, "tokenization_path_logprob_std": 60.92023943640682, "trajectory_balance_rmse": 920.5500449708646} {"canonical_length": 286, "canonical_nll": 1.4077418400691106, "generation_rouge_l_f1": 0.25984251968503935, "generation_token_f1": 0.4566929133858268, "prompt_id": "0e8185255effe518", "reading_canonical_log_prob": -402.6141662597656, "reading_distances": [30, 29, 26, 25], "reading_log_probs": [-403.8355712890625, -406.37286376953125, -404.47406005859375, -401.34027099609375], "reading_logprob_std": 2.0781627309661594, "tokenization_path_logprob_std": 221.53218193609743, "trajectory_balance_rmse": 6727.343138451263} {"canonical_length": 51, "canonical_nll": 1.0870537851371018, "generation_rouge_l_f1": 0.47058823529411764, "generation_token_f1": 0.6117647058823529, "prompt_id": "d03a9ea989bf62b9", "reading_canonical_log_prob": -55.43974304199219, "reading_distances": [30, 36, 34, 33], "reading_log_probs": [-54.68505859375, -51.275962829589844, -56.29521560668945, -53.66404724121094], "reading_logprob_std": 2.103090905518414, "tokenization_path_logprob_std": 179.7431774338849, "trajectory_balance_rmse": 1240.6556598404336} {"canonical_length": 69, "canonical_nll": 0.741579028143399, "generation_rouge_l_f1": 0.3695652173913043, "generation_token_f1": 0.5326086956521738, "prompt_id": "fd11f4c3417e5ef0", "reading_canonical_log_prob": -51.16895294189453, "reading_distances": [44, 42, 39, 41], "reading_log_probs": [-47.069828033447266, -52.1970329284668, -51.97439956665039, -54.224693298339844], "reading_logprob_std": 3.0380907242709365, "tokenization_path_logprob_std": 121.43663751175045, "trajectory_balance_rmse": 1855.5208342655033} {"canonical_length": 40, "canonical_nll": 1.7723638534545898, "generation_rouge_l_f1": 0.23157894736842105, "generation_token_f1": 0.29473684210526313, "prompt_id": "43fe74e6e2773223", "reading_canonical_log_prob": -70.8945541381836, "reading_distances": [16, 14, 13, 20], "reading_log_probs": [-71.77238464355469, -68.89543151855469, -74.30182647705078, -72.98078155517578], "reading_logprob_std": 2.3057809491053574, "tokenization_path_logprob_std": 143.6234622295377, "trajectory_balance_rmse": 879.7091093074142} {"canonical_length": 82, "canonical_nll": 1.3182049262814406, "generation_rouge_l_f1": 0.1246458923512748, "generation_token_f1": 0.1869688385269122, "prompt_id": "7eeeafb10f4366ae", "reading_canonical_log_prob": -108.09280395507812, "reading_distances": [30, 28, 34, 28], "reading_log_probs": [-106.76575469970703, -108.08560943603516, -106.47454071044922, -107.59259033203125], "reading_logprob_std": 0.741570658985702, "tokenization_path_logprob_std": 47.70673439630889, "trajectory_balance_rmse": 1411.6395333220323} {"canonical_length": 3, "canonical_nll": 3.710138956705729, "generation_rouge_l_f1": 0.042553191489361694, "generation_token_f1": 0.042553191489361694, "prompt_id": "6b50f3d234f6e0b4", "reading_canonical_log_prob": -11.130416870117188, "reading_distances": [20, 25, 26, 21], "reading_log_probs": [-13.046289443969727, -13.392536163330078, -12.506168365478516, -12.67972183227539], "reading_logprob_std": 0.39474277843341526, "tokenization_path_logprob_std": 21.466720102931017, "trajectory_balance_rmse": 33.73255106685643} {"canonical_length": 348, "canonical_nll": 1.2529084655060165, "generation_rouge_l_f1": 0.22866894197952217, "generation_token_f1": 