diff --git "a/test_metrics.per_example.jsonl" "b/test_metrics.per_example.jsonl" new file mode 100644--- /dev/null +++ "b/test_metrics.per_example.jsonl" @@ -0,0 +1,2477 @@ +{"canonical_length": 14, "canonical_nll": 1.82518127986363, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.3157894736842105, "prompt_id": "54ef1334aadf1dfd", "reading_canonical_log_prob": -25.55253791809082, "reading_distances": [37, 36, 33, 38], "reading_log_probs": [-28.16619110107422, -28.17404556274414, -28.287145614624023, -27.271472930908203], "reading_logprob_std": 0.472072758239518, "tokenization_path_logprob_std": 35.70947493396841, "trajectory_balance_rmse": 263.1766089553883} +{"canonical_length": 23, "canonical_nll": 0.7400279667066492, "generation_rouge_l_f1": 0.0916030534351145, "generation_token_f1": 0.09923664122137404, "prompt_id": "d109cb9b45bb2b84", "reading_canonical_log_prob": -17.02064323425293, "reading_distances": [12, 12, 9, 10], "reading_log_probs": [-21.5074520111084, -21.187236785888672, -20.575363159179688, -21.360126495361328], "reading_logprob_std": 0.4095898259703365, "tokenization_path_logprob_std": 49.198752958950315, "trajectory_balance_rmse": 237.44418480581024} +{"canonical_length": 79, "canonical_nll": 1.7332842863058742, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.42857142857142855, "prompt_id": "f461f81672c6f974", "reading_canonical_log_prob": -136.92945861816406, "reading_distances": [24, 27, 34, 27], "reading_log_probs": [-134.41493225097656, -133.47084045410156, -133.977783203125, -133.48020935058594], "reading_logprob_std": 0.45284047754315904, "tokenization_path_logprob_std": 173.8838342374747, "trajectory_balance_rmse": 2032.8125486657716} +{"canonical_length": 22, "canonical_nll": 0.8214158144864169, "generation_rouge_l_f1": 0.14851485148514854, "generation_token_f1": 0.14851485148514854, "prompt_id": "db2b92b024f2b9fd", "reading_canonical_log_prob": -18.071147918701172, "reading_distances": [20, 13, 17, 14], "reading_log_probs": [-18.741168975830078, -17.828441619873047, -18.654094696044922, -17.99125099182129], "reading_logprob_std": 0.46103161580555685, "tokenization_path_logprob_std": 54.87061665743557, "trajectory_balance_rmse": 200.5920947361285} +{"canonical_length": 22, "canonical_nll": 1.8255851052024148, "generation_rouge_l_f1": 0.4666666666666667, "generation_token_f1": 0.4666666666666667, "prompt_id": "a3ca3e528a4c44fb", "reading_canonical_log_prob": -40.162872314453125, "reading_distances": [27, 30, 28, 27], "reading_log_probs": [-42.11420440673828, -42.45229721069336, -42.38743591308594, -41.70882034301758], "reading_logprob_std": 0.33798204734145293, "tokenization_path_logprob_std": 110.15211166287826, "trajectory_balance_rmse": 707.9207704486764} +{"canonical_length": 46, "canonical_nll": 1.324197935021442, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.3157894736842105, "prompt_id": "bea2699cff758312", "reading_canonical_log_prob": -60.91310501098633, "reading_distances": [21, 15, 20, 25], "reading_log_probs": [-59.030967712402344, -60.36042404174805, -59.75444030761719, -58.828094482421875], "reading_logprob_std": 0.7015097425897272, "tokenization_path_logprob_std": 51.62216683652098, "trajectory_balance_rmse": 1365.5883307957556} +{"canonical_length": 41, "canonical_nll": 1.1907001588402726, "generation_rouge_l_f1": 0.16988416988416988, "generation_token_f1": 0.17760617760617758, "prompt_id": "cf9306b8e22601b5", "reading_canonical_log_prob": -48.81870651245117, "reading_distances": [16, 21, 21, 18], "reading_log_probs": [-48.48314666748047, -49.19281005859375, -48.352874755859375, -49.16201400756836], "reading_logprob_std": 0.4418332833468075, "tokenization_path_logprob_std": 56.80758758264843, "trajectory_balance_rmse": 411.17751820124863} +{"canonical_length": 231, "canonical_nll": 1.2500288001386635, "generation_rouge_l_f1": 0.2566191446028513, "generation_token_f1": 0.46843177189409374, "prompt_id": "ff6c469935e35f5e", "reading_canonical_log_prob": -288.75665283203125, "reading_distances": [17, 21, 23, 17], "reading_log_probs": [-284.0778503417969, -285.37603759765625, -284.4306640625, -282.06561279296875], "reading_logprob_std": 1.3935858615361478, "tokenization_path_logprob_std": 283.1248023571745, "trajectory_balance_rmse": 5211.32955283666} +{"canonical_length": 203, "canonical_nll": 1.4845062406192273, "generation_rouge_l_f1": 0.21786492374727673, "generation_token_f1": 0.44880174291938996, "prompt_id": "41c1bf9aacb45501", "reading_canonical_log_prob": -301.3547668457031, "reading_distances": [23, 22, 21, 22], "reading_log_probs": [-302.44183349609375, -302.0894775390625, -301.19732666015625, -302.073486328125], "reading_logprob_std": 0.5301317677896008, "tokenization_path_logprob_std": 125.9139307723619, "trajectory_balance_rmse": 5112.981442754103} +{"canonical_length": 33, "canonical_nll": 1.3606444850112454, "generation_rouge_l_f1": 0.1142857142857143, "generation_token_f1": 0.1142857142857143, "prompt_id": "aff22e4e1c81f916", "reading_canonical_log_prob": -44.901268005371094, "reading_distances": [25, 29, 34, 25], "reading_log_probs": [-47.93975830078125, -45.254364013671875, -46.329627990722656, -46.668190002441406], "reading_logprob_std": 1.1064383102472768, "tokenization_path_logprob_std": 153.36478634438706, "trajectory_balance_rmse": 878.2383583718257} +{"canonical_length": 10, "canonical_nll": 1.3421123504638672, "generation_rouge_l_f1": 0.031007751937984496, "generation_token_f1": 0.031007751937984496, "prompt_id": "f599bb889055c49d", "reading_canonical_log_prob": -13.421123504638672, "reading_distances": [63, 53, 60, 54], "reading_log_probs": [-14.23855972290039, -14.666946411132812, -14.761606216430664, -14.027799606323242], "reading_logprob_std": 0.34850321300089016, "tokenization_path_logprob_std": 56.18276097897103, "trajectory_balance_rmse": 214.06620464452016} +{"canonical_length": 90, "canonical_nll": 1.2612045288085938, "generation_rouge_l_f1": 0.17679558011049726, "generation_token_f1": 0.3259668508287293, "prompt_id": "7fbc599a6c4c6b54", "reading_canonical_log_prob": -113.50840759277344, "reading_distances": [21, 34, 28, 33], "reading_log_probs": [-122.42788696289062, -124.18341064453125, -122.1878662109375, -124.900390625], "reading_logprob_std": 1.3262350939482332, "tokenization_path_logprob_std": 100.88403791924938, "trajectory_balance_rmse": 2083.8790754740467} +{"canonical_length": 294, "canonical_nll": 1.5990262193744684, "generation_rouge_l_f1": 0.040268456375838924, "generation_token_f1": 0.040268456375838924, "prompt_id": "9e624d36638d2bbd", "reading_canonical_log_prob": -470.11370849609375, "reading_distances": [20, 13, 22, 16], "reading_log_probs": [-465.2415466308594, -466.26324462890625, -464.5204162597656, -465.3994140625], "reading_logprob_std": 0.7156063306880969, "tokenization_path_logprob_std": 260.0448486189809, "trajectory_balance_rmse": 6397.72304647467} +{"canonical_length": 27, "canonical_nll": 0.8465338812934028, "generation_rouge_l_f1": 0.07586206896551724, "generation_token_f1": 0.07586206896551724, "prompt_id": "3982276a331d29f5", "reading_canonical_log_prob": -22.856414794921875, "reading_distances": [26, 23, 24, 25], "reading_log_probs": [-21.302274703979492, -21.58027458190918, -22.27667236328125, -21.75732421875], "reading_logprob_std": 0.41026858382731024, "tokenization_path_logprob_std": 36.39817606738327, "trajectory_balance_rmse": 342.8456708550408} +{"canonical_length": 10, "canonical_nll": 1.2497442245483399, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.5, "prompt_id": "3b79dbc85baf9416", "reading_canonical_log_prob": -12.497442245483398, "reading_distances": [21, 25, 21, 19], "reading_log_probs": [-12.817241668701172, -12.873235702514648, -12.226289749145508, -12.476105690002441], "reading_logprob_std": 0.3037807017044884, "tokenization_path_logprob_std": 26.953457706424455, "trajectory_balance_rmse": 211.36543771294828} +{"canonical_length": 174, "canonical_nll": 1.4130349871756016, "generation_rouge_l_f1": 0.25404157043879905, "generation_token_f1": 0.4803695150115473, "prompt_id": "bc017f1a688cb1fb", "reading_canonical_log_prob": -245.8680877685547, "reading_distances": [23, 20, 25, 28], "reading_log_probs": [-244.94082641601562, -244.705078125, -244.68228149414062, -245.13499450683594], "reading_logprob_std": 0.21416909877673201, "tokenization_path_logprob_std": 260.1382162540632, "trajectory_balance_rmse": 4488.3725577906125} +{"canonical_length": 64, "canonical_nll": 0.8578239679336548, "generation_rouge_l_f1": 0.2054054054054054, "generation_token_f1": 0.2918918918918919, "prompt_id": "2060bf62bca6703a", "reading_canonical_log_prob": -54.900733947753906, "reading_distances": [17, 14, 23, 18], "reading_log_probs": [-57.405216217041016, -55.50359344482422, -55.531707763671875, -56.74268341064453], "reading_logprob_std": 0.9384272831042203, "tokenization_path_logprob_std": 55.14570004455029, "trajectory_balance_rmse": 1308.4801421196748} +{"canonical_length": 12, "canonical_nll": 1.0494049390157063, "generation_rouge_l_f1": 0.04477611940298508, "generation_token_f1": 0.04477611940298508, "prompt_id": "8300e5ea4c06c29e", "reading_canonical_log_prob": -12.592859268188477, "reading_distances": [27, 29, 30, 27], "reading_log_probs": [-12.136682510375977, -12.099482536315918, -14.163726806640625, -12.213272094726562], "reading_logprob_std": 1.0080709480620047, "tokenization_path_logprob_std": 16.284736168552357, "trajectory_balance_rmse": 142.88708669412864} +{"canonical_length": 142, "canonical_nll": 1.300172617737676, "generation_rouge_l_f1": 0.2723004694835681, "generation_token_f1": 0.4694835680751173, "prompt_id": "5c25240178b5a183", "reading_canonical_log_prob": -184.62451171875, "reading_distances": [20, 19, 20, 21], "reading_log_probs": [-189.88682556152344, -190.16522216796875, -191.43502807617188, -189.55657958984375], "reading_logprob_std": 0.821326145473349, "tokenization_path_logprob_std": 84.09535350745048, "trajectory_balance_rmse": 2987.5606280251573} +{"canonical_length": 14, "canonical_nll": 1.1363656180245536, "generation_rouge_l_f1": 0.5454545454545454, "generation_token_f1": 0.5454545454545454, "prompt_id": "571e075c79645e81", "reading_canonical_log_prob": -15.90911865234375, "reading_distances": [28, 15, 24, 19], "reading_log_probs": [-16.863937377929688, -15.38131046295166, -16.31487464904785, -17.338472366333008], "reading_logprob_std": 0.8403676277342106, "tokenization_path_logprob_std": 15.283228818556482, "trajectory_balance_rmse": 357.98578881932366} +{"canonical_length": 153, "canonical_nll": 1.4640832040824143, "generation_rouge_l_f1": 0.20833333333333331, "generation_token_f1": 0.48214285714285715, "prompt_id": "fc69046f95700385", "reading_canonical_log_prob": -224.00473022460938, "reading_distances": [39, 41, 39, 38], "reading_log_probs": [-233.97943115234375, -228.4720458984375, -234.04476928710938, -232.86265563964844], "reading_logprob_std": 2.6349073591294574, "tokenization_path_logprob_std": 212.0817326584769, "trajectory_balance_rmse": 3832.000061827535} +{"canonical_length": 56, "canonical_nll": 1.7144794464111328, "generation_rouge_l_f1": 0.08783783783783784, "generation_token_f1": 0.14189189189189189, "prompt_id": "fcbb32fd544b170a", "reading_canonical_log_prob": -96.01084899902344, "reading_distances": [24, 22, 24, 25], "reading_log_probs": [-96.724609375, -95.51791381835938, -97.06680297851562, -95.51378631591797], "reading_logprob_std": 0.8088179919971645, "tokenization_path_logprob_std": 62.37344192850416, "trajectory_balance_rmse": 1208.7507999226796} +{"canonical_length": 83, "canonical_nll": 0.9046916731868881, "generation_rouge_l_f1": 0.2643171806167401, "generation_token_f1": 0.42290748898678415, "prompt_id": "f6d789446e38536d", "reading_canonical_log_prob": -75.08940887451172, "reading_distances": [8, 7, 8, 9], "reading_log_probs": [-73.16914367675781, -72.72250366210938, -73.30656433105469, -72.09463500976562], "reading_logprob_std": 0.5459746028707773, "tokenization_path_logprob_std": 149.43069678219422, "trajectory_balance_rmse": 1818.9981744675736} +{"canonical_length": 270, "canonical_nll": 1.342265037254051, "generation_rouge_l_f1": 0.14545454545454548, "generation_token_f1": 0.2772727272727273, "prompt_id": "86891b80d6ccb140", "reading_canonical_log_prob": -362.41156005859375, "reading_distances": [20, 24, 27, 23], "reading_log_probs": [-361.6643371582031, -364.34332275390625, -361.95465087890625, -363.9759216308594], "reading_logprob_std": 1.370247575469851, "tokenization_path_logprob_std": 124.63540206446271, "trajectory_balance_rmse": 3677.9813545406782} +{"canonical_length": 27, "canonical_nll": 1.4985772592050057, "generation_rouge_l_f1": 0.25974025974025977, "generation_token_f1": 0.38961038961038963, "prompt_id": "01a2b8a844748c41", "reading_canonical_log_prob": -40.461585998535156, "reading_distances": [33, 37, 33, 36], "reading_log_probs": [-39.13014221191406, -38.48267364501953, -39.59670639038086, -39.38654327392578], "reading_logprob_std": 0.4834646644843967, "tokenization_path_logprob_std": 38.0415523782024, "trajectory_balance_rmse": 727.7361068994584} +{"canonical_length": 66, "canonical_nll": 1.161428740530303, "generation_rouge_l_f1": 0.1703470031545741, "generation_token_f1": 0.26498422712933756, "prompt_id": "fc8d39645d6796a9", "reading_canonical_log_prob": -76.654296875, "reading_distances": [28, 32, 30, 26], "reading_log_probs": [-77.10763549804688, -77.84645080566406, -77.40747833251953, -77.212646484375], "reading_logprob_std": 0.32648957236007947, "tokenization_path_logprob_std": 85.92125290925519, "trajectory_balance_rmse": 1636.4888507927215} +{"canonical_length": 340, "canonical_nll": 1.290681547277114, "generation_rouge_l_f1": 0.19769357495881384, "generation_token_f1": 0.3986820428336079, "prompt_id": "7a3c4e3465a24fae", "reading_canonical_log_prob": -438.83172607421875, "reading_distances": [28, 28, 21, 20], "reading_log_probs": [-436.8099060058594, -435.77978515625, -436.8724365234375, -437.32342529296875], "reading_logprob_std": 0.6524865158766983, "tokenization_path_logprob_std": 203.5038277617911, "trajectory_balance_rmse": 7723.903821626305} +{"canonical_length": 22, "canonical_nll": 0.7476166811856356, "generation_rouge_l_f1": 0.6666666666666667, "generation_token_f1": 0.6666666666666667, "prompt_id": "9652191d0c5b1125", "reading_canonical_log_prob": -16.447566986083984, "reading_distances": [20, 17, 15, 23], "reading_log_probs": [-16.70171546936035, -17.190073013305664, -16.422954559326172, -16.47645378112793], "reading_logprob_std": 0.3497056931168362, "tokenization_path_logprob_std": 47.45677614910138, "trajectory_balance_rmse": 451.6119037932925} +{"canonical_length": 229, "canonical_nll": 0.9236271620837883, "generation_rouge_l_f1": 0.39323467230443976, "generation_token_f1": 0.6004228329809724, "prompt_id": "1fb3dd895fb36ca1", "reading_canonical_log_prob": -211.5106201171875, "reading_distances": [46, 47, 47, 53], "reading_log_probs": [-212.48956298828125, -212.03817749023438, -211.3835906982422, -210.39488220214844], "reading_logprob_std": 0.9092606397654123, "tokenization_path_logprob_std": 199.21787617211277, "trajectory_balance_rmse": 5054.027574861342} +{"canonical_length": 84, "canonical_nll": 0.21686463128952754, "generation_rouge_l_f1": 0.43312101910828027, "generation_token_f1": 0.7261146496815287, "prompt_id": "fba6cc8156910498", "reading_canonical_log_prob": -18.216629028320312, "reading_distances": [22, 14, 24, 27], "reading_log_probs": [-18.55182456970215, -18.357318878173828, -19.469703674316406, -20.06879997253418], "reading_logprob_std": 0.8014067180642473, "tokenization_path_logprob_std": 167.53365794080355, "trajectory_balance_rmse": 1529.4977507082838} +{"canonical_length": 24, "canonical_nll": 0.383142630259196, "generation_rouge_l_f1": 0.14285714285714285, "generation_token_f1": 0.15, "prompt_id": "02b01e693f3f83a0", "reading_canonical_log_prob": -9.195423126220703, "reading_distances": [34, 32, 25, 24], "reading_log_probs": [-12.83285140991211, -12.369693756103516, -13.138306617736816, -12.067972183227539], "reading_logprob_std": 0.4761193449813364, "tokenization_path_logprob_std": 55.67411741593165, "trajectory_balance_rmse": 388.9987361381444} +{"canonical_length": 331, "canonical_nll": 0.846163378021148, "generation_rouge_l_f1": 0.26024955436720143, "generation_token_f1": 0.5383244206773619, "prompt_id": "40bc9c44485ded59", "reading_canonical_log_prob": -280.080078125, "reading_distances": [27, 34, 28, 33], "reading_log_probs": [-282.51947021484375, -282.9703674316406, -283.5382995605469, -283.56884765625], "reading_logprob_std": 0.5020103932155284, "tokenization_path_logprob_std": 184.1929704118379, "trajectory_balance_rmse": 4991.499755629705} +{"canonical_length": 11, "canonical_nll": 1.602501088922674, "generation_rouge_l_f1": 0.06060606060606061, "generation_token_f1": 0.06060606060606061, "prompt_id": "f0c9ccf7c1a3a5c6", "reading_canonical_log_prob": -17.627511978149414, "reading_distances": [22, 26, 27, 28], "reading_log_probs": [-16.450040817260742, -16.66683578491211, -16.915313720703125, -16.397567749023438], "reading_logprob_std": 0.2360331555953792, "tokenization_path_logprob_std": 33.195116708409564, "trajectory_balance_rmse": 262.7603091217916} +{"canonical_length": 149, "canonical_nll": 1.8742651203334733, "generation_rouge_l_f1": 0.2113022113022113, "generation_token_f1": 0.43243243243243246, "prompt_id": "7b4b39937f27934d", "reading_canonical_log_prob": -279.2655029296875, "reading_distances": [27, 27, 23, 30], "reading_log_probs": [-278.17681884765625, -277.7474670410156, -279.17578125, -278.8111267089844], "reading_logprob_std": 0.6382947086598526, "tokenization_path_logprob_std": 151.91999541898247, "trajectory_balance_rmse": 3474.43294497452} +{"canonical_length": 10, "canonical_nll": 0.9164741516113282, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "177155c261ac8893", "reading_canonical_log_prob": -9.164741516113281, "reading_distances": [18, 15, 11, 15], "reading_log_probs": [-8.708270072937012, -8.309673309326172, -8.082223892211914, -8.582111358642578], "reading_logprob_std": 0.2802636301635412, "tokenization_path_logprob_std": 29.880386916550947, "trajectory_balance_rmse": 222.9778218365747} +{"canonical_length": 9, "canonical_nll": 0.5257841216193305, "generation_rouge_l_f1": 0.7058823529411765, "generation_token_f1": 0.823529411764706, "prompt_id": "316f7da66f20e71c", "reading_canonical_log_prob": -4.732057094573975, "reading_distances": [31, 31, 30, 26], "reading_log_probs": [-4.649623870849609, -4.572661399841309, -4.6741108894348145, -4.6605305671691895], "reading_logprob_std": 0.04549652615495263, "tokenization_path_logprob_std": 36.230713366427246, "trajectory_balance_rmse": 219.21541667343757} +{"canonical_length": 10, "canonical_nll": 1.4501893997192383, "generation_rouge_l_f1": 0.05947955390334573, "generation_token_f1": 0.06691449814126393, "prompt_id": "4c771fd22dee11bd", "reading_canonical_log_prob": -14.501893997192383, "reading_distances": [25, 31, 29, 30], "reading_log_probs": [-14.3262300491333, -14.314483642578125, -14.485509872436523, -14.381367683410645], "reading_logprob_std": 0.07805833323276744, "tokenization_path_logprob_std": 34.761898888968204, "trajectory_balance_rmse": 307.55614413133463} +{"canonical_length": 150, "canonical_nll": 1.1199886067708333, "generation_rouge_l_f1": 0.20398009950248755, "generation_token_f1": 0.3333333333333333, "prompt_id": "fb7e151d45f5c282", "reading_canonical_log_prob": -167.998291015625, "reading_distances": [37, 33, 38, 39], "reading_log_probs": [-167.36468505859375, -170.35366821289062, -169.56338500976562, -169.46763610839844], "reading_logprob_std": 1.2783281129817483, "tokenization_path_logprob_std": 164.0291979088846, "trajectory_balance_rmse": 3122.9915311055524} +{"canonical_length": 186, "canonical_nll": 1.1845188756142893, "generation_rouge_l_f1": 0.29, "generation_token_f1": 0.5, "prompt_id": "c6ce7ec89de605a5", "reading_canonical_log_prob": -220.3205108642578, "reading_distances": [17, 19, 20, 19], "reading_log_probs": [-220.51959228515625, -222.2515411376953, -221.85438537597656, -220.36468505859375], "reading_logprob_std": 0.9461536727664662, "tokenization_path_logprob_std": 177.03047346840958, "trajectory_balance_rmse": 4295.968671781641} +{"canonical_length": 220, "canonical_nll": 0.9591040871360085, "generation_rouge_l_f1": 0.29166666666666663, "generation_token_f1": 0.5625, "prompt_id": "320fa6e957218c02", "reading_canonical_log_prob": -211.00289916992188, "reading_distances": [14, 13, 14, 13], "reading_log_probs": [-210.8326416015625, -208.67816162109375, -210.24014282226562, -211.70187377929688], "reading_logprob_std": 1.2736980688707515, "tokenization_path_logprob_std": 221.4339656526243, "trajectory_balance_rmse": 4323.2421153415025} +{"canonical_length": 58, "canonical_nll": 1.1452252618197738, "generation_rouge_l_f1": 0.046511627906976744, "generation_token_f1": 0.046511627906976744, "prompt_id": "d3de64d71666f0d5", "reading_canonical_log_prob": -66.42306518554688, "reading_distances": [22, 20, 23, 23], "reading_log_probs": [-75.17107391357422, -72.51451110839844, -77.41305541992188, -73.42868041992188], "reading_logprob_std": 2.1568964613939556, "tokenization_path_logprob_std": 40.40292748978906, "trajectory_balance_rmse": 906.7701669240472} +{"canonical_length": 11, "canonical_nll": 0.17950043895027853, "generation_rouge_l_f1": 0.6, "generation_token_f1": 0.9, "prompt_id": "bd4efb1cb33c7494", "reading_canonical_log_prob": -1.974504828453064, "reading_distances": [44, 47, 47, 43], "reading_log_probs": [-4.734643936157227, -5.204246520996094, -5.171130657196045, -5.434939384460449], "reading_logprob_std": 0.29231430893735033, "tokenization_path_logprob_std": 46.96978099182184, "trajectory_balance_rmse": 203.16152299727784} +{"canonical_length": 38, "canonical_nll": 0.6662417963931435, "generation_rouge_l_f1": 0.37623762376237624, "generation_token_f1": 0.5346534653465347, "prompt_id": "d19ca6fde647262d", "reading_canonical_log_prob": -25.317188262939453, "reading_distances": [17, 15, 15, 17], "reading_log_probs": [-26.797195434570312, -24.117328643798828, -26.20765495300293, -24.890419006347656], "reading_logprob_std": 1.220221416713979, "tokenization_path_logprob_std": 65.81926135671802, "trajectory_balance_rmse": 985.9699057077586} +{"canonical_length": 88, "canonical_nll": 1.2625347484241833, "generation_rouge_l_f1": 0.1774193548387097, "generation_token_f1": 0.3333333333333333, "prompt_id": "b0e33c086e3feda9", "reading_canonical_log_prob": -111.10305786132812, "reading_distances": [14, 17, 19, 20], "reading_log_probs": [-114.105224609375, -119.13320922851562, -116.41365814208984, -115.16644287109375], "reading_logprob_std": 2.1683791620036903, "tokenization_path_logprob_std": 69.79271419725893, "trajectory_balance_rmse": 2020.5014475360506} +{"canonical_length": 53, "canonical_nll": 0.7300766279112618, "generation_rouge_l_f1": 0.6352941176470588, "generation_token_f1": 0.8235294117647058, "prompt_id": "85224f0bc07faa4c", "reading_canonical_log_prob": -38.694061279296875, "reading_distances": [18, 20, 17, 19], "reading_log_probs": [-40.87367630004883, -40.20135498046875, -39.994041442871094, -39.765342712402344], "reading_logprob_std": 0.4778044832288449, "tokenization_path_logprob_std": 96.89179158899003, "trajectory_balance_rmse": 1070.8452901873197} +{"canonical_length": 26, "canonical_nll": 1.2641868591308594, "generation_rouge_l_f1": 0.3137254901960785, "generation_token_f1": 0.47058823529411764, "prompt_id": "7327a63351d4e2c9", "reading_canonical_log_prob": -32.868858337402344, "reading_distances": [40, 29, 31, 36], "reading_log_probs": [-37.24516296386719, -39.128089904785156, -35.1734619140625, -40.19424819946289], "reading_logprob_std": 2.208289945662839, "tokenization_path_logprob_std": 94.16124481446816, "trajectory_balance_rmse": 684.6565997125361} +{"canonical_length": 84, "canonical_nll": 1.2587504613967169, "generation_rouge_l_f1": 0.30088495575221236, "generation_token_f1": 0.4070796460176992, "prompt_id": "7fabfcdc4a3bff58", "reading_canonical_log_prob": -105.73503875732422, "reading_distances": [29, 25, 32, 33], "reading_log_probs": [-107.59197998046875, -106.7469253540039, -107.69296264648438, -108.69198608398438], "reading_logprob_std": 0.7963784631624137, "tokenization_path_logprob_std": 118.07855901149992, "trajectory_balance_rmse": 1901.3778171453607} +{"canonical_length": 13, "canonical_nll": 0.726116840655987, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "173c8d85c343f33f", "reading_canonical_log_prob": -9.439518928527832, "reading_distances": [24, 32, 28, 32], "reading_log_probs": [-11.140069007873535, -11.149518966674805, -10.60527515411377, -10.591703414916992], "reading_logprob_std": 0.31548141792581824, "tokenization_path_logprob_std": 54.15747819375042, "trajectory_balance_rmse": 219.85801376351742} +{"canonical_length": 118, "canonical_nll": 1.5068894725734905, "generation_rouge_l_f1": 0.29110512129380056, "generation_token_f1": 0.43665768194070076, "prompt_id": "e1f635b187b10d1b", "reading_canonical_log_prob": -177.81295776367188, "reading_distances": [29, 26, 23, 20], "reading_log_probs": [-172.7666473388672, -167.3433837890625, -171.01040649414062, -169.97378540039062], "reading_logprob_std": 2.268202274400746, "tokenization_path_logprob_std": 141.13081527063662, "trajectory_balance_rmse": 2506.7377701310484} +{"canonical_length": 242, "canonical_nll": 0.7489560340061661, "generation_rouge_l_f1": 0.33925049309664695, "generation_token_f1": 0.5246548323471399, "prompt_id": "5ff715a6459080e1", "reading_canonical_log_prob": -181.2473602294922, "reading_distances": [20, 30, 28, 26], "reading_log_probs": [-178.15342712402344, -180.9729766845703, -179.52389526367188, -179.42001342773438], "reading_logprob_std": 1.15306138851584, "tokenization_path_logprob_std": 149.00477853913804, "trajectory_balance_rmse": 3141.7090520162965} +{"canonical_length": 147, "canonical_nll": 0.7626999316572332, "generation_rouge_l_f1": 0.2754098360655738, "generation_token_f1": 0.38688524590163936, "prompt_id": "3e82cf9b74d0c14b", "reading_canonical_log_prob": -112.11688995361328, "reading_distances": [24, 26, 25, 23], "reading_log_probs": [-111.18487548828125, -109.1173324584961, -110.49526977539062, -109.5663833618164], "reading_logprob_std": 0.9279455315857412, "tokenization_path_logprob_std": 133.55337196035322, "trajectory_balance_rmse": 1681.0407936194936} +{"canonical_length": 42, "canonical_nll": 1.3281542460123699, "generation_rouge_l_f1": 0.23668639053254437, "generation_token_f1": 0.35502958579881655, "prompt_id": "b1354ebe20eb5cb8", "reading_canonical_log_prob": -55.78247833251953, "reading_distances": [46, 54, 47, 50], "reading_log_probs": [-56.66537094116211, -55.43878173828125, -57.96266555786133, -55.71253204345703], "reading_logprob_std": 1.140300616690174, "tokenization_path_logprob_std": 38.43330554740291, "trajectory_balance_rmse": 941.9543597367665} +{"canonical_length": 124, "canonical_nll": 1.1147836254489036, "generation_rouge_l_f1": 0.35684647302904565, "generation_token_f1": 0.5809128630705395, "prompt_id": "751bafc67843f928", "reading_canonical_log_prob": -138.23316955566406, "reading_distances": [31, 31, 34, 36], "reading_log_probs": [-139.5167999267578, -140.4167022705078, -138.48013305664062, -139.32362365722656], "reading_logprob_std": 0.7946916103738597, "tokenization_path_logprob_std": 72.22075115761788, "trajectory_balance_rmse": 2707.961797056673} +{"canonical_length": 283, "canonical_nll": 1.1077285604847615, "generation_rouge_l_f1": 0.2851985559566787, "generation_token_f1": 0.4693140794223826, "prompt_id": "2c0f185491268c7a", "reading_canonical_log_prob": -313.4871826171875, "reading_distances": [23, 30, 38, 21], "reading_log_probs": [-311.2418518066406, -310.744140625, -312.291748046875, -310.4131164550781], "reading_logprob_std": 0.8200977182390176, "tokenization_path_logprob_std": 139.23979248134285, "trajectory_balance_rmse": 4690.517578728901} +{"canonical_length": 384, "canonical_nll": 1.3905949592590332, "generation_rouge_l_f1": 0.2237061769616027, "generation_token_f1": 0.39398998330550916, "prompt_id": "d93394062daf791b", "reading_canonical_log_prob": -533.9884643554688, "reading_distances": [25, 28, 29, 25], "reading_log_probs": [-540.3312377929688, -532.230712890625, -534.4793090820312, -532.3314208984375], "reading_logprob_std": 3.8028541283869868, "tokenization_path_logprob_std": 148.1027576917947, "trajectory_balance_rmse": 7299.026988902289} +{"canonical_length": 13, "canonical_nll": 4.0437325697678785, "generation_rouge_l_f1": 0.17391304347826086, "generation_token_f1": 0.34782608695652173, "prompt_id": "1bc766d97da83b42", "reading_canonical_log_prob": -52.56852340698242, "reading_distances": [20, 20, 19, 19], "reading_log_probs": [-51.29052734375, -50.47003936767578, -50.71237564086914, -50.68330383300781], "reading_logprob_std": 0.3513348443099311, "tokenization_path_logprob_std": 30.42186128654449, "trajectory_balance_rmse": 323.61219722797966} +{"canonical_length": 5, "canonical_nll": 0.5903768062591552, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "03cbf10fd71003f1", "reading_canonical_log_prob": -2.9518840312957764, "reading_distances": [27, 20, 23, 23], "reading_log_probs": [-3.8631880283355713, -4.564795017242432, -4.275982856750488, -4.080570697784424], "reading_logprob_std": 0.2980462393796381, "tokenization_path_logprob_std": 17.406073392455504, "trajectory_balance_rmse": 102.74834394377757} +{"canonical_length": 218, "canonical_nll": 1.1974586696799743, "generation_rouge_l_f1": 0.235827664399093, "generation_token_f1": 0.380952380952381, "prompt_id": "3b33c4e6381082ec", "reading_canonical_log_prob": -261.0459899902344, "reading_distances": [23, 20, 19, 25], "reading_log_probs": [-260.881591796875, -265.70758056640625, -261.02850341796875, -262.0505676269531], "reading_logprob_std": 2.2544470306301543, "tokenization_path_logprob_std": 161.47094640824648, "trajectory_balance_rmse": 3773.033584560797} +{"canonical_length": 14, "canonical_nll": 0.7027025903974261, "generation_rouge_l_f1": 0.608695652173913, "generation_token_f1": 0.608695652173913, "prompt_id": "275fffbea930b52f", "reading_canonical_log_prob": -9.837836265563965, "reading_distances": [12, 17, 19, 17], "reading_log_probs": [-11.140911102294922, -13.20708179473877, -12.92264175415039, -12.354710578918457], "reading_logprob_std": 0.9150101666819525, "tokenization_path_logprob_std": 16.993615745685734, "trajectory_balance_rmse": 301.048439934596} +{"canonical_length": 181, "canonical_nll": 0.9010421162810773, "generation_rouge_l_f1": 0.2850877192982456, "generation_token_f1": 0.42543859649122806, "prompt_id": "a7f0f685867c31e2", "reading_canonical_log_prob": -163.088623046875, "reading_distances": [81, 74, 72, 75], "reading_log_probs": [-165.81533813476562, -169.92898559570312, -169.40469360351562, -167.13075256347656], "reading_logprob_std": 1.9324354513254711, "tokenization_path_logprob_std": 105.13841782676356, "trajectory_balance_rmse": 3012.965741046867} +{"canonical_length": 128, "canonical_nll": 1.1262218952178955, "generation_rouge_l_f1": 0.3891402714932127, "generation_token_f1": 0.5339366515837104, "prompt_id": "858a5416234d0d2a", "reading_canonical_log_prob": -144.15640258789062, "reading_distances": [19, 21, 21, 19], "reading_log_probs": [-147.364013671875, -146.63702392578125, -145.6570587158203, -147.38714599609375], "reading_logprob_std": 0.814399643286212, "tokenization_path_logprob_std": 106.55836962277041, "trajectory_balance_rmse": 2942.9738745284226} +{"canonical_length": 205, "canonical_nll": 0.7910611036347179, "generation_rouge_l_f1": 0.2936708860759493, "generation_token_f1": 0.5316455696202532, "prompt_id": "17197512e608a44f", "reading_canonical_log_prob": -162.1675262451172, "reading_distances": [16, 14, 12, 13], "reading_log_probs": [-174.32040405273438, -165.09494018554688, -168.2275390625, -180.02252197265625], "reading_logprob_std": 6.6239076565643735, "tokenization_path_logprob_std": 163.28574921336565, "trajectory_balance_rmse": 3928.0843939638066} +{"canonical_length": 160, "canonical_nll": 1.5074356079101563, "generation_rouge_l_f1": 0.22670025188916876, "generation_token_f1": 0.397984886649874, "prompt_id": "7c90d4dc6f1c3493", "reading_canonical_log_prob": -241.189697265625, "reading_distances": [22, 24, 15, 17], "reading_log_probs": [-242.6934814453125, -243.6467742919922, -242.58682250976562, -241.80245971679688], "reading_logprob_std": 0.7557712465128198, "tokenization_path_logprob_std": 107.41947298809198, "trajectory_balance_rmse": 3329.9631134281362} +{"canonical_length": 23, "canonical_nll": 3.2692356938901157, "generation_rouge_l_f1": 0.054474708171206226, "generation_token_f1": 0.07782101167315175, "prompt_id": "d1e684b980513357", "reading_canonical_log_prob": -75.19242095947266, "reading_distances": [9, 19, 23, 13], "reading_log_probs": [-76.18940734863281, -72.58426666259766, -73.75814819335938, -73.89433288574219], "reading_logprob_std": 1.507985545469703, "tokenization_path_logprob_std": 70.94645726507339, "trajectory_balance_rmse": 413.74210400733705} +{"canonical_length": 392, "canonical_nll": 0.9285427015654895, "generation_rouge_l_f1": 0.2950310559006212, "generation_token_f1": 0.4968944099378882, "prompt_id": "5b2aa008c5db43fd", "reading_canonical_log_prob": -363.9887390136719, "reading_distances": [7, 12, 10, 9], "reading_log_probs": [-363.82257080078125, -360.5792541503906, -359.9194030761719, -361.6790771484375], "reading_logprob_std": 1.710015914315704, "tokenization_path_logprob_std": 180.10414054286483, "trajectory_balance_rmse": 7690.298546120328} +{"canonical_length": 82, "canonical_nll": 1.109368301019436, "generation_rouge_l_f1": 0.3120567375886525, "generation_token_f1": 0.3900709219858156, "prompt_id": "180245122ca6704c", "reading_canonical_log_prob": -90.96820068359375, "reading_distances": [29, 24, 35, 31], "reading_log_probs": [-93.08665466308594, -92.2902603149414, -91.63113403320312, -89.90756225585938], "reading_logprob_std": 1.3522145405205952, "tokenization_path_logprob_std": 64.09741262419487, "trajectory_balance_rmse": 1310.5940665281535} +{"canonical_length": 99, "canonical_nll": 1.9367820662681503, "generation_rouge_l_f1": 0.1602209944751381, "generation_token_f1": 0.27624309392265195, "prompt_id": "0280ead696ae0c9e", "reading_canonical_log_prob": -191.74142456054688, "reading_distances": [34, 24, 24, 32], "reading_log_probs": [-190.4311065673828, -191.51028442382812, -190.99610900878906, -191.75845336914062], "reading_logprob_std": 0.5882768086667449, "tokenization_path_logprob_std": 116.36532673928957, "trajectory_balance_rmse": 2697.495421502087} +{"canonical_length": 148, "canonical_nll": 1.4884845630542651, "generation_rouge_l_f1": 0.16393442622950818, "generation_token_f1": 0.26229508196721313, "prompt_id": "1d5af645718b1db0", "reading_canonical_log_prob": -220.29571533203125, "reading_distances": [58, 46, 51, 52], "reading_log_probs": [-219.96060180664062, -215.12986755371094, -220.13870239257812, -217.76011657714844], "reading_logprob_std": 2.342970965392366, "tokenization_path_logprob_std": 107.7251870646667, "trajectory_balance_rmse": 2679.675811988624} +{"canonical_length": 120, "canonical_nll": 1.5371945699055989, "generation_rouge_l_f1": 0.1515151515151515, "generation_token_f1": 0.1515151515151515, "prompt_id": "0211ac7bc9c6651a", "reading_canonical_log_prob": -184.46334838867188, "reading_distances": [17, 21, 20, 24], "reading_log_probs": [-186.93612670898438, -186.89447021484375, -189.1927490234375, -188.68045043945312], "reading_logprob_std": 1.185713761627272, "tokenization_path_logprob_std": 123.3875692753299, "trajectory_balance_rmse": 2641.2241622192414} +{"canonical_length": 115, "canonical_nll": 0.6735907513162364, "generation_rouge_l_f1": 0.16374269005847955, "generation_token_f1": 0.3157894736842105, "prompt_id": "0da3d3229c8dfac7", "reading_canonical_log_prob": -77.46293640136719, "reading_distances": [14, 15, 19, 15], "reading_log_probs": [-77.82320404052734, -77.36042785644531, -77.00749969482422, -78.67137908935547], "reading_logprob_std": 0.7194084578947869, "tokenization_path_logprob_std": 105.82160082739607, "trajectory_balance_rmse": 1602.4983854774307} +{"canonical_length": 196, "canonical_nll": 1.6610288036112884, "generation_rouge_l_f1": 0.23357664233576642, "generation_token_f1": 0.437956204379562, "prompt_id": "dd26708537c260a2", "reading_canonical_log_prob": -325.5616455078125, "reading_distances": [35, 38, 37, 36], "reading_log_probs": [-321.9931640625, -321.64154052734375, -321.095947265625, -321.96392822265625], "reading_logprob_std": 0.4167821773713136, "tokenization_path_logprob_std": 196.53240514054767, "trajectory_balance_rmse": 4851.750539849221} +{"canonical_length": 95, "canonical_nll": 1.7081342195209703, "generation_rouge_l_f1": 0.25842696629213485, "generation_token_f1": 0.4606741573033708, "prompt_id": "9fc2e86bc1fbb199", "reading_canonical_log_prob": -162.2727508544922, "reading_distances": [13, 14, 14, 17], "reading_log_probs": [-161.15386962890625, -162.25282287597656, -162.648681640625, -161.88552856445312], "reading_logprob_std": 0.6358392012455982, "tokenization_path_logprob_std": 93.7354144343545, "trajectory_balance_rmse": 2473.4115527150147} +{"canonical_length": 178, "canonical_nll": 1.1598709620786516, "generation_rouge_l_f1": 0.2993197278911564, "generation_token_f1": 0.5260770975056689, "prompt_id": "0d0deb69605ca531", "reading_canonical_log_prob": -206.45703125, "reading_distances": [17, 19, 18, 14], "reading_log_probs": [-209.84071350097656, -208.01278686523438, -208.80409240722656, -206.55007934570312], "reading_logprob_std": 1.3871561090805653, "tokenization_path_logprob_std": 81.48565474090746, "trajectory_balance_rmse": 4396.380179702067} +{"canonical_length": 20, "canonical_nll": 0.7186723709106445, "generation_rouge_l_f1": 0.0606060606060606, "generation_token_f1": 0.0606060606060606, "prompt_id": "5b98bbd785876818", "reading_canonical_log_prob": -14.37344741821289, "reading_distances": [25, 35, 29, 29], "reading_log_probs": [-15.505332946777344, -14.833136558532715, -15.348066329956055, -15.622241020202637], "reading_logprob_std": 0.34800126346332316, "tokenization_path_logprob_std": 18.600095120061475, "trajectory_balance_rmse": 277.82063916538414} +{"canonical_length": 322, "canonical_nll": 1.0789842309418671, "generation_rouge_l_f1": 0.25177304964539005, "generation_token_f1": 0.4609929078014184, "prompt_id": "9d8ebc1fd91528b6", "reading_canonical_log_prob": -347.43292236328125, "reading_distances": [21, 16, 23, 24], "reading_log_probs": [-346.5060729980469, -342.8590087890625, -347.99609375, -347.618896484375], "reading_logprob_std": 2.3442849983612954, "tokenization_path_logprob_std": 202.33799137414533, "trajectory_balance_rmse": 6001.901928815149} +{"canonical_length": 202, "canonical_nll": 1.0413788143951115, "generation_rouge_l_f1": 0.29090909090909084, "generation_token_f1": 0.5181818181818182, "prompt_id": "5f17b0be639454cf", "reading_canonical_log_prob": -210.3585205078125, "reading_distances": [20, 15, 22, 24], "reading_log_probs": [-209.71224975585938, -210.72811889648438, -211.2782440185547, -207.9333038330078], "reading_logprob_std": 1.4705789809899634, "tokenization_path_logprob_std": 266.74855556781455, "trajectory_balance_rmse": 4601.0581589002295} +{"canonical_length": 11, "canonical_nll": 0.46857491406527435, "generation_rouge_l_f1": 0.9, "generation_token_f1": 0.9, "prompt_id": "1932b722766373ae", "reading_canonical_log_prob": -5.154324054718018, "reading_distances": [39, 43, 40, 40], "reading_log_probs": [-6.4814887046813965, -6.625847816467285, -7.482376575469971, -8.181530952453613], "reading_logprob_std": 0.7934822044162905, "tokenization_path_logprob_std": 19.115421754054584, "trajectory_balance_rmse": 202.03935641986857} +{"canonical_length": 31, "canonical_nll": 1.260591568485383, "generation_rouge_l_f1": 0.2058823529411765, "generation_token_f1": 0.2647058823529412, "prompt_id": "3f8935161866e686", "reading_canonical_log_prob": -39.078338623046875, "reading_distances": [29, 21, 32, 27], "reading_log_probs": [-40.56043243408203, -40.41806411743164, -41.372520446777344, -38.929080963134766], "reading_logprob_std": 1.0181485454957997, "tokenization_path_logprob_std": 59.66868644920327, "trajectory_balance_rmse": 568.745789413112} +{"canonical_length": 45, "canonical_nll": 0.8561247931586371, "generation_rouge_l_f1": 0.4516129032258065, "generation_token_f1": 0.4838709677419355, "prompt_id": "88f54467af0c5001", "reading_canonical_log_prob": -38.52561569213867, "reading_distances": [23, 28, 28, 26], "reading_log_probs": [-43.610862731933594, -49.44493103027344, -43.547672271728516, -44.16679000854492], "reading_logprob_std": 2.8485250768799073, "tokenization_path_logprob_std": 54.255237536771, "trajectory_balance_rmse": 927.6851528385678} +{"canonical_length": 272, "canonical_nll": 1.3345431159524357, "generation_rouge_l_f1": 0.26356589147286824, "generation_token_f1": 0.4999999999999999, "prompt_id": "05940c01ebeedfd9", "reading_canonical_log_prob": -362.9957275390625, "reading_distances": [21, 29, 30, 24], "reading_log_probs": [-363.69964599609375, -364.8519287109375, -366.1372985839844, -368.1602783203125], "reading_logprob_std": 1.911738872249232, "tokenization_path_logprob_std": 159.38804377933354, "trajectory_balance_rmse": 6699.19620991941} +{"canonical_length": 26, "canonical_nll": 1.0913989727313702, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.3428571428571428, "prompt_id": "cc3b11b7124977c4", "reading_canonical_log_prob": -28.376373291015625, "reading_distances": [22, 25, 18, 24], "reading_log_probs": [-28.297788619995117, -28.776142120361328, -28.422651290893555, -27.57420539855957], "reading_logprob_std": 0.5047650871272797, "tokenization_path_logprob_std": 54.66467562334765, "trajectory_balance_rmse": 591.5422825620257} +{"canonical_length": 25, "canonical_nll": 0.9283495330810547, "generation_rouge_l_f1": 0.0909090909090909, "generation_token_f1": 0.0984848484848485, "prompt_id": "9335f10e0202a2ca", "reading_canonical_log_prob": -23.208738327026367, "reading_distances": [21, 24, 28, 25], "reading_log_probs": [-24.722850799560547, -24.466690063476562, -24.083721160888672, -24.286865234375], "reading_logprob_std": 0.27148635553398276, "tokenization_path_logprob_std": 14.581950288760702, "trajectory_balance_rmse": 249.65521970940114} +{"canonical_length": 160, "canonical_nll": 1.2445453643798827, "generation_rouge_l_f1": 0.2620320855614973, "generation_token_f1": 0.4331550802139037, "prompt_id": "ee0b55cb338c65fa", "reading_canonical_log_prob": -199.12725830078125, "reading_distances": [31, 27, 24, 29], "reading_log_probs": [-205.67977905273438, -209.21678161621094, -206.08749389648438, -206.72409057617188], "reading_logprob_std": 1.5858333947893626, "tokenization_path_logprob_std": 204.6292604147756, "trajectory_balance_rmse": 3314.325417634875} +{"canonical_length": 205, "canonical_nll": 0.5099056336937882, "generation_rouge_l_f1": 0.4152542372881356, "generation_token_f1": 0.5805084745762712, "prompt_id": "9ed2cebe3982e4a6", "reading_canonical_log_prob": -104.53065490722656, "reading_distances": [11, 9, 12, 13], "reading_log_probs": [-103.52909851074219, -103.2897720336914, -106.77639770507812, -104.22137451171875], "reading_logprob_std": 1.5977604260788594, "tokenization_path_logprob_std": 235.45232787750342, "trajectory_balance_rmse": 2259.2387734478248} +{"canonical_length": 67, "canonical_nll": 1.460343545942164, "generation_rouge_l_f1": 0.14241486068111456, "generation_token_f1": 0.27244582043343657, "prompt_id": "57c5e7bb82e63751", "reading_canonical_log_prob": -97.843017578125, "reading_distances": [25, 31, 31, 31], "reading_log_probs": [-95.93915557861328, -93.79755401611328, -96.0873031616211, -93.1622543334961], "reading_logprob_std": 1.486664414245012, "tokenization_path_logprob_std": 138.92430750651846, "trajectory_balance_rmse": 1484.4290272403366} +{"canonical_length": 233, "canonical_nll": 1.3447864188656786, "generation_rouge_l_f1": 0.22587268993839837, "generation_token_f1": 0.5626283367556468, "prompt_id": "2049d32643760e75", "reading_canonical_log_prob": -313.3352355957031, "reading_distances": [33, 23, 35, 30], "reading_log_probs": [-310.40899658203125, -310.14959716796875, -310.1063232421875, -309.84979248046875], "reading_logprob_std": 0.22897815559567344, "tokenization_path_logprob_std": 246.06652969077803, "trajectory_balance_rmse": 4670.976003183652} +{"canonical_length": 65, "canonical_nll": 1.2941307654747596, "generation_rouge_l_f1": 0.23236514522821577, "generation_token_f1": 0.34024896265560167, "prompt_id": "9327dc9ed2ef5314", "reading_canonical_log_prob": -84.11849975585938, "reading_distances": [14, 17, 18, 19], "reading_log_probs": [-84.76839447021484, -85.23985290527344, -84.5746841430664, -85.94200134277344], "reading_logprob_std": 0.6084247470081393, "tokenization_path_logprob_std": 87.0488798287729, "trajectory_balance_rmse": 1507.3857309065606} +{"canonical_length": 27, "canonical_nll": 1.1482647083423756, "generation_rouge_l_f1": 0.5652173913043479, "generation_token_f1": 0.608695652173913, "prompt_id": "fb2aa3ee59bdbf8a", "reading_canonical_log_prob": -31.00314712524414, "reading_distances": [103, 96, 95, 98], "reading_log_probs": [-32.095245361328125, -35.16835403442383, -32.66355514526367, -33.42433547973633], "reading_logprob_std": 1.3362839762308443, "tokenization_path_logprob_std": 131.28014732059623, "trajectory_balance_rmse": 707.7648822457029} +{"canonical_length": 176, "canonical_nll": 0.6951377175071023, "generation_rouge_l_f1": 0.4297994269340974, "generation_token_f1": 0.5902578796561605, "prompt_id": "24055c0763c8403a", "reading_canonical_log_prob": -122.34423828125, "reading_distances": [140, 136, 147, 136], "reading_log_probs": [-125.15481567382812, -127.10918426513672, -127.89213562011719, -128.7578125], "reading_logprob_std": 1.5377052044596602, "tokenization_path_logprob_std": 88.51293393090522, "trajectory_balance_rmse": 3357.033804833752} +{"canonical_length": 73, "canonical_nll": 0.482000272567958, "generation_rouge_l_f1": 0.4134615384615385, "generation_token_f1": 0.5, "prompt_id": "c59968b8c3d9c69d", "reading_canonical_log_prob": -35.18601989746094, "reading_distances": [15, 15, 12, 15], "reading_log_probs": [-35.47978591918945, -35.39038848876953, -35.01097869873047, -35.174560546875], "reading_logprob_std": 0.21178352720544533, "tokenization_path_logprob_std": 142.3530361892222, "trajectory_balance_rmse": 1163.228999019998} +{"canonical_length": 283, "canonical_nll": 1.3876510997543066, "generation_rouge_l_f1": 0.30827067669172936, "generation_token_f1": 0.4699248120300752, "prompt_id": "c3213521900f0885", "reading_canonical_log_prob": -392.70526123046875, "reading_distances": [39, 42, 35, 42], "reading_log_probs": [-392.3345031738281, -391.34552001953125, -389.78387451171875, -390.78515625], "reading_logprob_std": 1.0661292724744298, "tokenization_path_logprob_std": 157.98697342186884, "trajectory_balance_rmse": 5187.522570091132} +{"canonical_length": 70, "canonical_nll": 1.4911444527762276, "generation_rouge_l_f1": 0.2616033755274262, "generation_token_f1": 0.35443037974683544, "prompt_id": "48bb22e3d5050293", "reading_canonical_log_prob": -104.38011169433594, "reading_distances": [16, 17, 20, 23], "reading_log_probs": [-105.12055206298828, -102.81201171875, -108.24586486816406, -110.34652709960938], "reading_logprob_std": 3.33061828117772, "tokenization_path_logprob_std": 58.889598021582074, "trajectory_balance_rmse": 1226.4251048158076} +{"canonical_length": 183, "canonical_nll": 1.2878488842906848, "generation_rouge_l_f1": 0.24644549763033172, "generation_token_f1": 0.49289099526066343, "prompt_id": "5a342c18ce379992", "reading_canonical_log_prob": -235.6763458251953, "reading_distances": [30, 32, 26, 26], "reading_log_probs": [-231.5443878173828, -229.9441375732422, -231.56808471679688, -229.439208984375], "reading_logprob_std": 1.0961071037565824, "tokenization_path_logprob_std": 175.1415132031037, "trajectory_balance_rmse": 3958.966507428709} +{"canonical_length": 280, "canonical_nll": 1.5423692975725447, "generation_rouge_l_f1": 0.2441860465116279, "generation_token_f1": 0.46124031007751937, "prompt_id": "66f2043754e3c9e5", "reading_canonical_log_prob": -431.8634033203125, "reading_distances": [21, 15, 28, 21], "reading_log_probs": [-433.1808166503906, -431.24591064453125, -431.46990966796875, -430.3541259765625], "reading_logprob_std": 1.1815129009752268, "tokenization_path_logprob_std": 252.96424824877792, "trajectory_balance_rmse": 5800.755697749773} +{"canonical_length": 61, "canonical_nll": 1.6429178206647028, "generation_rouge_l_f1": 0.37623762376237624, "generation_token_f1": 0.4356435643564357, "prompt_id": "b6ca714dbce654ce", "reading_canonical_log_prob": -100.21798706054688, "reading_distances": [44, 40, 44, 45], "reading_log_probs": [-102.15438079833984, -102.67030334472656, -102.3347396850586, -101.48480987548828], "reading_logprob_std": 0.49894965002405134, "tokenization_path_logprob_std": 115.29381600362183, "trajectory_balance_rmse": 1649.7184603483693} +{"canonical_length": 38, "canonical_nll": 1.2644258800305819, "generation_rouge_l_f1": 0.26262626262626265, "generation_token_f1": 0.4040404040404041, "prompt_id": "4112128a24dc114d", "reading_canonical_log_prob": -48.04818344116211, "reading_distances": [28, 33, 29, 29], "reading_log_probs": [-48.606651306152344, -47.88606262207031, -48.23846435546875, -48.8439826965332], "reading_logprob_std": 0.4202763161765849, "tokenization_path_logprob_std": 143.3491746567815, "trajectory_balance_rmse": 913.7234279357875} +{"canonical_length": 24, "canonical_nll": 0.8634421825408936, "generation_rouge_l_f1": 0.5652173913043478, "generation_token_f1": 0.6086956521739131, "prompt_id": "07948a71ef97c7bb", "reading_canonical_log_prob": -20.722612380981445, "reading_distances": [22, 17, 20, 21], "reading_log_probs": [-22.16758155822754, -21.79458999633789, -22.157577514648438, -21.605558395385742], "reading_logprob_std": 0.2779854334221568, "tokenization_path_logprob_std": 94.10996842870672, "trajectory_balance_rmse": 658.0216960670087} +{"canonical_length": 130, "canonical_nll": 1.5623792208158054, "generation_rouge_l_f1": 0.16957605985037408, "generation_token_f1": 0.32418952618453867, "prompt_id": "d894cb33b4a83ecb", "reading_canonical_log_prob": -203.1092987060547, "reading_distances": [22, 32, 25, 33], "reading_log_probs": [-206.76412963867188, -206.71435546875, -209.05772399902344, -207.2558135986328], "reading_logprob_std": 1.1006149697797274, "tokenization_path_logprob_std": 85.46862927727017, "trajectory_balance_rmse": 3046.3174279617874} +{"canonical_length": 49, "canonical_nll": 0.46263932208625635, "generation_rouge_l_f1": 0.11904761904761903, "generation_token_f1": 0.13095238095238093, "prompt_id": "98f5a9670dc63974", "reading_canonical_log_prob": -22.669326782226562, "reading_distances": [14, 19, 21, 21], "reading_log_probs": [-24.699806213378906, -23.660829544067383, -25.09151840209961, -24.988168716430664], "reading_logprob_std": 0.6541817367169317, "tokenization_path_logprob_std": 68.21150935033809, "trajectory_balance_rmse": 753.2840830997179} +{"canonical_length": 155, "canonical_nll": 1.4077530399445564, "generation_rouge_l_f1": 0.1921182266009852, "generation_token_f1": 0.3399014778325123, "prompt_id": "9866aa5bc569efc5", "reading_canonical_log_prob": -218.20172119140625, "reading_distances": [11, 8, 10, 13], "reading_log_probs": [-217.56439208984375, -217.28009033203125, -217.06431579589844, -217.40086364746094], "reading_logprob_std": 0.21056563531039604, "tokenization_path_logprob_std": 121.06413583104901, "trajectory_balance_rmse": 3481.0357993329903} +{"canonical_length": 173, "canonical_nll": 1.8886313024972905, "generation_rouge_l_f1": 0.1400437636761488, "generation_token_f1": 0.3107221006564551, "prompt_id": "63dbcd016e93c526", "reading_canonical_log_prob": -326.73321533203125, "reading_distances": [23, 18, 27, 28], "reading_log_probs": [-331.3431396484375, -334.913330078125, -328.9732666015625, -331.61651611328125], "reading_logprob_std": 2.442290541336985, "tokenization_path_logprob_std": 161.13712120352676, "trajectory_balance_rmse": 3762.942182153215} +{"canonical_length": 45, "canonical_nll": 0.8517703586154514, "generation_rouge_l_f1": 0.34615384615384615, "generation_token_f1": 0.38461538461538464, "prompt_id": "bb113220673ce1f4", "reading_canonical_log_prob": -38.32966613769531, "reading_distances": [24, 30, 27, 22], "reading_log_probs": [-41.81959533691406, -42.32059097290039, -42.608943939208984, -42.363094329833984], "reading_logprob_std": 0.33101548270520154, "tokenization_path_logprob_std": 33.813926192873616, "trajectory_balance_rmse": 604.8969642140146} +{"canonical_length": 16, "canonical_nll": 1.9084739685058594, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.6428571428571429, "prompt_id": "5d38e8c442eebce3", "reading_canonical_log_prob": -30.53558349609375, "reading_distances": [37, 40, 42, 45], "reading_log_probs": [-30.836994171142578, -31.15134048461914, -31.675844192504883, -31.601299285888672], "reading_logprob_std": 0.3947329558375625, "tokenization_path_logprob_std": 42.89582934748391, "trajectory_balance_rmse": 383.78324639289855} +{"canonical_length": 7, "canonical_nll": 1.028784956250872, "generation_rouge_l_f1": 0.058479532163742694, "generation_token_f1": 0.058479532163742694, "prompt_id": "8ee1f7d9e89653dc", "reading_canonical_log_prob": -7.2014946937561035, "reading_distances": [21, 23, 24, 19], "reading_log_probs": [-6.281256198883057, -6.862571716308594, -6.5832929611206055, -6.599993705749512], "reading_logprob_std": 0.2376920774913387, "tokenization_path_logprob_std": 39.47035313379113, "trajectory_balance_rmse": 119.79204166063818} +{"canonical_length": 220, "canonical_nll": 1.6660758278586647, "generation_rouge_l_f1": 0.23750000000000002, "generation_token_f1": 0.4791666666666667, "prompt_id": "8768fc8d898485e9", "reading_canonical_log_prob": -366.53668212890625, "reading_distances": [16, 15, 12, 16], "reading_log_probs": [-364.92596435546875, -365.928466796875, -365.84979248046875, -365.4540100097656], "reading_logprob_std": 0.45873416572131165, "tokenization_path_logprob_std": 67.19466866315356, "trajectory_balance_rmse": 4820.397480095358} +{"canonical_length": 238, "canonical_nll": 1.1711448861771272, "generation_rouge_l_f1": 0.26046511627906976, "generation_token_f1": 0.5720930232558139, "prompt_id": "f7a53c2bee8bfe30", "reading_canonical_log_prob": -278.73248291015625, "reading_distances": [14, 15, 19, 24], "reading_log_probs": [-277.8546447753906, -279.67437744140625, -278.5523681640625, -281.4734191894531], "reading_logprob_std": 1.5790758976200319, "tokenization_path_logprob_std": 52.33944862019752, "trajectory_balance_rmse": 4521.798855685026} +{"canonical_length": 218, "canonical_nll": 1.1882640593642488, "generation_rouge_l_f1": 0.23554603854389722, "generation_token_f1": 0.4496788008565311, "prompt_id": "90b0a660ae2019c3", "reading_canonical_log_prob": -259.04156494140625, "reading_distances": [31, 29, 40, 26], "reading_log_probs": [-257.0905456542969, -257.0218505859375, -254.66494750976562, -256.84356689453125], "reading_logprob_std": 1.164846323473335, "tokenization_path_logprob_std": 144.15788985650428, "trajectory_balance_rmse": 4012.416808475061} +{"canonical_length": 153, "canonical_nll": 1.076313791711346, "generation_rouge_l_f1": 0.20833333333333331, "generation_token_f1": 0.3055555555555555, "prompt_id": "e11b424a8c471248", "reading_canonical_log_prob": -164.67601013183594, "reading_distances": [38, 51, 52, 57], "reading_log_probs": [-161.06353759765625, -160.64859008789062, -159.53567504882812, -158.60614013671875], "reading_logprob_std": 1.1112892109706787, "tokenization_path_logprob_std": 202.21396129508528, "trajectory_balance_rmse": 2925.6336144068855} +{"canonical_length": 69, "canonical_nll": 1.7608944436778193, "generation_rouge_l_f1": 0.10876132930513593, "generation_token_f1": 0.1933534743202417, "prompt_id": "cc633ec6fcbbe3de", "reading_canonical_log_prob": -121.50171661376953, "reading_distances": [18, 20, 20, 22], "reading_log_probs": [-122.39750671386719, -123.0452880859375, -125.4382095336914, -123.11941528320312], "reading_logprob_std": 1.3321355348825412, "tokenization_path_logprob_std": 125.11650619563504, "trajectory_balance_rmse": 1314.3967237473312} +{"canonical_length": 51, "canonical_nll": 1.5520849788890165, "generation_rouge_l_f1": 0.21212121212121213, "generation_token_f1": 0.29292929292929293, "prompt_id": "66e12e5aa4182ef3", "reading_canonical_log_prob": -79.15633392333984, "reading_distances": [33, 39, 39, 49], "reading_log_probs": [-78.83528900146484, -76.52114868164062, -79.39509582519531, -77.65420532226562], "reading_logprob_std": 1.2792470778744371, "tokenization_path_logprob_std": 85.40494215629147, "trajectory_balance_rmse": 956.2936022129518} +{"canonical_length": 147, "canonical_nll": 1.3929864792596727, "generation_rouge_l_f1": 0.14953271028037382, "generation_token_f1": 0.3130841121495327, "prompt_id": "831cd880037811e2", "reading_canonical_log_prob": -204.76901245117188, "reading_distances": [19, 19, 28, 19], "reading_log_probs": [-202.9534912109375, -203.82861328125, -202.86331176757812, -202.61349487304688], "reading_logprob_std": 0.5291736261045948, "tokenization_path_logprob_std": 118.39693579070145, "trajectory_balance_rmse": 3713.3675159617515} +{"canonical_length": 239, "canonical_nll": 1.1091124722149581, "generation_rouge_l_f1": 0.2535211267605634, "generation_token_f1": 0.5070422535211268, "prompt_id": "61ee3d334ec189e4", "reading_canonical_log_prob": -265.077880859375, "reading_distances": [22, 26, 31, 29], "reading_log_probs": [-262.1043395996094, -263.152587890625, -264.2024841308594, -262.0018310546875], "reading_logprob_std": 1.0320263371878873, "tokenization_path_logprob_std": 257.5699131310434, "trajectory_balance_rmse": 5649.944540005572} +{"canonical_length": 239, "canonical_nll": 1.4500469383335512, "generation_rouge_l_f1": 0.3002832861189802, "generation_token_f1": 0.47592067988668546, "prompt_id": "978f2098f6623381", "reading_canonical_log_prob": -346.56121826171875, "reading_distances": [16, 19, 23, 20], "reading_log_probs": [-348.96624755859375, -348.02691650390625, -348.5379638671875, -346.7447814941406], "reading_logprob_std": 0.9626882430608568, "tokenization_path_logprob_std": 383.43713868799574, "trajectory_balance_rmse": 4968.016817543844} +{"canonical_length": 366, "canonical_nll": 1.9123119917072233, "generation_rouge_l_f1": 0.24543946932006633, "generation_token_f1": 0.46102819237147596, "prompt_id": "f4a2dbf6cf78f4f4", "reading_canonical_log_prob": -699.9061889648438, "reading_distances": [41, 37, 37, 36], "reading_log_probs": [-699.2919921875, -697.0692138671875, -699.6319580078125, -698.1536254882812], "reading_logprob_std": 1.1648058527317173, "tokenization_path_logprob_std": 137.68264017577485, "trajectory_balance_rmse": 7685.513153021726} +{"canonical_length": 323, "canonical_nll": 1.6293533372436146, "generation_rouge_l_f1": 0.19202898550724637, "generation_token_f1": 0.40579710144927533, "prompt_id": "607d603874554f5a", "reading_canonical_log_prob": -526.2811279296875, "reading_distances": [21, 20, 22, 23], "reading_log_probs": [-530.5587158203125, -528.3143310546875, -526.591064453125, -524.8148193359375], "reading_logprob_std": 2.4519234615737853, "tokenization_path_logprob_std": 250.43828385115145, "trajectory_balance_rmse": 6279.655867265133} +{"canonical_length": 16, "canonical_nll": 1.0717720985412598, "generation_rouge_l_f1": 0.5000000000000001, "generation_token_f1": 0.6111111111111112, "prompt_id": "65ee5494d4aae49e", "reading_canonical_log_prob": -17.148353576660156, "reading_distances": [15, 17, 22, 18], "reading_log_probs": [-18.868152618408203, -18.8985652923584, -19.77974510192871, -19.307950973510742], "reading_logprob_std": 0.42739643699140717, "tokenization_path_logprob_std": 54.189833639546556, "trajectory_balance_rmse": 362.43832433488996} +{"canonical_length": 120, "canonical_nll": 1.7785818735758463, "generation_rouge_l_f1": 0.2346938775510204, "generation_token_f1": 0.3877551020408163, "prompt_id": "bc654a75738c95db", "reading_canonical_log_prob": -213.42982482910156, "reading_distances": [27, 29, 26, 27], "reading_log_probs": [-206.45611572265625, -207.7424774169922, -208.04205322265625, -205.36752319335938], "reading_logprob_std": 1.2328227083507517, "tokenization_path_logprob_std": 157.56709404201584, "trajectory_balance_rmse": 2420.3589359239077} +{"canonical_length": 166, "canonical_nll": 0.992441016507436, "generation_rouge_l_f1": 0.26004728132387706, "generation_token_f1": 0.4728132387706856, "prompt_id": "1277e399b9abf4ff", "reading_canonical_log_prob": -164.74520874023438, "reading_distances": [28, 32, 24, 24], "reading_log_probs": [-166.32972717285156, -164.51031494140625, -164.5582733154297, -165.1568603515625], "reading_logprob_std": 0.8466873503172877, "tokenization_path_logprob_std": 133.39937707766524, "trajectory_balance_rmse": 4155.970106089527} +{"canonical_length": 13, "canonical_nll": 1.8470563154954176, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.2857142857142857, "prompt_id": "888761fd7cf1d9af", "reading_canonical_log_prob": -24.01173210144043, "reading_distances": [29, 23, 26, 29], "reading_log_probs": [-23.569778442382812, -24.41339111328125, -23.89109230041504, -23.53246307373047], "reading_logprob_std": 0.4076111866331912, "tokenization_path_logprob_std": 67.69084683182447, "trajectory_balance_rmse": 372.4436299833676} +{"canonical_length": 196, "canonical_nll": 1.281341552734375, "generation_rouge_l_f1": 0.22362869198312235, "generation_token_f1": 0.4767932489451477, "prompt_id": "4049144029e8d577", "reading_canonical_log_prob": -251.1429443359375, "reading_distances": [20, 24, 26, 23], "reading_log_probs": [-247.70199584960938, -251.1967315673828, -251.33065795898438, -250.036865234375], "reading_logprob_std": 1.6800076212147474, "tokenization_path_logprob_std": 160.4145316024196, "trajectory_balance_rmse": 5167.194613548282} +{"canonical_length": 105, "canonical_nll": 1.0017947242373513, "generation_rouge_l_f1": 0.2242152466367713, "generation_token_f1": 0.32286995515695066, "prompt_id": "35fc3a98ce123749", "reading_canonical_log_prob": -105.18844604492188, "reading_distances": [37, 34, 41, 33], "reading_log_probs": [-107.93128204345703, -105.93260192871094, -106.17591094970703, -106.54359436035156], "reading_logprob_std": 0.8930019179421617, "tokenization_path_logprob_std": 205.14293199416977, "trajectory_balance_rmse": 2180.4043580317966} +{"canonical_length": 17, "canonical_nll": 3.3269377315745636, "generation_rouge_l_f1": 0.16666666666666669, "generation_token_f1": 0.16666666666666669, "prompt_id": "30594b15ee7b3362", "reading_canonical_log_prob": -56.55794143676758, "reading_distances": [20, 22, 21, 20], "reading_log_probs": [-52.915748596191406, -52.39881134033203, -53.88936996459961, -53.69218063354492], "reading_logprob_std": 0.6923066623873939, "tokenization_path_logprob_std": 87.46460034953262, "trajectory_balance_rmse": 378.38974205424154} +{"canonical_length": 89, "canonical_nll": 1.0906048035353757, "generation_rouge_l_f1": 0.25205479452054796, "generation_token_f1": 0.3013698630136986, "prompt_id": "55622d0a348ebb7e", "reading_canonical_log_prob": -97.06382751464844, "reading_distances": [18, 15, 16, 18], "reading_log_probs": [-93.97134399414062, -93.41341400146484, -94.35063934326172, -93.71080017089844], "reading_logprob_std": 0.3978335041183403, "tokenization_path_logprob_std": 133.65860970398282, "trajectory_balance_rmse": 1725.2852256790525} +{"canonical_length": 181, "canonical_nll": 1.2343619330811895, "generation_rouge_l_f1": 0.2757009345794392, "generation_token_f1": 0.425233644859813, "prompt_id": "96bafc4191db5d29", "reading_canonical_log_prob": -223.4195098876953, "reading_distances": [35, 26, 23, 33], "reading_log_probs": [-229.41334533691406, -224.30007934570312, -226.53318786621094, -229.61285400390625], "reading_logprob_std": 2.5360285818250596, "tokenization_path_logprob_std": 99.47876279312888, "trajectory_balance_rmse": 3743.4921587084696} +{"canonical_length": 186, "canonical_nll": 1.2339743337323588, "generation_rouge_l_f1": 0.21836228287841192, "generation_token_f1": 0.44168734491315137, "prompt_id": "a69ab9afafd56708", "reading_canonical_log_prob": -229.51922607421875, "reading_distances": [26, 30, 32, 31], "reading_log_probs": [-229.82733154296875, -229.8692169189453, -231.51052856445312, -230.88031005859375], "reading_logprob_std": 0.8194031858652633, "tokenization_path_logprob_std": 123.78257397464704, "trajectory_balance_rmse": 3636.3075261063314} +{"canonical_length": 7, "canonical_nll": 1.393878391810826, "generation_rouge_l_f1": 0.06153846153846154, "generation_token_f1": 0.06153846153846154, "prompt_id": "7d277a000cf626ea", "reading_canonical_log_prob": -9.757148742675781, "reading_distances": [35, 27, 28, 28], "reading_log_probs": [-10.748661041259766, -10.73945426940918, -11.274781227111816, -10.559833526611328], "reading_logprob_std": 0.30856277349384403, "tokenization_path_logprob_std": 50.258171153341976, "trajectory_balance_rmse": 187.65306409149377} +{"canonical_length": 8, "canonical_nll": 1.204237461090088, "generation_rouge_l_f1": 0.19047619047619047, "generation_token_f1": 0.19047619047619047, "prompt_id": "4b1173c764ce0a81", "reading_canonical_log_prob": -9.633899688720703, "reading_distances": [45, 57, 50, 50], "reading_log_probs": [-10.083990097045898, -9.56413459777832, -9.845085144042969, -10.153255462646484], "reading_logprob_std": 0.2666284174179283, "tokenization_path_logprob_std": 32.29685364885077, "trajectory_balance_rmse": 140.77623120797986} +{"canonical_length": 48, "canonical_nll": 0.972632884979248, "generation_rouge_l_f1": 0.18775510204081633, "generation_token_f1": 0.22040816326530613, "prompt_id": "c86574795b337627", "reading_canonical_log_prob": -46.686378479003906, "reading_distances": [8, 10, 9, 8], "reading_log_probs": [-48.36790084838867, -47.117835998535156, -47.607215881347656, -46.042686462402344], "reading_logprob_std": 0.9742996282036307, "tokenization_path_logprob_std": 7.091917108800722, "trajectory_balance_rmse": 524.2914764388132} +{"canonical_length": 251, "canonical_nll": 1.1452985664762825, "generation_rouge_l_f1": 0.10855949895615867, "generation_token_f1": 0.18371607515657618, "prompt_id": "f0455cfad098aced", "reading_canonical_log_prob": -287.4699401855469, "reading_distances": [40, 40, 42, 42], "reading_log_probs": [-285.90814208984375, -284.80792236328125, -284.6827697753906, -285.4915771484375], "reading_logprob_std": 0.57899075326324, "tokenization_path_logprob_std": 190.91528313428387, "trajectory_balance_rmse": 5341.041067846469} +{"canonical_length": 92, "canonical_nll": 1.1702835248864216, "generation_rouge_l_f1": 0.16883116883116883, "generation_token_f1": 0.2727272727272727, "prompt_id": "765c339341bf24b5", "reading_canonical_log_prob": -107.66608428955078, "reading_distances": [73, 64, 58, 62], "reading_log_probs": [-109.06132507324219, -110.2220458984375, -109.39440155029297, -111.37193298339844], "reading_logprob_std": 1.0293601464088828, "tokenization_path_logprob_std": 129.822601100932, "trajectory_balance_rmse": 1366.8496365811231} +{"canonical_length": 25, "canonical_nll": 1.9993939208984375, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.43333333333333335, "prompt_id": "ef02d7967a69b355", "reading_canonical_log_prob": -49.98484802246094, "reading_distances": [34, 29, 29, 42], "reading_log_probs": [-52.200782775878906, -52.40920639038086, -52.4090690612793, -51.93800354003906], "reading_logprob_std": 0.2235716471453416, "tokenization_path_logprob_std": 25.298391795878587, "trajectory_balance_rmse": 495.0665955190788} +{"canonical_length": 26, "canonical_nll": 1.284024898822491, "generation_rouge_l_f1": 0.16666666666666666, "generation_token_f1": 0.2777777777777778, "prompt_id": "27392a7bee3d725d", "reading_canonical_log_prob": -33.384647369384766, "reading_distances": [30, 32, 35, 30], "reading_log_probs": [-34.07268142700195, -33.855079650878906, -32.989166259765625, -33.11641311645508], "reading_logprob_std": 0.5359905739907811, "tokenization_path_logprob_std": 33.24013002532303, "trajectory_balance_rmse": 602.7309858009735} +{"canonical_length": 53, "canonical_nll": 1.4624603559386056, "generation_rouge_l_f1": 0.18571428571428572, "generation_token_f1": 0.22857142857142856, "prompt_id": "deecb5d2b0490e50", "reading_canonical_log_prob": -77.5103988647461, "reading_distances": [15, 23, 21, 17], "reading_log_probs": [-81.91160583496094, -84.61601257324219, -84.64472961425781, -82.9953384399414], "reading_logprob_std": 1.3324838435175803, "tokenization_path_logprob_std": 104.23571109558725, "trajectory_balance_rmse": 1210.8606435790073} +{"canonical_length": 124, "canonical_nll": 1.171598619030368, "generation_rouge_l_f1": 0.2099737532808399, "generation_token_f1": 0.3779527559055118, "prompt_id": "8c1c19dea7da1533", "reading_canonical_log_prob": -145.27822875976562, "reading_distances": [26, 29, 26, 20], "reading_log_probs": [-142.1409912109375, -147.8467254638672, -144.27093505859375, -141.34515380859375], "reading_logprob_std": 2.9060615998309633, "tokenization_path_logprob_std": 205.38104033515464, "trajectory_balance_rmse": 2152.805875125991} +{"canonical_length": 81, "canonical_nll": 1.2923838297526042, "generation_rouge_l_f1": 0.2641509433962264, "generation_token_f1": 0.49056603773584906, "prompt_id": "8ddcb00e61289cea", "reading_canonical_log_prob": -104.68309020996094, "reading_distances": [17, 20, 20, 22], "reading_log_probs": [-103.40791320800781, -103.22183990478516, -103.75424194335938, -104.25971984863281], "reading_logprob_std": 0.45609721563811295, "tokenization_path_logprob_std": 79.60655442398098, "trajectory_balance_rmse": 1809.9560169613746} +{"canonical_length": 78, "canonical_nll": 0.6304295857747396, "generation_rouge_l_f1": 0.34267912772585674, "generation_token_f1": 0.39252336448598135, "prompt_id": "e509fb17129966c3", "reading_canonical_log_prob": -49.17350769042969, "reading_distances": [48, 40, 43, 52], "reading_log_probs": [-54.03274154663086, -51.1756591796875, -55.743629455566406, -54.47206115722656], "reading_logprob_std": 1.928569486797447, "tokenization_path_logprob_std": 34.45099193436905, "trajectory_balance_rmse": 1218.2200745567127} +{"canonical_length": 170, "canonical_nll": 0.9995795754825367, "generation_rouge_l_f1": 0.4112676056338028, "generation_token_f1": 0.5915492957746478, "prompt_id": "07a3f091145dd3c4", "reading_canonical_log_prob": -169.92852783203125, "reading_distances": [27, 25, 30, 29], "reading_log_probs": [-167.92144775390625, -167.70516967773438, -167.49742126464844, -169.29356384277344], "reading_logprob_std": 0.8114575784135036, "tokenization_path_logprob_std": 106.03586856314463, "trajectory_balance_rmse": 4140.934664179913} +{"canonical_length": 159, "canonical_nll": 1.47109639869546, "generation_rouge_l_f1": 0.2572347266881029, "generation_token_f1": 0.5337620578778135, "prompt_id": "1cfb5bc2275fa797", "reading_canonical_log_prob": -233.90432739257812, "reading_distances": [24, 27, 18, 22], "reading_log_probs": [-230.48468017578125, -231.46624755859375, -232.38011169433594, -232.0481719970703], "reading_logprob_std": 0.8308935710568157, "tokenization_path_logprob_std": 53.15953697695834, "trajectory_balance_rmse": 3248.91515642168} +{"canonical_length": 11, "canonical_nll": 0.04573030363429676, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "c9f832251ee1421d", "reading_canonical_log_prob": -0.5030333399772644, "reading_distances": [20, 22, 20, 18], "reading_log_probs": [-0.9955158829689026, -1.129530906677246, -1.1233421564102173, -1.2001837491989136], "reading_logprob_std": 0.08520723433544748, "tokenization_path_logprob_std": 35.10601399972329, "trajectory_balance_rmse": 137.5796882553705} +{"canonical_length": 174, "canonical_nll": 0.9764292048311781, "generation_rouge_l_f1": 0.2795180722891566, "generation_token_f1": 0.5060240963855421, "prompt_id": "13d2c1dd5e0685fa", "reading_canonical_log_prob": -169.898681640625, "reading_distances": [54, 61, 58, 50], "reading_log_probs": [-168.2510528564453, -169.30303955078125, -166.59597778320312, -166.28781127929688], "reading_logprob_std": 1.420534480695659, "tokenization_path_logprob_std": 99.94108500488352, "trajectory_balance_rmse": 2054.865538726544} +{"canonical_length": 266, "canonical_nll": 1.1695857227296758, "generation_rouge_l_f1": 0.2418426103646833, "generation_token_f1": 0.44529750479846447, "prompt_id": "e2b16f584507a85d", "reading_canonical_log_prob": -311.10980224609375, "reading_distances": [32, 28, 27, 23], "reading_log_probs": [-312.0440673828125, -308.15740966796875, -310.77197265625, -308.2151794433594], "reading_logprob_std": 1.9313463468677876, "tokenization_path_logprob_std": 472.0256927284956, "trajectory_balance_rmse": 6429.356004796373} +{"canonical_length": 279, "canonical_nll": 1.2324332507280467, "generation_rouge_l_f1": 0.3412844036697248, "generation_token_f1": 0.5724770642201835, "prompt_id": "76e96744ab3b125b", "reading_canonical_log_prob": -343.848876953125, "reading_distances": [18, 25, 16, 24], "reading_log_probs": [-342.3276062011719, -344.35211181640625, -343.0646057128906, -344.281982421875], "reading_logprob_std": 0.9834452123607749, "tokenization_path_logprob_std": 240.45922095930024, "trajectory_balance_rmse": 6927.887978633905} +{"canonical_length": 157, "canonical_nll": 1.4583170702502986, "generation_rouge_l_f1": 0.31428571428571433, "generation_token_f1": 0.480952380952381, "prompt_id": "2fad87f3c29a8aa9", "reading_canonical_log_prob": -228.95578002929688, "reading_distances": [27, 26, 28, 22], "reading_log_probs": [-225.5895538330078, -226.46011352539062, -226.09844970703125, -225.68797302246094], "reading_logprob_std": 0.4002112262921268, "tokenization_path_logprob_std": 63.866391131834376, "trajectory_balance_rmse": 3658.8007884534036} +{"canonical_length": 135, "canonical_nll": 2.0388658311631946, "generation_rouge_l_f1": 0.13775510204081634, "generation_token_f1": 