0.45051194539249145, "prompt_id": "b62b5836a5079a74", "reading_canonical_log_prob": -436.01214599609375, "reading_distances": [31, 37, 29, 30], "reading_log_probs": [-435.08160400390625, -437.33203125, -438.59942626953125, -435.1235656738281], "reading_logprob_std": 1.7322065475779058, "tokenization_path_logprob_std": 649.3171337445317, "trajectory_balance_rmse": 9959.144410400291} {"canonical_length": 39, "canonical_nll": 1.1814692570612981, "generation_rouge_l_f1": 0.4912280701754386, "generation_token_f1": 0.5263157894736843, "prompt_id": "495dc4d7314e6352", "reading_canonical_log_prob": -46.077301025390625, "reading_distances": [27, 21, 27, 27], "reading_log_probs": [-46.943748474121094, -47.37027359008789, -44.865989685058594, -47.31033706665039], "reading_logprob_std": 1.1861442029384819, "tokenization_path_logprob_std": 139.38260525046803, "trajectory_balance_rmse": 1317.735025463865} {"canonical_length": 54, "canonical_nll": 1.3006345960828993, "generation_rouge_l_f1": 0.4444444444444445, "generation_token_f1": 0.4691358024691358, "prompt_id": "0c8a53a7cd7794b9", "reading_canonical_log_prob": -70.23426818847656, "reading_distances": [24, 25, 26, 30], "reading_log_probs": [-75.13040161132812, -68.35975646972656, -68.38762664794922, -71.12962341308594], "reading_logprob_std": 3.1951051354971307, "tokenization_path_logprob_std": 75.48664408712594, "trajectory_balance_rmse": 1467.187443940913} {"canonical_length": 185, "canonical_nll": 1.9652846877639358, "generation_rouge_l_f1": 0.263681592039801, "generation_token_f1": 0.5323383084577115, "prompt_id": "bfd0b674c803f580", "reading_canonical_log_prob": -363.5776672363281, "reading_distances": [31, 33, 30, 37], "reading_log_probs": [-363.3009033203125, -368.67572021484375, -365.25701904296875, -364.62884521484375], "reading_logprob_std": 2.290156333309266, "tokenization_path_logprob_std": 271.8682664636611, "trajectory_balance_rmse": 5879.769880642262} {"canonical_length": 215, "canonical_nll": 1.6051735101744187, "generation_rouge_l_f1": 0.19354838709677416, "generation_token_f1": 0.43870967741935485, "prompt_id": "f65fb83f66989dab", "reading_canonical_log_prob": -345.1123046875, "reading_distances": [23, 31, 22, 26], "reading_log_probs": [-352.22540283203125, -353.61419677734375, -353.6289978027344, -352.2074890136719], "reading_logprob_std": 0.8113199846195733, "tokenization_path_logprob_std": 558.8217602698213, "trajectory_balance_rmse": 5155.240892468745} {"canonical_length": 20, "canonical_nll": 0.5430198192596436, "generation_rouge_l_f1": 0.7272727272727272, "generation_token_f1": 0.7272727272727272, "prompt_id": "ed802f2ffb8ef795", "reading_canonical_log_prob": -10.860396385192871, "reading_distances": [22, 18, 19, 22], "reading_log_probs": [-11.372060775756836, -11.415281295776367, -12.333749771118164, -11.202007293701172], "reading_logprob_std": 0.5103547050219829, "tokenization_path_logprob_std": 68.5334664886228, "trajectory_balance_rmse": 608.2826087915298} {"canonical_length": 17, "canonical_nll": 0.5303370531867532, "generation_rouge_l_f1": 0.1656050955414013, "generation_token_f1": 0.19108280254777069, "prompt_id": "578c6b312be28f5a", "reading_canonical_log_prob": -9.015729904174805, "reading_distances": [29, 21, 29, 32], "reading_log_probs": [-12.046658515930176, -11.991169929504395, -9.811100006103516, -8.765899658203125], "reading_logprob_std": 1.6332915762794367, "tokenization_path_logprob_std": 63.31972670481235, "trajectory_balance_rmse": 548.7099362682156} {"canonical_length": 20, "canonical_nll": 0.9083683967590332, "generation_rouge_l_f1": 0.3076923076923077, "generation_token_f1": 0.3076923076923077, "prompt_id": "016449fd389e919a", "reading_canonical_log_prob": -18.167367935180664, "reading_distances": [13, 19, 14, 16], "reading_log_probs": [-20.500957489013672, -19.37791633605957, -19.27174186706543, -20.64576530456543], "reading_logprob_std": 0.7245582683908077, "tokenization_path_logprob_std": 35.44168794866103, "trajectory_balance_rmse": 607.2020533066033} {"canonical_length": 172, "canonical_nll": 1.4111211022665335, "generation_rouge_l_f1": 0.29166666666666663, "generation_token_f1": 0.5231481481481481, "prompt_id": "977004fae436eabd", "reading_canonical_log_prob": -242.71282958984375, "reading_distances": [26, 27, 25, 25], "reading_log_probs": [-244.25637817382812, -244.04177856445312, -243.3173828125, -245.71742248535156], "reading_logprob_std": 1.0064433229417216, "tokenization_path_logprob_std": 309.44675211451295, "trajectory_balance_rmse": 4833.537289211001} {"canonical_length": 223, "canonical_nll": 1.7102926622057175, "generation_rouge_l_f1": 0.18556701030927833, "generation_token_f1": 0.44948453608247424, "prompt_id": "3d21a3aabdc36c10", "reading_canonical_log_prob": -381.395263671875, "reading_distances": [32, 23, 34, 31], "reading_log_probs": [-382.75164794921875, -376.2861328125, -384.52392578125, -380.46630859375], "reading_logprob_std": 3.558627901138471, "tokenization_path_logprob_std": 330.65148612792177, "trajectory_balance_rmse": 5518.247323834242} {"canonical_length": 248, "canonical_nll": 0.4292257370487336, "generation_rouge_l_f1": 0.42857142857142855, "generation_token_f1": 0.6134453781512605, "prompt_id": "99f1a0976d255260", "reading_canonical_log_prob": -106.44798278808594, "reading_distances": [34, 32, 34, 32], "reading_log_probs": [-105.33452606201172, -105.00304412841797, -107.83325958251953, -105.45906066894531], "reading_logprob_std": 1.298202264358273, "tokenization_path_logprob_std": 261.7888638554283, "trajectory_balance_rmse": 5200.844473184483} {"canonical_length": 183, "canonical_nll": 1.2534346450221994, "generation_rouge_l_f1": 0.40865384615384615, "generation_token_f1": 0.625, "prompt_id": "24640c4b7c2fb283", "reading_canonical_log_prob": -229.3785400390625, "reading_distances": [72, 61, 66, 65], "reading_log_probs": [-215.25604248046875, -220.461669921875, -221.50314331054688, -221.35302734375], "reading_logprob_std": 2.9608524946274053, "tokenization_path_logprob_std": 292.1191541153206, "trajectory_balance_rmse": 4258.749571127683} {"canonical_length": 6, "canonical_nll": 0.003996663416425387, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "94e6b38d533edcd5", "reading_canonical_log_prob": -0.023979980498552322, "reading_distances": [35, 35, 38, 30], "reading_log_probs": [-0.22313223779201508, -0.2354028970003128, -0.22433282434940338, -0.3384454548358917], "reading_logprob_std": 0.05568599179090213, "tokenization_path_logprob_std": 57.65707730854956, "trajectory_balance_rmse": 163.90116845955572} {"canonical_length": 252, "canonical_nll": 1.5384798806811135, "generation_rouge_l_f1": 0.26510721247563346, "generation_token_f1": 0.49902534113060426, "prompt_id": "22a8bdc8dd7cf586", "reading_canonical_log_prob": -387.6969299316406, "reading_distances": [27, 