0.23469387755102042, "prompt_id": "057edeee1c342070", "reading_canonical_log_prob": -275.24688720703125, "reading_distances": [19, 19, 23, 21], "reading_log_probs": [-272.70770263671875, -273.63458251953125, -273.24993896484375, -273.2934875488281], "reading_logprob_std": 0.38323876786374206, "tokenization_path_logprob_std": 62.972082661791184, "trajectory_balance_rmse": 3374.6668770242504} +{"canonical_length": 273, "canonical_nll": 0.9385381565862523, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.43609022556390975, "prompt_id": "b42b291887a21049", "reading_canonical_log_prob": -256.2209167480469, "reading_distances": [31, 28, 27, 29], "reading_log_probs": [-257.0408935546875, -255.37118530273438, -258.4559326171875, -259.5789489746094], "reading_logprob_std": 1.8192052605705302, "tokenization_path_logprob_std": 103.76977579959289, "trajectory_balance_rmse": 5997.853148775095} +{"canonical_length": 8, "canonical_nll": 0.4131544828414917, "generation_rouge_l_f1": 0.923076923076923, "generation_token_f1": 0.923076923076923, "prompt_id": "751d7c04030e09ea", "reading_canonical_log_prob": -3.3052358627319336, "reading_distances": [22, 26, 26, 24], "reading_log_probs": [-2.1737544536590576, -1.9401462078094482, -1.8975865840911865, -2.6487648487091064], "reading_logprob_std": 0.3445645639973371, "tokenization_path_logprob_std": 70.47620468609401, "trajectory_balance_rmse": 212.1540128574505} +{"canonical_length": 45, "canonical_nll": 1.151190185546875, "generation_rouge_l_f1": 0.5744680851063829, "generation_token_f1": 0.7446808510638299, "prompt_id": "bdda8d0e92726300", "reading_canonical_log_prob": -51.803558349609375, "reading_distances": [18, 23, 23, 24], "reading_log_probs": [-51.337791442871094, -51.866180419921875, -50.74052047729492, -51.526248931884766], "reading_logprob_std": 0.47182947748635706, "tokenization_path_logprob_std": 58.789658179945846, "trajectory_balance_rmse": 1066.6475144702842} +{"canonical_length": 225, "canonical_nll": 0.7678415256076389, "generation_rouge_l_f1": 0.3483606557377049, "generation_token_f1": 0.5532786885245902, "prompt_id": "ce2f11428dfe632d", "reading_canonical_log_prob": -172.76434326171875, "reading_distances": [35, 33, 26, 26], "reading_log_probs": [-171.04608154296875, -172.07766723632812, -169.85267639160156, -171.85699462890625], "reading_logprob_std": 1.006749604178952, "tokenization_path_logprob_std": 49.97076936843772, "trajectory_balance_rmse": 3679.8810983815365} +{"canonical_length": 371, "canonical_nll": 1.0251517488628705, "generation_rouge_l_f1": 0.26362038664323373, "generation_token_f1": 0.460456942003515, "prompt_id": "3ce75f08c1522ce9", "reading_canonical_log_prob": -380.331298828125, "reading_distances": [20, 17, 24, 20], "reading_log_probs": [-386.6888427734375, -392.283203125, -385.59478759765625, -389.372314453125], "reading_logprob_std": 2.988488037407376, "tokenization_path_logprob_std": 319.2144302054405, "trajectory_balance_rmse": 6420.455609978281} +{"canonical_length": 3, "canonical_nll": 2.650111675262451, "generation_rouge_l_f1": 0.015151515151515152, "generation_token_f1": 0.015151515151515152, "prompt_id": "cc1031d1c53d6dff", "reading_canonical_log_prob": -7.9503350257873535, "reading_distances": [40, 29, 40, 40], "reading_log_probs": [-9.101804733276367, -8.406535148620605, -9.825488090515137, -9.209968566894531], "reading_logprob_std": 0.5814216631573105, "tokenization_path_logprob_std": 13.63965029454424, "trajectory_balance_rmse": 34.241701627534674} +{"canonical_length": 14, "canonical_nll": 3.454601560320173, "generation_rouge_l_f1": 0.09523809523809525, "generation_token_f1": 0.09523809523809525, "prompt_id": "6b9afce682e6088d", "reading_canonical_log_prob": -48.36442184448242, "reading_distances": [55, 50, 45, 53], "reading_log_probs": [-44.644168853759766, -44.860137939453125, -45.109107971191406, -45.205204010009766], "reading_logprob_std": 0.25295958074225044, "tokenization_path_logprob_std": 87.61572143562496, "trajectory_balance_rmse": 386.18744258097854} +{"canonical_length": 15, "canonical_nll": 0.5377274831136067, "generation_rouge_l_f1": 0.10408921933085502, "generation_token_f1": 0.11152416356877325, "prompt_id": "16fc84421f512267", "reading_canonical_log_prob": -8.065912246704102, "reading_distances": [24, 30, 36, 24], "reading_log_probs": [-9.255023002624512, -8.251853942871094, -8.754288673400879, -9.033886909484863], "reading_logprob_std": 0.4328371128894036, "tokenization_path_logprob_std": 13.143503522892615, "trajectory_balance_rmse": 308.4708038593291} +{"canonical_length": 12, "canonical_nll": 0.2989414532979329, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "7550a792cc616b59", "reading_canonical_log_prob": -3.5872974395751953, "reading_distances": [24, 37, 26, 31], "reading_log_probs": [-4.772709846496582, -5.8069257736206055, -5.344020366668701, -5.238140106201172], "reading_logprob_std": 0.4244243947030666, "tokenization_path_logprob_std": 30.418654506647542, "trajectory_balance_rmse": 195.04776948305656} +{"canonical_length": 242, "canonical_nll": 1.4964868214504778, "generation_rouge_l_f1": 0.23597678916827855, "generation_token_f1": 0.42166344294003877, "prompt_id": "d8ba10df8798bb3c", "reading_canonical_log_prob": -362.1498107910156, "reading_distances": [33, 38, 35, 38], "reading_log_probs": [-364.6158142089844, -365.050048828125, -365.23785400390625, -364.5579833984375], "reading_logprob_std": 0.3314690564868481, "tokenization_path_logprob_std": 463.1877934056624, "trajectory_balance_rmse": 5632.930583623793} +{"canonical_length": 15, "canonical_nll": 0.9774740854899089, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.3157894736842105, "prompt_id": "c62abdde12be0560", "reading_canonical_log_prob": -14.662111282348633, "reading_distances": [25, 28, 29, 29], "reading_log_probs": [-15.309329986572266, -15.73598575592041, -15.413885116577148, -15.304594039916992], "reading_logprob_std": 0.2030561342458378, "tokenization_path_logprob_std": 61.53316810326393, "trajectory_balance_rmse": 291.8487753993858} +{"canonical_length": 318, "canonical_nll": 1.49570047630454, "generation_rouge_l_f1": 0.24518388791593695, "generation_token_f1": 0.47635726795096317, "prompt_id": "e65b179e41f3017d", "reading_canonical_log_prob": -475.63275146484375, "reading_distances": [21, 27, 27, 30], "reading_log_probs": [-475.3590087890625, -474.432373046875, -474.62255859375, -476.3215637207031], "reading_logprob_std": 0.8572900642447494, "tokenization_path_logprob_std": 302.55160743718415, "trajectory_balance_rmse": 7084.473264593801} +{"canonical_length": 83, "canonical_nll": 1.2463549878223832, "generation_rouge_l_f1": 0.1781609195402299, "generation_token_f1": 0.3563218390804598, "prompt_id": "688f5dc10c3a0a47", "reading_canonical_log_prob": -103.44746398925781, "reading_distances": [33, 34, 24, 33], "reading_log_probs": [-109.78915405273438, -110.17285919189453, -109.98262023925781, -109.28584289550781], "reading_logprob_std": 0.3814957793621718, "tokenization_path_logprob_std": 22.164996644449936, "trajectory_balance_rmse": 1829.6252515163917} +{"canonical_length": 324, "canonical_nll": 1.5411573810341916, "generation_rouge_l_f1": 0.303886925795053, "generation_token_f1": 0.5123674911660778, "prompt_id": "d7836a3fff052985", "reading_canonical_log_prob": -499.3349914550781, "reading_distances": [22, 21, 22, 22], "reading_log_probs": [-501.2102966308594, -499.64630126953125, -501.85009765625, -501.1036376953125], "reading_logprob_std": 0.9311531120151096, "tokenization_path_logprob_std": 331.7276821764876, "trajectory_balance_rmse": 6355.503111312408} +{"canonical_length": 18, "canonical_nll": 1.2161165873209636, "generation_rouge_l_f1": 0.4210526315789473, "generation_token_f1": 0.4210526315789473, "prompt_id": "97d1bc52cae659b0", "reading_canonical_log_prob": -21.890098571777344, "reading_distances": [9, 12, 13, 10], "reading_log_probs": [-23.967153549194336, -23.035991668701172, -23.43821907043457, -22.82970428466797], "reading_logprob_std": 0.5012697686782571, "tokenization_path_logprob_std": 26.45616351139347, "trajectory_balance_rmse": 224.99725951490876} +{"canonical_length": 311, "canonical_nll": 0.959753987106863, "generation_rouge_l_f1": 0.31578947368421056, "generation_token_f1": 0.543859649122807, "prompt_id": "d655055a06f3b252", "reading_canonical_log_prob": -298.4834899902344, "reading_distances": [30, 34, 35, 31], "reading_log_probs": [-300.25714111328125, -297.83123779296875, -299.61529541015625, -299.750732421875], "reading_logprob_std": 1.0582645772223431, "tokenization_path_logprob_std": 349.26352181162116, "trajectory_balance_rmse": 6049.630193716903} +{"canonical_length": 270, "canonical_nll": 1.2515437373408564, "generation_rouge_l_f1": 0.33641404805914976, "generation_token_f1": 0.5064695009242144, "prompt_id": "62d551bf467199ab", "reading_canonical_log_prob": -337.91680908203125, "reading_distances": [18, 19, 24, 20], "reading_log_probs": [-339.18634033203125, -337.7006530761719, -338.61431884765625, -334.99365234375], "reading_logprob_std": 1.857080410386844, "tokenization_path_logprob_std": 306.7500023028346, "trajectory_balance_rmse": 5335.20121610649} +{"canonical_length": 11, "canonical_nll": 0.45763206481933594, "generation_rouge_l_f1": 0.6451612903225806, "generation_token_f1": 0.6451612903225806, "prompt_id": "1237b2c3cf7851d2", "reading_canonical_log_prob": -5.033952713012695, "reading_distances": [20, 20, 16, 18], "reading_log_probs": [-4.252781867980957, -4.9229302406311035, -5.098301410675049, -5.394358158111572], "reading_logprob_std": 0.48372232832112233, "tokenization_path_logprob_std": 63.89081657809997, "trajectory_balance_rmse": 303.5579491636677} +{"canonical_length": 3, "canonical_nll": 1.9668108622233074, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "9af597ad9c807de8", "reading_canonical_log_prob": -5.900432586669922, "reading_distances": [17, 19, 21, 18], "reading_log_probs": [-5.366464138031006, -5.501500606536865, -8.107806205749512, -8.42924976348877], "reading_logprob_std": 1.6427039936973902, "tokenization_path_logprob_std": 13.316884687936563, "trajectory_balance_rmse": 34.733183083413714} +{"canonical_length": 158, "canonical_nll": 1.2707252019568334, "generation_rouge_l_f1": 0.33507853403141363, "generation_token_f1": 0.5602094240837697, "prompt_id": "0366eeefae975865", "reading_canonical_log_prob": -200.7745819091797, "reading_distances": [31, 34, 31, 32], "reading_log_probs": [-201.01895141601562, -201.294921875, -201.00405883789062, -198.6641845703125], "reading_logprob_std": 1.2281999037019962, "tokenization_path_logprob_std": 157.22402273907656, "trajectory_balance_rmse": 3052.4982533673883} +{"canonical_length": 94, "canonical_nll": 1.8398192385409742, "generation_rouge_l_f1": 0.16853932584269662, "generation_token_f1": 0.3258426966292135, "prompt_id": "e99bd769bc8f61be", "reading_canonical_log_prob": -172.94300842285156, "reading_distances": [18, 16, 21, 22], "reading_log_probs": [-171.69107055664062, -171.1650390625, -171.16213989257812, -173.02633666992188], "reading_logprob_std": 0.8793501002980405, "tokenization_path_logprob_std": 262.6923701382607, "trajectory_balance_rmse": 2433.262104112952} +{"canonical_length": 287, "canonical_nll": 0.3019681658063616, "generation_rouge_l_f1": 0.6628571428571428, "generation_token_f1": 0.7600000000000001, "prompt_id": "e1ea57924db21e38", "reading_canonical_log_prob": -86.66486358642578, "reading_distances": [20, 19, 16, 21], "reading_log_probs": [-88.16041564941406, -87.81839752197266, -88.95042419433594, -88.72278594970703], "reading_logprob_std": 0.5170889866018531, "tokenization_path_logprob_std": 98.48376207918994, "trajectory_balance_rmse": 2617.989174085428} +{"canonical_length": 248, "canonical_nll": 0.8009809063326928, "generation_rouge_l_f1": 0.38515901060070673, "generation_token_f1": 0.6819787985865724, "prompt_id": "d13dad8f3ef63531", "reading_canonical_log_prob": -198.6432647705078, "reading_distances": [29, 34, 32, 32], "reading_log_probs": [-197.98680114746094, -197.48593139648438, -199.722412109375, -198.6842498779297], "reading_logprob_std": 0.9689016389317993, "tokenization_path_logprob_std": 188.38332466904137, "trajectory_balance_rmse": 3643.9643742333956} +{"canonical_length": 21, "canonical_nll": 1.1781079428536552, "generation_rouge_l_f1": 0.07434944237918216, "generation_token_f1": 0.0892193308550186, "prompt_id": "146de30da91fe96b", "reading_canonical_log_prob": -24.740266799926758, "reading_distances": [19, 19, 18, 14], "reading_log_probs": [-23.516056060791016, -24.500947952270508, -22.303035736083984, -24.046302795410156], "reading_logprob_std": 0.9486434311899858, "tokenization_path_logprob_std": 51.45764835489152, "trajectory_balance_rmse": 482.63822163420025} +{"canonical_length": 141, "canonical_nll": 1.163565994154477, "generation_rouge_l_f1": 0.37037037037037035, "generation_token_f1": 0.5454545454545454, "prompt_id": "e8d40d5d31440017", "reading_canonical_log_prob": -164.06280517578125, "reading_distances": [76, 71, 76, 68], "reading_log_probs": [-159.41043090820312, -160.51402282714844, -160.15484619140625, -161.87704467773438], "reading_logprob_std": 1.0331608769780896, "tokenization_path_logprob_std": 133.41746295596116, "trajectory_balance_rmse": 2795.275811927125} +{"canonical_length": 7, "canonical_nll": 0.12895712682179042, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "e77edca13c4c71bd", "reading_canonical_log_prob": -0.902699887752533, "reading_distances": [13, 20, 17, 13], "reading_log_probs": [-2.604191780090332, -2.392853021621704, -2.1680350303649902, -2.008674383163452], "reading_logprob_std": 0.2602994945765095, "tokenization_path_logprob_std": 27.95174244569268, "trajectory_balance_rmse": 219.214396781997} +{"canonical_length": 255, "canonical_nll": 1.280444096583946, "generation_rouge_l_f1": 0.19379844961240308, "generation_token_f1": 0.3992248062015504, "prompt_id": "5dc83250d862853a", "reading_canonical_log_prob": -326.51324462890625, "reading_distances": [31, 29, 31, 27], "reading_log_probs": [-329.83648681640625, -330.0259704589844, -330.5528259277344, -326.875244140625], "reading_logprob_std": 1.659499542460616, "tokenization_path_logprob_std": 297.702308141114, "trajectory_balance_rmse": 5756.617084175569} +{"canonical_length": 25, "canonical_nll": 1.2921649169921876, "generation_rouge_l_f1": 0.11881188118811882, "generation_token_f1": 0.19801980198019803, "prompt_id": "a9f2944f6ac15039", "reading_canonical_log_prob": -32.30412292480469, "reading_distances": [26, 27, 34, 28], "reading_log_probs": [-32.33442687988281, -33.28888702392578, -31.016252517700195, -31.72069549560547], "reading_logprob_std": 0.96374167550369, "tokenization_path_logprob_std": 69.46860758851135, "trajectory_balance_rmse": 486.2265654993639} +{"canonical_length": 51, "canonical_nll": 1.2968100753484988, "generation_rouge_l_f1": 0.2608695652173913, "generation_token_f1": 0.4, "prompt_id": "0e9e30fa66e7e5e1", "reading_canonical_log_prob": -66.13731384277344, "reading_distances": [14, 18, 21, 19], "reading_log_probs": [-69.96162414550781, -72.84107971191406, -69.36653137207031, -69.0982894897461], "reading_logprob_std": 1.721038083450524, "tokenization_path_logprob_std": 63.60296975328928, "trajectory_balance_rmse": 867.6343362935709} +{"canonical_length": 102, "canonical_nll": 2.081586949965533, "generation_rouge_l_f1": 0.15615615615615616, "generation_token_f1": 0.26426426426426425, "prompt_id": "f6e26ca6d4af6e82", "reading_canonical_log_prob": -212.32186889648438, "reading_distances": [27, 34, 25, 22], "reading_log_probs": [-210.10812377929688, -222.76083374023438, -222.8682098388672, -219.0739288330078], "reading_logprob_std": 5.99512226478959, "tokenization_path_logprob_std": 50.317592960753586, "trajectory_balance_rmse": 2142.6429238534597} +{"canonical_length": 92, "canonical_nll": 1.7331257695737092, "generation_rouge_l_f1": 0.17318435754189945, "generation_token_f1": 0.2569832402234637, "prompt_id": "e5d1110bf220108e", "reading_canonical_log_prob": -159.44757080078125, "reading_distances": [30, 29, 24, 27], "reading_log_probs": [-159.5550994873047, -159.68551635742188, -158.49472045898438, -159.47483825683594], "reading_logprob_std": 0.5455016911291017, "tokenization_path_logprob_std": 133.04664260872335, "trajectory_balance_rmse": 2015.873173651376} +{"canonical_length": 10, "canonical_nll": 1.0975784301757812, "generation_rouge_l_f1": 0.05529953917050691, "generation_token_f1": 0.06451612903225806, "prompt_id": "dd04d981a73d4900", "reading_canonical_log_prob": -10.975784301757812, "reading_distances": [20, 14, 16, 15], "reading_log_probs": [-13.768452644348145, -12.774032592773438, -12.928414344787598, -13.371968269348145], "reading_logprob_std": 0.4499848447439125, "tokenization_path_logprob_std": 27.39596785170931, "trajectory_balance_rmse": 142.55612990420747} +{"canonical_length": 299, "canonical_nll": 1.5596511483591138, "generation_rouge_l_f1": 0.2642857142857143, "generation_token_f1": 0.5285714285714286, "prompt_id": "d9940a473d55b366", "reading_canonical_log_prob": -466.335693359375, "reading_distances": [26, 27, 16, 26], "reading_log_probs": [-465.48468017578125, -467.9107971191406, -465.882080078125, -467.59271240234375], "reading_logprob_std": 1.2121229187426117, "tokenization_path_logprob_std": 286.45618894160486, "trajectory_balance_rmse": 6185.112904228426} +{"canonical_length": 416, "canonical_nll": 1.1943778991699219, "generation_rouge_l_f1": 0.28177641653905056, "generation_token_f1": 0.450229709035222, "prompt_id": "7c0396dbe5594b7c", "reading_canonical_log_prob": -496.8612060546875, "reading_distances": [25, 21, 23, 25], "reading_log_probs": [-495.712890625, -494.5560302734375, -493.41290283203125, -495.4362487792969], "reading_logprob_std": 1.0360282420252092, "tokenization_path_logprob_std": 68.18112738213607, "trajectory_balance_rmse": 6352.276930992854} +{"canonical_length": 138, "canonical_nll": 1.4499705217886663, "generation_rouge_l_f1": 0.25061425061425063, "generation_token_f1": 0.4127764127764128, "prompt_id": "5ee117416302abcd", "reading_canonical_log_prob": -200.09593200683594, "reading_distances": [20, 17, 19, 22], "reading_log_probs": [-214.3057098388672, -207.88958740234375, -212.67398071289062, -216.77484130859375], "reading_logprob_std": 3.7481216543505815, "tokenization_path_logprob_std": 66.9639642887235, "trajectory_balance_rmse": 2758.197666608187} +{"canonical_length": 326, "canonical_nll": 1.4414520263671875, "generation_rouge_l_f1": 0.23117338003502627, "generation_token_f1": 0.4903677758318739, "prompt_id": "7d6d0adc840aad0a", "reading_canonical_log_prob": -469.9133605957031, "reading_distances": [26, 24, 27, 19], "reading_log_probs": [-472.7107238769531, -473.49139404296875, -472.7183532714844, -472.28277587890625], "reading_logprob_std": 0.5033819930069865, "tokenization_path_logprob_std": 338.01041188312, "trajectory_balance_rmse": 7121.838736220915} +{"canonical_length": 17, "canonical_nll": 1.5937977959127987, "generation_rouge_l_f1": 0.20689655172413796, "generation_token_f1": 0.3448275862068965, "prompt_id": "305ea9936db72d06", "reading_canonical_log_prob": -27.094562530517578, "reading_distances": [62, 54, 57, 53], "reading_log_probs": [-25.533870697021484, -24.096309661865234, -24.683971405029297, -25.56906509399414], "reading_logprob_std": 0.7122672080169902, "tokenization_path_logprob_std": 16.711805619455074, "trajectory_balance_rmse": 375.76087430642775} +{"canonical_length": 8, "canonical_nll": 1.3639633655548096, "generation_rouge_l_f1": 0.4000000000000001, "generation_token_f1": 0.4000000000000001, "prompt_id": "f9a917228bf6b5fe", "reading_canonical_log_prob": -10.911706924438477, "reading_distances": [32, 38, 38, 37], "reading_log_probs": [-12.238996505737305, -11.343742370605469, -11.893513679504395, -11.350411415100098], "reading_logprob_std": 0.43852735099413964, "tokenization_path_logprob_std": 26.168967612576097, "trajectory_balance_rmse": 122.19606220326715} +{"canonical_length": 206, "canonical_nll": 1.1676849068947208, "generation_rouge_l_f1": 0.46393762183235865, "generation_token_f1": 0.608187134502924, "prompt_id": "6fb0440b40107b60", "reading_canonical_log_prob": -240.5430908203125, "reading_distances": [13, 11, 13, 12], "reading_log_probs": [-240.42184448242188, -241.22389221191406, -241.51889038085938, -240.80111694335938], "reading_logprob_std": 0.48058986484066674, "tokenization_path_logprob_std": 189.87074466955002, "trajectory_balance_rmse": 2981.4215859968385} +{"canonical_length": 94, "canonical_nll": 1.7406866195354056, "generation_rouge_l_f1": 0.3253588516746412, "generation_token_f1": 0.4880382775119617, "prompt_id": "16cf88f25ee512b8", "reading_canonical_log_prob": -163.62454223632812, "reading_distances": [17, 17, 16, 22], "reading_log_probs": [-163.2355194091797, -164.20440673828125, -164.74826049804688, -163.73605346679688], "reading_logprob_std": 0.6466166673738926, "tokenization_path_logprob_std": 85.21487428039615, "trajectory_balance_rmse": 2390.773744433377} +{"canonical_length": 12, "canonical_nll": 1.1396481196085613, "generation_rouge_l_f1": 0.3055555555555556, "generation_token_f1": 0.3055555555555556, "prompt_id": "024090bc27e03da4", "reading_canonical_log_prob": -13.675777435302734, "reading_distances": [26, 19, 17, 29], "reading_log_probs": [-13.803211212158203, -13.426335334777832, -13.477482795715332, -16.155460357666016], "reading_logprob_std": 1.3039527645585285, "tokenization_path_logprob_std": 60.66704973232444, "trajectory_balance_rmse": 343.91272594902466} +{"canonical_length": 58, "canonical_nll": 1.948968163852034, "generation_rouge_l_f1": 0.22857142857142856, "generation_token_f1": 0.3142857142857142, "prompt_id": "d97dcaaab25ed5d3", "reading_canonical_log_prob": -113.04015350341797, "reading_distances": [11, 17, 10, 15], "reading_log_probs": [-119.13941192626953, -122.5887451171875, -127.52603149414062, -123.16758728027344], "reading_logprob_std": 3.4419879781890637, "tokenization_path_logprob_std": 102.94122234018715, "trajectory_balance_rmse": 1433.4075719429213} +{"canonical_length": 386, "canonical_nll": 0.9625920112886577, "generation_rouge_l_f1": 0.276595744680851, "generation_token_f1": 0.4042553191489362, "prompt_id": "5a59813799d04978", "reading_canonical_log_prob": -371.5605163574219, "reading_distances": [9, 11, 12, 14], "reading_log_probs": [-373.03729248046875, -372.58740234375, -372.4500427246094, -373.3538818359375], "reading_logprob_std": 0.4154090200182937, "tokenization_path_logprob_std": 246.29254078570105, "trajectory_balance_rmse": 6051.756776985138} +{"canonical_length": 120, "canonical_nll": 1.0999139149983723, "generation_rouge_l_f1": 0.3310344827586207, "generation_token_f1": 0.3310344827586207, "prompt_id": "84eecbd6c7c38a00", "reading_canonical_log_prob": -131.9896697998047, "reading_distances": [20, 24, 27, 23], "reading_log_probs": [-137.89154052734375, -133.85916137695312, -132.846923828125, -135.97901916503906], "reading_logprob_std": 2.2489705110023577, "tokenization_path_logprob_std": 130.82912000568348, "trajectory_balance_rmse": 2219.5245579240154} +{"canonical_length": 12, "canonical_nll": 0.04790291686852773, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1914d368729c6a5a", "reading_canonical_log_prob": -0.5748350024223328, "reading_distances": [23, 22, 23, 27], "reading_log_probs": [-0.8383536338806152, -0.8791568279266357, -0.9370701909065247, -1.1121970415115356], "reading_logprob_std": 0.12066868681797252, "tokenization_path_logprob_std": 22.049065110588465, "trajectory_balance_rmse": 302.28267580563113} +{"canonical_length": 313, "canonical_nll": 1.0869888451914438, "generation_rouge_l_f1": 0.23426573426573427, "generation_token_f1": 0.4580419580419581, "prompt_id": "dc582c1cebaf17e1", "reading_canonical_log_prob": -340.2275085449219, "reading_distances": [43, 44, 34, 50], "reading_log_probs": [-344.7008056640625, -348.3106689453125, -344.3065185546875, -344.744384765625], "reading_logprob_std": 1.8737619110526857, "tokenization_path_logprob_std": 215.6814498244411, "trajectory_balance_rmse": 7216.362799417433} +{"canonical_length": 34, "canonical_nll": 1.2276324103860294, "generation_rouge_l_f1": 0.09876543209876543, "generation_token_f1": 0.1234567901234568, "prompt_id": "c32c06c0fe76b404", "reading_canonical_log_prob": -41.739501953125, "reading_distances": [19, 19, 20, 22], "reading_log_probs": [-41.766273498535156, -41.61802673339844, -40.04486083984375, -38.88838195800781], "reading_logprob_std": 1.3702418672296788, "tokenization_path_logprob_std": 59.02411349388036, "trajectory_balance_rmse": 582.5898359579146} +{"canonical_length": 124, "canonical_nll": 1.7719936986123361, "generation_rouge_l_f1": 0.3644859813084112, "generation_token_f1": 0.5420560747663551, "prompt_id": "3415a33aecea8511", "reading_canonical_log_prob": -219.7272186279297, "reading_distances": [22, 20, 19, 23], "reading_log_probs": [-218.92959594726562, -219.26467895507812, -218.24188232421875, -219.65216064453125], "reading_logprob_std": 0.598085162531619, "tokenization_path_logprob_std": 84.77570092792932, "trajectory_balance_rmse": 3133.8364083649526} +{"canonical_length": 333, "canonical_nll": 1.40802148584131, "generation_rouge_l_f1": 0.19205298013245034, "generation_token_f1": 0.413907284768212, "prompt_id": "f7e2c3707f2870f0", "reading_canonical_log_prob": -468.87115478515625, "reading_distances": [58, 53, 59, 53], "reading_log_probs": [-463.3029479980469, -465.3028564453125, -466.10845947265625, -464.278564453125], "reading_logprob_std": 1.2202814567569582, "tokenization_path_logprob_std": 207.09155301912278, "trajectory_balance_rmse": 6892.436683542414} +{"canonical_length": 39, "canonical_nll": 1.4678673377403846, "generation_rouge_l_f1": 0.16842105263157894, "generation_token_f1": 0.25263157894736843, "prompt_id": "ec9f4061e1384314", "reading_canonical_log_prob": -57.246826171875, "reading_distances": [19, 18, 19, 16], "reading_log_probs": [-55.75981521606445, -56.28331756591797, -56.945098876953125, -57.40638732910156], "reading_logprob_std": 0.7246941322149562, "tokenization_path_logprob_std": 56.88197814913849, "trajectory_balance_rmse": 875.5897672263512} +{"canonical_length": 14, "canonical_nll": 1.3972697939191545, "generation_rouge_l_f1": 0.4827586206896552, "generation_token_f1": 0.4827586206896552, "prompt_id": "a05868d3246d849c", "reading_canonical_log_prob": -19.561777114868164, "reading_distances": [10, 9, 8, 9], "reading_log_probs": [-18.221939086914062, -19.306907653808594, -19.327360153198242, -17.735658645629883], "reading_logprob_std": 0.7978270819408393, "tokenization_path_logprob_std": 97.57488465732669, "trajectory_balance_rmse": 418.7756453225759} +{"canonical_length": 127, "canonical_nll": 1.4703498900406005, "generation_rouge_l_f1": 0.22167487684729067, "generation_token_f1": 0.4285714285714286, "prompt_id": "420ee5406fba3002", "reading_canonical_log_prob": -186.73443603515625, "reading_distances": [21, 24, 21, 25], "reading_log_probs": [-184.9928741455078, -185.3399658203125, -184.39096069335938, -184.96990966796875], "reading_logprob_std": 0.393280434708771, "tokenization_path_logprob_std": 170.99583995893937, "trajectory_balance_rmse": 2950.8115698419547} +{"canonical_length": 9, "canonical_nll": 1.9014388190375433, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "0a4e72c4105acd5f", "reading_canonical_log_prob": -17.11294937133789, "reading_distances": [16, 8, 15, 16], "reading_log_probs": [-18.074481964111328, -18.05744171142578, -18.252771377563477, -18.480789184570312], "reading_logprob_std": 0.1971741878731709, "tokenization_path_logprob_std": 23.655809626366974, "trajectory_balance_rmse": 146.5540545418956} +{"canonical_length": 16, "canonical_nll": 1.4939874410629272, "generation_rouge_l_f1": 0.29629629629629634, "generation_token_f1": 0.4444444444444445, "prompt_id": "0ce850d0751da7a2", "reading_canonical_log_prob": -23.903799057006836, "reading_distances": [61, 60, 73, 64], "reading_log_probs": [-28.619997024536133, -27.50874137878418, -28.58626365661621, -30.369050979614258], "reading_logprob_std": 1.1837782977599323, "tokenization_path_logprob_std": 29.293259565698758, "trajectory_balance_rmse": 395.15522831015755} +{"canonical_length": 9, "canonical_nll": 3.7909634908040366, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.22222222222222224, "prompt_id": "eeee24292d24cae3", "reading_canonical_log_prob": -34.11867141723633, "reading_distances": [19, 19, 26, 23], "reading_log_probs": [-33.73274230957031, -35.523014068603516, -34.74517822265625, -34.0833854675293], "reading_logprob_std": 0.788911636186809, "tokenization_path_logprob_std": 29.509306216906324, "trajectory_balance_rmse": 184.59747453588568} +{"canonical_length": 20, "canonical_nll": 2.0216192245483398, "generation_rouge_l_f1": 0.4444444444444445, "generation_token_f1": 0.4888888888888889, "prompt_id": "aa417b16bf366ab6", "reading_canonical_log_prob": -40.4323844909668, "reading_distances": [31, 33, 27, 38], "reading_log_probs": [-37.82493591308594, -36.06245422363281, -37.415828704833984, -36.44501495361328], "reading_logprob_std": 0.8214999901573407, "tokenization_path_logprob_std": 65.06509955116388, "trajectory_balance_rmse": 438.57201447788464} +{"canonical_length": 277, "canonical_nll": 1.413549099588222, "generation_rouge_l_f1": 0.20809248554913296, "generation_token_f1": 0.4932562620423892, "prompt_id": "31ffef5f718912de", "reading_canonical_log_prob": -391.5531005859375, "reading_distances": [15, 18, 20, 15], "reading_log_probs": [-391.7793273925781, -397.96746826171875, -398.12274169921875, -393.6853332519531], "reading_logprob_std": 3.1651255192081393, "tokenization_path_logprob_std": 221.28056945556062, "trajectory_balance_rmse": 6405.935006886755} +{"canonical_length": 170, "canonical_nll": 1.0648475198184744, "generation_rouge_l_f1": 0.1267605633802817, "generation_token_f1": 0.2112676056338028, "prompt_id": "fe5c1f86ece4d243", "reading_canonical_log_prob": -181.02407836914062, "reading_distances": [18, 14, 18, 19], "reading_log_probs": [-184.02349853515625, -177.85157775878906, -184.26751708984375, -181.79678344726562], "reading_logprob_std": 2.971305287862476, "tokenization_path_logprob_std": 52.281706349647465, "trajectory_balance_rmse": 3252.804998886058} +{"canonical_length": 166, "canonical_nll": 1.2791435517460468, "generation_rouge_l_f1": 0.36597938144329895, "generation_token_f1": 0.5463917525773196, "prompt_id": "8db7426fa0df6c62", "reading_canonical_log_prob": -212.33782958984375, "reading_distances": [18, 17, 24, 22], "reading_log_probs": [-217.40728759765625, -214.41552734375, -215.04098510742188, -215.80560302734375], "reading_logprob_std": 1.2917545799751542, "tokenization_path_logprob_std": 134.43881533419835, "trajectory_balance_rmse": 2345.5648277479304} +{"canonical_length": 39, "canonical_nll": 0.5295024774013422, "generation_rouge_l_f1": 0.380952380952381, "generation_token_f1": 0.47619047619047616, "prompt_id": "6c3aa6c41d3147aa", "reading_canonical_log_prob": -20.650596618652344, "reading_distances": [49, 49, 41, 43], "reading_log_probs": [-20.441844940185547, -20.50550079345703, -20.894086837768555, -19.565834045410156], "reading_logprob_std": 0.5608177125841136, "tokenization_path_logprob_std": 84.53656605703493, "trajectory_balance_rmse": 752.6542704564122} +{"canonical_length": 8, "canonical_nll": 2.1114208698272705, "generation_rouge_l_f1": 0.5454545454545454, "generation_token_f1": 0.5454545454545454, "prompt_id": "e07ed69174530f07", "reading_canonical_log_prob": -16.891366958618164, "reading_distances": [35, 38, 33, 38], "reading_log_probs": [-17.686437606811523, -16.620403289794922, -17.903657913208008, -16.842199325561523], "reading_logprob_std": 0.6270953107633971, "tokenization_path_logprob_std": 46.52870084093584, "trajectory_balance_rmse": 125.65379896107785} +{"canonical_length": 57, "canonical_nll": 2.029311732241982, "generation_rouge_l_f1": 0.11564625850340135, "generation_token_f1": 0.21088435374149658, "prompt_id": "93c96f6a3b9968fc", "reading_canonical_log_prob": -115.67076873779297, "reading_distances": [25, 18, 29, 22], "reading_log_probs": [-116.86607360839844, -115.9607162475586, -116.11737060546875, -115.22577667236328], "reading_logprob_std": 0.6727071356578662, "tokenization_path_logprob_std": 139.09283825173964, "trajectory_balance_rmse": 1327.5902373876797} +{"canonical_length": 264, "canonical_nll": 1.3235045924331204, "generation_rouge_l_f1": 0.2765531062124249, "generation_token_f1": 0.5130260521042084, "prompt_id": "0a9cdb872ed6cdad", "reading_canonical_log_prob": -349.40521240234375, "reading_distances": [23, 20, 21, 19], "reading_log_probs": [-348.28424072265625, -346.6661376953125, -348.43743896484375, -349.014404296875], "reading_logprob_std": 1.0066288279635436, "tokenization_path_logprob_std": 294.83177703590115, "trajectory_balance_rmse": 5213.308480147397} +{"canonical_length": 141, "canonical_nll": 1.5221055483987145, "generation_rouge_l_f1": 0.14444444444444443, "generation_token_f1": 0.2222222222222222, "prompt_id": "c6291fd5f11c978c", "reading_canonical_log_prob": -214.61688232421875, "reading_distances": [23, 17, 21, 19], "reading_log_probs": [-224.10751342773438, -221.92674255371094, -220.967041015625, -219.58724975585938], "reading_logprob_std": 1.9006400183756378, "tokenization_path_logprob_std": 171.36848433916674, "trajectory_balance_rmse": 3185.2099371540967} +{"canonical_length": 9, "canonical_nll": 0.16817369725969103, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "c5b68b2a56c5748c", "reading_canonical_log_prob": -1.5135632753372192, "reading_distances": [48, 43, 38, 41], "reading_log_probs": [-2.2147891521453857, -2.10184907913208, -2.512486457824707, -2.2651562690734863], "reading_logprob_std": 0.17329659298979594, "tokenization_path_logprob_std": 50.59787791908074, "trajectory_balance_rmse": 124.39036782094428} +{"canonical_length": 101, "canonical_nll": 1.7117232520981591, "generation_rouge_l_f1": 0.234375, "generation_token_f1": 0.39062499999999994, "prompt_id": "4478e3a9a8e3ebd2", "reading_canonical_log_prob": -172.88404846191406, "reading_distances": [28, 21, 23, 19], "reading_log_probs": [-172.67466735839844, -173.791259765625, -173.46368408203125, -173.63507080078125], "reading_logprob_std": 0.4960489722777539, "tokenization_path_logprob_std": 124.15755241977757, "trajectory_balance_rmse": 1308.1261217640906} +{"canonical_length": 153, "canonical_nll": 1.6194181753919015, "generation_rouge_l_f1": 0.31143552311435524, "generation_token_f1": 0.4963503649635037, "prompt_id": "1679ec014dbdb2b7", "reading_canonical_log_prob": -247.77098083496094, "reading_distances": [25, 25, 20, 28], "reading_log_probs": [-247.41506958007812, -244.24993896484375, -243.96041870117188, -247.1278076171875], "reading_logprob_std": 1.8356081292599722, "tokenization_path_logprob_std": 141.08049074062754, "trajectory_balance_rmse": 3166.047235965894} +{"canonical_length": 6, "canonical_nll": 0.5999074379603068, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "d608b8ecb5eb0aa8", "reading_canonical_log_prob": -3.599444627761841, "reading_distances": [6, 12, 17, 13], "reading_log_probs": [-3.2693920135498047, -3.000856876373291, -3.037569999694824, -3.447465419769287], "reading_logprob_std": 0.20944052534174876, "tokenization_path_logprob_std": 32.29375909514235, "trajectory_balance_rmse": 136.68632927971927} +{"canonical_length": 8, "canonical_nll": 0.24463129043579102, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.8571428571428571, "prompt_id": "c7c5da8841670598", "reading_canonical_log_prob": -1.9570503234863281, "reading_distances": [23, 23, 28, 23], "reading_log_probs": [-1.8982810974121094, -3.553969383239746, -3.422269582748413, -2.1696572303771973], "reading_logprob_std": 0.8485375796104013, "tokenization_path_logprob_std": 28.866212618728753, "trajectory_balance_rmse": 222.0738504801947} +{"canonical_length": 72, "canonical_nll": 1.5275607638888888, "generation_rouge_l_f1": 0.20754716981132074, "generation_token_f1": 0.2830188679245283, "prompt_id": "8b986ced31f57ea9", "reading_canonical_log_prob": -109.984375, "reading_distances": [31, 31, 35, 33], "reading_log_probs": [-110.537109375, -109.14954376220703, -111.87310791015625, -109.21109008789062], "reading_logprob_std": 1.2902353167087568, "tokenization_path_logprob_std": 147.36702184884032, "trajectory_balance_rmse": 1582.7206397916536} +{"canonical_length": 288, "canonical_nll": 1.2581406699286566, "generation_rouge_l_f1": 0.2567049808429119, "generation_token_f1": 0.5019157088122606, "prompt_id": "3d6847202e1bf0e3", "reading_canonical_log_prob": -362.3445129394531, "reading_distances": [24, 20, 25, 23], "reading_log_probs": [-366.65716552734375, -367.34857177734375, -366.29400634765625, -368.44439697265625], "reading_logprob_std": 0.9460992009168365, "tokenization_path_logprob_std": 201.20644495876033, "trajectory_balance_rmse": 5891.45723528919} +{"canonical_length": 6, "canonical_nll": 1.307737112045288, "generation_rouge_l_f1": 0.8000000000000002, "generation_token_f1": 0.8000000000000002, "prompt_id": "81930576a5952ad4", "reading_canonical_log_prob": -7.8464226722717285, "reading_distances": [48, 43, 45, 39], "reading_log_probs": [-4.844667434692383, -5.244795322418213, -5.469240188598633, -4.378463268280029], "reading_logprob_std": 0.4794302862182691, "tokenization_path_logprob_std": 23.528662945736787, "trajectory_balance_rmse": 96.95791055586976} +{"canonical_length": 271, "canonical_nll": 1.139877600863411, "generation_rouge_l_f1": 0.23463687150837992, "generation_token_f1": 0.5437616387337058, "prompt_id": "52dc7c85e884efbf", "reading_canonical_log_prob": -308.9068298339844, "reading_distances": [16, 11, 11, 19], "reading_log_probs": [-306.03973388671875, -306.8634033203125, -308.274658203125, -309.19464111328125], "reading_logprob_std": 1.4112475624639231, "tokenization_path_logprob_std": 103.98982437293984, "trajectory_balance_rmse": 5950.660928298041} +{"canonical_length": 220, "canonical_nll": 1.1339433843439275, "generation_rouge_l_f1": 0.2646420824295011, "generation_token_f1": 0.41214750542299344, "prompt_id": "02a68089e544097d", "reading_canonical_log_prob": -249.46754455566406, "reading_distances": [27, 21, 23, 25], "reading_log_probs": [-248.79165649414062, -247.474609375, -247.4312744140625, -247.18304443359375], "reading_logprob_std": 0.7257979193200272, "tokenization_path_logprob_std": 361.386665109452, "trajectory_balance_rmse": 4162.42277446972} +{"canonical_length": 116, "canonical_nll": 1.6301172190699085, "generation_rouge_l_f1": 0.27169811320754716, "generation_token_f1": 0.430188679245283, "prompt_id": "4fcd7cfb63238642", "reading_canonical_log_prob": -189.09359741210938, "reading_distances": [28, 26, 36, 35], "reading_log_probs": [-186.73028564453125, -188.18087768554688, -186.76663208007812, -187.17869567871094], "reading_logprob_std": 0.6758245401186779, "tokenization_path_logprob_std": 83.29713956218059, "trajectory_balance_rmse": 3146.578106409483} +{"canonical_length": 260, "canonical_nll": 1.3865875244140624, "generation_rouge_l_f1": 0.19999999999999996, "generation_token_f1": 0.38490566037735846, "prompt_id": "0b49b7e13d1bd9b0", "reading_canonical_log_prob": -360.51275634765625, "reading_distances": [35, 29, 27, 24], "reading_log_probs": [-367.7022399902344, -365.93609619140625, -366.76959228515625, -367.0447998046875], "reading_logprob_std": 0.7314940120912585, "tokenization_path_logprob_std": 254.51767117873734, "trajectory_balance_rmse": 5091.500649846675} +{"canonical_length": 169, "canonical_nll": 1.1888100889307507, "generation_rouge_l_f1": 0.30120481927710846, "generation_token_f1": 0.5481927710843373, "prompt_id": "112cdc2afc089b69", "reading_canonical_log_prob": -200.90890502929688, "reading_distances": [23, 27, 25, 31], "reading_log_probs": [-203.86392211914062, -200.94932556152344, -202.96685791015625, -204.1682891845703], "reading_logprob_std": 1.4510868711260334, "tokenization_path_logprob_std": 96.30858674978657, "trajectory_balance_rmse": 3213.559628850447} +{"canonical_length": 250, "canonical_nll": 1.1810908203125, "generation_rouge_l_f1": 0.2809224318658281, "generation_token_f1": 0.47379454926624737, "prompt_id": "845212ac338f2658", "reading_canonical_log_prob": -295.272705078125, "reading_distances": [41, 42, 28, 40], "reading_log_probs": [-298.3289489746094, -300.26556396484375, -297.97021484375, -299.08282470703125], "reading_logprob_std": 1.0145974926264263, "tokenization_path_logprob_std": 242.55675327294816, "trajectory_balance_rmse": 5417.784720999052} +{"canonical_length": 30, "canonical_nll": 0.7002779642740885, "generation_rouge_l_f1": 0.5454545454545455, "generation_token_f1": 0.6060606060606061, "prompt_id": "9ca4c748ffa8cd48", "reading_canonical_log_prob": -21.008338928222656, "reading_distances": [46, 53, 45, 52], "reading_log_probs": [-23.3682918548584, -24.79416275024414, -23.470279693603516, -24.991592407226562], "reading_logprob_std": 0.8556016987392685, "tokenization_path_logprob_std": 119.3963638303181, "trajectory_balance_rmse": 760.3274994816126} +{"canonical_length": 501, "canonical_nll": 0.5595343736355414, "generation_rouge_l_f1": 0.2502951593860685, "generation_token_f1": 0.38488783943329397, "prompt_id": "585bee96b6b0ece7", "reading_canonical_log_prob": -280.32672119140625, "reading_distances": [19, 22, 22, 17], "reading_log_probs": [-282.8166809082031, -285.3328857421875, -282.01287841796875, -282.7152404785156], "reading_logprob_std": 1.4536019398387034, "tokenization_path_logprob_std": 183.83673331719714, "trajectory_balance_rmse": 8532.266291494278} +{"canonical_length": 168, "canonical_nll": 1.183152335030692, "generation_rouge_l_f1": 0.23310023310023312, "generation_token_f1": 0.41025641025641024, "prompt_id": "b3cb4c0e99b9a63e", "reading_canonical_log_prob": -198.76959228515625, "reading_distances": [21, 25, 25, 26], "reading_log_probs": [-197.40750122070312, -196.54969787597656, -196.413330078125, -197.90536499023438], "reading_logprob_std": 0.7103208695075333, "tokenization_path_logprob_std": 87.66905645823684, "trajectory_balance_rmse": 3686.4083758190095} +{"canonical_length": 430, "canonical_nll": 0.9661129973655523, "generation_rouge_l_f1": 0.20272314674735248, "generation_token_f1": 0.4145234493192133, "prompt_id": "2fa618fc412d053e", "reading_canonical_log_prob": -415.4285888671875, "reading_distances": [27, 29, 26, 25], "reading_log_probs": [-412.2284240722656, -414.02862548828125, -409.56085205078125, -411.1019287109375], "reading_logprob_std": 1.8825321647307003, "tokenization_path_logprob_std": 223.09170204458195, "trajectory_balance_rmse": 7040.890236789861} +{"canonical_length": 160, "canonical_nll": 1.1058063507080078, "generation_rouge_l_f1": 0.2405660377358491, "generation_token_f1": 0.47641509433962265, "prompt_id": "72f894b5c3eed051", "reading_canonical_log_prob": -176.92901611328125, "reading_distances": [27, 26, 28, 25], "reading_log_probs": [-181.01779174804688, -177.18359375, -180.2826385498047, -181.11721801757812], "reading_logprob_std": 1.8489954825501487, "tokenization_path_logprob_std": 222.72106808677586, "trajectory_balance_rmse": 3385.2135351918773} +{"canonical_length": 98, "canonical_nll": 1.3922031947544642, "generation_rouge_l_f1": 0.33170731707317075, "generation_token_f1": 0.5658536585365854, "prompt_id": "9c00b56f3aa02b5b", "reading_canonical_log_prob": -136.4359130859375, "reading_distances": [37, 31, 42, 34], "reading_log_probs": [-136.240234375, -134.12840270996094, -134.87391662597656, -135.75491333007812], "reading_logprob_std": 0.9371800199875759, "tokenization_path_logprob_std": 188.6359407869775, "trajectory_balance_rmse": 2255.3293021508844} +{"canonical_length": 10, "canonical_nll": 0.30001590251922605, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "363eae7c1280c1a6", "reading_canonical_log_prob": -3.0001590251922607, "reading_distances": [65, 53, 54, 43], "reading_log_probs": [-2.686849594116211, -3.8538501262664795, -3.063321113586426, -3.7461764812469482], "reading_logprob_std": 0.5574208230526855, "tokenization_path_logprob_std": 65.31665514907066, "trajectory_balance_rmse": 236.11912441838146} +{"canonical_length": 7, "canonical_nll": 1.1120469229561942, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "86c615edc3ae2a5f", "reading_canonical_log_prob": -7.784328460693359, "reading_distances": [42, 39, 44, 43], "reading_log_probs": [-11.096833229064941, -10.477363586425781, -10.069452285766602, -9.500194549560547], "reading_logprob_std": 0.6729174441525528, "tokenization_path_logprob_std": 28.585368957157407, "trajectory_balance_rmse": 132.01343504632933} +{"canonical_length": 28, "canonical_nll": 0.994321346282959, "generation_rouge_l_f1": 0.1473684210526316, "generation_token_f1": 0.1614035087719298, "prompt_id": "b424089c85e11874", "reading_canonical_log_prob": -27.84099769592285, "reading_distances": [23, 24, 20, 16], "reading_log_probs": [-26.65364646911621, -27.394609451293945, -26.6865291595459, -26.910184860229492], "reading_logprob_std": 0.34180722086687476, "tokenization_path_logprob_std": 91.8090326396446, "trajectory_balance_rmse": 603.8673885422667} +{"canonical_length": 7, "canonical_nll": 0.21134441239493235, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.33333333333333337, "prompt_id": "69e0fa630dc3c290", "reading_canonical_log_prob": -1.4794108867645264, "reading_distances": [41, 32, 25, 23], "reading_log_probs": [-3.4858882427215576, -4.7868218421936035, -3.3234310150146484, -3.021448850631714], "reading_logprob_std": 0.7790898680515712, "tokenization_path_logprob_std": 12.802686971031171, "trajectory_balance_rmse": 96.12867108734989} +{"canonical_length": 158, "canonical_nll": 1.4468754635581487, "generation_rouge_l_f1": 0.19387755102040816, "generation_token_f1": 0.3673469387755102, "prompt_id": "edfb4a427bd0504a", "reading_canonical_log_prob": -228.6063232421875, "reading_distances": [31, 25, 27, 23], "reading_log_probs": [-227.701904296875, -226.20535278320312, -223.79702758789062, -225.20877075195312], "reading_logprob_std": 1.6454398644668822, "tokenization_path_logprob_std": 94.82100608184166, "trajectory_balance_rmse": 3275.032508945971} +{"canonical_length": 9, "canonical_nll": 0.1919121609793769, "generation_rouge_l_f1": 0.875, "generation_token_f1": 0.875, "prompt_id": "9a5bfd989c2d5e52", "reading_canonical_log_prob": -1.727209448814392, "reading_distances": [37, 40, 36, 44], "reading_log_probs": [-1.7378424406051636, -2.661193370819092, -2.207930564880371, -2.308882236480713], "reading_logprob_std": 0.38072384786638547, "tokenization_path_logprob_std": 39.487506061382575, "trajectory_balance_rmse": 134.94713606624182} +{"canonical_length": 272, "canonical_nll": 0.8487581926233628, "generation_rouge_l_f1": 0.33771106941838647, "generation_token_f1": 0.5103189493433397, "prompt_id": "a2cdcfce41950902", "reading_canonical_log_prob": -230.8622283935547, "reading_distances": [29, 18, 27, 24], "reading_log_probs": [-232.47268676757812, -230.6267547607422, -234.35528564453125, -231.18153381347656], "reading_logprob_std": 1.6558266437567368, "tokenization_path_logprob_std": 186.79659335547146, "trajectory_balance_rmse": 3879.328362008153} +{"canonical_length": 72, "canonical_nll": 1.1735187106662326, "generation_rouge_l_f1": 0.1836065573770492, "generation_token_f1": 0.30163934426229505, "prompt_id": "41dfa00e84da9369", "reading_canonical_log_prob": -84.49334716796875, "reading_distances": [44, 37, 39, 41], "reading_log_probs": [-90.04232788085938, -89.6455078125, -93.73512268066406, -93.45807647705078], "reading_logprob_std": 2.175601875789404, "tokenization_path_logprob_std": 68.42796222291473, "trajectory_balance_rmse": 889.0025087544291} +{"canonical_length": 13, "canonical_nll": 0.9552095853365384, "generation_rouge_l_f1": 0.8695652173913043, "generation_token_f1": 0.8695652173913043, "prompt_id": "398d73f7e39625a8", "reading_canonical_log_prob": -12.417724609375, "reading_distances": [27, 27, 24, 31], "reading_log_probs": [-11.720916748046875, -11.19649887084961, -12.57223892211914, -11.461289405822754], "reading_logprob_std": 0.5961091521035394, "tokenization_path_logprob_std": 33.65334052824493, "trajectory_balance_rmse": 146.87520899924579} +{"canonical_length": 89, "canonical_nll": 1.4389476990431882, "generation_rouge_l_f1": 0.26582278481012656, "generation_token_f1": 0.37341772151898733, "prompt_id": "d769bfcaa2a347d7", "reading_canonical_log_prob": -128.06634521484375, "reading_distances": [67, 74, 63, 79], "reading_log_probs": [-126.28427124023438, -126.84832763671875, -127.38697052001953, -125.57118225097656], "reading_logprob_std": 0.7778669957380016, "tokenization_path_logprob_std": 95.87275154872924, "trajectory_balance_rmse": 2171.1736276839156} +{"canonical_length": 351, "canonical_nll": 1.3244378505608974, "generation_rouge_l_f1": 0.28837209302325584, "generation_token_f1": 0.5116279069767442, "prompt_id": "026d0e5d489e80b2", "reading_canonical_log_prob": -464.877685546875, "reading_distances": [7, 12, 10, 9], "reading_log_probs": [-461.45703125, -460.0986328125, -459.5722961425781, -458.23583984375], "reading_logprob_std": 1.332500474954765, "tokenization_path_logprob_std": 88.50585634176213, "trajectory_balance_rmse": 7460.665208190696} +{"canonical_length": 55, "canonical_nll": 1.3325145374644887, "generation_rouge_l_f1": 0.10828025477707007, "generation_token_f1": 0.1337579617834395, "prompt_id": "4c12c24de5859788", "reading_canonical_log_prob": -73.28829956054688, "reading_distances": [20, 25, 24, 27], "reading_log_probs": [-69.45137023925781, -68.28792572021484, -68.27372741699219, -69.4591064453125], "reading_logprob_std": 0.6780790845005517, "tokenization_path_logprob_std": 123.65231750205109, "trajectory_balance_rmse": 1335.6301139214454} +{"canonical_length": 298, "canonical_nll": 1.682607868373794, "generation_rouge_l_f1": 0.17614678899082573, "generation_token_f1": 0.34862385321100914, "prompt_id": "1259351bdd54ab2c", "reading_canonical_log_prob": -501.4171447753906, "reading_distances": [18, 17, 17, 15], "reading_log_probs": [-497.83990478515625, -502.5337829589844, -500.4757080078125, -502.02587890625], "reading_logprob_std": 2.109482541753612, "tokenization_path_logprob_std": 120.27500178273927, "trajectory_balance_rmse": 5618.497881795104} +{"canonical_length": 17, "canonical_nll": 0.8944998348460478, "generation_rouge_l_f1": 0.6153846153846153, "generation_token_f1": 0.6153846153846153, "prompt_id": "1e2afb46510ad456", "reading_canonical_log_prob": -15.206497192382812, "reading_distances": [26, 30, 36, 28], "reading_log_probs": [-19.422426223754883, -17.57632827758789, -22.812986373901367, -25.67607879638672], "reading_logprob_std": 3.59673452548684, "tokenization_path_logprob_std": 64.80561128338067, "trajectory_balance_rmse": 313.9145384778095} +{"canonical_length": 23, "canonical_nll": 1.0727386474609375, "generation_rouge_l_f1": 0.14096916299559473, "generation_token_f1": 0.15859030837004406, "prompt_id": "dfa52b61eae824fc", "reading_canonical_log_prob": -24.672988891601562, "reading_distances": [14, 16, 20, 16], "reading_log_probs": [-21.56032371520996, -21.42220115661621, -22.451919555664062, -21.872573852539062], "reading_logprob_std": 0.4573764061169486, "tokenization_path_logprob_std": 41.97277640353546, "trajectory_balance_rmse": 287.92533096186474} +{"canonical_length": 5, "canonical_nll": 0.9801555633544922, "generation_rouge_l_f1": 0.024896265560165977, "generation_token_f1": 0.024896265560165977, "prompt_id": "30fd5bfb1635c7a5", "reading_canonical_log_prob": -4.900777816772461, "reading_distances": [19, 19, 28, 23], "reading_log_probs": [-7.882452011108398, -7.761914253234863, -7.194730758666992, -7.603824615478516], "reading_logprob_std": 0.2998847461227576, "tokenization_path_logprob_std": 9.882796162067354, "trajectory_balance_rmse": 14.31873107444378} +{"canonical_length": 11, "canonical_nll": 0.06336017088456587, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "bd172f5e24a2f430", "reading_canonical_log_prob": -0.6969618797302246, "reading_distances": [36, 40, 44, 39], "reading_log_probs": [-2.0901081562042236, -1.3191585540771484, -1.0549581050872803, -1.012451410293579], "reading_logprob_std": 0.4994097305612091, "tokenization_path_logprob_std": 24.80354527431888, "trajectory_balance_rmse": 194.12958767265644} +{"canonical_length": 8, "canonical_nll": 0.3459022045135498, "generation_rouge_l_f1": 0.11864406779661016, "generation_token_f1": 0.11864406779661016, "prompt_id": "74eb19d7e1933dab", "reading_canonical_log_prob": -2.7672176361083984, "reading_distances": [27, 23, 22, 23], "reading_log_probs": [-2.4946327209472656, -2.7385730743408203, -2.4713644981384277, -2.484247922897339], "reading_logprob_std": 0.1279335239891438, "tokenization_path_logprob_std": 28.40965429915088, "trajectory_balance_rmse": 165.85620873520293} +{"canonical_length": 2, "canonical_nll": 0.20861470699310303, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "69770982bd8468eb", "reading_canonical_log_prob": -0.41722941398620605, "reading_distances": [27, 24, 27, 30], "reading_log_probs": [-0.41409140825271606, -0.4499626159667969, -0.7200177907943726, -0.4123935103416443], "reading_logprob_std": 0.14828309383515154, "tokenization_path_logprob_std": 18.35633256077618, "trajectory_balance_rmse": 25.453689610642144} +{"canonical_length": 33, "canonical_nll": 0.7393144549745502, "generation_rouge_l_f1": 0.17582417582417584, "generation_token_f1": 0.1978021978021978, "prompt_id": "decae069198bb200", "reading_canonical_log_prob": -24.397377014160156, "reading_distances": [16, 19, 18, 15], "reading_log_probs": [-25.29798126220703, -27.168664932250977, -27.907878875732422, -26.137287139892578], "reading_logprob_std": 1.1460304607137028, "tokenization_path_logprob_std": 80.01026044895339, "trajectory_balance_rmse": 529.8544010808587} +{"canonical_length": 296, "canonical_nll": 1.5459578230574325, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.4492753623188406, "prompt_id": "78378b5209c8e413", "reading_canonical_log_prob": -457.603515625, "reading_distances": [15, 20, 20, 20], "reading_log_probs": [-456.7516174316406, -455.744140625, -455.91778564453125, -455.5347900390625], "reading_logprob_std": 0.5331994438364536, "tokenization_path_logprob_std": 212.08409172183568, "trajectory_balance_rmse": 5915.261281799704} +{"canonical_length": 46, "canonical_nll": 0.8988710486370585, "generation_rouge_l_f1": 0.19730941704035876, "generation_token_f1": 0.3049327354260089, "prompt_id": "6c87b1ea5d43b53e", "reading_canonical_log_prob": -41.34806823730469, "reading_distances": [22, 18, 15, 22], "reading_log_probs": [-41.756202697753906, -44.344032287597656, -44.63404083251953, -41.99614334106445], "reading_logprob_std": 1.5163436413815727, "tokenization_path_logprob_std": 78.56670367977351, "trajectory_balance_rmse": 666.9735603849819} +{"canonical_length": 120, "canonical_nll": 2.0136969248453775, "generation_rouge_l_f1": 0.17553191489361702, "generation_token_f1": 0.30851063829787234, "prompt_id": "c823f34b5de71522", "reading_canonical_log_prob": -241.6436309814453, "reading_distances": [21, 26, 20, 22], "reading_log_probs": [-240.61537170410156, -240.65921020507812, -241.4354248046875, -240.48646545410156], "reading_logprob_std": 0.43049384683356445, "tokenization_path_logprob_std": 92.39315668047307, "trajectory_balance_rmse": 2858.200425676669} +{"canonical_length": 23, "canonical_nll": 1.1416063723356829, "generation_rouge_l_f1": 0.23076923076923075, "generation_token_f1": 0.23076923076923075, "prompt_id": "95ee759ef85d853e", "reading_canonical_log_prob": -26.256946563720703, "reading_distances": [29, 27, 29, 26], "reading_log_probs": [-27.642005920410156, -27.988525390625, -27.904163360595703, -28.609783172607422], "reading_logprob_std": 0.409915903920237, "tokenization_path_logprob_std": 48.160131663038875, "trajectory_balance_rmse": 230.81759247997033} +{"canonical_length": 11, "canonical_nll": 0.5370205965909091, "generation_rouge_l_f1": 0.823529411764706, "generation_token_f1": 0.823529411764706, "prompt_id": "036acca12e6542a8", "reading_canonical_log_prob": -5.9072265625, "reading_distances": [18, 20, 25, 21], "reading_log_probs": [-6.4940571784973145, -7.03102445602417, -6.9458770751953125, -7.363227844238281], "reading_logprob_std": 0.3582042660265108, "tokenization_path_logprob_std": 17.385521977682718, "trajectory_balance_rmse": 195.54418167222659} +{"canonical_length": 13, "canonical_nll": 1.2437638502854567, "generation_rouge_l_f1": 0.4444444444444445, "generation_token_f1": 0.4444444444444445, "prompt_id": "5e5a1ac1eccda1ac", "reading_canonical_log_prob": -16.168930053710938, "reading_distances": [20, 36, 27, 27], "reading_log_probs": [-16.441986083984375, -17.08297348022461, -17.07217025756836, -17.21775245666504], "reading_logprob_std": 0.3475253777976264, "tokenization_path_logprob_std": 107.30683175706893, "trajectory_balance_rmse": 375.0651619035885} +{"canonical_length": 8, "canonical_nll": 0.024519413709640503, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "05b4960bf40b9d90", "reading_canonical_log_prob": -0.19615530967712402, "reading_distances": [24, 30, 24, 25], "reading_log_probs": [-0.44842657446861267, -0.45874691009521484, -0.45361945033073425, -0.30112767219543457], "reading_logprob_std": 0.07635132573431942, "tokenization_path_logprob_std": 39.979519135181924, "trajectory_balance_rmse": 157.84629528282045} +{"canonical_length": 199, "canonical_nll": 1.217737399153973, "generation_rouge_l_f1": 0.3097345132743363, "generation_token_f1": 0.5486725663716814, "prompt_id": "55f01821168c15c2", "reading_canonical_log_prob": -242.32974243164062, "reading_distances": [29, 33, 37, 34], "reading_log_probs": [-239.47381591796875, -239.24560546875, -240.0124969482422, -238.05722045898438], "reading_logprob_std": 0.8252518594405965, "tokenization_path_logprob_std": 105.22726722611769, "trajectory_balance_rmse": 3910.445596548817} +{"canonical_length": 181, "canonical_nll": 0.8526382867802572, "generation_rouge_l_f1": 0.20967741935483872, "generation_token_f1": 0.25, "prompt_id": "28aed3c6f636dfb4", "reading_canonical_log_prob": -154.32752990722656, "reading_distances": [13, 12, 11, 15], "reading_log_probs": [-154.4078369140625, -153.8711700439453, -153.6189727783203, -153.28912353515625], "reading_logprob_std": 0.4719657209201085, "tokenization_path_logprob_std": 142.14076475424366, "trajectory_balance_rmse": 3254.1668478068646} +{"canonical_length": 102, "canonical_nll": 1.982143476897595, "generation_rouge_l_f1": 0.22516556291390727, "generation_token_f1": 0.47019867549668876, "prompt_id": "33e9d73a4126f71a", "reading_canonical_log_prob": -202.1786346435547, "reading_distances": [18, 20, 12, 20], "reading_log_probs": [-199.20697021484375, -200.1559295654297, -201.0293731689453, -199.89601135253906], "reading_logprob_std": 0.753404731779967, "tokenization_path_logprob_std": 198.27752328945957, "trajectory_balance_rmse": 2620.2969677291603} +{"canonical_length": 169, "canonical_nll": 1.8284883216993344, "generation_rouge_l_f1": 0.18981481481481483, "generation_token_f1": 0.33796296296296297, "prompt_id": "ba9c16c03160cc50", "reading_canonical_log_prob": -309.0145263671875, "reading_distances": [23, 21, 23, 16], "reading_log_probs": [-306.0939025878906, -305.1983337402344, -303.814697265625, -303.77557373046875], "reading_logprob_std": 1.129590476800928, "tokenization_path_logprob_std": 282.5974497597385, "trajectory_balance_rmse": 3430.7183604234033} +{"canonical_length": 278, "canonical_nll": 0.6647166519713916, "generation_rouge_l_f1": 0.2601941747572815, "generation_token_f1": 0.41165048543689325, "prompt_id": "d04a82b0b7e2a434", "reading_canonical_log_prob": -184.79122924804688, "reading_distances": [23, 27, 24, 28], "reading_log_probs": [-185.36846923828125, -185.87313842773438, -188.8494873046875, -188.3828887939453], "reading_logprob_std": 1.7520016346849012, "tokenization_path_logprob_std": 126.24917375810388, "trajectory_balance_rmse": 3548.7920774319227} +{"canonical_length": 38, "canonical_nll": 0.5487648813348067, "generation_rouge_l_f1": 0.20765027322404372, "generation_token_f1": 0.28415300546448086, "prompt_id": "e1a492712f46f15d", "reading_canonical_log_prob": -20.853065490722656, "reading_distances": [19, 21, 19, 17], "reading_log_probs": [-20.774356842041016, -21.457477569580078, -20.630420684814453, -21.158172607421875], "reading_logprob_std": 0.3749238511942699, "tokenization_path_logprob_std": 81.72480298445349, "trajectory_balance_rmse": 581.7217589073684} +{"canonical_length": 25, "canonical_nll": 1.0328157043457031, "generation_rouge_l_f1": 0.13333333333333333, "generation_token_f1": 0.14666666666666664, "prompt_id": "fcc6cd8702ae37ce", "reading_canonical_log_prob": -25.820392608642578, "reading_distances": [23, 26, 26, 22], "reading_log_probs": [-25.290855407714844, -26.490989685058594, -26.07891082763672, -28.008296966552734], "reading_logprob_std": 1.1416509791552416, "tokenization_path_logprob_std": 27.94101831797326, "trajectory_balance_rmse": 327.8811996216632} +{"canonical_length": 342, "canonical_nll": 1.1457484730503016, "generation_rouge_l_f1": 0.23225806451612901, "generation_token_f1": 0.4451612903225806, "prompt_id": "38014139c1d9865d", "reading_canonical_log_prob": -391.8459777832031, "reading_distances": [32, 30, 25, 38], "reading_log_probs": [-395.0211181640625, -397.42230224609375, -397.0304260253906, -397.55108642578125], "reading_logprob_std": 1.177744386264295, "tokenization_path_logprob_std": 217.8961204255711, "trajectory_balance_rmse": 6026.982032752405} +{"canonical_length": 6, "canonical_nll": 0.7823030948638916, "generation_rouge_l_f1": 0.888888888888889, "generation_token_f1": 0.888888888888889, "prompt_id": "bd70ea68bde60a24", "reading_canonical_log_prob": -4.69381856918335, "reading_distances": [25, 25, 19, 25], "reading_log_probs": [-4.8912577629089355, -5.139690399169922, -4.66944694519043, -4.640644550323486], "reading_logprob_std": 0.23179211935669658, "tokenization_path_logprob_std": 17.231909276873097, "trajectory_balance_rmse": 116.27680716927858} +{"canonical_length": 11, "canonical_nll": 2.496721614490856, "generation_rouge_l_f1": 0.049999999999999996, "generation_token_f1": 0.075, "prompt_id": "57765c15fcf7839d", "reading_canonical_log_prob": -27.463937759399414, "reading_distances": [10, 12, 12, 14], "reading_log_probs": [-27.065933227539062, -27.991260528564453, -27.98297119140625, -28.06001091003418], "reading_logprob_std": 0.4739998352971321, "tokenization_path_logprob_std": 48.638302085459, "trajectory_balance_rmse": 239.77487683651145} +{"canonical_length": 50, "canonical_nll": 1.1493325805664063, "generation_rouge_l_f1": 0.3684210526315789, "generation_token_f1": 0.4210526315789474, "prompt_id": "253987951dace9e0", "reading_canonical_log_prob": -57.46662902832031, "reading_distances": [32, 23, 29, 26], "reading_log_probs": [-57.403907775878906, -56.82802963256836, -56.41850280761719, -55.920989990234375], "reading_logprob_std": 0.6284674403511752, "tokenization_path_logprob_std": 151.27288491210987, "trajectory_balance_rmse": 1193.8088055691696} +{"canonical_length": 38, "canonical_nll": 1.1620527568616366, "generation_rouge_l_f1": 0.3484848484848485, "generation_token_f1": 0.409090909090909, "prompt_id": "e17273b705c2c377", "reading_canonical_log_prob": -44.15800476074219, "reading_distances": [29, 22, 28, 29], "reading_log_probs": [-43.54844665527344, -42.6026725769043, -42.177181243896484, -44.1829833984375], "reading_logprob_std": 0.9073386800813019, "tokenization_path_logprob_std": 85.6252910867407, "trajectory_balance_rmse": 849.8574028435542} +{"canonical_length": 344, "canonical_nll": 1.6096414965252543, "generation_rouge_l_f1": 0.23890784982935157, "generation_token_f1": 0.40955631399317405, "prompt_id": "3e31d044adc4af42", "reading_canonical_log_prob": -553.7166748046875, "reading_distances": [23, 27, 31, 26], "reading_log_probs": [-545.2733154296875, -548.439697265625, -547.1146240234375, -547.0836181640625], "reading_logprob_std": 1.3002986275156683, "tokenization_path_logprob_std": 120.15007246298106, "trajectory_balance_rmse": 8241.445302415861} +{"canonical_length": 23, "canonical_nll": 0.8090263034986414, "generation_rouge_l_f1": 0.45, "generation_token_f1": 0.45, "prompt_id": "31f5a769dd1b152b", "reading_canonical_log_prob": -18.60760498046875, "reading_distances": [24, 19, 24, 19], "reading_log_probs": [-19.63236427307129, -19.418140411376953, -19.147977828979492, -19.126483917236328], "reading_logprob_std": 0.2406502206354907, "tokenization_path_logprob_std": 48.468512473332936, "trajectory_balance_rmse": 530.2550241242268} +{"canonical_length": 22, "canonical_nll": 1.081559267911044, "generation_rouge_l_f1": 0.1896551724137931, "generation_token_f1": 0.1896551724137931, "prompt_id": "74fb912570f58841", "reading_canonical_log_prob": -23.79430389404297, "reading_distances": [26, 24, 17, 29], "reading_log_probs": [-22.616777420043945, -23.156721115112305, -21.84688949584961, -22.02511978149414], "reading_logprob_std": 0.5959776180575311, "tokenization_path_logprob_std": 42.711094693579035, "trajectory_balance_rmse": 223.19358957312878} +{"canonical_length": 10, "canonical_nll": 0.6065518379211425, "generation_rouge_l_f1": 0.18823529411764706, "generation_token_f1": 0.18823529411764706, "prompt_id": "434a6990c55f121a", "reading_canonical_log_prob": -6.065518379211426, "reading_distances": [14, 20, 19, 15], "reading_log_probs": [-5.843220233917236, -5.950335502624512, -5.473353862762451, -5.632436752319336], "reading_logprob_std": 0.21342113207163016, "tokenization_path_logprob_std": 40.18667694506275, "trajectory_balance_rmse": 269.75927221597385} +{"canonical_length": 222, "canonical_nll": 1.1254008525126689, "generation_rouge_l_f1": 0.21010101010101012, "generation_token_f1": 0.44848484848484854, "prompt_id": "2e39c9a4bb908fa5", "reading_canonical_log_prob": -249.8389892578125, "reading_distances": [16, 23, 18, 23], "reading_log_probs": [-246.85455322265625, -247.08168029785156, -247.986083984375, -248.18115234375], "reading_logprob_std": 0.655530972687271, "tokenization_path_logprob_std": 140.0379189164107, "trajectory_balance_rmse": 5612.3892514548525} +{"canonical_length": 226, "canonical_nll": 1.3512565140175608, "generation_rouge_l_f1": 0.19867549668874174, "generation_token_f1": 0.2958057395143488, "prompt_id": "1a576b185bace9d2", "reading_canonical_log_prob": -305.38397216796875, "reading_distances": [10, 16, 12, 16], "reading_log_probs": [-302.0302734375, -303.4794921875, -302.9117431640625, -307.651611328125], "reading_logprob_std": 2.494526188054073, "tokenization_path_logprob_std": 216.86809773083877, "trajectory_balance_rmse": 4389.729271015221} +{"canonical_length": 270, "canonical_nll": 1.1845967610677084, "generation_rouge_l_f1": 0.26217228464419473, "generation_token_f1": 0.49063670411985016, "prompt_id": "c2df8f1e4778ecd3", "reading_canonical_log_prob": -319.84112548828125, "reading_distances": [35, 28, 35, 35], "reading_log_probs": [-320.62896728515625, -325.8403015136719, -327.9068298339844, -327.080810546875], "reading_logprob_std": 3.2690885454703835, "tokenization_path_logprob_std": 236.5508367291232, "trajectory_balance_rmse": 5775.49915252994} +{"canonical_length": 2, "canonical_nll": 3.1604530811309814, "generation_rouge_l_f1": 0.007662835249042146, "generation_token_f1": 0.007662835249042146, "prompt_id": "42c831675952234b", "reading_canonical_log_prob": -6.320906162261963, "reading_distances": [68, 67, 67, 49], "reading_log_probs": [-6.498906135559082, -6.564741611480713, -6.401036262512207, -7.191558837890625], "reading_logprob_std": 0.35803931427233776, "tokenization_path_logprob_std": 8.168514400600694, "trajectory_balance_rmse": 29.368493703109518} +{"canonical_length": 16, "canonical_nll": 3.7216148376464844, "generation_rouge_l_f1": 0.08, "generation_token_f1": 0.08, "prompt_id": "a9f4c15e6cb406fd", "reading_canonical_log_prob": -59.54583740234375, "reading_distances": [21, 18, 20, 19], "reading_log_probs": [-60.05070114135742, -61.60713195800781, -61.4248046875, -61.08458709716797], "reading_logprob_std": 0.6953177922177739, "tokenization_path_logprob_std": 57.78812047289528, "trajectory_balance_rmse": 468.6719215427934} +{"canonical_length": 161, "canonical_nll": 1.622930301642566, "generation_rouge_l_f1": 0.1323529411764706, "generation_token_f1": 0.2450980392156863, "prompt_id": "25e2da0cacffb5d9", "reading_canonical_log_prob": -261.2917785644531, "reading_distances": [23, 21, 19, 23], "reading_log_probs": [-260.3769226074219, -259.3305358886719, -260.8597412109375, -261.80426025390625], "reading_logprob_std": 1.0293699165621921, "tokenization_path_logprob_std": 194.8336578515288, "trajectory_balance_rmse": 3555.3885744477016} +{"canonical_length": 116, "canonical_nll": 1.7597145869814117, "generation_rouge_l_f1": 0.21024258760107817, "generation_token_f1": 0.2641509433962264, "prompt_id": "d71e3178e4fc08e6", "reading_canonical_log_prob": -204.12689208984375, "reading_distances": [36, 35, 42, 35], "reading_log_probs": [-201.55433654785156, -202.77874755859375, -200.1842041015625, -201.84280395507812], "reading_logprob_std": 1.0730897113048652, "tokenization_path_logprob_std": 108.38257469494506, "trajectory_balance_rmse": 3207.984480107025} +{"canonical_length": 9, "canonical_nll": 1.224426481458876, "generation_rouge_l_f1": 0.044444444444444446, "generation_token_f1": 0.044444444444444446, "prompt_id": "c69244953386b7ea", "reading_canonical_log_prob": -11.019838333129883, "reading_distances": [16, 19, 15, 5], "reading_log_probs": [-12.258146286010742, -12.614431381225586, -12.210402488708496, -10.735963821411133], "reading_logprob_std": 0.8322713448586097, "tokenization_path_logprob_std": 16.973851601027203, "trajectory_balance_rmse": 242.65086373261948} +{"canonical_length": 16, "canonical_nll": 2.0346264839172363, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "44b99fb81cd83d68", "reading_canonical_log_prob": -32.55402374267578, "reading_distances": [25, 21, 24, 30], "reading_log_probs": [-34.312583923339844, -33.092445373535156, -33.82185363769531, -33.838016510009766], "reading_logprob_std": 0.5035658263469366, "tokenization_path_logprob_std": 73.12036737181415, "trajectory_balance_rmse": 425.48468857493583} +{"canonical_length": 23, "canonical_nll": 0.41165476259977923, "generation_rouge_l_f1": 0.04065040650406504, "generation_token_f1": 0.08130081300813008, "prompt_id": "f3f998d8d70ee428", "reading_canonical_log_prob": -9.468059539794922, "reading_distances": [51, 41, 40, 42], "reading_log_probs": [-11.431680679321289, -12.50308609008789, -11.293275833129883, -11.986971855163574], "reading_logprob_std": 0.554274501635186, "tokenization_path_logprob_std": 21.227687366526002, "trajectory_balance_rmse": 308.8780615954836} +{"canonical_length": 45, "canonical_nll": 0.6357311672634549, "generation_rouge_l_f1": 0.36363636363636365, "generation_token_f1": 0.43939393939393945, "prompt_id": "8565123aabdca8e0", "reading_canonical_log_prob": -28.60790252685547, "reading_distances": [21, 22, 24, 22], "reading_log_probs": [-31.68829345703125, -31.729284286499023, -32.634586334228516, -31.553138732910156], "reading_logprob_std": 0.4945996044829728, "tokenization_path_logprob_std": 59.74886277703016, "trajectory_balance_rmse": 1039.5465321270995} +{"canonical_length": 293, "canonical_nll": 1.6672390361694753, "generation_rouge_l_f1": 0.22426470588235295, "generation_token_f1": 0.45588235294117646, "prompt_id": "c754674f366c507b", "reading_canonical_log_prob": -488.50103759765625, "reading_distances": [27, 30, 36, 32], "reading_log_probs": [-487.34661865234375, -487.6001281738281, -489.54241943359375, -489.0863952636719], "reading_logprob_std": 1.0840538904736703, "tokenization_path_logprob_std": 21.1289926340156, "trajectory_balance_rmse": 6438.274495561672} +{"canonical_length": 15, "canonical_nll": 1.210242462158203, "generation_rouge_l_f1": 0.09448818897637797, "generation_token_f1": 0.11023622047244096, "prompt_id": "f7c03b54285f64b4", "reading_canonical_log_prob": -18.153636932373047, "reading_distances": [26, 26, 29, 25], "reading_log_probs": [-20.703014373779297, -24.47693634033203, -23.028732299804688, -22.18932342529297], "reading_logprob_std": 1.5783863206964164, "tokenization_path_logprob_std": 77.78110786971851, "trajectory_balance_rmse": 264.9224793685933} +{"canonical_length": 33, "canonical_nll": 0.33246615438750293, "generation_rouge_l_f1": 0.06315789473684211, "generation_token_f1": 0.09824561403508772, "prompt_id": "278cabb1c777cfb5", "reading_canonical_log_prob": -10.971383094787598, "reading_distances": [58, 44, 50, 56], "reading_log_probs": [-15.600282669067383, -17.492891311645508, -13.898518562316895, -15.750940322875977], "reading_logprob_std": 1.468730783279135, "tokenization_path_logprob_std": 65.50052377566497, "trajectory_balance_rmse": 455.62592611153667} +{"canonical_length": 82, "canonical_nll": 0.46331061386480565, "generation_rouge_l_f1": 0.25531914893617025, "generation_token_f1": 0.3333333333333333, "prompt_id": "26496139eab8348f", "reading_canonical_log_prob": -37.99147033691406, "reading_distances": [17, 20, 18, 17], "reading_log_probs": [-36.56575012207031, -37.13780212402344, -36.673622131347656, -36.02301788330078], "reading_logprob_std": 0.4577964433807357, "tokenization_path_logprob_std": 33.64808242809569, "trajectory_balance_rmse": 873.5776558914956} +{"canonical_length": 130, "canonical_nll": 1.7139596792367788, "generation_rouge_l_f1": 0.19638242894056845, "generation_token_f1": 0.3565891472868217, "prompt_id": "5e25b104005e6f72", "reading_canonical_log_prob": -222.81475830078125, "reading_distances": [25, 23, 23, 19], "reading_log_probs": [-220.76473999023438, -221.0616455078125, -214.81735229492188, -212.304931640625], "reading_logprob_std": 4.36855643225279, "tokenization_path_logprob_std": 246.04688969147142, "trajectory_balance_rmse": 1798.7962154960958} +{"canonical_length": 107, "canonical_nll": 2.0801283221378504, "generation_rouge_l_f1": 0.2391304347826087, "generation_token_f1": 0.39855072463768115, "prompt_id": "14f8273006e150ca", "reading_canonical_log_prob": -222.57373046875, "reading_distances": [25, 25, 34, 30], "reading_log_probs": [-226.40679931640625, -223.9805908203125, -224.20785522460938, -224.0640869140625], "reading_logprob_std": 1.1650977863005938, "tokenization_path_logprob_std": 133.39353103595596, "trajectory_balance_rmse": 2084.1736078822896} +{"canonical_length": 185, "canonical_nll": 1.5755460172086149, "generation_rouge_l_f1": 0.3057324840764331, "generation_token_f1": 0.43312101910828027, "prompt_id": "c0e6a5cad32d62b6", "reading_canonical_log_prob": -291.47601318359375, "reading_distances": [13, 17, 17, 12], "reading_log_probs": [-295.18060302734375, -294.05194091796875, -296.76165771484375, -294.3377685546875], "reading_logprob_std": 1.2173565366606172, "tokenization_path_logprob_std": 205.80179869413655, "trajectory_balance_rmse": 3498.659439189505} +{"canonical_length": 16, "canonical_nll": 1.4458541870117188, "generation_rouge_l_f1": 0.3555555555555555, "generation_token_f1": 0.4888888888888889, "prompt_id": "937bfa51bfbd0542", "reading_canonical_log_prob": -23.1336669921875, "reading_distances": [31, 34, 37, 39], "reading_log_probs": [-24.542011260986328, -23.312015533447266, -23.91097068786621, -22.854150772094727], "reading_logprob_std": 0.732872799087843, "tokenization_path_logprob_std": 31.66282951922854, "trajectory_balance_rmse": 448.25772960541735} +{"canonical_length": 75, "canonical_nll": 0.4272786966959635, "generation_rouge_l_f1": 0.28800000000000003, "generation_token_f1": 0.38400000000000006, "prompt_id": "19b52394b9bc5537", "reading_canonical_log_prob": -32.045902252197266, "reading_distances": [13, 13, 17, 12], "reading_log_probs": [-31.938642501831055, -31.122386932373047, -31.753122329711914, -31.628271102905273], "reading_logprob_std": 0.3495658090750938, "tokenization_path_logprob_std": 60.861417804229724, "trajectory_balance_rmse": 600.7719049615893} +{"canonical_length": 339, "canonical_nll": 1.4162793904982485, "generation_rouge_l_f1": 0.19889502762430936, "generation_token_f1": 0.36095764272559855, "prompt_id": "c03437022c1ce69c", "reading_canonical_log_prob": -480.11871337890625, "reading_distances": [29, 29, 24, 28], "reading_log_probs": [-482.73626708984375, -481.126220703125, -485.4203796386719, -480.33367919921875], "reading_logprob_std": 2.2455805202334704, "tokenization_path_logprob_std": 133.5776100902128, "trajectory_balance_rmse": 6148.0306923034605} +{"canonical_length": 22, "canonical_nll": 1.0509041872891514, "generation_rouge_l_f1": 0.08270676691729324, "generation_token_f1": 0.09774436090225563, "prompt_id": "01ef5d8df6df3761", "reading_canonical_log_prob": -23.119892120361328, "reading_distances": [21, 20, 21, 26], "reading_log_probs": [-22.499605178833008, -24.082256317138672, -24.223438262939453, -23.563615798950195], "reading_logprob_std": 0.7817069858155546, "tokenization_path_logprob_std": 46.46848637631512, "trajectory_balance_rmse": 362.66486602717265} +{"canonical_length": 110, "canonical_nll": 1.5392403342507102, "generation_rouge_l_f1": 0.22811671087533156, "generation_token_f1": 0.3448275862068966, "prompt_id": "1275798cfa3be44f", "reading_canonical_log_prob": -169.31643676757812, "reading_distances": [36, 34, 37, 38], "reading_log_probs": [-168.7587890625, -169.76002502441406, -172.24740600585938, -169.21493530273438], "reading_logprob_std": 1.5561988020561555, "tokenization_path_logprob_std": 135.52677993221002, "trajectory_balance_rmse": 2677.03994296528} +{"canonical_length": 4, "canonical_nll": 0.19926530122756958, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1f2fdcd7eec76cc5", "reading_canonical_log_prob": -0.7970612049102783, "reading_distances": [29, 29, 25, 33], "reading_log_probs": [-3.5141615867614746, -1.3642137050628662, -1.3657556772232056, -2.0711100101470947], "reading_logprob_std": 1.0131449056330706, "tokenization_path_logprob_std": 29.75894163522397, "trajectory_balance_rmse": 51.00162888639922} +{"canonical_length": 128, "canonical_nll": 1.189184308052063, "generation_rouge_l_f1": 0.20281690140845074, "generation_token_f1": 0.3211267605633803, "prompt_id": "b8b97c6b0ef87a98", "reading_canonical_log_prob": -152.21559143066406, "reading_distances": [32, 28, 24, 26], "reading_log_probs": [-147.78843688964844, -148.5888671875, -147.4681396484375, -149.33409118652344], "reading_logprob_std": 0.8379285151447923, "tokenization_path_logprob_std": 79.61569076671768, "trajectory_balance_rmse": 1680.4958396639217} +{"canonical_length": 168, "canonical_nll": 1.0507629031226748, "generation_rouge_l_f1": 0.22055137844611528, "generation_token_f1": 0.40601503759398494, "prompt_id": "7ec56cfda55797c0", "reading_canonical_log_prob": -176.52816772460938, "reading_distances": [27, 35, 24, 27], "reading_log_probs": [-171.9344482421875, -172.57766723632812, -173.6243438720703, -173.12460327148438], "reading_logprob_std": 0.7263125235459347, "tokenization_path_logprob_std": 70.19854550764673, "trajectory_balance_rmse": 3426.626418777498} +{"canonical_length": 32, "canonical_nll": 1.8807587623596191, "generation_rouge_l_f1": 0.15384615384615385, "generation_token_f1": 0.3076923076923077, "prompt_id": "c6a2889afcb5222e", "reading_canonical_log_prob": -60.18428039550781, "reading_distances": [15, 16, 13, 9], "reading_log_probs": [-60.33412170410156, -59.94072723388672, -60.934722900390625, -60.08015441894531], "reading_logprob_std": 0.4394821765963858, "tokenization_path_logprob_std": 43.15191519985355, "trajectory_balance_rmse": 504.57299147630715} +{"canonical_length": 32, "canonical_nll": 1.6364033222198486, "generation_rouge_l_f1": 0.326530612244898, "generation_token_f1": 0.326530612244898, "prompt_id": "5371fa87b29b339c", "reading_canonical_log_prob": -52.364906311035156, "reading_distances": [12, 17, 19, 16], "reading_log_probs": [-52.05732727050781, -51.36339569091797, -52.869468688964844, -53.34284973144531], "reading_logprob_std": 0.8757980410120287, "tokenization_path_logprob_std": 104.88757942993709, "trajectory_balance_rmse": 782.8188291657547} +{"canonical_length": 63, "canonical_nll": 1.239548819405692, "generation_rouge_l_f1": 0.28169014084507044, "generation_token_f1": 0.40845070422535207, "prompt_id": "c89f55288eb64d0c", "reading_canonical_log_prob": -78.0915756225586, "reading_distances": [22, 29, 25, 31], "reading_log_probs": [-80.67753601074219, -79.58385467529297, -80.95643615722656, -81.01749420166016], "reading_logprob_std": 0.6666162081544509, "tokenization_path_logprob_std": 33.81287430413501, "trajectory_balance_rmse": 1565.5160303757973} +{"canonical_length": 221, "canonical_nll": 1.3227634343626273, "generation_rouge_l_f1": 0.2248995983935743, "generation_token_f1": 0.4176706827309237, "prompt_id": "0e1e524655779cac", "reading_canonical_log_prob": -292.3307189941406, "reading_distances": [18, 24, 18, 17], "reading_log_probs": [-291.1323547363281, -290.5114440917969, -289.246337890625, -290.02239990234375], "reading_logprob_std": 0.7966868618309103, "tokenization_path_logprob_std": 128.99444411901942, "trajectory_balance_rmse": 4731.149350916531} +{"canonical_length": 9, "canonical_nll": 0.4760504828559028, "generation_rouge_l_f1": 0.8235294117647058, "generation_token_f1": 0.8235294117647058, "prompt_id": "5f85bd9588d13639", "reading_canonical_log_prob": -4.284454345703125, "reading_distances": [21, 13, 20, 14], "reading_log_probs": [-3.889939308166504, -3.9397904872894287, -3.9313442707061768, -3.8647005558013916], "reading_logprob_std": 0.035340871906765284, "tokenization_path_logprob_std": 40.251928594261216, "trajectory_balance_rmse": 211.7706971668507} +{"canonical_length": 26, "canonical_nll": 0.8122124305138221, "generation_rouge_l_f1": 0.08540925266903915, "generation_token_f1": 0.14234875444839856, "prompt_id": "570dbf2f8a32567a", "reading_canonical_log_prob": -21.117523193359375, "reading_distances": [20, 14, 21, 23], "reading_log_probs": [-22.482460021972656, -21.12981605529785, -21.720821380615234, -22.04454803466797], "reading_logprob_std": 0.5695266210046527, "tokenization_path_logprob_std": 99.92434605682024, "trajectory_balance_rmse": 566.4693809803181} +{"canonical_length": 26, "canonical_nll": 0.9467344284057617, "generation_rouge_l_f1": 0.14617940199335547, "generation_token_f1": 0.16611295681063123, "prompt_id": "b4dad9a8bbd784f6", "reading_canonical_log_prob": -24.615095138549805, "reading_distances": [40, 29, 38, 37], "reading_log_probs": [-25.15993881225586, -25.3287353515625, -24.891334533691406, -25.219892501831055], "reading_logprob_std": 0.1860446977831594, "tokenization_path_logprob_std": 50.987574283456176, "trajectory_balance_rmse": 647.770452386648} +{"canonical_length": 10, "canonical_nll": 1.8483531951904297, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.39999999999999997, "prompt_id": "9e76d45f4e1b2541", "reading_canonical_log_prob": -18.483531951904297, "reading_distances": [28, 41, 31, 40], "reading_log_probs": [-17.83451271057129, -17.25299072265625, -17.610578536987305, -18.31214714050293], "reading_logprob_std": 0.4433141148446794, "tokenization_path_logprob_std": 35.26237520616951, "trajectory_balance_rmse": 332.45882917855215} +{"canonical_length": 256, "canonical_nll": 1.3973262310028076, "generation_rouge_l_f1": 0.2375478927203065, "generation_token_f1": 0.46360153256704983, "prompt_id": "6a9973c4375e58ef", "reading_canonical_log_prob": -357.71551513671875, "reading_distances": [32, 35, 29, 33], "reading_log_probs": [-350.02099609375, -352.053955078125, -353.65972900390625, -350.4505310058594], "reading_logprob_std": 1.658468925224756, "tokenization_path_logprob_std": 290.0785787061901, "trajectory_balance_rmse": 6097.83683739255} +{"canonical_length": 277, "canonical_nll": 1.0996137818704874, "generation_rouge_l_f1": 0.241635687732342, "generation_token_f1": 0.5092936802973977, "prompt_id": "4a7ad2175af88b68", "reading_canonical_log_prob": -304.593017578125, "reading_distances": [30, 31, 28, 36], "reading_log_probs": [-306.03125, -303.66888427734375, -306.92364501953125, -306.0549621582031], "reading_logprob_std": 1.396995227887594, "tokenization_path_logprob_std": 268.7479525762755, "trajectory_balance_rmse": 5611.036024366176} +{"canonical_length": 24, "canonical_nll": 0.2676961024602254, "generation_rouge_l_f1": 0.37037037037037035, "generation_token_f1": 0.37037037037037035, "prompt_id": "7271cafab8ca2bf4", "reading_canonical_log_prob": -6.42470645904541, "reading_distances": [28, 34, 34, 26], "reading_log_probs": [-8.834948539733887, -8.686628341674805, -9.67519760131836, -8.328669548034668], "reading_logprob_std": 0.5702979675440378, "tokenization_path_logprob_std": 86.8842238464931, "trajectory_balance_rmse": 544.589044830836} +{"canonical_length": 11, "canonical_nll": 0.7899636355313387, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "7798c57a62239065", "reading_canonical_log_prob": -8.689599990844727, "reading_distances": [14, 14, 17, 15], "reading_log_probs": [-10.12091064453125, -9.836594581604004, -9.916155815124512, -10.539182662963867], "reading_logprob_std": 0.31435570355416514, "tokenization_path_logprob_std": 56.17112610860064, "trajectory_balance_rmse": 269.9892068034008} +{"canonical_length": 97, "canonical_nll": 0.9961094807103738, "generation_rouge_l_f1": 0.29629629629629634, "generation_token_f1": 0.4159544159544159, "prompt_id": "fd06dfbeb5280c45", "reading_canonical_log_prob": -96.62261962890625, "reading_distances": [39, 44, 39, 45], "reading_log_probs": [-98.97681427001953, -98.00112915039062, -97.01554870605469, -97.89305114746094], "reading_logprob_std": 0.8023988395495818, "tokenization_path_logprob_std": 244.95279379461311, "trajectory_balance_rmse": 2156.412961024329} +{"canonical_length": 210, "canonical_nll": 1.030728004092262, "generation_rouge_l_f1": 0.27638190954773867, "generation_token_f1": 0.5577889447236181, "prompt_id": "3efd48165ac7ab7c", "reading_canonical_log_prob": -216.452880859375, "reading_distances": [16, 19, 17, 19], "reading_log_probs": [-216.0217742919922, -221.65077209472656, -216.6798858642578, -218.49105834960938], "reading_logprob_std": 2.519754245638875, "tokenization_path_logprob_std": 132.9559293870094, "trajectory_balance_rmse": 2292.332728417475} +{"canonical_length": 69, "canonical_nll": 1.4032978666001472, "generation_rouge_l_f1": 0.175, "generation_token_f1": 0.19999999999999998, "prompt_id": "18e11e5d4d3c48a7", "reading_canonical_log_prob": -96.82755279541016, "reading_distances": [32, 40, 35, 39], "reading_log_probs": [-98.20585632324219, -99.8766860961914, -97.60124206542969, -98.44567108154297], "reading_logprob_std": 0.9640734550919434, "tokenization_path_logprob_std": 117.99137975956832, "trajectory_balance_rmse": 1916.7552030067236} +{"canonical_length": 17, "canonical_nll": 0.27336962082806754, "generation_rouge_l_f1": 0.7692307692307693, "generation_token_f1": 1.0, "prompt_id": "b05a564949c24a61", "reading_canonical_log_prob": -4.647283554077148, "reading_distances": [70, 64, 51, 67], "reading_log_probs": [-5.3651227951049805, -6.164666652679443, -5.182036399841309, -6.031816482543945], "reading_logprob_std": 0.48499218650428877, "tokenization_path_logprob_std": 53.29305992824764, "trajectory_balance_rmse": 144.33055425045114} +{"canonical_length": 129, "canonical_nll": 0.7155916672344356, "generation_rouge_l_f1": 0.26044226044226043, "generation_token_f1": 0.42751842751842745, "prompt_id": "977aa631a6e3fb74", "reading_canonical_log_prob": -92.31132507324219, "reading_distances": [9, 15, 14, 11], "reading_log_probs": [-94.8642807006836, -95.03228759765625, -95.50408935546875, -93.46141815185547], "reading_logprob_std": 0.8788410039918704, "tokenization_path_logprob_std": 103.25155473780958, "trajectory_balance_rmse": 1361.5616913548545} +{"canonical_length": 85, "canonical_nll": 1.1774798224954044, "generation_rouge_l_f1": 0.21839080459770116, "generation_token_f1": 0.3333333333333333, "prompt_id": "308d6723913890bf", "reading_canonical_log_prob": -100.08578491210938, "reading_distances": [24, 23, 23, 21], "reading_log_probs": [-100.21681213378906, -99.36581420898438, -100.36996459960938, -99.59107971191406], "reading_logprob_std": 0.48346996043161117, "tokenization_path_logprob_std": 44.19597829120884, "trajectory_balance_rmse": 1886.369435756774} +{"canonical_length": 189, "canonical_nll": 1.1520652972831928, "generation_rouge_l_f1": 0.2600896860986547, "generation_token_f1": 0.47982062780269064, "prompt_id": "ff140f7a02786106", "reading_canonical_log_prob": -217.74034118652344, "reading_distances": [15, 15, 20, 21], "reading_log_probs": [-223.61825561523438, -220.24673461914062, -216.19021606445312, -215.71435546875], "reading_logprob_std": 3.721993303041505, "tokenization_path_logprob_std": 157.1841156409232, "trajectory_balance_rmse": 3652.1259193222936} +{"canonical_length": 27, "canonical_nll": 1.6191916289152923, "generation_rouge_l_f1": 0.24390243902439027, "generation_token_f1": 0.2926829268292683, "prompt_id": "536f5c5bf3146493", "reading_canonical_log_prob": -43.71817398071289, "reading_distances": [31, 42, 32, 33], "reading_log_probs": [-50.97358322143555, -52.40620803833008, -48.511192321777344, -50.125694274902344], "reading_logprob_std": 1.6282200273195928, "tokenization_path_logprob_std": 32.00301304044059, "trajectory_balance_rmse": 790.1792466415362} +{"canonical_length": 210, "canonical_nll": 1.3608389718191964, "generation_rouge_l_f1": 0.2340862422997947, "generation_token_f1": 0.517453798767967, "prompt_id": "18e930a2b5a84e85", "reading_canonical_log_prob": -285.77618408203125, "reading_distances": [14, 19, 17, 14], "reading_log_probs": [-285.31842041015625, -285.7586669921875, -284.94476318359375, -286.70556640625], "reading_logprob_std": 0.7592263984556611, "tokenization_path_logprob_std": 240.0111011997402, "trajectory_balance_rmse": 4562.928278838876} +{"canonical_length": 21, "canonical_nll": 1.620196751185826, "generation_rouge_l_f1": 0.0790513833992095, "generation_token_f1": 0.11067193675889328, "prompt_id": "2631f3bcf5083818", "reading_canonical_log_prob": -34.024131774902344, "reading_distances": [32, 21, 35, 28], "reading_log_probs": [-35.43988800048828, -32.666263580322266, -33.16020202636719, -36.10967254638672], "reading_logprob_std": 1.6866877982994932, "tokenization_path_logprob_std": 56.47817622150809, "trajectory_balance_rmse": 287.9501131741202} +{"canonical_length": 169, "canonical_nll": 0.8242787919806306, "generation_rouge_l_f1": 0.21897810218978103, "generation_token_f1": 0.3746958637469586, "prompt_id": "62ad42126bc04505", "reading_canonical_log_prob": -139.30311584472656, "reading_distances": [43, 45, 43, 38], "reading_log_probs": [-134.5091552734375, -133.9176483154297, -134.84622192382812, -134.34246826171875], "reading_logprob_std": 0.3859801002125068, "tokenization_path_logprob_std": 185.56817456786632, "trajectory_balance_rmse": 3067.1638652667407} +{"canonical_length": 221, "canonical_nll": 1.2524568721719458, "generation_rouge_l_f1": 0.26326963906581746, "generation_token_f1": 0.5605095541401275, "prompt_id": "44a0d1ef8b9a47d0", "reading_canonical_log_prob": -276.79296875, "reading_distances": [15, 7, 13, 19], "reading_log_probs": [-274.90576171875, -276.44708251953125, -276.4246826171875, -275.309814453125], "reading_logprob_std": 0.7843713274824813, "tokenization_path_logprob_std": 205.8881030156341, "trajectory_balance_rmse": 5382.709209498901} +{"canonical_length": 84, "canonical_nll": 1.1368474506196522, "generation_rouge_l_f1": 0.23853211009174316, "generation_token_f1": 0.3302752293577982, "prompt_id": "1bc849b3b651f6ed", "reading_canonical_log_prob": -95.49518585205078, "reading_distances": [23, 20, 21, 26], "reading_log_probs": [-97.64572143554688, -97.140869140625, -96.87367248535156, -97.9709701538086], "reading_logprob_std": 0.49339352682852095, "tokenization_path_logprob_std": 166.1673950491142, "trajectory_balance_rmse": 2103.7539008912477} +{"canonical_length": 225, "canonical_nll": 1.1908013237847221, "generation_rouge_l_f1": 0.22580645161290322, "generation_token_f1": 0.49193548387096775, "prompt_id": "a8febb0397bb60c3", "reading_canonical_log_prob": -267.9302978515625, "reading_distances": [16, 15, 18, 12], "reading_log_probs": [-265.8659973144531, -265.0548095703125, -265.8490905761719, -265.02325439453125], "reading_logprob_std": 0.4727940414366535, "tokenization_path_logprob_std": 77.61594044528763, "trajectory_balance_rmse": 4555.385200468637} +{"canonical_length": 7, "canonical_nll": 1.7982664108276367, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.5714285714285715, "prompt_id": "37af528e072ecf7b", "reading_canonical_log_prob": -12.587864875793457, "reading_distances": [9, 15, 22, 20], "reading_log_probs": [-11.590763092041016, -11.265894889831543, -11.839576721191406, -10.76644515991211], "reading_logprob_std": 0.4634220402118913, "tokenization_path_logprob_std": 53.77175521821489, "trajectory_balance_rmse": 141.50336525294244} +{"canonical_length": 252, "canonical_nll": 1.471619621155754, "generation_rouge_l_f1": 0.20161290322580647, "generation_token_f1": 0.47580645161290325, "prompt_id": "192d126d4e011ab4", "reading_canonical_log_prob": -370.84814453125, "reading_distances": [21, 25, 16, 20], "reading_log_probs": [-369.77362060546875, -369.5334167480469, -369.74920654296875, -370.9065856933594], "reading_logprob_std": 0.6200529893800144, "tokenization_path_logprob_std": 178.0229910109466, "trajectory_balance_rmse": 6304.868421229431} +{"canonical_length": 182, "canonical_nll": 1.2096892975188873, "generation_rouge_l_f1": 0.23776223776223776, "generation_token_f1": 0.46153846153846156, "prompt_id": "55712d2518889f49", "reading_canonical_log_prob": -220.1634521484375, "reading_distances": [23, 25, 23, 19], "reading_log_probs": [-241.02435302734375, -240.59201049804688, -238.09249877929688, -227.02987670898438], "reading_logprob_std": 6.564987278705085, "tokenization_path_logprob_std": 73.24291938880606, "trajectory_balance_rmse": 3515.7442708090807} +{"canonical_length": 12, "canonical_nll": 0.9703704516092936, "generation_rouge_l_f1": 0.588235294117647, "generation_token_f1": 0.588235294117647, "prompt_id": "587eb61a8e257f2b", "reading_canonical_log_prob": -11.644445419311523, "reading_distances": [21, 25, 23, 29], "reading_log_probs": [-12.545417785644531, -12.594732284545898, -12.223499298095703, -12.734277725219727], "reading_logprob_std": 0.21600790322363198, "tokenization_path_logprob_std": 36.296154164293085, "trajectory_balance_rmse": 359.1072212807403} +{"canonical_length": 109, "canonical_nll": 0.8064033438306336, "generation_rouge_l_f1": 0.13609467455621302, "generation_token_f1": 0.242603550295858, "prompt_id": "fff5e81d18ab40f8", "reading_canonical_log_prob": -87.89796447753906, "reading_distances": [15, 15, 14, 16], "reading_log_probs": [-90.56842041015625, -90.39175415039062, -91.138427734375, -91.07991027832031], "reading_logprob_std": 0.371062190691292, "tokenization_path_logprob_std": 137.2808029676384, "trajectory_balance_rmse": 1296.919330231911} +{"canonical_length": 15, "canonical_nll": 0.7880585988362631, "generation_rouge_l_f1": 0.4210526315789473, "generation_token_f1": 0.4210526315789473, "prompt_id": "e02d75efb09e53b0", "reading_canonical_log_prob": -11.820878982543945, "reading_distances": [25, 28, 20, 18], "reading_log_probs": [-12.475625038146973, -13.493803024291992, -11.656454086303711, -11.89773941040039], "reading_logprob_std": 0.8176780155675587, "tokenization_path_logprob_std": 26.46342203434699, "trajectory_balance_rmse": 251.09230311387438} +{"canonical_length": 27, "canonical_nll": 1.233022760461878, "generation_rouge_l_f1": 0.11666666666666667, "generation_token_f1": 0.15, "prompt_id": "93316a155b82ba8e", "reading_canonical_log_prob": -33.2916145324707, "reading_distances": [27, 27, 28, 24], "reading_log_probs": [-35.81471633911133, -34.7672233581543, -33.02922439575195, -33.06839370727539], "reading_logprob_std": 1.3634115214169777, "tokenization_path_logprob_std": 145.69615462092648, "trajectory_balance_rmse": 598.8302095907281} +{"canonical_length": 110, "canonical_nll": 1.2105171897194602, "generation_rouge_l_f1": 0.4244897959183674, "generation_token_f1": 0.6122448979591837, "prompt_id": "37f1255c888facef", "reading_canonical_log_prob": -133.15689086914062, "reading_distances": [14, 19, 18, 12], "reading_log_probs": [-129.38742065429688, -129.14944458007812, -130.03700256347656, -130.67535400390625], "reading_logprob_std": 0.6868413746267862, "tokenization_path_logprob_std": 188.94007239034917, "trajectory_balance_rmse": 2639.061265321019} +{"canonical_length": 150, "canonical_nll": 0.8201908365885416, "generation_rouge_l_f1": 0.4966887417218543, "generation_token_f1": 0.6026490066225165, "prompt_id": "cf1ad5289e6ad557", "reading_canonical_log_prob": -123.02862548828125, "reading_distances": [19, 21, 21, 19], "reading_log_probs": [-121.6607666015625, -122.28771209716797, -122.4957275390625, -122.22577667236328], "reading_logprob_std": 0.357005683085151, "tokenization_path_logprob_std": 71.228408689967, "trajectory_balance_rmse": 3023.5487279390522} +{"canonical_length": 314, "canonical_nll": 1.0119290685957405, "generation_rouge_l_f1": 0.21674876847290642, "generation_token_f1": 0.47619047619047616, "prompt_id": "ff9bcfc5d56ada93", "reading_canonical_log_prob": -317.7457275390625, "reading_distances": [41, 34, 36, 40], "reading_log_probs": [-329.4039001464844, -325.80096435546875, -327.95819091796875, -330.6014404296875], "reading_logprob_std": 2.06539522407929, "tokenization_path_logprob_std": 215.82856322936996, "trajectory_balance_rmse": 5750.780168047703} +{"canonical_length": 12, "canonical_nll": 2.010833422342936, "generation_rouge_l_f1": 0.10526315789473685, "generation_token_f1": 0.10526315789473685, "prompt_id": "6f26ffa7e9fb10c0", "reading_canonical_log_prob": -24.130001068115234, "reading_distances": [22, 21, 20, 22], "reading_log_probs": [-24.057334899902344, -23.910926818847656, -23.97588539123535, -24.08120346069336], "reading_logprob_std": 0.0779661804319154, "tokenization_path_logprob_std": 45.98509024314367, "trajectory_balance_rmse": 253.67434353055438} +{"canonical_length": 41, "canonical_nll": 0.9392374550423971, "generation_rouge_l_f1": 0.17346938775510204, "generation_token_f1": 0.25510204081632654, "prompt_id": "07e67bead15cc6c8", "reading_canonical_log_prob": -38.50873565673828, "reading_distances": [31, 38, 35, 24], "reading_log_probs": [-37.6229362487793, -37.29322052001953, -37.94573211669922, -39.24823760986328], "reading_logprob_std": 0.856295213579229, "tokenization_path_logprob_std": 113.83135596286735, "trajectory_balance_rmse": 806.4042416514764} +{"canonical_length": 413, "canonical_nll": 1.1605127071352905, "generation_rouge_l_f1": 0.23493975903614459, "generation_token_f1": 0.43373493975903615, "prompt_id": "52cb3482c2bf1893", "reading_canonical_log_prob": -479.291748046875, "reading_distances": [26, 24, 22, 33], "reading_log_probs": [-476.49481201171875, -475.0337219238281, -473.7402648925781, -475.8963623046875], "reading_logprob_std": 1.1953518204984657, "tokenization_path_logprob_std": 83.78634947136212, "trajectory_balance_rmse": 7596.176071500834} +{"canonical_length": 265, "canonical_nll": 1.6548914495504128, "generation_rouge_l_f1": 0.22440944881889766, "generation_token_f1": 0.4566929133858268, "prompt_id": "3b670d936478ce43", "reading_canonical_log_prob": -438.5462341308594, "reading_distances": [31, 26, 30, 29], "reading_log_probs": [-439.86376953125, -437.6618347167969, -438.377197265625, -438.2661437988281], "reading_logprob_std": 0.9354182631449609, "tokenization_path_logprob_std": 330.3898183639745, "trajectory_balance_rmse": 6450.607139644743} +{"canonical_length": 78, "canonical_nll": 2.1856967241336136, "generation_rouge_l_f1": 0.16216216216216217, "generation_token_f1": 0.25225225225225223, "prompt_id": "d940482038c5842a", "reading_canonical_log_prob": -170.48434448242188, "reading_distances": [20, 30, 28, 26], "reading_log_probs": [-170.34088134765625, -170.69764709472656, -167.99334716796875, -169.38339233398438], "reading_logprob_std": 1.208570538830387, "tokenization_path_logprob_std": 155.7403641872678, "trajectory_balance_rmse": 1951.303923785623} +{"canonical_length": 291, "canonical_nll": 0.9690356697003866, "generation_rouge_l_f1": 0.2434782608695652, "generation_token_f1": 0.4556521739130435, "prompt_id": "07d10dd03fee840c", "reading_canonical_log_prob": -281.9893798828125, "reading_distances": [26, 27, 19, 24], "reading_log_probs": [-284.6432189941406, -284.5799865722656, -282.893798828125, -282.82574462890625], "reading_logprob_std": 1.0121309040453974, "tokenization_path_logprob_std": 261.0137326150722, "trajectory_balance_rmse": 6112.74786648918} +{"canonical_length": 4, "canonical_nll": 1.3402361869812012, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "7ef3a945dcbdd086", "reading_canonical_log_prob": -5.360944747924805, "reading_distances": [21, 23, 17, 24], "reading_log_probs": [-5.596384048461914, -5.414483070373535, -5.402379512786865, -5.114780426025391], "reading_logprob_std": 0.19902872084985207, "tokenization_path_logprob_std": 16.468736038723364, "trajectory_balance_rmse": 47.39373672705303} +{"canonical_length": 156, "canonical_nll": 0.9809192755283453, "generation_rouge_l_f1": 0.22613065326633167, "generation_token_f1": 0.3969849246231156, "prompt_id": "ef63cfbdbb1633a2", "reading_canonical_log_prob": -153.02340698242188, "reading_distances": [25, 27, 34, 30], "reading_log_probs": [-154.42051696777344, -153.33749389648438, -155.01849365234375, -154.99319458007812], "reading_logprob_std": 0.7866712417828834, "tokenization_path_logprob_std": 184.82301756125744, "trajectory_balance_rmse": 2915.5826164107975} +{"canonical_length": 152, "canonical_nll": 1.468431874325401, "generation_rouge_l_f1": 0.2783018867924528, "generation_token_f1": 0.4716981132075472, "prompt_id": "1d1550150b11ce18", "reading_canonical_log_prob": -223.20164489746094, "reading_distances": [24, 30, 29, 27], "reading_log_probs": [-221.7872314453125, -222.05389404296875, -221.5081787109375, -221.71234130859375], "reading_logprob_std": 0.225598050077171, "tokenization_path_logprob_std": 259.68918552192144, "trajectory_balance_rmse": 4003.315483229709} +{"canonical_length": 436, "canonical_nll": 0.8139761128556837, "generation_rouge_l_f1": 0.2579505300353357, "generation_token_f1": 0.3674911660777385, "prompt_id": "becce7dcacc67b08", "reading_canonical_log_prob": -354.8935852050781, "reading_distances": [22, 25, 20, 29], "reading_log_probs": [-357.2000732421875, -355.18280029296875, -353.9143371582031, -354.3531494140625], "reading_logprob_std": 1.4566028556603414, "tokenization_path_logprob_std": 64.49472511577851, "trajectory_balance_rmse": 5728.755051233959} +{"canonical_length": 289, "canonical_nll": 1.0855908245363863, "generation_rouge_l_f1": 0.1693121693121693, "generation_token_f1": 0.3597883597883598, "prompt_id": "11042f17f722f10a", "reading_canonical_log_prob": -313.7357482910156, "reading_distances": [23, 29, 28, 26], "reading_log_probs": [-311.71307373046875, -312.2574768066406, -312.0758361816406, -313.14813232421875], "reading_logprob_std": 0.6098800468023767, "tokenization_path_logprob_std": 171.4324283473176, "trajectory_balance_rmse": 6556.974603397722} +{"canonical_length": 31, "canonical_nll": 1.2576995357390373, "generation_rouge_l_f1": 0.11805555555555555, "generation_token_f1": 0.14583333333333334, "prompt_id": "b29cd52aed559d3c", "reading_canonical_log_prob": -38.988685607910156, "reading_distances": [21, 19, 19, 18], "reading_log_probs": [-37.56181335449219, -37.423866271972656, -39.044097900390625, -38.14134979248047], "reading_logprob_std": 0.7363732428713674, "tokenization_path_logprob_std": 41.83812978891123, "trajectory_balance_rmse": 523.4830761811469} +{"canonical_length": 192, "canonical_nll": 0.8486124674479166, "generation_rouge_l_f1": 0.2053789731051345, "generation_token_f1": 0.36674816625916873, "prompt_id": "3d6b176b83ba1bb0", "reading_canonical_log_prob": -162.93359375, "reading_distances": [20, 22, 20, 16], "reading_log_probs": [-168.24098205566406, -165.47039794921875, -165.64328002929688, -164.64431762695312], "reading_logprob_std": 1.5564492152325742, "tokenization_path_logprob_std": 81.31742148712301, "trajectory_balance_rmse": 2338.0528280345256} +{"canonical_length": 146, "canonical_nll": 1.863083930864726, "generation_rouge_l_f1": 0.23376623376623376, "generation_token_f1": 0.474025974025974, "prompt_id": "ed74cc90058602c2", "reading_canonical_log_prob": -272.01025390625, "reading_distances": [28, 29, 22, 30], "reading_log_probs": [-270.30987548828125, -270.77349853515625, -270.2422790527344, -270.7071533203125], "reading_logprob_std": 0.2708088492844596, "tokenization_path_logprob_std": 193.66815202664162, "trajectory_balance_rmse": 3638.8440275718544} +{"canonical_length": 217, "canonical_nll": 1.5214909847980271, "generation_rouge_l_f1": 0.27572016460905346, "generation_token_f1": 0.5555555555555556, "prompt_id": "b2e99fb7bff1b5b7", "reading_canonical_log_prob": -330.1635437011719, "reading_distances": [26, 37, 39, 36], "reading_log_probs": [-328.9293212890625, -330.6322937011719, -332.03753662109375, -329.41455078125], "reading_logprob_std": 1.3884711449948444, "tokenization_path_logprob_std": 197.95538533277977, "trajectory_balance_rmse": 5216.086110095861} +{"canonical_length": 318, "canonical_nll": 0.71204064327216, "generation_rouge_l_f1": 0.3377483443708609, "generation_token_f1": 0.5264900662251656, "prompt_id": "5a32b0efbf18f769", "reading_canonical_log_prob": -226.42892456054688, "reading_distances": [26, 29, 29, 25], "reading_log_probs": [-222.65567016601562, -221.7147674560547, -222.346923828125, -222.17848205566406], "reading_logprob_std": 0.39278452986162016, "tokenization_path_logprob_std": 105.52847233480085, "trajectory_balance_rmse": 5407.594381897984} +{"canonical_length": 25, "canonical_nll": 0.523201789855957, "generation_rouge_l_f1": 0.15602836879432622, "generation_token_f1": 0.18439716312056736, "prompt_id": "3697c04353d3a1f1", "reading_canonical_log_prob": -13.080044746398926, "reading_distances": [29, 29, 32, 26], "reading_log_probs": [-14.542649269104004, -12.974695205688477, -14.073173522949219, -14.693061828613281], "reading_logprob_std": 0.7770248459691212, "tokenization_path_logprob_std": 51.34191010612893, "trajectory_balance_rmse": 287.5651724174058} +{"canonical_length": 83, "canonical_nll": 1.7180389036615211, "generation_rouge_l_f1": 0.25510204081632654, "generation_token_f1": 0.5714285714285714, "prompt_id": "4983a8fb02a849d4", "reading_canonical_log_prob": -142.59722900390625, "reading_distances": [18, 19, 13, 10], "reading_log_probs": [-141.634521484375, -141.91192626953125, -141.88812255859375, -141.74566650390625], "reading_logprob_std": 0.12978276364637661, "tokenization_path_logprob_std": 146.70550966657478, "trajectory_balance_rmse": 1511.2788122872719} +{"canonical_length": 244, "canonical_nll": 1.6765236776383197, "generation_rouge_l_f1": 0.23668639053254437, "generation_token_f1": 0.47337278106508873, "prompt_id": "fe89f27185bbc3b2", "reading_canonical_log_prob": -409.07177734375, "reading_distances": [43, 39, 29, 41], "reading_log_probs": [-404.8743591308594, -404.65185546875, -405.4835205078125, -402.8121032714844], "reading_logprob_std": 1.1505907767839512, "tokenization_path_logprob_std": 84.74663862475376, "trajectory_balance_rmse": 6423.2451132104925} +{"canonical_length": 87, "canonical_nll": 1.4906903847880748, "generation_rouge_l_f1": 0.2021276595744681, "generation_token_f1": 0.37234042553191493, "prompt_id": "6b767f6a353fd065", "reading_canonical_log_prob": -129.6900634765625, "reading_distances": [20, 17, 23, 18], "reading_log_probs": [-133.93508911132812, -132.1754913330078, -136.13145446777344, -135.70968627929688], "reading_logprob_std": 1.8117088359666649, "tokenization_path_logprob_std": 108.32617029750463, "trajectory_balance_rmse": 1735.0175076898886} +{"canonical_length": 136, "canonical_nll": 1.3133768193862017, "generation_rouge_l_f1": 0.2618181818181818, "generation_token_f1": 0.5963636363636364, "prompt_id": "e77b6aefa4b6596d", "reading_canonical_log_prob": -178.61924743652344, "reading_distances": [31, 25, 29, 28], "reading_log_probs": [-179.01702880859375, -180.317138671875, -177.85531616210938, -178.30892944335938], "reading_logprob_std": 1.0739532122280235, "tokenization_path_logprob_std": 150.6540273832489, "trajectory_balance_rmse": 2697.7388445592765} +{"canonical_length": 9, "canonical_nll": 1.7400115331013997, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.8571428571428571, "prompt_id": "383239282305a169", "reading_canonical_log_prob": -15.660103797912598, "reading_distances": [31, 34, 35, 32], "reading_log_probs": [-13.668015480041504, -13.729391098022461, -14.189329147338867, -12.78554916381836], "reading_logprob_std": 0.586462598520853, "tokenization_path_logprob_std": 22.17227873900344, "trajectory_balance_rmse": 132.02354529326175} +{"canonical_length": 10, "canonical_nll": 1.1981875419616699, "generation_rouge_l_f1": 0.4615384615384615, "generation_token_f1": 0.6153846153846153, "prompt_id": "29653e40ee0ea781", "reading_canonical_log_prob": -11.9818754196167, "reading_distances": [30, 24, 20, 25], "reading_log_probs": [-12.064559936523438, -12.774056434631348, -14.042411804199219, -12.31965446472168], "reading_logprob_std": 0.8786095005796476, "tokenization_path_logprob_std": 26.506792001653046, "trajectory_balance_rmse": 175.66156054449792} +{"canonical_length": 251, "canonical_nll": 1.5493645535047311, "generation_rouge_l_f1": 0.22891566265060243, "generation_token_f1": 0.46987951807228917, "prompt_id": "b8e0545bd0cc139e", "reading_canonical_log_prob": -388.8905029296875, "reading_distances": [48, 52, 34, 45], "reading_log_probs": [-387.7122802734375, -387.76800537109375, -385.4039306640625, -385.927734375], "reading_logprob_std": 1.2167581297694736, "tokenization_path_logprob_std": 85.90276881480207, "trajectory_balance_rmse": 6043.994260884911} +{"canonical_length": 430, "canonical_nll": 1.248706196629724, "generation_rouge_l_f1": 0.1984251968503937, "generation_token_f1": 0.368503937007874, "prompt_id": "a7d20d598bd1e5dd", "reading_canonical_log_prob": -536.9436645507812, "reading_distances": [36, 35, 38, 29], "reading_log_probs": [-596.438720703125, -594.40771484375, -597.56494140625, -601.8734130859375], "reading_logprob_std": 3.151678087328895, "tokenization_path_logprob_std": 320.89448248715456, "trajectory_balance_rmse": 8161.129336621819} +{"canonical_length": 47, "canonical_nll": 2.8276428871966424, "generation_rouge_l_f1": 0.13402061855670105, "generation_token_f1": 0.2061855670103093, "prompt_id": "ecbe9c8ff4f546fd", "reading_canonical_log_prob": -132.8992156982422, "reading_distances": [12, 14, 10, 13], "reading_log_probs": [-130.41928100585938, -137.25970458984375, -136.9503173828125, -137.16201782226562], "reading_logprob_std": 3.3548523435619093, "tokenization_path_logprob_std": 77.20543348921447, "trajectory_balance_rmse": 912.7488223404611} +{"canonical_length": 12, "canonical_nll": 0.7616353034973145, "generation_rouge_l_f1": 0.5263157894736842, "generation_token_f1": 0.5263157894736842, "prompt_id": "641a36ce0ce7d08c", "reading_canonical_log_prob": -9.139623641967773, "reading_distances": [25, 23, 26, 29], "reading_log_probs": [-12.235533714294434, -12.084529876708984, -14.139921188354492, -11.650961875915527], "reading_logprob_std": 1.1029789646991082, "tokenization_path_logprob_std": 50.34661839911783, "trajectory_balance_rmse": 325.2457468041315} +{"canonical_length": 105, "canonical_nll": 1.723705328078497, "generation_rouge_l_f1": 0.23115577889447234, "generation_token_f1": 0.35175879396984927, "prompt_id": "32d37daf5bb72f67", "reading_canonical_log_prob": -180.9890594482422, "reading_distances": [19, 20, 24, 20], "reading_log_probs": [-178.69418334960938, -179.51715087890625, -179.47695922851562, -180.10586547851562], "reading_logprob_std": 0.5792655827130843, "tokenization_path_logprob_std": 63.59983578347321, "trajectory_balance_rmse": 2602.0287526079687} +{"canonical_length": 19, "canonical_nll": 2.4412472373560856, "generation_rouge_l_f1": 0.23076923076923078, "generation_token_f1": 0.23076923076923078, "prompt_id": "b21a0da1ea5ee3b7", "reading_canonical_log_prob": -46.383697509765625, "reading_distances": [20, 26, 23, 24], "reading_log_probs": [-45.82587814331055, -47.93424987792969, -46.01447677612305, -46.93977355957031], "reading_logprob_std": 0.9683426503421515, "tokenization_path_logprob_std": 67.70455420232813, "trajectory_balance_rmse": 449.0492169190545} +{"canonical_length": 281, "canonical_nll": 1.4616682928227869, "generation_rouge_l_f1": 0.24574669187145556, "generation_token_f1": 0.5217391304347826, "prompt_id": "11a80e6e30ed7f8f", "reading_canonical_log_prob": -410.7287902832031, "reading_distances": [19, 17, 19, 21], "reading_log_probs": [-409.70697021484375, -412.8659973144531, -410.7562255859375, -410.88232421875], "reading_logprob_std": 1.318580127971984, "tokenization_path_logprob_std": 112.02789765028201, "trajectory_balance_rmse": 6551.548920022535} +{"canonical_length": 119, "canonical_nll": 1.8698472736262475, "generation_rouge_l_f1": 0.17571059431524547, "generation_token_f1": 0.37209302325581395, "prompt_id": "e2dff42d7f8d6988", "reading_canonical_log_prob": -222.51182556152344, "reading_distances": [23, 21, 26, 21], "reading_log_probs": [-220.36148071289062, -220.12625122070312, -219.02391052246094, -218.40135192871094], "reading_logprob_std": 0.9248668176219951, "tokenization_path_logprob_std": 150.11404719347533, "trajectory_balance_rmse": 2676.675649060738} +{"canonical_length": 338, "canonical_nll": 1.6746009962093196, "generation_rouge_l_f1": 0.1927710843373494, "generation_token_f1": 0.41996557659208267, "prompt_id": "bd63a203fd3cb2b2", "reading_canonical_log_prob": -566.01513671875, "reading_distances": [29, 29, 31, 30], "reading_log_probs": [-560.4450073242188, -561.3995361328125, -561.9534301757812, -561.06494140625], "reading_logprob_std": 0.6310670558056234, "tokenization_path_logprob_std": 134.1615097849889, "trajectory_balance_rmse": 6254.504408756814} +{"canonical_length": 17, "canonical_nll": 0.8248701656565947, "generation_rouge_l_f1": 0.19834710743801653, "generation_token_f1": 0.21487603305785125, "prompt_id": "f5560d8544cd62d8", "reading_canonical_log_prob": -14.02279281616211, "reading_distances": [25, 21, 22, 18], "reading_log_probs": [-14.430169105529785, -14.407952308654785, -14.0986909866333, -13.984052658081055], "reading_logprob_std": 0.22320891549604982, "tokenization_path_logprob_std": 46.911465274083156, "trajectory_balance_rmse": 447.05329119822613} +{"canonical_length": 143, "canonical_nll": 1.5591498955146417, "generation_rouge_l_f1": 0.22388059701492538, "generation_token_f1": 0.43781094527363185, "prompt_id": "d5ad06fe63008db3", "reading_canonical_log_prob": -222.95843505859375, "reading_distances": [21, 24, 21, 26], "reading_log_probs": [-226.25486755371094, -227.05482482910156, -227.68087768554688, -226.93690490722656], "reading_logprob_std": 0.5843767529071552, "tokenization_path_logprob_std": 229.14433092803816, "trajectory_balance_rmse": 3421.3792365443055} +{"canonical_length": 68, "canonical_nll": 1.2685266382554, "generation_rouge_l_f1": 0.2337662337662338, "generation_token_f1": 0.4025974025974026, "prompt_id": "32cf7dbccdbc97bc", "reading_canonical_log_prob": -86.25981140136719, "reading_distances": [16, 20, 16, 23], "reading_log_probs": [-86.42056274414062, -86.14582824707031, -85.18936157226562, -85.94467163085938], "reading_logprob_std": 0.5278567900858766, "tokenization_path_logprob_std": 49.44594840194697, "trajectory_balance_rmse": 1838.1649185551237} +{"canonical_length": 47, "canonical_nll": 1.8338892510596743, "generation_rouge_l_f1": 0.2531645569620253, "generation_token_f1": 0.32911392405063294, "prompt_id": "c180433ec97360b0", "reading_canonical_log_prob": -86.19279479980469, "reading_distances": [32, 24, 28, 36], "reading_log_probs": [-86.09010314941406, -86.63472747802734, -86.41381072998047, -86.24736022949219], "reading_logprob_std": 0.2332193906651033, "tokenization_path_logprob_std": 9.384309938937447, "trajectory_balance_rmse": 921.7235509746823} +{"canonical_length": 122, "canonical_nll": 0.9642519090996414, "generation_rouge_l_f1": 0.2905405405405405, "generation_token_f1": 0.46621621621621623, "prompt_id": "aaf9ee6d37011f66", "reading_canonical_log_prob": -117.63873291015625, "reading_distances": [35, 34, 46, 32], "reading_log_probs": [-130.53591918945312, -125.86567687988281, -121.63522338867188, -129.50987243652344], "reading_logprob_std": 4.033952946867696, "tokenization_path_logprob_std": 177.44584640289983, "trajectory_balance_rmse": 2226.790978004824} +{"canonical_length": 19, "canonical_nll": 0.3367518876728259, "generation_rouge_l_f1": 0.10572687224669602, "generation_token_f1": 0.1145374449339207, "prompt_id": "f459f20b3852c4a3", "reading_canonical_log_prob": -6.398285865783691, "reading_distances": [15, 21, 20, 22], "reading_log_probs": [-5.09444522857666, -5.754433631896973, -5.87432861328125, -5.1185221672058105], "reading_logprob_std": 0.41174264480835154, "tokenization_path_logprob_std": 80.1088665325252, "trajectory_balance_rmse": 267.22685626226286} +{"canonical_length": 19, "canonical_nll": 1.4905648482473273, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "a05d7928d7de0d6c", "reading_canonical_log_prob": -28.32073211669922, "reading_distances": [26, 28, 28, 20], "reading_log_probs": [-28.324548721313477, -28.447280883789062, -29.97850799560547, -27.779457092285156], "reading_logprob_std": 0.9431436488597289, "tokenization_path_logprob_std": 20.081053197019564, "trajectory_balance_rmse": 387.5524743781169} +{"canonical_length": 15, "canonical_nll": 0.9471144994099935, "generation_rouge_l_f1": 0.8181818181818181, "generation_token_f1": 0.8181818181818181, "prompt_id": "2c22dccbca7a08c7", "reading_canonical_log_prob": -14.206717491149902, "reading_distances": [26, 28, 26, 28], "reading_log_probs": [-15.55804443359375, -15.25539779663086, -14.5297269821167, -15.443641662597656], "reading_logprob_std": 0.46182551622150986, "tokenization_path_logprob_std": 46.03959579900019, "trajectory_balance_rmse": 389.09270477002343} +{"canonical_length": 166, "canonical_nll": 1.7799651364246047, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.532051282051282, "prompt_id": "6e818f0ad7674362", "reading_canonical_log_prob": -295.4742126464844, "reading_distances": [25, 24, 28, 23], "reading_log_probs": [-293.173095703125, -293.35626220703125, -293.88348388671875, -292.9141845703125], "reading_logprob_std": 0.410098851441341, "tokenization_path_logprob_std": 227.81103827303423, "trajectory_balance_rmse": 3303.7221091391966} +{"canonical_length": 70, "canonical_nll": 0.9900804792131697, "generation_rouge_l_f1": 0.25316455696202533, "generation_token_f1": 0.5189873417721518, "prompt_id": "8ebd94ecdb15c9f4", "reading_canonical_log_prob": -69.30563354492188, "reading_distances": [22, 16, 20, 23], "reading_log_probs": [-67.7603530883789, -67.10694122314453, -69.41880798339844, -67.77154541015625], "reading_logprob_std": 0.9864686851033737, "tokenization_path_logprob_std": 74.99990374382617, "trajectory_balance_rmse": 1834.5928309316257} +{"canonical_length": 3, "canonical_nll": 0.17315646012624106, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "2148a4bb4107090b", "reading_canonical_log_prob": -0.5194693803787231, "reading_distances": [15, 22, 21, 16], "reading_log_probs": [-0.5904853343963623, -0.5646833181381226, -0.5492292046546936, -0.5889236330986023], "reading_logprob_std": 0.019942252751435914, "tokenization_path_logprob_std": 20.516744396663977, "trajectory_balance_rmse": 47.504522870637075} +{"canonical_length": 368, "canonical_nll": 0.9292364535124406, "generation_rouge_l_f1": 0.2080745341614907, "generation_token_f1": 0.4409937888198758, "prompt_id": "5d2176dd2ade1ce9", "reading_canonical_log_prob": -341.9590148925781, "reading_distances": [25, 27, 20, 22], "reading_log_probs": [-335.4672546386719, -335.92401123046875, -336.96710205078125, -336.5644836425781], "reading_logprob_std": 0.6659845646041412, "tokenization_path_logprob_std": 207.55732899264308, "trajectory_balance_rmse": 7586.1825618882} +{"canonical_length": 12, "canonical_nll": 0.9330000082651774, "generation_rouge_l_f1": 0.06249999999999999, "generation_token_f1": 0.06944444444444443, "prompt_id": "9d6ee475e283101d", "reading_canonical_log_prob": -11.196000099182129, "reading_distances": [30, 28, 33, 33], "reading_log_probs": [-10.50064754486084, -10.043460845947266, -10.133840560913086, -10.500605583190918], "reading_logprob_std": 0.24069922935246563, "tokenization_path_logprob_std": 35.00000017857889, "trajectory_balance_rmse": 315.8991130195078} +{"canonical_length": 12, "canonical_nll": 0.7922300497690836, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.28571428571428575, "prompt_id": "a41c0de97fa957ce", "reading_canonical_log_prob": -9.506760597229004, "reading_distances": [20, 17, 24, 23], "reading_log_probs": [-10.197906494140625, -9.586118698120117, -10.350275039672852, -9.738357543945312], "reading_logprob_std": 0.36403300225205776, "tokenization_path_logprob_std": 18.733525998145257, "trajectory_balance_rmse": 210.77325871326693} +{"canonical_length": 9, "canonical_nll": 0.09712394740846422, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "916ba9265d9ad9c6", "reading_canonical_log_prob": -0.874115526676178, "reading_distances": [28, 36, 39, 32], "reading_log_probs": [-0.9534990191459656, -0.9982243180274963, -1.0510720014572144, -0.9182727932929993], "reading_logprob_std": 0.05743291729291577, "tokenization_path_logprob_std": 17.79599801508573, "trajectory_balance_rmse": 191.98547652649657} +{"canonical_length": 99, "canonical_nll": 0.7597077494919903, "generation_rouge_l_f1": 0.5136612021857925, "generation_token_f1": 0.6338797814207651, "prompt_id": "5baddab4b656da75", "reading_canonical_log_prob": -75.21106719970703, "reading_distances": [20, 25, 26, 30], "reading_log_probs": [-76.05669403076172, -75.08885192871094, -76.27159118652344, -76.44149017333984], "reading_logprob_std": 0.604726882700256, "tokenization_path_logprob_std": 104.83644295266018, "trajectory_balance_rmse": 1965.990138659432} +{"canonical_length": 54, "canonical_nll": 0.6574285295274522, "generation_rouge_l_f1": 0.4039408866995074, "generation_token_f1": 0.4236453201970444, "prompt_id": "22a09297862230a9", "reading_canonical_log_prob": -35.50114059448242, "reading_distances": [31, 27, 27, 34], "reading_log_probs": [-36.7639274597168, -37.20521545410156, -36.76029586791992, -38.904266357421875], "reading_logprob_std": 1.0188692977010247, "tokenization_path_logprob_std": 81.93963512396229, "trajectory_balance_rmse": 866.9852498206899} +{"canonical_length": 40, "canonical_nll": 0.21306970119476318, "generation_rouge_l_f1": 0.15720524017467247, "generation_token_f1": 0.19213973799126635, "prompt_id": "c6508f3625213115", "reading_canonical_log_prob": -8.522788047790527, "reading_distances": [10, 15, 15, 12], "reading_log_probs": [-8.06262493133545, -7.103915691375732, -7.092610836029053, -6.988400459289551], "reading_logprob_std": 0.5031848847953434, "tokenization_path_logprob_std": 41.32688093403453, "trajectory_balance_rmse": 244.86440477958513} +{"canonical_length": 272, "canonical_nll": 1.485611635095933, "generation_rouge_l_f1": 0.273921200750469, "generation_token_f1": 0.45778611632270166, "prompt_id": "a4b74ba3880275a6", "reading_canonical_log_prob": -404.08636474609375, "reading_distances": [20, 15, 19, 14], "reading_log_probs": [-402.71624755859375, -403.319091796875, -404.9504089355469, -403.1751708984375], "reading_logprob_std": 0.9746318871794162, "tokenization_path_logprob_std": 170.90550659114726, "trajectory_balance_rmse": 6175.8296213600615} +{"canonical_length": 17, "canonical_nll": 0.31773780373965993, "generation_rouge_l_f1": 0.10830324909747292, "generation_token_f1": 0.10830324909747292, "prompt_id": "5f38035dc4f5627c", "reading_canonical_log_prob": -5.401542663574219, "reading_distances": [20, 17, 20, 22], "reading_log_probs": [-6.085668563842773, -5.635015964508057, -6.070394515991211, -5.998819351196289], "reading_logprob_std": 0.2117180198272851, "tokenization_path_logprob_std": 31.314300084408533, "trajectory_balance_rmse": 434.21874663186134} +{"canonical_length": 34, "canonical_nll": 0.9685944949879366, "generation_rouge_l_f1": 0.14084507042253522, "generation_token_f1": 0.19718309859154926, "prompt_id": "0e027c7a776ba5aa", "reading_canonical_log_prob": -32.932212829589844, "reading_distances": [18, 25, 24, 27], "reading_log_probs": [-32.16961669921875, -32.719139099121094, -35.71367645263672, -32.90658950805664], "reading_logprob_std": 1.5886994696029482, "tokenization_path_logprob_std": 80.41467226403105, "trajectory_balance_rmse": 593.0530135355643} +{"canonical_length": 23, "canonical_nll": 2.7478380617888076, "generation_rouge_l_f1": 0.14814814814814814, "generation_token_f1": 0.2962962962962963, "prompt_id": "9d315888010aca18", "reading_canonical_log_prob": -63.20027542114258, "reading_distances": [19, 24, 26, 19], "reading_log_probs": [-63.49591827392578, -58.60700988769531, -59.753746032714844, -59.36513137817383], "reading_logprob_std": 2.179630167411894, "tokenization_path_logprob_std": 64.83772798003417, "trajectory_balance_rmse": 453.635709829911} +{"canonical_length": 390, "canonical_nll": 0.9820616893279247, "generation_rouge_l_f1": 0.2776911076443057, "generation_token_f1": 0.46801872074883, "prompt_id": "b3017655183fdef3", "reading_canonical_log_prob": -383.0040588378906, "reading_distances": [23, 22, 20, 23], "reading_log_probs": [-381.162109375, -380.2960510253906, -381.2397766113281, -379.2200927734375], "reading_logprob_std": 0.942290856185385, "tokenization_path_logprob_std": 254.1930563575549, "trajectory_balance_rmse": 7783.27883807533} +{"canonical_length": 24, "canonical_nll": 0.7402307192484537, "generation_rouge_l_f1": 0.6285714285714286, "generation_token_f1": 0.6285714285714286, "prompt_id": "70f7bc293fb9b6e2", "reading_canonical_log_prob": -17.76553726196289, "reading_distances": [36, 38, 32, 32], "reading_log_probs": [-18.18386459350586, -20.0456485748291, -21.380035400390625, -17.96622085571289], "reading_logprob_std": 1.6198727779573974, "tokenization_path_logprob_std": 55.17783726791563, "trajectory_balance_rmse": 437.3551165736009} +{"canonical_length": 85, "canonical_nll": 1.3256574742934284, "generation_rouge_l_f1": 0.1834862385321101, "generation_token_f1": 0.27522935779816515, "prompt_id": "6a1786b00d1d1907", "reading_canonical_log_prob": -112.6808853149414, "reading_distances": [30, 26, 28, 25], "reading_log_probs": [-111.28363800048828, -111.581298828125, -110.9487075805664, -112.41755676269531], "reading_logprob_std": 0.6287268255002755, "tokenization_path_logprob_std": 93.44367451745507, "trajectory_balance_rmse": 1333.632946748537} +{"canonical_length": 108, "canonical_nll": 1.3907320940936054, "generation_rouge_l_f1": 0.14906832298136646, "generation_token_f1": 0.31055900621118016, "prompt_id": "7988aedbe67fdb66", "reading_canonical_log_prob": -150.19906616210938, "reading_distances": [15, 15, 17, 16], "reading_log_probs": [-149.19827270507812, -152.12939453125, -153.0199432373047, -151.78390502929688], "reading_logprob_std": 1.641197880062647, "tokenization_path_logprob_std": 79.25386841183789, "trajectory_balance_rmse": 2336.684056258786} +{"canonical_length": 98, "canonical_nll": 2.2694672565070952, "generation_rouge_l_f1": 0.39106145251396646, "generation_token_f1": 0.5139664804469274, "prompt_id": "0b1fe757c0710416", "reading_canonical_log_prob": -222.4077911376953, "reading_distances": [12, 14, 16, 18], "reading_log_probs": [-217.28919982910156, -217.35501098632812, -219.704833984375, -217.16920471191406], "reading_logprob_std": 1.2192766369795651, "tokenization_path_logprob_std": 146.33709238338065, "trajectory_balance_rmse": 2210.7795558524085} +{"canonical_length": 13, "canonical_nll": 0.8028059739332932, "generation_rouge_l_f1": 0.11627906976744186, "generation_token_f1": 0.11627906976744186, "prompt_id": "78fa2b118c48dd30", "reading_canonical_log_prob": -10.436477661132812, "reading_distances": [22, 33, 29, 27], "reading_log_probs": [-9.182145118713379, -9.732612609863281, -9.904532432556152, -9.956106185913086], "reading_logprob_std": 0.35426441155822613, "tokenization_path_logprob_std": 31.81184694936705, "trajectory_balance_rmse": 256.5094533489726} +{"canonical_length": 424, "canonical_nll": 1.1236262771318544, "generation_rouge_l_f1": 0.22471910112359547, "generation_token_f1": 0.4622792937399679, "prompt_id": "995817411c15ddee", "reading_canonical_log_prob": -476.41754150390625, "reading_distances": [18, 16, 10, 10], "reading_log_probs": [-474.18682861328125, -474.96453857421875, -477.5870666503906, -477.3482360839844], "reading_logprob_std": 1.7023918481552869, "tokenization_path_logprob_std": 162.34930406186697, "trajectory_balance_rmse": 8198.995412358016} +{"canonical_length": 64, "canonical_nll": 1.254360556602478, "generation_rouge_l_f1": 0.273224043715847, "generation_token_f1": 0.44808743169398907, "prompt_id": "fd32715308ec2a93", "reading_canonical_log_prob": -80.2790756225586, "reading_distances": [37, 47, 42, 46], "reading_log_probs": [-79.51786041259766, -80.91680145263672, -81.63249206542969, -80.21940612792969], "reading_logprob_std": 0.9090404561170022, "tokenization_path_logprob_std": 195.77041643928797, "trajectory_balance_rmse": 1628.1649323231072} +{"canonical_length": 333, "canonical_nll": 1.2909480830928586, "generation_rouge_l_f1": 0.20332717190388172, "generation_token_f1": 0.3733826247689464, "prompt_id": "6c1da32736122296", "reading_canonical_log_prob": -429.8857116699219, "reading_distances": [35, 35, 36, 41], "reading_log_probs": [-431.84429931640625, -430.76300048828125, -428.9122314453125, -434.328369140625], "reading_logprob_std": 2.2621630367878525, "tokenization_path_logprob_std": 160.8597954149698, "trajectory_balance_rmse": 5684.103742174751} +{"canonical_length": 53, "canonical_nll": 0.7662772412570018, "generation_rouge_l_f1": 0.2923976608187135, "generation_token_f1": 0.33918128654970753, "prompt_id": "ffb0593302e5d193", "reading_canonical_log_prob": -40.612693786621094, "reading_distances": [40, 37, 35, 40], "reading_log_probs": [-39.30503463745117, -38.844932556152344, -39.93450927734375, -39.95404052734375], "reading_logprob_std": 0.5359431161861425, "tokenization_path_logprob_std": 108.38793745173152, "trajectory_balance_rmse": 934.4173934862863} +{"canonical_length": 142, "canonical_nll": 1.392455195037412, "generation_rouge_l_f1": 0.2670299727520436, "generation_token_f1": 0.46321525885558584, "prompt_id": "780311bcf262bb9d", "reading_canonical_log_prob": -197.7286376953125, "reading_distances": [35, 31, 31, 37], "reading_log_probs": [-191.8950653076172, -196.9464111328125, -193.05361938476562, -197.0359344482422], "reading_logprob_std": 2.6505903901911725, "tokenization_path_logprob_std": 159.8509347239232, "trajectory_balance_rmse": 3355.5311204071386} +{"canonical_length": 32, "canonical_nll": 1.750192403793335, "generation_rouge_l_f1": 0.13684210526315788, "generation_token_f1": 0.17894736842105263, "prompt_id": "ddc773449a2312dd", "reading_canonical_log_prob": -56.00615692138672, "reading_distances": [34, 28, 33, 31], "reading_log_probs": [-58.352657318115234, -54.960426330566406, -55.32026672363281, -56.47007369995117], "reading_logprob_std": 1.5269064924691835, "tokenization_path_logprob_std": 81.39698804689978, "trajectory_balance_rmse": 849.3910374626179} +{"canonical_length": 259, "canonical_nll": 0.6826827001387549, "generation_rouge_l_f1": 0.4130019120458891, "generation_token_f1": 0.6424474187380497, "prompt_id": "fa8dcc0f054c9984", "reading_canonical_log_prob": -176.8148193359375, "reading_distances": [21, 28, 27, 21], "reading_log_probs": [-174.03524780273438, -174.1194610595703, -173.77044677734375, -173.91815185546875], "reading_logprob_std": 0.15140345424805526, "tokenization_path_logprob_std": 185.7284824165155, "trajectory_balance_rmse": 3258.150146863437} +{"canonical_length": 79, "canonical_nll": 1.1211801842798161, "generation_rouge_l_f1": 0.27272727272727276, "generation_token_f1": 0.36363636363636365, "prompt_id": "2d9359e9dd847731", "reading_canonical_log_prob": -88.57323455810547, "reading_distances": [21, 13, 17, 16], "reading_log_probs": [-92.09428405761719, -92.08675384521484, -92.71599578857422, -94.6203384399414], "reading_logprob_std": 1.197533930580621, "tokenization_path_logprob_std": 168.4902617625758, "trajectory_balance_rmse": 1655.431294477658} +{"canonical_length": 91, "canonical_nll": 1.5738454965444713, "generation_rouge_l_f1": 0.14794520547945206, "generation_token_f1": 0.28493150684931506, "prompt_id": "5cb4ccbc96ab0eb7", "reading_canonical_log_prob": -143.21994018554688, "reading_distances": [27, 35, 32, 31], "reading_log_probs": [-143.9949951171875, -144.75059509277344, -142.5830535888672, -144.68475341796875], "reading_logprob_std": 1.0066441851079755, "tokenization_path_logprob_std": 92.90032644906887, "trajectory_balance_rmse": 2186.347920260494} +{"canonical_length": 76, "canonical_nll": 0.5074507060803866, "generation_rouge_l_f1": 0.5774647887323944, "generation_token_f1": 0.647887323943662, "prompt_id": "23716ea232c0bbd4", "reading_canonical_log_prob": -38.566253662109375, "reading_distances": [4, 4, 2, 5], "reading_log_probs": [-37.84721755981445, -37.52234649658203, -38.25764083862305, -37.60285949707031], "reading_logprob_std": 0.3303505784607953, "tokenization_path_logprob_std": 42.46952793712852, "trajectory_balance_rmse": 1050.8227123417164} +{"canonical_length": 7, "canonical_nll": 2.085645948137556, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "d26a8398a22a7919", "reading_canonical_log_prob": -14.59952163696289, "reading_distances": [39, 35, 40, 40], "reading_log_probs": [-14.142903327941895, -13.941286087036133, -14.224532127380371, -14.198685646057129], "reading_logprob_std": 0.1283145005751539, "tokenization_path_logprob_std": 23.811276800073532, "trajectory_balance_rmse": 204.69068155611154} +{"canonical_length": 177, "canonical_nll": 1.8946674583995409, "generation_rouge_l_f1": 0.1670428893905192, "generation_token_f1": 0.31151241534988716, "prompt_id": "cf319e0486bfdfdc", "reading_canonical_log_prob": -335.35614013671875, "reading_distances": [31, 27, 24, 18], "reading_log_probs": [-334.7933349609375, -335.360595703125, -336.29156494140625, -333.73138427734375], "reading_logprob_std": 1.0712055398611022, "tokenization_path_logprob_std": 191.58259693453337, "trajectory_balance_rmse": 4697.573798213183} +{"canonical_length": 8, "canonical_nll": 1.8747648000717163, "generation_rouge_l_f1": 0.19999999999999998, "generation_token_f1": 0.19999999999999998, "prompt_id": "2145da07c0ca4001", "reading_canonical_log_prob": -14.99811840057373, "reading_distances": [22, 22, 21, 25], "reading_log_probs": [-14.96540355682373, -15.622753143310547, -15.672908782958984, -16.01740264892578], "reading_logprob_std": 0.4393474382352258, "tokenization_path_logprob_std": 21.17007340837606, "trajectory_balance_rmse": 241.71903047496923} +{"canonical_length": 272, "canonical_nll": 1.2000425002154183, "generation_rouge_l_f1": 0.25708884688090744, "generation_token_f1": 0.4952741020793951, "prompt_id": "924cae405c1c753d", "reading_canonical_log_prob": -326.41156005859375, "reading_distances": [27, 22, 24, 21], "reading_log_probs": [-328.2065124511719, -327.9236145019531, -328.3896179199219, -331.1920166015625], "reading_logprob_std": 1.5215081261561016, "tokenization_path_logprob_std": 53.00628588321063, "trajectory_balance_rmse": 6202.077000571966} +{"canonical_length": 147, "canonical_nll": 1.2859494994286778, "generation_rouge_l_f1": 0.19851116625310172, "generation_token_f1": 0.401985111662531, "prompt_id": "eb989912613f5027", "reading_canonical_log_prob": -189.03457641601562, "reading_distances": [30, 30, 34, 36], "reading_log_probs": [-188.81512451171875, -189.63650512695312, -188.98097229003906, -190.83132934570312], "reading_logprob_std": 0.9150775459247734, "tokenization_path_logprob_std": 159.17574866171262, "trajectory_balance_rmse": 3329.382497420114} +{"canonical_length": 7, "canonical_nll": 1.681469508579799, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.888888888888889, "prompt_id": "1a61687e59394e25", "reading_canonical_log_prob": -11.770286560058594, "reading_distances": [19, 19, 20, 19], "reading_log_probs": [-13.849874496459961, -12.982954025268555, -12.262882232666016, -11.889025688171387], "reading_logprob_std": 0.8645796139717632, "tokenization_path_logprob_std": 14.299689886913136, "trajectory_balance_rmse": 72.35319508309607} +{"canonical_length": 12, "canonical_nll": 1.679092248280843, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.34782608695652173, "prompt_id": "687a259eb6971a8a", "reading_canonical_log_prob": -20.149106979370117, "reading_distances": [22, 21, 20, 22], "reading_log_probs": [-19.458477020263672, -19.412864685058594, -19.14860725402832, -18.053028106689453], "reading_logprob_std": 0.6578163107095155, "tokenization_path_logprob_std": 30.961269145369204, "trajectory_balance_rmse": 242.69413093120204} +{"canonical_length": 177, "canonical_nll": 1.1541129936606198, "generation_rouge_l_f1": 0.22857142857142856, "generation_token_f1": 0.3952380952380952, "prompt_id": "ae453e9c5b7b5add", "reading_canonical_log_prob": -204.2779998779297, "reading_distances": [27, 28, 24, 25], "reading_log_probs": [-203.45286560058594, -199.0944366455078, -200.1674346923828, -198.95346069335938], "reading_logprob_std": 2.095223449225264, "tokenization_path_logprob_std": 181.25927353455043, "trajectory_balance_rmse": 3394.8125243996174} +{"canonical_length": 5, "canonical_nll": 1.8641891479492188, "generation_rouge_l_f1": 0.015267175572519085, "generation_token_f1": 0.015267175572519085, "prompt_id": "ffdd99b122d4c9bc", "reading_canonical_log_prob": -9.320945739746094, "reading_distances": [42, 43, 43, 40], "reading_log_probs": [-10.789813995361328, -11.07456111907959, -11.999002456665039, -13.311574935913086], "reading_logprob_std": 1.135934454437126, "tokenization_path_logprob_std": 32.09590103596618, "trajectory_balance_rmse": 75.4940478404063} +{"canonical_length": 350, "canonical_nll": 1.0284316580636161, "generation_rouge_l_f1": 0.2605633802816901, "generation_token_f1": 0.5492957746478874, "prompt_id": "56f7657e17072dd3", "reading_canonical_log_prob": -359.9510803222656, "reading_distances": [16, 12, 11, 15], "reading_log_probs": [-355.85626220703125, -357.330078125, -357.3539733886719, -356.8419189453125], "reading_logprob_std": 0.7004677883797762, "tokenization_path_logprob_std": 283.3307369934127, "trajectory_balance_rmse": 7595.032360149585} +{"canonical_length": 172, "canonical_nll": 1.1946991321652434, "generation_rouge_l_f1": 0.3055555555555555, "generation_token_f1": 0.5462962962962963, "prompt_id": "f0c44899ac8f3454", "reading_canonical_log_prob": -205.48825073242188, "reading_distances": [26, 32, 27, 32], "reading_log_probs": [-201.20773315429688, -200.42727661132812, -200.87896728515625, -201.21485900878906], "reading_logprob_std": 0.37130220593601515, "tokenization_path_logprob_std": 227.09734177967488, "trajectory_balance_rmse": 3856.269815816546} +{"canonical_length": 316, "canonical_nll": 1.1181717884691456, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.4392857142857142, "prompt_id": "fb86a83720d2950d", "reading_canonical_log_prob": -353.34228515625, "reading_distances": [25, 22, 27, 26], "reading_log_probs": [-358.14410400390625, -356.70013427734375, -359.0355224609375, -359.4036560058594], "reading_logprob_std": 1.2029386481276887, "tokenization_path_logprob_std": 178.1055182887113, "trajectory_balance_rmse": 6111.018737216136} +{"canonical_length": 60, "canonical_nll": 1.4734814961751301, "generation_rouge_l_f1": 0.12101910828025476, "generation_token_f1": 0.20382165605095542, "prompt_id": "4f164749bc77ca14", "reading_canonical_log_prob": -88.40888977050781, "reading_distances": [26, 25, 19, 23], "reading_log_probs": [-87.86424255371094, -85.39592742919922, -86.2121353149414, -87.554931640625], "reading_logprob_std": 1.156442374742951, "tokenization_path_logprob_std": 59.677162152643525, "trajectory_balance_rmse": 1540.4235080372891} +{"canonical_length": 161, "canonical_nll": 1.3603173486934685, "generation_rouge_l_f1": 0.2816229116945107, "generation_token_f1": 0.4486873508353222, "prompt_id": "8d15f97380395bd9", "reading_canonical_log_prob": -219.01109313964844, "reading_distances": [26, 31, 27, 30], "reading_log_probs": [-218.2779541015625, -218.73545837402344, -219.09628295898438, -218.52020263671875], "reading_logprob_std": 0.347137650681882, "tokenization_path_logprob_std": 227.90715171804138, "trajectory_balance_rmse": 3218.617834113326} +{"canonical_length": 9, "canonical_nll": 0.9159992006089952, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.8571428571428571, "prompt_id": "a75869a5f1810362", "reading_canonical_log_prob": -8.243992805480957, "reading_distances": [44, 44, 46, 48], "reading_log_probs": [-11.988344192504883, -12.25425910949707, -12.138614654541016, -14.219383239746094], "reading_logprob_std": 1.0518044644633293, "tokenization_path_logprob_std": 19.236034290990528, "trajectory_balance_rmse": 157.09788046117703} +{"canonical_length": 228, "canonical_nll": 1.3958274439761513, "generation_rouge_l_f1": 0.21311475409836064, "generation_token_f1": 0.45901639344262296, "prompt_id": "f3ee3068b8bfca4b", "reading_canonical_log_prob": -318.2486572265625, "reading_distances": [22, 24, 28, 26], "reading_log_probs": [-317.0052490234375, -318.7168884277344, -316.65728759765625, -316.6201171875], "reading_logprob_std": 0.9932648181523124, "tokenization_path_logprob_std": 75.56963868032858, "trajectory_balance_rmse": 5157.111090253172} +{"canonical_length": 126, "canonical_nll": 1.3288181849888392, "generation_rouge_l_f1": 0.0851063829787234, "generation_token_f1": 0.1276595744680851, "prompt_id": "6a31e531aca13031", "reading_canonical_log_prob": -167.43109130859375, "reading_distances": [22, 30, 22, 21], "reading_log_probs": [-164.5591583251953, -164.875244140625, -165.03773498535156, -163.6483612060547], "reading_logprob_std": 0.6205173757273387, "tokenization_path_logprob_std": 167.31077347637557, "trajectory_balance_rmse": 2480.091941738451} +{"canonical_length": 11, "canonical_nll": 1.0224813981489702, "generation_rouge_l_f1": 0.1311475409836066, "generation_token_f1": 0.1311475409836066, "prompt_id": "01a0ba7b191885ca", "reading_canonical_log_prob": -11.247295379638672, "reading_distances": [12, 6, 7, 9], "reading_log_probs": [-11.410094261169434, -10.253183364868164, -11.752394676208496, -11.737142562866211], "reading_logprob_std": 0.7078473865703961, "tokenization_path_logprob_std": 29.16843162364549, "trajectory_balance_rmse": 183.89383621576687} +{"canonical_length": 9, "canonical_nll": 2.025467130872938, "generation_rouge_l_f1": 0.0321285140562249, "generation_token_f1": 0.0321285140562249, "prompt_id": "8b612a049e9d2b42", "reading_canonical_log_prob": -18.229204177856445, "reading_distances": [16, 11, 14, 13], "reading_log_probs": [-15.91193962097168, -15.730340957641602, -16.61279296875, -14.744422912597656], "reading_logprob_std": 0.7707588989716082, "tokenization_path_logprob_std": 49.331088229685136, "trajectory_balance_rmse": 173.11808282628547} +{"canonical_length": 168, "canonical_nll": 0.5630790619623094, "generation_rouge_l_f1": 0.2800982800982801, "generation_token_f1": 0.5012285012285013, "prompt_id": "1792c7fe245165d5", "reading_canonical_log_prob": -94.59728240966797, "reading_distances": [53, 51, 52, 52], "reading_log_probs": [-95.17559814453125, -94.75901794433594, -94.27261352539062, -94.32685852050781], "reading_logprob_std": 0.42185826814770777, "tokenization_path_logprob_std": 115.58952221714998, "trajectory_balance_rmse": 2447.249923200593} +{"canonical_length": 240, "canonical_nll": 1.4776561737060547, "generation_rouge_l_f1": 0.2948402948402949, "generation_token_f1": 0.5454545454545454, "prompt_id": "9148af04e8535f5f", "reading_canonical_log_prob": -354.6374816894531, "reading_distances": [24, 22, 23, 22], "reading_log_probs": [-348.60601806640625, -350.6846008300781, -350.05389404296875, -351.1732482910156], "reading_logprob_std": 1.1141904198451151, "tokenization_path_logprob_std": 176.11779983802808, "trajectory_balance_rmse": 4557.471138858659} +{"canonical_length": 11, "canonical_nll": 0.9305384375832297, "generation_rouge_l_f1": 0.09389671361502348, "generation_token_f1": 0.09389671361502348, "prompt_id": "4f964e48306df1f7", "reading_canonical_log_prob": -10.235922813415527, "reading_distances": [19, 16, 11, 12], "reading_log_probs": [-10.416431427001953, -9.551511764526367, -10.525564193725586, -9.523391723632812], "reading_logprob_std": 0.5409431906743908, "tokenization_path_logprob_std": 29.822662721904823, "trajectory_balance_rmse": 180.5906907687477} +{"canonical_length": 13, "canonical_nll": 1.5881720322829027, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.4444444444444444, "prompt_id": "65511a559eb975a4", "reading_canonical_log_prob": -20.646236419677734, "reading_distances": [22, 21, 18, 18], "reading_log_probs": [-21.68617057800293, -22.306777954101562, -22.105663299560547, -21.52968978881836], "reading_logprob_std": 0.360748354357845, "tokenization_path_logprob_std": 9.390352802188247, "trajectory_balance_rmse": 347.37758467342104} +{"canonical_length": 273, "canonical_nll": 1.057725843492445, "generation_rouge_l_f1": 0.28158844765342955, "generation_token_f1": 0.5487364620938628, "prompt_id": "dfd43bd8f8a6af69", "reading_canonical_log_prob": -288.7591552734375, "reading_distances": [16, 9, 16, 19], "reading_log_probs": [-287.57196044921875, -289.16876220703125, -288.45513916015625, -288.6324157714844], "reading_logprob_std": 0.6634944315646003, "tokenization_path_logprob_std": 262.92402758478204, "trajectory_balance_rmse": 4947.7977021529505} +{"canonical_length": 137, "canonical_nll": 1.768645015076129, "generation_rouge_l_f1": 0.17073170731707316, "generation_token_f1": 0.3170731707317073, "prompt_id": "9ebcabae9c8c462a", "reading_canonical_log_prob": -242.3043670654297, "reading_distances": [13, 16, 16, 21], "reading_log_probs": [-240.13890075683594, -236.64840698242188, -237.30853271484375, -236.751220703125], "reading_logprob_std": 1.6438734620513753, "tokenization_path_logprob_std": 227.26528187678113, "trajectory_balance_rmse": 3371.185360560462} +{"canonical_length": 224, "canonical_nll": 0.9899557658604213, "generation_rouge_l_f1": 0.18218623481781376, "generation_token_f1": 0.27530364372469635, "prompt_id": "4e90bb5af9e629db", "reading_canonical_log_prob": -221.75009155273438, "reading_distances": [57, 54, 44, 47], "reading_log_probs": [-220.444091796875, -218.10348510742188, -220.6341552734375, -222.1986846923828], "reading_logprob_std": 1.6885857855360644, "tokenization_path_logprob_std": 89.5245744812932, "trajectory_balance_rmse": 4540.4970853647865} +{"canonical_length": 17, "canonical_nll": 0.7542954052195829, "generation_rouge_l_f1": 0.10606060606060606, "generation_token_f1": 0.10606060606060606, "prompt_id": "3779413f4fba856c", "reading_canonical_log_prob": -12.82302188873291, "reading_distances": [12, 16, 21, 18], "reading_log_probs": [-13.287599563598633, -13.202421188354492, -14.027362823486328, -12.826177597045898], "reading_logprob_std": 0.5026885348718939, "tokenization_path_logprob_std": 79.97309706389433, "trajectory_balance_rmse": 461.95501844993805} +{"canonical_length": 125, "canonical_nll": 1.0926090087890625, "generation_rouge_l_f1": 0.34254143646408847, "generation_token_f1": 0.4419889502762431, "prompt_id": "f749a091b4c13176", "reading_canonical_log_prob": -136.5761260986328, "reading_distances": [18, 22, 15, 19], "reading_log_probs": [-133.6520233154297, -136.65325927734375, -136.77597045898438, -132.09490966796875], "reading_logprob_std": 2.307542077812297, "tokenization_path_logprob_std": 70.75800098799688, "trajectory_balance_rmse": 2727.370248467484} +{"canonical_length": 45, "canonical_nll": 1.519642130533854, "generation_rouge_l_f1": 0.3617021276595745, "generation_token_f1": 0.44680851063829785, "prompt_id": "6378420685c9b069", "reading_canonical_log_prob": -68.38389587402344, "reading_distances": [23, 