22, 23, 22], "reading_log_probs": [-390.6044616699219, -392.34600830078125, -391.2863464355469, -390.9796142578125], "reading_logprob_std": 0.7484813259416412, "tokenization_path_logprob_std": 133.85310891333407, "trajectory_balance_rmse": 7647.290941803512} {"canonical_length": 27, "canonical_nll": 1.7374064127604167, "generation_rouge_l_f1": 0.1590909090909091, "generation_token_f1": 0.18181818181818182, "prompt_id": "6585d8646962bd31", "reading_canonical_log_prob": -46.90997314453125, "reading_distances": [16, 14, 18, 12], "reading_log_probs": [-42.49705505371094, -42.655860900878906, -41.77478790283203, -42.19593048095703], "reading_logprob_std": 0.38759427163124716, "tokenization_path_logprob_std": 90.41757207232979, "trajectory_balance_rmse": 394.8742153446689} {"canonical_length": 81, "canonical_nll": 1.4333331261152102, "generation_rouge_l_f1": 0.20809248554913293, "generation_token_f1": 0.32947976878612717, "prompt_id": "04a5aa53ab7903f1", "reading_canonical_log_prob": -116.09998321533203, "reading_distances": [22, 17, 27, 27], "reading_log_probs": [-115.85982513427734, -114.69671630859375, -114.81867980957031, -114.949951171875], "reading_logprob_std": 0.5292220325717885, "tokenization_path_logprob_std": 142.43102647209778, "trajectory_balance_rmse": 2601.4981532007337} {"canonical_length": 198, "canonical_nll": 1.0687198060931582, "generation_rouge_l_f1": 0.2424242424242424, "generation_token_f1": 0.4978354978354978, "prompt_id": "f2143d23c11c6f8b", "reading_canonical_log_prob": -211.6065216064453, "reading_distances": [20, 24, 23, 20], "reading_log_probs": [-209.96885681152344, -211.6246795654297, -215.1112060546875, -213.7130126953125], "reading_logprob_std": 2.26708614203708, "tokenization_path_logprob_std": 262.8357446612734, "trajectory_balance_rmse": 5146.994680196662} {"canonical_length": 12, "canonical_nll": 1.3335366249084473, "generation_rouge_l_f1": 0.761904761904762, "generation_token_f1": 0.761904761904762, "prompt_id": "5633164a0de34e8b", "reading_canonical_log_prob": -16.002439498901367, "reading_distances": [36, 42, 40, 33], "reading_log_probs": [-13.152137756347656, -15.579936981201172, -16.634571075439453, -13.21589469909668], "reading_logprob_std": 1.7419795491609662, "tokenization_path_logprob_std": 54.11301177629544, "trajectory_balance_rmse": 329.2543249257388} {"canonical_length": 134, "canonical_nll": 1.417812176604769, "generation_rouge_l_f1": 0.16494845360824745, "generation_token_f1": 0.2938144329896907, "prompt_id": "fe842e52ca168c74", "reading_canonical_log_prob": -189.98683166503906, "reading_distances": [26, 25, 19, 27], "reading_log_probs": [-178.2610626220703, -175.24942016601562, -182.39950561523438, -170.69439697265625], "reading_logprob_std": 4.9356919161900965, "tokenization_path_logprob_std": 294.68282892719407, "trajectory_balance_rmse": 3053.8953210126497} {"canonical_length": 46, "canonical_nll": 0.38311381961988367, "generation_rouge_l_f1": 0.3957219251336898, "generation_token_f1": 0.3957219251336898, "prompt_id": "6327942806f18cd4", "reading_canonical_log_prob": -17.62323570251465, "reading_distances": [7, 13, 10, 12], "reading_log_probs": [-17.98184585571289, -17.962078094482422, -17.251523971557617, -17.792415618896484], "reading_logprob_std": 0.34106217967657476, "tokenization_path_logprob_std": 47.768823940263424, "trajectory_balance_rmse": 