19, 22, 21], "reading_log_probs": [-68.77828979492188, -68.57638549804688, -68.40422058105469, -68.866455078125], "reading_logprob_std": 0.20734594466606654, "tokenization_path_logprob_std": 151.67696034471996, "trajectory_balance_rmse": 1261.2492790238032} +{"canonical_length": 119, "canonical_nll": 0.8958246567670036, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.5526315789473685, "prompt_id": "634524f69adfb3f0", "reading_canonical_log_prob": -106.60313415527344, "reading_distances": [28, 25, 28, 25], "reading_log_probs": [-107.01408386230469, -108.53684997558594, -107.19156646728516, -108.23548126220703], "reading_logprob_std": 0.7545692610955295, "tokenization_path_logprob_std": 117.72242104699797, "trajectory_balance_rmse": 2166.9303895240814} +{"canonical_length": 162, "canonical_nll": 1.5671420627170138, "generation_rouge_l_f1": 0.1523809523809524, "generation_token_f1": 0.30000000000000004, "prompt_id": "c5c8a284983e4726", "reading_canonical_log_prob": -253.87701416015625, "reading_distances": [20, 22, 23, 19], "reading_log_probs": [-253.61239624023438, -254.5235137939453, -253.8895721435547, -254.4821014404297], "reading_logprob_std": 0.44889092773999467, "tokenization_path_logprob_std": 38.52232798010907, "trajectory_balance_rmse": 2994.8135415677125} +{"canonical_length": 88, "canonical_nll": 1.1512237028642134, "generation_rouge_l_f1": 0.23780487804878048, "generation_token_f1": 0.3292682926829268, "prompt_id": "626ba72af7ac67f2", "reading_canonical_log_prob": -101.30768585205078, "reading_distances": [31, 26, 29, 27], "reading_log_probs": [-96.98126983642578, -100.61444854736328, -99.56452178955078, -99.3344955444336], "reading_logprob_std": 1.5330902178194585, "tokenization_path_logprob_std": 61.864415887136715, "trajectory_balance_rmse": 1476.656273406821} +{"canonical_length": 297, "canonical_nll": 0.9005838002420034, "generation_rouge_l_f1": 0.2867383512544803, "generation_token_f1": 0.4874551971326165, "prompt_id": "c0257d516383b045", "reading_canonical_log_prob": -267.473388671875, "reading_distances": [26, 29, 34, 23], "reading_log_probs": [-265.81231689453125, -264.91845703125, -264.8460693359375, -266.7464904785156], "reading_logprob_std": 0.8927414177035539, "tokenization_path_logprob_std": 223.2627364766962, "trajectory_balance_rmse": 6284.814113741807} +{"canonical_length": 38, "canonical_nll": 0.44043671457391037, "generation_rouge_l_f1": 0.16666666666666666, "generation_token_f1": 0.2101449275362319, "prompt_id": "dae9d23ce12cb989", "reading_canonical_log_prob": -16.736595153808594, "reading_distances": [21, 14, 18, 13], "reading_log_probs": [-16.988805770874023, -15.684305191040039, -17.68621826171875, -17.64755630493164], "reading_logprob_std": 0.9347710877066878, "tokenization_path_logprob_std": 34.288393676027766, "trajectory_balance_rmse": 290.7568318869761} +{"canonical_length": 10, "canonical_nll": 0.542584753036499, "generation_rouge_l_f1": 0.06222222222222222, "generation_token_f1": 0.06222222222222222, "prompt_id": "cde09f27bb3893ac", "reading_canonical_log_prob": -5.42584753036499, "reading_distances": [29, 28, 22, 26], "reading_log_probs": [-5.5432353019714355, -5.203476905822754, -5.300348281860352, -5.698705196380615], "reading_logprob_std": 0.2258177943553076, "tokenization_path_logprob_std": 34.65662353853639, "trajectory_balance_rmse": 224.18898790605041} +{"canonical_length": 174, "canonical_nll": 1.6247099076194325, "generation_rouge_l_f1": 0.17494089834515367, "generation_token_f1": 0.4822695035460993, "prompt_id": "64bb51344f196fcb", "reading_canonical_log_prob": -282.69952392578125, "reading_distances": [27, 18, 20, 18], "reading_log_probs": [-287.0478820800781, -285.755615234375, -283.5249328613281, -287.778564453125], "reading_logprob_std": 1.8658481234743232, "tokenization_path_logprob_std": 168.23318117483774, "trajectory_balance_rmse": 3543.264855137779} +{"canonical_length": 312, "canonical_nll": 1.2745453272110376, "generation_rouge_l_f1": 0.2280701754385965, "generation_token_f1": 0.45964912280701753, "prompt_id": "bc3436553628cc88", "reading_canonical_log_prob": -397.65814208984375, "reading_distances": [37, 30, 30, 31], "reading_log_probs": [-400.2803649902344, -398.5277404785156, -401.0413818359375, -399.0703125], "reading_logprob_std": 1.140648863109508, "tokenization_path_logprob_std": 418.54819723034274, "trajectory_balance_rmse": 7485.026307259175} +{"canonical_length": 39, "canonical_nll": 1.2291995317508013, "generation_rouge_l_f1": 0.3300970873786408, "generation_token_f1": 0.48543689320388345, "prompt_id": "b3b9c2d52d21969e", "reading_canonical_log_prob": -47.93878173828125, "reading_distances": [27, 30, 29, 26], "reading_log_probs": [-45.43427276611328, -45.72711944580078, -45.372737884521484, -45.88158416748047], "reading_logprob_std": 0.24117825187833875, "tokenization_path_logprob_std": 18.8244765779151, "trajectory_balance_rmse": 960.7834146651663} +{"canonical_length": 21, "canonical_nll": 0.49412427629743305, "generation_rouge_l_f1": 0.08968609865470852, "generation_token_f1": 0.13452914798206275, "prompt_id": "cc11564e8f134b4b", "reading_canonical_log_prob": -10.376609802246094, "reading_distances": [15, 14, 14, 14], "reading_log_probs": [-11.189501762390137, -10.89277172088623, -11.811300277709961, -11.58697509765625], "reading_logprob_std": 0.40912528519779534, "tokenization_path_logprob_std": 16.117485637602055, "trajectory_balance_rmse": 168.1655074495996} +{"canonical_length": 20, "canonical_nll": 1.0491954803466796, "generation_rouge_l_f1": 0.08239700374531835, "generation_token_f1": 0.09737827715355805, "prompt_id": "e07efb6f1797853d", "reading_canonical_log_prob": -20.983909606933594, "reading_distances": [27, 28, 25, 24], "reading_log_probs": [-24.01296615600586, -24.91201400756836, -25.964736938476562, -26.84869384765625], "reading_logprob_std": 1.234888050170528, "tokenization_path_logprob_std": 44.80168334781158, "trajectory_balance_rmse": 373.6947086188238} +{"canonical_length": 10, "canonical_nll": 0.06371049880981446, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "302fe191c6bb7b5b", "reading_canonical_log_prob": -0.6371049880981445, "reading_distances": [24, 30, 26, 22], "reading_log_probs": [-1.2246239185333252, -1.6710848808288574, -1.6740214824676514, -0.7392866611480713], "reading_logprob_std": 0.44523622444973093, "tokenization_path_logprob_std": 32.86676419017872, "trajectory_balance_rmse": 177.2965929378738} +{"canonical_length": 22, "canonical_nll": 2.0097971829501065, "generation_rouge_l_f1": 0.6060606060606061, "generation_token_f1": 0.6060606060606061, "prompt_id": "de0e448f061c6724", "reading_canonical_log_prob": -44.215538024902344, "reading_distances": [32, 25, 34, 28], "reading_log_probs": [-43.175724029541016, -43.972923278808594, -42.834346771240234, -42.80180740356445], "reading_logprob_std": 0.5447328386453612, "tokenization_path_logprob_std": 36.179249590991326, "trajectory_balance_rmse": 665.7791818874142} +{"canonical_length": 79, "canonical_nll": 1.3979044081289558, "generation_rouge_l_f1": 0.2879377431906615, "generation_token_f1": 0.3735408560311284, "prompt_id": "f979fa68d4ceeb01", "reading_canonical_log_prob": -110.4344482421875, "reading_distances": [26, 27, 25, 24], "reading_log_probs": [-109.279541015625, -109.4613265991211, -109.70201873779297, -108.32024383544922], "reading_logprob_std": 0.605604998046239, "tokenization_path_logprob_std": 59.59731915864851, "trajectory_balance_rmse": 1729.7469975791971} +{"canonical_length": 193, "canonical_nll": 1.3803461104477008, "generation_rouge_l_f1": 0.20512820512820512, "generation_token_f1": 0.43589743589743585, "prompt_id": "e2f73d0c042e6f14", "reading_canonical_log_prob": -266.40679931640625, "reading_distances": [12, 13, 12, 16], "reading_log_probs": [-263.78839111328125, -265.34375, -263.6373291015625, -264.2838134765625], "reading_logprob_std": 0.7713993820420805, "tokenization_path_logprob_std": 313.922107989386, "trajectory_balance_rmse": 4593.093673893998} +{"canonical_length": 216, "canonical_nll": 1.645983236807364, "generation_rouge_l_f1": 0.273109243697479, "generation_token_f1": 0.5084033613445378, "prompt_id": "409538e18660e490", "reading_canonical_log_prob": -355.5323791503906, "reading_distances": [28, 38, 30, 31], "reading_log_probs": [-359.45172119140625, -358.82427978515625, -357.5791015625, -357.8956298828125], "reading_logprob_std": 0.8580431602578458, "tokenization_path_logprob_std": 77.68452188732743, "trajectory_balance_rmse": 5264.349028721517} +{"canonical_length": 64, "canonical_nll": 1.9417567253112793, "generation_rouge_l_f1": 0.1917808219178082, "generation_token_f1": 0.2191780821917808, "prompt_id": "cc38da16fec412a8", "reading_canonical_log_prob": -124.27243041992188, "reading_distances": [32, 31, 26, 30], "reading_log_probs": [-125.63587188720703, -123.31631469726562, -122.86033630371094, -125.41458892822266], "reading_logprob_std": 1.4220814262349981, "tokenization_path_logprob_std": 99.06382180666692, "trajectory_balance_rmse": 1262.861516509495} +{"canonical_length": 53, "canonical_nll": 0.7196910066424675, "generation_rouge_l_f1": 0.5346534653465347, "generation_token_f1": 0.6534653465346534, "prompt_id": "0e93241c78cc22bd", "reading_canonical_log_prob": -38.14362335205078, "reading_distances": [32, 28, 26, 35], "reading_log_probs": [-38.84383773803711, -38.708099365234375, -39.13280487060547, -38.72356414794922], "reading_logprob_std": 0.19674106097984204, "tokenization_path_logprob_std": 142.7466893268748, "trajectory_balance_rmse": 969.634809108608} +{"canonical_length": 19, "canonical_nll": 0.3933925377695184, "generation_rouge_l_f1": 0.21739130434782608, "generation_token_f1": 0.3260869565217392, "prompt_id": "32e3b4bcc84e208d", "reading_canonical_log_prob": -7.47445821762085, "reading_distances": [31, 28, 36, 29], "reading_log_probs": [-6.470606327056885, -6.32391357421875, -7.12056303024292, -6.278166770935059], "reading_logprob_std": 0.39023177365105893, "tokenization_path_logprob_std": 25.55312669672916, "trajectory_balance_rmse": 408.2940688919209} +{"canonical_length": 28, "canonical_nll": 1.4790751593453544, "generation_rouge_l_f1": 0.23255813953488372, "generation_token_f1": 0.32558139534883723, "prompt_id": "08748791c0916706", "reading_canonical_log_prob": -41.41410446166992, "reading_distances": [22, 19, 23, 21], "reading_log_probs": [-46.74452590942383, -48.029815673828125, -48.988189697265625, -45.67353057861328], "reading_logprob_std": 1.4517389934594989, "tokenization_path_logprob_std": 117.3821281380084, "trajectory_balance_rmse": 937.3904166567879} +{"canonical_length": 268, "canonical_nll": 1.5971861881996268, "generation_rouge_l_f1": 0.20183486238532108, "generation_token_f1": 0.3963302752293578, "prompt_id": "340800ab7642e87c", "reading_canonical_log_prob": -428.0458984375, "reading_distances": [20, 20, 25, 20], "reading_log_probs": [-429.69580078125, -430.4837341308594, -429.08135986328125, -427.5647888183594], "reading_logprob_std": 1.2358017522075266, "tokenization_path_logprob_std": 316.93199401615976, "trajectory_balance_rmse": 5257.75467833911} +{"canonical_length": 335, "canonical_nll": 1.238217390829058, "generation_rouge_l_f1": 0.22622950819672136, "generation_token_f1": 0.4557377049180328, "prompt_id": "ba9377f6ac227fec", "reading_canonical_log_prob": -414.8028259277344, "reading_distances": [21, 25, 30, 26], "reading_log_probs": [-411.7226257324219, -411.16986083984375, -412.31878662109375, -413.0068664550781], "reading_logprob_std": 0.7894251602979382, "tokenization_path_logprob_std": 76.53575023017675, "trajectory_balance_rmse": 7397.059522995046} +{"canonical_length": 12, "canonical_nll": 0.3087989290555318, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "39897df6fcbea057", "reading_canonical_log_prob": -3.705587148666382, "reading_distances": [22, 20, 26, 18], "reading_log_probs": [-3.380948066711426, -2.882669687271118, -3.957998752593994, -2.797238349914551], "reading_logprob_std": 0.5348661589233902, "tokenization_path_logprob_std": 62.205497206376585, "trajectory_balance_rmse": 237.04517714346701} +{"canonical_length": 55, "canonical_nll": 0.8041144631125711, "generation_rouge_l_f1": 0.24427480916030536, "generation_token_f1": 0.32061068702290074, "prompt_id": "d9f80153d939b659", "reading_canonical_log_prob": -44.226295471191406, "reading_distances": [21, 17, 22, 21], "reading_log_probs": [-45.62076187133789, -46.077972412109375, -46.505958557128906, -45.83601760864258], "reading_logprob_std": 0.3796363199693702, "tokenization_path_logprob_std": 61.44391708845838, "trajectory_balance_rmse": 1183.22978967728} +{"canonical_length": 105, "canonical_nll": 1.3704833984375, "generation_rouge_l_f1": 0.2395543175487465, "generation_token_f1": 0.3398328690807799, "prompt_id": "ea105e2e33c793e5", "reading_canonical_log_prob": -143.9007568359375, "reading_distances": [34, 44, 32, 30], "reading_log_probs": [-148.98348999023438, -149.74087524414062, -149.45274353027344, -146.36575317382812], "reading_logprob_std": 1.5451633053519913, "tokenization_path_logprob_std": 76.99157198087829, "trajectory_balance_rmse": 2504.6032488161054} +{"canonical_length": 311, "canonical_nll": 1.4492090354011757, "generation_rouge_l_f1": 0.22099447513812154, "generation_token_f1": 0.4825046040515654, "prompt_id": "2fb44fa5d32cf514", "reading_canonical_log_prob": -450.7040100097656, "reading_distances": [33, 30, 30, 31], "reading_log_probs": [-458.7381591796875, -456.1831359863281, -458.17010498046875, -457.6793518066406], "reading_logprob_std": 1.095424133510426, "tokenization_path_logprob_std": 182.6264904745462, "trajectory_balance_rmse": 6577.834388961733} +{"canonical_length": 134, "canonical_nll": 0.9738031643540112, "generation_rouge_l_f1": 0.19559902200488996, "generation_token_f1": 0.3178484107579462, "prompt_id": "f67931e5043d993b", "reading_canonical_log_prob": -130.4896240234375, "reading_distances": [27, 26, 23, 29], "reading_log_probs": [-132.40814208984375, -132.90957641601562, -131.7470703125, -132.83441162109375], "reading_logprob_std": 0.5330368420676207, "tokenization_path_logprob_std": 214.23954160690937, "trajectory_balance_rmse": 2774.9960633102282} +{"canonical_length": 42, "canonical_nll": 1.4906625293550038, "generation_rouge_l_f1": 0.08974358974358973, "generation_token_f1": 0.10256410256410256, "prompt_id": "f4dd2f85181ef8ef", "reading_canonical_log_prob": -62.607826232910156, "reading_distances": [23, 21, 17, 16], "reading_log_probs": [-59.936702728271484, -61.078887939453125, -61.06008529663086, -60.85620880126953], "reading_logprob_std": 0.5403372429276732, "tokenization_path_logprob_std": 42.58213389563961, "trajectory_balance_rmse": 1091.953056446833} +{"canonical_length": 386, "canonical_nll": 0.8608278264653497, "generation_rouge_l_f1": 0.26991869918699185, "generation_token_f1": 0.40975609756097553, "prompt_id": "044b1de06d0cdd9b", "reading_canonical_log_prob": -332.279541015625, "reading_distances": [44, 47, 39, 43], "reading_log_probs": [-328.82147216796875, -329.812255859375, -329.2151794433594, -329.78668212890625], "reading_logprob_std": 0.47889261082074525, "tokenization_path_logprob_std": 288.8066508333332, "trajectory_balance_rmse": 7024.565360770537} +{"canonical_length": 120, "canonical_nll": 0.7791748046875, "generation_rouge_l_f1": 0.40160642570281124, "generation_token_f1": 0.5461847389558233, "prompt_id": "c7dd76b068a7ae00", "reading_canonical_log_prob": -93.5009765625, "reading_distances": [41, 43, 39, 47], "reading_log_probs": [-95.4565658569336, -94.87147521972656, -94.88707733154297, -95.05282592773438], "reading_logprob_std": 0.2723750886578098, "tokenization_path_logprob_std": 240.9641896526844, "trajectory_balance_rmse": 2273.2407936494983} +{"canonical_length": 35, "canonical_nll": 2.3536189488002233, "generation_rouge_l_f1": 0.3582089552238806, "generation_token_f1": 0.3582089552238806, "prompt_id": "2fa90c11cfdc9fd0", "reading_canonical_log_prob": -82.37666320800781, "reading_distances": [20, 19, 24, 27], "reading_log_probs": [-82.8714599609375, -85.76484680175781, -84.25447082519531, -83.97074127197266], "reading_logprob_std": 1.1928040689915365, "tokenization_path_logprob_std": 8.278644045281295, "trajectory_balance_rmse": 728.1193490150253} +{"canonical_length": 33, "canonical_nll": 1.2554707382664536, "generation_rouge_l_f1": 0.5490196078431372, "generation_token_f1": 0.5882352941176471, "prompt_id": "acac7a68437d400e", "reading_canonical_log_prob": -41.43053436279297, "reading_distances": [24, 27, 34, 27], "reading_log_probs": [-41.861236572265625, -43.12994384765625, -41.738304138183594, -44.07421875], "reading_logprob_std": 1.1108117864435616, "tokenization_path_logprob_std": 78.46116369230556, "trajectory_balance_rmse": 572.2970818934657} +{"canonical_length": 41, "canonical_nll": 0.49475041831411964, "generation_rouge_l_f1": 0.11920529801324503, "generation_token_f1": 0.17880794701986755, "prompt_id": "98a669e9058b6ddc", "reading_canonical_log_prob": -20.284767150878906, "reading_distances": [12, 12, 9, 12], "reading_log_probs": [-19.95912742614746, -21.003582000732422, -20.87131118774414, -21.07587432861328], "reading_logprob_std": 0.5191865468130881, "tokenization_path_logprob_std": 59.6767714047517, "trajectory_balance_rmse": 583.1383955012182} +{"canonical_length": 133, "canonical_nll": 1.3929971106966634, "generation_rouge_l_f1": 0.26455026455026454, "generation_token_f1": 0.43386243386243384, "prompt_id": "610adc3864696d51", "reading_canonical_log_prob": -185.26861572265625, "reading_distances": [21, 19, 21, 21], "reading_log_probs": [-185.77743530273438, -186.80587768554688, -186.09010314941406, -186.23284912109375], "reading_logprob_std": 0.43049661223513724, "tokenization_path_logprob_std": 211.89136319265938, "trajectory_balance_rmse": 3536.5109205225954} +{"canonical_length": 34, "canonical_nll": 0.6138545765596277, "generation_rouge_l_f1": 0.13793103448275862, "generation_token_f1": 0.18620689655172415, "prompt_id": "74b66fda46123ad6", "reading_canonical_log_prob": -20.871055603027344, "reading_distances": [38, 42, 44, 39], "reading_log_probs": [-19.948314666748047, -19.209964752197266, -20.768550872802734, -20.95121955871582], "reading_logprob_std": 0.8019919522535857, "tokenization_path_logprob_std": 63.14593473228293, "trajectory_balance_rmse": 692.5020607350181} +{"canonical_length": 23, "canonical_nll": 0.7699822964875594, "generation_rouge_l_f1": 0.33070866141732286, "generation_token_f1": 0.33070866141732286, "prompt_id": "9ce2e67b60ca246c", "reading_canonical_log_prob": -17.709592819213867, "reading_distances": [19, 18, 17, 19], "reading_log_probs": [-17.517736434936523, -16.816566467285156, -17.30219268798828, -17.148235321044922], "reading_logprob_std": 0.2949815572168158, "tokenization_path_logprob_std": 78.98506320226863, "trajectory_balance_rmse": 438.27355536258017} +{"canonical_length": 9, "canonical_nll": 1.2927151785956488, "generation_rouge_l_f1": 0.5555555555555556, "generation_token_f1": 0.5555555555555556, "prompt_id": "f37797ee8ed6a36e", "reading_canonical_log_prob": -11.63443660736084, "reading_distances": [42, 32, 42, 48], "reading_log_probs": [-10.461441040039062, -9.946821212768555, -11.974794387817383, -10.159358024597168], "reading_logprob_std": 0.9174231282089516, "tokenization_path_logprob_std": 20.403983773951683, "trajectory_balance_rmse": 181.38731518683835} +{"canonical_length": 25, "canonical_nll": 1.1796452331542968, "generation_rouge_l_f1": 0.18897637795275588, "generation_token_f1": 0.2834645669291339, "prompt_id": "b9be7345d7f0f09d", "reading_canonical_log_prob": -29.491130828857422, "reading_distances": [29, 23, 24, 28], "reading_log_probs": [-28.110095977783203, -28.091537475585938, -28.450611114501953, -28.493694305419922], "reading_logprob_std": 0.21524459121935868, "tokenization_path_logprob_std": 17.00713783731465, "trajectory_balance_rmse": 601.3356680787488} +{"canonical_length": 102, "canonical_nll": 0.8005725636201746, "generation_rouge_l_f1": 0.3525641025641026, "generation_token_f1": 0.4743589743589744, "prompt_id": "101280c93f2c70da", "reading_canonical_log_prob": -81.65840148925781, "reading_distances": [18, 16, 10, 18], "reading_log_probs": [-84.84671020507812, -83.1643295288086, -83.17294311523438, -82.91590118408203], "reading_logprob_std": 0.8891843583304428, "tokenization_path_logprob_std": 91.68651406732204, "trajectory_balance_rmse": 932.0884395714197} +{"canonical_length": 309, "canonical_nll": 1.099853120575445, "generation_rouge_l_f1": 0.288659793814433, "generation_token_f1": 0.5257731958762886, "prompt_id": "bc039a3de3397314", "reading_canonical_log_prob": -339.8546142578125, "reading_distances": [35, 38, 27, 35], "reading_log_probs": [-341.91748046875, -341.71295166015625, -340.7906494140625, -340.572021484375], "reading_logprob_std": 0.6659580447736918, "tokenization_path_logprob_std": 323.9645369428632, "trajectory_balance_rmse": 7213.538108060617} +{"canonical_length": 144, "canonical_nll": 1.0256995095147028, "generation_rouge_l_f1": 0.2663043478260869, "generation_token_f1": 0.38586956521739135, "prompt_id": "3bf14cf65dcd8fce", "reading_canonical_log_prob": -147.7007293701172, "reading_distances": [35, 36, 35, 39], "reading_log_probs": [-146.65872192382812, -146.62564086914062, -146.8072509765625, -147.7626953125], "reading_logprob_std": 0.5385676525081138, "tokenization_path_logprob_std": 184.55388471974902, "trajectory_balance_rmse": 2436.8810167505303} +{"canonical_length": 38, "canonical_nll": 1.5527034558747943, "generation_rouge_l_f1": 0.08108108108108109, "generation_token_f1": 0.1081081081081081, "prompt_id": "ab60c5be348f9273", "reading_canonical_log_prob": -59.00273132324219, "reading_distances": [15, 20, 25, 22], "reading_log_probs": [-59.06083297729492, -58.29168701171875, -58.395713806152344, -58.31253433227539], "reading_logprob_std": 0.36652590858065837, "tokenization_path_logprob_std": 108.56135857674198, "trajectory_balance_rmse": 1024.944955923252} +{"canonical_length": 12, "canonical_nll": 2.692002614339193, "generation_rouge_l_f1": 0.4705882352941177, "generation_token_f1": 0.4705882352941177, "prompt_id": "8392d28ac48cc769", "reading_canonical_log_prob": -32.30403137207031, "reading_distances": [38, 35, 33, 32], "reading_log_probs": [-33.417362213134766, -35.94242858886719, -34.359352111816406, -35.590049743652344], "reading_logprob_std": 1.1593387094982328, "tokenization_path_logprob_std": 69.39343359955458, "trajectory_balance_rmse": 368.7290273161431} +{"canonical_length": 90, "canonical_nll": 1.8878716362847223, "generation_rouge_l_f1": 0.1643835616438356, "generation_token_f1": 0.28767123287671237, "prompt_id": "2a39144255d482e3", "reading_canonical_log_prob": -169.908447265625, "reading_distances": [29, 29, 25, 23], "reading_log_probs": [-177.74960327148438, -177.15493774414062, -175.91387939453125, -183.5814208984375], "reading_logprob_std": 3.4078917034532945, "tokenization_path_logprob_std": 159.70384357038066, "trajectory_balance_rmse": 2017.1167820118833} +{"canonical_length": 292, "canonical_nll": 1.375071695406143, "generation_rouge_l_f1": 0.216998191681736, "generation_token_f1": 0.4810126582278481, "prompt_id": "7938846720b7f660", "reading_canonical_log_prob": -401.52093505859375, "reading_distances": [20, 13, 22, 17], "reading_log_probs": [-404.30865478515625, -401.938720703125, -402.2784423828125, -402.7588806152344], "reading_logprob_std": 1.0471804972486172, "tokenization_path_logprob_std": 264.72505779595804, "trajectory_balance_rmse": 5957.775533685044} +{"canonical_length": 196, "canonical_nll": 1.3934134658502073, "generation_rouge_l_f1": 0.27865168539325846, "generation_token_f1": 0.5752808988764045, "prompt_id": "2a7c37e2ef87098f", "reading_canonical_log_prob": -273.1090393066406, "reading_distances": [27, 30, 28, 31], "reading_log_probs": [-274.8376770019531, -274.353515625, -273.4328918457031, -275.7379150390625], "reading_logprob_std": 0.9615743991063185, "tokenization_path_logprob_std": 144.48370020307868, "trajectory_balance_rmse": 4667.345285583347} +{"canonical_length": 284, "canonical_nll": 1.4168774242132482, "generation_rouge_l_f1": 0.2610294117647059, "generation_token_f1": 0.463235294117647, "prompt_id": "f9f1a4c90d1c6936", "reading_canonical_log_prob": -402.3931884765625, "reading_distances": [27, 25, 28, 28], "reading_log_probs": [-399.56829833984375, -400.799072265625, -402.9710998535156, -401.6505432128906], "reading_logprob_std": 1.43225296784631, "tokenization_path_logprob_std": 214.83479978078722, "trajectory_balance_rmse": 6608.409655867903} +{"canonical_length": 149, "canonical_nll": 1.2601118663813444, "generation_rouge_l_f1": 0.2666666666666667, "generation_token_f1": 0.4380952380952381, "prompt_id": "4c7fd497c60d26bd", "reading_canonical_log_prob": -187.7566680908203, "reading_distances": [20, 22, 27, 23], "reading_log_probs": [-195.48379516601562, -187.8150177001953, -193.29946899414062, -188.94818115234375], "reading_logprob_std": 3.6123936024496324, "tokenization_path_logprob_std": 121.22402056405272, "trajectory_balance_rmse": 3113.234846344877} +{"canonical_length": 25, "canonical_nll": 1.0731486511230468, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.3333333333333333, "prompt_id": "c2e4f5be5a2ad0b2", "reading_canonical_log_prob": -26.828716278076172, "reading_distances": [13, 17, 20, 16], "reading_log_probs": [-26.477840423583984, -27.17061424255371, -27.611083984375, -26.21384048461914], "reading_logprob_std": 0.6387221301839058, "tokenization_path_logprob_std": 105.60300639962095, "trajectory_balance_rmse": 534.4958980804374} +{"canonical_length": 14, "canonical_nll": 1.3667118889944894, "generation_rouge_l_f1": 0.2608695652173913, "generation_token_f1": 0.2608695652173913, "prompt_id": "619e2fb6adbffc57", "reading_canonical_log_prob": -19.13396644592285, "reading_distances": [36, 30, 30, 30], "reading_log_probs": [-19.91805648803711, -20.345922470092773, -20.245262145996094, -20.159761428833008], "reading_logprob_std": 0.18272313445290875, "tokenization_path_logprob_std": 57.73811852646059, "trajectory_balance_rmse": 375.45241722708454} +{"canonical_length": 136, "canonical_nll": 1.338028627283433, "generation_rouge_l_f1": 0.22959183673469388, "generation_token_f1": 0.4693877551020408, "prompt_id": "237ba288f6603bf1", "reading_canonical_log_prob": -181.97189331054688, "reading_distances": [18, 17, 18, 21], "reading_log_probs": [-182.29013061523438, -181.67628479003906, -183.39755249023438, -181.7362060546875], "reading_logprob_std": 0.7977284081331771, "tokenization_path_logprob_std": 154.61895359702368, "trajectory_balance_rmse": 2876.567270843921} +{"canonical_length": 9, "canonical_nll": 1.1204074223836262, "generation_rouge_l_f1": 0.39999999999999997, "generation_token_f1": 0.39999999999999997, "prompt_id": "52afc3f6a37f4218", "reading_canonical_log_prob": -10.083666801452637, "reading_distances": [27, 26, 24, 22], "reading_log_probs": [-8.387955665588379, -8.526360511779785, -8.528453826904297, -8.176969528198242], "reading_logprob_std": 0.1655874488415929, "tokenization_path_logprob_std": 30.08034315474147, "trajectory_balance_rmse": 234.96314798094716} +{"canonical_length": 115, "canonical_nll": 1.517487899116848, "generation_rouge_l_f1": 0.21108179419525067, "generation_token_f1": 0.34828496042216356, "prompt_id": "0096b48ae77be6ff", "reading_canonical_log_prob": -174.5111083984375, "reading_distances": [35, 31, 30, 25], "reading_log_probs": [-186.58277893066406, -183.62957763671875, -183.02679443359375, -192.01275634765625], "reading_logprob_std": 4.105277323006297, "tokenization_path_logprob_std": 88.72134897993864, "trajectory_balance_rmse": 2926.1646720697718} +{"canonical_length": 252, "canonical_nll": 1.089722648499504, "generation_rouge_l_f1": 0.32113821138211385, "generation_token_f1": 0.540650406504065, "prompt_id": "2061c66b7116786e", "reading_canonical_log_prob": -274.610107421875, "reading_distances": [31, 38, 33, 29], "reading_log_probs": [-270.27642822265625, -269.18798828125, -269.7763977050781, -271.8050842285156], "reading_logprob_std": 1.1211046288679103, "tokenization_path_logprob_std": 193.5619814416416, "trajectory_balance_rmse": 5222.719775819711} +{"canonical_length": 94, "canonical_nll": 1.4214117171916556, "generation_rouge_l_f1": 0.22580645161290322, "generation_token_f1": 0.3483870967741935, "prompt_id": "5622ae2789dbdaf4", "reading_canonical_log_prob": -133.61270141601562, "reading_distances": [18, 18, 16, 17], "reading_log_probs": [-132.0923309326172, -131.8538818359375, -132.12875366210938, -131.3050537109375], "reading_logprob_std": 0.3800480728233967, "tokenization_path_logprob_std": 72.497666080359, "trajectory_balance_rmse": 2036.2209785788525} +{"canonical_length": 279, "canonical_nll": 1.223971240409386, "generation_rouge_l_f1": 0.2462686567164179, "generation_token_f1": 0.5, "prompt_id": "531c2e72ca7ad121", "reading_canonical_log_prob": -341.48797607421875, "reading_distances": [27, 16, 22, 25], "reading_log_probs": [-343.7203369140625, -340.64984130859375, -342.505126953125, -339.87677001953125], "reading_logprob_std": 1.7470372347741585, "tokenization_path_logprob_std": 115.3583648562839, "trajectory_balance_rmse": 5957.8581147678} +{"canonical_length": 33, "canonical_nll": 0.7033474084102747, "generation_rouge_l_f1": 0.6909090909090909, "generation_token_f1": 0.7636363636363637, "prompt_id": "50acd3d956ee080e", "reading_canonical_log_prob": -23.210464477539062, "reading_distances": [26, 23, 37, 36], "reading_log_probs": [-22.133275985717773, -21.66590690612793, -21.77090072631836, -21.057575225830078], "reading_logprob_std": 0.4469159331074339, "tokenization_path_logprob_std": 51.394950243040824, "trajectory_balance_rmse": 724.2554639729931} +{"canonical_length": 26, "canonical_nll": 1.442614041841947, "generation_rouge_l_f1": 0.2191780821917808, "generation_token_f1": 0.24657534246575347, "prompt_id": "3fcc8e674911ba73", "reading_canonical_log_prob": -37.507965087890625, "reading_distances": [16, 15, 15, 15], "reading_log_probs": [-35.37324523925781, -36.80755615234375, -37.61418151855469, -33.75225830078125], "reading_logprob_std": 1.6981995917523176, "tokenization_path_logprob_std": 46.80664149649874, "trajectory_balance_rmse": 561.7292478477684} +{"canonical_length": 347, "canonical_nll": 1.0122971768337987, "generation_rouge_l_f1": 0.2016260162601626, "generation_token_f1": 0.383739837398374, "prompt_id": "915d77693c098234", "reading_canonical_log_prob": -351.2671203613281, "reading_distances": [17, 13, 11, 17], "reading_log_probs": [-348.2861328125, -345.751708984375, -349.928466796875, -346.8229675292969], "reading_logprob_std": 1.814259355266126, "tokenization_path_logprob_std": 212.54269764586655, "trajectory_balance_rmse": 6108.024102836586} +{"canonical_length": 16, "canonical_nll": 0.1836235523223877, "generation_rouge_l_f1": 0.9333333333333333, "generation_token_f1": 0.9333333333333333, "prompt_id": "d4ba094ba07a6bba", "reading_canonical_log_prob": -2.937976837158203, "reading_distances": [21, 21, 28, 21], "reading_log_probs": [-2.0328781604766846, -2.029406785964966, -2.153503894805908, -2.4863200187683105], "reading_logprob_std": 0.21507915290602309, "tokenization_path_logprob_std": 47.077882553971705, "trajectory_balance_rmse": 270.8392794091019} +{"canonical_length": 35, "canonical_nll": 0.950351061139788, "generation_rouge_l_f1": 0.4313725490196078, "generation_token_f1": 0.47058823529411764, "prompt_id": "40f13ac2314179d4", "reading_canonical_log_prob": -33.26228713989258, "reading_distances": [20, 25, 28, 30], "reading_log_probs": [-34.048675537109375, -33.65160369873047, -33.593894958496094, -34.78843307495117], "reading_logprob_std": 0.550330603312911, "tokenization_path_logprob_std": 101.99505411133204, "trajectory_balance_rmse": 834.794455692342} +{"canonical_length": 228, "canonical_nll": 1.1213497362638776, "generation_rouge_l_f1": 0.3031578947368421, "generation_token_f1": 0.5642105263157895, "prompt_id": "c60f4d0722caa17f", "reading_canonical_log_prob": -255.66773986816406, "reading_distances": [62, 53, 67, 59], "reading_log_probs": [-246.72921752929688, -247.8370819091797, -246.8842010498047, -249.1087646484375], "reading_logprob_std": 1.094968416359738, "tokenization_path_logprob_std": 185.0013221326185, "trajectory_balance_rmse": 4992.105879465824} +{"canonical_length": 20, "canonical_nll": 2.1478099822998047, "generation_rouge_l_f1": 0.16969696969696968, "generation_token_f1": 0.20606060606060606, "prompt_id": "2ca8b957d3b2b5bc", "reading_canonical_log_prob": -42.956199645996094, "reading_distances": [24, 23, 30, 30], "reading_log_probs": [-45.69554138183594, -46.989646911621094, -46.117984771728516, -45.65231704711914], "reading_logprob_std": 0.620492302218566, "tokenization_path_logprob_std": 60.14239915118973, "trajectory_balance_rmse": 486.7203303791696} +{"canonical_length": 159, "canonical_nll": 1.2268964659492925, "generation_rouge_l_f1": 0.2956120092378753, "generation_token_f1": 0.5127020785219399, "prompt_id": "762bfb8c99f50dd6", "reading_canonical_log_prob": -195.0765380859375, "reading_distances": [22, 20, 31, 20], "reading_log_probs": [-196.46566772460938, -197.0849609375, -195.1263427734375, -196.03770446777344], "reading_logprob_std": 0.8227990500646828, "tokenization_path_logprob_std": 89.41534115893778, "trajectory_balance_rmse": 3140.9581252799476} +{"canonical_length": 231, "canonical_nll": 1.0374810024773404, "generation_rouge_l_f1": 0.2454728370221328, "generation_token_f1": 0.482897384305835, "prompt_id": "9f2ef447478c22f4", "reading_canonical_log_prob": -239.65811157226562, "reading_distances": [29, 27, 19, 24], "reading_log_probs": [-241.11102294921875, -239.387451171875, -239.43508911132812, -237.81153869628906], "reading_logprob_std": 1.3474585859476944, "tokenization_path_logprob_std": 177.51057370776613, "trajectory_balance_rmse": 5085.844031752195} +{"canonical_length": 361, "canonical_nll": 1.1970058451729138, "generation_rouge_l_f1": 0.22297297297297294, "generation_token_f1": 0.46283783783783783, "prompt_id": "dc0376ea25a8c1bc", "reading_canonical_log_prob": -432.1191101074219, "reading_distances": [27, 26, 30, 30], "reading_log_probs": [-432.351806640625, -429.49481201171875, -427.1922607421875, -431.18084716796875], "reading_logprob_std": 2.2399359523422238, "tokenization_path_logprob_std": 221.63508111893344, "trajectory_balance_rmse": 6442.49962196706} +{"canonical_length": 18, "canonical_nll": 3.0441120995415583, "generation_rouge_l_f1": 0.3125, "generation_token_f1": 0.375, "prompt_id": "9cf805580809e94d", "reading_canonical_log_prob": -54.79401779174805, "reading_distances": [31, 22, 23, 20], "reading_log_probs": [-57.54280090332031, -56.62348175048828, -57.94046401977539, -57.11030960083008], "reading_logprob_std": 0.5664900998297717, "tokenization_path_logprob_std": 28.55777406198148, "trajectory_balance_rmse": 410.77216562805114} +{"canonical_length": 280, "canonical_nll": 1.6158040727887835, "generation_rouge_l_f1": 0.24444444444444444, "generation_token_f1": 0.5148148148148147, "prompt_id": "d3ad116b6d046506", "reading_canonical_log_prob": -452.4251403808594, "reading_distances": [24, 22, 22, 24], "reading_log_probs": [-450.570068359375, -451.7412414550781, -451.9830017089844, -452.64825439453125], "reading_logprob_std": 0.8665336065606076, "tokenization_path_logprob_std": 107.56446332199961, "trajectory_balance_rmse": 6970.52514230927} +{"canonical_length": 141, "canonical_nll": 1.2345827792553192, "generation_rouge_l_f1": 0.32105263157894737, "generation_token_f1": 0.4842105263157895, "prompt_id": "d4c38137aba16167", "reading_canonical_log_prob": -174.076171875, "reading_distances": [12, 14, 9, 16], "reading_log_probs": [-184.12770080566406, -196.09078979492188, -182.490966796875, -187.0397186279297], "reading_logprob_std": 6.067965489947367, "tokenization_path_logprob_std": 34.98724337607487, "trajectory_balance_rmse": 1438.376254283185} +{"canonical_length": 41, "canonical_nll": 1.7641277778439406, "generation_rouge_l_f1": 0.17721518987341772, "generation_token_f1": 0.27848101265822783, "prompt_id": "84547b634d24f63a", "reading_canonical_log_prob": -72.32923889160156, "reading_distances": [24, 28, 29, 24], "reading_log_probs": [-71.4807357788086, -71.09384155273438, -70.32106018066406, -71.24981689453125], "reading_logprob_std": 0.5026568209721753, "tokenization_path_logprob_std": 166.42117591858803, "trajectory_balance_rmse": 1061.1271475898457} +{"canonical_length": 6, "canonical_nll": 0.12074508269627889, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "58c9afa198f60472", "reading_canonical_log_prob": -0.7244704961776733, "reading_distances": [43, 40, 52, 40], "reading_log_probs": [-2.167353868484497, -1.3440011739730835, -1.4139468669891357, -1.517350435256958], "reading_logprob_std": 0.37789668969162976, "tokenization_path_logprob_std": 9.158387897995373, "trajectory_balance_rmse": 60.424344016275136} +{"canonical_length": 238, "canonical_nll": 1.5493102514443278, "generation_rouge_l_f1": 0.2593320235756385, "generation_token_f1": 0.5108055009823183, "prompt_id": "74608c6aad72d41d", "reading_canonical_log_prob": -368.73583984375, "reading_distances": [37, 39, 36, 32], "reading_log_probs": [-369.747314453125, -373.1445007324219, -374.42596435546875, -372.8437805175781], "reading_logprob_std": 1.984410680255483, "tokenization_path_logprob_std": 166.75976898976958, "trajectory_balance_rmse": 5309.028703089837} +{"canonical_length": 296, "canonical_nll": 1.3919584944441512, "generation_rouge_l_f1": 0.24444444444444446, "generation_token_f1": 0.48148148148148157, "prompt_id": "5c242c11a7bd0e44", "reading_canonical_log_prob": -412.01971435546875, "reading_distances": [26, 35, 28, 36], "reading_log_probs": [-413.1495666503906, -412.2997131347656, -412.887451171875, -412.525634765625], "reading_logprob_std": 0.37723050140816183, "tokenization_path_logprob_std": 220.8138854953918, "trajectory_balance_rmse": 7225.176835071632} +{"canonical_length": 42, "canonical_nll": 0.7884801228841146, "generation_rouge_l_f1": 0.211864406779661, "generation_token_f1": 0.23728813559322032, "prompt_id": "c3513f19426dd701", "reading_canonical_log_prob": -33.11616516113281, "reading_distances": [23, 27, 26, 33], "reading_log_probs": [-32.49533462524414, -33.144493103027344, -31.715715408325195, -33.71976089477539], "reading_logprob_std": 0.8620213448173131, "tokenization_path_logprob_std": 33.40244705481777, "trajectory_balance_rmse": 373.7678382330518} +{"canonical_length": 356, "canonical_nll": 1.131748756665862, "generation_rouge_l_f1": 0.21368948247078465, "generation_token_f1": 0.41402337228714525, "prompt_id": "80a15e1bafbe7e14", "reading_canonical_log_prob": -402.9025573730469, "reading_distances": [20, 20, 17, 16], "reading_log_probs": [-402.3079833984375, -402.190185546875, -403.3753662109375, -403.3714904785156], "reading_logprob_std": 0.6509210961474354, "tokenization_path_logprob_std": 167.97917779369462, "trajectory_balance_rmse": 7474.659055593762} +{"canonical_length": 51, "canonical_nll": 1.4640679452933518, "generation_rouge_l_f1": 0.4782608695652174, "generation_token_f1": 0.5434782608695652, "prompt_id": "74f44ff0ad2c7451", "reading_canonical_log_prob": -74.66746520996094, "reading_distances": [23, 25, 25, 22], "reading_log_probs": [-75.705078125, -76.15301513671875, -77.20361328125, -76.65711975097656], "reading_logprob_std": 0.6460889735307411, "tokenization_path_logprob_std": 66.4984004840989, "trajectory_balance_rmse": 1256.263450216832} +{"canonical_length": 21, "canonical_nll": 1.3293615068708147, "generation_rouge_l_f1": 0.358974358974359, "generation_token_f1": 0.3846153846153846, "prompt_id": "0b30260636e7066b", "reading_canonical_log_prob": -27.91659164428711, "reading_distances": [32, 34, 23, 31], "reading_log_probs": [-29.337038040161133, -28.569252014160156, -27.078935623168945, -28.984981536865234], "reading_logprob_std": 0.9932837247938966, "tokenization_path_logprob_std": 84.20196502942854, "trajectory_balance_rmse": 553.241575683162} +{"canonical_length": 142, "canonical_nll": 1.2571148939535652, "generation_rouge_l_f1": 0.33962264150943394, "generation_token_f1": 0.5175202156334232, "prompt_id": "270a3b8ee410411a", "reading_canonical_log_prob": -178.51031494140625, "reading_distances": [24, 31, 31, 29], "reading_log_probs": [-176.76771545410156, -175.82321166992188, -176.27786254882812, -175.93881225585938], "reading_logprob_std": 0.4236891350105856, "tokenization_path_logprob_std": 145.40621575271027, "trajectory_balance_rmse": 2643.892619902735} +{"canonical_length": 2, "canonical_nll": 1.7690788507461548, "generation_rouge_l_f1": 0.0078125, "generation_token_f1": 0.0078125, "prompt_id": "aa51f5831004ecdf", "reading_canonical_log_prob": -3.5381577014923096, "reading_distances": [35, 33, 39, 28], "reading_log_probs": [-4.298358917236328, -3.9897103309631348, -4.216250419616699, -4.244388580322266], "reading_logprob_std": 0.13598153398167084, "tokenization_path_logprob_std": 27.05581280284099, "trajectory_balance_rmse": 47.79692594346786} +{"canonical_length": 53, "canonical_nll": 0.9386573647553066, "generation_rouge_l_f1": 0.23863636363636365, "generation_token_f1": 0.3409090909090909, "prompt_id": "d23e5c49d3b87636", "reading_canonical_log_prob": -49.74884033203125, "reading_distances": [20, 16, 18, 13], "reading_log_probs": [-46.46188735961914, -47.717063903808594, -45.91385269165039, -46.09324645996094], "reading_logprob_std": 0.8130314753022304, "tokenization_path_logprob_std": 94.62948615470904, "trajectory_balance_rmse": 912.6466007434353} +{"canonical_length": 78, "canonical_nll": 1.7394819993239183, "generation_rouge_l_f1": 0.19178082191780824, "generation_token_f1": 0.3287671232876712, "prompt_id": "1cf9ae29e6e4ee3d", "reading_canonical_log_prob": -135.67959594726562, "reading_distances": [19, 22, 27, 21], "reading_log_probs": [-137.4658203125, -136.12603759765625, -136.5139617919922, -137.75921630859375], "reading_logprob_std": 0.7722019527574219, "tokenization_path_logprob_std": 88.84074129965721, "trajectory_balance_rmse": 1742.939587325547} +{"canonical_length": 113, "canonical_nll": 1.3784332275390625, "generation_rouge_l_f1": 0.15633423180592992, "generation_token_f1": 0.3018867924528302, "prompt_id": "4b26a788bce312c3", "reading_canonical_log_prob": -155.76295471191406, "reading_distances": [27, 37, 31, 34], "reading_log_probs": [-155.67935180664062, -156.171630859375, -156.42227172851562, -155.78854370117188], "reading_logprob_std": 0.3436785531345446, "tokenization_path_logprob_std": 170.98786434793823, "trajectory_balance_rmse": 2494.579038354063} +{"canonical_length": 320, "canonical_nll": 1.4246746063232423, "generation_rouge_l_f1": 0.2996515679442509, "generation_token_f1": 0.505226480836237, "prompt_id": "c59752dc85c30f5b", "reading_canonical_log_prob": -455.8958740234375, "reading_distances": [35, 38, 41, 35], "reading_log_probs": [-459.034912109375, -455.4695129394531, -460.6022644042969, -456.41265869140625], "reading_logprob_std": 2.3599494165499593, "tokenization_path_logprob_std": 306.04929349695493, "trajectory_balance_rmse": 7986.425563923422} +{"canonical_length": 9, "canonical_nll": 1.362831221686469, "generation_rouge_l_f1": 0.7692307692307693, "generation_token_f1": 0.7692307692307693, "prompt_id": "b4e868a234198084", "reading_canonical_log_prob": -12.265480995178223, "reading_distances": [25, 32, 30, 31], "reading_log_probs": [-12.751197814941406, -13.066766738891602, -12.838553428649902, -13.004693984985352], "reading_logprob_std": 0.1457772511530444, "tokenization_path_logprob_std": 34.2660168933551, "trajectory_balance_rmse": 248.09617119618065} +{"canonical_length": 68, "canonical_nll": 1.6481139239142923, "generation_rouge_l_f1": 0.208955223880597, "generation_token_f1": 0.3582089552238806, "prompt_id": "466e7627e6013e86", "reading_canonical_log_prob": -112.07174682617188, "reading_distances": [33, 36, 37, 37], "reading_log_probs": [-111.18132019042969, -111.24968719482422, -112.32299041748047, -111.90048217773438], "reading_logprob_std": 0.546146046463059, "tokenization_path_logprob_std": 105.05248492617774, "trajectory_balance_rmse": 1590.1191868864785} +{"canonical_length": 14, "canonical_nll": 0.9779755047389439, "generation_rouge_l_f1": 0.06920415224913495, "generation_token_f1": 0.08304498269896195, "prompt_id": "ca45cac53e43d02f", "reading_canonical_log_prob": -13.691657066345215, "reading_distances": [24, 24, 23, 29], "reading_log_probs": [-12.527402877807617, -11.040550231933594, -12.854265213012695, -12.608543395996094], "reading_logprob_std": 0.8232404148093492, "tokenization_path_logprob_std": 50.83216413593819, "trajectory_balance_rmse": 197.58030924433297} +{"canonical_length": 20, "canonical_nll": 2.195535659790039, "generation_rouge_l_f1": 0.0743801652892562, "generation_token_f1": 0.0909090909090909, "prompt_id": "685a9c22a80a27a9", "reading_canonical_log_prob": -43.91071319580078, "reading_distances": [24, 21, 20, 16], "reading_log_probs": [-39.6331787109375, -36.999351501464844, -41.38990783691406, -35.53224182128906], "reading_logprob_std": 2.623332379235094, "tokenization_path_logprob_std": 49.62176415251093, "trajectory_balance_rmse": 429.93399134518233} +{"canonical_length": 12, "canonical_nll": 1.5655029614766438, "generation_rouge_l_f1": 0.34146341463414637, "generation_token_f1": 0.3902439024390244, "prompt_id": "426d94bf331d4f8d", "reading_canonical_log_prob": -18.786035537719727, "reading_distances": [23, 26, 23, 31], "reading_log_probs": [-17.76291275024414, -15.874770164489746, -16.450674057006836, -13.591245651245117], "reading_logprob_std": 1.7419419120797777, "tokenization_path_logprob_std": 43.88976480363732, "trajectory_balance_rmse": 164.79226644846565} +{"canonical_length": 123, "canonical_nll": 1.1617302623221544, "generation_rouge_l_f1": 0.1842818428184282, "generation_token_f1": 0.3089430894308943, "prompt_id": "efba508a45047679", "reading_canonical_log_prob": -142.892822265625, "reading_distances": [29, 34, 36, 30], "reading_log_probs": [-143.20925903320312, -145.38143920898438, -144.8052215576172, -144.8765869140625], "reading_logprob_std": 0.9415165511564199, "tokenization_path_logprob_std": 138.8136804249928, "trajectory_balance_rmse": 2603.381864573676} +{"canonical_length": 138, "canonical_nll": 1.1973763065061707, "generation_rouge_l_f1": 0.22596153846153846, "generation_token_f1": 0.3461538461538462, "prompt_id": "79eb49a82dc5be50", "reading_canonical_log_prob": -165.23793029785156, "reading_distances": [20, 28, 20, 15], "reading_log_probs": [-164.80917358398438, -166.3077850341797, -164.15170288085938, -167.00531005859375], "reading_logprob_std": 1.315909848596565, "tokenization_path_logprob_std": 202.93774433236803, "trajectory_balance_rmse": 3015.7525787825207} +{"canonical_length": 355, "canonical_nll": 1.1423276229643486, "generation_rouge_l_f1": 0.2403846153846154, "generation_token_f1": 0.4487179487179487, "prompt_id": "144d91c0bdbf405c", "reading_canonical_log_prob": -405.52630615234375, "reading_distances": [30, 28, 30, 35], "reading_log_probs": [-408.6094970703125, -409.6082763671875, -408.5076904296875, -409.25604248046875], "reading_logprob_std": 0.5260968345608878, "tokenization_path_logprob_std": 235.16265395864124, "trajectory_balance_rmse": 6551.564239246774} +{"canonical_length": 194, "canonical_nll": 1.0873474435707957, "generation_rouge_l_f1": 0.308390022675737, "generation_token_f1": 0.49886621315192753, "prompt_id": "6bebc48cdc1b16ab", "reading_canonical_log_prob": -210.94540405273438, "reading_distances": [58, 61, 53, 54], "reading_log_probs": [-208.49569702148438, -207.75791931152344, -208.80088806152344, -208.31381225585938], "reading_logprob_std": 0.43823350013664014, "tokenization_path_logprob_std": 144.37856514693138, "trajectory_balance_rmse": 3826.5333785631606} +{"canonical_length": 136, "canonical_nll": 2.673277911017923, "generation_rouge_l_f1": 0.22895622895622897, "generation_token_f1": 0.39730639730639733, "prompt_id": "f19548b23e650d6d", "reading_canonical_log_prob": -363.5657958984375, "reading_distances": [20, 12, 22, 19], "reading_log_probs": [-383.7130126953125, -375.0596923828125, -380.39117431640625, -380.1639404296875], "reading_logprob_std": 3.5711828055102566, "tokenization_path_logprob_std": 175.4879713403523, "trajectory_balance_rmse": 2942.9541922883077} +{"canonical_length": 283, "canonical_nll": 1.0596082707597174, "generation_rouge_l_f1": 0.18861209964412812, "generation_token_f1": 0.4056939501779359, "prompt_id": "b85ad6ca4cfd648f", "reading_canonical_log_prob": -299.869140625, "reading_distances": [25, 23, 24, 23], "reading_log_probs": [-296.6615295410156, -296.3095703125, -297.07965087890625, -296.67510986328125], "reading_logprob_std": 0.31479912274264066, "tokenization_path_logprob_std": 356.33173720807946, "trajectory_balance_rmse": 5429.822478915852} +{"canonical_length": 146, "canonical_nll": 1.369704782146297, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.4642857142857143, "prompt_id": "6b7b364a0fc8e714", "reading_canonical_log_prob": -199.97689819335938, "reading_distances": [33, 22, 34, 27], "reading_log_probs": [-200.3417510986328, -205.8473358154297, -197.76541137695312, -198.05166625976562], "reading_logprob_std": 3.7457240767843927, "tokenization_path_logprob_std": 69.64407569319323, "trajectory_balance_rmse": 3328.3504645805865} +{"canonical_length": 112, "canonical_nll": 2.5303540910993303, "generation_rouge_l_f1": 0.23121387283236994, "generation_token_f1": 0.30057803468208094, "prompt_id": "1d91ffc1f9553c07", "reading_canonical_log_prob": -283.399658203125, "reading_distances": [12, 16, 19, 16], "reading_log_probs": [-282.54638671875, -283.5144958496094, -284.764404296875, -281.98638916015625], "reading_logprob_std": 1.2174147407687186, "tokenization_path_logprob_std": 171.34566608396335, "trajectory_balance_rmse": 2523.043539958157} +{"canonical_length": 37, "canonical_nll": 0.9674181551546663, "generation_rouge_l_f1": 0.14666666666666667, "generation_token_f1": 0.16666666666666666, "prompt_id": "1ac7a2e92a59b9ce", "reading_canonical_log_prob": -35.794471740722656, "reading_distances": [13, 20, 20, 13], "reading_log_probs": [-38.35517120361328, -38.531185150146484, -39.01478958129883, -38.630252838134766], "reading_logprob_std": 0.2788822001434876, "tokenization_path_logprob_std": 118.3961949416874, "trajectory_balance_rmse": 898.3760965813055} +{"canonical_length": 43, "canonical_nll": 1.8091790842455486, "generation_rouge_l_f1": 0.27338129496402874, "generation_token_f1": 0.3741007194244605, "prompt_id": "66c94013abca2b3a", "reading_canonical_log_prob": -77.7947006225586, "reading_distances": [36, 24, 28, 26], "reading_log_probs": [-74.99778747558594, -77.1010513305664, -76.44998168945312, -77.2868423461914], "reading_logprob_std": 1.038071321287786, "tokenization_path_logprob_std": 119.45111401991882, "trajectory_balance_rmse": 903.4630485696076} +{"canonical_length": 15, "canonical_nll": 1.1946534474690755, "generation_rouge_l_f1": 0.12999999999999998, "generation_token_f1": 0.14, "prompt_id": "785090c8f7a0ea91", "reading_canonical_log_prob": -17.919801712036133, "reading_distances": [17, 16, 18, 13], "reading_log_probs": [-18.612777709960938, -18.750516891479492, -18.156463623046875, -18.21464729309082], "reading_logprob_std": 0.2928513998246784, "tokenization_path_logprob_std": 55.11880449322281, "trajectory_balance_rmse": 403.33515426039077} +{"canonical_length": 217, "canonical_nll": 1.098355974469866, "generation_rouge_l_f1": 0.1968680089485459, "generation_token_f1": 0.3847874720357942, "prompt_id": "30ed32c8494397b3", "reading_canonical_log_prob": -238.34324645996094, "reading_distances": [23, 23, 22, 15], "reading_log_probs": [-233.8546600341797, -235.87060546875, -236.581298828125, -235.08949279785156], "reading_logprob_std": 1.1677654712351264, "tokenization_path_logprob_std": 49.50297330987964, "trajectory_balance_rmse": 3806.169414269125} +{"canonical_length": 16, "canonical_nll": 1.0191646814346313, "generation_rouge_l_f1": 0.37333333333333335, "generation_token_f1": 0.37333333333333335, "prompt_id": "e8d7fb5246616f5e", "reading_canonical_log_prob": -16.3066349029541, "reading_distances": [79, 78, 81, 78], "reading_log_probs": [-16.00996971130371, -16.389102935791016, -15.575626373291016, -16.220165252685547], "reading_logprob_std": 0.3514603733429006, "tokenization_path_logprob_std": 34.832553365752204, "trajectory_balance_rmse": 340.5551912108564} +{"canonical_length": 6, "canonical_nll": 2.0505654017130532, "generation_rouge_l_f1": 0.6666666666666665, "generation_token_f1": 0.6666666666666665, "prompt_id": "96f83a49dbeabc97", "reading_canonical_log_prob": -12.30339241027832, "reading_distances": [31, 28, 32, 30], "reading_log_probs": [-10.956663131713867, -11.69884204864502, -11.116588592529297, -12.186073303222656], "reading_logprob_std": 0.563328352781207, "tokenization_path_logprob_std": 10.895964112513724, "trajectory_balance_rmse": 121.09557860339734} +{"canonical_length": 80, "canonical_nll": 1.1360504150390625, "generation_rouge_l_f1": 0.16286644951140064, "generation_token_f1": 0.29315960912052114, "prompt_id": "dea31d77a681f384", "reading_canonical_log_prob": -90.884033203125, "reading_distances": [13, 18, 15, 14], "reading_log_probs": [-91.10916137695312, -90.87606048583984, -91.80191040039062, -91.16325378417969], "reading_logprob_std": 0.39630833448065744, "tokenization_path_logprob_std": 91.723184239632, "trajectory_balance_rmse": 1599.1719222885356} +{"canonical_length": 9, "canonical_nll": 1.1541292402479384, "generation_rouge_l_f1": 0.42857142857142855, "generation_token_f1": 0.42857142857142855, "prompt_id": "428bb97daf254741", "reading_canonical_log_prob": -10.387163162231445, "reading_distances": [21, 22, 25, 22], "reading_log_probs": [-10.835675239562988, -10.706932067871094, -10.39018440246582, -10.014625549316406], "reading_logprob_std": 0.36626735856483517, "tokenization_path_logprob_std": 68.9480983263852, "trajectory_balance_rmse": 285.7131139061195} +{"canonical_length": 249, "canonical_nll": 1.4451252274723896, "generation_rouge_l_f1": 0.22727272727272727, "generation_token_f1": 0.46694214876033063, "prompt_id": "ed06fac728bea8f5", "reading_canonical_log_prob": -359.836181640625, "reading_distances": [21, 18, 19, 17], "reading_log_probs": [-355.8934326171875, -358.28936767578125, -356.505126953125, -357.93878173828125], "reading_logprob_std": 1.142363022121788, "tokenization_path_logprob_std": 204.2080249282174, "trajectory_balance_rmse": 5786.2837775108665} +{"canonical_length": 42, "canonical_nll": 1.6416586013067336, "generation_rouge_l_f1": 0.28205128205128205, "generation_token_f1": 0.35897435897435903, "prompt_id": "2455341860ece852", "reading_canonical_log_prob": -68.94966125488281, "reading_distances": [18, 16, 24, 16], "reading_log_probs": [-75.39632415771484, -75.88742065429688, -80.17622375488281, -73.39339447021484], "reading_logprob_std": 2.8536689104914017, "tokenization_path_logprob_std": 120.37561058011802, "trajectory_balance_rmse": 974.7808285411703} +{"canonical_length": 56, "canonical_nll": 1.7140452521187919, "generation_rouge_l_f1": 0.3088235294117647, "generation_token_f1": 0.39705882352941174, "prompt_id": "2b713b1aba655ea9", "reading_canonical_log_prob": -95.98653411865234, "reading_distances": [19, 18, 10, 18], "reading_log_probs": [-100.97003936767578, -100.25030517578125, -97.36294555664062, -100.41588592529297], "reading_logprob_std": 1.620723515277842, "tokenization_path_logprob_std": 133.79613747072437, "trajectory_balance_rmse": 1291.0019892633754} +{"canonical_length": 21, "canonical_nll": 2.9066874186197915, "generation_rouge_l_f1": 0.08, "generation_token_f1": 0.08, "prompt_id": "12ac8fa81a314c1b", "reading_canonical_log_prob": -61.040435791015625, "reading_distances": [27, 26, 32, 32], "reading_log_probs": [-61.01522445678711, -61.16493606567383, -62.065467834472656, -62.00212097167969], "reading_logprob_std": 0.5488805507424945, "tokenization_path_logprob_std": 45.13298043454385, "trajectory_balance_rmse": 463.83192385766716} +{"canonical_length": 174, "canonical_nll": 1.1715110691114403, "generation_rouge_l_f1": 0.24122807017543862, "generation_token_f1": 0.4298245614035088, "prompt_id": "46239be323c13eb4", "reading_canonical_log_prob": -203.84292602539062, "reading_distances": [21, 15, 22, 21], "reading_log_probs": [-221.268798828125, -207.12442016601562, -218.31353759765625, -227.4033203125], "reading_logprob_std": 8.49256370976449, "tokenization_path_logprob_std": 197.1251975830872, "trajectory_balance_rmse": 3177.679588120487} +{"canonical_length": 298, "canonical_nll": 1.1398481330615562, "generation_rouge_l_f1": 0.18214936247723135, "generation_token_f1": 0.42622950819672134, "prompt_id": "3fde994857b8ea58", "reading_canonical_log_prob": -339.67474365234375, "reading_distances": [39, 36, 35, 35], "reading_log_probs": [-338.70196533203125, -333.57513427734375, -337.1594543457031, -337.2860412597656], "reading_logprob_std": 2.1852304038726946, "tokenization_path_logprob_std": 181.60540811191245, "trajectory_balance_rmse": 6670.648934504306} +{"canonical_length": 111, "canonical_nll": 1.1911446511208474, "generation_rouge_l_f1": 0.2644628099173554, "generation_token_f1": 0.4573002754820936, "prompt_id": "366758412bdbb007", "reading_canonical_log_prob": -132.21705627441406, "reading_distances": [39, 36, 33, 39], "reading_log_probs": [-130.727783203125, -131.15164184570312, -129.90921020507812, -133.44436645507812], "reading_logprob_std": 1.5145707503774177, "tokenization_path_logprob_std": 86.97131703780107, "trajectory_balance_rmse": 2457.5885440258967} +{"canonical_length": 17, "canonical_nll": 1.4594176797305836, "generation_rouge_l_f1": 0.34146341463414637, "generation_token_f1": 0.43902439024390244, "prompt_id": "f77e57c7031a2b6c", "reading_canonical_log_prob": -24.810100555419922, "reading_distances": [10, 15, 16, 13], "reading_log_probs": [-24.50716209411621, -24.29438018798828, -24.521183013916016, -24.993431091308594], "reading_logprob_std": 0.2951073529639526, "tokenization_path_logprob_std": 18.19605420261703, "trajectory_balance_rmse": 497.6386853758948} +{"canonical_length": 8, "canonical_nll": 1.8843098878860474, "generation_rouge_l_f1": 0.30769230769230765, "generation_token_f1": 0.30769230769230765, "prompt_id": "1f37960a14b0dc02", "reading_canonical_log_prob": -15.074479103088379, "reading_distances": [24, 23, 21, 30], "reading_log_probs": [-14.3577880859375, -14.80713176727295, -14.648399353027344, -15.225531578063965], "reading_logprob_std": 0.36201742670086656, "tokenization_path_logprob_std": 71.83755151996172, "trajectory_balance_rmse": 207.5758590726791} +{"canonical_length": 28, "canonical_nll": 2.13618837084089, "generation_rouge_l_f1": 0.25641025641025644, "generation_token_f1": 0.41025641025641024, "prompt_id": "8d8a8351f2dceabb", "reading_canonical_log_prob": -59.81327438354492, "reading_distances": [20, 20, 21, 19], "reading_log_probs": [-62.602169036865234, -65.33509063720703, -64.59335327148438, -62.2744140625], "reading_logprob_std": 1.4954512396672106, "tokenization_path_logprob_std": 85.13685187977481, "trajectory_balance_rmse": 674.1679237634302} +{"canonical_length": 2, "canonical_nll": 1.4040008783340454, "generation_rouge_l_f1": 0.007782101167315175, "generation_token_f1": 0.007782101167315175, "prompt_id": "7a9b64f9112cd0c5", "reading_canonical_log_prob": -2.808001756668091, "reading_distances": [45, 42, 52, 41], "reading_log_probs": [-4.390259742736816, -4.6417717933654785, -5.126852989196777, -4.926835536956787], "reading_logprob_std": 0.3227890940620313, "tokenization_path_logprob_std": 0.0, "trajectory_balance_rmse": 2.8215315341949463} +{"canonical_length": 5, "canonical_nll": 1.7340415954589843, "generation_rouge_l_f1": 0.022900763358778626, "generation_token_f1": 0.022900763358778626, "prompt_id": "2797e2b7991dcf3e", "reading_canonical_log_prob": -8.670207977294922, "reading_distances": [28, 20, 20, 23], "reading_log_probs": [-9.29603099822998, -8.753725051879883, -8.791521072387695, -9.535866737365723], "reading_logprob_std": 0.38442292365736225, "tokenization_path_logprob_std": 48.80344878074779, "trajectory_balance_rmse": 79.40348091540616} +{"canonical_length": 313, "canonical_nll": 1.66881349216254, "generation_rouge_l_f1": 0.1851851851851852, "generation_token_f1": 0.4377104377104377, "prompt_id": "d353ae8508e0bb05", "reading_canonical_log_prob": -522.338623046875, "reading_distances": [42, 41, 40, 45], "reading_log_probs": [-527.9732666015625, -530.717529296875, -532.5902709960938, -528.1170654296875], "reading_logprob_std": 2.2201262587752972, "tokenization_path_logprob_std": 233.7577317298709, "trajectory_balance_rmse": 6780.708114048667} +{"canonical_length": 7, "canonical_nll": 0.11016269241060529, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1099a8369d87bee2", "reading_canonical_log_prob": -0.7711388468742371, "reading_distances": [20, 19, 25, 19], "reading_log_probs": [-1.5506483316421509, -1.6425586938858032, -1.64814031124115, -1.5987462997436523], "reading_logprob_std": 0.045328489950184465, "tokenization_path_logprob_std": 23.558962855611103, "trajectory_balance_rmse": 184.22293249815522} +{"canonical_length": 264, "canonical_nll": 1.27315498120857, "generation_rouge_l_f1": 0.2780952380952381, "generation_token_f1": 0.537142857142857, "prompt_id": "d8d494771b6c2d81", "reading_canonical_log_prob": -336.1129150390625, "reading_distances": [25, 25, 31, 25], "reading_log_probs": [-337.7918701171875, -337.73138427734375, -338.19561767578125, -337.945068359375], "reading_logprob_std": 0.20698003615312055, "tokenization_path_logprob_std": 307.3458131016933, "trajectory_balance_rmse": 6380.195386435332} +{"canonical_length": 372, "canonical_nll": 1.1803748017998152, "generation_rouge_l_f1": 0.26898734177215183, "generation_token_f1": 0.41455696202531644, "prompt_id": "27da9cdd8cf02e41", "reading_canonical_log_prob": -439.09942626953125, "reading_distances": [33, 34, 34, 31], "reading_log_probs": [-444.55767822265625, -441.5777282714844, -443.1871337890625, -444.2048034667969], "reading_logprob_std": 1.3357387507818783, "tokenization_path_logprob_std": 217.48784256277494, "trajectory_balance_rmse": 7566.947021964846} +{"canonical_length": 51, "canonical_nll": 1.055084228515625, "generation_rouge_l_f1": 0.09345794392523364, "generation_token_f1": 0.11214953271028037, "prompt_id": "47574188e8957081", "reading_canonical_log_prob": -53.809295654296875, "reading_distances": [33, 28, 28, 33], "reading_log_probs": [-51.30348205566406, -49.973052978515625, -48.23027801513672, -50.43103790283203], "reading_logprob_std": 1.2931274186707387, "tokenization_path_logprob_std": 116.67299406975009, "trajectory_balance_rmse": 1068.5132433629988} +{"canonical_length": 238, "canonical_nll": 1.99327279740021, "generation_rouge_l_f1": 0.16736401673640167, "generation_token_f1": 0.33891213389121333, "prompt_id": "b7c16abbc7061a65", "reading_canonical_log_prob": -474.39892578125, "reading_distances": [27, 30, 34, 27], "reading_log_probs": [-480.4040222167969, -484.4084777832031, -484.0081787109375, -489.57196044921875], "reading_logprob_std": 3.773307122315618, "tokenization_path_logprob_std": 131.99436570758238, "trajectory_balance_rmse": 5710.407966206206} +{"canonical_length": 334, "canonical_nll": 1.293790074879538, "generation_rouge_l_f1": 0.22033898305084745, "generation_token_f1": 0.47796610169491527, "prompt_id": "01fef5f51abd900f", "reading_canonical_log_prob": -432.1258850097656, "reading_distances": [20, 31, 34, 33], "reading_log_probs": [-433.241943359375, -436.0415954589844, -434.0262756347656, -432.8251953125], "reading_logprob_std": 1.4281761611804127, "tokenization_path_logprob_std": 370.3709801756847, "trajectory_balance_rmse": 7516.217338681549} +{"canonical_length": 60, "canonical_nll": 1.002896499633789, "generation_rouge_l_f1": 0.30303030303030304, "generation_token_f1": 0.3333333333333333, "prompt_id": "1d0839cb7b1b8564", "reading_canonical_log_prob": -60.173789978027344, "reading_distances": [29, 34, 29, 30], "reading_log_probs": [-63.60926818847656, -62.362152099609375, -63.64817428588867, -62.50695037841797], "reading_logprob_std": 0.6921661884999644, "tokenization_path_logprob_std": 77.04439501505247, "trajectory_balance_rmse": 1157.05768856716} +{"canonical_length": 198, "canonical_nll": 1.4898601493450125, "generation_rouge_l_f1": 0.29370629370629375, "generation_token_f1": 0.49883449883449876, "prompt_id": "b900859423a57fd5", "reading_canonical_log_prob": -294.9923095703125, "reading_distances": [19, 22, 22, 25], "reading_log_probs": [-301.29058837890625, -300.6447448730469, -295.4171447753906, -297.3802490234375], "reading_logprob_std": 2.76953247700912, "tokenization_path_logprob_std": 225.2855941848194, "trajectory_balance_rmse": 4590.95480096717} +{"canonical_length": 48, "canonical_nll": 1.4756843249003093, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.320855614973262, "prompt_id": "5fa2bbf4741331df", "reading_canonical_log_prob": -70.83284759521484, "reading_distances": [18, 16, 18, 16], "reading_log_probs": [-69.97366333007812, -71.06706237792969, -70.21969604492188, -71.26221466064453], "reading_logprob_std": 0.6297735893433416, "tokenization_path_logprob_std": 31.433974808704917, "trajectory_balance_rmse": 1213.206571039996} +{"canonical_length": 501, "canonical_nll": 0.5529289321747131, "generation_rouge_l_f1": 0.200339558573854, "generation_token_f1": 0.29541595925297115, "prompt_id": "97c9524265f7d241", "reading_canonical_log_prob": -277.01739501953125, "reading_distances": [22, 18, 21, 19], "reading_log_probs": [-275.38494873046875, -274.1275634765625, -273.71148681640625, -274.59954833984375], "reading_logprob_std": 0.7178027233541752, "tokenization_path_logprob_std": 194.04021107132675, "trajectory_balance_rmse": 7188.569458587966} +{"canonical_length": 11, "canonical_nll": 1.243779876015403, "generation_rouge_l_f1": 0.588235294117647, "generation_token_f1": 0.588235294117647, "prompt_id": "990e2b768a18fc67", "reading_canonical_log_prob": -13.681578636169434, "reading_distances": [25, 26, 23, 31], "reading_log_probs": [-13.795609474182129, -13.590222358703613, -13.873703956604004, -15.482583999633789], "reading_logprob_std": 0.8729287415640814, "tokenization_path_logprob_std": 51.60176980579003, "trajectory_balance_rmse": 215.08886353884114} +{"canonical_length": 30, "canonical_nll": 1.5775960286458333, "generation_rouge_l_f1": 0.0738255033557047, "generation_token_f1": 0.11409395973154361, "prompt_id": "4a59564b57b1ad24", "reading_canonical_log_prob": -47.327880859375, "reading_distances": [31, 30, 32, 32], "reading_log_probs": [-49.63419723510742, -48.73374938964844, -48.75983428955078, -48.32279968261719], "reading_logprob_std": 0.5519389877967473, "tokenization_path_logprob_std": 34.12870036311096, "trajectory_balance_rmse": 674.6585577057918} +{"canonical_length": 118, "canonical_nll": 1.5484439397262315, "generation_rouge_l_f1": 0.2530612244897959, "generation_token_f1": 0.3510204081632653, "prompt_id": "fc01869b83f5b227", "reading_canonical_log_prob": -182.7163848876953, "reading_distances": [26, 26, 28, 33], "reading_log_probs": [-184.640380859375, -189.50534057617188, -186.83050537109375, -184.93826293945312], "reading_logprob_std": 2.238800554812822, "tokenization_path_logprob_std": 164.64335667322888, "trajectory_balance_rmse": 2762.845827303166} +{"canonical_length": 10, "canonical_nll": 0.3182713508605957, "generation_rouge_l_f1": 0.3636363636363636, "generation_token_f1": 0.3636363636363636, "prompt_id": "e35706a135002934", "reading_canonical_log_prob": -3.182713508605957, "reading_distances": [32, 30, 33, 36], "reading_log_probs": [-3.6643710136413574, -3.6757755279541016, -3.6981043815612793, -4.078771114349365], "reading_logprob_std": 0.20016797702882919, "tokenization_path_logprob_std": 13.707889412770847, "trajectory_balance_rmse": 228.0267200437341} +{"canonical_length": 123, "canonical_nll": 1.0868700229055512, "generation_rouge_l_f1": 0.2041884816753927, "generation_token_f1": 0.38219895287958117, "prompt_id": "5df9638e97e9c3b6", "reading_canonical_log_prob": -133.6850128173828, "reading_distances": [11, 11, 10, 14], "reading_log_probs": [-131.55072021484375, -131.68032836914062, -131.53912353515625, -131.36041259765625], "reading_logprob_std": 0.1314572969214489, "tokenization_path_logprob_std": 91.83412825480117, "trajectory_balance_rmse": 2427.1024281613873} +{"canonical_length": 25, "canonical_nll": 0.6060862731933594, "generation_rouge_l_f1": 0.23170731707317074, "generation_token_f1": 0.25609756097560976, "prompt_id": "6f6b5ba97837e4b7", "reading_canonical_log_prob": -15.152156829833984, "reading_distances": [27, 27, 33, 26], "reading_log_probs": [-14.510360717773438, -13.636153221130371, -13.867581367492676, -14.87594985961914], "reading_logprob_std": 0.5714397874827466, "tokenization_path_logprob_std": 62.7727863169805, "trajectory_balance_rmse": 479.76176668357} +{"canonical_length": 211, "canonical_nll": 0.8457897602099378, "generation_rouge_l_f1": 0.3813387423935091, "generation_token_f1": 0.5233265720081136, "prompt_id": "111777e66227bbb0", "reading_canonical_log_prob": -178.46163940429688, "reading_distances": [40, 43, 44, 36], "reading_log_probs": [-176.41558837890625, -175.78091430664062, -175.10122680664062, -173.26434326171875], "reading_logprob_std": 1.361064209502165, "tokenization_path_logprob_std": 123.6406403924849, "trajectory_balance_rmse": 3841.756303710176} +{"canonical_length": 264, "canonical_nll": 1.0892683086973247, "generation_rouge_l_f1": 0.2608695652173913, "generation_token_f1": 0.4584980237154151, "prompt_id": "7eae8a9735d3d72c", "reading_canonical_log_prob": -287.56683349609375, "reading_distances": [31, 30, 35, 31], "reading_log_probs": [-290.76873779296875, -288.8182373046875, -291.1396789550781, -290.59521484375], "reading_logprob_std": 1.033414300327774, "tokenization_path_logprob_std": 118.23894771837716, "trajectory_balance_rmse": 5964.522919562526} +{"canonical_length": 6, "canonical_nll": 0.15322275956471762, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.6666666666666666, "prompt_id": "88814d5eacafd109", "reading_canonical_log_prob": -0.9193365573883057, "reading_distances": [38, 42, 36, 41], "reading_log_probs": [-4.540360450744629, -5.208441734313965, -3.4088633060455322, -3.331629991531372], "reading_logprob_std": 0.9107925822762973, "tokenization_path_logprob_std": 21.26317846279086, "trajectory_balance_rmse": 117.74245256201105} +{"canonical_length": 150, "canonical_nll": 1.312946573893229, "generation_rouge_l_f1": 0.21052631578947367, "generation_token_f1": 0.41403508771929826, "prompt_id": "c6b7c5355ef042f7", "reading_canonical_log_prob": -196.94198608398438, "reading_distances": [40, 40, 38, 41], "reading_log_probs": [-197.11380004882812, -196.18975830078125, -196.94229125976562, -197.7767333984375], "reading_logprob_std": 0.6521655298959543, "tokenization_path_logprob_std": 93.46209314427907, "trajectory_balance_rmse": 3183.6820781449496} +{"canonical_length": 21, "canonical_nll": 1.780559358142671, "generation_rouge_l_f1": 0.17647058823529413, "generation_token_f1": 0.17647058823529413, "prompt_id": "62bbc871380d57bb", "reading_canonical_log_prob": -37.391746520996094, "reading_distances": [13, 17, 17, 13], "reading_log_probs": [-38.348079681396484, -37.64259338378906, -37.62635040283203, -37.045265197753906], "reading_logprob_std": 0.5331240051386559, "tokenization_path_logprob_std": 46.133035936943, "trajectory_balance_rmse": 436.4266037986213} +{"canonical_length": 28, "canonical_nll": 1.7175737108503069, "generation_rouge_l_f1": 0.28070175438596495, "generation_token_f1": 0.3508771929824561, "prompt_id": "085c601e3edc073f", "reading_canonical_log_prob": -48.092063903808594, "reading_distances": [15, 11, 13, 11], "reading_log_probs": [-49.34524917602539, -50.30438232421875, -49.74357604980469, -49.82052993774414], "reading_logprob_std": 0.39359784964969863, "tokenization_path_logprob_std": 44.58864077454225, "trajectory_balance_rmse": 643.4355177371272} +{"canonical_length": 314, "canonical_nll": 1.3916754267018312, "generation_rouge_l_f1": 0.23570190641247835, "generation_token_f1": 0.48180242634315423, "prompt_id": "575d828d632e226f", "reading_canonical_log_prob": -436.986083984375, "reading_distances": [37, 32, 28, 34], "reading_log_probs": [-435.13128662109375, -438.6860656738281, -438.3807373046875, -438.30926513671875], "reading_logprob_std": 1.6717068032703186, "tokenization_path_logprob_std": 309.89662802847045, "trajectory_balance_rmse": 7111.128885519623} +{"canonical_length": 382, "canonical_nll": 0.9688436298470222, "generation_rouge_l_f1": 0.2673611111111111, "generation_token_f1": 0.5, "prompt_id": "acf1157605893744", "reading_canonical_log_prob": -370.0982666015625, "reading_distances": [20, 17, 14, 14], "reading_log_probs": [-369.62420654296875, -369.50311279296875, -368.7198486328125, -368.6213684082031], "reading_logprob_std": 0.5195258663580942, "tokenization_path_logprob_std": 237.40744610367503, "trajectory_balance_rmse": 8011.626666937277} +{"canonical_length": 19, "canonical_nll": 2.165213133159437, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "c0621d0076f956e9", "reading_canonical_log_prob": -41.1390495300293, "reading_distances": [34, 34, 32, 34], "reading_log_probs": [-44.46510314941406, -43.504425048828125, -43.425025939941406, -42.77510070800781], "reading_logprob_std": 0.6965027091522953, "tokenization_path_logprob_std": 33.48784042277402, "trajectory_balance_rmse": 333.57959162732163} +{"canonical_length": 252, "canonical_nll": 1.170471675812252, "generation_rouge_l_f1": 0.24662813102119457, "generation_token_f1": 0.5433526011560693, "prompt_id": "71ec6348f5f044b8", "reading_canonical_log_prob": -294.9588623046875, "reading_distances": [19, 20, 17, 19], "reading_log_probs": [-301.57861328125, -299.05755615234375, -295.22747802734375, -302.2835693359375], "reading_logprob_std": 3.1892357386370387, "tokenization_path_logprob_std": 207.55589010580636, "trajectory_balance_rmse": 5826.706970669832} +{"canonical_length": 13, "canonical_nll": 0.617885736318735, "generation_rouge_l_f1": 0.5599999999999999, "generation_token_f1": 0.5599999999999999, "prompt_id": "73a9260fbda0ad11", "reading_canonical_log_prob": -8.032514572143555, "reading_distances": [40, 30, 36, 38], "reading_log_probs": [-10.359578132629395, -10.499128341674805, -9.949027061462402, -10.530180931091309], "reading_logprob_std": 0.26746452408208343, "tokenization_path_logprob_std": 41.7067334859676, "trajectory_balance_rmse": 174.88884141866902} +{"canonical_length": 142, "canonical_nll": 0.954924355090504, "generation_rouge_l_f1": 0.30851063829787234, "generation_token_f1": 0.44148936170212766, "prompt_id": "0e246b28f0fca81c", "reading_canonical_log_prob": -135.59925842285156, "reading_distances": [10, 7, 10, 8], "reading_log_probs": [-134.74899291992188, -133.7326202392578, -135.5926513671875, -135.22994995117188], "reading_logprob_std": 0.8067092127784483, "tokenization_path_logprob_std": 83.59098218477187, "trajectory_balance_rmse": 2989.3439505160545} +{"canonical_length": 38, "canonical_nll": 1.2490077771638568, "generation_rouge_l_f1": 0.10256410256410256, "generation_token_f1": 0.10989010989010989, "prompt_id": "7dd4910bb043bf96", "reading_canonical_log_prob": -47.46229553222656, "reading_distances": [14, 9, 11, 12], "reading_log_probs": [-50.20496368408203, -49.140869140625, -50.59637451171875, -50.04370880126953], "reading_logprob_std": 0.6158028293928578, "tokenization_path_logprob_std": 85.3159706775784, "trajectory_balance_rmse": 565.0527297889832} +{"canonical_length": 187, "canonical_nll": 1.2404292305523061, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.4235294117647058, "prompt_id": "b7fe1fa68d367af3", "reading_canonical_log_prob": -231.96026611328125, "reading_distances": [17, 13, 15, 15], "reading_log_probs": [-228.89639282226562, -229.5076904296875, -228.92214965820312, -228.76675415039062], "reading_logprob_std": 0.33004377153894965, "tokenization_path_logprob_std": 232.61622428261154, "trajectory_balance_rmse": 4130.382623601214} +{"canonical_length": 96, "canonical_nll": 1.3413556416829426, "generation_rouge_l_f1": 0.33070866141732286, "generation_token_f1": 0.49606299212598426, "prompt_id": "b435d26419c9a382", "reading_canonical_log_prob": -128.7701416015625, "reading_distances": [41, 46, 48, 52], "reading_log_probs": [-128.23480224609375, -129.1026611328125, -129.400390625, -129.53201293945312], "reading_logprob_std": 0.5834410251052022, "tokenization_path_logprob_std": 180.19378683206642, "trajectory_balance_rmse": 2388.818507707576} +{"canonical_length": 113, "canonical_nll": 1.3664726324840986, "generation_rouge_l_f1": 0.18285714285714286, "generation_token_f1": 0.3142857142857143, "prompt_id": "fc2765baa4652639", "reading_canonical_log_prob": -154.41140747070312, "reading_distances": [19, 26, 24, 21], "reading_log_probs": [-156.83050537109375, -156.88040161132812, -157.58026123046875, -155.67913818359375], "reading_logprob_std": 0.787261923408352, "tokenization_path_logprob_std": 212.04511411801167, "trajectory_balance_rmse": 2462.2829075435825} +{"canonical_length": 216, "canonical_nll": 1.4518196671097368, "generation_rouge_l_f1": 0.21008403361344538, "generation_token_f1": 0.4453781512605042, "prompt_id": "6389a5bbf9eaba60", "reading_canonical_log_prob": -313.5930480957031, "reading_distances": [34, 29, 31, 26], "reading_log_probs": [-312.20916748046875, -313.02410888671875, -312.86883544921875, -313.71173095703125], "reading_logprob_std": 0.6167384044835833, "tokenization_path_logprob_std": 116.0285226474137, "trajectory_balance_rmse": 4786.036861246125} +{"canonical_length": 13, "canonical_nll": 1.1617220365084135, "generation_rouge_l_f1": 0.13333333333333336, "generation_token_f1": 0.13333333333333336, "prompt_id": "286cf1906a688fcb", "reading_canonical_log_prob": -15.102386474609375, "reading_distances": [25, 25, 24, 24], "reading_log_probs": [-13.935741424560547, -13.765182495117188, -14.253731727600098, -11.178764343261719], "reading_logprob_std": 1.417591691085006, "tokenization_path_logprob_std": 25.70177812373054, "trajectory_balance_rmse": 279.8963261007194} +{"canonical_length": 93, "canonical_nll": 1.4070785686533938, "generation_rouge_l_f1": 0.3287671232876712, "generation_token_f1": 0.410958904109589, "prompt_id": "33333486aa163f12", "reading_canonical_log_prob": -130.85830688476562, "reading_distances": [34, 26, 32, 28], "reading_log_probs": [-128.01434326171875, -128.44207763671875, -129.46177673339844, -129.24249267578125], "reading_logprob_std": 0.6779200613032771, "tokenization_path_logprob_std": 142.10840620717647, "trajectory_balance_rmse": 2310.13262624317} +{"canonical_length": 60, "canonical_nll": 1.8131600697835286, "generation_rouge_l_f1": 0.23343848580441642, "generation_token_f1": 0.25867507886435326, "prompt_id": "c5556f6cb2c9f8dd", "reading_canonical_log_prob": -108.78960418701172, "reading_distances": [25, 23, 27, 23], "reading_log_probs": [-118.33981323242188, -119.52958679199219, -113.34144592285156, -115.02037048339844], "reading_logprob_std": 2.8702909820913907, "tokenization_path_logprob_std": 40.205197870214256, "trajectory_balance_rmse": 359.5785838252434} +{"canonical_length": 11, "canonical_nll": 3.3304772810502485, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.25, "prompt_id": "b1950a5b21116b51", "reading_canonical_log_prob": -36.635250091552734, "reading_distances": [17, 18, 12, 19], "reading_log_probs": [-36.2546501159668, -37.65177917480469, -37.677459716796875, -35.05888366699219], "reading_logprob_std": 1.2578725856593087, "tokenization_path_logprob_std": 27.089603164145593, "trajectory_balance_rmse": 228.05923445039778} +{"canonical_length": 263, "canonical_nll": 1.4479553454729088, "generation_rouge_l_f1": 0.22433460076045628, "generation_token_f1": 0.4790874524714829, "prompt_id": "355011c0ec753a64", "reading_canonical_log_prob": -380.812255859375, "reading_distances": [18, 21, 20, 18], "reading_log_probs": [-380.98712158203125, -383.5208435058594, -382.7337646484375, -382.4898376464844], "reading_logprob_std": 1.059505818421406, "tokenization_path_logprob_std": 233.9471804810239, "trajectory_balance_rmse": 6394.191915281657} +{"canonical_length": 29, "canonical_nll": 1.156238818990773, "generation_rouge_l_f1": 0.325, "generation_token_f1": 0.39999999999999997, "prompt_id": "c2e4ff98064c8812", "reading_canonical_log_prob": -33.53092575073242, "reading_distances": [25, 22, 22, 29], "reading_log_probs": [-33.93408966064453, -33.050716400146484, -32.40497970581055, -33.56122589111328], "reading_logprob_std": 0.6628256446061438, "tokenization_path_logprob_std": 83.70765107383149, "trajectory_balance_rmse": 660.4736423956124} +{"canonical_length": 57, "canonical_nll": 1.7573710659094024, "generation_rouge_l_f1": 0.11920529801324502, "generation_token_f1": 0.18543046357615894, "prompt_id": "e95bb93274990ecf", "reading_canonical_log_prob": -100.17015075683594, "reading_distances": [28, 30, 27, 28], "reading_log_probs": [-102.1337661743164, -98.56314086914062, -99.43721008300781, -101.15420532226562], "reading_logprob_std": 1.6177660504771676, "tokenization_path_logprob_std": 141.35204069894013, "trajectory_balance_rmse": 1170.8104403628179} +{"canonical_length": 40, "canonical_nll": 1.7971992492675781, "generation_rouge_l_f1": 0.21538461538461537, "generation_token_f1": 0.2769230769230769, "prompt_id": "0eb7b4a395752869", "reading_canonical_log_prob": -71.88796997070312, "reading_distances": [30, 29, 36, 29], "reading_log_probs": [-74.19122314453125, -72.35844421386719, -74.71154022216797, -73.26107788085938], "reading_logprob_std": 1.038854659634068, "tokenization_path_logprob_std": 78.2214632521168, "trajectory_balance_rmse": 726.1309882462657} +{"canonical_length": 124, "canonical_nll": 1.9266138384419103, "generation_rouge_l_f1": 0.3206751054852321, "generation_token_f1": 0.44725738396624476, "prompt_id": "5d63d37b660c138f", "reading_canonical_log_prob": -238.90011596679688, "reading_distances": [38, 28, 37, 39], "reading_log_probs": [-240.27911376953125, -239.5877227783203, -240.54432678222656, -240.48605346679688], "reading_logprob_std": 0.4393810522176211, "tokenization_path_logprob_std": 60.89866870023057, "trajectory_balance_rmse": 2980.4770015483386} +{"canonical_length": 241, "canonical_nll": 1.238248579729642, "generation_rouge_l_f1": 0.27680311890838205, "generation_token_f1": 0.530214424951267, "prompt_id": "17f90ddfcc762d15", "reading_canonical_log_prob": -298.41790771484375, "reading_distances": [23, 25, 31, 31], "reading_log_probs": [-293.61944580078125, -293.12530517578125, -292.44049072265625, -291.68548583984375], "reading_logprob_std": 0.8409515489630682, "tokenization_path_logprob_std": 70.38040595129333, "trajectory_balance_rmse": 5486.0117301994305} +{"canonical_length": 61, "canonical_nll": 1.4501296496782146, "generation_rouge_l_f1": 0.2916666666666667, "generation_token_f1": 0.33333333333333337, "prompt_id": "9e46c646f8e1ae11", "reading_canonical_log_prob": -88.4579086303711, "reading_distances": [21, 19, 23, 26], "reading_log_probs": [-92.93342590332031, -90.57077026367188, -91.79147338867188, -90.96196746826172], "reading_logprob_std": 1.044991109320127, "tokenization_path_logprob_std": 51.6404063526513, "trajectory_balance_rmse": 1336.1190793444607} +{"canonical_length": 317, "canonical_nll": 1.1021628394863958, "generation_rouge_l_f1": 0.22030981067125643, "generation_token_f1": 0.495697074010327, "prompt_id": "c405755d9e3e8c9b", "reading_canonical_log_prob": -349.3856201171875, "reading_distances": [23, 26, 23, 21], "reading_log_probs": [-343.1683349609375, -343.5927734375, -343.98291015625, -342.37359619140625], "reading_logprob_std": 0.689431029615001, "tokenization_path_logprob_std": 166.87717863701423, "trajectory_balance_rmse": 6233.551488924517} +{"canonical_length": 225, "canonical_nll": 0.527074686686198, "generation_rouge_l_f1": 0.2832980972515856, "generation_token_f1": 0.46934460887949264, "prompt_id": "d5c04061128f24af", "reading_canonical_log_prob": -118.59180450439453, "reading_distances": [27, 26, 23, 22], "reading_log_probs": [-120.64326477050781, -118.35419464111328, -118.70472717285156, -120.28642272949219], "reading_logprob_std": 1.1359007289471978, "tokenization_path_logprob_std": 169.18424298301022, "trajectory_balance_rmse": 3489.350314223531} +{"canonical_length": 61, "canonical_nll": 0.8874180277840036, "generation_rouge_l_f1": 0.4150943396226416, "generation_token_f1": 0.490566037735849, "prompt_id": "1419b413999c869f", "reading_canonical_log_prob": -54.13249969482422, "reading_distances": [17, 13, 12, 12], "reading_log_probs": [-55.75307083129883, -58.8343505859375, -57.513031005859375, -55.422584533691406], "reading_logprob_std": 1.593135415259477, "tokenization_path_logprob_std": 29.886794206471407, "trajectory_balance_rmse": 753.1360460404584} +{"canonical_length": 116, "canonical_nll": 1.397442653261382, "generation_rouge_l_f1": 0.2908011869436202, "generation_token_f1": 0.4154302670623145, "prompt_id": "3501affcfd13e96c", "reading_canonical_log_prob": -162.1033477783203, "reading_distances": [24, 21, 25, 18], "reading_log_probs": [-164.84115600585938, -163.02743530273438, -162.8044891357422, -165.6965789794922], "reading_logprob_std": 1.4055710529138032, "tokenization_path_logprob_std": 92.00331963211053, "trajectory_balance_rmse": 2546.483421496666} +{"canonical_length": 26, "canonical_nll": 1.6881041893592248, "generation_rouge_l_f1": 0.27999999999999997, "generation_token_f1": 0.32, "prompt_id": "6fdd3868e470d9e0", "reading_canonical_log_prob": -43.890708923339844, "reading_distances": [57, 60, 47, 39], "reading_log_probs": [-47.8370361328125, -46.219879150390625, -47.19474792480469, -50.15983581542969], "reading_logprob_std": 1.6755223827797838, "tokenization_path_logprob_std": 46.91952961881307, "trajectory_balance_rmse": 636.0346810890262} +{"canonical_length": 304, "canonical_nll": 0.8123632732190584, "generation_rouge_l_f1": 0.3073770491803279, "generation_token_f1": 0.46311475409836067, "prompt_id": "1d8b10c0b2c4c1d8", "reading_canonical_log_prob": -246.95843505859375, "reading_distances": [27, 19, 19, 23], "reading_log_probs": [-245.30349731445312, -244.67279052734375, -242.73312377929688, -247.13296508789062], "reading_logprob_std": 1.814867603027664, "tokenization_path_logprob_std": 192.53044700295152, "trajectory_balance_rmse": 5025.324337744855} +{"canonical_length": 31, "canonical_nll": 0.8214729678246283, "generation_rouge_l_f1": 0.27450980392156865, "generation_token_f1": 0.37254901960784315, "prompt_id": "6f83d657c17ec17d", "reading_canonical_log_prob": -25.465662002563477, "reading_distances": [16, 19, 23, 24], "reading_log_probs": [-23.037677764892578, -26.8294734954834, -24.611164093017578, -24.658771514892578], "reading_logprob_std": 1.5576860293888626, "tokenization_path_logprob_std": 100.13237930245666, "trajectory_balance_rmse": 549.7251315902408} +{"canonical_length": 39, "canonical_nll": 1.2463110899313903, "generation_rouge_l_f1": 0.14814814814814814, "generation_token_f1": 0.18518518518518517, "prompt_id": "6587efe9dc00b7ac", "reading_canonical_log_prob": -48.60613250732422, "reading_distances": [24, 22, 21, 23], "reading_log_probs": [-46.008052825927734, -46.953365325927734, -47.14173126220703, -47.35907745361328], "reading_logprob_std": 0.5952193914553765, "tokenization_path_logprob_std": 107.9563628052935, "trajectory_balance_rmse": 981.2814561983264} +{"canonical_length": 15, "canonical_nll": 0.8853932698567708, "generation_rouge_l_f1": 0.64, "generation_token_f1": 0.64, "prompt_id": "ea95653fa930225e", "reading_canonical_log_prob": -13.280899047851562, "reading_distances": [15, 23, 15, 16], "reading_log_probs": [-13.290837287902832, -13.9151611328125, -12.94714641571045, -14.237403869628906], "reading_logprob_std": 0.5852028952950825, "tokenization_path_logprob_std": 55.51037774495607, "trajectory_balance_rmse": 442.6183128395485} +{"canonical_length": 218, "canonical_nll": 1.5437440084754874, "generation_rouge_l_f1": 0.2966101694915254, "generation_token_f1": 0.5296610169491526, "prompt_id": "dbfaa5f0bb7fb12c", "reading_canonical_log_prob": -336.53619384765625, "reading_distances": [38, 42, 40, 45], "reading_log_probs": [-334.9379577636719, -335.12957763671875, -333.74066162109375, -334.76171875], "reading_logprob_std": 0.619694947683055, "tokenization_path_logprob_std": 138.1795000281462, "trajectory_balance_rmse": 5270.889839190626} +{"canonical_length": 236, "canonical_nll": 1.1902804617154397, "generation_rouge_l_f1": 0.23647294589178355, "generation_token_f1": 0.5250501002004008, "prompt_id": "6ef66ee37f69a6e9", "reading_canonical_log_prob": -280.90618896484375, "reading_distances": [37, 45, 46, 48], "reading_log_probs": [-277.5804443359375, -282.7792663574219, -280.27850341796875, -278.60784912109375], "reading_logprob_std": 2.269516406497, "tokenization_path_logprob_std": 265.2816382681026, "trajectory_balance_rmse": 5898.552246650889} +{"canonical_length": 254, "canonical_nll": 0.3187504265252061, "generation_rouge_l_f1": 0.4659090909090909, "generation_token_f1": 0.6515151515151515, "prompt_id": "e2bfc09d77b30452", "reading_canonical_log_prob": -80.96260833740234, "reading_distances": [21, 26, 24, 23], "reading_log_probs": [-78.76698303222656, -79.68357849121094, -80.13273620605469, -79.82231140136719], "reading_logprob_std": 0.5871186681934311, "tokenization_path_logprob_std": 90.19354046101995, "trajectory_balance_rmse": 3473.9387269312156} +{"canonical_length": 315, "canonical_nll": 0.8981214735243056, "generation_rouge_l_f1": 0.17832167832167836, "generation_token_f1": 0.37762237762237766, "prompt_id": "3024b6f11a619ed8", "reading_canonical_log_prob": -282.90826416015625, "reading_distances": [16, 13, 15, 12], "reading_log_probs": [-282.05389404296875, -281.7874755859375, -281.99072265625, -282.1365966796875], "reading_logprob_std": 0.14896400148827074, "tokenization_path_logprob_std": 106.09234621011598, "trajectory_balance_rmse": 5997.961388948732} +{"canonical_length": 147, "canonical_nll": 1.5151230169802297, "generation_rouge_l_f1": 0.2024691358024691, "generation_token_f1": 0.37530864197530867, "prompt_id": "2027b774c4dd5ef1", "reading_canonical_log_prob": -222.72308349609375, "reading_distances": [79, 81, 68, 82], "reading_log_probs": [-221.69338989257812, -218.90467834472656, -225.35818481445312, -229.14456176757812], "reading_logprob_std": 4.449411992301731, "tokenization_path_logprob_std": 17.37954339351171, "trajectory_balance_rmse": 3237.8500451544546} +{"canonical_length": 320, "canonical_nll": 1.0291313171386718, "generation_rouge_l_f1": 0.19813084112149534, "generation_token_f1": 0.3962616822429907, "prompt_id": "ea4a6ac8af4f585f", "reading_canonical_log_prob": -329.322021484375, "reading_distances": [29, 33, 37, 34], "reading_log_probs": [-329.638671875, -327.6676940917969, -329.38677978515625, -328.823486328125], "reading_logprob_std": 0.8766005055536757, "tokenization_path_logprob_std": 116.50497378111487, "trajectory_balance_rmse": 6764.393853866954} +{"canonical_length": 142, "canonical_nll": 1.2772567104285872, "generation_rouge_l_f1": 0.27100271002710025, "generation_token_f1": 0.42276422764227645, "prompt_id": "e9a730206e7c88d2", "reading_canonical_log_prob": -181.37045288085938, "reading_distances": [33, 22, 21, 22], "reading_log_probs": [-181.12701416015625, -183.40460205078125, -184.03121948242188, -183.17234802246094], "reading_logprob_std": 1.257956586616107, "tokenization_path_logprob_std": 67.45382374459086, "trajectory_balance_rmse": 2944.9846262774} +{"canonical_length": 324, "canonical_nll": 0.9652402901355132, "generation_rouge_l_f1": 0.2309124767225326, "generation_token_f1": 0.44320297951582865, "prompt_id": "21922a55d95c3a47", "reading_canonical_log_prob": -312.73785400390625, "reading_distances": [18, 19, 18, 19], "reading_log_probs": [-305.088134765625, -304.0833740234375, -305.5461120605469, -307.2598876953125], "reading_logprob_std": 1.3261054111709683, "tokenization_path_logprob_std": 97.99058071948973, "trajectory_balance_rmse": 6943.77289824865} +{"canonical_length": 17, "canonical_nll": 0.42429469613467946, "generation_rouge_l_f1": 0.41379310344827586, "generation_token_f1": 0.41379310344827586, "prompt_id": "732a8e545760e4c1", "reading_canonical_log_prob": -7.213009834289551, "reading_distances": [24, 22, 16, 17], "reading_log_probs": [-8.770832061767578, -8.163907051086426, -9.077950477600098, -7.809226036071777], "reading_logprob_std": 0.5743330592293037, "tokenization_path_logprob_std": 55.037125721791206, "trajectory_balance_rmse": 357.27502320250693} +{"canonical_length": 8, "canonical_nll": 0.6484816074371338, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.5714285714285715, "prompt_id": "530dfa24f61035c2", "reading_canonical_log_prob": -5.18785285949707, "reading_distances": [34, 32, 26, 35], "reading_log_probs": [-4.328393936157227, -5.533730983734131, -4.447519779205322, -5.269660949707031], "reading_logprob_std": 0.5971122049420594, "tokenization_path_logprob_std": 33.23657830672138, "trajectory_balance_rmse": 120.93319366749455} +{"canonical_length": 8, "canonical_nll": 1.8311448097229004, "generation_rouge_l_f1": 0.39999999999999997, "generation_token_f1": 0.39999999999999997, "prompt_id": "ab024353313a6e65", "reading_canonical_log_prob": -14.649158477783203, "reading_distances": [27, 29, 36, 33], "reading_log_probs": [-12.99822998046875, -16.441980361938477, -17.54690933227539, -16.469623565673828], "reading_logprob_std": 1.9786917133540702, "tokenization_path_logprob_std": 8.274022792959569, "trajectory_balance_rmse": 146.83044065365894} +{"canonical_length": 12, "canonical_nll": 1.4880717595418294, "generation_rouge_l_f1": 0.03125000000000001, "generation_token_f1": 0.03125000000000001, "prompt_id": "2321d8df80c63fef", "reading_canonical_log_prob": -17.856861114501953, "reading_distances": [20, 20, 20, 16], "reading_log_probs": [-18.090238571166992, -19.520427703857422, -18.612449645996094, -17.911298751831055], "reading_logprob_std": 0.7220051670567349, "tokenization_path_logprob_std": 48.51163448657957, "trajectory_balance_rmse": 256.35079985851223} +{"canonical_length": 16, "canonical_nll": 0.47296690940856934, "generation_rouge_l_f1": 0.35294117647058826, "generation_token_f1": 0.35294117647058826, "prompt_id": "f3f8000f2b6cf3b2", "reading_canonical_log_prob": -7.567470550537109, "reading_distances": [5, 7, 4, 12], "reading_log_probs": [-8.417765617370605, -7.893529415130615, -8.438138961791992, -8.370965957641602], "reading_logprob_std": 0.25924349879003955, "tokenization_path_logprob_std": 18.650112543125424, "trajectory_balance_rmse": 195.5327595344084} +{"canonical_length": 8, "canonical_nll": 0.16960377991199493, "generation_rouge_l_f1": 0.9333333333333333, "generation_token_f1": 0.9333333333333333, "prompt_id": "30a8ec8b58b46182", "reading_canonical_log_prob": -1.3568302392959595, "reading_distances": [28, 24, 26, 26], "reading_log_probs": [-1.4369568824768066, -1.4359946250915527, -1.6734776496887207, -1.5183124542236328], "reading_logprob_std": 0.11173345446664472, "tokenization_path_logprob_std": 23.892006301047726, "trajectory_balance_rmse": 56.65588274561612} +{"canonical_length": 267, "canonical_nll": 1.382497037394663, "generation_rouge_l_f1": 0.23507462686567163, "generation_token_f1": 0.5111940298507462, "prompt_id": "f1f8d8a792fc708b", "reading_canonical_log_prob": -369.126708984375, "reading_distances": [24, 28, 29, 30], "reading_log_probs": [-364.74078369140625, -368.613037109375, -365.34576416015625, -365.1505432128906], "reading_logprob_std": 1.7848972540953667, "tokenization_path_logprob_std": 234.82884165485518, "trajectory_balance_rmse": 6061.301493071675} +{"canonical_length": 194, "canonical_nll": 0.9131323430956024, "generation_rouge_l_f1": 0.2806236080178174, "generation_token_f1": 0.5478841870824054, "prompt_id": "1d19c88736fcd9d4", "reading_canonical_log_prob": -177.14767456054688, "reading_distances": [52, 48, 29, 47], "reading_log_probs": [-180.4954833984375, -179.95608520507812, -177.97286987304688, -181.69921875], "reading_logprob_std": 1.5535132701796144, "tokenization_path_logprob_std": 187.25397339132303, "trajectory_balance_rmse": 3796.607492700562} +{"canonical_length": 15, "canonical_nll": 2.8495437622070314, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "1dc2cb66ddb80668", "reading_canonical_log_prob": -42.74315643310547, "reading_distances": [15, 15, 15, 13], "reading_log_probs": [-43.31911087036133, -43.73969650268555, -44.54909133911133, -44.444766998291016], "reading_logprob_std": 0.5859446730894864, "tokenization_path_logprob_std": 47.83501653853431, "trajectory_balance_rmse": 346.19381609515887} +{"canonical_length": 16, "canonical_nll": 1.6151809692382812, "generation_rouge_l_f1": 0.125, "generation_token_f1": 0.16071428571428573, "prompt_id": "bf66528e05d6425b", "reading_canonical_log_prob": -25.8428955078125, "reading_distances": [29, 37, 31, 36], "reading_log_probs": [-26.14146614074707, -27.363250732421875, -27.036258697509766, -26.72234344482422], "reading_logprob_std": 0.5201809363532562, "tokenization_path_logprob_std": 81.67021382386213, "trajectory_balance_rmse": 583.2488275078778} +{"canonical_length": 118, "canonical_nll": 1.1752993050268141, "generation_rouge_l_f1": 0.26119402985074625, "generation_token_f1": 0.44029850746268656, "prompt_id": "471e2be2592f5add", "reading_canonical_log_prob": -138.68531799316406, "reading_distances": [20, 22, 21, 20], "reading_log_probs": [-139.8756561279297, -135.71749877929688, -136.53411865234375, -137.02981567382812], "reading_logprob_std": 1.8071550457549523, "tokenization_path_logprob_std": 120.05371107541889, "trajectory_balance_rmse": 2707.240458546125} +{"canonical_length": 19, "canonical_nll": 3.717860171669408, "generation_rouge_l_f1": 0.03745318352059925, "generation_token_f1": 0.052434456928838954, "prompt_id": "57103ee2e4cc89e4", "reading_canonical_log_prob": -70.63934326171875, "reading_distances": [36, 32, 40, 39], "reading_log_probs": [-70.4271240234375, -70.304931640625, -68.78533935546875, -70.36605834960938], "reading_logprob_std": 0.791922061453932, "tokenization_path_logprob_std": 30.73673817350998, "trajectory_balance_rmse": 281.11893492755786} +{"canonical_length": 23, "canonical_nll": 1.6436502207880435, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.380952380952381, "prompt_id": "aa0821e4d154fd43", "reading_canonical_log_prob": -37.803955078125, "reading_distances": [20, 22, 25, 21], "reading_log_probs": [-39.99870300292969, -40.06041717529297, -38.346275329589844, -38.5179328918457], "reading_logprob_std": 0.9252933015680962, "tokenization_path_logprob_std": 40.38622457381602, "trajectory_balance_rmse": 537.7361242865945} +{"canonical_length": 24, "canonical_nll": 1.497607707977295, "generation_rouge_l_f1": 0.3050847457627119, "generation_token_f1": 0.3050847457627119, "prompt_id": "2aaa42d342162266", "reading_canonical_log_prob": -35.94258499145508, "reading_distances": [31, 32, 27, 28], "reading_log_probs": [-39.04584503173828, -38.196327209472656, -38.15443801879883, -38.76524353027344], "reading_logprob_std": 0.43718112692809774, "tokenization_path_logprob_std": 81.82755785898406, "trajectory_balance_rmse": 707.3855071925208} +{"canonical_length": 10, "canonical_nll": 0.5068389415740967, "generation_rouge_l_f1": 0.12121212121212123, "generation_token_f1": 0.14141414141414144, "prompt_id": "b5604eaaebf6e52c", "reading_canonical_log_prob": -5.068389415740967, "reading_distances": [19, 20, 12, 15], "reading_log_probs": [-6.361121654510498, -5.599759101867676, -5.041544437408447, -6.712644100189209], "reading_logprob_std": 0.7520646975897948, "tokenization_path_logprob_std": 48.359325039540686, "trajectory_balance_rmse": 242.21968741935822} +{"canonical_length": 223, "canonical_nll": 1.296432974092629, "generation_rouge_l_f1": 0.21698113207547173, "generation_token_f1": 0.43396226415094347, "prompt_id": "5dbcfc1c200ae9f7", "reading_canonical_log_prob": -289.10455322265625, "reading_distances": [23, 24, 25, 24], "reading_log_probs": [-287.68463134765625, -292.27166748046875, -286.736083984375, -287.42596435546875], "reading_logprob_std": 2.5266415380441836, "tokenization_path_logprob_std": 259.66548468775164, "trajectory_balance_rmse": 5313.1884141131} +{"canonical_length": 163, "canonical_nll": 1.2715568308449963, "generation_rouge_l_f1": 0.23111111111111113, "generation_token_f1": 0.33777777777777773, "prompt_id": "d526cee0379fafdd", "reading_canonical_log_prob": -207.26376342773438, "reading_distances": [34, 29, 22, 30], "reading_log_probs": [-205.12847900390625, -205.22244262695312, -205.6220703125, -206.19467163085938], "reading_logprob_std": 0.48494261284399903, "tokenization_path_logprob_std": 136.83346534206981, "trajectory_balance_rmse": 4330.003986393989} +{"canonical_length": 85, "canonical_nll": 1.4942421408260569, "generation_rouge_l_f1": 0.1901840490797546, "generation_token_f1": 0.3374233128834356, "prompt_id": "b641694c0de50f89", "reading_canonical_log_prob": -127.01058197021484, "reading_distances": [35, 25, 31, 37], "reading_log_probs": [-133.0115509033203, -130.1949462890625, -132.33413696289062, -132.00233459472656], "reading_logprob_std": 1.2028975709802487, "tokenization_path_logprob_std": 67.15801055319973, "trajectory_balance_rmse": 1899.6484318034675} +{"canonical_length": 13, "canonical_nll": 0.7719710423396184, "generation_rouge_l_f1": 0.21568627450980393, "generation_token_f1": 0.21568627450980393, "prompt_id": "2e5bfbd760499e1d", "reading_canonical_log_prob": -10.035623550415039, "reading_distances": [22, 24, 25, 19], "reading_log_probs": [-9.788800239562988, -9.200963020324707, -9.417037010192871, -9.920287132263184], "reading_logprob_std": 0.3314645695574486, "tokenization_path_logprob_std": 27.45378931018075, "trajectory_balance_rmse": 380.657045849978} +{"canonical_length": 363, "canonical_nll": 1.3146577525073175, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.46386554621848736, "prompt_id": "1d0a6d1714584f14", "reading_canonical_log_prob": -477.22076416015625, "reading_distances": [36, 39, 33, 28], "reading_log_probs": [-473.474365234375, -475.21612548828125, -473.997314453125, -476.2505187988281], "reading_logprob_std": 1.246549670247266, "tokenization_path_logprob_std": 146.48085191401344, "trajectory_balance_rmse": 7419.118240133673} +{"canonical_length": 37, "canonical_nll": 1.6905125798405827, "generation_rouge_l_f1": 0.05405405405405406, "generation_token_f1": 0.05405405405405406, "prompt_id": "c7b4cbd5453e19a7", "reading_canonical_log_prob": -62.54896545410156, "reading_distances": [24, 21, 16, 21], "reading_log_probs": [-58.46968460083008, -59.50751495361328, -58.802589416503906, -58.32017135620117], "reading_logprob_std": 0.5283388814535547, "tokenization_path_logprob_std": 56.38849000998463, "trajectory_balance_rmse": 779.1067083909483} +{"canonical_length": 10, "canonical_nll": 0.27713062763214114, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "88beef2f88591814", "reading_canonical_log_prob": -2.771306276321411, "reading_distances": [62, 63, 57, 70], "reading_log_probs": [-6.315895080566406, -8.628902435302734, -5.253261566162109, -6.87729549407959], "reading_logprob_std": 1.4111144468785715, "tokenization_path_logprob_std": 27.764871250210643, "trajectory_balance_rmse": 289.08870530913305} +{"canonical_length": 65, "canonical_nll": 0.9475210336538461, "generation_rouge_l_f1": 0.5555555555555555, "generation_token_f1": 0.6161616161616161, "prompt_id": "cfc50376a0f13c13", "reading_canonical_log_prob": -61.5888671875, "reading_distances": [28, 24, 30, 17], "reading_log_probs": [-61.71051788330078, -62.05835723876953, -62.46890640258789, -61.63974380493164], "reading_logprob_std": 0.3799619246080208, "tokenization_path_logprob_std": 48.46941338143713, "trajectory_balance_rmse": 676.619977362771} +{"canonical_length": 90, "canonical_nll": 1.5534064398871528, "generation_rouge_l_f1": 0.22710622710622713, "generation_token_f1": 0.3516483516483516, "prompt_id": "6a7a68d7cf446159", "reading_canonical_log_prob": -139.80657958984375, "reading_distances": [21, 18, 14, 18], "reading_log_probs": [-140.8164825439453, -142.73202514648438, -142.09442138671875, -140.52841186523438], "reading_logprob_std": 1.0448385733807795, "tokenization_path_logprob_std": 86.33646468073873, "trajectory_balance_rmse": 2195.733019358883} +{"canonical_length": 216, "canonical_nll": 1.6070789761013455, "generation_rouge_l_f1": 0.1768421052631579, "generation_token_f1": 0.4000000000000001, "prompt_id": "a929bd728f13cca9", "reading_canonical_log_prob": -347.1290588378906, "reading_distances": [35, 36, 29, 38], "reading_log_probs": [-339.97869873046875, -338.81329345703125, -342.056884765625, -338.7053527832031], "reading_logprob_std": 1.5562706208791866, "tokenization_path_logprob_std": 242.24398853740828, "trajectory_balance_rmse": 5736.229146415323} +{"canonical_length": 321, "canonical_nll": 0.9630893220039914, "generation_rouge_l_f1": 0.259567387687188, "generation_token_f1": 0.502495840266223, "prompt_id": "167bbbea26df8184", "reading_canonical_log_prob": -309.15167236328125, "reading_distances": [31, 34, 34, 37], "reading_log_probs": [-309.67913818359375, -310.72467041015625, -309.1728820800781, -308.76287841796875], "reading_logprob_std": 0.8472378002540322, "tokenization_path_logprob_std": 177.75348424469848, "trajectory_balance_rmse": 6489.039581997167} +{"canonical_length": 195, "canonical_nll": 1.7469078650841345, "generation_rouge_l_f1": 0.23640661938534277, "generation_token_f1": 0.45390070921985815, "prompt_id": "00b0a2b88466f41c", "reading_canonical_log_prob": -340.64703369140625, "reading_distances": [21, 26, 16, 25], "reading_log_probs": [-339.15472412109375, -339.04840087890625, -338.853759765625, -339.157958984375], "reading_logprob_std": 0.1426884559608236, "tokenization_path_logprob_std": 102.17562171297544, "trajectory_balance_rmse": 4249.084889970672} +{"canonical_length": 89, "canonical_nll": 1.0358069773470417, "generation_rouge_l_f1": 0.33492822966507174, "generation_token_f1": 0.44976076555023925, "prompt_id": "a26fac121253e753", "reading_canonical_log_prob": -92.18682098388672, "reading_distances": [29, 27, 29, 31], "reading_log_probs": [-93.79595947265625, -93.34640502929688, -94.91590118408203, -93.24034881591797], "reading_logprob_std": 0.7663285672150507, "tokenization_path_logprob_std": 104.50695536031145, "trajectory_balance_rmse": 2213.9749176052433} +{"canonical_length": 223, "canonical_nll": 1.07794278405707, "generation_rouge_l_f1": 0.2585034013605442, "generation_token_f1": 0.39455782312925164, "prompt_id": "109439f51e6651d1", "reading_canonical_log_prob": -240.38124084472656, "reading_distances": [38, 33, 38, 41], "reading_log_probs": [-238.82566833496094, -238.51390075683594, -237.77032470703125, -236.57235717773438], "reading_logprob_std": 1.0019141421842968, "tokenization_path_logprob_std": 270.2848492234258, "trajectory_balance_rmse": 3892.1131088498205} +{"canonical_length": 174, "canonical_nll": 1.0006267503760327, "generation_rouge_l_f1": 0.28899082568807344, "generation_token_f1": 0.5045871559633027, "prompt_id": "25f04ac4bbf2d3c3", "reading_canonical_log_prob": -174.1090545654297, "reading_distances": [22, 20, 16, 20], "reading_log_probs": [-173.7569580078125, -172.3530731201172, -173.96163940429688, -173.85629272460938], "reading_logprob_std": 0.7572377010543907, "tokenization_path_logprob_std": 126.89961885024783, "trajectory_balance_rmse": 3747.613783213431} +{"canonical_length": 13, "canonical_nll": 0.8712684924785907, "generation_rouge_l_f1": 0.06521739130434784, "generation_token_f1": 0.08695652173913043, "prompt_id": "68cd67de83285a28", "reading_canonical_log_prob": -11.32649040222168, "reading_distances": [10, 9, 11, 10], "reading_log_probs": [-10.588942527770996, -10.089775085449219, -9.609841346740723, -10.595996856689453], "reading_logprob_std": 0.4714301456522117, "tokenization_path_logprob_std": 18.227874551692608, "trajectory_balance_rmse": 187.6586106123824} +{"canonical_length": 12, "canonical_nll": 0.6210424105326334, "generation_rouge_l_f1": 0.17391304347826084, "generation_token_f1": 0.17391304347826084, "prompt_id": "41752c3a8cc3cb46", "reading_canonical_log_prob": -7.452508926391602, "reading_distances": [20, 21, 21, 19], "reading_log_probs": [-8.052483558654785, -7.454339027404785, -7.685448169708252, -7.800356864929199], "reading_logprob_std": 0.248730696817379, "tokenization_path_logprob_std": 38.410411445504344, "trajectory_balance_rmse": 237.96475935660843} +{"canonical_length": 33, "canonical_nll": 0.4446426449400006, "generation_rouge_l_f1": 0.2891566265060241, "generation_token_f1": 0.5301204819277109, "prompt_id": "6595c3b5c5d331ea", "reading_canonical_log_prob": -14.67320728302002, "reading_distances": [24, 21, 22, 23], "reading_log_probs": [-16.481136322021484, -15.993350982666016, -17.76504898071289, -18.150447845458984], "reading_logprob_std": 1.025243067578836, "tokenization_path_logprob_std": 60.060230278329364, "trajectory_balance_rmse": 373.6215214936893} +{"canonical_length": 7, "canonical_nll": 0.4420816898345947, "generation_rouge_l_f1": 0.007633587786259542, "generation_token_f1": 0.022900763358778626, "prompt_id": "77f51b75ebf88292", "reading_canonical_log_prob": -3.094571828842163, "reading_distances": [19, 22, 19, 26], "reading_log_probs": [-4.758390426635742, -3.6307566165924072, -4.644341945648193, -3.562429189682007], "reading_logprob_std": 0.6401462992897271, "tokenization_path_logprob_std": 38.112672827112824, "trajectory_balance_rmse": 142.56719567518323} +{"canonical_length": 250, "canonical_nll": 1.09366259765625, "generation_rouge_l_f1": 0.204, "generation_token_f1": 0.48000000000000004, "prompt_id": "3d42fc04af88620e", "reading_canonical_log_prob": -273.4156494140625, "reading_distances": [29, 25, 33, 29], "reading_log_probs": [-271.193115234375, -269.47900390625, -268.95465087890625, -268.9371337890625], "reading_logprob_std": 1.0648641841070157, "tokenization_path_logprob_std": 80.41009121931509, "trajectory_balance_rmse": 4991.735340126275} +{"canonical_length": 311, "canonical_nll": 0.627308051302502, "generation_rouge_l_f1": 0.27039627039627034, "generation_token_f1": 0.40559440559440557, "prompt_id": "391687d0204ea44e", "reading_canonical_log_prob": -195.09280395507812, "reading_distances": [21, 24, 23, 26], "reading_log_probs": [-198.76637268066406, -201.72866821289062, -197.48651123046875, -198.50698852539062], "reading_logprob_std": 1.8233899307280257, "tokenization_path_logprob_std": 139.7783146604095, "trajectory_balance_rmse": 4511.665364824812} +{"canonical_length": 45, "canonical_nll": 1.6297320895724827, "generation_rouge_l_f1": 0.12546125461254612, "generation_token_f1": 0.16974169741697417, "prompt_id": "9043379228e84740", "reading_canonical_log_prob": -73.33794403076172, "reading_distances": [32, 31, 36, 33], "reading_log_probs": [-70.4664306640625, -73.68777465820312, -71.97529602050781, -72.72200012207031], "reading_logprob_std": 1.3590497883588855, "tokenization_path_logprob_std": 86.04307228573036, "trajectory_balance_rmse": 1117.4333696081458} +{"canonical_length": 14, "canonical_nll": 1.3450843266078405, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.33333333333333337, "prompt_id": "82a1a337beb39341", "reading_canonical_log_prob": -18.831180572509766, "reading_distances": [26, 23, 25, 34], "reading_log_probs": [-18.70366859436035, -17.751184463500977, -23.23491096496582, -19.99713897705078], "reading_logprob_std": 2.392891983513123, "tokenization_path_logprob_std": 59.37712164336867, "trajectory_balance_rmse": 371.95818350744537} +{"canonical_length": 15, "canonical_nll": 1.2520490010579428, "generation_rouge_l_f1": 0.05240174672489083, "generation_token_f1": 0.06986899563318777, "prompt_id": "32a6cf563f35ecba", "reading_canonical_log_prob": -18.78073501586914, "reading_distances": [33, 25, 30, 33], "reading_log_probs": [-21.18328094482422, -20.357707977294922, -16.695362091064453, -15.207244873046875], "reading_logprob_std": 2.867788548792163, "tokenization_path_logprob_std": 30.73238994668343, "trajectory_balance_rmse": 198.6955412093822} +{"canonical_length": 147, "canonical_nll": 1.4597117106119792, "generation_rouge_l_f1": 0.2402826855123675, "generation_token_f1": 0.4381625441696113, "prompt_id": "2655c987f350c6c7", "reading_canonical_log_prob": -214.57762145996094, "reading_distances": [14, 17, 14, 15], "reading_log_probs": [-214.42861938476562, -215.39144897460938, -215.09860229492188, -215.82406616210938], "reading_logprob_std": 0.5861174696465309, "tokenization_path_logprob_std": 62.20198297123005, "trajectory_balance_rmse": 3222.1164727421165} +{"canonical_length": 392, "canonical_nll": 0.9795366404007892, "generation_rouge_l_f1": 0.19816513761467888, "generation_token_f1": 0.3119266055045872, "prompt_id": "dd04ff95cd33338f", "reading_canonical_log_prob": -383.9783630371094, "reading_distances": [25, 30, 24, 23], "reading_log_probs": [-388.3602294921875, -394.707763671875, -387.6715393066406, -387.87030029296875], "reading_logprob_std": 3.3826072249536656, "tokenization_path_logprob_std": 132.73152335795118, "trajectory_balance_rmse": 4453.911683679335} +{"canonical_length": 197, "canonical_nll": 0.5887563482758963, "generation_rouge_l_f1": 0.2880324543610548, "generation_token_f1": 0.47058823529411764, "prompt_id": "f51cb0e119af1f71", "reading_canonical_log_prob": -115.98500061035156, "reading_distances": [38, 44, 38, 39], "reading_log_probs": [-114.60447692871094, -114.30337524414062, -113.53413391113281, -116.77949523925781], "reading_logprob_std": 1.3911108768312812, "tokenization_path_logprob_std": 190.6978312979583, "trajectory_balance_rmse": 3779.507143882104} +{"canonical_length": 11, "canonical_nll": 1.4898967742919922, "generation_rouge_l_f1": 0.03731343283582089, "generation_token_f1": 0.06716417910447763, "prompt_id": "40f09143e7b23121", "reading_canonical_log_prob": -16.388864517211914, "reading_distances": [42, 47, 49, 42], "reading_log_probs": [-18.88812255859375, -19.67928123474121, -19.016368865966797, -19.046062469482422], "reading_logprob_std": 0.3545683649511356, "tokenization_path_logprob_std": 87.42880639605445, "trajectory_balance_rmse": 321.31687934579406} +{"canonical_length": 98, "canonical_nll": 1.3320113201530612, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.3641456582633053, "prompt_id": "09e3575b08c40649", "reading_canonical_log_prob": -130.537109375, "reading_distances": [16, 17, 13, 13], "reading_log_probs": [-133.18963623046875, -135.07383728027344, -132.31021118164062, -133.05035400390625], "reading_logprob_std": 1.1769638457829066, "tokenization_path_logprob_std": 64.06588931259189, "trajectory_balance_rmse": 2121.674931689771} +{"canonical_length": 11, "canonical_nll": 0.8951065757057883, "generation_rouge_l_f1": 0.09271523178807949, "generation_token_f1": 0.10596026490066225, "prompt_id": "fc90d5db5297a0af", "reading_canonical_log_prob": -9.846172332763672, "reading_distances": [18, 22, 20, 20], "reading_log_probs": [-10.466665267944336, -10.022377967834473, -9.388185501098633, -9.813222885131836], "reading_logprob_std": 0.4485253125057693, "tokenization_path_logprob_std": 36.724112541034984, "trajectory_balance_rmse": 287.02634467365056} +{"canonical_length": 244, "canonical_nll": 1.041660684054015, "generation_rouge_l_f1": 0.35655737704918034, "generation_token_f1": 0.6188524590163934, "prompt_id": "e15ab7087e38fd8b", "reading_canonical_log_prob": -254.1652069091797, "reading_distances": [21, 16, 16, 15], "reading_log_probs": [-254.15025329589844, -252.9658660888672, -253.44912719726562, -253.1430206298828], "reading_logprob_std": 0.5218153222756854, "tokenization_path_logprob_std": 90.05786695471807, "trajectory_balance_rmse": 4857.419791851077} +{"canonical_length": 83, "canonical_nll": 1.0038141457431287, "generation_rouge_l_f1": 0.4503311258278146, "generation_token_f1": 0.5827814569536424, "prompt_id": "514750b55e58033f", "reading_canonical_log_prob": -83.31657409667969, "reading_distances": [17, 18, 16, 13], "reading_log_probs": [-84.05999755859375, -81.23567962646484, -83.81570434570312, -82.43264770507812], "reading_logprob_std": 1.3129744953610518, "tokenization_path_logprob_std": 195.61962358831272, "trajectory_balance_rmse": 1956.9557173032829} +{"canonical_length": 145, "canonical_nll": 1.2634502542429957, "generation_rouge_l_f1": 0.18004866180048662, "generation_token_f1": 0.340632603406326, "prompt_id": "612827ad6442cf24", "reading_canonical_log_prob": -183.20028686523438, "reading_distances": [27, 26, 28, 29], "reading_log_probs": [-182.78182983398438, -182.62533569335938, -182.41952514648438, -182.50367736816406], "reading_logprob_std": 0.1574177622824831, "tokenization_path_logprob_std": 235.92929946727878, "trajectory_balance_rmse": 3862.0320854614265} +{"canonical_length": 184, "canonical_nll": 0.7411635025687839, "generation_rouge_l_f1": 0.20323325635103928, "generation_token_f1": 0.40646651270207856, "prompt_id": "637d5437d0c071a7", "reading_canonical_log_prob": -136.37408447265625, "reading_distances": [249, 215, 215, 227], "reading_log_probs": [-140.0035400390625, -142.89083862304688, -137.208984375, -140.31533813476562], "reading_logprob_std": 2.3239577711865698, "tokenization_path_logprob_std": 124.4988376536522, "trajectory_balance_rmse": 3131.988711759372} +{"canonical_length": 65, "canonical_nll": 1.0108121431790866, "generation_rouge_l_f1": 0.17763157894736845, "generation_token_f1": 0.24342105263157893, "prompt_id": "c97d2c584d8df586", "reading_canonical_log_prob": -65.70278930664062, "reading_distances": [15, 19, 16, 17], "reading_log_probs": [-66.4390869140625, -66.57890319824219, -66.66018676757812, -67.05473327636719], "reading_logprob_std": 0.2639669189272765, "tokenization_path_logprob_std": 60.27480577076538, "trajectory_balance_rmse": 995.6963066048596} +{"canonical_length": 17, "canonical_nll": 0.8292714848237879, "generation_rouge_l_f1": 0.7333333333333333, "generation_token_f1": 0.8000000000000002, "prompt_id": "a17a07dc87573da6", "reading_canonical_log_prob": -14.097615242004395, "reading_distances": [29, 25, 21, 24], "reading_log_probs": [-14.822043418884277, -14.338953971862793, -14.794332504272461, -14.403488159179688], "reading_logprob_std": 0.2539070343717948, "tokenization_path_logprob_std": 90.7021647972394, "trajectory_balance_rmse": 486.03673778186936} +{"canonical_length": 13, "canonical_nll": 0.8461294907789964, "generation_rouge_l_f1": 0.13698630136986303, "generation_token_f1": 0.19178082191780818, "prompt_id": "4b95b246a8b6c8d6", "reading_canonical_log_prob": -10.999683380126953, "reading_distances": [10, 12, 14, 12], "reading_log_probs": [-12.004959106445312, -11.004716873168945, -11.400771141052246, -12.72983169555664], "reading_logprob_std": 0.7522327866796679, "tokenization_path_logprob_std": 27.43786712538929, "trajectory_balance_rmse": 351.12092780221457} +{"canonical_length": 99, "canonical_nll": 1.1795099431818181, "generation_rouge_l_f1": 0.44827586206896547, "generation_token_f1": 0.6321839080459771, "prompt_id": "0c93e5e51c22b30d", "reading_canonical_log_prob": -116.771484375, "reading_distances": [17, 18, 18, 15], "reading_log_probs": [-118.45502471923828, -120.12156677246094, -120.20025634765625, -120.87773132324219], "reading_logprob_std": 1.0299535734165384, "tokenization_path_logprob_std": 109.51700274060791, "trajectory_balance_rmse": 2152.6642470105226} +{"canonical_length": 306, "canonical_nll": 1.1540238124872344, "generation_rouge_l_f1": 0.1840277777777778, "generation_token_f1": 0.40624999999999994, "prompt_id": "6c10022ee752e034", "reading_canonical_log_prob": -353.13128662109375, "reading_distances": [29, 34, 30, 36], "reading_log_probs": [-360.137451171875, -358.75201416015625, -359.16668701171875, -361.8253173828125], "reading_logprob_std": 1.3661433806041652, "tokenization_path_logprob_std": 210.19453426283889, "trajectory_balance_rmse": 6687.268973756561} +{"canonical_length": 69, "canonical_nll": 1.0761325117470562, "generation_rouge_l_f1": 0.30344827586206896, "generation_token_f1": 0.5517241379310345, "prompt_id": "7884c89879ae1c04", "reading_canonical_log_prob": -74.25314331054688, "reading_distances": [35, 28, 26, 37], "reading_log_probs": [-76.31687927246094, -75.66316223144531, -76.6555404663086, -76.73092651367188], "reading_logprob_std": 0.48683421814251093, "tokenization_path_logprob_std": 126.51399358610088, "trajectory_balance_rmse": 1348.6878580977952} +{"canonical_length": 46, "canonical_nll": 0.9491802713145381, "generation_rouge_l_f1": 0.4507042253521127, "generation_token_f1": 0.5352112676056338, "prompt_id": "803eb754dd6450de", "reading_canonical_log_prob": -43.66229248046875, "reading_distances": [14, 15, 17, 13], "reading_log_probs": [-41.87321472167969, -47.373294830322266, -45.231895446777344, -44.9771728515625], "reading_logprob_std": 2.264914438121763, "tokenization_path_logprob_std": 70.67796946049782, "trajectory_balance_rmse": 1170.4949389223195} +{"canonical_length": 12, "canonical_nll": 1.1384929021199544, "generation_rouge_l_f1": 0.28571428571428564, "generation_token_f1": 0.28571428571428564, "prompt_id": "7f58078c29e97e0d", "reading_canonical_log_prob": -13.661914825439453, "reading_distances": [39, 46, 39, 39], "reading_log_probs": [-18.145999908447266, -14.609289169311523, -14.380669593811035, -19.908214569091797], "reading_logprob_std": 2.7153278883853256, "tokenization_path_logprob_std": 15.521989877870267, "trajectory_balance_rmse": 199.24633976239375} +{"canonical_length": 135, "canonical_nll": 1.414033677842882, "generation_rouge_l_f1": 0.2074074074074074, "generation_token_f1": 0.3111111111111111, "prompt_id": "979bfe699ee39dca", "reading_canonical_log_prob": -190.89454650878906, "reading_distances": [22, 21, 20, 29], "reading_log_probs": [-188.2098846435547, -190.09329223632812, -188.18231201171875, -187.61474609375], "reading_logprob_std": 1.080869317377735, "tokenization_path_logprob_std": 176.3153426198575, "trajectory_balance_rmse": 2682.518323423148} +{"canonical_length": 13, "canonical_nll": 1.0363433544452374, "generation_rouge_l_f1": 0.051502145922746795, "generation_token_f1": 0.07725321888412018, "prompt_id": "f30d075da952702b", "reading_canonical_log_prob": -13.472463607788086, "reading_distances": [34, 26, 31, 29], "reading_log_probs": [-17.139190673828125, -15.136229515075684, -14.020369529724121, -15.847829818725586], "reading_logprob_std": 1.3069570637366348, "tokenization_path_logprob_std": 67.81866104047081, "trajectory_balance_rmse": 201.12257372920087} +{"canonical_length": 117, "canonical_nll": 1.1853695078792734, "generation_rouge_l_f1": 0.21989528795811517, "generation_token_f1": 0.3455497382198953, "prompt_id": "ad627413d7298d7b", "reading_canonical_log_prob": -138.688232421875, "reading_distances": [24, 17, 22, 29], "reading_log_probs": [-140.95169067382812, -138.92962646484375, -139.01580810546875, -140.20733642578125], "reading_logprob_std": 0.9768212898600055, "tokenization_path_logprob_std": 88.88393446598175, "trajectory_balance_rmse": 2382.362767945746} +{"canonical_length": 314, "canonical_nll": 1.5134996547820463, "generation_rouge_l_f1": 0.28142589118198874, "generation_token_f1": 0.5290806754221389, "prompt_id": "802f028e684a8238", "reading_canonical_log_prob": -475.2388916015625, "reading_distances": [24, 22, 23, 19], "reading_log_probs": [-474.3558349609375, -474.86639404296875, -475.8834228515625, -474.3661193847656], "reading_logprob_std": 0.7177007994692668, "tokenization_path_logprob_std": 241.9892883942813, "trajectory_balance_rmse": 6283.536572787437} +{"canonical_length": 190, "canonical_nll": 1.455556929738898, "generation_rouge_l_f1": 0.26200873362445415, "generation_token_f1": 0.4279475982532751, "prompt_id": "2e2423f516c40a85", "reading_canonical_log_prob": -276.5558166503906, "reading_distances": [8, 16, 21, 16], "reading_log_probs": [-275.81292724609375, -272.54156494140625, -273.2654724121094, -273.4859619140625], "reading_logprob_std": 1.416293531392463, "tokenization_path_logprob_std": 156.89172296184495, "trajectory_balance_rmse": 4290.170150272457} +{"canonical_length": 297, "canonical_nll": 1.4233768347537878, "generation_rouge_l_f1": 0.25043478260869567, "generation_token_f1": 0.497391304347826, "prompt_id": "8d1b09dca5de1a98", "reading_canonical_log_prob": -422.742919921875, "reading_distances": [26, 21, 28, 23], "reading_log_probs": [-421.3770751953125, -420.4093017578125, -421.38372802734375, -419.71759033203125], "reading_logprob_std": 0.811095235315309, "tokenization_path_logprob_std": 187.8313213749486, "trajectory_balance_rmse": 7474.155540668431} +{"canonical_length": 82, "canonical_nll": 0.6579124171559404, "generation_rouge_l_f1": 0.282258064516129, "generation_token_f1": 0.4274193548387097, "prompt_id": "0609bdeeadeb9baf", "reading_canonical_log_prob": -53.94881820678711, "reading_distances": [35, 36, 39, 34], "reading_log_probs": [-52.442657470703125, -51.84752655029297, -52.46222686767578, -51.6274528503418], "reading_logprob_std": 0.4225181377173074, "tokenization_path_logprob_std": 99.79605895216893, "trajectory_balance_rmse": 1070.9839566612427} +{"canonical_length": 222, "canonical_nll": 1.0779330966708895, "generation_rouge_l_f1": 0.392, "generation_token_f1": 0.6240000000000001, "prompt_id": "e9f992cd65ad5461", "reading_canonical_log_prob": -239.3011474609375, "reading_distances": [25, 32, 29, 24], "reading_log_probs": [-240.3997802734375, -238.71780395507812, -239.665283203125, -240.0350341796875], "reading_logprob_std": 0.7229049136123267, "tokenization_path_logprob_std": 123.22893785400086, "trajectory_balance_rmse": 3336.1490384274816} +{"canonical_length": 6, "canonical_nll": 1.74409023920695, "generation_rouge_l_f1": 0.023255813953488372, "generation_token_f1": 0.023255813953488372, "prompt_id": "15eeee68ceb20913", "reading_canonical_log_prob": -10.4645414352417, "reading_distances": [32, 38, 36, 31], "reading_log_probs": [-10.923139572143555, -11.334630012512207, -10.811198234558105, -10.543825149536133], "reading_logprob_std": 0.3287065815239805, "tokenization_path_logprob_std": 44.073791794187855, "trajectory_balance_rmse": 162.29896199288245} +{"canonical_length": 105, "canonical_nll": 1.447126697358631, "generation_rouge_l_f1": 0.20994475138121546, "generation_token_f1": 0.3591160220994475, "prompt_id": "728a7ca9e481ba81", "reading_canonical_log_prob": -151.94830322265625, "reading_distances": [29, 30, 30, 24], "reading_log_probs": [-144.43746948242188, -140.4705810546875, -140.47303771972656, -140.45291137695312], "reading_logprob_std": 1.9859999500126895, "tokenization_path_logprob_std": 94.32103758130678, "trajectory_balance_rmse": 2160.831821168952} +{"canonical_length": 123, "canonical_nll": 1.374558363503557, "generation_rouge_l_f1": 0.08673469387755103, "generation_token_f1": 0.1377551020408163, "prompt_id": "92cf4663b97aed32", "reading_canonical_log_prob": -169.0706787109375, "reading_distances": [26, 24, 30, 22], "reading_log_probs": [-167.91336059570312, -167.73907470703125, -168.14749145507812, -167.589111328125], "reading_logprob_std": 0.24003680952115167, "tokenization_path_logprob_std": 153.63752407623127, "trajectory_balance_rmse": 1931.0937424949407} +{"canonical_length": 66, "canonical_nll": 0.8892419988458807, "generation_rouge_l_f1": 0.2686567164179105, "generation_token_f1": 0.537313432835821, "prompt_id": "3226fde653ee5231", "reading_canonical_log_prob": -58.689971923828125, "reading_distances": [8, 13, 12, 12], "reading_log_probs": [-60.0252685546875, -59.60712432861328, -59.84517288208008, -59.38144302368164], "reading_logprob_std": 0.2805403872053038, "tokenization_path_logprob_std": 79.853678581218, "trajectory_balance_rmse": 1563.3613751702164} +{"canonical_length": 259, "canonical_nll": 1.1399583264207287, "generation_rouge_l_f1": 0.2679900744416873, "generation_token_f1": 0.47146401985111663, "prompt_id": "d40a7f2ec7ebb044", "reading_canonical_log_prob": -295.24920654296875, "reading_distances": [22, 24, 26, 20], "reading_log_probs": [-297.0320739746094, -297.68402099609375, -298.60302734375, -297.45587158203125], "reading_logprob_std": 0.6636468050795293, "tokenization_path_logprob_std": 97.65178757830154, "trajectory_balance_rmse": 4774.300411579884} +{"canonical_length": 13, "canonical_nll": 1.2002443900475135, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "6844823da97b3fe4", "reading_canonical_log_prob": -15.603177070617676, "reading_distances": [20, 24, 22, 21], "reading_log_probs": [-15.822404861450195, -16.29254722595215, -17.33347511291504, -17.049129486083984], "reading_logprob_std": 0.691979074234763, "tokenization_path_logprob_std": 73.67792857399664, "trajectory_balance_rmse": 294.51649101354184} +{"canonical_length": 118, "canonical_nll": 0.8134009797694319, "generation_rouge_l_f1": 0.11940298507462689, "generation_token_f1": 0.19402985074626863, "prompt_id": "49597fd7a90f1c14", "reading_canonical_log_prob": -95.98131561279297, "reading_distances": [24, 25, 31, 24], "reading_log_probs": [-96.90055847167969, -98.05302429199219, -101.36809539794922, -98.16802978515625], "reading_logprob_std": 1.9178305845461059, "tokenization_path_logprob_std": 166.57428281408434, "trajectory_balance_rmse": 2023.2271908897972} +{"canonical_length": 258, "canonical_nll": 1.6719007861706638, "generation_rouge_l_f1": 0.13545816733067728, "generation_token_f1": 0.26294820717131473, "prompt_id": "06dc86d1f57bb10a", "reading_canonical_log_prob": -431.35040283203125, "reading_distances": [29, 34, 39, 30], "reading_log_probs": [-432.08642578125, -427.7000732421875, -426.6080322265625, -432.56915283203125], "reading_logprob_std": 3.0265678918471597, "tokenization_path_logprob_std": 159.4116956499434, "trajectory_balance_rmse": 5543.406230768456} +{"canonical_length": 250, "canonical_nll": 1.411145751953125, "generation_rouge_l_f1": 0.2547528517110267, "generation_token_f1": 0.5095057034220534, "prompt_id": "104414445caa5ddc", "reading_canonical_log_prob": -352.78643798828125, "reading_distances": [37, 32, 36, 35], "reading_log_probs": [-354.7935791015625, -353.13238525390625, -354.1431884765625, -353.7337646484375], "reading_logprob_std": 0.6986169576245551, "tokenization_path_logprob_std": 262.0950495558935, "trajectory_balance_rmse": 5904.88226880386} +{"canonical_length": 274, "canonical_nll": 0.930479676183993, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.541044776119403, "prompt_id": "96cee7006ba2cc41", "reading_canonical_log_prob": -254.95143127441406, "reading_distances": [20, 20, 21, 18], "reading_log_probs": [-258.59051513671875, -260.2747802734375, -258.9111328125, -258.9779052734375], "reading_logprob_std": 0.7436113225168168, "tokenization_path_logprob_std": 55.1942606614039, "trajectory_balance_rmse": 4994.817966513561} +{"canonical_length": 150, "canonical_nll": 2.1226924641927085, "generation_rouge_l_f1": 0.22871046228710462, "generation_token_f1": 0.42335766423357657, "prompt_id": "c61662a19e060e22", "reading_canonical_log_prob": -318.40386962890625, "reading_distances": [28, 29, 33, 33], "reading_log_probs": [-314.79150390625, -313.800537109375, -315.203857421875, -315.2409973144531], "reading_logprob_std": 0.6708022443241985, "tokenization_path_logprob_std": 282.3448691075444, "trajectory_balance_rmse": 3491.710911992862} +{"canonical_length": 80, "canonical_nll": 1.1765228271484376, "generation_rouge_l_f1": 0.024390243902439025, "generation_token_f1": 0.024390243902439025, "prompt_id": "448e404116b3a283", "reading_canonical_log_prob": -94.121826171875, "reading_distances": [18, 27, 26, 24], "reading_log_probs": [-92.49848937988281, -91.42167663574219, -92.5613021850586, -92.16477966308594], "reading_logprob_std": 0.523052891036213, "tokenization_path_logprob_std": 82.48899172296302, "trajectory_balance_rmse": 1428.4056319432989} +{"canonical_length": 40, "canonical_nll": 2.249708366394043, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.42857142857142855, "prompt_id": "62e94a0db3b86cbd", "reading_canonical_log_prob": -89.98833465576172, "reading_distances": [16, 17, 15, 16], "reading_log_probs": [-88.07939910888672, -87.77521514892578, -87.57146453857422, -86.80613708496094], "reading_logprob_std": 0.5429912758137642, "tokenization_path_logprob_std": 78.51961920827881, "trajectory_balance_rmse": 1099.383501258859} +{"canonical_length": 203, "canonical_nll": 1.0333687918526786, "generation_rouge_l_f1": 0.34101382488479265, "generation_token_f1": 0.5345622119815667, "prompt_id": "6996ebbe576a8626", "reading_canonical_log_prob": -209.77386474609375, "reading_distances": [33, 32, 37, 29], "reading_log_probs": [-208.44931030273438, -208.82711791992188, -210.20785522460938, -208.03916931152344], "reading_logprob_std": 0.9413612295579763, "tokenization_path_logprob_std": 284.1287733559723, "trajectory_balance_rmse": 4276.494244648574} +{"canonical_length": 97, "canonical_nll": 1.2380739978908264, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.45, "prompt_id": "539a3dd889cf333b", "reading_canonical_log_prob": -120.09317779541016, "reading_distances": [39, 25, 33, 33], "reading_log_probs": [-123.52420043945312, -123.86006164550781, -124.43628692626953, -124.31547546386719], "reading_logprob_std": 0.4207981902504314, "tokenization_path_logprob_std": 170.47394051661038, "trajectory_balance_rmse": 2630.757585484689} +{"canonical_length": 83, "canonical_nll": 1.7531429428652108, "generation_rouge_l_f1": 0.16231884057971013, "generation_token_f1": 0.23768115942028983, "prompt_id": "22dd62a830e20aab", "reading_canonical_log_prob": -145.5108642578125, "reading_distances": [24, 37, 29, 27], "reading_log_probs": [-142.0135498046875, -143.66787719726562, -142.10516357421875, -142.87135314941406], "reading_logprob_std": 0.7716103115361106, "tokenization_path_logprob_std": 149.63150137997079, "trajectory_balance_rmse": 1871.933621036991} +{"canonical_length": 26, "canonical_nll": 1.0774393815260668, "generation_rouge_l_f1": 0.2777777777777778, "generation_token_f1": 0.3333333333333333, "prompt_id": "23f859fe909ddbe9", "reading_canonical_log_prob": -28.013423919677734, "reading_distances": [17, 14, 14, 16], "reading_log_probs": [-37.06364822387695, -31.599477767944336, -32.36117172241211, -31.089155197143555], "reading_logprob_std": 2.740493322255132, "tokenization_path_logprob_std": 104.57661323509768, "trajectory_balance_rmse": 557.5202958365694} +{"canonical_length": 97, "canonical_nll": 0.6305421652253141, "generation_rouge_l_f1": 0.3462897526501767, "generation_token_f1": 0.45936395759717313, "prompt_id": "9658518adf176971", "reading_canonical_log_prob": -61.16259002685547, "reading_distances": [16, 13, 18, 14], "reading_log_probs": [-57.857940673828125, -59.378299713134766, -56.75630187988281, -58.96142578125], "reading_logprob_std": 1.178063628589744, "tokenization_path_logprob_std": 64.64956451980845, "trajectory_balance_rmse": 1211.8209029672507} +{"canonical_length": 243, "canonical_nll": 1.3079855334121013, "generation_rouge_l_f1": 0.26694045174537984, "generation_token_f1": 0.48049281314168374, "prompt_id": "c496b50e51f7e82e", "reading_canonical_log_prob": -317.8404846191406, "reading_distances": [30, 37, 26, 33], "reading_log_probs": [-319.58758544921875, -320.8333740234375, -319.37109375, -319.931640625], "reading_logprob_std": 0.6443880372415904, "tokenization_path_logprob_std": 175.39603309840484, "trajectory_balance_rmse": 5705.475375809416} +{"canonical_length": 157, "canonical_nll": 1.0449035061392815, "generation_rouge_l_f1": 0.23529411764705885, "generation_token_f1": 0.4852941176470588, "prompt_id": "9433be6b41e84522", "reading_canonical_log_prob": -164.0498504638672, "reading_distances": [41, 34, 38, 35], "reading_log_probs": [-168.77069091796875, -169.908935546875, -170.68081665039062, -168.26370239257812], "reading_logprob_std": 1.093398845308156, "tokenization_path_logprob_std": 205.80980292305028, "trajectory_balance_rmse": 3532.9393019940653} +{"canonical_length": 14, "canonical_nll": 1.1490716934204102, "generation_rouge_l_f1": 0.3636363636363637, "generation_token_f1": 0.3636363636363637, "prompt_id": "22dd55d64faca0d8", "reading_canonical_log_prob": -16.087003707885742, "reading_distances": [40, 42, 49, 49], "reading_log_probs": [-19.692001342773438, -19.909391403198242, -19.84968376159668, -18.835315704345703], "reading_logprob_std": 0.4993477939245764, "tokenization_path_logprob_std": 30.013558793501623, "trajectory_balance_rmse": 430.9576038334961} +{"canonical_length": 52, "canonical_nll": 1.2294596892136793, "generation_rouge_l_f1": 0.23809523809523808, "generation_token_f1": 0.38095238095238093, "prompt_id": "9ce142b4fe68303a", "reading_canonical_log_prob": -63.93190383911133, "reading_distances": [31, 27, 33, 38], "reading_log_probs": [-63.923973083496094, -63.63108444213867, -63.652252197265625, -62.47455596923828], "reading_logprob_std": 0.6445541453432345, "tokenization_path_logprob_std": 89.02210322664568, "trajectory_balance_rmse": 1317.0173866461587} +{"canonical_length": 28, "canonical_nll": 1.2900446483067103, "generation_rouge_l_f1": 0.5625, "generation_token_f1": 0.59375, "prompt_id": "e62a412929959b77", "reading_canonical_log_prob": -36.12125015258789, "reading_distances": [27, 24, 28, 24], "reading_log_probs": [-35.41356658935547, -36.610618591308594, -35.44934844970703, -36.80876922607422], "reading_logprob_std": 0.7425542122447613, "tokenization_path_logprob_std": 45.11237771144319, "trajectory_balance_rmse": 796.6601732265184} +{"canonical_length": 21, "canonical_nll": 0.9848848978678385, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4210526315789474, "prompt_id": "273ee8648fccd69e", "reading_canonical_log_prob": -20.68258285522461, "reading_distances": [31, 32, 30, 28], "reading_log_probs": [-21.99519157409668, -21.127614974975586, -21.194326400756836, -20.081279754638672], "reading_logprob_std": 0.7850304281764287, "tokenization_path_logprob_std": 50.02273693882011, "trajectory_balance_rmse": 455.1977596821784} +{"canonical_length": 109, "canonical_nll": 1.1896873264137757, "generation_rouge_l_f1": 0.42735042735042733, "generation_token_f1": 0.6666666666666665, "prompt_id": "0ffd3304b685d298", "reading_canonical_log_prob": -129.67591857910156, "reading_distances": [22, 24, 24, 19], "reading_log_probs": [-130.43582153320312, -130.2041778564453, -128.69149780273438, -132.4290008544922], "reading_logprob_std": 1.5350367871787522, "tokenization_path_logprob_std": 131.0761983635474, "trajectory_balance_rmse": 2378.3645694012444} +{"canonical_length": 288, "canonical_nll": 1.1499623192681208, "generation_rouge_l_f1": 0.3565891472868217, "generation_token_f1": 0.6046511627906976, "prompt_id": "6b90af69594f1bfc", "reading_canonical_log_prob": -331.18914794921875, "reading_distances": [30, 33, 33, 33], "reading_log_probs": [-336.23529052734375, -337.5120849609375, -337.279296875, -338.15106201171875], "reading_logprob_std": 0.7964917910188366, "tokenization_path_logprob_std": 188.172698061554, "trajectory_balance_rmse": 7211.256115184508} +{"canonical_length": 256, "canonical_nll": 1.3844068050384521, "generation_rouge_l_f1": 0.2623574144486693, "generation_token_f1": 0.5247148288973386, "prompt_id": "7021b5d065291a9c", "reading_canonical_log_prob": -354.40814208984375, "reading_distances": [18, 24, 19, 23], "reading_log_probs": [-347.53411865234375, -347.5740051269531, -347.86328125, -347.81488037109375], "reading_logprob_std": 0.16653585523638312, "tokenization_path_logprob_std": 153.44272584031233, "trajectory_balance_rmse": 5794.591865815751} +{"canonical_length": 68, "canonical_nll": 0.7427719340604895, "generation_rouge_l_f1": 0.3867403314917127, "generation_token_f1": 0.5082872928176796, "prompt_id": "56589cd90846fa45", "reading_canonical_log_prob": -50.50849151611328, "reading_distances": [39, 45, 32, 27], "reading_log_probs": [-52.1949577331543, -50.10783386230469, -51.667991638183594, -49.719451904296875], "reading_logprob_std": 1.1952561912167172, "tokenization_path_logprob_std": 37.63174773834889, "trajectory_balance_rmse": 1123.579961544239} +{"canonical_length": 187, "canonical_nll": 1.5463077320772058, "generation_rouge_l_f1": 0.19597989949748745, "generation_token_f1": 0.4472361809045226, "prompt_id": "b2517684dcd5d425", "reading_canonical_log_prob": -289.1595458984375, "reading_distances": [47, 36, 38, 36], "reading_log_probs": [-289.95257568359375, -291.81640625, -291.6220397949219, -289.2866516113281], "reading_logprob_std": 1.2448528858234782, "tokenization_path_logprob_std": 66.60023283316319, "trajectory_balance_rmse": 4718.881373141599} +{"canonical_length": 2, "canonical_nll": 9.074554443359375, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "44fca97732ef92a8", "reading_canonical_log_prob": -18.14910888671875, "reading_distances": [22, 24, 22, 18], "reading_log_probs": [-14.892066955566406, -17.129859924316406, -16.205677032470703, -15.511451721191406], "reading_logprob_std": 0.9605655541295799, "tokenization_path_logprob_std": 19.695246790843125, "trajectory_balance_rmse": 39.30662322995726} +{"canonical_length": 14, "canonical_nll": 0.31231883593967985, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.4, "prompt_id": "907693271b754ffe", "reading_canonical_log_prob": -4.372463703155518, "reading_distances": [5, 6, 7, 8], "reading_log_probs": [-4.9557108879089355, -5.014664649963379, -4.869003772735596, -5.069265365600586], "reading_logprob_std": 0.08572792435255264, "tokenization_path_logprob_std": 20.951228343300013, "trajectory_balance_rmse": 192.23263476166667} +{"canonical_length": 220, "canonical_nll": 1.1559899763627486, "generation_rouge_l_f1": 0.1925601750547046, "generation_token_f1": 0.4026258205689279, "prompt_id": "3d885c9990d3f7a8", "reading_canonical_log_prob": -254.3177947998047, "reading_distances": [43, 52, 46, 50], "reading_log_probs": [-255.6055145263672, -260.6979675292969, -255.75389099121094, -264.03741455078125], "reading_logprob_std": 4.095369264687761, "tokenization_path_logprob_std": 114.91060023424053, "trajectory_balance_rmse": 4407.889416124785} +{"canonical_length": 12, "canonical_nll": 1.5245952606201172, "generation_rouge_l_f1": 0.05714285714285714, "generation_token_f1": 0.06428571428571428, "prompt_id": "946f59f054cd3b7a", "reading_canonical_log_prob": -18.295143127441406, "reading_distances": [27, 25, 33, 27], "reading_log_probs": [-20.35483169555664, -19.958206176757812, -19.49034881591797, -20.470014572143555], "reading_logprob_std": 0.44332953168890515, "tokenization_path_logprob_std": 55.97914318982798, "trajectory_balance_rmse": 292.55026798095787} +{"canonical_length": 114, "canonical_nll": 1.3236810784590871, "generation_rouge_l_f1": 0.19104477611940301, "generation_token_f1": 0.3880597014925373, "prompt_id": "1e094a0368e082c2", "reading_canonical_log_prob": -150.89964294433594, "reading_distances": [30, 31, 29, 21], "reading_log_probs": [-150.57315063476562, -148.38668823242188, -147.55108642578125, -147.13436889648438], "reading_logprob_std": 1.532397287511496, "tokenization_path_logprob_std": 185.74532562407447, "trajectory_balance_rmse": 2721.1592266639827} +{"canonical_length": 8, "canonical_nll": 0.04865289479494095, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "f01addec53ef2695", "reading_canonical_log_prob": -0.3892231583595276, "reading_distances": [24, 25, 23, 24], "reading_log_probs": [-1.3219250440597534, -0.5315701961517334, -0.6233285069465637, -1.4462388753890991], "reading_logprob_std": 0.4699620277928498, "tokenization_path_logprob_std": 45.931553164879766, "trajectory_balance_rmse": 155.17588456442112} +{"canonical_length": 43, "canonical_nll": 1.2396880748660066, "generation_rouge_l_f1": 0.3235294117647059, "generation_token_f1": 0.4411764705882353, "prompt_id": "bd70996cafa5e523", "reading_canonical_log_prob": -53.30658721923828, "reading_distances": [32, 29, 25, 18], "reading_log_probs": [-55.911346435546875, -58.34497833251953, -58.640281677246094, -57.29658508300781], "reading_logprob_std": 1.2342471012107645, "tokenization_path_logprob_std": 123.08097396234704, "trajectory_balance_rmse": 879.8551388207235} +{"canonical_length": 280, "canonical_nll": 0.9874285016741071, "generation_rouge_l_f1": 0.21468926553672316, "generation_token_f1": 0.4670433145009416, "prompt_id": "40b3b0fca502cfb8", "reading_canonical_log_prob": -276.47998046875, "reading_distances": [29, 29, 26, 34], "reading_log_probs": [-274.658447265625, -276.2810974121094, -277.235595703125, -275.78466796875], "reading_logprob_std": 1.0726043503666767, "tokenization_path_logprob_std": 144.2121478899519, "trajectory_balance_rmse": 5545.446087272443} +{"canonical_length": 74, "canonical_nll": 1.2615224477407094, "generation_rouge_l_f1": 0.19883040935672514, "generation_token_f1": 0.30409356725146197, "prompt_id": "88f331fe7934b488", "reading_canonical_log_prob": -93.3526611328125, "reading_distances": [15, 10, 13, 16], "reading_log_probs": [-94.5266342163086, -94.09828186035156, -93.96221923828125, -93.95381164550781], "reading_logprob_std": 0.26920114437786946, "tokenization_path_logprob_std": 99.71777133801119, "trajectory_balance_rmse": 1633.4285725910156} +{"canonical_length": 33, "canonical_nll": 0.549414027820934, "generation_rouge_l_f1": 0.14400000000000002, "generation_token_f1": 0.22399999999999998, "prompt_id": "8b4c34e2f1db7af4", "reading_canonical_log_prob": -18.13066291809082, "reading_distances": [18, 17, 20, 21], "reading_log_probs": [-17.45806884765625, -17.824674606323242, -18.727523803710938, -17.546092987060547], "reading_logprob_std": 0.5803868740764874, "tokenization_path_logprob_std": 74.52676226639005, "trajectory_balance_rmse": 887.274707702455} +{"canonical_length": 155, "canonical_nll": 1.9171026414440524, "generation_rouge_l_f1": 0.13053613053613053, "generation_token_f1": 0.23776223776223773, "prompt_id": "7a99289ed340c72d", "reading_canonical_log_prob": -297.1509094238281, "reading_distances": [13, 21, 17, 17], "reading_log_probs": [-299.577392578125, -300.23907470703125, -300.7362060546875, -300.1880798339844], "reading_logprob_std": 0.4746747347190939, "tokenization_path_logprob_std": 129.03912054634395, "trajectory_balance_rmse": 2651.307456258209} +{"canonical_length": 21, "canonical_nll": 0.8706043788364956, "generation_rouge_l_f1": 0.8372093023255814, "generation_token_f1": 0.8372093023255814, "prompt_id": "321ea8e8622b4791", "reading_canonical_log_prob": -18.282691955566406, "reading_distances": [26, 33, 35, 27], "reading_log_probs": [-18.169147491455078, -18.027931213378906, -17.582670211791992, -17.239057540893555], "reading_logprob_std": 0.42501114212520974, "tokenization_path_logprob_std": 66.80739178748247, "trajectory_balance_rmse": 473.42105614403425} +{"canonical_length": 176, "canonical_nll": 1.9099245938387783, "generation_rouge_l_f1": 0.1826923076923077, "generation_token_f1": 0.3894230769230769, "prompt_id": "e9ba47f9f768bdc0", "reading_canonical_log_prob": -336.146728515625, "reading_distances": [25, 29, 29, 20], "reading_log_probs": [-336.9380798339844, -338.943603515625, -338.7061767578125, -337.7974548339844], "reading_logprob_std": 0.9166342946205714, "tokenization_path_logprob_std": 114.48929468745627, "trajectory_balance_rmse": 4552.411587080645} +{"canonical_length": 7, "canonical_nll": 0.582751955304827, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "cd4e165565532ef7", "reading_canonical_log_prob": -4.079263687133789, "reading_distances": [31, 32, 36, 37], "reading_log_probs": [-4.873748779296875, -4.684781074523926, -5.1513214111328125, -4.625367641448975], "reading_logprob_std": 0.23668997982375378, "tokenization_path_logprob_std": 11.922311688973544, "trajectory_balance_rmse": 153.54857024982286} +{"canonical_length": 270, "canonical_nll": 0.9277338098596644, "generation_rouge_l_f1": 0.24113475177304963, "generation_token_f1": 0.524822695035461, "prompt_id": "df9cf174d34cff3e", "reading_canonical_log_prob": -250.48812866210938, "reading_distances": [32, 36, 42, 30], "reading_log_probs": [-251.6967010498047, -251.31866455078125, -249.369873046875, -250.837158203125], "reading_logprob_std": 1.0197392016823021, "tokenization_path_logprob_std": 205.6146190508064, "trajectory_balance_rmse": 4306.711642596588} +{"canonical_length": 143, "canonical_nll": 1.5850842882703233, "generation_rouge_l_f1": 0.43097643097643096, "generation_token_f1": 