653.2852490457097} {"canonical_length": 54, "canonical_nll": 1.0596438513861761, "generation_rouge_l_f1": 0.18106995884773663, "generation_token_f1": 0.27983539094650206, "prompt_id": "95494e563c3fc530", "reading_canonical_log_prob": -57.220767974853516, "reading_distances": [29, 29, 30, 30], "reading_log_probs": [-60.3541259765625, -60.392391204833984, -63.171104431152344, -60.69454574584961], "reading_logprob_std": 1.3539634701854928, "tokenization_path_logprob_std": 68.38058180322895, "trajectory_balance_rmse": 1544.0381364938942} {"canonical_length": 44, "canonical_nll": 1.992135308005593, "generation_rouge_l_f1": 0.15646258503401358, "generation_token_f1": 0.21088435374149658, "prompt_id": "15a442e093e950e5", "reading_canonical_log_prob": -87.6539535522461, "reading_distances": [23, 29, 30, 26], "reading_log_probs": [-94.99427032470703, -89.85549926757812, -94.53655242919922, -91.60966491699219], "reading_logprob_std": 2.443154616338048, "tokenization_path_logprob_std": 102.61784558159611, "trajectory_balance_rmse": 1066.8772746319405} {"canonical_length": 303, "canonical_nll": 0.7703249084674092, "generation_rouge_l_f1": 0.23182711198428288, "generation_token_f1": 0.40078585461689586, "prompt_id": "452a884e801c54f0", "reading_canonical_log_prob": -233.408447265625, "reading_distances": [15, 24, 29, 28], "reading_log_probs": [-240.03726196289062, -236.71884155273438, -236.1664276123047, -234.37095642089844], "reading_logprob_std": 2.365441002513698, "tokenization_path_logprob_std": 162.37040696678855, "trajectory_balance_rmse": 7093.396641819773} {"canonical_length": 340, "canonical_nll": 1.2784547693589154, "generation_rouge_l_f1": 0.22847100175746923, "generation_token_f1": 0.3796133567662566, "prompt_id": "6f1aa7f491e54018", "reading_canonical_log_prob": -434.67462158203125, "reading_distances": [25, 23, 16, 18], "reading_log_probs": [-434.02764892578125, -435.0032958984375, -432.2874450683594, -436.2356872558594], "reading_logprob_std": 1.66680533526395, "tokenization_path_logprob_std": 284.4721271718539, "trajectory_balance_rmse": 9221.506092730277} {"canonical_length": 197, "canonical_nll": 0.8966727474619289, "generation_rouge_l_f1": 0.281767955801105, "generation_token_f1": 0.585635359116022, "prompt_id": "b6dbd4688fd838f3", "reading_canonical_log_prob": -176.64453125, "reading_distances": [20, 19, 18, 22], "reading_log_probs": [-175.55767822265625, -174.3721923828125, -174.04742431640625, -174.11917114257812], "reading_logprob_std": 0.7029810221052113, "tokenization_path_logprob_std": 448.90932340905374, "trajectory_balance_rmse": 6058.745793700401} {"canonical_length": 138, "canonical_nll": 1.8437146173007246, "generation_rouge_l_f1": 0.20454545454545453, "generation_token_f1": 0.3636363636363636, "prompt_id": "4e91817c9d8f7b21", "reading_canonical_log_prob": -254.4326171875, "reading_distances": [17, 20, 19, 20], "reading_log_probs": [-261.39910888671875, -262.50958251953125, -262.78350830078125, -260.97137451171875], "reading_logprob_std": 0.8687931380550724, "tokenization_path_logprob_std": 57.44784501389167, "trajectory_balance_rmse": 2698.6373728785293} {"canonical_length": 10, "canonical_nll": 1.2951017379760743, "generation_rouge_l_f1": 0.5454545454545454, "generation_token_f1": 0.5454545454545454, "prompt_id": "8f6ce2573cd1b9d1", "reading_canonical_log_prob": -12.951017379760742, "reading_distances": [37, 