0.5252525252525252, "prompt_id": "1fba2562017bea7b", "reading_canonical_log_prob": -226.66705322265625, "reading_distances": [29, 34, 33, 29], "reading_log_probs": [-226.30245971679688, -228.13400268554688, -226.4139404296875, -226.15341186523438], "reading_logprob_std": 0.9281891937267671, "tokenization_path_logprob_std": 110.52887733374321, "trajectory_balance_rmse": 3145.62037783135} +{"canonical_length": 28, "canonical_nll": 1.1652743475777763, "generation_rouge_l_f1": 0.45714285714285713, "generation_token_f1": 0.45714285714285713, "prompt_id": "56997da805bc232c", "reading_canonical_log_prob": -32.627681732177734, "reading_distances": [25, 26, 29, 36], "reading_log_probs": [-31.244083404541016, -31.586801528930664, -32.24507141113281, -31.369598388671875], "reading_logprob_std": 0.4455461226009366, "tokenization_path_logprob_std": 30.201665006934892, "trajectory_balance_rmse": 716.6926223670524} +{"canonical_length": 110, "canonical_nll": 1.4741005637428977, "generation_rouge_l_f1": 0.2113821138211382, "generation_token_f1": 0.3252032520325203, "prompt_id": "02d66047fd43a021", "reading_canonical_log_prob": -162.15106201171875, "reading_distances": [22, 19, 16, 18], "reading_log_probs": [-162.5646514892578, -160.69229125976562, -164.07427978515625, -163.0966339111328], "reading_logprob_std": 1.4213320379136591, "tokenization_path_logprob_std": 156.59191027879945, "trajectory_balance_rmse": 2333.062328764792} +{"canonical_length": 79, "canonical_nll": 1.5990376532832278, "generation_rouge_l_f1": 0.15015015015015015, "generation_token_f1": 0.22822822822822822, "prompt_id": "83a9bce5dad053f6", "reading_canonical_log_prob": -126.323974609375, "reading_distances": [20, 21, 22, 24], "reading_log_probs": [-135.5526123046875, -134.9068603515625, -134.7788848876953, -134.99993896484375], "reading_logprob_std": 0.34095500362590975, "tokenization_path_logprob_std": 113.90072013089126, "trajectory_balance_rmse": 1514.3850548767068} +{"canonical_length": 21, "canonical_nll": 6.645237513950893, "generation_rouge_l_f1": 0.016, "generation_token_f1": 0.016, "prompt_id": "35f4df3b6373154b", "reading_canonical_log_prob": -139.54998779296875, "reading_distances": [20, 17, 23, 20], "reading_log_probs": [-134.6644744873047, -135.81301879882812, -133.35653686523438, -134.936767578125], "reading_logprob_std": 1.016663135945534, "tokenization_path_logprob_std": 21.810506074268556, "trajectory_balance_rmse": 253.87970315196858} +{"canonical_length": 228, "canonical_nll": 1.1970174689041941, "generation_rouge_l_f1": 0.19693654266958427, "generation_token_f1": 0.35886214442013126, "prompt_id": "7c0757fa3b0cff4b", "reading_canonical_log_prob": -272.91998291015625, "reading_distances": [21, 29, 32, 26], "reading_log_probs": [-267.8269958496094, -268.1163024902344, -266.8996276855469, -267.85443115234375], "reading_logprob_std": 0.5326808439425231, "tokenization_path_logprob_std": 213.0051599151784, "trajectory_balance_rmse": 3244.554504057011} +{"canonical_length": 211, "canonical_nll": 1.527929513940314, "generation_rouge_l_f1": 0.23093681917211328, "generation_token_f1": 0.5054466230936818, "prompt_id": "5fa3ea0df830456f", "reading_canonical_log_prob": -322.39312744140625, "reading_distances": [20, 27, 23, 19], "reading_log_probs": [-327.76416015625, -326.41510009765625, -323.1372985839844, -324.3465270996094], "reading_logprob_std": 2.069485390435727, "tokenization_path_logprob_std": 90.82022191973581, "trajectory_balance_rmse": 4854.197490258226} +{"canonical_length": 145, "canonical_nll": 1.033246165308459, "generation_rouge_l_f1": 0.25839793281653745, "generation_token_f1": 0.4186046511627907, "prompt_id": "70669892361b488c", "reading_canonical_log_prob": -149.82069396972656, "reading_distances": [35, 36, 29, 32], "reading_log_probs": [-155.71646118164062, -154.8851318359375, -155.5133056640625, -156.90597534179688], "reading_logprob_std": 0.8448502947937899, "tokenization_path_logprob_std": 136.02032975355684, "trajectory_balance_rmse": 2556.226584088281} +{"canonical_length": 12, "canonical_nll": 1.0049448013305664, "generation_rouge_l_f1": 0.6666666666666667, "generation_token_f1": 0.8333333333333333, "prompt_id": "e90b400c44335d9a", "reading_canonical_log_prob": -12.059337615966797, "reading_distances": [19, 18, 21, 25], "reading_log_probs": [-13.188583374023438, -13.113142013549805, -12.630623817443848, -12.870796203613281], "reading_logprob_std": 0.2528565185323083, "tokenization_path_logprob_std": 48.21765902694778, "trajectory_balance_rmse": 235.20985846231375} +{"canonical_length": 232, "canonical_nll": 1.3014048872322872, "generation_rouge_l_f1": 0.2336065573770492, "generation_token_f1": 0.45901639344262296, "prompt_id": "9f0a3bc3aaf621a1", "reading_canonical_log_prob": -301.9259338378906, "reading_distances": [20, 31, 25, 26], "reading_log_probs": [-298.3871154785156, -298.58502197265625, -297.76873779296875, -300.7477722167969], "reading_logprob_std": 1.297841565744493, "tokenization_path_logprob_std": 226.42739363004426, "trajectory_balance_rmse": 5254.538912055753} +{"canonical_length": 93, "canonical_nll": 0.5765742640341481, "generation_rouge_l_f1": 0.18430034129692832, "generation_token_f1": 0.3208191126279864, "prompt_id": "f2be334fa1e04283", "reading_canonical_log_prob": -53.62140655517578, "reading_distances": [25, 15, 23, 24], "reading_log_probs": [-54.19321823120117, -54.849212646484375, -55.948402404785156, -55.48240280151367], "reading_logprob_std": 0.7637244561927427, "tokenization_path_logprob_std": 57.94406773342339, "trajectory_balance_rmse": 1809.3373347899067} +{"canonical_length": 54, "canonical_nll": 1.486406820791739, "generation_rouge_l_f1": 0.3252032520325203, "generation_token_f1": 0.4065040650406504, "prompt_id": "1a42eb844c9543a9", "reading_canonical_log_prob": -80.2659683227539, "reading_distances": [45, 31, 40, 36], "reading_log_probs": [-81.68047332763672, -80.51795959472656, -80.5386734008789, -81.51548767089844], "reading_logprob_std": 0.6212905225891027, "tokenization_path_logprob_std": 42.81898361098108, "trajectory_balance_rmse": 1173.1824884094879} +{"canonical_length": 40, "canonical_nll": 0.4973738670349121, "generation_rouge_l_f1": 0.1548821548821549, "generation_token_f1": 0.20875420875420878, "prompt_id": "3afed29700596cb5", "reading_canonical_log_prob": -19.894954681396484, "reading_distances": [11, 9, 16, 14], "reading_log_probs": [-21.16944694519043, -21.434001922607422, -22.387699127197266, -23.32480239868164], "reading_logprob_std": 0.9816020997042929, "tokenization_path_logprob_std": 46.800857651301655, "trajectory_balance_rmse": 470.5166317363991} +{"canonical_length": 28, "canonical_nll": 0.7684531893048968, "generation_rouge_l_f1": 0.2941176470588235, "generation_token_f1": 0.4313725490196079, "prompt_id": "e24f8c298fcc6b27", "reading_canonical_log_prob": -21.51668930053711, "reading_distances": [13, 12, 13, 12], "reading_log_probs": [-21.177371978759766, -21.510448455810547, -20.974472045898438, -21.471378326416016], "reading_logprob_std": 0.25401133396511594, "tokenization_path_logprob_std": 31.66074575673137, "trajectory_balance_rmse": 627.7051619025264} +{"canonical_length": 232, "canonical_nll": 1.441320353540881, "generation_rouge_l_f1": 0.22857142857142856, "generation_token_f1": 0.39591836734693875, "prompt_id": "d2492eb6d00f3204", "reading_canonical_log_prob": -334.3863220214844, "reading_distances": [31, 27, 20, 34], "reading_log_probs": [-334.09417724609375, -334.5227355957031, -332.72943115234375, -333.73187255859375], "reading_logprob_std": 0.7650573393006287, "tokenization_path_logprob_std": 235.70351994790838, "trajectory_balance_rmse": 5929.7213872349475} +{"canonical_length": 83, "canonical_nll": 1.439985987651779, "generation_rouge_l_f1": 0.28104575163398693, "generation_token_f1": 0.37254901960784315, "prompt_id": "d1d32462252d2ba1", "reading_canonical_log_prob": -119.51883697509766, "reading_distances": [18, 20, 14, 14], "reading_log_probs": [-120.5308837890625, -123.03999328613281, -123.190185546875, -120.52122497558594], "reading_logprob_std": 1.496042324729053, "tokenization_path_logprob_std": 111.57370513799847, "trajectory_balance_rmse": 1827.5476856656426} +{"canonical_length": 171, "canonical_nll": 1.1351890340883133, "generation_rouge_l_f1": 0.2517482517482518, "generation_token_f1": 0.4801864801864802, "prompt_id": "981d7ddcc229e43f", "reading_canonical_log_prob": -194.11732482910156, "reading_distances": [22, 19, 20, 21], "reading_log_probs": [-188.1584930419922, -191.35169982910156, -189.76763916015625, -189.927978515625], "reading_logprob_std": 1.306360687383734, "tokenization_path_logprob_std": 71.96321773108735, "trajectory_balance_rmse": 3268.992080747682} +{"canonical_length": 14, "canonical_nll": 0.8996915136064801, "generation_rouge_l_f1": 0.5217391304347826, "generation_token_f1": 0.7826086956521738, "prompt_id": "a353a5c04a14d88b", "reading_canonical_log_prob": -12.595681190490723, "reading_distances": [35, 31, 31, 35], "reading_log_probs": [-11.737707138061523, -11.585251808166504, -12.049409866333008, -11.39266586303711], "reading_logprob_std": 0.27738352435198876, "tokenization_path_logprob_std": 35.794055290777365, "trajectory_balance_rmse": 325.5645009033364} +{"canonical_length": 37, "canonical_nll": 0.7409886540593328, "generation_rouge_l_f1": 0.2931937172774869, "generation_token_f1": 0.31413612565445026, "prompt_id": "104ca723f309949c", "reading_canonical_log_prob": -27.416580200195312, "reading_distances": [17, 15, 21, 14], "reading_log_probs": [-27.79606819152832, -28.283687591552734, -27.612220764160156, -27.21666717529297], "reading_logprob_std": 0.44282624546900917, "tokenization_path_logprob_std": 85.3710943717374, "trajectory_balance_rmse": 626.5735147602193} +{"canonical_length": 270, "canonical_nll": 0.9659326623987269, "generation_rouge_l_f1": 0.16414686825053995, "generation_token_f1": 0.2764578833693304, "prompt_id": "5768cefb5b534d32", "reading_canonical_log_prob": -260.80181884765625, "reading_distances": [22, 20, 25, 20], "reading_log_probs": [-263.53375244140625, -261.1512145996094, -260.22900390625, -262.09747314453125], "reading_logprob_std": 1.41119933502295, "tokenization_path_logprob_std": 133.55089437292557, "trajectory_balance_rmse": 4659.2588527329035} +{"canonical_length": 460, "canonical_nll": 0.7045751157014266, "generation_rouge_l_f1": 0.35999999999999993, "generation_token_f1": 0.5085714285714286, "prompt_id": "2c20880d9a9f5257", "reading_canonical_log_prob": -324.10455322265625, "reading_distances": [31, 27, 35, 21], "reading_log_probs": [-322.7557067871094, -323.7548828125, -323.2329406738281, -322.833984375], "reading_logprob_std": 0.4575183010729384, "tokenization_path_logprob_std": 233.732303748338, "trajectory_balance_rmse": 8039.294144021547} +{"canonical_length": 43, "canonical_nll": 1.1103596354639806, "generation_rouge_l_f1": 0.2253521126760563, "generation_token_f1": 0.323943661971831, "prompt_id": "1a2898fe9876d2bc", "reading_canonical_log_prob": -47.74546432495117, "reading_distances": [19, 17, 12, 16], "reading_log_probs": [-49.7535285949707, -53.98976516723633, -54.16476058959961, -53.191444396972656], "reading_logprob_std": 2.058300612341922, "tokenization_path_logprob_std": 101.77881697509669, "trajectory_balance_rmse": 965.4727662032111} +{"canonical_length": 85, "canonical_nll": 0.9098711799172794, "generation_rouge_l_f1": 0.0689655172413793, "generation_token_f1": 0.0689655172413793, "prompt_id": "52568183284a12f5", "reading_canonical_log_prob": -77.33905029296875, "reading_distances": [17, 20, 15, 17], "reading_log_probs": [-76.77143859863281, -75.96125793457031, -76.34832763671875, -75.9651870727539], "reading_logprob_std": 0.38536694652608977, "tokenization_path_logprob_std": 106.66324786786434, "trajectory_balance_rmse": 1842.7129728278705} +{"canonical_length": 202, "canonical_nll": 1.66779893931776, "generation_rouge_l_f1": 0.23376623376623376, "generation_token_f1": 0.4675324675324675, "prompt_id": "ea255680cda4b618", "reading_canonical_log_prob": -336.8953857421875, "reading_distances": [29, 17, 25, 26], "reading_log_probs": [-336.62750244140625, -337.4407043457031, -337.63616943359375, -336.35052490234375], "reading_logprob_std": 0.6214903241045028, "tokenization_path_logprob_std": 207.98954327197117, "trajectory_balance_rmse": 4388.876019990626} +{"canonical_length": 19, "canonical_nll": 1.2000671185945209, "generation_rouge_l_f1": 0.36842105263157887, "generation_token_f1": 0.5789473684210527, "prompt_id": "559aadd3a6bc5deb", "reading_canonical_log_prob": -22.8012752532959, "reading_distances": [32, 35, 35, 39], "reading_log_probs": [-22.068042755126953, -21.485549926757812, -21.250194549560547, -22.01302146911621], "reading_logprob_std": 0.4006998268796259, "tokenization_path_logprob_std": 88.7117219411937, "trajectory_balance_rmse": 436.2028777501267} +{"canonical_length": 24, "canonical_nll": 0.9010748863220215, "generation_rouge_l_f1": 0.40476190476190477, "generation_token_f1": 0.42857142857142855, "prompt_id": "4196e10bb50daa8f", "reading_canonical_log_prob": -21.625797271728516, "reading_distances": [25, 23, 20, 18], "reading_log_probs": [-22.67180633544922, -22.534122467041016, -22.50223159790039, -22.535770416259766], "reading_logprob_std": 0.07547781470540198, "tokenization_path_logprob_std": 14.051282177119736, "trajectory_balance_rmse": 533.8195587536364} +{"canonical_length": 293, "canonical_nll": 1.4550691676221204, "generation_rouge_l_f1": 0.24399260628465805, "generation_token_f1": 0.41774491682070247, "prompt_id": "0d2dbbf4ae30ce19", "reading_canonical_log_prob": -426.33526611328125, "reading_distances": [18, 13, 20, 21], "reading_log_probs": [-425.69573974609375, -422.1595458984375, -421.34857177734375, -421.69482421875], "reading_logprob_std": 2.0083865280265685, "tokenization_path_logprob_std": 216.53102503477803, "trajectory_balance_rmse": 7257.873070966498} +{"canonical_length": 227, "canonical_nll": 1.1499402554549836, "generation_rouge_l_f1": 0.2345679012345679, "generation_token_f1": 0.3395061728395062, "prompt_id": "cd0a37baf57f0d68", "reading_canonical_log_prob": -261.03643798828125, "reading_distances": [23, 29, 26, 27], "reading_log_probs": [-257.3583984375, -257.50994873046875, -261.77880859375, -259.7178649902344], "reading_logprob_std": 2.0911644111513628, "tokenization_path_logprob_std": 272.5820692793637, "trajectory_balance_rmse": 4603.807792812206} +{"canonical_length": 27, "canonical_nll": 1.6775148179796007, "generation_rouge_l_f1": 0.10738255033557047, "generation_token_f1": 0.12751677852348994, "prompt_id": "f0dd5c7b12d6ff90", "reading_canonical_log_prob": -45.29290008544922, "reading_distances": [23, 23, 19, 24], "reading_log_probs": [-50.661102294921875, -49.9781379699707, -50.26445007324219, -49.021121978759766], "reading_logprob_std": 0.6986312612268541, "tokenization_path_logprob_std": 101.18103641997331, "trajectory_balance_rmse": 593.9363167658523} +{"canonical_length": 29, "canonical_nll": 0.9543739845012796, "generation_rouge_l_f1": 0.1413427561837456, "generation_token_f1": 0.15547703180212014, "prompt_id": "c737e978a37038bd", "reading_canonical_log_prob": -27.67684555053711, "reading_distances": [23, 22, 27, 26], "reading_log_probs": [-31.00909423828125, -32.50468444824219, -32.089683532714844, -29.861465454101562], "reading_logprob_std": 1.1848102289110345, "tokenization_path_logprob_std": 43.57987068849173, "trajectory_balance_rmse": 498.4547624520902} +{"canonical_length": 90, "canonical_nll": 1.2872806125217013, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.5128205128205128, "prompt_id": "6fe0c1612b081574", "reading_canonical_log_prob": -115.85525512695312, "reading_distances": [20, 25, 24, 27], "reading_log_probs": [-120.3924789428711, -119.29081726074219, -116.50263977050781, -118.57334899902344], "reading_logprob_std": 1.6388581424783049, "tokenization_path_logprob_std": 190.98740278862556, "trajectory_balance_rmse": 2124.2786805155606} +{"canonical_length": 276, "canonical_nll": 1.2726711328478828, "generation_rouge_l_f1": 0.3173277661795407, "generation_token_f1": 0.5469728601252609, "prompt_id": "a45d6c8378dedb7b", "reading_canonical_log_prob": -351.2572326660156, "reading_distances": [13, 14, 15, 14], "reading_log_probs": [-350.66595458984375, -347.791015625, -349.970458984375, -353.16656494140625], "reading_logprob_std": 2.214791890478555, "tokenization_path_logprob_std": 202.43687457888723, "trajectory_balance_rmse": 5913.033413237537} +{"canonical_length": 103, "canonical_nll": 1.1561074858730278, "generation_rouge_l_f1": 0.2648648648648649, "generation_token_f1": 0.3405405405405405, "prompt_id": "2e7570c8d7724f9e", "reading_canonical_log_prob": -119.07907104492188, "reading_distances": [25, 22, 17, 25], "reading_log_probs": [-117.50515747070312, -117.9093017578125, -116.60922241210938, -117.85610961914062], "reading_logprob_std": 0.6011768361799472, "tokenization_path_logprob_std": 86.87464684190483, "trajectory_balance_rmse": 1590.7769596552941} +{"canonical_length": 30, "canonical_nll": 1.4691216786702475, "generation_rouge_l_f1": 0.14184397163120568, "generation_token_f1": 0.1631205673758865, "prompt_id": "7aa76ccd6b778b98", "reading_canonical_log_prob": -44.07365036010742, "reading_distances": [40, 36, 27, 35], "reading_log_probs": [-42.54680252075195, -41.78789520263672, -41.8560791015625, -42.485984802246094], "reading_logprob_std": 0.4026472069420773, "tokenization_path_logprob_std": 87.24853021115885, "trajectory_balance_rmse": 779.5655267593763} +{"canonical_length": 252, "canonical_nll": 1.3392550756060888, "generation_rouge_l_f1": 0.30979498861047833, "generation_token_f1": 0.6104783599088838, "prompt_id": "0667a89ba9993054", "reading_canonical_log_prob": -337.4922790527344, "reading_distances": [28, 27, 25, 26], "reading_log_probs": [-336.732177734375, -337.78643798828125, -340.47918701171875, -335.59466552734375], "reading_logprob_std": 2.088835225344504, "tokenization_path_logprob_std": 126.07177212989446, "trajectory_balance_rmse": 5037.564790297424} +{"canonical_length": 248, "canonical_nll": 1.0880551492014239, "generation_rouge_l_f1": 0.2514506769825918, "generation_token_f1": 0.5299806576402321, "prompt_id": "45c2d1c36166f6b6", "reading_canonical_log_prob": -269.8376770019531, "reading_distances": [24, 24, 22, 32], "reading_log_probs": [-267.09283447265625, -266.650146484375, -268.11968994140625, -266.4242858886719], "reading_logprob_std": 0.7517938492166956, "tokenization_path_logprob_std": 100.01031917288084, "trajectory_balance_rmse": 4604.549836523454} +{"canonical_length": 20, "canonical_nll": 1.1637869834899903, "generation_rouge_l_f1": 0.3870967741935484, "generation_token_f1": 0.5161290322580646, "prompt_id": "e5f7552dff198384", "reading_canonical_log_prob": -23.275739669799805, "reading_distances": [29, 20, 21, 27], "reading_log_probs": [-23.750133514404297, -23.279150009155273, -24.752826690673828, -24.17734718322754], "reading_logprob_std": 0.6271222884093341, "tokenization_path_logprob_std": 61.774925774416296, "trajectory_balance_rmse": 498.11298815593517} +{"canonical_length": 117, "canonical_nll": 1.652948232797476, "generation_rouge_l_f1": 0.26174496644295303, "generation_token_f1": 0.4161073825503356, "prompt_id": "4a930bccec9843cd", "reading_canonical_log_prob": -193.3949432373047, "reading_distances": [34, 32, 26, 27], "reading_log_probs": [-198.05126953125, -192.80718994140625, -196.710205078125, -194.78134155273438], "reading_logprob_std": 2.2884232169457857, "tokenization_path_logprob_std": 242.76078633278763, "trajectory_balance_rmse": 2797.3945351897587} +{"canonical_length": 49, "canonical_nll": 1.1369111975845025, "generation_rouge_l_f1": 0.0967741935483871, "generation_token_f1": 0.0967741935483871, "prompt_id": "35925631dfc295f8", "reading_canonical_log_prob": -55.708648681640625, "reading_distances": [33, 29, 36, 28], "reading_log_probs": [-54.81757354736328, -57.609474182128906, -56.84730529785156, -55.929718017578125], "reading_logprob_std": 1.2040152979856231, "tokenization_path_logprob_std": 140.00913982083873, "trajectory_balance_rmse": 1069.0862965068784} +{"canonical_length": 295, "canonical_nll": 1.1656442415916313, "generation_rouge_l_f1": 0.257602862254025, "generation_token_f1": 0.5688729874776386, "prompt_id": "f6597a51ed7cc783", "reading_canonical_log_prob": -343.86505126953125, "reading_distances": [21, 21, 19, 23], "reading_log_probs": [-344.04815673828125, -343.5234069824219, -343.60272216796875, -342.7418518066406], "reading_logprob_std": 0.5430181095502697, "tokenization_path_logprob_std": 211.3781331171973, "trajectory_balance_rmse": 5957.594662609131} +{"canonical_length": 213, "canonical_nll": 1.22113252021897, "generation_rouge_l_f1": 0.19831223628691982, "generation_token_f1": 0.39662447257383965, "prompt_id": "87270c0f1888627a", "reading_canonical_log_prob": -260.1012268066406, "reading_distances": [13, 15, 9, 10], "reading_log_probs": [-259.23822021484375, -260.85137939453125, -258.35009765625, -260.3218994140625], "reading_logprob_std": 1.1176674513244638, "tokenization_path_logprob_std": 413.3142837303207, "trajectory_balance_rmse": 4990.568083025632} +{"canonical_length": 14, "canonical_nll": 1.2862839017595564, "generation_rouge_l_f1": 0.15384615384615385, "generation_token_f1": 0.15384615384615385, "prompt_id": "a53e27e32808bf25", "reading_canonical_log_prob": -18.00797462463379, "reading_distances": [25, 20, 27, 27], "reading_log_probs": [-16.434797286987305, -16.54958724975586, -16.136350631713867, -16.23398208618164], "reading_logprob_std": 0.18763370991062903, "tokenization_path_logprob_std": 30.252289210705122, "trajectory_balance_rmse": 276.1254982871842} +{"canonical_length": 128, "canonical_nll": 1.241076111793518, "generation_rouge_l_f1": 0.3232323232323232, "generation_token_f1": 0.5353535353535354, "prompt_id": "1e1844cf1025dc97", "reading_canonical_log_prob": -158.8577423095703, "reading_distances": [15, 16, 18, 16], "reading_log_probs": [-156.99891662597656, -157.24188232421875, -156.75674438476562, -155.47531127929688], "reading_logprob_std": 0.7872555630273885, "tokenization_path_logprob_std": 121.349391189748, "trajectory_balance_rmse": 3064.88459172641} +{"canonical_length": 155, "canonical_nll": 1.5511944186302924, "generation_rouge_l_f1": 0.2125340599455041, "generation_token_f1": 0.35967302452316074, "prompt_id": "87981277aeac45ee", "reading_canonical_log_prob": -240.4351348876953, "reading_distances": [21, 21, 23, 18], "reading_log_probs": [-241.1837615966797, -241.2380828857422, -241.90133666992188, -240.880615234375], "reading_logprob_std": 0.4300507196976915, "tokenization_path_logprob_std": 65.73613406875039, "trajectory_balance_rmse": 3250.8512848367313} +{"canonical_length": 26, "canonical_nll": 0.2933824979341947, "generation_rouge_l_f1": 0.0794701986754967, "generation_token_f1": 0.09933774834437085, "prompt_id": "40b80040395896ae", "reading_canonical_log_prob": -7.6279449462890625, "reading_distances": [39, 46, 45, 47], "reading_log_probs": [-5.134502410888672, -5.09372615814209, -4.99440860748291, -7.008519172668457], "reading_logprob_std": 0.9689412363121018, "tokenization_path_logprob_std": 96.00541268845963, "trajectory_balance_rmse": 323.3845659709398} +{"canonical_length": 356, "canonical_nll": 1.3620437450623244, "generation_rouge_l_f1": 0.202401372212693, "generation_token_f1": 0.44596912521440824, "prompt_id": "63d17b19976fa874", "reading_canonical_log_prob": -484.8875732421875, "reading_distances": [29, 29, 27, 20], "reading_log_probs": [-481.490478515625, -481.29693603515625, -479.6568603515625, -481.9686279296875], "reading_logprob_std": 1.0047192273573522, "tokenization_path_logprob_std": 234.50138736004698, "trajectory_balance_rmse": 8365.626828516939} +{"canonical_length": 19, "canonical_nll": 0.6362417120682565, "generation_rouge_l_f1": 0.14659685863874347, "generation_token_f1": 0.14659685863874347, "prompt_id": "53bd5d756df134c2", "reading_canonical_log_prob": -12.088592529296875, "reading_distances": [27, 21, 25, 25], "reading_log_probs": [-10.299463272094727, -9.144851684570312, -10.067995071411133, -10.464988708496094], "reading_logprob_std": 0.5892554297983965, "tokenization_path_logprob_std": 73.29625558971804, "trajectory_balance_rmse": 238.23995290418384} +{"canonical_length": 114, "canonical_nll": 1.0092598764519942, "generation_rouge_l_f1": 0.1935483870967742, "generation_token_f1": 0.3519061583577712, "prompt_id": "1cf0a058de57af94", "reading_canonical_log_prob": -115.05562591552734, "reading_distances": [38, 28, 33, 36], "reading_log_probs": [-122.44441986083984, -121.26456451416016, -124.31167602539062, -125.05142211914062], "reading_logprob_std": 1.728379982416505, "tokenization_path_logprob_std": 87.85800908655838, "trajectory_balance_rmse": 1615.4201655808058} +{"canonical_length": 257, "canonical_nll": 1.4242955070525292, "generation_rouge_l_f1": 0.22090729783037477, "generation_token_f1": 0.47337278106508873, "prompt_id": "a5e93d67499f0604", "reading_canonical_log_prob": -366.0439453125, "reading_distances": [15, 19, 20, 19], "reading_log_probs": [-365.049560546875, -365.6172790527344, -365.43267822265625, -365.69049072265625], "reading_logprob_std": 0.286615253992285, "tokenization_path_logprob_std": 274.9616605994205, "trajectory_balance_rmse": 4822.172508799527} +{"canonical_length": 118, "canonical_nll": 1.2818996623411016, "generation_rouge_l_f1": 0.3923444976076555, "generation_token_f1": 0.5645933014354068, "prompt_id": "9493ef83593c3f2d", "reading_canonical_log_prob": -151.26416015625, "reading_distances": [32, 36, 37, 32], "reading_log_probs": [-148.00213623046875, -145.2617950439453, -146.47589111328125, -144.93113708496094], "reading_logprob_std": 1.391628503092498, "tokenization_path_logprob_std": 56.0234288514992, "trajectory_balance_rmse": 2647.087272224696} +{"canonical_length": 22, "canonical_nll": 2.1564884185791016, "generation_rouge_l_f1": 0.21428571428571427, "generation_token_f1": 0.21428571428571427, "prompt_id": "be453082a5e24aed", "reading_canonical_log_prob": -47.442745208740234, "reading_distances": [35, 24, 25, 37], "reading_log_probs": [-53.365264892578125, -51.74790954589844, -56.524932861328125, -55.07743835449219], "reading_logprob_std": 2.0722739551958314, "tokenization_path_logprob_std": 44.60886506134182, "trajectory_balance_rmse": 430.6073023652237} +{"canonical_length": 280, "canonical_nll": 1.100600106375558, "generation_rouge_l_f1": 0.30797101449275366, "generation_token_f1": 0.5543478260869565, "prompt_id": "6b6295e1afd98591", "reading_canonical_log_prob": -308.16802978515625, "reading_distances": [14, 17, 17, 20], "reading_log_probs": [-308.3017578125, -307.31048583984375, -308.39642333984375, -306.83258056640625], "reading_logprob_std": 0.7639437910496106, "tokenization_path_logprob_std": 225.74660012317207, "trajectory_balance_rmse": 6227.398423425962} +{"canonical_length": 200, "canonical_nll": 1.351807861328125, "generation_rouge_l_f1": 0.2251082251082251, "generation_token_f1": 0.3636363636363636, "prompt_id": "900ff807cdf33dd9", "reading_canonical_log_prob": -270.361572265625, "reading_distances": [23, 20, 27, 21], "reading_log_probs": [-267.2532958984375, -269.4966125488281, -272.3997497558594, -268.21942138671875], "reading_logprob_std": 2.235814980807901, "tokenization_path_logprob_std": 163.65941697783984, "trajectory_balance_rmse": 3766.2072529585166} +{"canonical_length": 120, "canonical_nll": 1.4147604624430339, "generation_rouge_l_f1": 0.28699551569506726, "generation_token_f1": 0.39461883408071746, "prompt_id": "6b5f709d21a7a023", "reading_canonical_log_prob": -169.77125549316406, "reading_distances": [13, 19, 16, 18], "reading_log_probs": [-165.0753173828125, -165.86798095703125, -166.3765411376953, -165.1749267578125], "reading_logprob_std": 0.6133416345524343, "tokenization_path_logprob_std": 116.42904952639664, "trajectory_balance_rmse": 2481.3236441335835} +{"canonical_length": 52, "canonical_nll": 1.0105705994826097, "generation_rouge_l_f1": 0.3956043956043956, "generation_token_f1": 0.6373626373626374, "prompt_id": "e072285f618dcbd9", "reading_canonical_log_prob": -52.5496711730957, "reading_distances": [28, 34, 21, 27], "reading_log_probs": [-52.55963134765625, -52.488616943359375, -52.31954574584961, -51.80353546142578], "reading_logprob_std": 0.34138851905783363, "tokenization_path_logprob_std": 40.31002663735148, "trajectory_balance_rmse": 1194.4780349333896} +{"canonical_length": 147, "canonical_nll": 1.308165050688244, "generation_rouge_l_f1": 0.1778846153846154, "generation_token_f1": 0.3173076923076923, "prompt_id": "21874077bcbf7b19", "reading_canonical_log_prob": -192.30026245117188, "reading_distances": [27, 20, 22, 23], "reading_log_probs": [-189.8476104736328, -192.36697387695312, -191.57008361816406, -190.93588256835938], "reading_logprob_std": 1.0639434991828145, "tokenization_path_logprob_std": 196.1777379352271, "trajectory_balance_rmse": 4408.469317347639} +{"canonical_length": 108, "canonical_nll": 1.094332800971137, "generation_rouge_l_f1": 0.29596412556053814, "generation_token_f1": 0.4753363228699552, "prompt_id": "8ba54ff004576b32", "reading_canonical_log_prob": -118.18794250488281, "reading_distances": [20, 23, 22, 21], "reading_log_probs": [-116.89466857910156, -116.68743133544922, -116.87388610839844, -116.87989807128906], "reading_logprob_std": 0.09808258620688547, "tokenization_path_logprob_std": 145.11558837069438, "trajectory_balance_rmse": 2784.4091047684547} +{"canonical_length": 8, "canonical_nll": 0.07379648834466934, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "bccef35e52f5dc4b", "reading_canonical_log_prob": -0.5903719067573547, "reading_distances": [34, 34, 34, 34], "reading_log_probs": [-2.6518166065216064, -3.7663493156433105, -2.3845555782318115, -0.9291405081748962], "reading_logprob_std": 1.1675674617495158, "tokenization_path_logprob_std": 34.2116810257486, "trajectory_balance_rmse": 155.11193442150164} +{"canonical_length": 53, "canonical_nll": 0.9684728946325913, "generation_rouge_l_f1": 0.1925133689839572, "generation_token_f1": 0.31016042780748665, "prompt_id": "492f8c42a0de57b2", "reading_canonical_log_prob": -51.329063415527344, "reading_distances": [47, 39, 37, 45], "reading_log_probs": [-48.107791900634766, -50.34430694580078, -49.64216613769531, -51.70561218261719], "reading_logprob_std": 1.4973470392260344, "tokenization_path_logprob_std": 90.6093596942509, "trajectory_balance_rmse": 1016.1188915973926} +{"canonical_length": 211, "canonical_nll": 1.2538081887773993, "generation_rouge_l_f1": 0.21247113163972287, "generation_token_f1": 0.3926096997690531, "prompt_id": "6aa6f6caad65a47f", "reading_canonical_log_prob": -264.55352783203125, "reading_distances": [21, 21, 29, 27], "reading_log_probs": [-271.4853515625, -270.16424560546875, -269.60162353515625, -271.9405212402344], "reading_logprob_std": 1.0970822236625777, "tokenization_path_logprob_std": 213.45193568920993, "trajectory_balance_rmse": 4078.7344932559276} +{"canonical_length": 107, "canonical_nll": 1.5826555769020152, "generation_rouge_l_f1": 0.27927927927927926, "generation_token_f1": 0.45045045045045046, "prompt_id": "83dae6df424d2709", "reading_canonical_log_prob": -169.34414672851562, "reading_distances": [31, 32, 29, 25], "reading_log_probs": [-164.09396362304688, -165.30946350097656, -166.82826232910156, -164.68821716308594], "reading_logprob_std": 1.17542392502562, "tokenization_path_logprob_std": 66.80546803082206, "trajectory_balance_rmse": 2242.801231349393} +{"canonical_length": 104, "canonical_nll": 1.0786399841308594, "generation_rouge_l_f1": 0.1881720430107527, "generation_token_f1": 0.38172043010752693, "prompt_id": "d7aad76280327bd4", "reading_canonical_log_prob": -112.17855834960938, "reading_distances": [28, 32, 36, 32], "reading_log_probs": [-114.73301696777344, -114.57601928710938, -113.84675598144531, -112.78521728515625], "reading_logprob_std": 0.8883360660119612, "tokenization_path_logprob_std": 78.28663964854671, "trajectory_balance_rmse": 2476.1867045193} +{"canonical_length": 362, "canonical_nll": 1.5777490099490676, "generation_rouge_l_f1": 0.16129032258064518, "generation_token_f1": 0.3612903225806452, "prompt_id": "117db47d424e536b", "reading_canonical_log_prob": -571.1451416015625, "reading_distances": [27, 15, 15, 23], "reading_log_probs": [-571.0584106445312, -569.9920654296875, -573.7237548828125, -570.7982177734375], "reading_logprob_std": 1.618718822354978, "tokenization_path_logprob_std": 250.37471180493907, "trajectory_balance_rmse": 7173.302738408934} +{"canonical_length": 207, "canonical_nll": 1.4006686740451388, "generation_rouge_l_f1": 0.3446808510638298, "generation_token_f1": 0.48936170212765956, "prompt_id": "f593668e2032bc26", "reading_canonical_log_prob": -289.93841552734375, "reading_distances": [32, 31, 31, 33], "reading_log_probs": [-293.53143310546875, -292.2291259765625, -290.6002197265625, -292.4971008300781], "reading_logprob_std": 1.213849352297414, "tokenization_path_logprob_std": 214.03426475938673, "trajectory_balance_rmse": 4646.098353789778} +{"canonical_length": 243, "canonical_nll": 0.9812158655237269, "generation_rouge_l_f1": 0.3275109170305677, "generation_token_f1": 0.5851528384279475, "prompt_id": "62a2987fac290e0c", "reading_canonical_log_prob": -238.43545532226562, "reading_distances": [13, 16, 18, 14], "reading_log_probs": [-236.7681121826172, -238.5118865966797, -237.55404663085938, -238.138671875], "reading_logprob_std": 0.7602307401454304, "tokenization_path_logprob_std": 222.76414342212172, "trajectory_balance_rmse": 5338.585624775886} +{"canonical_length": 28, "canonical_nll": 0.6745708329336983, "generation_rouge_l_f1": 0.24691358024691354, "generation_token_f1": 0.2962962962962963, "prompt_id": "0597c4917add4bdd", "reading_canonical_log_prob": -18.887983322143555, "reading_distances": [17, 16, 17, 20], "reading_log_probs": [-16.8520450592041, -21.36598777770996, -17.790855407714844, -20.532272338867188], "reading_logprob_std": 2.1562532367658744, "tokenization_path_logprob_std": 39.12639504235898, "trajectory_balance_rmse": 372.35999046262714} +{"canonical_length": 51, "canonical_nll": 1.602396946327359, "generation_rouge_l_f1": 0.09150326797385622, "generation_token_f1": 0.14379084967320263, "prompt_id": "ea67625f5161437c", "reading_canonical_log_prob": -81.72224426269531, "reading_distances": [19, 21, 20, 20], "reading_log_probs": [-84.18551635742188, -87.08775329589844, -84.7391128540039, -85.67984008789062], "reading_logprob_std": 1.2697030819148656, "tokenization_path_logprob_std": 103.28781052711932, "trajectory_balance_rmse": 1136.2211393064895} +{"canonical_length": 98, "canonical_nll": 0.9521470361826371, "generation_rouge_l_f1": 0.3181818181818182, "generation_token_f1": 0.5363636363636365, "prompt_id": "f9bf7abf6deb13c3", "reading_canonical_log_prob": -93.31040954589844, "reading_distances": [17, 20, 14, 24], "reading_log_probs": [-92.54690551757812, -93.40864562988281, -94.23365783691406, -93.12397766113281], "reading_logprob_std": 0.7020097787490566, "tokenization_path_logprob_std": 111.98342035041998, "trajectory_balance_rmse": 2268.880907200329} +{"canonical_length": 144, "canonical_nll": 1.4811583624945746, "generation_rouge_l_f1": 0.1846153846153846, "generation_token_f1": 0.3384615384615385, "prompt_id": "dd61adf3859cf501", "reading_canonical_log_prob": -213.28680419921875, "reading_distances": [40, 41, 39, 36], "reading_log_probs": [-233.87612915039062, -232.44061279296875, -229.0924072265625, -227.24725341796875], "reading_logprob_std": 3.0341493978682816, "tokenization_path_logprob_std": 187.62956130446764, "trajectory_balance_rmse": 2614.3218125290546} +{"canonical_length": 38, "canonical_nll": 2.4189224243164062, "generation_rouge_l_f1": 0.2452830188679245, "generation_token_f1": 0.32075471698113206, "prompt_id": "932b65e0c79cc6cd", "reading_canonical_log_prob": -91.91905212402344, "reading_distances": [31, 23, 29, 24], "reading_log_probs": [-94.33631896972656, -93.01154327392578, -97.91928100585938, -95.06919860839844], "reading_logprob_std": 2.073092797299452, "tokenization_path_logprob_std": 48.87044066341133, "trajectory_balance_rmse": 654.4387428141242} +{"canonical_length": 246, "canonical_nll": 1.3062389342765497, "generation_rouge_l_f1": 0.2035225048923679, "generation_token_f1": 0.36790606653620356, "prompt_id": "5a38778847ffa7e8", "reading_canonical_log_prob": -321.33477783203125, "reading_distances": [28, 29, 34, 30], "reading_log_probs": [-325.50579833984375, -324.0666809082031, -323.61529541015625, -325.2943115234375], "reading_logprob_std": 0.9228446910995918, "tokenization_path_logprob_std": 205.25797928954697, "trajectory_balance_rmse": 5779.500411400421} +{"canonical_length": 163, "canonical_nll": 0.8233995496129697, "generation_rouge_l_f1": 0.22714681440443216, "generation_token_f1": 0.39889196675900274, "prompt_id": "bea3ed585e39be34", "reading_canonical_log_prob": -134.21412658691406, "reading_distances": [19, 15, 16, 14], "reading_log_probs": [-133.4850616455078, -134.55160522460938, -135.2969970703125, -134.102783203125], "reading_logprob_std": 0.7629658185726538, "tokenization_path_logprob_std": 203.3277840896911, "trajectory_balance_rmse": 3161.703966416567} +{"canonical_length": 256, "canonical_nll": 0.6126832962036133, "generation_rouge_l_f1": 0.34947368421052627, "generation_token_f1": 0.5894736842105263, "prompt_id": "e77970903b119833", "reading_canonical_log_prob": -156.846923828125, "reading_distances": [21, 26, 26, 33], "reading_log_probs": [-155.72972106933594, -153.44924926757812, -155.01058959960938, -158.1840362548828], "reading_logprob_std": 1.9720572984579037, "tokenization_path_logprob_std": 147.9937418876342, "trajectory_balance_rmse": 2472.078219619203} +{"canonical_length": 28, "canonical_nll": 1.420715059552874, "generation_rouge_l_f1": 0.08097165991902834, "generation_token_f1": 0.11336032388663969, "prompt_id": "4ea6cdd31bfaf581", "reading_canonical_log_prob": -39.78002166748047, "reading_distances": [24, 17, 19, 19], "reading_log_probs": [-39.05029296875, -38.24494171142578, -38.31022644042969, -38.91666793823242], "reading_logprob_std": 0.41204715270957637, "tokenization_path_logprob_std": 6.297416111322068, "trajectory_balance_rmse": 259.9381662337088} +{"canonical_length": 92, "canonical_nll": 0.7298232368800951, "generation_rouge_l_f1": 0.44808743169398907, "generation_token_f1": 0.6666666666666667, "prompt_id": "c7f44052e9719f7b", "reading_canonical_log_prob": -67.14373779296875, "reading_distances": [30, 36, 40, 40], "reading_log_probs": [-63.88495635986328, -64.13872528076172, -64.71527099609375, -64.7537841796875], "reading_logprob_std": 0.4302003910174732, "tokenization_path_logprob_std": 163.23054664919596, "trajectory_balance_rmse": 1857.1302539651256} +{"canonical_length": 22, "canonical_nll": 2.1630325317382812, "generation_rouge_l_f1": 0.33333333333333326, "generation_token_f1": 0.4444444444444445, "prompt_id": "4d4bec1ba05bf153", "reading_canonical_log_prob": -47.58671569824219, "reading_distances": [41, 31, 45, 38], "reading_log_probs": [-50.072208404541016, -51.39176559448242, -53.601619720458984, -50.55177307128906], "reading_logprob_std": 1.5630782920415518, "tokenization_path_logprob_std": 76.36426824164104, "trajectory_balance_rmse": 708.4642352483024} +{"canonical_length": 56, "canonical_nll": 1.9496165684291296, "generation_rouge_l_f1": 0.3235294117647059, "generation_token_f1": 0.3235294117647059, "prompt_id": "2dc7b73ea6e1f7f9", "reading_canonical_log_prob": -109.17852783203125, "reading_distances": [45, 47, 42, 49], "reading_log_probs": [-111.91667175292969, -114.98307800292969, -111.098876953125, -113.41294860839844], "reading_logprob_std": 1.7131290034479052, "tokenization_path_logprob_std": 138.59096273138883, "trajectory_balance_rmse": 1235.5119320262204} +{"canonical_length": 14, "canonical_nll": 1.8101092747279577, "generation_rouge_l_f1": 0.41379310344827586, "generation_token_f1": 0.5517241379310345, "prompt_id": "cc99639a6604cf81", "reading_canonical_log_prob": -25.341529846191406, "reading_distances": [37, 35, 37, 33], "reading_log_probs": [-29.385976791381836, -32.618988037109375, -27.507150650024414, -29.86121368408203], "reading_logprob_std": 2.1112005492760835, "tokenization_path_logprob_std": 20.702954555468132, "trajectory_balance_rmse": 326.9585469648176} +{"canonical_length": 41, "canonical_nll": 1.1594725818168827, "generation_rouge_l_f1": 0.19512195121951217, "generation_token_f1": 0.2764227642276423, "prompt_id": "34af9dd76e319507", "reading_canonical_log_prob": -47.53837585449219, "reading_distances": [29, 35, 37, 35], "reading_log_probs": [-47.569976806640625, -48.17551040649414, -49.68620300292969, -47.929691314697266], "reading_logprob_std": 0.9310581453124624, "tokenization_path_logprob_std": 110.82444528798, "trajectory_balance_rmse": 921.231292798826} +{"canonical_length": 319, "canonical_nll": 1.3253735389829056, "generation_rouge_l_f1": 0.3053173241852487, "generation_token_f1": 0.5042881646655231, "prompt_id": "761d4b8b4ce8e2b9", "reading_canonical_log_prob": -422.7941589355469, "reading_distances": [29, 23, 27, 31], "reading_log_probs": [-424.61175537109375, -423.09619140625, -423.4797668457031, -424.41595458984375], "reading_logprob_std": 0.7292703299581564, "tokenization_path_logprob_std": 175.8569998487126, "trajectory_balance_rmse": 7239.266301491786} +{"canonical_length": 30, "canonical_nll": 0.6220765431722005, "generation_rouge_l_f1": 0.14814814814814814, "generation_token_f1": 0.14814814814814814, "prompt_id": "667ddd428394372e", "reading_canonical_log_prob": -18.662296295166016, "reading_distances": [32, 33, 31, 34], "reading_log_probs": [-19.45659828186035, -19.126827239990234, -18.796188354492188, -18.996660232543945], "reading_logprob_std": 0.2773216720933801, "tokenization_path_logprob_std": 55.75013646397671, "trajectory_balance_rmse": 648.9314458553297} +{"canonical_length": 15, "canonical_nll": 0.7397505442301432, "generation_rouge_l_f1": 0.20155038759689922, "generation_token_f1": 0.21705426356589147, "prompt_id": "eaab92bbea496d12", "reading_canonical_log_prob": -11.096258163452148, "reading_distances": [28, 24, 20, 23], "reading_log_probs": [-10.756086349487305, -11.872762680053711, -10.944040298461914, -11.297611236572266], "reading_logprob_std": 0.49107694329226953, "tokenization_path_logprob_std": 37.46328610515629, "trajectory_balance_rmse": 351.8386723051004} +{"canonical_length": 77, "canonical_nll": 0.9370334179370434, "generation_rouge_l_f1": 0.7173913043478262, "generation_token_f1": 0.8043478260869565, "prompt_id": "7cb40862b1b15473", "reading_canonical_log_prob": -72.15157318115234, "reading_distances": [33, 25, 25, 25], "reading_log_probs": [-71.60620880126953, -76.84501647949219, -78.32688903808594, -71.08999633789062], "reading_logprob_std": 3.6579587102027733, "tokenization_path_logprob_std": 102.83503207368396, "trajectory_balance_rmse": 770.9737489914404} +{"canonical_length": 37, "canonical_nll": 1.5015560871845968, "generation_rouge_l_f1": 0.2589928057553957, "generation_token_f1": 0.38848920863309355, "prompt_id": "b218ec9e88427665", "reading_canonical_log_prob": -55.55757522583008, "reading_distances": [28, 35, 29, 27], "reading_log_probs": [-57.34763717651367, -56.35003662109375, -58.46199417114258, -57.457542419433594], "reading_logprob_std": 0.8633720060586891, "tokenization_path_logprob_std": 81.0589746139078, "trajectory_balance_rmse": 1012.9701711789745} +{"canonical_length": 68, "canonical_nll": 1.382959478041705, "generation_rouge_l_f1": 0.22686567164179103, "generation_token_f1": 0.3044776119402985, "prompt_id": "605c5bbf05c7cb07", "reading_canonical_log_prob": -94.04124450683594, "reading_distances": [34, 31, 35, 28], "reading_log_probs": [-90.29475402832031, -90.67405700683594, -91.2431869506836, -90.49849700927734], "reading_logprob_std": 0.4076559287132849, "tokenization_path_logprob_std": 86.89701918752031, "trajectory_balance_rmse": 1929.7562780999417} +{"canonical_length": 163, "canonical_nll": 2.0197589148773005, "generation_rouge_l_f1": 0.1895734597156398, "generation_token_f1": 0.4028436018957346, "prompt_id": "371fd9b264246fc1", "reading_canonical_log_prob": -329.220703125, "reading_distances": [18, 15, 18, 17], "reading_log_probs": [-331.6519470214844, -329.4468078613281, -330.1543884277344, -329.50933837890625], "reading_logprob_std": 1.025376440699427, "tokenization_path_logprob_std": 179.3532480101623, "trajectory_balance_rmse": 3534.000388985284} +{"canonical_length": 9, "canonical_nll": 2.372654596964518, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.3157894736842105, "prompt_id": "ed3c13f24a96d3b3", "reading_canonical_log_prob": -21.353891372680664, "reading_distances": [28, 24, 28, 29], "reading_log_probs": [-23.67765235900879, -23.29538917541504, -23.975849151611328, -22.548601150512695], "reading_logprob_std": 0.6169520571621159, "tokenization_path_logprob_std": 28.745667453324508, "trajectory_balance_rmse": 175.4697146016012} +{"canonical_length": 27, "canonical_nll": 1.0228700284604673, "generation_rouge_l_f1": 0.4745762711864407, "generation_token_f1": 0.4745762711864407, "prompt_id": "9e311e757597dd6a", "reading_canonical_log_prob": -27.617490768432617, "reading_distances": [32, 34, 33, 34], "reading_log_probs": [-27.1921443939209, -28.39319610595703, -27.75556755065918, -27.850839614868164], "reading_logprob_std": 0.4919051201543442, "tokenization_path_logprob_std": 52.123494323689314, "trajectory_balance_rmse": 647.6456970771168} +{"canonical_length": 23, "canonical_nll": 0.913055419921875, "generation_rouge_l_f1": 0.20512820512820512, "generation_token_f1": 0.23931623931623933, "prompt_id": "6d2468c671bba59a", "reading_canonical_log_prob": -21.000274658203125, "reading_distances": [38, 41, 45, 45], "reading_log_probs": [-21.025325775146484, -21.148544311523438, -20.482948303222656, -20.67238426208496], "reading_logprob_std": 0.3081607079183395, "tokenization_path_logprob_std": 54.53465194991536, "trajectory_balance_rmse": 472.04660674206565} +{"canonical_length": 34, "canonical_nll": 2.009814542882583, "generation_rouge_l_f1": 0.21428571428571427, "generation_token_f1": 0.3, "prompt_id": "d8ebdfdb02245d42", "reading_canonical_log_prob": -68.33369445800781, "reading_distances": [20, 20, 21, 20], "reading_log_probs": [-68.57972717285156, -66.85675048828125, -68.40461730957031, -71.255859375], "reading_logprob_std": 1.826623198064721, "tokenization_path_logprob_std": 66.20511409336743, "trajectory_balance_rmse": 987.0229219219887} +{"canonical_length": 282, "canonical_nll": 1.1251705521387412, "generation_rouge_l_f1": 0.21863799283154123, "generation_token_f1": 0.39784946236559143, "prompt_id": "0659b1d0d1c9b272", "reading_canonical_log_prob": -317.298095703125, "reading_distances": [22, 21, 19, 21], "reading_log_probs": [-315.8409423828125, -318.35235595703125, -318.69964599609375, -315.9510803222656], "reading_logprob_std": 1.5256928252371607, "tokenization_path_logprob_std": 135.5934665014224, "trajectory_balance_rmse": 5339.700458779529} +{"canonical_length": 126, "canonical_nll": 1.4649983966161335, "generation_rouge_l_f1": 0.2953586497890296, "generation_token_f1": 0.5232067510548524, "prompt_id": "11e44461a25c9604", "reading_canonical_log_prob": -184.5897979736328, "reading_distances": [13, 16, 12, 13], "reading_log_probs": [-184.27304077148438, -183.88632202148438, -184.26773071289062, -183.52825927734375], "reading_logprob_std": 0.3564616137526443, "tokenization_path_logprob_std": 180.20961525225903, "trajectory_balance_rmse": 3405.30314698811} +{"canonical_length": 4, "canonical_nll": 0.6753737926483154, "generation_rouge_l_f1": 0.0163265306122449, "generation_token_f1": 0.02448979591836735, "prompt_id": "72cbdeacc12e1d9a", "reading_canonical_log_prob": -2.7014951705932617, "reading_distances": [20, 24, 19, 21], "reading_log_probs": [-3.26790452003479, -3.6020352840423584, -3.3316261768341064, -3.287367820739746], "reading_logprob_std": 0.1555040559177645, "tokenization_path_logprob_std": 29.94568944904992, "trajectory_balance_rmse": 78.7064341320424} +{"canonical_length": 9, "canonical_nll": 1.6543286641438801, "generation_rouge_l_f1": 0.875, "generation_token_f1": 0.875, "prompt_id": "e46df451982e05b1", "reading_canonical_log_prob": -14.888957977294922, "reading_distances": [16, 22, 15, 22], "reading_log_probs": [-15.329198837280273, -15.874237060546875, -16.262405395507812, -16.063457489013672], "reading_logprob_std": 0.40136602259240434, "tokenization_path_logprob_std": 34.01864291071876, "trajectory_balance_rmse": 129.23247401701414} +{"canonical_length": 112, "canonical_nll": 1.0884534972054618, "generation_rouge_l_f1": 0.3280423280423281, "generation_token_f1": 0.5026455026455027, "prompt_id": "ad4b8c7e2797f458", "reading_canonical_log_prob": -121.90679168701172, "reading_distances": [36, 35, 37, 30], "reading_log_probs": [-111.8682632446289, -112.21101379394531, -111.38055419921875, -115.2169189453125], "reading_logprob_std": 1.732331195150831, "tokenization_path_logprob_std": 104.6578698190183, "trajectory_balance_rmse": 1924.5130312358044} +{"canonical_length": 229, "canonical_nll": 1.297989358027429, "generation_rouge_l_f1": 0.255863539445629, "generation_token_f1": 0.4690831556503199, "prompt_id": "9d2ca51ff7ce1141", "reading_canonical_log_prob": -297.23956298828125, "reading_distances": [29, 33, 32, 26], "reading_log_probs": [-294.03643798828125, -296.22674560546875, -293.0088195800781, -294.30780029296875], "reading_logprob_std": 1.3432496188009473, "tokenization_path_logprob_std": 196.05172865507708, "trajectory_balance_rmse": 5494.632271960523} +{"canonical_length": 3, "canonical_nll": 1.937777042388916, "generation_rouge_l_f1": 0.01639344262295082, "generation_token_f1": 0.01639344262295082, "prompt_id": "f22d1f7967fa63cf", "reading_canonical_log_prob": -5.813331127166748, "reading_distances": [24, 31, 24, 32], "reading_log_probs": [-6.677822589874268, -6.736556053161621, -6.352509498596191, -6.675229072570801], "reading_logprob_std": 0.17432862440587746, "tokenization_path_logprob_std": 36.25143085262078, "trajectory_balance_rmse": 60.498766336163065} +{"canonical_length": 341, "canonical_nll": 1.3756934023322946, "generation_rouge_l_f1": 0.19281045751633985, "generation_token_f1": 0.38888888888888884, "prompt_id": "a546d753c3fe436f", "reading_canonical_log_prob": -469.1114501953125, "reading_distances": [20, 22, 25, 22], "reading_log_probs": [-474.7778015136719, -474.236328125, -474.140869140625, -473.1587219238281], "reading_logprob_std": 0.6742437550500169, "tokenization_path_logprob_std": 150.83247622813693, "trajectory_balance_rmse": 7544.358230779027} +{"canonical_length": 175, "canonical_nll": 1.5855876813616072, "generation_rouge_l_f1": 0.23500000000000001, "generation_token_f1": 0.385, "prompt_id": "55985253cc61ade7", "reading_canonical_log_prob": -277.47784423828125, "reading_distances": [25, 30, 26, 24], "reading_log_probs": [-274.689208984375, -274.90850830078125, -274.0775146484375, -275.5773620605469], "reading_logprob_std": 0.6190406623227077, "tokenization_path_logprob_std": 124.04910946558647, "trajectory_balance_rmse": 3425.6824669825032} +{"canonical_length": 134, "canonical_nll": 1.4465641761893657, "generation_rouge_l_f1": 0.3453815261044177, "generation_token_f1": 0.5622489959839357, "prompt_id": "e4db8e939e6495ed", "reading_canonical_log_prob": -193.839599609375, "reading_distances": [17, 15, 20, 15], "reading_log_probs": [-193.64956665039062, -193.12002563476562, -192.31048583984375, -192.27874755859375], "reading_logprob_std": 0.6656329623214, "tokenization_path_logprob_std": 94.18690506679829, "trajectory_balance_rmse": 3153.660649394038} +{"canonical_length": 36, "canonical_nll": 0.7154826058281792, "generation_rouge_l_f1": 0.19444444444444445, "generation_token_f1": 0.24074074074074076, "prompt_id": "8ba984bcebc3168b", "reading_canonical_log_prob": -25.757373809814453, "reading_distances": [16, 17, 15, 15], "reading_log_probs": [-32.195960998535156, -32.88325881958008, -26.953733444213867, -30.14447593688965], "reading_logprob_std": 2.6614995693914407, "tokenization_path_logprob_std": 62.596650948496034, "trajectory_balance_rmse": 793.2173485641118} +{"canonical_length": 4, "canonical_nll": 2.529656410217285, "generation_rouge_l_f1": 0.015094339622641508, "generation_token_f1": 0.015094339622641508, "prompt_id": "fe48fe173961fdd5", "reading_canonical_log_prob": -10.11862564086914, "reading_distances": [18, 27, 24, 23], "reading_log_probs": [-11.190165519714355, -11.275017738342285, -11.818737030029297, -10.576738357543945], "reading_logprob_std": 0.5086239925958931, "tokenization_path_logprob_std": 18.721179033501514, "trajectory_balance_rmse": 43.67323990758188} +{"canonical_length": 267, "canonical_nll": 1.727040494425913, "generation_rouge_l_f1": 0.13618677042801555, "generation_token_f1": 0.19844357976653695, "prompt_id": "51d00b55119644b0", "reading_canonical_log_prob": -461.11981201171875, "reading_distances": [25, 28, 26, 25], "reading_log_probs": [-464.46966552734375, -467.25506591796875, -466.35528564453125, -465.2402038574219], "reading_logprob_std": 1.2254396808219894, "tokenization_path_logprob_std": 180.4938012986226, "trajectory_balance_rmse": 5986.789901800253} +{"canonical_length": 7, "canonical_nll": 0.09090966837746757, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "e708a1adc69b9f0e", "reading_canonical_log_prob": -0.636367678642273, "reading_distances": [22, 28, 24, 25], "reading_log_probs": [-0.6151078939437866, -0.5675818920135498, -0.7120668292045593, -0.6839572787284851], "reading_logprob_std": 0.06558027416057535, "tokenization_path_logprob_std": 15.773963792591282, "trajectory_balance_rmse": 119.64950026411272} +{"canonical_length": 22, "canonical_nll": 1.126007686961781, "generation_rouge_l_f1": 0.11764705882352941, "generation_token_f1": 0.11764705882352941, "prompt_id": "d7a89d7235636e3d", "reading_canonical_log_prob": -24.77216911315918, "reading_distances": [14, 16, 15, 14], "reading_log_probs": [-24.036588668823242, -25.020273208618164, -24.612163543701172, -24.846195220947266], "reading_logprob_std": 0.4287590572424554, "tokenization_path_logprob_std": 49.12946044843485, "trajectory_balance_rmse": 573.9752243605367} +{"canonical_length": 12, "canonical_nll": 0.5179082552591959, "generation_rouge_l_f1": 0.8695652173913043, "generation_token_f1": 0.8695652173913043, "prompt_id": "17e2c9ee9612ad40", "reading_canonical_log_prob": -6.214899063110352, "reading_distances": [21, 22, 23, 29], "reading_log_probs": [-6.265376567840576, -6.561233043670654, -6.770655632019043, -6.573461055755615], "reading_logprob_std": 0.20829608021182533, "tokenization_path_logprob_std": 28.329682612333396, "trajectory_balance_rmse": 274.4224281998838} +{"canonical_length": 324, "canonical_nll": 1.089334558557581, "generation_rouge_l_f1": 0.1835205992509363, "generation_token_f1": 0.38951310861423216, "prompt_id": "36541d1f10617d25", "reading_canonical_log_prob": -352.94439697265625, "reading_distances": [11, 11, 15, 17], "reading_log_probs": [-359.75238037109375, -365.37506103515625, -357.32177734375, -362.8626708984375], "reading_logprob_std": 3.5245006725394, "tokenization_path_logprob_std": 221.81383242112426, "trajectory_balance_rmse": 6656.434182702882} +{"canonical_length": 315, "canonical_nll": 1.1320226275731646, "generation_rouge_l_f1": 0.24825174825174826, "generation_token_f1": 0.5734265734265734, "prompt_id": "e15e866d282a743f", "reading_canonical_log_prob": -356.5871276855469, "reading_distances": [24, 27, 30, 33], "reading_log_probs": [-352.89898681640625, -354.8369140625, -355.1068115234375, -356.4198303222656], "reading_logprob_std": 1.4528385386261735, "tokenization_path_logprob_std": 217.54450772457875, "trajectory_balance_rmse": 6200.431803089495} +{"canonical_length": 10, "canonical_nll": 1.5699334144592285, "generation_rouge_l_f1": 0.057971014492753624, "generation_token_f1": 0.07246376811594203, "prompt_id": "718643fa581e9e44", "reading_canonical_log_prob": -15.699334144592285, "reading_distances": [18, 21, 22, 21], "reading_log_probs": [-17.915081024169922, -18.117650985717773, -17.538097381591797, -18.08159828186035], "reading_logprob_std": 0.26511863057024115, "tokenization_path_logprob_std": 24.61271360939395, "trajectory_balance_rmse": 130.91053202614606} +{"canonical_length": 159, "canonical_nll": 0.9248518074083628, "generation_rouge_l_f1": 0.3076923076923077, "generation_token_f1": 0.5174825174825175, "prompt_id": "482bbea97f7ae0fa", "reading_canonical_log_prob": -147.0514373779297, "reading_distances": [12, 10, 15, 12], "reading_log_probs": [-147.66073608398438, -146.99981689453125, -146.83090209960938, -147.064697265625], "reading_logprob_std": 0.36149028799210026, "tokenization_path_logprob_std": 190.04023609458957, "trajectory_balance_rmse": 3368.748420480166} +{"canonical_length": 11, "canonical_nll": 1.4559253345836292, "generation_rouge_l_f1": 0.057971014492753624, "generation_token_f1": 0.06521739130434782, "prompt_id": "f9bf285092a25c3b", "reading_canonical_log_prob": -16.015178680419922, "reading_distances": [47, 46, 43, 42], "reading_log_probs": [-14.762838363647461, -15.404489517211914, -16.187440872192383, -14.106996536254883], "reading_logprob_std": 0.8895732360526242, "tokenization_path_logprob_std": 29.74992095955344, "trajectory_balance_rmse": 238.3867135009727} +{"canonical_length": 150, "canonical_nll": 0.902379150390625, "generation_rouge_l_f1": 0.39598997493734334, "generation_token_f1": 0.5363408521303258, "prompt_id": "9d4eb4020bd4213f", "reading_canonical_log_prob": -135.35687255859375, "reading_distances": [31, 18, 28, 24], "reading_log_probs": [-134.20164489746094, -134.08084106445312, -134.8432159423828, -134.2315673828125], "reading_logprob_std": 0.34219270458447537, "tokenization_path_logprob_std": 110.73331817408646, "trajectory_balance_rmse": 2936.1254510953136} +{"canonical_length": 134, "canonical_nll": 1.0808567787284282, "generation_rouge_l_f1": 0.2535211267605634, "generation_token_f1": 0.3887323943661971, "prompt_id": "13778e3d12464b89", "reading_canonical_log_prob": -144.83480834960938, "reading_distances": [18, 19, 21, 19], "reading_log_probs": [-146.27381896972656, -145.75198364257812, -145.34982299804688, -146.76719665527344], "reading_logprob_std": 0.6171737285327016, "tokenization_path_logprob_std": 109.58922268685221, "trajectory_balance_rmse": 1652.1105925424874} +{"canonical_length": 395, "canonical_nll": 1.3168994758702532, "generation_rouge_l_f1": 0.309375, "generation_token_f1": 0.46249999999999997, "prompt_id": "513a1cb695ee9eb0", "reading_canonical_log_prob": -520.17529296875, "reading_distances": [25, 24, 16, 27], "reading_log_probs": [-526.7518920898438, -525.35693359375, -522.478515625, -529.6927490234375], "reading_logprob_std": 2.9998061616577996, "tokenization_path_logprob_std": 382.5450092674971, "trajectory_balance_rmse": 6046.189364056378} +{"canonical_length": 193, "canonical_nll": 1.0890551156948267, "generation_rouge_l_f1": 0.34989200863930886, "generation_token_f1": 0.531317494600432, "prompt_id": "444facd57b3cdbe1", "reading_canonical_log_prob": -210.18763732910156, "reading_distances": [24, 22, 22, 14], "reading_log_probs": [-215.3566131591797, -212.2391815185547, -217.8266143798828, -217.51190185546875], "reading_logprob_std": 2.575276443620112, "tokenization_path_logprob_std": 186.68848142141377, "trajectory_balance_rmse": 4154.965588408623} +{"canonical_length": 226, "canonical_nll": 1.1190323281077157, "generation_rouge_l_f1": 0.26495726495726496, "generation_token_f1": 0.6025641025641025, "prompt_id": "5870502ae5fa3ae7", "reading_canonical_log_prob": -252.90130615234375, "reading_distances": [27, 20, 23, 23], "reading_log_probs": [-256.1521301269531, -254.95098876953125, -255.06057739257812, -254.9532470703125], "reading_logprob_std": 0.5841719980901846, "tokenization_path_logprob_std": 222.19190349349856, "trajectory_balance_rmse": 2944.926141291934} +{"canonical_length": 17, "canonical_nll": 1.0413013907039868, "generation_rouge_l_f1": 0.8387096774193549, "generation_token_f1": 0.9032258064516129, "prompt_id": "8c61a7a56532124f", "reading_canonical_log_prob": -17.702123641967773, "reading_distances": [29, 40, 31, 29], "reading_log_probs": [-16.509607315063477, -18.462913513183594, -15.77542495727539, -14.434966087341309], "reading_logprob_std": 1.680831893422081, "tokenization_path_logprob_std": 84.26326156223162, "trajectory_balance_rmse": 396.3358282864397} +{"canonical_length": 12, "canonical_nll": 0.6218854586283366, "generation_rouge_l_f1": 0.05622489959839358, "generation_token_f1": 0.05622489959839358, "prompt_id": "fb0ad86c22711d8d", "reading_canonical_log_prob": -7.462625503540039, "reading_distances": [9, 12, 14, 16], "reading_log_probs": [-6.870209693908691, -6.078676223754883, -6.706172466278076, -6.549316883087158], "reading_logprob_std": 0.34110891235242113, "tokenization_path_logprob_std": 38.669268303481225, "trajectory_balance_rmse": 182.94777982279692} +{"canonical_length": 219, "canonical_nll": 1.374899110837614, "generation_rouge_l_f1": 0.18067226890756302, "generation_token_f1": 0.3739495798319328, "prompt_id": "a18ee8ea214b4b09", "reading_canonical_log_prob": -301.1029052734375, "reading_distances": [36, 42, 52, 45], "reading_log_probs": [-303.25323486328125, -301.7581787109375, -300.94427490234375, -301.74249267578125], "reading_logprob_std": 0.9638753391102822, "tokenization_path_logprob_std": 175.86342661576862, "trajectory_balance_rmse": 5171.696982114435} +{"canonical_length": 283, "canonical_nll": 1.3057539976948984, "generation_rouge_l_f1": 0.21505376344086022, "generation_token_f1": 0.4551971326164875, "prompt_id": "6933d659ab69515c", "reading_canonical_log_prob": -369.52838134765625, "reading_distances": [32, 27, 25, 23], "reading_log_probs": [-371.70556640625, -372.61614990234375, -370.32159423828125, -364.88067626953125], "reading_logprob_std": 3.4644599507647733, "tokenization_path_logprob_std": 399.7080345436352, "trajectory_balance_rmse": 5705.926698711274} +{"canonical_length": 49, "canonical_nll": 1.3465633781588808, "generation_rouge_l_f1": 0.16666666666666669, "generation_token_f1": 0.24242424242424243, "prompt_id": "e748f45af7778fcd", "reading_canonical_log_prob": -65.98160552978516, "reading_distances": [21, 31, 38, 33], "reading_log_probs": [-70.61395263671875, -69.97591400146484, -70.94036865234375, -70.42973327636719], "reading_logprob_std": 0.4025388887150194, "tokenization_path_logprob_std": 84.11845508543084, "trajectory_balance_rmse": 940.2141981677402} +{"canonical_length": 8, "canonical_nll": 1.3554458618164062, "generation_rouge_l_f1": 0.631578947368421, "generation_token_f1": 0.631578947368421, "prompt_id": "3f21812c287e05df", "reading_canonical_log_prob": -10.84356689453125, "reading_distances": [41, 27, 42, 32], "reading_log_probs": [-14.199145317077637, -13.978394508361816, -14.012706756591797, -14.990800857543945], "reading_logprob_std": 0.4737281892689096, "tokenization_path_logprob_std": 27.85618313354053, "trajectory_balance_rmse": 187.5908292125197} +{"canonical_length": 38, "canonical_nll": 1.6021519711143093, "generation_rouge_l_f1": 0.09523809523809523, "generation_token_f1": 0.14965986394557823, "prompt_id": "cc2b221229d6b7ba", "reading_canonical_log_prob": -60.88177490234375, "reading_distances": [27, 19, 21, 26], "reading_log_probs": [-63.43495178222656, -63.86545181274414, -64.81475830078125, -60.808433532714844], "reading_logprob_std": 1.7147637383379457, "tokenization_path_logprob_std": 216.2326620852851, "trajectory_balance_rmse": 1074.1030798863274} +{"canonical_length": 57, "canonical_nll": 0.9392225700512267, "generation_rouge_l_f1": 0.23000000000000004, "generation_token_f1": 0.36, "prompt_id": "6c3587030700c60e", "reading_canonical_log_prob": -53.53568649291992, "reading_distances": [33, 25, 30, 23], "reading_log_probs": [-56.096195220947266, -55.596473693847656, -54.737281799316406, -54.6457405090332], "reading_logprob_std": 0.6982520003319782, "tokenization_path_logprob_std": 74.44103761027618, "trajectory_balance_rmse": 1061.372651681869} +{"canonical_length": 108, "canonical_nll": 1.6972810250741464, "generation_rouge_l_f1": 0.18604651162790697, "generation_token_f1": 0.3023255813953488, "prompt_id": "3f751ca73a73cf7d", "reading_canonical_log_prob": -183.3063507080078, "reading_distances": [26, 21, 13, 23], "reading_log_probs": [-179.3701171875, -180.07058715820312, -181.68637084960938, -180.2462615966797], "reading_logprob_std": 0.9720618695566192, "tokenization_path_logprob_std": 49.163983977304255, "trajectory_balance_rmse": 1867.7438216028536} +{"canonical_length": 77, "canonical_nll": 0.6958151978331727, "generation_rouge_l_f1": 0.24489795918367346, "generation_token_f1": 0.3401360544217687, "prompt_id": "7c7baba7466fe6f6", "reading_canonical_log_prob": -53.5777702331543, "reading_distances": [8, 7, 7, 11], "reading_log_probs": [-54.983272552490234, -53.24934005737305, -53.931236267089844, -54.88029479980469], "reading_logprob_std": 0.8240962005016859, "tokenization_path_logprob_std": 142.14408319522718, "trajectory_balance_rmse": 1632.023397393824} +{"canonical_length": 294, "canonical_nll": 0.7623708296795281, "generation_rouge_l_f1": 0.2333931777378815, "generation_token_f1": 0.5134649910233393, "prompt_id": "fd3d2e71f4e4243f", "reading_canonical_log_prob": -224.13702392578125, "reading_distances": [15, 19, 17, 20], "reading_log_probs": [-220.2464599609375, -221.12884521484375, -220.2597198486328, -220.2129364013672], "reading_logprob_std": 0.44500561210998135, "tokenization_path_logprob_std": 197.3129820708024, "trajectory_balance_rmse": 5200.020911809938} +{"canonical_length": 399, "canonical_nll": 0.8846158670602287, "generation_rouge_l_f1": 0.2208398133748056, "generation_token_f1": 0.46967340590979784, "prompt_id": "5bc59064e72854cb", "reading_canonical_log_prob": -352.96173095703125, "reading_distances": [22, 20, 21, 24], "reading_log_probs": [-353.40753173828125, -355.42034912109375, -356.4289245605469, -355.2777404785156], "reading_logprob_std": 1.2596532448709719, "tokenization_path_logprob_std": 76.34308273267415, "trajectory_balance_rmse": 7977.170311627548} +{"canonical_length": 280, "canonical_nll": 1.1940475463867188, "generation_rouge_l_f1": 0.297951582867784, "generation_token_f1": 0.5512104283054003, "prompt_id": "cd25f5705dfa5cbb", "reading_canonical_log_prob": -334.33331298828125, "reading_distances": [22, 18, 25, 28], "reading_log_probs": [-332.5599670410156, -333.01593017578125, -333.7487487792969, -336.1809997558594], "reading_logprob_std": 1.6125423945044237, "tokenization_path_logprob_std": 225.57418122490174, "trajectory_balance_rmse": 6478.6093830995} +{"canonical_length": 365, "canonical_nll": 0.8337658189747431, "generation_rouge_l_f1": 0.2545454545454546, "generation_token_f1": 0.4925619834710743, "prompt_id": "ee3200ffbd24e2bc", "reading_canonical_log_prob": -304.32452392578125, "reading_distances": [26, 28, 28, 31], "reading_log_probs": [-307.00762939453125, -303.1528015136719, -309.1146240234375, -306.1865234375], "reading_logprob_std": 2.4714015716524402, "tokenization_path_logprob_std": 237.69863310238034, "trajectory_balance_rmse": 5485.104119071643} +{"canonical_length": 115, "canonical_nll": 2.0189511506453806, "generation_rouge_l_f1": 0.3886255924170617, "generation_token_f1": 0.5497630331753556, "prompt_id": "4180e393ac001d5f", "reading_canonical_log_prob": -232.17938232421875, "reading_distances": [23, 25, 20, 17], "reading_log_probs": [-234.4092559814453, -236.419921875, -237.98486328125, -236.8501739501953], "reading_logprob_std": 1.4918529797789, "tokenization_path_logprob_std": 87.25835407407332, "trajectory_balance_rmse": 2647.221951032653} +{"canonical_length": 9, "canonical_nll": 1.0167359246148004, "generation_rouge_l_f1": 0.4615384615384615, "generation_token_f1": 0.4615384615384615, "prompt_id": "79896777ff1e0459", "reading_canonical_log_prob": -9.150623321533203, "reading_distances": [36, 51, 52, 45], "reading_log_probs": [-7.860236167907715, -8.8201265335083, -8.507654190063477, -8.332767486572266], "reading_logprob_std": 0.40099549266067064, "tokenization_path_logprob_std": 41.614902229521086, "trajectory_balance_rmse": 193.3633908025289} +{"canonical_length": 147, "canonical_nll": 1.4444273864330888, "generation_rouge_l_f1": 0.3007518796992481, "generation_token_f1": 0.4761904761904762, "prompt_id": "4be1dda9c6193970", "reading_canonical_log_prob": -212.33082580566406, "reading_distances": [43, 40, 45, 43], "reading_log_probs": [-211.9818115234375, -211.81991577148438, -212.51742553710938, -211.30081176757812], "reading_logprob_std": 0.5010814286661627, "tokenization_path_logprob_std": 206.6729323821853, "trajectory_balance_rmse": 4182.824799437978} +{"canonical_length": 9, "canonical_nll": 1.8138595157199435, "generation_rouge_l_f1": 0.019230769230769232, "generation_token_f1": 0.019230769230769232, "prompt_id": "e275be5afb765439", "reading_canonical_log_prob": -16.324735641479492, "reading_distances": [32, 31, 30, 24], "reading_log_probs": [-25.28191566467285, -25.031835556030273, -19.411611557006836, -25.725696563720703], "reading_logprob_std": 2.981274646757378, "tokenization_path_logprob_std": 33.29404036661821, "trajectory_balance_rmse": 126.8262911874006} +{"canonical_length": 8, "canonical_nll": 1.6020323038101196, "generation_rouge_l_f1": 0.01818181818181818, "generation_token_f1": 0.01818181818181818, "prompt_id": "b4100d452af762d4", "reading_canonical_log_prob": -12.816258430480957, "reading_distances": [17, 18, 23, 23], "reading_log_probs": [-15.304988861083984, -15.907352447509766, -15.657504081726074, -16.117565155029297], "reading_logprob_std": 0.3494828183633587, "tokenization_path_logprob_std": 8.32487693716447, "trajectory_balance_rmse": 111.24102993233926} +{"canonical_length": 65, "canonical_nll": 0.6436298076923077, "generation_rouge_l_f1": 0.2539682539682539, "generation_token_f1": 0.3571428571428571, "prompt_id": "660009a4d953661a", "reading_canonical_log_prob": -41.8359375, "reading_distances": [24, 31, 30, 34], "reading_log_probs": [-41.92665481567383, -42.27499008178711, -42.20320129394531, -41.173973083496094], "reading_logprob_std": 0.5034090427596876, "tokenization_path_logprob_std": 58.208972675616835, "trajectory_balance_rmse": 895.8487149062375} +{"canonical_length": 245, "canonical_nll": 0.9266600316884567, "generation_rouge_l_f1": 0.3478260869565218, "generation_token_f1": 0.6864988558352402, "prompt_id": "9c1c5be112145501", "reading_canonical_log_prob": -227.03170776367188, "reading_distances": [26, 30, 16, 16], "reading_log_probs": [-227.136962890625, -228.11563110351562, -225.83090209960938, -226.84384155273438], "reading_logprob_std": 0.9404337650760628, "tokenization_path_logprob_std": 72.38878953221848, "trajectory_balance_rmse": 4377.619364192726} +{"canonical_length": 41, "canonical_nll": 0.6631022662651248, "generation_rouge_l_f1": 0.13377926421404682, "generation_token_f1": 0.17391304347826086, "prompt_id": "8a1926c471e546ee", "reading_canonical_log_prob": -27.187192916870117, "reading_distances": [21, 15, 27, 24], "reading_log_probs": [-23.954700469970703, -25.40302085876465, -24.001182556152344, -23.11853790283203], "reading_logprob_std": 0.9470143855942073, "tokenization_path_logprob_std": 41.59941467266021, "trajectory_balance_rmse": 489.25359215701394} +{"canonical_length": 311, "canonical_nll": 1.197593370051246, "generation_rouge_l_f1": 0.21989528795811517, "generation_token_f1": 0.4712041884816754, "prompt_id": "a1e93e4a3a2191cd", "reading_canonical_log_prob": -372.4515380859375, "reading_distances": [37, 35, 36, 42], "reading_log_probs": [-374.4319152832031, -376.78875732421875, -374.08489990234375, -375.02490234375], "reading_logprob_std": 1.2018177785695559, "tokenization_path_logprob_std": 307.49122218420007, "trajectory_balance_rmse": 6424.022468795137} +{"canonical_length": 130, "canonical_nll": 1.4602382953350361, "generation_rouge_l_f1": 0.21750663129973477, "generation_token_f1": 0.41909814323607425, "prompt_id": "c4f0abd89b25e9b2", "reading_canonical_log_prob": -189.8309783935547, "reading_distances": [22, 21, 18, 21], "reading_log_probs": [-190.469970703125, -188.5046844482422, -190.96774291992188, -189.66629028320312], "reading_logprob_std": 1.0749338575451417, "tokenization_path_logprob_std": 172.72715710898174, "trajectory_balance_rmse": 3273.996865138175} +{"canonical_length": 226, "canonical_nll": 1.1586251047860205, "generation_rouge_l_f1": 0.25986078886310904, "generation_token_f1": 0.4501160092807424, "prompt_id": "8d709bdbfc4890e5", "reading_canonical_log_prob": -261.8492736816406, "reading_distances": [23, 27, 30, 20], "reading_log_probs": [-262.8813171386719, -257.6772766113281, -259.1809387207031, -260.07037353515625], "reading_logprob_std": 2.188135604872283, "tokenization_path_logprob_std": 103.45274178152084, "trajectory_balance_rmse": 3897.540253244391} +{"canonical_length": 410, "canonical_nll": 1.1731932105087652, "generation_rouge_l_f1": 0.25249169435215946, "generation_token_f1": 0.40531561461794025, "prompt_id": "fbcfdeca84611d04", "reading_canonical_log_prob": -481.00921630859375, "reading_distances": [30, 19, 29, 28], "reading_log_probs": [-484.3851013183594, -484.16253662109375, -492.2215576171875, -491.0412902832031], "reading_logprob_std": 4.276121063950689, "tokenization_path_logprob_std": 185.7472565201469, "trajectory_balance_rmse": 7747.121758295403} +{"canonical_length": 45, "canonical_nll": 1.3403600056966145, "generation_rouge_l_f1": 0.22222222222222218, "generation_token_f1": 0.25396825396825395, "prompt_id": "8c6898d598b2732f", "reading_canonical_log_prob": -60.316200256347656, "reading_distances": [31, 31, 33, 26], "reading_log_probs": [-65.74589538574219, -66.5399169921875, -66.03729248046875, -65.20452880859375], "reading_logprob_std": 0.5581105537904617, "tokenization_path_logprob_std": 9.8784380035426, "trajectory_balance_rmse": 990.9423276864432} +{"canonical_length": 93, "canonical_nll": 0.6091103912681661, "generation_rouge_l_f1": 0.34545454545454546, "generation_token_f1": 0.5909090909090909, "prompt_id": "f5259298cfe12698", "reading_canonical_log_prob": -56.64726638793945, "reading_distances": [19, 16, 17, 21], "reading_log_probs": [-57.69274139404297, -57.54126739501953, -57.64112091064453, -57.61016082763672], "reading_logprob_std": 0.06331413284483402, "tokenization_path_logprob_std": 54.30265068112739, "trajectory_balance_rmse": 1255.896264497505} +{"canonical_length": 183, "canonical_nll": 0.3711944329934042, "generation_rouge_l_f1": 0.20197044334975367, "generation_token_f1": 0.354679802955665, "prompt_id": "af8c4385ce2b0480", "reading_canonical_log_prob": -67.92858123779297, "reading_distances": [31, 31, 37, 35], "reading_log_probs": [-69.51457214355469, -71.14958190917969, -72.12380981445312, -67.82076263427734], "reading_logprob_std": 1.8906952341629202, "tokenization_path_logprob_std": 48.970007530890044, "trajectory_balance_rmse": 2806.74006662652} +{"canonical_length": 164, "canonical_nll": 1.6510906684689406, "generation_rouge_l_f1": 0.20673076923076925, "generation_token_f1": 0.3798076923076923, "prompt_id": "8db3810d387bcbf4", "reading_canonical_log_prob": -270.77886962890625, "reading_distances": [24, 22, 19, 25], "reading_log_probs": [-278.6978454589844, -279.0435485839844, -275.71319580078125, -277.5609436035156], "reading_logprob_std": 1.5006589994905666, "tokenization_path_logprob_std": 286.753346620935, "trajectory_balance_rmse": 3825.2117095816548} +{"canonical_length": 170, "canonical_nll": 0.7290286793428309, "generation_rouge_l_f1": 0.33816425120772947, "generation_token_f1": 0.5120772946859904, "prompt_id": "d92a9462a284aee7", "reading_canonical_log_prob": -123.93487548828125, "reading_distances": [12, 16, 13, 14], "reading_log_probs": [-125.35206604003906, -123.79472351074219, -125.95323181152344, -122.92719268798828], "reading_logprob_std": 1.3915042041493033, "tokenization_path_logprob_std": 96.26252221449143, "trajectory_balance_rmse": 3459.004739250435} +{"canonical_length": 76, "canonical_nll": 0.9614550941868832, "generation_rouge_l_f1": 0.35135135135135137, "generation_token_f1": 0.5045045045045045, "prompt_id": "32c20ba8cc10f798", "reading_canonical_log_prob": -73.07058715820312, "reading_distances": [33, 33, 28, 28], "reading_log_probs": [-71.30232238769531, -71.79497528076172, -71.76688385009766, -71.37687683105469], "reading_logprob_std": 0.2568695368352856, "tokenization_path_logprob_std": 121.5498601939218, "trajectory_balance_rmse": 1906.4385456949794} +{"canonical_length": 21, "canonical_nll": 0.5854852767217726, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "2e575375b1e838f0", "reading_canonical_log_prob": -12.295190811157227, "reading_distances": [25, 23, 27, 29], "reading_log_probs": [-17.585758209228516, -12.890462875366211, -12.649206161499023, -12.143698692321777], "reading_logprob_std": 2.5315139660906567, "tokenization_path_logprob_std": 81.37096018846123, "trajectory_balance_rmse": 243.87509515432637} +{"canonical_length": 9, "canonical_nll": 0.010631465249591403, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "f19b5cf2d17aa856", "reading_canonical_log_prob": -0.09568318724632263, "reading_distances": [37, 36, 30, 42], "reading_log_probs": [-0.32272446155548096, -0.7885379195213318, -0.5738234519958496, -0.4094484746456146], "reading_logprob_std": 0.20501710864148112, "tokenization_path_logprob_std": 18.330171435930218, "trajectory_balance_rmse": 197.46871110817747} +{"canonical_length": 3, "canonical_nll": 1.6515963872273762, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "745badf44c39e597", "reading_canonical_log_prob": -4.954789161682129, "reading_distances": [24, 26, 19, 20], "reading_log_probs": [-6.654106140136719, -5.4945068359375, -5.051887512207031, -5.487052917480469], "reading_logprob_std": 0.6867268172936576, "tokenization_path_logprob_std": 6.205847343488223, "trajectory_balance_rmse": 40.00158434934426} +{"canonical_length": 272, "canonical_nll": 1.3650516061221851, "generation_rouge_l_f1": 0.297029702970297, "generation_token_f1": 0.4871287128712872, "prompt_id": "8f529e5554e9c902", "reading_canonical_log_prob": -371.2940368652344, "reading_distances": [21, 22, 18, 19], "reading_log_probs": [-368.1552734375, -369.40155029296875, -372.83380126953125, -369.18048095703125], "reading_logprob_std": 2.0344705874994173, "tokenization_path_logprob_std": 81.74539164618379, "trajectory_balance_rmse": 5617.670405352796} +{"canonical_length": 324, "canonical_nll": 0.9554629855685763, "generation_rouge_l_f1": 0.20107719928186715, "generation_token_f1": 0.42728904847396765, "prompt_id": "642f20171ea337c8", "reading_canonical_log_prob": -309.57000732421875, "reading_distances": [26, 21, 25, 27], "reading_log_probs": [-311.1183776855469, -314.0841369628906, -310.12408447265625, -310.35394287109375], "reading_logprob_std": 1.8261504648010913, "tokenization_path_logprob_std": 195.82577571166365, "trajectory_balance_rmse": 5205.7452946885105} +{"canonical_length": 14, "canonical_nll": 2.082329886300223, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4210526315789474, "prompt_id": "c076cc444fb03f81", "reading_canonical_log_prob": -29.152618408203125, "reading_distances": [196, 181, 192, 186], "reading_log_probs": [-30.1695499420166, -29.09292984008789, -30.316797256469727, -27.54403305053711], "reading_logprob_std": 1.2799507028982857, "tokenization_path_logprob_std": 34.605866061129376, "trajectory_balance_rmse": 424.8931065658997} +{"canonical_length": 183, "canonical_nll": 1.23947018482646, "generation_rouge_l_f1": 0.17801047120418848, "generation_token_f1": 0.27225130890052357, "prompt_id": "8082ddc02c38399e", "reading_canonical_log_prob": -226.8230438232422, "reading_distances": [18, 26, 25, 25], "reading_log_probs": [-226.12716674804688, -225.10009765625, -225.49285888671875, -226.23350524902344], "reading_logprob_std": 0.5366529332017134, "tokenization_path_logprob_std": 142.54453511444336, "trajectory_balance_rmse": 1644.4121817117743} +{"canonical_length": 7, "canonical_nll": 0.4345573697771345, "generation_rouge_l_f1": 0.08928571428571429, "generation_token_f1": 0.08928571428571429, "prompt_id": "3af881f54016c0d1", "reading_canonical_log_prob": -3.0419015884399414, "reading_distances": [35, 38, 36, 40], "reading_log_probs": [-4.334273338317871, -3.3594868183135986, -4.205879211425781, -3.8208370208740234], "reading_logprob_std": 0.43853850500823854, "tokenization_path_logprob_std": 13.597560683626543, "trajectory_balance_rmse": 80.45892883753412} +{"canonical_length": 241, "canonical_nll": 0.6485107548503954, "generation_rouge_l_f1": 0.31434184675834975, "generation_token_f1": 0.5383104125736738, "prompt_id": "edd1c37dcb8ffb27", "reading_canonical_log_prob": -156.2910919189453, "reading_distances": [14, 14, 15, 16], "reading_log_probs": [-158.40673828125, -157.87454223632812, -160.27784729003906, -157.4423370361328], "reading_logprob_std": 1.2489043942212426, "tokenization_path_logprob_std": 90.3635099337731, "trajectory_balance_rmse": 3908.835991743039} +{"canonical_length": 277, "canonical_nll": 1.6466848876071751, "generation_rouge_l_f1": 0.2025782688766114, "generation_token_f1": 0.3683241252302026, "prompt_id": "b7a747a09567587c", "reading_canonical_log_prob": -456.1317138671875, "reading_distances": [36, 29, 37, 27], "reading_log_probs": [-454.45477294921875, -453.3538513183594, -454.2879638671875, -454.71063232421875], "reading_logprob_std": 0.5914239387011723, "tokenization_path_logprob_std": 328.4559723792178, "trajectory_balance_rmse": 5915.6762038763745} +{"canonical_length": 203, "canonical_nll": 1.4590550314616688, "generation_rouge_l_f1": 0.21739130434782608, "generation_token_f1": 0.48260869565217385, "prompt_id": "85ba85bd1384cec6", "reading_canonical_log_prob": -296.18817138671875, "reading_distances": [19, 22, 16, 16], "reading_log_probs": [-289.6205749511719, -289.3431701660156, -288.9600524902344, -290.1170654296875], "reading_logprob_std": 0.48683656619195964, "tokenization_path_logprob_std": 188.35651901318224, "trajectory_balance_rmse": 4752.04735867697} +{"canonical_length": 25, "canonical_nll": 0.44203819274902345, "generation_rouge_l_f1": 0.7272727272727272, "generation_token_f1": 0.9090909090909091, "prompt_id": "ff4cb1405a9ba9bb", "reading_canonical_log_prob": -11.050954818725586, "reading_distances": [16, 15, 16, 22], "reading_log_probs": [-10.951650619506836, -10.912254333496094, -11.044904708862305, -11.547987937927246], "reading_logprob_std": 0.294492878115362, "tokenization_path_logprob_std": 78.31343977485116, "trajectory_balance_rmse": 350.18528206752234} +{"canonical_length": 103, "canonical_nll": 1.5199500519095115, "generation_rouge_l_f1": 0.253968253968254, "generation_token_f1": 0.39153439153439157, "prompt_id": "ea60f11b47ae285e", "reading_canonical_log_prob": -156.5548553466797, "reading_distances": [22, 29, 23, 29], "reading_log_probs": [-154.93280029296875, -153.7489776611328, -154.22402954101562, -153.72427368164062], "reading_logprob_std": 0.5657126086422618, "tokenization_path_logprob_std": 179.1861748388994, "trajectory_balance_rmse": 2643.977211328605} +{"canonical_length": 22, "canonical_nll": 0.3828803842717951, "generation_rouge_l_f1": 0.12592592592592594, "generation_token_f1": 0.12592592592592594, "prompt_id": "6245bcac7beebc0c", "reading_canonical_log_prob": -8.423368453979492, "reading_distances": [25, 30, 21, 25], "reading_log_probs": [-7.887392997741699, -7.865744590759277, -7.538455963134766, -7.747463226318359], "reading_logprob_std": 0.1598432595767863, "tokenization_path_logprob_std": 67.97765491454734, "trajectory_balance_rmse": 510.41562288658594} +{"canonical_length": 156, "canonical_nll": 1.433299529246795, "generation_rouge_l_f1": 0.18894009216589863, "generation_token_f1": 0.36405529953917054, "prompt_id": "5016da9f4e182301", "reading_canonical_log_prob": -223.5947265625, "reading_distances": [14, 16, 20, 18], "reading_log_probs": [-224.471923828125, -224.19796752929688, -225.0926513671875, -226.146484375], "reading_logprob_std": 0.8646891942915668, "tokenization_path_logprob_std": 144.1146380340873, "trajectory_balance_rmse": 3936.699750885918} +{"canonical_length": 103, "canonical_nll": 1.7304785274764867, "generation_rouge_l_f1": 0.29850746268656714, "generation_token_f1": 0.4477611940298507, "prompt_id": "d6b153dd541c60bc", "reading_canonical_log_prob": -178.23928833007812, "reading_distances": [30, 33, 31, 34], "reading_log_probs": [-180.77767944335938, -181.56539916992188, -181.31655883789062, -182.31597900390625], "reading_logprob_std": 0.6391003555593407, "tokenization_path_logprob_std": 134.86342483182773, "trajectory_balance_rmse": 1823.255579430018} +{"canonical_length": 50, "canonical_nll": 1.1168247222900392, "generation_rouge_l_f1": 0.16194331983805668, "generation_token_f1": 0.24291497975708504, "prompt_id": "f50b16ccb1ed30ed", "reading_canonical_log_prob": -55.84123611450195, "reading_distances": [14, 17, 13, 16], "reading_log_probs": [-63.83704376220703, -60.532588958740234, -59.02292251586914, -60.678951263427734], "reading_logprob_std": 2.0230274149392153, "tokenization_path_logprob_std": 58.96816925449307, "trajectory_balance_rmse": 1026.1185091250977} +{"canonical_length": 26, "canonical_nll": 0.5076295045705942, "generation_rouge_l_f1": 0.37254901960784315, "generation_token_f1": 0.39215686274509803, "prompt_id": "ac38f5f5447a0903", "reading_canonical_log_prob": -13.19836711883545, "reading_distances": [35, 43, 36, 36], "reading_log_probs": [-12.026491165161133, -12.651592254638672, -12.437027931213379, -11.62043571472168], "reading_logprob_std": 0.45646443307169743, "tokenization_path_logprob_std": 108.63197827644699, "trajectory_balance_rmse": 624.9540256647908} +{"canonical_length": 10, "canonical_nll": 1.7748414993286132, "generation_rouge_l_f1": 0.05204460966542751, "generation_token_f1": 0.05947955390334573, "prompt_id": "0027f65734731d27", "reading_canonical_log_prob": -17.748414993286133, "reading_distances": [25, 23, 18, 21], "reading_log_probs": [-24.05259895324707, -19.990962982177734, -20.306215286254883, -23.282058715820312], "reading_logprob_std": 2.059780507315402, "tokenization_path_logprob_std": 35.08084145294178, "trajectory_balance_rmse": 285.8845423676065} +{"canonical_length": 146, "canonical_nll": 1.4338426981886772, "generation_rouge_l_f1": 0.19499999999999998, "generation_token_f1": 0.26499999999999996, "prompt_id": "488c8f3398b00ac7", "reading_canonical_log_prob": -209.34103393554688, "reading_distances": [15, 14, 19, 12], "reading_log_probs": [-207.3146209716797, -206.66421508789062, -206.87869262695312, -207.6714630126953], "reading_logprob_std": 0.4499473542292792, "tokenization_path_logprob_std": 150.01432847490915, "trajectory_balance_rmse": 2547.1258286285574} +{"canonical_length": 4, "canonical_nll": 5.3622541427612305, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "91adc07e055d938f", "reading_canonical_log_prob": -21.449016571044922, "reading_distances": [31, 31, 34, 30], "reading_log_probs": [-20.437969207763672, -20.571308135986328, -21.88838005065918, -21.538175582885742], "reading_logprob_std": 0.7143808046349154, "tokenization_path_logprob_std": 6.158729140842383, "trajectory_balance_rmse": 38.17789764055351} +{"canonical_length": 263, "canonical_nll": 1.240886267600404, "generation_rouge_l_f1": 0.2674418604651163, "generation_token_f1": 0.45736434108527135, "prompt_id": "5f3070e7c5d93f6b", "reading_canonical_log_prob": -326.35308837890625, "reading_distances": [19, 21, 16, 20], "reading_log_probs": [-327.321044921875, -327.9415283203125, -326.87872314453125, -329.659423828125], "reading_logprob_std": 1.2200300618158473, "tokenization_path_logprob_std": 214.05246257425065, "trajectory_balance_rmse": 5969.792346701565} +{"canonical_length": 17, "canonical_nll": 0.2576001672183766, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.2857142857142857, "prompt_id": "f969cf0286bd838d", "reading_canonical_log_prob": -4.379202842712402, "reading_distances": [18, 22, 21, 15], "reading_log_probs": [-4.6784586906433105, -4.06069278717041, -4.00302267074585, -4.121233940124512], "reading_logprob_std": 0.312158203586592, "tokenization_path_logprob_std": 41.10630111638969, "trajectory_balance_rmse": 263.7083517663603} +{"canonical_length": 291, "canonical_nll": 0.6721364444064111, "generation_rouge_l_f1": 0.19459459459459458, "generation_token_f1": 0.31891891891891894, "prompt_id": "60292acfc5a172b6", "reading_canonical_log_prob": -195.59170532226562, "reading_distances": [25, 36, 33, 23], "reading_log_probs": [-195.52880859375, -195.31930541992188, -196.25357055664062, -196.94363403320312], "reading_logprob_std": 0.7392780768041366, "tokenization_path_logprob_std": 230.94897205901438, "trajectory_balance_rmse": 5125.595009780744} +{"canonical_length": 174, "canonical_nll": 0.8244271113954741, "generation_rouge_l_f1": 0.27011494252873564, "generation_token_f1": 0.44252873563218387, "prompt_id": "2b8f8cb7903f03af", "reading_canonical_log_prob": -143.4503173828125, "reading_distances": [37, 33, 41, 42], "reading_log_probs": [-152.99575805664062, -151.22799682617188, -153.99996948242188, -146.57293701171875], "reading_logprob_std": 3.290140272272115, "tokenization_path_logprob_std": 321.54939189996463, "trajectory_balance_rmse": 3607.9619642965813} +{"canonical_length": 71, "canonical_nll": 1.1934967041015625, "generation_rouge_l_f1": 0.3473053892215569, "generation_token_f1": 0.45508982035928147, "prompt_id": "2a8ed34f73d4ba31", "reading_canonical_log_prob": -84.73826599121094, "reading_distances": [19, 14, 21, 19], "reading_log_probs": [-84.48261260986328, -84.8900146484375, -83.84606170654297, -84.60585021972656], "reading_logprob_std": 0.4410429622154877, "tokenization_path_logprob_std": 22.570385481192762, "trajectory_balance_rmse": 1235.4178168044537} +{"canonical_length": 370, "canonical_nll": 0.8425026558540963, "generation_rouge_l_f1": 0.4613003095975233, "generation_token_f1": 0.5696594427244581, "prompt_id": "a2f2bd1ddbd7ac6c", "reading_canonical_log_prob": -311.7259826660156, "reading_distances": [22, 19, 21, 22], "reading_log_probs": [-313.7330322265625, -312.53631591796875, -313.03948974609375, -311.5137939453125], "reading_logprob_std": 0.9338376200224707, "tokenization_path_logprob_std": 185.57399185415932, "trajectory_balance_rmse": 7604.5587879466275} +{"canonical_length": 292, "canonical_nll": 1.3785716017631635, "generation_rouge_l_f1": 0.20377358490566036, "generation_token_f1": 0.3773584905660377, "prompt_id": "439bf3be7d027444", "reading_canonical_log_prob": -402.54290771484375, "reading_distances": [22, 23, 22, 22], "reading_log_probs": [-402.765380859375, -399.7956237792969, -401.14984130859375, -401.8976745605469], "reading_logprob_std": 1.2581109037022562, "tokenization_path_logprob_std": 137.41899066903164, "trajectory_balance_rmse": 5712.100181297937} +{"canonical_length": 152, "canonical_nll": 1.2990734702662419, "generation_rouge_l_f1": 0.18233618233618232, "generation_token_f1": 0.2564102564102564, "prompt_id": "53583cdc2f7841b7", "reading_canonical_log_prob": -197.45916748046875, "reading_distances": [23, 21, 25, 25], "reading_log_probs": [-193.81729125976562, -195.27386474609375, -192.850341796875, -193.42388916015625], "reading_logprob_std": 1.0342546643484065, "tokenization_path_logprob_std": 123.65935197296662, "trajectory_balance_rmse": 2805.992385798591} +{"canonical_length": 73, "canonical_nll": 1.1256000048493686, "generation_rouge_l_f1": 0.1415929203539823, "generation_token_f1": 0.24778761061946902, "prompt_id": "11b8b10c9b5e1733", "reading_canonical_log_prob": -82.1688003540039, "reading_distances": [17, 16, 12, 20], "reading_log_probs": [-81.25025177001953, -80.6050796508789, -81.18824768066406, -80.52311706542969], "reading_logprob_std": 0.3805717191204181, "tokenization_path_logprob_std": 152.28529699067175, "trajectory_balance_rmse": 1777.7311653827928} +{"canonical_length": 17, "canonical_nll": 0.8579579521627987, "generation_rouge_l_f1": 0.07929515418502203, "generation_token_f1": 0.0881057268722467, "prompt_id": "d62e67048e61fa64", "reading_canonical_log_prob": -14.585285186767578, "reading_distances": [15, 9, 17, 13], "reading_log_probs": [-15.234295845031738, -13.995515823364258, -16.27100372314453, -14.540632247924805], "reading_logprob_std": 0.9814823352630597, "tokenization_path_logprob_std": 70.24928246649011, "trajectory_balance_rmse": 284.3962742249203} +{"canonical_length": 17, "canonical_nll": 1.2760623483096851, "generation_rouge_l_f1": 0.08118081180811808, "generation_token_f1": 0.08118081180811808, "prompt_id": "15d2843edae95663", "reading_canonical_log_prob": -21.69305992126465, "reading_distances": [21, 22, 23, 26], "reading_log_probs": [-25.15290069580078, -24.44202995300293, -22.18484878540039, -25.4437313079834], "reading_logprob_std": 1.4752920480571, "tokenization_path_logprob_std": 30.054672087243812, "trajectory_balance_rmse": 340.01554558815803} +{"canonical_length": 243, "canonical_nll": 1.3208336849762088, "generation_rouge_l_f1": 0.18585858585858586, "generation_token_f1": 0.4121212121212121, "prompt_id": "960951e022fd8e15", "reading_canonical_log_prob": -320.96258544921875, "reading_distances": [33, 34, 33, 30], "reading_log_probs": [-325.9946594238281, -327.7749938964844, -326.5867919921875, -326.4974365234375], "reading_logprob_std": 0.7541528156110215, "tokenization_path_logprob_std": 250.59911753517846, "trajectory_balance_rmse": 6031.801449076308} +{"canonical_length": 192, "canonical_nll": 1.2443981170654297, "generation_rouge_l_f1": 0.28, "generation_token_f1": 0.5700000000000001, "prompt_id": "e17ce1bfa59fa95a", "reading_canonical_log_prob": -238.9244384765625, "reading_distances": [18, 25, 15, 25], "reading_log_probs": [-239.57376098632812, -241.35067749023438, -238.44287109375, -240.56259155273438], "reading_logprob_std": 1.2577679439058802, "tokenization_path_logprob_std": 152.99983588761262, "trajectory_balance_rmse": 4362.764998010211} +{"canonical_length": 37, "canonical_nll": 1.3023106343037374, "generation_rouge_l_f1": 0.11221122112211221, "generation_token_f1": 0.132013201320132, "prompt_id": "305cbe322d895404", "reading_canonical_log_prob": -48.18549346923828, "reading_distances": [13, 24, 14, 13], "reading_log_probs": [-48.02460479736328, -47.772396087646484, -48.52830505371094, -48.04563903808594], "reading_logprob_std": 0.3158048155592952, "tokenization_path_logprob_std": 28.315078906640377, "trajectory_balance_rmse": 577.0652464565643} +{"canonical_length": 25, "canonical_nll": 1.188307647705078, "generation_rouge_l_f1": 0.12307692307692307, "generation_token_f1": 0.14615384615384613, "prompt_id": "fee5a1b64fc319ac", "reading_canonical_log_prob": -29.707691192626953, "reading_distances": [27, 26, 24, 24], "reading_log_probs": [-32.68286895751953, -32.516632080078125, -32.01921844482422, -31.713605880737305], "reading_logprob_std": 0.4465803450600664, "tokenization_path_logprob_std": 44.99123540438733, "trajectory_balance_rmse": 650.3255454321012} +{"canonical_length": 12, "canonical_nll": 1.533490816752116, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "dce2143d59fcc65f", "reading_canonical_log_prob": -18.40188980102539, "reading_distances": [11, 11, 13, 20], "reading_log_probs": [-18.061161041259766, -18.00218963623047, -18.225799560546875, -19.209402084350586], "reading_logprob_std": 0.564496691728161, "tokenization_path_logprob_std": 36.38516115028749, "trajectory_balance_rmse": 258.0313556931421} +{"canonical_length": 30, "canonical_nll": 0.3739190419514974, "generation_rouge_l_f1": 0.12709030100334448, "generation_token_f1": 0.14715719063545152, "prompt_id": "bc27bf2b83eeb772", "reading_canonical_log_prob": -11.217571258544922, "reading_distances": [35, 42, 39, 37], "reading_log_probs": [-11.558137893676758, -12.109164237976074, -12.200273513793945, -11.714008331298828], "reading_logprob_std": 0.3083783793881353, "tokenization_path_logprob_std": 43.15902560622191, "trajectory_balance_rmse": 393.965252186} +{"canonical_length": 171, "canonical_nll": 1.9631740280062135, "generation_rouge_l_f1": 0.24154589371980675, "generation_token_f1": 0.38164251207729466, "prompt_id": "8e8e4001993842cc", "reading_canonical_log_prob": -335.7027587890625, "reading_distances": [23, 25, 30, 23], "reading_log_probs": [-335.6540832519531, -335.08355712890625, -334.685546875, -335.635009765625], "reading_logprob_std": 0.46796710371150846, "tokenization_path_logprob_std": 175.4800835711998, "trajectory_balance_rmse": 2803.881864530268} +{"canonical_length": 93, "canonical_nll": 1.8711229549941195, "generation_rouge_l_f1": 0.21804511278195488, "generation_token_f1": 0.3157894736842105, "prompt_id": "ae4b947d6478d1ef", "reading_canonical_log_prob": -174.01443481445312, "reading_distances": [21, 24, 18, 27], "reading_log_probs": [-175.8039093017578, -175.31680297851562, -173.75454711914062, -175.688720703125], "reading_logprob_std": 0.9473832362509593, "tokenization_path_logprob_std": 61.65251016303319, "trajectory_balance_rmse": 2067.4176633467087} +{"canonical_length": 153, "canonical_nll": 2.1074817133884802, "generation_rouge_l_f1": 0.3014705882352941, "generation_token_f1": 0.4632352941176471, "prompt_id": "e8d78f3e32f3fc0a", "reading_canonical_log_prob": -322.4447021484375, "reading_distances": [18, 20, 23, 24], "reading_log_probs": [-323.1842041015625, -323.8085021972656, -325.4846496582031, -323.4396667480469], "reading_logprob_std": 1.035797761685556, "tokenization_path_logprob_std": 120.12188911933445, "trajectory_balance_rmse": 4142.971321367772} +{"canonical_length": 14, "canonical_nll": 1.4638808114188058, "generation_rouge_l_f1": 0.5333333333333333, "generation_token_f1": 0.5333333333333333, "prompt_id": "0652d72e6ba6a111", "reading_canonical_log_prob": -20.49433135986328, "reading_distances": [35, 27, 25, 28], "reading_log_probs": [-20.513986587524414, -20.658288955688477, -20.492277145385742, -20.69746208190918], "reading_logprob_std": 0.10253172010386365, "tokenization_path_logprob_std": 29.954688250919062, "trajectory_balance_rmse": 202.91751851166225} +{"canonical_length": 35, "canonical_nll": 1.5690056937081474, "generation_rouge_l_f1": 0.047619047619047616, "generation_token_f1": 0.047619047619047616, "prompt_id": "abb413900ce0e656", "reading_canonical_log_prob": -54.915199279785156, "reading_distances": [28, 24, 23, 27], "reading_log_probs": [-54.848812103271484, -55.259368896484375, -54.80958557128906, -54.5621337890625], "reading_logprob_std": 0.2889570723992829, "tokenization_path_logprob_std": 93.4486103360475, "trajectory_balance_rmse": 923.8248935137954} +{"canonical_length": 294, "canonical_nll": 0.8823578503667092, "generation_rouge_l_f1": 0.14099216710182766, "generation_token_f1": 0.26631853785900783, "prompt_id": "7607920df7068b2c", "reading_canonical_log_prob": -259.4132080078125, "reading_distances": [14, 8, 13, 9], "reading_log_probs": [-256.30792236328125, -258.3722229003906, -257.60601806640625, -258.38336181640625], "reading_logprob_std": 0.9766138487270071, "tokenization_path_logprob_std": 161.50298870664258, "trajectory_balance_rmse": 5633.190396194588} +{"canonical_length": 50, "canonical_nll": 2.599837646484375, "generation_rouge_l_f1": 0.19672131147540983, "generation_token_f1": 0.32786885245901637, "prompt_id": "6e3d23fca9ec801f", "reading_canonical_log_prob": -129.99188232421875, "reading_distances": [28, 32, 27, 24], "reading_log_probs": [-131.36383056640625, -129.33868408203125, -128.61549377441406, -130.34896850585938], "reading_logprob_std": 1.1983715301035383, "tokenization_path_logprob_std": 111.83052585798062, "trajectory_balance_rmse": 1326.7767070502368} +{"canonical_length": 18, "canonical_nll": 1.637648688422309, "generation_rouge_l_f1": 0.5384615384615384, "generation_token_f1": 0.5384615384615384, "prompt_id": "4afc9f7951fa42e2", "reading_canonical_log_prob": -29.477676391601562, "reading_distances": [18, 20, 24, 21], "reading_log_probs": [-29.17437744140625, -29.987939834594727, -28.913040161132812, -29.205097198486328], "reading_logprob_std": 0.4641007307266141, "tokenization_path_logprob_std": 41.10900721813325, "trajectory_balance_rmse": 410.4248413543868} +{"canonical_length": 501, "canonical_nll": 0.6185499484429579, "generation_rouge_l_f1": 0.22310756972111553, "generation_token_f1": 0.297476759628154, "prompt_id": "031d7ff67f65087b", "reading_canonical_log_prob": -309.8935241699219, "reading_distances": [34, 27, 22, 24], "reading_log_probs": [-309.7279357910156, -310.5274658203125, -311.716796875, -309.5667419433594], "reading_logprob_std": 0.9823880373067664, "tokenization_path_logprob_std": 196.57194463946954, "trajectory_balance_rmse": 6185.568934306981} +{"canonical_length": 128, "canonical_nll": 1.0863282680511475, "generation_rouge_l_f1": 0.24880382775119617, "generation_token_f1": 0.45933014354066987, "prompt_id": "81ff4f6d7865cbd6", "reading_canonical_log_prob": -139.05001831054688, "reading_distances": [20, 19, 18, 19], "reading_log_probs": [-149.01220703125, -150.3588409423828, -149.85733032226562, -151.87490844726562], "reading_logprob_std": 1.2021962796211163, "tokenization_path_logprob_std": 124.28737114122694, "trajectory_balance_rmse": 2418.487812118412} +{"canonical_length": 112, "canonical_nll": 1.3205968311854772, "generation_rouge_l_f1": 0.2808219178082192, "generation_token_f1": 0.3835616438356164, "prompt_id": "fa3e1b2e7d03f031", "reading_canonical_log_prob": -147.90684509277344, "reading_distances": [26, 29, 30, 28], "reading_log_probs": [-146.63746643066406, -147.04835510253906, -147.29269409179688, -145.09759521484375], "reading_logprob_std": 0.9854347987315943, "tokenization_path_logprob_std": 167.14031784905416, "trajectory_balance_rmse": 2827.448211802364} +{"canonical_length": 136, "canonical_nll": 1.2027107687557446, "generation_rouge_l_f1": 0.2780748663101604, "generation_token_f1": 0.4598930481283422, "prompt_id": "c4484ce7220f53d7", "reading_canonical_log_prob": -163.56866455078125, "reading_distances": [19, 15, 13, 15], "reading_log_probs": [-162.58987426757812, -163.9212646484375, -164.05673217773438, -163.27389526367188], "reading_logprob_std": 0.6734506542040553, "tokenization_path_logprob_std": 108.59293567178334, "trajectory_balance_rmse": 2909.3494715458205} +{"canonical_length": 5, "canonical_nll": 0.3910529375076294, "generation_rouge_l_f1": 0.6666666666666665, "generation_token_f1": 0.6666666666666665, "prompt_id": "e265516b7f250f1e", "reading_canonical_log_prob": -1.955264687538147, "reading_distances": [36, 36, 40, 41], "reading_log_probs": [-3.8916866779327393, -2.7628588676452637, -2.977513551712036, -3.321455240249634], "reading_logprob_std": 0.4925724184064505, "tokenization_path_logprob_std": 38.61655904283832, "trajectory_balance_rmse": 118.88774735949248} +{"canonical_length": 20, "canonical_nll": 0.541668176651001, "generation_rouge_l_f1": 0.1037037037037037, "generation_token_f1": 0.11111111111111112, "prompt_id": "d3f0dad3a37f19c3", "reading_canonical_log_prob": -10.83336353302002, "reading_distances": [16, 14, 16, 17], "reading_log_probs": [-10.573593139648438, -10.544771194458008, -10.105900764465332, -10.522040367126465], "reading_logprob_std": 0.22145741961299903, "tokenization_path_logprob_std": 47.86674506349162, "trajectory_balance_rmse": 516.7429518013662} +{"canonical_length": 23, "canonical_nll": 1.761057480521824, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.35714285714285715, "prompt_id": "448ac2a15956c368", "reading_canonical_log_prob": -40.50432205200195, "reading_distances": [29, 35, 41, 31], "reading_log_probs": [-37.899024963378906, -38.58477783203125, -38.388710021972656, -37.569190979003906], "reading_logprob_std": 0.46190814515218126, "tokenization_path_logprob_std": 26.22658254959572, "trajectory_balance_rmse": 482.1349147332577} +{"canonical_length": 165, "canonical_nll": 1.4513700543027936, "generation_rouge_l_f1": 0.316622691292876, "generation_token_f1": 0.45382585751978893, "prompt_id": "e00cf91341eb5691", "reading_canonical_log_prob": -239.47605895996094, "reading_distances": [14, 14, 9, 17], "reading_log_probs": [-239.21163940429688, -238.8033447265625, -240.65365600585938, -240.10577392578125], "reading_logprob_std": 0.8399276451228669, "tokenization_path_logprob_std": 67.23494284981997, "trajectory_balance_rmse": 4093.142505732039} +{"canonical_length": 224, "canonical_nll": 1.0083773476736886, "generation_rouge_l_f1": 0.27755102040816326, "generation_token_f1": 0.5142857142857142, "prompt_id": "bf8645b4ff7492c2", "reading_canonical_log_prob": -225.87652587890625, "reading_distances": [52, 55, 60, 48], "reading_log_probs": [-227.53492736816406, -228.29971313476562, -228.6416015625, -227.29220581054688], "reading_logprob_std": 0.6338616040127765, "tokenization_path_logprob_std": 199.01129128101613, "trajectory_balance_rmse": 4846.362425297266} +{"canonical_length": 261, "canonical_nll": 1.2985007333572798, "generation_rouge_l_f1": 0.2091254752851711, "generation_token_f1": 0.4448669201520913, "prompt_id": "93af13c34520a089", "reading_canonical_log_prob": -338.90869140625, "reading_distances": [25, 28, 26, 22], "reading_log_probs": [-341.52294921875, -341.6876220703125, -341.5359191894531, -340.5789794921875], "reading_logprob_std": 0.5071324553427629, "tokenization_path_logprob_std": 266.77904545841034, "trajectory_balance_rmse": 6067.597692012468} +{"canonical_length": 107, "canonical_nll": 1.1187503850348641, "generation_rouge_l_f1": 0.31607629427792916, "generation_token_f1": 0.38692098092643046, "prompt_id": "8f7942ab9e010ef0", "reading_canonical_log_prob": -119.70629119873047, "reading_distances": [34, 31, 34, 35], "reading_log_probs": [-120.6055679321289, -120.97044372558594, -120.38844299316406, -120.9593505859375], "reading_logprob_std": 0.28434473497271706, "tokenization_path_logprob_std": 115.13272310985046, "trajectory_balance_rmse": 2029.781683928591} +{"canonical_length": 39, "canonical_nll": 2.4510079408303285, "generation_rouge_l_f1": 0.09302325581395349, "generation_token_f1": 0.09302325581395349, "prompt_id": "ba9caab6c5bf7b20", "reading_canonical_log_prob": -95.58930969238281, "reading_distances": [38, 37, 40, 36], "reading_log_probs": [-104.21434020996094, -100.14240264892578, -103.38189697265625, -102.43533325195312], "reading_logprob_std": 1.757989858286922, "tokenization_path_logprob_std": 12.38445617619006, "trajectory_balance_rmse": 782.9085007147586} +{"canonical_length": 292, "canonical_nll": 1.4346288393621576, "generation_rouge_l_f1": 0.18996415770609318, "generation_token_f1": 0.36200716845878134, "prompt_id": "90e4d685d3873c08", "reading_canonical_log_prob": -418.91162109375, "reading_distances": [30, 30, 22, 20], "reading_log_probs": [-422.0000305175781, -421.7912292480469, -422.8392028808594, -421.7321472167969], "reading_logprob_std": 0.512093851514713, "tokenization_path_logprob_std": 259.1461944887428, "trajectory_balance_rmse": 6848.303190768975} +{"canonical_length": 15, "canonical_nll": 1.9964890797932944, "generation_rouge_l_f1": 0.08, "generation_token_f1": 0.08, "prompt_id": "ec455005996bdab4", "reading_canonical_log_prob": -29.947336196899414, "reading_distances": [42, 34, 48, 44], "reading_log_probs": [-28.853214263916016, -29.117687225341797, -29.645885467529297, -28.87251853942871], "reading_logprob_std": 0.36921580325718406, "tokenization_path_logprob_std": 68.60768656848099, "trajectory_balance_rmse": 289.48306050418245} +{"canonical_length": 243, "canonical_nll": 1.5950321150414737, "generation_rouge_l_f1": 0.17333333333333334, "generation_token_f1": 0.39555555555555555, "prompt_id": "ab90737e4575d4fd", "reading_canonical_log_prob": -387.5928039550781, "reading_distances": [11, 17, 14, 17], "reading_log_probs": [-387.8646240234375, -395.76763916015625, -391.7403564453125, -391.40167236328125], "reading_logprob_std": 3.2324531455702714, "tokenization_path_logprob_std": 243.99828936590018, "trajectory_balance_rmse": 5412.865661735745} +{"canonical_length": 240, "canonical_nll": 1.440386962890625, "generation_rouge_l_f1": 0.30434782608695654, "generation_token_f1": 0.541501976284585, "prompt_id": "ec5a98de132967a0", "reading_canonical_log_prob": -345.69287109375, "reading_distances": [33, 32, 29, 37], "reading_log_probs": [-346.3058776855469, -341.6698913574219, -341.14910888671875, -349.4963684082031], "reading_logprob_std": 3.9735119191307913, "tokenization_path_logprob_std": 247.51814995898533, "trajectory_balance_rmse": 5400.586021411866} +{"canonical_length": 37, "canonical_nll": 0.9792442321777344, "generation_rouge_l_f1": 0.14634146341463417, "generation_token_f1": 0.24390243902439024, "prompt_id": "71251a192c44eeb3", "reading_canonical_log_prob": -36.23203659057617, "reading_distances": [22, 24, 25, 22], "reading_log_probs": [-37.40028762817383, -41.32695770263672, -41.451744079589844, -39.98609924316406], "reading_logprob_std": 1.8815119539520888, "tokenization_path_logprob_std": 94.02654328727989, "trajectory_balance_rmse": 740.6008812234247} +{"canonical_length": 221, "canonical_nll": 1.7795073220093325, "generation_rouge_l_f1": 0.27571115973741794, "generation_token_f1": 0.48140043763676144, "prompt_id": "646766dc50567beb", "reading_canonical_log_prob": -393.2711181640625, "reading_distances": [44, 41, 39, 54], "reading_log_probs": [-393.35284423828125, -389.6094970703125, -392.781005859375, -394.4051513671875], "reading_logprob_std": 2.06441303622968, "tokenization_path_logprob_std": 192.92620590537982, "trajectory_balance_rmse": 4241.231169484972} +{"canonical_length": 87, "canonical_nll": 1.163714704842403, "generation_rouge_l_f1": 0.08602150537634409, "generation_token_f1": 0.12903225806451613, "prompt_id": "3175f6135d7796a6", "reading_canonical_log_prob": -101.24317932128906, "reading_distances": [40, 40, 42, 47], "reading_log_probs": [-103.68601989746094, -105.00933074951172, -102.43023681640625, -105.08615112304688], "reading_logprob_std": 1.258306331822966, "tokenization_path_logprob_std": 50.40193984984103, "trajectory_balance_rmse": 1584.0167471765453} +{"canonical_length": 315, "canonical_nll": 1.0623821924603174, "generation_rouge_l_f1": 0.32638888888888884, "generation_token_f1": 0.545138888888889, "prompt_id": "364270b26598f5f6", "reading_canonical_log_prob": -334.650390625, "reading_distances": [25, 17, 25, 26], "reading_log_probs": [-335.00079345703125, -333.30023193359375, -333.60003662109375, -333.4239196777344], "reading_logprob_std": 0.7893435463879429, "tokenization_path_logprob_std": 248.0346665633904, "trajectory_balance_rmse": 5778.5866919417} +{"canonical_length": 132, "canonical_nll": 1.6210275129838423, "generation_rouge_l_f1": 0.1935483870967742, "generation_token_f1": 0.3771712158808933, "prompt_id": "3b49b131c410b62c", "reading_canonical_log_prob": -213.9756317138672, "reading_distances": [18, 13, 10, 23], "reading_log_probs": [-212.04104614257812, -213.07533264160156, -212.80178833007812, -213.70162963867188], "reading_logprob_std": 0.6881620197405804, "tokenization_path_logprob_std": 192.61668333683733, "trajectory_balance_rmse": 2734.732687819942} +{"canonical_length": 9, "canonical_nll": 0.49406422509087455, "generation_rouge_l_f1": 0.044444444444444446, "generation_token_f1": 0.05185185185185186, "prompt_id": "300deaab60201ff0", "reading_canonical_log_prob": -4.446578025817871, "reading_distances": [31, 19, 23, 20], "reading_log_probs": [-5.097175121307373, -5.767045021057129, -5.0457963943481445, -4.740085601806641], "reading_logprob_std": 0.4327415221685663, "tokenization_path_logprob_std": 74.75788124533904, "trajectory_balance_rmse": 202.9530106582947} +{"canonical_length": 239, "canonical_nll": 1.3468779879135069, "generation_rouge_l_f1": 0.2658486707566462, "generation_token_f1": 0.5480572597137013, "prompt_id": "166e7bed707f11c1", "reading_canonical_log_prob": -321.9038391113281, "reading_distances": [21, 26, 23, 26], "reading_log_probs": [-321.5426940917969, -323.0349426269531, -323.17962646484375, -322.5204162597656], "reading_logprob_std": 0.7406372713357188, "tokenization_path_logprob_std": 283.3952883142004, "trajectory_balance_rmse": 5740.089380476448} +{"canonical_length": 278, "canonical_nll": 1.2850791876264613, "generation_rouge_l_f1": 0.21072088724584104, "generation_token_f1": 0.41035120147874304, "prompt_id": "af085ada31a636ce", "reading_canonical_log_prob": -357.25201416015625, "reading_distances": [29, 31, 32, 32], "reading_log_probs": [-355.5769958496094, -358.2568664550781, -357.11907958984375, -359.1334228515625], "reading_logprob_std": 1.536456708011468, "tokenization_path_logprob_std": 31.863392314535716, "trajectory_balance_rmse": 5639.00224921088} +{"canonical_length": 157, "canonical_nll": 1.560959056684166, "generation_rouge_l_f1": 0.22815533980582522, "generation_token_f1": 0.4077669902912621, "prompt_id": "f0f7445e4deb0e6e", "reading_canonical_log_prob": -245.07057189941406, "reading_distances": [28, 29, 13, 35], "reading_log_probs": [-243.06631469726562, -244.7781982421875, -242.6448516845703, -248.3101806640625], "reading_logprob_std": 2.5776010557199, "tokenization_path_logprob_std": 199.00410804390117, "trajectory_balance_rmse": 4157.8593367571575} +{"canonical_length": 20, "canonical_nll": 1.3858312606811523, "generation_rouge_l_f1": 0.1509433962264151, "generation_token_f1": 0.21383647798742136, "prompt_id": "b5d6efcaf5d67694", "reading_canonical_log_prob": -27.716625213623047, "reading_distances": [41, 35, 40, 28], "reading_log_probs": [-25.57929229736328, -26.122575759887695, -26.39207649230957, -25.914384841918945], "reading_logprob_std": 0.3430531418475023, "tokenization_path_logprob_std": 63.45447017581878, "trajectory_balance_rmse": 515.7355924535059} +{"canonical_length": 91, "canonical_nll": 1.8583803281679259, "generation_rouge_l_f1": 0.39520958083832336, "generation_token_f1": 0.5389221556886228, "prompt_id": "b7652a39606177d7", "reading_canonical_log_prob": -169.11260986328125, "reading_distances": [24, 13, 22, 24], "reading_log_probs": [-170.13124084472656, -168.58499145507812, -166.45826721191406, -167.70986938476562], "reading_logprob_std": 1.5438038755223498, "tokenization_path_logprob_std": 118.0166185295614, "trajectory_balance_rmse": 1967.4828999919882} +{"canonical_length": 18, "canonical_nll": 0.9273490905761719, "generation_rouge_l_f1": 0.0879120879120879, "generation_token_f1": 0.09523809523809525, "prompt_id": "f3723ab195d4cd31", "reading_canonical_log_prob": -16.692283630371094, "reading_distances": [25, 26, 22, 20], "reading_log_probs": [-15.14880657196045, -15.07048225402832, -14.9536714553833, -14.440284729003906], "reading_logprob_std": 0.31892726815806327, "tokenization_path_logprob_std": 75.95642772257249, "trajectory_balance_rmse": 413.26384414452406} +{"canonical_length": 7, "canonical_nll": 1.0136969430106026, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "ff3fea8e176d412d", "reading_canonical_log_prob": -7.095878601074219, "reading_distances": [15, 23, 26, 19], "reading_log_probs": [-10.3740234375, -10.838739395141602, -10.446722030639648, -9.903318405151367], "reading_logprob_std": 0.3837085240331887, "tokenization_path_logprob_std": 28.5727015348236, "trajectory_balance_rmse": 132.95738506081304} +{"canonical_length": 69, "canonical_nll": 1.3413205354110054, "generation_rouge_l_f1": 0.4626865671641791, "generation_token_f1": 0.5671641791044776, "prompt_id": "f5b5083636173713", "reading_canonical_log_prob": -92.55111694335938, "reading_distances": [16, 25, 19, 24], "reading_log_probs": [-99.28092956542969, -101.88591003417969, -101.57988739013672, -100.17294311523438], "reading_logprob_std": 1.220458117953198, "tokenization_path_logprob_std": 131.65858010300414, "trajectory_balance_rmse": 1145.4644354155903} +{"canonical_length": 291, "canonical_nll": 0.9992342290190077, "generation_rouge_l_f1": 0.24344569288389514, "generation_token_f1": 0.4681647940074906, "prompt_id": "640dfdad1fff517f", "reading_canonical_log_prob": -290.77716064453125, "reading_distances": [22, 25, 24, 22], "reading_log_probs": [-289.0762939453125, -290.4732666015625, -289.24560546875, -287.57220458984375], "reading_logprob_std": 1.1890495832769699, "tokenization_path_logprob_std": 385.4870546485745, "trajectory_balance_rmse": 6473.057955854666} +{"canonical_length": 17, "canonical_nll": 1.611347983865177, "generation_rouge_l_f1": 0.14814814814814817, "generation_token_f1": 0.14814814814814817, "prompt_id": "f39da6ed7768e21d", "reading_canonical_log_prob": -27.392915725708008, "reading_distances": [34, 32, 27, 33], "reading_log_probs": [-28.30059242248535, -28.998916625976562, -28.945919036865234, -27.691818237304688], "reading_logprob_std": 0.6163598041816191, "tokenization_path_logprob_std": 31.250306207164968, "trajectory_balance_rmse": 405.41951590870326} +{"canonical_length": 16, "canonical_nll": 2.044109344482422, "generation_rouge_l_f1": 0.2424242424242424, "generation_token_f1": 0.2424242424242424, "prompt_id": "968e82be4a52a789", "reading_canonical_log_prob": -32.70574951171875, "reading_distances": [34, 29, 30, 27], "reading_log_probs": [-32.480045318603516, -32.42814636230469, -32.17233657836914, -32.28643798828125], "reading_logprob_std": 0.13946355838399072, "tokenization_path_logprob_std": 53.02205040558088, "trajectory_balance_rmse": 449.4736756912262} +{"canonical_length": 15, "canonical_nll": 2.016009521484375, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.56, "prompt_id": "e36a9df5c4b484e7", "reading_canonical_log_prob": -30.240142822265625, "reading_distances": [23, 25, 23, 26], "reading_log_probs": [-31.155391693115234, -29.207073211669922, -31.865859985351562, -30.199251174926758], "reading_logprob_std": 1.156361532840501, "tokenization_path_logprob_std": 84.52934545222661, "trajectory_balance_rmse": 354.5786086758984} +{"canonical_length": 243, "canonical_nll": 1.1129064991640947, "generation_rouge_l_f1": 0.25193798449612403, "generation_token_f1": 0.5310077519379846, "prompt_id": "87e83988cc685281", "reading_canonical_log_prob": -270.436279296875, "reading_distances": [28, 33, 37, 39], "reading_log_probs": [-274.46063232421875, -273.509521484375, -272.4578857421875, -273.34100341796875], "reading_logprob_std": 0.8207419149487828, "tokenization_path_logprob_std": 237.55114059567848, "trajectory_balance_rmse": 5478.795602111627} +{"canonical_length": 106, "canonical_nll": 0.8039140521355395, "generation_rouge_l_f1": 0.24464831804281348, "generation_token_f1": 0.38532110091743116, "prompt_id": "753de9bc3063e720", "reading_canonical_log_prob": -85.21488952636719, "reading_distances": [21, 16, 16, 21], "reading_log_probs": [-84.73591613769531, -83.58200073242188, -84.1830062866211, -84.52284240722656], "reading_logprob_std": 0.5036954935485923, "tokenization_path_logprob_std": 108.22858669688117, "trajectory_balance_rmse": 2543.200870760378} +{"canonical_length": 18, "canonical_nll": 1.5182680553860135, "generation_rouge_l_f1": 0.56, "generation_token_f1": 0.56, "prompt_id": "52785eaaf4af8786", "reading_canonical_log_prob": -27.328824996948242, "reading_distances": [20, 24, 24, 27], "reading_log_probs": [-24.83161163330078, -21.671606063842773, -26.290332794189453, -24.692123413085938], "reading_logprob_std": 1.9395745770734083, "tokenization_path_logprob_std": 67.56310743558535, "trajectory_balance_rmse": 410.62740767656226} +{"canonical_length": 293, "canonical_nll": 1.7015282953151665, "generation_rouge_l_f1": 0.21254355400696867, "generation_token_f1": 0.43902439024390244, "prompt_id": "9296c91d93887868", "reading_canonical_log_prob": -498.54779052734375, "reading_distances": [19, 19, 23, 17], "reading_log_probs": [-493.0677185058594, -495.0457763671875, -493.6026611328125, -491.94146728515625], "reading_logprob_std": 1.289256884959532, "tokenization_path_logprob_std": 198.838518061979, "trajectory_balance_rmse": 6212.524619599453} +{"canonical_length": 130, "canonical_nll": 1.571245633638822, "generation_rouge_l_f1": 0.2676767676767676, "generation_token_f1": 0.43434343434343436, "prompt_id": "dd7d0916ee5ef6b2", "reading_canonical_log_prob": -204.26193237304688, "reading_distances": [54, 60, 62, 57], "reading_log_probs": [-202.40478515625, -203.4158477783203, -205.02752685546875, -202.42144775390625], "reading_logprob_std": 1.2342098831709682, "tokenization_path_logprob_std": 123.08447927629928, "trajectory_balance_rmse": 2713.6273896944645} +{"canonical_length": 172, "canonical_nll": 1.7713573367096658, "generation_rouge_l_f1": 0.21951219512195125, "generation_token_f1": 0.4292682926829269, "prompt_id": "0d595f7cde9d737e", "reading_canonical_log_prob": -304.6734619140625, "reading_distances": [17, 18, 19, 20], "reading_log_probs": [-300.7552490234375, -302.83782958984375, -301.687744140625, -301.4686584472656], "reading_logprob_std": 0.8641453708899536, "tokenization_path_logprob_std": 297.5983784170231, "trajectory_balance_rmse": 3664.9580076904954} +{"canonical_length": 154, "canonical_nll": 0.8024149312601461, "generation_rouge_l_f1": 0.2179176755447942, "generation_token_f1": 0.4358353510895884, "prompt_id": "8c2391cc4e076841", "reading_canonical_log_prob": -123.5718994140625, "reading_distances": [233, 246, 247, 230], "reading_log_probs": [-120.77554321289062, -123.69595336914062, -121.90282440185547, -123.80716705322266], "reading_logprob_std": 1.467551883797529, "tokenization_path_logprob_std": 107.70142975806357, "trajectory_balance_rmse": 2805.7476560695} +{"canonical_length": 395, "canonical_nll": 0.7989743003362342, "generation_rouge_l_f1": 0.14205607476635515, "generation_token_f1": 0.20186915887850468, "prompt_id": "a3d283a16147427b", "reading_canonical_log_prob": -315.5948486328125, "reading_distances": [14, 17, 17, 14], "reading_log_probs": [-313.6956787109375, -314.8850402832031, -315.5116271972656, -313.53277587890625], "reading_logprob_std": 0.9520097122858374, "tokenization_path_logprob_std": 212.09755537209824, "trajectory_balance_rmse": 6190.140104243439} +{"canonical_length": 11, "canonical_nll": 1.237927263433283, "generation_rouge_l_f1": 0.3773584905660377, "generation_token_f1": 0.41509433962264153, "prompt_id": "977c41664864ae25", "reading_canonical_log_prob": -13.617199897766113, "reading_distances": [30, 32, 35, 27], "reading_log_probs": [-13.995083808898926, -14.202783584594727, -13.7266845703125, -13.171671867370605], "reading_logprob_std": 0.4463822463962872, "tokenization_path_logprob_std": 20.31237468250411, "trajectory_balance_rmse": 275.1806252905894} +{"canonical_length": 20, "canonical_nll": 1.081063175201416, "generation_rouge_l_f1": 0.12941176470588237, "generation_token_f1": 0.15294117647058825, "prompt_id": "4c8a56abb260b4fb", "reading_canonical_log_prob": -21.62126350402832, "reading_distances": [28, 29, 25, 28], "reading_log_probs": [-23.3360538482666, -22.902198791503906, -23.048023223876953, -23.26436996459961], "reading_logprob_std": 0.19907474851635304, "tokenization_path_logprob_std": 53.84300928271522, "trajectory_balance_rmse": 254.17091690730683} +{"canonical_length": 244, "canonical_nll": 0.7817182697233607, "generation_rouge_l_f1": 0.3353293413173653, "generation_token_f1": 0.5988023952095809, "prompt_id": "7a98c8dec132eec9", "reading_canonical_log_prob": -190.7392578125, "reading_distances": [32, 35, 31, 33], "reading_log_probs": [-196.7659149169922, -194.78179931640625, -197.45672607421875, -197.29295349121094], "reading_logprob_std": 1.2308416611678465, "tokenization_path_logprob_std": 221.73888686544385, "trajectory_balance_rmse": 4913.980506502252} +{"canonical_length": 419, "canonical_nll": 1.1571215354172881, "generation_rouge_l_f1": 0.21551724137931033, "generation_token_f1": 0.3649425287356322, "prompt_id": "7f79088a335cbb01", "reading_canonical_log_prob": -484.83392333984375, "reading_distances": [23, 13, 20, 16], "reading_log_probs": [-469.3155822753906, -473.3508605957031, -470.9941711425781, -472.191650390625], "reading_logprob_std": 1.7249301806746768, "tokenization_path_logprob_std": 72.78819712973716, "trajectory_balance_rmse": 7865.935811342744} +{"canonical_length": 12, "canonical_nll": 2.006214459737142, "generation_rouge_l_f1": 0.4444444444444445, "generation_token_f1": 0.4444444444444445, "prompt_id": "48b64f1a7aabf6e4", "reading_canonical_log_prob": -24.074573516845703, "reading_distances": [30, 23, 21, 27], "reading_log_probs": [-23.05025863647461, -24.180831909179688, -22.87753677368164, -22.742645263671875], "reading_logprob_std": 0.6575085240874207, "tokenization_path_logprob_std": 49.22248377375953, "trajectory_balance_rmse": 327.76094444024534} +{"canonical_length": 319, "canonical_nll": 1.4971810941785855, "generation_rouge_l_f1": 0.252991452991453, "generation_token_f1": 0.4478632478632478, "prompt_id": "735c6c6b04d8c3fe", "reading_canonical_log_prob": -477.60076904296875, "reading_distances": [35, 38, 37, 39], "reading_log_probs": [-480.1153869628906, -480.18096923828125, -479.16339111328125, -481.7421875], "reading_logprob_std": 1.0677131178916364, "tokenization_path_logprob_std": 286.4592379414002, "trajectory_balance_rmse": 6737.858958433047} +{"canonical_length": 57, "canonical_nll": 1.319793701171875, "generation_rouge_l_f1": 0.208955223880597, "generation_token_f1": 0.2537313432835821, "prompt_id": "eab393eed1fb7651", "reading_canonical_log_prob": -75.22824096679688, "reading_distances": [35, 35, 37, 35], "reading_log_probs": [-75.34500885009766, -76.85078430175781, -75.94420623779297, -75.37948608398438], "reading_logprob_std": 0.7031534317927378, "tokenization_path_logprob_std": 85.50341840701688, "trajectory_balance_rmse": 1313.8388946589248} +{"canonical_length": 20, "canonical_nll": 0.0069592341780662535, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "691dfdc26548dcb9", "reading_canonical_log_prob": -0.13918468356132507, "reading_distances": [30, 31, 33, 31], "reading_log_probs": [-0.23035100102424622, -0.22561341524124146, -0.23827876150608063, -0.2623448371887207], "reading_logprob_std": 0.01632402118270928, "tokenization_path_logprob_std": 38.93117089310236, "trajectory_balance_rmse": 325.16531414641514} +{"canonical_length": 230, "canonical_nll": 1.8108012323794158, "generation_rouge_l_f1": 0.25684210526315787, "generation_token_f1": 0.5094736842105264, "prompt_id": "2b073d802d00719d", "reading_canonical_log_prob": -416.4842834472656, "reading_distances": [24, 28, 25, 21], "reading_log_probs": [-414.5251159667969, -414.63970947265625, -414.656982421875, -413.73187255859375], "reading_logprob_std": 0.4415927624960971, "tokenization_path_logprob_std": 77.79951159480017, "trajectory_balance_rmse": 4337.5206065323255} +{"canonical_length": 11, "canonical_nll": 4.009003379128196, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "11a956113e5f3fb8", "reading_canonical_log_prob": -44.099037170410156, "reading_distances": [29, 30, 27, 26], "reading_log_probs": [-52.85093688964844, -49.421600341796875, -49.0755615234375, -51.26776123046875], "reading_logprob_std": 1.752475533072741, "tokenization_path_logprob_std": 0.0, "trajectory_balance_rmse": 44.135467529296875} +{"canonical_length": 162, "canonical_nll": 1.1385540432400174, "generation_rouge_l_f1": 0.45993031358885017, "generation_token_f1": 0.5644599303135889, "prompt_id": "b48c6b78ff2012e6", "reading_canonical_log_prob": -184.4457550048828, "reading_distances": [30, 20, 32, 22], "reading_log_probs": [-182.20114135742188, -181.09051513671875, -181.6356964111328, -182.26211547851562], "reading_logprob_std": 0.5491817598096395, "tokenization_path_logprob_std": 182.66695231004715, "trajectory_balance_rmse": 3023.5022969630736} +{"canonical_length": 142, "canonical_nll": 0.7847210521429357, "generation_rouge_l_f1": 0.37419354838709673, "generation_token_f1": 0.5225806451612903, "prompt_id": "c057418a3f03722b", "reading_canonical_log_prob": -111.43038940429688, "reading_distances": [34, 36, 34, 30], "reading_log_probs": [-114.9261703491211, -111.96532440185547, -111.64300537109375, -111.34681701660156], "reading_logprob_std": 1.6565959012370524, "tokenization_path_logprob_std": 169.51692017619433, "trajectory_balance_rmse": 2899.270285812166} +{"canonical_length": 239, "canonical_nll": 1.2583444507550994, "generation_rouge_l_f1": 0.2927835051546392, "generation_token_f1": 0.5319587628865978, "prompt_id": "14f504daadaa6ae8", "reading_canonical_log_prob": -300.74432373046875, "reading_distances": [23, 17, 20, 23], "reading_log_probs": [-302.25640869140625, -303.15509033203125, -302.4205322265625, -303.03863525390625], "reading_logprob_std": 0.4454990057051793, "tokenization_path_logprob_std": 263.279531479348, "trajectory_balance_rmse": 5977.279534228498} +{"canonical_length": 97, "canonical_nll": 1.3593765730710374, "generation_rouge_l_f1": 0.19375, "generation_token_f1": 0.33749999999999997, "prompt_id": "20abb31273f6c931", "reading_canonical_log_prob": -131.85952758789062, "reading_distances": [20, 23, 30, 29], "reading_log_probs": [-129.31307983398438, -132.78936767578125, -132.72093200683594, -130.5234832763672], "reading_logprob_std": 1.711013504712358, "tokenization_path_logprob_std": 212.68601576908162, "trajectory_balance_rmse": 2410.4913752139237} +{"canonical_length": 5, "canonical_nll": 1.14029541015625, "generation_rouge_l_f1": 0.00881057268722467, "generation_token_f1": 0.00881057268722467, "prompt_id": "609097a0b2afbf5d", "reading_canonical_log_prob": -5.70147705078125, "reading_distances": [26, 26, 23, 33], "reading_log_probs": [-5.8042168617248535, -5.268927574157715, -6.607482433319092, -5.297617435455322], "reading_logprob_std": 0.6256138126101859, "tokenization_path_logprob_std": 35.684058825562374, "trajectory_balance_rmse": 79.87134337664428} +{"canonical_length": 70, "canonical_nll": 2.225093296595982, "generation_rouge_l_f1": 0.08124999999999999, "generation_token_f1": 0.1375, "prompt_id": "b13cf3764c885d75", "reading_canonical_log_prob": -155.75653076171875, "reading_distances": [30, 22, 25, 35], "reading_log_probs": [-153.96737670898438, -154.83155822753906, -156.89051818847656, -154.2316131591797], "reading_logprob_std": 1.323833224275997, "tokenization_path_logprob_std": 73.31861183559799, "trajectory_balance_rmse": 1603.9363947457832} +{"canonical_length": 42, "canonical_nll": 0.9672066824776786, "generation_rouge_l_f1": 0.1245136186770428, "generation_token_f1": 0.16342412451361868, "prompt_id": "d289e388118f2de0", "reading_canonical_log_prob": -40.6226806640625, "reading_distances": [21, 22, 26, 20], "reading_log_probs": [-38.83055877685547, -39.74464416503906, -41.19381332397461, -38.227935791015625], "reading_logprob_std": 1.290368283927313, "tokenization_path_logprob_std": 88.17543082248658, "trajectory_balance_rmse": 831.8111299534208} +{"canonical_length": 10, "canonical_nll": 2.6498714447021485, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.25, "prompt_id": "58e5e1069ca06eb2", "reading_canonical_log_prob": -26.498714447021484, "reading_distances": [23, 12, 20, 22], "reading_log_probs": [-26.857532501220703, -25.516416549682617, -26.658138275146484, -25.914196014404297], "reading_logprob_std": 0.6287183304475737, "tokenization_path_logprob_std": 64.62983137295568, "trajectory_balance_rmse": 274.45519028512814} +{"canonical_length": 113, "canonical_nll": 1.400510399742464, "generation_rouge_l_f1": 0.3846153846153846, "generation_token_f1": 0.5288461538461539, "prompt_id": "ec3868510227235c", "reading_canonical_log_prob": -158.25767517089844, "reading_distances": [34, 28, 26, 26], "reading_log_probs": [-158.04893493652344, -158.4451904296875, -156.78286743164062, -156.11593627929688], "reading_logprob_std": 1.0851255288574007, "tokenization_path_logprob_std": 118.40152507512083, "trajectory_balance_rmse": 2564.5704057869866} +{"canonical_length": 16, "canonical_nll": 2.5522944927215576, "generation_rouge_l_f1": 0.24, "generation_token_f1": 0.32, "prompt_id": "d625c7ada7581821", "reading_canonical_log_prob": -40.83671188354492, "reading_distances": [11, 10, 11, 10], "reading_log_probs": [-41.186248779296875, -40.587730407714844, -42.29832077026367, -43.607688903808594], "reading_logprob_std": 1.3297560630470933, "tokenization_path_logprob_std": 46.46825208933357, "trajectory_balance_rmse": 437.63706517727655} +{"canonical_length": 7, "canonical_nll": 1.4588026319231306, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.5714285714285715, "prompt_id": "34ca0227975643ca", "reading_canonical_log_prob": -10.211618423461914, "reading_distances": [23, 19, 21, 16], "reading_log_probs": [-9.41035270690918, -9.629303932189941, -9.454123497009277, -9.643001556396484], "reading_logprob_std": 0.11920980782586922, "tokenization_path_logprob_std": 18.67832090799946, "trajectory_balance_rmse": 76.07861893467948} +{"canonical_length": 92, "canonical_nll": 0.4672715145608653, "generation_rouge_l_f1": 0.343042071197411, "generation_token_f1": 0.43365695792880254, "prompt_id": "03d8c4a193a46caa", "reading_canonical_log_prob": -42.98897933959961, "reading_distances": [29, 29, 32, 23], "reading_log_probs": [-42.738739013671875, -42.60014343261719, -43.14642333984375, -42.86381530761719], "reading_logprob_std": 0.23253516199807353, "tokenization_path_logprob_std": 128.6091364872101, "trajectory_balance_rmse": 1405.9485534045427} +{"canonical_length": 22, "canonical_nll": 0.9292023398659446, "generation_rouge_l_f1": 0.4186046511627907, "generation_token_f1": 0.6511627906976744, "prompt_id": "cfb9b15522adbd99", "reading_canonical_log_prob": -20.44245147705078, "reading_distances": [22, 14, 16, 16], "reading_log_probs": [-20.93605613708496, -21.288965225219727, -21.140073776245117, -21.16783905029297], "reading_logprob_std": 0.14648733298619057, "tokenization_path_logprob_std": 29.035491941589886, "trajectory_balance_rmse": 566.026279163057} +{"canonical_length": 115, "canonical_nll": 1.4461433742357337, "generation_rouge_l_f1": 0.21587301587301586, "generation_token_f1": 0.4380952380952381, "prompt_id": "7e152e3b35975991", "reading_canonical_log_prob": -166.30648803710938, "reading_distances": [25, 29, 33, 23], "reading_log_probs": [-163.43663024902344, -163.49652099609375, -164.17984008789062, -163.34182739257812], "reading_logprob_std": 0.3827592495563734, "tokenization_path_logprob_std": 114.08552609631823, "trajectory_balance_rmse": 3293.162094881917} +{"canonical_length": 10, "canonical_nll": 3.6648616790771484, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.2222222222222222, "prompt_id": "6df3b4a03782fc55", "reading_canonical_log_prob": -36.648616790771484, "reading_distances": [26, 26, 24, 27], "reading_log_probs": [-32.29587936401367, -33.329402923583984, -33.60430145263672, -32.46335220336914], "reading_logprob_std": 0.6413244565460883, "tokenization_path_logprob_std": 42.091126002035516, "trajectory_balance_rmse": 286.5270816146915} +{"canonical_length": 150, "canonical_nll": 1.5300838216145833, "generation_rouge_l_f1": 0.1897810218978102, "generation_token_f1": 0.40875912408759124, "prompt_id": "68f61446f4cfbca0", "reading_canonical_log_prob": -229.5125732421875, "reading_distances": [20, 26, 25, 32], "reading_log_probs": [-231.75445556640625, -231.88485717773438, -232.06092834472656, -231.63534545898438], "reading_logprob_std": 0.18245896982943205, "tokenization_path_logprob_std": 187.38850567938817, "trajectory_balance_rmse": 3060.324095026664} +{"canonical_length": 206, "canonical_nll": 1.3981862484830097, "generation_rouge_l_f1": 0.2545454545454546, "generation_token_f1": 0.37272727272727274, "prompt_id": "6a128630a7f46740", "reading_canonical_log_prob": -288.0263671875, "reading_distances": [26, 28, 25, 28], "reading_log_probs": [-286.50018310546875, -285.85400390625, -287.1044006347656, -285.880126953125], "reading_logprob_std": 0.5937252768249557, "tokenization_path_logprob_std": 86.84358643549227, "trajectory_balance_rmse": 4459.25304113499} +{"canonical_length": 324, "canonical_nll": 1.180803840543017, "generation_rouge_l_f1": 0.1903114186851211, "generation_token_f1": 0.36678200692041524, "prompt_id": "fc8ef010d8778b0f", "reading_canonical_log_prob": -382.5804443359375, "reading_distances": [41, 46, 39, 39], "reading_log_probs": [-380.23309326171875, -379.97930908203125, -381.6121826171875, -380.3852233886719], "reading_logprob_std": 0.7260572056400761, "tokenization_path_logprob_std": 330.96468323912336, "trajectory_balance_rmse": 7456.024835928569} +{"canonical_length": 54, "canonical_nll": 0.387207101892542, "generation_rouge_l_f1": 0.27586206896551724, "generation_token_f1": 0.3065134099616858, "prompt_id": "68eb421daa21c33e", "reading_canonical_log_prob": -20.909183502197266, "reading_distances": [20, 21, 24, 17], "reading_log_probs": [-23.93260383605957, -23.08151626586914, -24.82818031311035, -25.58587646484375], "reading_logprob_std": 1.086142885856149, "tokenization_path_logprob_std": 51.88135231928485, "trajectory_balance_rmse": 793.8620132322208} +{"canonical_length": 133, "canonical_nll": 0.9427004362407484, "generation_rouge_l_f1": 0.29207920792079206, "generation_token_f1": 0.5099009900990099, "prompt_id": "a12c87820d75347d", "reading_canonical_log_prob": -125.37915802001953, "reading_distances": [22, 28, 23, 20], "reading_log_probs": [-127.80026245117188, -128.52328491210938, -129.01072692871094, -127.7105712890625], "reading_logprob_std": 0.6180977589149887, "tokenization_path_logprob_std": 91.37580065791211, "trajectory_balance_rmse": 2756.9311136027172} +{"canonical_length": 220, "canonical_nll": 0.0428065993569114, "generation_rouge_l_f1": 0.9051490514905149, "generation_token_f1": 0.9051490514905149, "prompt_id": "c386599fdaa650e0", "reading_canonical_log_prob": -9.417451858520508, "reading_distances": [14, 10, 11, 12], "reading_log_probs": [-11.182275772094727, -10.261672019958496, -10.912300109863281, -10.664315223693848], "reading_logprob_std": 0.391111260917768, "tokenization_path_logprob_std": 162.12783290896982, "trajectory_balance_rmse": 2415.839436704003} +{"canonical_length": 13, "canonical_nll": 1.2675000704251802, "generation_rouge_l_f1": 0.5263157894736842, "generation_token_f1": 0.5263157894736842, "prompt_id": "28aa44aad6506016", "reading_canonical_log_prob": -16.477500915527344, "reading_distances": [24, 28, 30, 33], "reading_log_probs": [-16.166494369506836, -17.21092987060547, -16.936885833740234, -17.208049774169922], "reading_logprob_std": 0.49310449338755297, "tokenization_path_logprob_std": 41.16292421503554, "trajectory_balance_rmse": 313.890100810415} +{"canonical_length": 150, "canonical_nll": 1.4018941243489584, "generation_rouge_l_f1": 0.2755344418052256, "generation_token_f1": 0.48456057007125886, "prompt_id": "764869f6f55c420d", "reading_canonical_log_prob": -210.28411865234375, "reading_distances": [11, 13, 15, 13], "reading_log_probs": [-211.0135498046875, -208.59640502929688, -209.74253845214844, -210.6126708984375], "reading_logprob_std": 1.0706245440972524, "tokenization_path_logprob_std": 87.75437741747788, "trajectory_balance_rmse": 2709.0837442961115} +{"canonical_length": 266, "canonical_nll": 0.7659774436090225, "generation_rouge_l_f1": 0.27466150870406186, "generation_token_f1": 0.5106382978723404, "prompt_id": "8fb6a1df0bc7d753", "reading_canonical_log_prob": -203.75, "reading_distances": [44, 32, 42, 44], "reading_log_probs": [-207.29718017578125, -208.81906127929688, -209.87490844726562, -208.98123168945312], "reading_logprob_std": 1.0699145319074337, "tokenization_path_logprob_std": 154.56021659830088, "trajectory_balance_rmse": 4785.44746985385} +{"canonical_length": 81, "canonical_nll": 1.645562631112558, "generation_rouge_l_f1": 0.17058823529411765, "generation_token_f1": 0.2823529411764706, "prompt_id": "f5aeb1160ca2b326", "reading_canonical_log_prob": -133.2905731201172, "reading_distances": [52, 48, 48, 53], "reading_log_probs": [-127.16606140136719, -127.5578384399414, -128.71682739257812, -126.3552017211914], "reading_logprob_std": 0.9824596730382751, "tokenization_path_logprob_std": 110.12031657075931, "trajectory_balance_rmse": 1844.7135572798518} +{"canonical_length": 29, "canonical_nll": 0.7840959614720838, "generation_rouge_l_f1": 0.42857142857142866, "generation_token_f1": 0.44897959183673475, "prompt_id": "7d27494a4e6427f6", "reading_canonical_log_prob": -22.73878288269043, "reading_distances": [38, 36, 37, 27], "reading_log_probs": [-22.622407913208008, -23.769306182861328, -23.056373596191406, -24.102235794067383], "reading_logprob_std": 0.6712261881567585, "tokenization_path_logprob_std": 47.55731849995952, "trajectory_balance_rmse": 599.2846792949118} +{"canonical_length": 13, "canonical_nll": 0.32061925301185024, "generation_rouge_l_f1": 0.8421052631578948, "generation_token_f1": 0.8421052631578948, "prompt_id": "b015ce761d144f7f", "reading_canonical_log_prob": -4.168050289154053, "reading_distances": [29, 23, 31, 25], "reading_log_probs": [-5.002214431762695, -4.982389450073242, -4.9380717277526855, -5.0040178298950195], "reading_logprob_std": 0.030674778367429052, "tokenization_path_logprob_std": 55.396634774702136, "trajectory_balance_rmse": 187.74369849117326} +{"canonical_length": 246, "canonical_nll": 1.2613553923319996, "generation_rouge_l_f1": 0.22656250000000003, "generation_token_f1": 0.4609375, "prompt_id": "e7d47f51d6085f91", "reading_canonical_log_prob": -310.2934265136719, "reading_distances": [24, 16, 22, 18], "reading_log_probs": [-309.9610595703125, -311.38623046875, -309.8631286621094, -310.6457824707031], "reading_logprob_std": 0.7065289176768735, "tokenization_path_logprob_std": 149.74404106074965, "trajectory_balance_rmse": 5884.590940830618} +{"canonical_length": 205, "canonical_nll": 1.5247339760384908, "generation_rouge_l_f1": 0.27536231884057966, "generation_token_f1": 0.5120772946859903, "prompt_id": "321f893720a9d001", "reading_canonical_log_prob": -312.5704650878906, "reading_distances": [29, 23, 37, 23], "reading_log_probs": [-313.61309814453125, -312.13677978515625, -311.8411865234375, -312.2054748535156], "reading_logprob_std": 0.7919064757925421, "tokenization_path_logprob_std": 381.2313514062006, "trajectory_balance_rmse": 4922.50202371645} +{"canonical_length": 159, "canonical_nll": 1.0864912308986832, "generation_rouge_l_f1": 0.13095238095238096, "generation_token_f1": 0.13095238095238096, "prompt_id": "46617b87eb38bf31", "reading_canonical_log_prob": -172.75210571289062, "reading_distances": [34, 38, 37, 34], "reading_log_probs": [-175.24526977539062, -181.096923828125, -178.12472534179688, -181.50021362304688], "reading_logprob_std": 2.9161614279562604, "tokenization_path_logprob_std": 154.15848993616902, "trajectory_balance_rmse": 2129.0840371125223} +{"canonical_length": 4, "canonical_nll": 0.3138948678970337, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "03d95aafdc590416", "reading_canonical_log_prob": -1.2555794715881348, "reading_distances": [38, 34, 23, 35], "reading_log_probs": [-1.09666109085083, -1.1118690967559814, -1.0591907501220703, -1.2712433338165283], "reading_logprob_std": 0.09365527362402129, "tokenization_path_logprob_std": 21.641323853846153, "trajectory_balance_rmse": 67.10924080100835} +{"canonical_length": 153, "canonical_nll": 0.6699812147352431, "generation_rouge_l_f1": 0.29906542056074764, "generation_token_f1": 0.49065420560747663, "prompt_id": "3e79c37d02563d09", "reading_canonical_log_prob": -102.50712585449219, "reading_distances": [39, 36, 39, 30], "reading_log_probs": [-107.18773651123047, -109.61538696289062, -108.42208099365234, -109.4196548461914], "reading_logprob_std": 1.1126543435879952, "tokenization_path_logprob_std": 135.74244800575562, "trajectory_balance_rmse": 2275.247388109913} +{"canonical_length": 385, "canonical_nll": 0.9809838233056006, "generation_rouge_l_f1": 0.2137931034482759, "generation_token_f1": 0.33793103448275863, "prompt_id": "8a6a50a6d59a29d8", "reading_canonical_log_prob": -377.67877197265625, "reading_distances": [35, 25, 33, 31], "reading_log_probs": [-386.4287109375, -384.4874572753906, -381.36346435546875, -382.4098815917969], "reading_logprob_std": 2.249941101906588, "tokenization_path_logprob_std": 227.9015761319025, "trajectory_balance_rmse": 7133.382573358769} +{"canonical_length": 28, "canonical_nll": 0.712965897151402, "generation_rouge_l_f1": 0.19095477386934673, "generation_token_f1": 0.21105527638190955, "prompt_id": "280d0259825e43a6", "reading_canonical_log_prob": -19.963045120239258, "reading_distances": [32, 33, 36, 36], "reading_log_probs": [-20.545692443847656, -20.585342407226562, -18.865175247192383, -19.45621681213379], "reading_logprob_std": 0.8463597107313192, "tokenization_path_logprob_std": 93.31414313605349, "trajectory_balance_rmse": 430.307376329756} +{"canonical_length": 245, "canonical_nll": 1.6267697704081632, "generation_rouge_l_f1": 0.15444015444015444, "generation_token_f1": 0.33976833976833976, "prompt_id": "fcecb04380951f2a", "reading_canonical_log_prob": -398.55859375, "reading_distances": [26, 23, 25, 26], "reading_log_probs": [-392.4910888671875, -391.671875, -390.621826171875, -392.7711181640625], "reading_logprob_std": 0.9649725788096128, "tokenization_path_logprob_std": 199.75654949960762, "trajectory_balance_rmse": 5557.647999276965} +{"canonical_length": 126, "canonical_nll": 1.536636715843564, "generation_rouge_l_f1": 0.18630136986301374, "generation_token_f1": 0.30136986301369867, "prompt_id": "1ae061f91192abf8", "reading_canonical_log_prob": -193.61622619628906, "reading_distances": [29, 32, 30, 21], "reading_log_probs": [-191.14047241210938, -193.04171752929688, -192.6264190673828, -211.5789337158203], "reading_logprob_std": 9.689135406813696, "tokenization_path_logprob_std": 153.6819228924614, "trajectory_balance_rmse": 2500.6454227375793} +{"canonical_length": 205, "canonical_nll": 1.2638424757050304, "generation_rouge_l_f1": 0.24524312896405917, "generation_token_f1": 0.5158562367864693, "prompt_id": "eedaf90695c8e20b", "reading_canonical_log_prob": -259.08770751953125, "reading_distances": [16, 17, 16, 21], "reading_log_probs": [-258.9288024902344, -258.0395812988281, -258.4094543457031, -259.21856689453125], "reading_logprob_std": 0.5264564286199064, "tokenization_path_logprob_std": 124.92685715603378, "trajectory_balance_rmse": 4896.944925681963} +{"canonical_length": 119, "canonical_nll": 1.1280579126181722, "generation_rouge_l_f1": 0.2492753623188406, "generation_token_f1": 0.4521739130434782, "prompt_id": "a0b38c7913178787", "reading_canonical_log_prob": -134.2388916015625, "reading_distances": [48, 44, 45, 41], "reading_log_probs": [-124.40190124511719, -123.70986938476562, -124.78854370117188, -126.39372253417969], "reading_logprob_std": 1.137942500984373, "tokenization_path_logprob_std": 99.762786710748, "trajectory_balance_rmse": 1592.1634688040285} +{"canonical_length": 68, "canonical_nll": 1.5946590199190027, "generation_rouge_l_f1": 0.2538071065989848, "generation_token_f1": 0.4060913705583757, "prompt_id": "4101000b96f3bbde", "reading_canonical_log_prob": -108.43681335449219, "reading_distances": [18, 19, 23, 15], "reading_log_probs": [-105.51608276367188, -107.22846984863281, -105.87055969238281, -108.52714538574219], "reading_logprob_std": 1.3757527401709109, "tokenization_path_logprob_std": 39.39476496864024, "trajectory_balance_rmse": 1199.822102822046} +{"canonical_length": 138, "canonical_nll": 1.2660571112149004, "generation_rouge_l_f1": 0.248780487804878, "generation_token_f1": 0.41463414634146345, "prompt_id": "5539256ae94ad40f", "reading_canonical_log_prob": -174.71588134765625, "reading_distances": [63, 57, 65, 62], "reading_log_probs": [-178.00595092773438, -178.1317901611328, -177.4249725341797, -175.5821533203125], "reading_logprob_std": 1.1770155790891075, "tokenization_path_logprob_std": 243.58815537002167, "trajectory_balance_rmse": 2686.303223564932} +{"canonical_length": 86, "canonical_nll": 1.193320252174555, "generation_rouge_l_f1": 0.19827586206896552, "generation_token_f1": 0.37068965517241376, "prompt_id": "3b0f78eeffb87a9e", "reading_canonical_log_prob": -102.62554168701172, "reading_distances": [32, 33, 33, 35], "reading_log_probs": [-101.300537109375, -103.246826171875, -101.8636245727539, -101.70860290527344], "reading_logprob_std": 0.845332059958186, "tokenization_path_logprob_std": 49.10903801018805, "trajectory_balance_rmse": 1745.546168523969} +{"canonical_length": 302, "canonical_nll": 1.5024306947821813, "generation_rouge_l_f1": 0.2834951456310679, "generation_token_f1": 0.45436893203883494, "prompt_id": "3a8d8459792a6678", "reading_canonical_log_prob": -453.73406982421875, "reading_distances": [14, 