33, 38, 35], "reading_log_probs": [-14.126073837280273, -13.605907440185547, -15.545391082763672, -18.90749168395996], "reading_logprob_std": 2.386055040835706, "tokenization_path_logprob_std": 38.23086254796367, "trajectory_balance_rmse": 238.7194356464596} {"canonical_length": 9, "canonical_nll": 0.5354229079352485, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.33333333333333337, "prompt_id": "c246d0eff1fb5aac", "reading_canonical_log_prob": -4.818806171417236, "reading_distances": [12, 18, 17, 17], "reading_log_probs": [-5.029733657836914, -4.085846424102783, -6.0025177001953125, -4.360958576202393], "reading_logprob_std": 0.8528639910757617, "tokenization_path_logprob_std": 17.483752789806825, "trajectory_balance_rmse": 190.55761609077192} {"canonical_length": 222, "canonical_nll": 0.72038104083087, "generation_rouge_l_f1": 0.3938730853391685, "generation_token_f1": 0.5864332603938731, "prompt_id": "877651642570fc39", "reading_canonical_log_prob": -159.92459106445312, "reading_distances": [31, 33, 32, 30], "reading_log_probs": [-187.34130859375, -176.96578979492188, -180.0679168701172, -172.9409637451172], "reading_logprob_std": 6.086456462491025, "tokenization_path_logprob_std": 289.74487635338846, "trajectory_balance_rmse": 4419.582823316} {"canonical_length": 53, "canonical_nll": 0.9420101957501106, "generation_rouge_l_f1": 0.22140221402214022, "generation_token_f1": 0.2656826568265682, "prompt_id": "77a973a8d7f91484", "reading_canonical_log_prob": -49.92654037475586, "reading_distances": [31, 29, 31, 33], "reading_log_probs": [-49.805580139160156, -49.86518859863281, -52.18898010253906, -52.12822341918945], "reading_logprob_std": 1.341760202352084, "tokenization_path_logprob_std": 77.72014130278792, "trajectory_balance_rmse": 1478.50705621353} {"canonical_length": 21, "canonical_nll": 0.98091797601609, "generation_rouge_l_f1": 0.10238907849829351, "generation_token_f1": 0.1092150170648464, "prompt_id": "f9bfdd6eb133d424", "reading_canonical_log_prob": -20.59927749633789, "reading_distances": [24, 24, 22, 26], "reading_log_probs": [-25.573650360107422, -26.26983070373535, -23.94405746459961, -27.24385643005371], "reading_logprob_std": 1.3897376514967013, "tokenization_path_logprob_std": 61.942213199100216, "trajectory_balance_rmse": 555.011110235593} {"canonical_length": 247, "canonical_nll": 1.5375458875648407, "generation_rouge_l_f1": 0.20233463035019456, "generation_token_f1": 0.4163424124513619, "prompt_id": "3a822df2306dda39", "reading_canonical_log_prob": -379.7738342285156, "reading_distances": [39, 36, 48, 42], "reading_log_probs": [-395.03009033203125, -385.8055419921875, -393.8201904296875, -396.9734802246094], "reading_logprob_std": 4.909460124706944, "tokenization_path_logprob_std": 348.96299408129795, "trajectory_balance_rmse": 7409.700594469256} {"canonical_length": 88, "canonical_nll": 1.2097471410577947, "generation_rouge_l_f1": 0.1935483870967742, "generation_token_f1": 0.3790322580645161, "prompt_id": "c28f9f6ed8615ffa", "reading_canonical_log_prob": -106.45774841308594, "reading_distances": [26, 27, 29, 29], "reading_log_probs": [-104.14479064941406, -100.3467788696289, -101.44647979736328, -99.98611450195312], "reading_logprob_std": 1.8813215644361894, "tokenization_path_logprob_std": 217.85814535263683, "trajectory_balance_rmse": 2589.6697712786963}