16, 18, 17], "reading_log_probs": [-450.77886962890625, -448.06890869140625, -448.6070251464844, -445.62847900390625], "reading_logprob_std": 2.115504743377144, "tokenization_path_logprob_std": 162.32837505506745, "trajectory_balance_rmse": 6452.919580391399} +{"canonical_length": 289, "canonical_nll": 1.1713425791387326, "generation_rouge_l_f1": 0.36116504854368936, "generation_token_f1": 0.6291262135922331, "prompt_id": "a788a16b80a319d9", "reading_canonical_log_prob": -338.51800537109375, "reading_distances": [12, 20, 16, 20], "reading_log_probs": [-338.00421142578125, -337.321044921875, -335.91693115234375, -336.9039611816406], "reading_logprob_std": 0.8733903582477339, "tokenization_path_logprob_std": 184.95067506248049, "trajectory_balance_rmse": 5687.799506668776} +{"canonical_length": 252, "canonical_nll": 1.1732310946025546, "generation_rouge_l_f1": 0.2846153846153846, "generation_token_f1": 0.5769230769230771, "prompt_id": "79833cd8d7978d26", "reading_canonical_log_prob": -295.65423583984375, "reading_distances": [30, 30, 26, 31], "reading_log_probs": [-291.69244384765625, -293.1526794433594, -295.0028381347656, -291.9388427734375], "reading_logprob_std": 1.512069910767587, "tokenization_path_logprob_std": 203.581828397429, "trajectory_balance_rmse": 4465.411056345584} +{"canonical_length": 26, "canonical_nll": 1.4830504197340746, "generation_rouge_l_f1": 0.1176470588235294, "generation_token_f1": 0.15294117647058825, "prompt_id": "7c29635f759f46ec", "reading_canonical_log_prob": -38.55931091308594, "reading_distances": [18, 22, 22, 16], "reading_log_probs": [-37.577247619628906, -38.04298400878906, -39.91388702392578, -39.524391174316406], "reading_logprob_std": 1.129701819487788, "tokenization_path_logprob_std": 53.17562659518146, "trajectory_balance_rmse": 582.4027392249615} +{"canonical_length": 346, "canonical_nll": 1.2057612424640987, "generation_rouge_l_f1": 0.31736526946107785, "generation_token_f1": 0.5089820359281437, "prompt_id": "88945c2caccedba1", "reading_canonical_log_prob": -417.1933898925781, "reading_distances": [39, 36, 53, 44], "reading_log_probs": [-411.9376220703125, -415.5388488769531, -412.5898742675781, -419.09619140625], "reading_logprob_std": 3.270100434308652, "tokenization_path_logprob_std": 144.60296379783136, "trajectory_balance_rmse": 7556.691136070349} +{"canonical_length": 17, "canonical_nll": 1.627893560072955, "generation_rouge_l_f1": 0.2758620689655172, "generation_token_f1": 0.4827586206896552, "prompt_id": "f3be16dc1ce038fa", "reading_canonical_log_prob": -27.674190521240234, "reading_distances": [34, 49, 48, 47], "reading_log_probs": [-29.282651901245117, -28.56758689880371, -28.62774085998535, -28.101337432861328], "reading_logprob_std": 0.485955925172118, "tokenization_path_logprob_std": 69.05636120186992, "trajectory_balance_rmse": 488.7083461421638} +{"canonical_length": 56, "canonical_nll": 1.4947171892438615, "generation_rouge_l_f1": 0.2346368715083799, "generation_token_f1": 0.29050279329608936, "prompt_id": "08f613a2192f5ff2", "reading_canonical_log_prob": -83.70416259765625, "reading_distances": [23, 24, 20, 20], "reading_log_probs": [-86.44158935546875, -85.5196762084961, -85.57334899902344, -85.57666778564453], "reading_logprob_std": 0.44328267051265713, "tokenization_path_logprob_std": 21.248389435255664, "trajectory_balance_rmse": 1166.2560176681484} +{"canonical_length": 180, "canonical_nll": 1.679837375217014, "generation_rouge_l_f1": 0.19672131147540986, "generation_token_f1": 0.39812646370023425, "prompt_id": "ddaf5127b3985530", "reading_canonical_log_prob": -302.3707275390625, "reading_distances": [26, 27, 31, 27], "reading_log_probs": [-299.81951904296875, -300.9543762207031, -301.93841552734375, -302.221923828125], "reading_logprob_std": 1.087985873368774, "tokenization_path_logprob_std": 91.5748466803534, "trajectory_balance_rmse": 3769.627571302604} +{"canonical_length": 184, "canonical_nll": 1.016448974609375, "generation_rouge_l_f1": 0.3350785340314136, "generation_token_f1": 0.6073298429319371, "prompt_id": "6fc1da282516cfc1", "reading_canonical_log_prob": -187.026611328125, "reading_distances": [18, 12, 9, 14], "reading_log_probs": [-188.97463989257812, -186.913818359375, -187.38662719726562, -187.9413604736328], "reading_logprob_std": 0.8861694754895592, "tokenization_path_logprob_std": 209.26667047959734, "trajectory_balance_rmse": 3880.597968799075} +{"canonical_length": 80, "canonical_nll": 1.2984180450439453, "generation_rouge_l_f1": 0.14814814814814814, "generation_token_f1": 0.2222222222222222, "prompt_id": "9d2634370eff9e7e", "reading_canonical_log_prob": -103.87344360351562, "reading_distances": [39, 26, 33, 34], "reading_log_probs": [-109.07696533203125, -109.98155212402344, -107.75277709960938, -109.03179931640625], "reading_logprob_std": 0.9164768611457753, "tokenization_path_logprob_std": 137.53132244128932, "trajectory_balance_rmse": 1908.7045799725956} +{"canonical_length": 132, "canonical_nll": 0.7477546460700758, "generation_rouge_l_f1": 0.36746987951807225, "generation_token_f1": 0.5602409638554217, "prompt_id": "8a81f9867fa3fdd7", "reading_canonical_log_prob": -98.70361328125, "reading_distances": [33, 32, 31, 34], "reading_log_probs": [-96.09679412841797, -101.01953887939453, -96.664306640625, -98.07466888427734], "reading_logprob_std": 2.200320122253496, "tokenization_path_logprob_std": 59.13226352879833, "trajectory_balance_rmse": 1927.2680374781505} +{"canonical_length": 33, "canonical_nll": 0.8702930681633226, "generation_rouge_l_f1": 0.8363636363636363, "generation_token_f1": 0.8363636363636363, "prompt_id": "32094cae1c9e7173", "reading_canonical_log_prob": -28.71967124938965, "reading_distances": [44, 46, 39, 44], "reading_log_probs": [-29.745759963989258, -30.02827262878418, -29.84642791748047, -28.644704818725586], "reading_logprob_std": 0.6254155728847914, "tokenization_path_logprob_std": 47.236261696180904, "trajectory_balance_rmse": 636.4880705496503} +{"canonical_length": 23, "canonical_nll": 1.8069968845533289, "generation_rouge_l_f1": 0.3225806451612903, "generation_token_f1": 0.3225806451612903, "prompt_id": "896dee39b103c9ac", "reading_canonical_log_prob": -41.56092834472656, "reading_distances": [27, 29, 30, 26], "reading_log_probs": [-39.758949279785156, -51.680572509765625, -53.69951629638672, -52.68844985961914], "reading_logprob_std": 6.517608844068366, "tokenization_path_logprob_std": 103.01270297155277, "trajectory_balance_rmse": 655.8345085340198} +{"canonical_length": 105, "canonical_nll": 1.7239445277622767, "generation_rouge_l_f1": 0.24109589041095889, "generation_token_f1": 0.39999999999999997, "prompt_id": "1057782952469d89", "reading_canonical_log_prob": -181.01417541503906, "reading_distances": [11, 10, 10, 9], "reading_log_probs": [-179.40664672851562, -179.97862243652344, -180.50527954101562, -177.5985107421875], "reading_logprob_std": 1.2647499833753364, "tokenization_path_logprob_std": 168.57889175622228, "trajectory_balance_rmse": 2732.571136076562} +{"canonical_length": 25, "canonical_nll": 0.6965357971191406, "generation_rouge_l_f1": 0.20809248554913296, "generation_token_f1": 0.21965317919075145, "prompt_id": "03db11358469fd8f", "reading_canonical_log_prob": -17.413394927978516, "reading_distances": [19, 20, 25, 24], "reading_log_probs": [-19.194459915161133, -18.53464698791504, -19.217178344726562, -18.631275177001953], "reading_logprob_std": 0.3618833671692821, "tokenization_path_logprob_std": 34.428213534320165, "trajectory_balance_rmse": 243.80705824270342} +{"canonical_length": 240, "canonical_nll": 1.1174840291341146, "generation_rouge_l_f1": 0.22587268993839832, "generation_token_f1": 0.5256673511293634, "prompt_id": "85339fcc838b71e5", "reading_canonical_log_prob": -268.1961669921875, "reading_distances": [21, 19, 22, 25], "reading_log_probs": [-263.58282470703125, -265.0533447265625, -263.9731750488281, -263.43548583984375], "reading_logprob_std": 0.730858790194629, "tokenization_path_logprob_std": 204.2435139071634, "trajectory_balance_rmse": 5849.702854272142} +{"canonical_length": 28, "canonical_nll": 0.35433203833443777, "generation_rouge_l_f1": 0.15228426395939088, "generation_token_f1": 0.17258883248730966, "prompt_id": "5af4615607f58b8b", "reading_canonical_log_prob": -9.921297073364258, "reading_distances": [47, 45, 38, 39], "reading_log_probs": [-10.50269889831543, -11.127426147460938, -10.637837409973145, -10.857181549072266], "reading_logprob_std": 0.27310645498227865, "tokenization_path_logprob_std": 92.62655418900624, "trajectory_balance_rmse": 661.0780169357694} +{"canonical_length": 65, "canonical_nll": 1.786942584698017, "generation_rouge_l_f1": 0.2622950819672131, "generation_token_f1": 0.3497267759562842, "prompt_id": "13f18463e2c1c5f7", "reading_canonical_log_prob": -116.1512680053711, "reading_distances": [25, 18, 20, 18], "reading_log_probs": [-115.63465881347656, -116.47900390625, -116.39500427246094, -115.83584594726562], "reading_logprob_std": 0.4148178562943694, "tokenization_path_logprob_std": 42.934301305956545, "trajectory_balance_rmse": 1459.6874230565184} +{"canonical_length": 64, "canonical_nll": 1.7973535060882568, "generation_rouge_l_f1": 0.15807560137457044, "generation_token_f1": 0.29553264604810997, "prompt_id": "d02d88b60737f21f", "reading_canonical_log_prob": -115.03062438964844, "reading_distances": [21, 23, 16, 19], "reading_log_probs": [-119.66471862792969, -116.27487182617188, -117.63359069824219, -117.1869888305664], "reading_logprob_std": 1.4327483666635068, "tokenization_path_logprob_std": 60.82058913943416, "trajectory_balance_rmse": 1597.7862384356442} +{"canonical_length": 236, "canonical_nll": 1.5001205185712394, "generation_rouge_l_f1": 0.20907297830374755, "generation_token_f1": 0.41025641025641024, "prompt_id": "559f28b026670bdd", "reading_canonical_log_prob": -354.0284423828125, "reading_distances": [19, 13, 16, 14], "reading_log_probs": [-354.158935546875, -351.98138427734375, -356.1494140625, -354.36163330078125], "reading_logprob_std": 1.7073136400872815, "tokenization_path_logprob_std": 79.88669982206595, "trajectory_balance_rmse": 5217.799605168986} +{"canonical_length": 50, "canonical_nll": 0.7350397491455078, "generation_rouge_l_f1": 0.5354330708661418, "generation_token_f1": 0.5984251968503937, "prompt_id": "44cafb7b8f27398a", "reading_canonical_log_prob": -36.75198745727539, "reading_distances": [30, 23, 28, 26], "reading_log_probs": [-35.921051025390625, -35.03411102294922, -36.158119201660156, -35.820411682128906], "reading_logprob_std": 0.48722869773222793, "tokenization_path_logprob_std": 91.92378211343635, "trajectory_balance_rmse": 1096.020215995606} +{"canonical_length": 274, "canonical_nll": 1.1070561095745894, "generation_rouge_l_f1": 0.22509225092250926, "generation_token_f1": 0.49446494464944646, "prompt_id": "55f3c7490c4e7162", "reading_canonical_log_prob": -303.3333740234375, "reading_distances": [14, 17, 22, 21], "reading_log_probs": [-299.87347412109375, -300.2261962890625, -298.67340087890625, -299.384033203125], "reading_logprob_std": 0.672653432058176, "tokenization_path_logprob_std": 223.5386127245687, "trajectory_balance_rmse": 5100.591577844237} +{"canonical_length": 308, "canonical_nll": 1.0252644922826197, "generation_rouge_l_f1": 0.21164021164021166, "generation_token_f1": 0.4409171075837743, "prompt_id": "41cd8f9fa13781b8", "reading_canonical_log_prob": -315.7814636230469, "reading_distances": [28, 27, 31, 17], "reading_log_probs": [-320.1492919921875, -320.6250915527344, -319.3072204589844, -320.0042724609375], "reading_logprob_std": 0.5450215906426558, "tokenization_path_logprob_std": 187.93494178765454, "trajectory_balance_rmse": 6618.7120665125285} +{"canonical_length": 373, "canonical_nll": 1.2313506507362182, "generation_rouge_l_f1": 0.2558922558922559, "generation_token_f1": 0.48484848484848486, "prompt_id": "28de6d5b6a032c7b", "reading_canonical_log_prob": -459.2937927246094, "reading_distances": [26, 21, 23, 25], "reading_log_probs": [-454.74090576171875, -457.8006591796875, -459.3896484375, -457.74444580078125], "reading_logprob_std": 1.94141037075135, "tokenization_path_logprob_std": 254.67228572137589, "trajectory_balance_rmse": 6411.798503539282} +{"canonical_length": 132, "canonical_nll": 1.617357196229877, "generation_rouge_l_f1": 0.24264705882352938, "generation_token_f1": 0.39705882352941174, "prompt_id": "dc17c375b90b9957", "reading_canonical_log_prob": -213.49114990234375, "reading_distances": [23, 24, 22, 18], "reading_log_probs": [-212.64559936523438, -213.11343383789062, -211.27828979492188, -212.55218505859375], "reading_logprob_std": 0.7854241837520247, "tokenization_path_logprob_std": 203.6466417716135, "trajectory_balance_rmse": 2962.5988958738953} +{"canonical_length": 24, "canonical_nll": 0.5077614386876425, "generation_rouge_l_f1": 0.6842105263157895, "generation_token_f1": 0.6842105263157895, "prompt_id": "c1a810543bd499fe", "reading_canonical_log_prob": -12.186274528503418, "reading_distances": [28, 25, 24, 27], "reading_log_probs": [-13.836664199829102, -12.646666526794434, -12.707374572753906, -12.406854629516602], "reading_logprob_std": 0.6381778997145139, "tokenization_path_logprob_std": 45.835166779939755, "trajectory_balance_rmse": 418.38035179669305} +{"canonical_length": 15, "canonical_nll": 1.1327279408772786, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.34782608695652173, "prompt_id": "f21bbbf6325c37df", "reading_canonical_log_prob": -16.99091911315918, "reading_distances": [42, 39, 38, 36], "reading_log_probs": [-18.71881103515625, -18.551963806152344, -18.420761108398438, -18.13282012939453], "reading_logprob_std": 0.24763256977420992, "tokenization_path_logprob_std": 64.3010641060434, "trajectory_balance_rmse": 342.5027242292171} +{"canonical_length": 141, "canonical_nll": 1.6061085369570036, "generation_rouge_l_f1": 0.1836228287841191, "generation_token_f1": 0.39702233250620345, "prompt_id": "ff640b48674add94", "reading_canonical_log_prob": -226.4613037109375, "reading_distances": [48, 54, 50, 41], "reading_log_probs": [-229.37460327148438, -227.76141357421875, -228.9608154296875, -227.1962127685547], "reading_logprob_std": 1.0161539365806131, "tokenization_path_logprob_std": 126.53443347091033, "trajectory_balance_rmse": 3249.978865465746} +{"canonical_length": 133, "canonical_nll": 1.9893518892445958, "generation_rouge_l_f1": 0.1568627450980392, "generation_token_f1": 0.27941176470588236, "prompt_id": "294ad70d174d121a", "reading_canonical_log_prob": -264.58380126953125, "reading_distances": [29, 22, 27, 28], "reading_log_probs": [-263.6551513671875, -264.132080078125, -264.89532470703125, -264.83233642578125], "reading_logprob_std": 0.5935854621081686, "tokenization_path_logprob_std": 314.8786403945749, "trajectory_balance_rmse": 3571.5389565423957} +{"canonical_length": 49, "canonical_nll": 2.5730422662228953, "generation_rouge_l_f1": 0.32876712328767116, "generation_token_f1": 0.4383561643835616, "prompt_id": "fdd72207f77ed671", "reading_canonical_log_prob": -126.07907104492188, "reading_distances": [19, 23, 20, 23], "reading_log_probs": [-125.50381469726562, -126.09919738769531, -128.28744506835938, -127.03130340576172], "reading_logprob_std": 1.2135156146118993, "tokenization_path_logprob_std": 78.68080616647734, "trajectory_balance_rmse": 1029.4774113904512} +{"canonical_length": 9, "canonical_nll": 1.4172704484727647, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.4210526315789474, "prompt_id": "bead008713b25f77", "reading_canonical_log_prob": -12.755434036254883, "reading_distances": [36, 42, 33, 40], "reading_log_probs": [-20.407493591308594, -18.917613983154297, -15.229562759399414, -19.063234329223633], "reading_logprob_std": 2.2203172958318107, "tokenization_path_logprob_std": 68.66098830170418, "trajectory_balance_rmse": 246.18786015456345} +{"canonical_length": 353, "canonical_nll": 0.9813970722490262, "generation_rouge_l_f1": 0.196875, "generation_token_f1": 0.375, "prompt_id": "5e76d06be1ed709a", "reading_canonical_log_prob": -346.43316650390625, "reading_distances": [31, 31, 23, 25], "reading_log_probs": [-345.8423767089844, -345.71630859375, -345.0445556640625, -346.9610595703125], "reading_logprob_std": 0.7946439007765683, "tokenization_path_logprob_std": 380.37053368608395, "trajectory_balance_rmse": 6535.456630323656} +{"canonical_length": 238, "canonical_nll": 1.3440144001936711, "generation_rouge_l_f1": 0.21973929236499068, "generation_token_f1": 0.42830540037243947, "prompt_id": "f6f871799dc3c062", "reading_canonical_log_prob": -319.87542724609375, "reading_distances": [30, 29, 31, 29], "reading_log_probs": [-318.888427734375, -317.6286926269531, -318.2723083496094, -310.3052062988281], "reading_logprob_std": 4.01207172307352, "tokenization_path_logprob_std": 192.22196097625712, "trajectory_balance_rmse": 3999.886849598548} +{"canonical_length": 108, "canonical_nll": 0.7908253846345125, "generation_rouge_l_f1": 0.3668639053254438, "generation_token_f1": 0.43786982248520717, "prompt_id": "e5243c35b3e882de", "reading_canonical_log_prob": -85.40914154052734, "reading_distances": [26, 27, 23, 32], "reading_log_probs": [-79.96489715576172, -81.52057647705078, -81.05591583251953, -81.70974731445312], "reading_logprob_std": 0.7817796315178046, "tokenization_path_logprob_std": 112.25903672116971, "trajectory_balance_rmse": 1617.4677800549973} +{"canonical_length": 147, "canonical_nll": 1.5584528916547087, "generation_rouge_l_f1": 0.20947630922693267, "generation_token_f1": 0.38902743142144636, "prompt_id": "26ddb945cf9f0d7d", "reading_canonical_log_prob": -229.0925750732422, "reading_distances": [33, 39, 29, 28], "reading_log_probs": [-230.7732391357422, -230.10975646972656, -229.4725341796875, -229.5787353515625], "reading_logprob_std": 0.5956926140352983, "tokenization_path_logprob_std": 222.70872611025177, "trajectory_balance_rmse": 3853.0609011685488} +{"canonical_length": 48, "canonical_nll": 1.5615088144938152, "generation_rouge_l_f1": 0.11515151515151516, "generation_token_f1": 0.15757575757575756, "prompt_id": "e4581a40938db0d9", "reading_canonical_log_prob": -74.95242309570312, "reading_distances": [19, 20, 15, 22], "reading_log_probs": [-79.25387573242188, -76.616943359375, -76.5678482055664, -79.68681335449219], "reading_logprob_std": 1.6710786666330657, "tokenization_path_logprob_std": 73.64191544764128, "trajectory_balance_rmse": 1073.8870774047339} +{"canonical_length": 10, "canonical_nll": 0.08137767910957336, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "267c9641a2f8cfb9", "reading_canonical_log_prob": -0.8137767910957336, "reading_distances": [22, 25, 25, 23], "reading_log_probs": [-1.3705188035964966, -1.3224700689315796, -1.5044606924057007, -1.2420499324798584], "reading_logprob_std": 0.11000028546589133, "tokenization_path_logprob_std": 51.85989340007153, "trajectory_balance_rmse": 196.37376759229053} +{"canonical_length": 63, "canonical_nll": 0.9030414762951079, "generation_rouge_l_f1": 0.31671554252199413, "generation_token_f1": 0.39882697947214074, "prompt_id": "df9c5775b476f764", "reading_canonical_log_prob": -56.8916130065918, "reading_distances": [29, 29, 30, 25], "reading_log_probs": [-56.38041687011719, -55.85192108154297, -54.5845947265625, -56.12939453125], "reading_logprob_std": 0.7977474133623234, "tokenization_path_logprob_std": 81.63681780627513, "trajectory_balance_rmse": 920.883713055888} +{"canonical_length": 106, "canonical_nll": 1.9022769568101414, "generation_rouge_l_f1": 0.19292604501607716, "generation_token_f1": 0.33440514469453375, "prompt_id": "518ad8fe3c294839", "reading_canonical_log_prob": -201.641357421875, "reading_distances": [18, 25, 21, 17], "reading_log_probs": [-203.72427368164062, -211.8428497314453, -206.8474578857422, -212.42225646972656], "reading_logprob_std": 4.160213259192452, "tokenization_path_logprob_std": 12.996157032314063, "trajectory_balance_rmse": 2362.475244761723} +{"canonical_length": 203, "canonical_nll": 0.7076416015625, "generation_rouge_l_f1": 0.3567839195979899, "generation_token_f1": 0.5175879396984925, "prompt_id": "6369662548b008b0", "reading_canonical_log_prob": -143.6512451171875, "reading_distances": [23, 23, 26, 23], "reading_log_probs": [-147.43692016601562, -147.9307098388672, -145.321044921875, -145.47047424316406], "reading_logprob_std": 1.3376940714243732, "tokenization_path_logprob_std": 133.51162355837502, "trajectory_balance_rmse": 2134.3942323332863} +{"canonical_length": 68, "canonical_nll": 0.40999844494987936, "generation_rouge_l_f1": 0.20645161290322583, "generation_token_f1": 0.27741935483870966, "prompt_id": "3f0899fb36d7151b", "reading_canonical_log_prob": -27.879894256591797, "reading_distances": [32, 35, 29, 37], "reading_log_probs": [-30.110151290893555, -29.918441772460938, -30.42481803894043, -30.88285255432129], "reading_logprob_std": 0.4212219961996743, "tokenization_path_logprob_std": 95.24856114500132, "trajectory_balance_rmse": 1339.3741080737595} +{"canonical_length": 59, "canonical_nll": 0.913433462886487, "generation_rouge_l_f1": 0.16363636363636364, "generation_token_f1": 0.3090909090909091, "prompt_id": "ca44a0ae267d5905", "reading_canonical_log_prob": -53.892574310302734, "reading_distances": [16, 19, 18, 20], "reading_log_probs": [-59.37114334106445, -60.212303161621094, -57.92009735107422, -59.53132629394531], "reading_logprob_std": 0.964051285190918, "tokenization_path_logprob_std": 113.92400752042231, "trajectory_balance_rmse": 1253.6426975514444} +{"canonical_length": 144, "canonical_nll": 2.5895457797580295, "generation_rouge_l_f1": 0.2637362637362637, "generation_token_f1": 0.40293040293040294, "prompt_id": "b3b2b51b9f629425", "reading_canonical_log_prob": -372.89459228515625, "reading_distances": [9, 12, 15, 11], "reading_log_probs": [-381.8276672363281, -391.2401428222656, -386.67388916015625, -386.93243408203125], "reading_logprob_std": 3.8472184834619796, "tokenization_path_logprob_std": 95.80172394734902, "trajectory_balance_rmse": 3071.5560956234985} +{"canonical_length": 120, "canonical_nll": 1.0189203898111978, "generation_rouge_l_f1": 0.05925925925925927, "generation_token_f1": 0.07407407407407407, "prompt_id": "4412ea6c558d478d", "reading_canonical_log_prob": -122.27044677734375, "reading_distances": [24, 26, 31, 24], "reading_log_probs": [-124.92556762695312, -123.42230224609375, -125.41567993164062, -125.02202606201172], "reading_logprob_std": 0.8754501394385397, "tokenization_path_logprob_std": 46.82058725987753, "trajectory_balance_rmse": 2550.5737672131118} +{"canonical_length": 30, "canonical_nll": 0.840692138671875, "generation_rouge_l_f1": 0.11764705882352942, "generation_token_f1": 0.2032085561497326, "prompt_id": "7938d8548e06b8bf", "reading_canonical_log_prob": -25.22076416015625, "reading_distances": [8, 9, 6, 8], "reading_log_probs": [-25.53807830810547, -25.077903747558594, -25.89090919494629, -25.923053741455078], "reading_logprob_std": 0.39377933612427096, "tokenization_path_logprob_std": 82.12111185189669, "trajectory_balance_rmse": 358.26198520407985} +{"canonical_length": 332, "canonical_nll": 1.1485015685299793, "generation_rouge_l_f1": 0.23423423423423423, "generation_token_f1": 0.46846846846846846, "prompt_id": "8829bdef994324cc", "reading_canonical_log_prob": -381.3025207519531, "reading_distances": [34, 33, 33, 35], "reading_log_probs": [-376.1643371582031, -375.7103271484375, -378.1880798339844, -377.07391357421875], "reading_logprob_std": 1.0942646514527166, "tokenization_path_logprob_std": 398.2287840561804, "trajectory_balance_rmse": 7226.315635125209} +{"canonical_length": 14, "canonical_nll": 1.0507426943097795, "generation_rouge_l_f1": 0.06694560669456066, "generation_token_f1": 0.06694560669456066, "prompt_id": "75996f4d9b7d5f4d", "reading_canonical_log_prob": -14.710397720336914, "reading_distances": [10, 18, 19, 23], "reading_log_probs": [-12.295442581176758, -11.95822811126709, -12.043508529663086, -11.990340232849121], "reading_logprob_std": 0.15313463952896256, "tokenization_path_logprob_std": 50.49109518154726, "trajectory_balance_rmse": 252.62258019819976} +{"canonical_length": 164, "canonical_nll": 1.2788962387457126, "generation_rouge_l_f1": 0.1826923076923077, "generation_token_f1": 0.3701923076923077, "prompt_id": "f5d8be75fb333542", "reading_canonical_log_prob": -209.73898315429688, "reading_distances": [18, 26, 19, 19], "reading_log_probs": [-211.38458251953125, -212.86016845703125, -210.6019287109375, -211.3157958984375], "reading_logprob_std": 0.948198825592488, "tokenization_path_logprob_std": 205.2891881981827, "trajectory_balance_rmse": 3935.2491298953714} +{"canonical_length": 45, "canonical_nll": 1.2966587490505643, "generation_rouge_l_f1": 0.14084507042253522, "generation_token_f1": 0.16901408450704225, "prompt_id": "61f6f6b3601f9e9d", "reading_canonical_log_prob": -58.34964370727539, "reading_distances": [19, 15, 27, 22], "reading_log_probs": [-59.17311096191406, -57.395912170410156, -60.21797180175781, -59.67002868652344], "reading_logprob_std": 1.222462170289498, "tokenization_path_logprob_std": 87.63437801603982, "trajectory_balance_rmse": 924.131577426799} +{"canonical_length": 26, "canonical_nll": 1.1841558309701772, "generation_rouge_l_f1": 0.2702702702702703, "generation_token_f1": 0.32432432432432434, "prompt_id": "4478590e9673bdfe", "reading_canonical_log_prob": -30.78805160522461, "reading_distances": [35, 29, 29, 31], "reading_log_probs": [-31.63953399658203, -31.300792694091797, -31.144325256347656, -30.416255950927734], "reading_logprob_std": 0.5158616389088941, "tokenization_path_logprob_std": 91.94041863127785, "trajectory_balance_rmse": 846.492680963468} +{"canonical_length": 90, "canonical_nll": 1.0725884331597222, "generation_rouge_l_f1": 0.13173652694610777, "generation_token_f1": 0.21556886227544908, "prompt_id": "b3f4ac8bd91cf595", "reading_canonical_log_prob": -96.532958984375, "reading_distances": [30, 32, 23, 32], "reading_log_probs": [-100.17247772216797, -99.67552947998047, -101.8968505859375, -101.07747650146484], "reading_logprob_std": 0.9836386072626512, "tokenization_path_logprob_std": 57.53865477521514, "trajectory_balance_rmse": 1851.943128774403} +{"canonical_length": 116, "canonical_nll": 1.0120197164601292, "generation_rouge_l_f1": 0.303030303030303, "generation_token_f1": 0.5378787878787877, "prompt_id": "47d49f308f57d626", "reading_canonical_log_prob": -117.394287109375, "reading_distances": [32, 35, 39, 36], "reading_log_probs": [-118.44026184082031, -120.03770446777344, -120.83557891845703, -116.4295654296875], "reading_logprob_std": 1.9450898164137949, "tokenization_path_logprob_std": 92.78949456819441, "trajectory_balance_rmse": 2881.497293651944} +{"canonical_length": 264, "canonical_nll": 1.2305658513849431, "generation_rouge_l_f1": 0.2142857142857143, "generation_token_f1": 0.48496240601503765, "prompt_id": "d8021c0fb98991e7", "reading_canonical_log_prob": -324.869384765625, "reading_distances": [15, 24, 20, 24], "reading_log_probs": [-326.29925537109375, -320.9302978515625, -322.3643798828125, -323.962158203125], "reading_logprob_std": 2.30167759746246, "tokenization_path_logprob_std": 221.63159185859575, "trajectory_balance_rmse": 5271.67476500118} +{"canonical_length": 165, "canonical_nll": 1.48824111476089, "generation_rouge_l_f1": 0.024096385542168672, "generation_token_f1": 0.024096385542168672, "prompt_id": "fb0e2e105f1d060e", "reading_canonical_log_prob": -245.55978393554688, "reading_distances": [35, 27, 23, 27], "reading_log_probs": [-242.78878784179688, -240.51620483398438, -239.9205322265625, -241.6036834716797], "reading_logprob_std": 1.2638047688893135, "tokenization_path_logprob_std": 153.52293218905754, "trajectory_balance_rmse": 3402.7711798217097} +{"canonical_length": 12, "canonical_nll": 0.606733481089274, "generation_rouge_l_f1": 0.5263157894736842, "generation_token_f1": 0.5263157894736842, "prompt_id": "d8467aeeb0933718", "reading_canonical_log_prob": -7.280801773071289, "reading_distances": [41, 42, 45, 36], "reading_log_probs": [-9.04745101928711, -9.304153442382812, -7.882077693939209, -7.240615367889404], "reading_logprob_std": 0.9738501308189923, "tokenization_path_logprob_std": 37.124845398307784, "trajectory_balance_rmse": 263.02313582235746} +{"canonical_length": 62, "canonical_nll": 0.4656058280698715, "generation_rouge_l_f1": 0.37190082644628103, "generation_token_f1": 0.42975206611570244, "prompt_id": "b80e624242a44cf6", "reading_canonical_log_prob": -28.86756134033203, "reading_distances": [23, 29, 22, 21], "reading_log_probs": [-31.631824493408203, -31.45306396484375, -30.791471481323242, -30.692195892333984], "reading_logprob_std": 0.46971009676486675, "tokenization_path_logprob_std": 135.29498079102166, "trajectory_balance_rmse": 1342.8018445302223} +{"canonical_length": 225, "canonical_nll": 1.1920035807291667, "generation_rouge_l_f1": 0.20085470085470084, "generation_token_f1": 0.30769230769230765, "prompt_id": "aad9f0c226359f45", "reading_canonical_log_prob": -268.2008056640625, "reading_distances": [13, 20, 17, 19], "reading_log_probs": [-270.41015625, -273.5290222167969, -271.8719787597656, -270.7245178222656], "reading_logprob_std": 1.410991899561621, "tokenization_path_logprob_std": 192.33430883585052, "trajectory_balance_rmse": 4631.318491128558} +{"canonical_length": 221, "canonical_nll": 0.9175152670743778, "generation_rouge_l_f1": 0.32377049180327866, "generation_token_f1": 0.5122950819672131, "prompt_id": "2fd9ba0f00f7f678", "reading_canonical_log_prob": -202.7708740234375, "reading_distances": [36, 26, 21, 28], "reading_log_probs": [-209.07357788085938, -206.96884155273438, -205.64129638671875, -205.23147583007812], "reading_logprob_std": 1.730149033688629, "tokenization_path_logprob_std": 254.28216648346853, "trajectory_balance_rmse": 4319.552760589546} +{"canonical_length": 8, "canonical_nll": 1.5873587131500244, "generation_rouge_l_f1": 0.02857142857142857, "generation_token_f1": 0.03571428571428571, "prompt_id": "b1a346c0437945a9", "reading_canonical_log_prob": -12.698869705200195, "reading_distances": [28, 31, 24, 31], "reading_log_probs": [-12.53235149383545, -12.877450942993164, -13.035024642944336, -12.973228454589844], "reading_logprob_std": 0.22434497073092835, "tokenization_path_logprob_std": 50.97875804013403, "trajectory_balance_rmse": 173.26602186458882} +{"canonical_length": 21, "canonical_nll": 2.3217867896670388, "generation_rouge_l_f1": 0.17301038062283738, "generation_token_f1": 0.18685121107266436, "prompt_id": "c2ce9d8e95420539", "reading_canonical_log_prob": -48.75752258300781, "reading_distances": [19, 16, 19, 22], "reading_log_probs": [-49.92293930053711, -48.18573760986328, -48.00699234008789, -48.55445861816406], "reading_logprob_std": 0.8674236866733006, "tokenization_path_logprob_std": 57.14362423792338, "trajectory_balance_rmse": 349.15645875590434} +{"canonical_length": 86, "canonical_nll": 1.2454849952875182, "generation_rouge_l_f1": 0.21203438395415475, "generation_token_f1": 0.3151862464183381, "prompt_id": "ecdc1eebf63862d2", "reading_canonical_log_prob": -107.11170959472656, "reading_distances": [32, 35, 28, 36], "reading_log_probs": [-110.8516616821289, -107.54605865478516, -111.37805938720703, -110.65503692626953], "reading_logprob_std": 1.7348271815413467, "tokenization_path_logprob_std": 146.76440646706982, "trajectory_balance_rmse": 2145.879384708199} +{"canonical_length": 149, "canonical_nll": 0.9015728227244127, "generation_rouge_l_f1": 0.21052631578947364, "generation_token_f1": 0.4578947368421052, "prompt_id": "a473b04dd30c4a84", "reading_canonical_log_prob": -134.3343505859375, "reading_distances": [50, 50, 49, 47], "reading_log_probs": [-132.1438446044922, -132.22979736328125, -132.4345703125, -135.0262451171875], "reading_logprob_std": 1.3838043070926294, "tokenization_path_logprob_std": 74.0815208543862, "trajectory_balance_rmse": 3704.33105160286} +{"canonical_length": 147, "canonical_nll": 1.5634800917437288, "generation_rouge_l_f1": 0.20708446866485014, "generation_token_f1": 0.40871934604904636, "prompt_id": "30cb3266dd228382", "reading_canonical_log_prob": -229.83157348632812, "reading_distances": [17, 20, 16, 12], "reading_log_probs": [-226.68206787109375, -227.14877319335938, -227.65859985351562, -226.82777404785156], "reading_logprob_std": 0.4326174267033747, "tokenization_path_logprob_std": 232.89756290445007, "trajectory_balance_rmse": 2691.0969936699016} +{"canonical_length": 151, "canonical_nll": 1.165701657731012, "generation_rouge_l_f1": 0.22712933753943215, "generation_token_f1": 0.5047318611987381, "prompt_id": "d49ee101a0892105", "reading_canonical_log_prob": -176.0209503173828, "reading_distances": [21, 17, 24, 23], "reading_log_probs": [-174.37559509277344, -176.34349060058594, -175.10037231445312, -176.19326782226562], "reading_logprob_std": 0.9338174094237552, "tokenization_path_logprob_std": 183.55520589473454, "trajectory_balance_rmse": 2774.304690741819} +{"canonical_length": 18, "canonical_nll": 0.844720204671224, "generation_rouge_l_f1": 0.09022556390977443, "generation_token_f1": 0.09774436090225563, "prompt_id": "c17572e82dc23485", "reading_canonical_log_prob": -15.204963684082031, "reading_distances": [26, 29, 24, 21], "reading_log_probs": [-14.31834888458252, -14.959285736083984, -16.135114669799805, -14.219205856323242], "reading_logprob_std": 0.881394056658179, "tokenization_path_logprob_std": 58.70403241833733, "trajectory_balance_rmse": 544.9222894938273} +{"canonical_length": 267, "canonical_nll": 1.411974503306414, "generation_rouge_l_f1": 0.23552123552123552, "generation_token_f1": 0.44787644787644787, "prompt_id": "50c4d271f8ba07ce", "reading_canonical_log_prob": -376.9971923828125, "reading_distances": [13, 16, 12, 13], "reading_log_probs": [-375.62762451171875, -376.5975646972656, -377.9434509277344, -376.4348449707031], "reading_logprob_std": 0.96043480649936, "tokenization_path_logprob_std": 170.70040678832424, "trajectory_balance_rmse": 5466.536847343834} +{"canonical_length": 27, "canonical_nll": 2.046004118742766, "generation_rouge_l_f1": 0.19607843137254902, "generation_token_f1": 0.23529411764705882, "prompt_id": "71a578aa99112e47", "reading_canonical_log_prob": -55.24211120605469, "reading_distances": [10, 13, 11, 16], "reading_log_probs": [-55.640625, -55.654014587402344, -55.87074661254883, -56.56348419189453], "reading_logprob_std": 0.4338584073687061, "tokenization_path_logprob_std": 42.29311319932069, "trajectory_balance_rmse": 717.2180827203908} +{"canonical_length": 260, "canonical_nll": 1.034362323467548, "generation_rouge_l_f1": 0.26285714285714284, "generation_token_f1": 0.5142857142857142, "prompt_id": "c22b91c04ec53dee", "reading_canonical_log_prob": -268.9342041015625, "reading_distances": [38, 39, 38, 43], "reading_log_probs": [-267.28753662109375, -267.354248046875, -268.1769714355469, -267.54205322265625], "reading_logprob_std": 0.405749214349223, "tokenization_path_logprob_std": 161.80263843908588, "trajectory_balance_rmse": 5640.587315040797} +{"canonical_length": 42, "canonical_nll": 1.6758261181059337, "generation_rouge_l_f1": 0.22950819672131148, "generation_token_f1": 0.29508196721311475, "prompt_id": "ce56aa7776fbfa4a", "reading_canonical_log_prob": -70.38469696044922, "reading_distances": [18, 19, 16, 16], "reading_log_probs": [-73.52606201171875, -70.51805114746094, -71.31646728515625, -69.07966613769531], "reading_logprob_std": 1.8576547405759547, "tokenization_path_logprob_std": 68.84048453162119, "trajectory_balance_rmse": 946.130183336851} +{"canonical_length": 92, "canonical_nll": 1.6428425001061482, "generation_rouge_l_f1": 0.02247191011235955, "generation_token_f1": 0.02247191011235955, "prompt_id": "61cafca95080c624", "reading_canonical_log_prob": -151.14151000976562, "reading_distances": [28, 27, 23, 27], "reading_log_probs": [-152.72792053222656, -149.945556640625, -147.64041137695312, -142.88958740234375], "reading_logprob_std": 4.164210829804694, "tokenization_path_logprob_std": 33.148980500880995, "trajectory_balance_rmse": 959.0111285147815} +{"canonical_length": 9, "canonical_nll": 1.7710721757676866, "generation_rouge_l_f1": 0.6153846153846153, "generation_token_f1": 0.7692307692307692, "prompt_id": "33cbcf32b9a54c98", "reading_canonical_log_prob": -15.93964958190918, "reading_distances": [24, 22, 17, 21], "reading_log_probs": [-15.89423942565918, -15.108848571777344, -15.05246353149414, -14.560853958129883], "reading_logprob_std": 0.5513997035087665, "tokenization_path_logprob_std": 26.476872986220677, "trajectory_balance_rmse": 172.1775764766754} +{"canonical_length": 13, "canonical_nll": 1.0280308356651893, "generation_rouge_l_f1": 0.09594095940959409, "generation_token_f1": 0.09594095940959409, "prompt_id": "5e0db48c41c73148", "reading_canonical_log_prob": -13.364400863647461, "reading_distances": [39, 37, 34, 28], "reading_log_probs": [-13.289483070373535, -13.247934341430664, -13.50258731842041, -13.103668212890625], "reading_logprob_std": 0.16494044025145904, "tokenization_path_logprob_std": 25.683562899020743, "trajectory_balance_rmse": 277.20121417239267} +{"canonical_length": 45, "canonical_nll": 1.5194178263346354, "generation_rouge_l_f1": 0.13058419243986252, "generation_token_f1": 0.18556701030927833, "prompt_id": "c70b50c5f56eb753", "reading_canonical_log_prob": -68.3738021850586, "reading_distances": [26, 33, 32, 25], "reading_log_probs": [-70.75971221923828, -72.27001953125, -75.72993469238281, -75.20805358886719], "reading_logprob_std": 2.3742945879106365, "tokenization_path_logprob_std": 101.01201799331552, "trajectory_balance_rmse": 909.2118624452922} +{"canonical_length": 28, "canonical_nll": 1.2557155064174108, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.2777777777777778, "prompt_id": "a944937cb2c4cc95", "reading_canonical_log_prob": -35.1600341796875, "reading_distances": [16, 13, 12, 14], "reading_log_probs": [-31.35089111328125, -31.621326446533203, -31.923095703125, -32.80696487426758], "reading_logprob_std": 0.6323719921047852, "tokenization_path_logprob_std": 64.92823676853128, "trajectory_balance_rmse": 701.5320127198222} +{"canonical_length": 158, "canonical_nll": 1.3027127422863924, "generation_rouge_l_f1": 0.24703087885985744, "generation_token_f1": 0.4085510688836105, "prompt_id": "b725cba739a4a837", "reading_canonical_log_prob": -205.82861328125, "reading_distances": [35, 37, 30, 31], "reading_log_probs": [-202.9168701171875, -204.5737762451172, -201.76248168945312, -203.96112060546875], "reading_logprob_std": 1.234272617632595, "tokenization_path_logprob_std": 114.45748270334005, "trajectory_balance_rmse": 3118.754312310051} +{"canonical_length": 501, "canonical_nll": 0.7372525456898703, "generation_rouge_l_f1": 0.19335347432024166, "generation_token_f1": 0.3444108761329305, "prompt_id": "c3741a95aac9caea", "reading_canonical_log_prob": -369.363525390625, "reading_distances": [33, 35, 25, 30], "reading_log_probs": [-371.93670654296875, -373.292724609375, -371.09716796875, -373.8726806640625], "reading_logprob_std": 1.263325456911513, "tokenization_path_logprob_std": 128.2887688419252, "trajectory_balance_rmse": 5588.622686098617} +{"canonical_length": 14, "canonical_nll": 0.8427217347281319, "generation_rouge_l_f1": 0.4888888888888889, "generation_token_f1": 0.4888888888888889, "prompt_id": "77d6fb3ebd2588d3", "reading_canonical_log_prob": -11.798104286193848, "reading_distances": [23, 18, 19, 26], "reading_log_probs": [-13.316052436828613, -13.347311019897461, -12.374031066894531, -12.853525161743164], "reading_logprob_std": 0.4585598623454374, "tokenization_path_logprob_std": 41.429531453411144, "trajectory_balance_rmse": 233.65636753374636} +{"canonical_length": 242, "canonical_nll": 1.366749408816503, "generation_rouge_l_f1": 0.2742616033755274, "generation_token_f1": 0.4388185654008439, "prompt_id": "8d10c80bdd9627ab", "reading_canonical_log_prob": -330.75335693359375, "reading_distances": [16, 20, 18, 18], "reading_log_probs": [-331.79779052734375, -331.8040771484375, -331.4797058105469, -331.36566162109375], "reading_logprob_std": 0.22330543673229622, "tokenization_path_logprob_std": 322.8616970617028, "trajectory_balance_rmse": 5539.004357747865} +{"canonical_length": 233, "canonical_nll": 1.2477634413559548, "generation_rouge_l_f1": 0.19461697722567287, "generation_token_f1": 0.463768115942029, "prompt_id": "a741547771fa6f03", "reading_canonical_log_prob": -290.7288818359375, "reading_distances": [22, 28, 26, 20], "reading_log_probs": [-287.688232421875, -287.70880126953125, -288.15606689453125, -287.96795654296875], "reading_logprob_std": 0.22363125427125227, "tokenization_path_logprob_std": 130.47240975026793, "trajectory_balance_rmse": 5158.420166869459} +{"canonical_length": 177, "canonical_nll": 1.0026774433373058, "generation_rouge_l_f1": 0.2540415704387991, "generation_token_f1": 0.46651270207852197, "prompt_id": "51a6d546f1967143", "reading_canonical_log_prob": -177.47390747070312, "reading_distances": [27, 33, 28, 28], "reading_log_probs": [-176.2642059326172, -176.9847412109375, -176.88507080078125, -178.79507446289062], "reading_logprob_std": 1.0895447999207457, "tokenization_path_logprob_std": 215.36251584427546, "trajectory_balance_rmse": 3332.367972983818} +{"canonical_length": 163, "canonical_nll": 1.125014977952454, "generation_rouge_l_f1": 0.24036281179138322, "generation_token_f1": 0.47619047619047616, "prompt_id": "715a0325621f71c4", "reading_canonical_log_prob": -183.37744140625, "reading_distances": [46, 42, 32, 36], "reading_log_probs": [-175.40298461914062, -176.13763427734375, -178.1328125, -176.55665588378906], "reading_logprob_std": 1.1533723268436908, "tokenization_path_logprob_std": 133.9570858085509, "trajectory_balance_rmse": 2811.119459125179} +{"canonical_length": 264, "canonical_nll": 1.2419239390980115, "generation_rouge_l_f1": 0.32000000000000006, "generation_token_f1": 0.48, "prompt_id": "0857b1f32ec14a13", "reading_canonical_log_prob": -327.867919921875, "reading_distances": [17, 23, 21, 17], "reading_log_probs": [-332.6920166015625, -332.277099609375, -329.45654296875, -331.501220703125], "reading_logprob_std": 1.4374942170420715, "tokenization_path_logprob_std": 87.23055598423001, "trajectory_balance_rmse": 4593.606244579136} +{"canonical_length": 2, "canonical_nll": 4.131918430328369, "generation_rouge_l_f1": 0.008032128514056224, "generation_token_f1": 0.008032128514056224, "prompt_id": "7c6d4e136a035811", "reading_canonical_log_prob": -8.263836860656738, "reading_distances": [65, 71, 83, 67], "reading_log_probs": [-9.291272163391113, -9.385777473449707, -8.847376823425293, -7.676666259765625], "reading_logprob_std": 0.7849854656322544, "tokenization_path_logprob_std": 0.0, "trajectory_balance_rmse": 8.34771728515625} +{"canonical_length": 13, "canonical_nll": 1.0124020209679236, "generation_rouge_l_f1": 0.5454545454545454, "generation_token_f1": 0.7272727272727272, "prompt_id": "d3b27b0b1b980505", "reading_canonical_log_prob": -13.161226272583008, "reading_distances": [31, 27, 34, 37], "reading_log_probs": [-13.942813873291016, -13.492536544799805, -13.892474174499512, -14.145306587219238], "reading_logprob_std": 0.27329320668977874, "tokenization_path_logprob_std": 20.141002447483004, "trajectory_balance_rmse": 234.68905643384537} +{"canonical_length": 131, "canonical_nll": 0.4824161092743619, "generation_rouge_l_f1": 0.33532934131736525, "generation_token_f1": 0.4491017964071856, "prompt_id": "37ec341c729bc100", "reading_canonical_log_prob": -63.196510314941406, "reading_distances": [9, 8, 14, 7], "reading_log_probs": [-59.66399002075195, -61.54206085205078, -61.71465301513672, -59.95265579223633], "reading_logprob_std": 1.0597298163514914, "tokenization_path_logprob_std": 93.84148604632068, "trajectory_balance_rmse": 1417.4371531683398} +{"canonical_length": 12, "canonical_nll": 3.5338242848714194, "generation_rouge_l_f1": 0.3, "generation_token_f1": 0.3, "prompt_id": "b0f40cc1f3136930", "reading_canonical_log_prob": -42.40589141845703, "reading_distances": [10, 12, 9, 13], "reading_log_probs": [-40.850791931152344, -40.74162673950195, -41.44200897216797, -41.846336364746094], "reading_logprob_std": 0.5185687527741738, "tokenization_path_logprob_std": 48.87282619494272, "trajectory_balance_rmse": 303.629083030594} +{"canonical_length": 264, "canonical_nll": 1.2168119025952888, "generation_rouge_l_f1": 0.26153846153846155, "generation_token_f1": 0.5346153846153846, "prompt_id": "dc37089a4982f612", "reading_canonical_log_prob": -321.23834228515625, "reading_distances": [16, 16, 14, 14], "reading_log_probs": [-328.82720947265625, -326.5777587890625, -323.84747314453125, -321.6884765625], "reading_logprob_std": 3.1203647819746463, "tokenization_path_logprob_std": 279.529747831439, "trajectory_balance_rmse": 5584.009572748956} +{"canonical_length": 3, "canonical_nll": 1.2652823130289714, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "9ba6dc4bdecfc908", "reading_canonical_log_prob": -3.795846939086914, "reading_distances": [24, 25, 31, 15], "reading_log_probs": [-3.5938148498535156, -3.645789384841919, -3.4062283039093018, -3.338818311691284], "reading_logprob_std": 0.14693483739426616, "tokenization_path_logprob_std": 12.718398213386536, "trajectory_balance_rmse": 14.467105661948782} +{"canonical_length": 261, "canonical_nll": 0.9600741214679118, "generation_rouge_l_f1": 0.44234404536862004, "generation_token_f1": 0.6011342155009451, "prompt_id": "5cf08716fd03e235", "reading_canonical_log_prob": -250.579345703125, "reading_distances": [30, 30, 30, 31], "reading_log_probs": [-249.02529907226562, -249.0052490234375, -249.90847778320312, -250.19317626953125], "reading_logprob_std": 0.6091243955056445, "tokenization_path_logprob_std": 95.81865106648904, "trajectory_balance_rmse": 4604.52325089578} +{"canonical_length": 102, "canonical_nll": 0.8789396846995634, "generation_rouge_l_f1": 0.3357664233576642, "generation_token_f1": 0.49635036496350365, "prompt_id": "8cfc1971edf74b1e", "reading_canonical_log_prob": -89.65184783935547, "reading_distances": [38, 31, 32, 39], "reading_log_probs": [-90.41163635253906, -94.515869140625, -88.54924011230469, -89.90341186523438], "reading_logprob_std": 2.570355745146089, "tokenization_path_logprob_std": 53.57282270169484, "trajectory_balance_rmse": 1928.7428449599358} +{"canonical_length": 42, "canonical_nll": 2.7169732593354725, "generation_rouge_l_f1": 0.15789473684210525, "generation_token_f1": 0.2631578947368421, "prompt_id": "73cb687de2d9ad9d", "reading_canonical_log_prob": -114.11287689208984, "reading_distances": [24, 27, 23, 23], "reading_log_probs": [-114.37176513671875, -115.48785400390625, -113.67167663574219, -114.1795654296875], "reading_logprob_std": 0.7659846195758045, "tokenization_path_logprob_std": 46.53720865047792, "trajectory_balance_rmse": 862.1994843894786} +{"canonical_length": 50, "canonical_nll": 1.1126780700683594, "generation_rouge_l_f1": 0.32380952380952377, "generation_token_f1": 0.4000000000000001, "prompt_id": "5b7c88eabb38a156", "reading_canonical_log_prob": -55.63390350341797, "reading_distances": [29, 26, 30, 26], "reading_log_probs": [-53.07207489013672, -53.18385314941406, -53.153350830078125, -52.97319412231445], "reading_logprob_std": 0.09426983453594946, "tokenization_path_logprob_std": 68.71403700003283, "trajectory_balance_rmse": 1050.6098076627036} +{"canonical_length": 297, "canonical_nll": 1.9038213350957491, "generation_rouge_l_f1": 0.24489795918367346, "generation_token_f1": 0.43042671614100186, "prompt_id": "7564700da05daae9", "reading_canonical_log_prob": -565.4349365234375, "reading_distances": [25, 22, 21, 30], "reading_log_probs": [-564.8405151367188, -564.0363159179688, -564.499755859375, -563.7012329101562], "reading_logprob_std": 0.502121612272985, "tokenization_path_logprob_std": 313.4175526045963, "trajectory_balance_rmse": 6142.048279287292} +{"canonical_length": 240, "canonical_nll": 1.4308275858561197, "generation_rouge_l_f1": 0.27089337175792505, "generation_token_f1": 0.4553314121037464, "prompt_id": "ca2d7d9e3b473a6f", "reading_canonical_log_prob": -343.39862060546875, "reading_distances": [17, 14, 21, 13], "reading_log_probs": [-348.83380126953125, -348.07135009765625, -347.83685302734375, -348.77691650390625], "reading_logprob_std": 0.5012489096609071, "tokenization_path_logprob_std": 204.56771622431214, "trajectory_balance_rmse": 5854.174844004011} +{"canonical_length": 282, "canonical_nll": 1.1784910377881206, "generation_rouge_l_f1": 0.30833333333333335, "generation_token_f1": 0.4, "prompt_id": "9e92f8d560ad3563", "reading_canonical_log_prob": -332.33447265625, "reading_distances": [22, 25, 24, 15], "reading_log_probs": [-337.18585205078125, -337.56378173828125, -336.7627868652344, -342.05364990234375], "reading_logprob_std": 2.463246697591611, "tokenization_path_logprob_std": 141.39784179832645, "trajectory_balance_rmse": 4416.366289513414} +{"canonical_length": 16, "canonical_nll": 1.9503144025802612, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "8efbf41d41230763", "reading_canonical_log_prob": -31.20503044128418, "reading_distances": [37, 36, 42, 28], "reading_log_probs": [-33.24768829345703, -30.422513961791992, -29.98388671875, -30.50354766845703], "reading_logprob_std": 1.4897798024020024, "tokenization_path_logprob_std": 40.942234426448266, "trajectory_balance_rmse": 429.399849911871} +{"canonical_length": 39, "canonical_nll": 2.148041553986378, "generation_rouge_l_f1": 0.15384615384615383, "generation_token_f1": 0.20512820512820512, "prompt_id": "3f9b43f9811260d6", "reading_canonical_log_prob": -83.77362060546875, "reading_distances": [32, 31, 28, 35], "reading_log_probs": [-87.09796142578125, -92.24406433105469, -90.00657653808594, -84.348876953125], "reading_logprob_std": 3.4381412416566284, "tokenization_path_logprob_std": 86.47469823010793, "trajectory_balance_rmse": 755.2381041709019} +{"canonical_length": 281, "canonical_nll": 0.972937641618939, "generation_rouge_l_f1": 0.21217391304347827, "generation_token_f1": 0.33739130434782605, "prompt_id": "9e7206bd2d909c4b", "reading_canonical_log_prob": -273.3954772949219, "reading_distances": [28, 37, 33, 32], "reading_log_probs": [-270.34075927734375, -270.62420654296875, -270.2328796386719, -269.41778564453125], "reading_logprob_std": 0.5177537256581554, "tokenization_path_logprob_std": 69.97044472731204, "trajectory_balance_rmse": 4552.766319971357} +{"canonical_length": 193, "canonical_nll": 1.2633976121022912, "generation_rouge_l_f1": 0.3829787234042554, "generation_token_f1": 0.5372340425531914, "prompt_id": "59cc1c6a894af7a8", "reading_canonical_log_prob": -243.8357391357422, "reading_distances": [20, 23, 27, 23], "reading_log_probs": [-242.81655883789062, -242.13375854492188, -242.01727294921875, -244.53831481933594], "reading_logprob_std": 1.1626345697475635, "tokenization_path_logprob_std": 112.83369275868829, "trajectory_balance_rmse": 3801.7153693218606} +{"canonical_length": 85, "canonical_nll": 1.0295807782341453, "generation_rouge_l_f1": 0.2251655629139073, "generation_token_f1": 0.4039735099337748, "prompt_id": "bb9e06240071dbbd", "reading_canonical_log_prob": -87.51436614990234, "reading_distances": [25, 25, 20, 24], "reading_log_probs": [-88.67900085449219, -87.89889526367188, -90.246826171875, -88.4102554321289], "reading_logprob_std": 1.0118516819165542, "tokenization_path_logprob_std": 49.988799609570776, "trajectory_balance_rmse": 1779.595098506121} +{"canonical_length": 59, "canonical_nll": 0.8603322950460143, "generation_rouge_l_f1": 0.19762845849802368, "generation_token_f1": 0.26877470355731226, "prompt_id": "fbd2f42b1f0643a2", "reading_canonical_log_prob": -50.759605407714844, "reading_distances": [17, 17, 13, 17], "reading_log_probs": [-50.23956298828125, -50.071739196777344, -50.52967834472656, -50.41325759887695], "reading_logprob_std": 0.20049905133845325, "tokenization_path_logprob_std": 71.23306473423457, "trajectory_balance_rmse": 981.6821479879709} +{"canonical_length": 73, "canonical_nll": 0.7807519991103917, "generation_rouge_l_f1": 0.21804511278195488, "generation_token_f1": 0.3157894736842105, "prompt_id": "1a0ff319dcddb777", "reading_canonical_log_prob": -56.994895935058594, "reading_distances": [48, 48, 55, 41], "reading_log_probs": [-56.44377136230469, -56.123268127441406, -57.32313537597656, -55.39978790283203], "reading_logprob_std": 0.7973014594927402, "tokenization_path_logprob_std": 102.74613454271616, "trajectory_balance_rmse": 680.5283848569118} +{"canonical_length": 15, "canonical_nll": 0.6471051534016927, "generation_rouge_l_f1": 0.24999999999999994, "generation_token_f1": 0.33333333333333337, "prompt_id": "9f8d22c249bf7888", "reading_canonical_log_prob": -9.70657730102539, "reading_distances": [20, 21, 24, 21], "reading_log_probs": [-9.350162506103516, -9.251385688781738, -9.485176086425781, -9.891570091247559], "reading_logprob_std": 0.28147791596460436, "tokenization_path_logprob_std": 33.41644159210753, "trajectory_balance_rmse": 346.8760725167218} +{"canonical_length": 168, "canonical_nll": 1.2399677094959078, "generation_rouge_l_f1": 0.2359882005899705, "generation_token_f1": 0.3775811209439528, "prompt_id": "9edad123b1e31f1d", "reading_canonical_log_prob": -208.3145751953125, "reading_distances": [28, 29, 28, 31], "reading_log_probs": [-206.161376953125, -207.14736938476562, -209.33120727539062, -205.19419860839844], "reading_logprob_std": 1.7714001999987639, "tokenization_path_logprob_std": 245.04234998541904, "trajectory_balance_rmse": 3425.5713812963345} +{"canonical_length": 82, "canonical_nll": 1.1608827172256098, "generation_rouge_l_f1": 0.19540229885057472, "generation_token_f1": 0.3045977011494253, "prompt_id": "e027ef71856c6f38", "reading_canonical_log_prob": -95.1923828125, "reading_distances": [29, 27, 25, 26], "reading_log_probs": [-101.7489013671875, -100.02057647705078, -99.36566925048828, -97.68510437011719], "reading_logprob_std": 1.6805004223974418, "tokenization_path_logprob_std": 85.08163947569567, "trajectory_balance_rmse": 2111.638590122513} +{"canonical_length": 27, "canonical_nll": 2.437202736183449, "generation_rouge_l_f1": 0.18749999999999997, "generation_token_f1": 0.21875, "prompt_id": "59a16e60dba35dc9", "reading_canonical_log_prob": -65.80447387695312, "reading_distances": [15, 14, 18, 15], "reading_log_probs": [-66.15371704101562, -65.65562438964844, -65.50320434570312, -67.6900634765625], "reading_logprob_std": 0.9990057787833092, "tokenization_path_logprob_std": 56.44419244850362, "trajectory_balance_rmse": 686.3562061820044} +{"canonical_length": 16, "canonical_nll": 0.3481554687023163, "generation_rouge_l_f1": 0.0851063829787234, "generation_token_f1": 0.0851063829787234, "prompt_id": "ca87800b07a107f6", "reading_canonical_log_prob": -5.5704874992370605, "reading_distances": [27, 32, 30, 26], "reading_log_probs": [-5.79811954498291, -5.38165283203125, -5.865479946136475, -5.67350959777832], "reading_logprob_std": 0.2140154294062194, "tokenization_path_logprob_std": 11.027490321443215, "trajectory_balance_rmse": 246.16053387003484} +{"canonical_length": 100, "canonical_nll": 1.5095748901367188, "generation_rouge_l_f1": 0.2017291066282421, "generation_token_f1": 0.3112391930835735, "prompt_id": "911e3e27940b6a5a", "reading_canonical_log_prob": -150.95748901367188, "reading_distances": [29, 27, 23, 30], "reading_log_probs": [-149.4145050048828, -150.932373046875, -150.16696166992188, -149.681396484375], "reading_logprob_std": 0.6663279956095947, "tokenization_path_logprob_std": 182.48753859431602, "trajectory_balance_rmse": 2128.3875239258955} +{"canonical_length": 190, "canonical_nll": 2.14306929738898, "generation_rouge_l_f1": 0.1934065934065934, "generation_token_f1": 0.3912087912087912, "prompt_id": "1e23f4d5f326fadb", "reading_canonical_log_prob": -407.18316650390625, "reading_distances": [23, 25, 31, 29], "reading_log_probs": [-400.6434020996094, -400.1880798339844, -403.5838623046875, -403.34130859375], "reading_logprob_std": 1.7716599216738653, "tokenization_path_logprob_std": 195.23541197026285, "trajectory_balance_rmse": 4431.500223231049} +{"canonical_length": 151, "canonical_nll": 1.4833770145643626, "generation_rouge_l_f1": 0.17391304347826086, "generation_token_f1": 0.43961352657004826, "prompt_id": "75c88d19daf68f00", "reading_canonical_log_prob": -223.98992919921875, "reading_distances": [18, 19, 24, 20], "reading_log_probs": [-223.10064697265625, -224.7630157470703, -220.53485107421875, -221.57879638671875], "reading_logprob_std": 1.8432141083711697, "tokenization_path_logprob_std": 182.43095706068132, "trajectory_balance_rmse": 3836.8902348232004} +{"canonical_length": 165, "canonical_nll": 1.3460382634943182, "generation_rouge_l_f1": 0.2462686567164179, "generation_token_f1": 0.44029850746268656, "prompt_id": "0efac862acbbc3cf", "reading_canonical_log_prob": -222.0963134765625, "reading_distances": [28, 33, 24, 31], "reading_log_probs": [-226.14309692382812, -224.73236083984375, -223.66497802734375, -224.32864379882812], "reading_logprob_std": 1.047466605691145, "tokenization_path_logprob_std": 141.67822712910848, "trajectory_balance_rmse": 3948.430271331689} +{"canonical_length": 225, "canonical_nll": 0.8601558430989583, "generation_rouge_l_f1": 0.27751196172248804, "generation_token_f1": 0.4641148325358852, "prompt_id": "867d14d7a7b70c9c", "reading_canonical_log_prob": -193.53506469726562, "reading_distances": [14, 16, 20, 20], "reading_log_probs": [-197.8153076171875, -193.457763671875, -192.95639038085938, -197.80975341796875], "reading_logprob_std": 2.6668273830367184, "tokenization_path_logprob_std": 79.69036939291843, "trajectory_balance_rmse": 3935.082070910413} +{"canonical_length": 263, "canonical_nll": 1.497095376366445, "generation_rouge_l_f1": 0.14084507042253525, "generation_token_f1": 0.24144869215291748, "prompt_id": "67b31fbc0903ce67", "reading_canonical_log_prob": -393.736083984375, "reading_distances": [27, 20, 20, 14], "reading_log_probs": [-395.43182373046875, -395.8636474609375, -392.11334228515625, -395.4452209472656], "reading_logprob_std": 1.7449989350606152, "tokenization_path_logprob_std": 283.5028222052269, "trajectory_balance_rmse": 5615.865400972925} +{"canonical_length": 38, "canonical_nll": 0.5569534301757812, "generation_rouge_l_f1": 0.17391304347826086, "generation_token_f1": 0.20289855072463767, "prompt_id": "c50601d0ada030f4", "reading_canonical_log_prob": -21.164230346679688, "reading_distances": [28, 26, 24, 26], "reading_log_probs": [-19.175533294677734, -21.057565689086914, -21.593887329101562, -22.518566131591797], "reading_logprob_std": 1.409596419796209, "tokenization_path_logprob_std": 48.01002840530109, "trajectory_balance_rmse": 265.1754789780037} +{"canonical_length": 35, "canonical_nll": 0.6718268803187779, "generation_rouge_l_f1": 0.5263157894736842, "generation_token_f1": 0.631578947368421, "prompt_id": "d0d4fbc6a94b1aa0", "reading_canonical_log_prob": -23.513940811157227, "reading_distances": [21, 19, 20, 21], "reading_log_probs": [-19.6982421875, -19.719646453857422, -19.507152557373047, -19.50940704345703], "reading_logprob_std": 0.11618643402674105, "tokenization_path_logprob_std": 89.83560988601897, "trajectory_balance_rmse": 620.5001036416863} +{"canonical_length": 74, "canonical_nll": 0.5306547525766734, "generation_rouge_l_f1": 0.23036649214659688, "generation_token_f1": 0.32460732984293195, "prompt_id": "a83eef42db047836", "reading_canonical_log_prob": -39.26845169067383, "reading_distances": [20, 25, 26, 29], "reading_log_probs": [-39.285701751708984, -40.31233596801758, -37.69457244873047, -38.20525360107422], "reading_logprob_std": 1.1657013504916165, "tokenization_path_logprob_std": 87.32118648948068, "trajectory_balance_rmse": 1246.5361156313106} +{"canonical_length": 107, "canonical_nll": 1.1700811653493721, "generation_rouge_l_f1": 0.19318181818181818, "generation_token_f1": 0.27272727272727276, "prompt_id": "8e875ab98664b402", "reading_canonical_log_prob": -125.19868469238281, "reading_distances": [18, 17, 18, 18], "reading_log_probs": [-127.32347106933594, -126.48704528808594, -126.11054229736328, -125.81681823730469], "reading_logprob_std": 0.6530722059852887, "tokenization_path_logprob_std": 114.07018494647329, "trajectory_balance_rmse": 2145.1764520906268} +{"canonical_length": 31, "canonical_nll": 0.7504790521437122, "generation_rouge_l_f1": 0.31999999999999995, "generation_token_f1": 0.42, "prompt_id": "54f5b26b0d406b0a", "reading_canonical_log_prob": -23.264850616455078, "reading_distances": [29, 31, 27, 34], "reading_log_probs": [-28.555952072143555, -28.30392837524414, -25.66415786743164, -28.395597457885742], "reading_logprob_std": 1.3811003179665335, "tokenization_path_logprob_std": 66.92785268729223, "trajectory_balance_rmse": 670.6071179061338} +{"canonical_length": 262, "canonical_nll": 1.1952871075113312, "generation_rouge_l_f1": 0.2578268876611418, "generation_token_f1": 0.4972375690607735, "prompt_id": "652a0c6a273ec7b9", "reading_canonical_log_prob": -313.16522216796875, "reading_distances": [16, 17, 13, 13], "reading_log_probs": [-314.00543212890625, -312.9806213378906, -313.860595703125, -311.9376220703125], "reading_logprob_std": 0.9533748447037546, "tokenization_path_logprob_std": 197.7661717857298, "trajectory_balance_rmse": 4551.575187625263} +{"canonical_length": 304, "canonical_nll": 1.6112215142501027, "generation_rouge_l_f1": 0.14855072463768118, "generation_token_f1": 0.2789855072463768, "prompt_id": "13a435b70dd47ea2", "reading_canonical_log_prob": -489.81134033203125, "reading_distances": [23, 14, 17, 11], "reading_log_probs": [-489.1878356933594, -490.72857666015625, -489.7630920410156, -489.49993896484375], "reading_logprob_std": 0.6654053728113711, "tokenization_path_logprob_std": 572.4722146281096, "trajectory_balance_rmse": 7238.468137877646} +{"canonical_length": 15, "canonical_nll": 0.6131463368733724, "generation_rouge_l_f1": 0.1509433962264151, "generation_token_f1": 0.18867924528301885, "prompt_id": "1f8b04daa461e185", "reading_canonical_log_prob": -9.197195053100586, "reading_distances": [24, 18, 25, 23], "reading_log_probs": [-10.352699279785156, -8.628887176513672, -9.352375030517578, -9.431885719299316], "reading_logprob_std": 0.7067907321401319, "tokenization_path_logprob_std": 19.013380859281803, "trajectory_balance_rmse": 211.6047295441682} +{"canonical_length": 18, "canonical_nll": 0.3103417290581597, "generation_rouge_l_f1": 0.3529411764705882, "generation_token_f1": 0.37647058823529417, "prompt_id": "713c6185b310a4ca", "reading_canonical_log_prob": -5.586151123046875, "reading_distances": [14, 27, 24, 30], "reading_log_probs": [-7.127962112426758, -7.960738658905029, -7.597780227661133, -7.155205726623535], "reading_logprob_std": 0.3970183002762479, "tokenization_path_logprob_std": 62.451757242793256, "trajectory_balance_rmse": 413.0482792576194} +{"canonical_length": 227, "canonical_nll": 0.765635486216272, "generation_rouge_l_f1": 0.2740286298568507, "generation_token_f1": 0.4826175869120654, "prompt_id": "a7de6528410fa4b7", "reading_canonical_log_prob": -173.79925537109375, "reading_distances": [21, 28, 25, 25], "reading_log_probs": [-175.19602966308594, -172.61697387695312, -173.9869842529297, -172.48538208007812], "reading_logprob_std": 1.2783447561812078, "tokenization_path_logprob_std": 109.54621267154775, "trajectory_balance_rmse": 3419.812033822188} +{"canonical_length": 154, "canonical_nll": 1.7342267717633928, "generation_rouge_l_f1": 0.22815533980582522, "generation_token_f1": 0.4077669902912622, "prompt_id": "b7245c913ba925d6", "reading_canonical_log_prob": -267.0709228515625, "reading_distances": [22, 28, 21, 28], "reading_log_probs": [-268.42266845703125, -268.02825927734375, -268.1127624511719, -267.408935546875], "reading_logprob_std": 0.42478717488331874, "tokenization_path_logprob_std": 168.38511267946896, "trajectory_balance_rmse": 3184.6618245679274} +{"canonical_length": 9, "canonical_nll": 0.473663330078125, "generation_rouge_l_f1": 0.8750000000000001, "generation_token_f1": 0.8750000000000001, "prompt_id": "402312ec99016f32", "reading_canonical_log_prob": -4.262969970703125, "reading_distances": [23, 21, 22, 20], "reading_log_probs": [-4.925297737121582, -4.937930583953857, -5.5445237159729, -5.887803077697754], "reading_logprob_std": 0.47417214127284135, "tokenization_path_logprob_std": 39.95493375449841, "trajectory_balance_rmse": 178.7487433913308} +{"canonical_length": 290, "canonical_nll": 1.418047780004041, "generation_rouge_l_f1": 0.23843416370106763, "generation_token_f1": 0.46619217081850534, "prompt_id": "083ca3a9f1344c38", "reading_canonical_log_prob": -411.2338562011719, "reading_distances": [30, 35, 32, 31], "reading_log_probs": [-413.25048828125, -414.3248291015625, -414.31658935546875, -414.34063720703125], "reading_logprob_std": 0.5385242541583042, "tokenization_path_logprob_std": 268.50222973616223, "trajectory_balance_rmse": 6242.139899622332} +{"canonical_length": 58, "canonical_nll": 1.5810846789129849, "generation_rouge_l_f1": 0.18888888888888886, "generation_token_f1": 0.34444444444444444, "prompt_id": "c06dd3bfbe8f1823", "reading_canonical_log_prob": -91.70291137695312, "reading_distances": [35, 29, 32, 34], "reading_log_probs": [-94.39746856689453, -95.82443237304688, -94.77114868164062, -95.08313751220703], "reading_logprob_std": 0.6056859795911524, "tokenization_path_logprob_std": 127.81109436125637, "trajectory_balance_rmse": 1235.0592141119193} +{"canonical_length": 117, "canonical_nll": 0.8531590649205395, "generation_rouge_l_f1": 0.22589531680440775, "generation_token_f1": 0.37465564738292007, "prompt_id": "22135eb61f4e8a41", "reading_canonical_log_prob": -99.81961059570312, "reading_distances": [23, 16, 22, 21], "reading_log_probs": [-101.92276763916016, -101.4326171875, -103.33421325683594, -102.37639617919922], "reading_logprob_std": 0.8094445111492624, "tokenization_path_logprob_std": 126.47472811253898, "trajectory_balance_rmse": 2408.811042672501} +{"canonical_length": 21, "canonical_nll": 1.7638458978562128, "generation_rouge_l_f1": 0.13793103448275862, "generation_token_f1": 0.13793103448275862, "prompt_id": "b36f1ed6b55c0058", "reading_canonical_log_prob": -37.04076385498047, "reading_distances": [27, 26, 24, 28], "reading_log_probs": [-39.43168258666992, -42.02682113647461, -41.26850128173828, -40.33931350708008], "reading_logprob_std": 1.1261498501844656, "tokenization_path_logprob_std": 82.24504767198395, "trajectory_balance_rmse": 465.54464191472147} +{"canonical_length": 25, "canonical_nll": 1.6010659790039063, "generation_rouge_l_f1": 0.2285714285714286, "generation_token_f1": 0.4571428571428572, "prompt_id": "017529d62a606652", "reading_canonical_log_prob": -40.026649475097656, "reading_distances": [25, 19, 26, 27], "reading_log_probs": [-39.798683166503906, -40.51521682739258, -40.28358840942383, -40.412662506103516], "reading_logprob_std": 0.317063800954998, "tokenization_path_logprob_std": 109.95471310098274, "trajectory_balance_rmse": 524.6710131277383} +{"canonical_length": 231, "canonical_nll": 1.585124226359578, "generation_rouge_l_f1": 0.008298755186721992, "generation_token_f1": 0.008298755186721992, "prompt_id": "47cd1ce3b4c7133a", "reading_canonical_log_prob": -366.1636962890625, "reading_distances": [15, 12, 12, 9], "reading_log_probs": [-372.041015625, -371.58380126953125, -372.29150390625, -371.95965576171875], "reading_logprob_std": 0.2930644959715147, "tokenization_path_logprob_std": 123.36115408273443, "trajectory_balance_rmse": 5355.69432404264} +{"canonical_length": 34, "canonical_nll": 0.7417945861816406, "generation_rouge_l_f1": 0.15217391304347824, "generation_token_f1": 0.1884057971014493, "prompt_id": "549cddbffb2c458b", "reading_canonical_log_prob": -25.22101593017578, "reading_distances": [11, 13, 17, 10], "reading_log_probs": [-25.327632904052734, -25.802391052246094, -25.695547103881836, -25.312795639038086], "reading_logprob_std": 0.2514283005685404, "tokenization_path_logprob_std": 106.32691849215215, "trajectory_balance_rmse": 513.2747644176277} +{"canonical_length": 356, "canonical_nll": 1.41656494140625, "generation_rouge_l_f1": 0.2358974358974359, "generation_token_f1": 0.4512820512820513, "prompt_id": "6b4ba85e93ef4f52", "reading_canonical_log_prob": -504.297119140625, "reading_distances": [29, 24, 24, 28], "reading_log_probs": [-504.1223449707031, -498.6793518066406, -502.183349609375, -501.821044921875], "reading_logprob_std": 2.253910683465561, "tokenization_path_logprob_std": 278.3096924684524, "trajectory_balance_rmse": 7994.302943721863} +{"canonical_length": 420, "canonical_nll": 1.261261712937128, "generation_rouge_l_f1": 0.18575851393188852, "generation_token_f1": 0.3374613003095976, "prompt_id": "71f47b6572a134fd", "reading_canonical_log_prob": -529.7299194335938, "reading_distances": [28, 37, 36, 36], "reading_log_probs": [-527.5496215820312, -528.8470458984375, -536.15478515625, -525.9954833984375], "reading_logprob_std": 4.499012571733934, "tokenization_path_logprob_std": 246.79855640044306, "trajectory_balance_rmse": 7878.88061466987} +{"canonical_length": 144, "canonical_nll": 1.3349451488918729, "generation_rouge_l_f1": 0.3405797101449275, "generation_token_f1": 0.5652173913043478, "prompt_id": "c1009daeb59a17f6", "reading_canonical_log_prob": -192.2321014404297, "reading_distances": [18, 17, 20, 19], "reading_log_probs": [-195.22320556640625, -195.0088348388672, -197.12013244628906, -195.0987548828125], "reading_logprob_std": 1.00877007723223, "tokenization_path_logprob_std": 118.90417854826909, "trajectory_balance_rmse": 2779.0016333379976} +{"canonical_length": 19, "canonical_nll": 2.2438476964047083, "generation_rouge_l_f1": 0.08695652173913045, "generation_token_f1": 0.08695652173913045, "prompt_id": "231e10798735d9e9", "reading_canonical_log_prob": -42.63310623168945, "reading_distances": [17, 20, 16, 21], "reading_log_probs": [-43.97887420654297, -42.852577209472656, -41.83769226074219, -43.82821273803711], "reading_logprob_std": 0.9924688006436464, "tokenization_path_logprob_std": 25.567671973956113, "trajectory_balance_rmse": 494.8425625655426} +{"canonical_length": 133, "canonical_nll": 1.6466293908599625, "generation_rouge_l_f1": 0.2808988764044944, "generation_token_f1": 0.37078651685393255, "prompt_id": "948eaf298aee4503", "reading_canonical_log_prob": -219.001708984375, "reading_distances": [24, 19, 19, 19], "reading_log_probs": [-217.9663543701172, -217.75592041015625, -219.10316467285156, -215.49368286132812], "reading_logprob_std": 1.5113975123677446, "tokenization_path_logprob_std": 171.9279923634321, "trajectory_balance_rmse": 2791.1662260027256} +{"canonical_length": 43, "canonical_nll": 0.5594776951989462, "generation_rouge_l_f1": 0.5476190476190477, "generation_token_f1": 0.7142857142857143, "prompt_id": "cc672b4406958c5b", "reading_canonical_log_prob": -24.057540893554688, "reading_distances": [44, 54, 40, 46], "reading_log_probs": [-31.656356811523438, -30.82425880432129, -28.84328842163086, -32.9862060546875], "reading_logprob_std": 1.735325760822757, "tokenization_path_logprob_std": 56.371868227228454, "trajectory_balance_rmse": 859.1188607795781} +{"canonical_length": 189, "canonical_nll": 1.1585120952949322, "generation_rouge_l_f1": 0.145, "generation_token_f1": 0.32000000000000006, "prompt_id": "f0c1c49e728adae3", "reading_canonical_log_prob": -218.9587860107422, "reading_distances": [17, 19, 17, 21], "reading_log_probs": [-212.50531005859375, -214.2952880859375, -212.65350341796875, -215.86338806152344], "reading_logprob_std": 1.5800955756492123, "tokenization_path_logprob_std": 35.39016354914744, "trajectory_balance_rmse": 3859.171005244504} +{"canonical_length": 202, "canonical_nll": 1.5364431249033106, "generation_rouge_l_f1": 0.22746781115879824, "generation_token_f1": 0.48068669527896996, "prompt_id": "91ba22c36f60a796", "reading_canonical_log_prob": -310.36151123046875, "reading_distances": [28, 31, 29, 23], "reading_log_probs": [-308.5681457519531, -313.49676513671875, -309.06964111328125, -309.513671875], "reading_logprob_std": 2.256443157096371, "tokenization_path_logprob_std": 223.13331581361612, "trajectory_balance_rmse": 4987.995391422788} +{"canonical_length": 91, "canonical_nll": 1.2029169103601476, "generation_rouge_l_f1": 0.27999999999999997, "generation_token_f1": 0.43, "prompt_id": "f1222deb29a5be25", "reading_canonical_log_prob": -109.46543884277344, "reading_distances": [13, 13, 9, 15], "reading_log_probs": [-108.83417510986328, -108.36136627197266, -109.279296875, -106.51216125488281], "reading_logprob_std": 1.2156145651834196, "tokenization_path_logprob_std": 54.02606236072729, "trajectory_balance_rmse": 2314.3600839577252} +{"canonical_length": 9, "canonical_nll": 0.15736797120836046, "generation_rouge_l_f1": 0.9333333333333333, "generation_token_f1": 0.9333333333333333, "prompt_id": "6b02223ad9dd90ef", "reading_canonical_log_prob": -1.4163117408752441, "reading_distances": [25, 29, 27, 25], "reading_log_probs": [-1.6301313638687134, -1.604924201965332, -1.7424108982086182, -1.48616361618042], "reading_logprob_std": 0.10513409894984806, "tokenization_path_logprob_std": 50.68948013970102, "trajectory_balance_rmse": 188.8934311148323} +{"canonical_length": 170, "canonical_nll": 1.2437903909122243, "generation_rouge_l_f1": 0.17326732673267325, "generation_token_f1": 0.3217821782178218, "prompt_id": "a027618ad6b82f3e", "reading_canonical_log_prob": -211.44436645507812, "reading_distances": [23, 19, 18, 19], "reading_log_probs": [-217.92996215820312, -217.71273803710938, -217.4520721435547, -218.0438232421875], "reading_logprob_std": 0.2608101185328069, "tokenization_path_logprob_std": 183.84259035495938, "trajectory_balance_rmse": 3909.772383139977} +{"canonical_length": 38, "canonical_nll": 1.4508188147293894, "generation_rouge_l_f1": 0.05405405405405406, "generation_token_f1": 0.05405405405405406, "prompt_id": "252eb77faa4d024a", "reading_canonical_log_prob": -55.1311149597168, "reading_distances": [20, 24, 22, 13], "reading_log_probs": [-57.647212982177734, -57.43714904785156, -58.77069091796875, -57.51224136352539], "reading_logprob_std": 0.6253142491888284, "tokenization_path_logprob_std": 24.43298226647681, "trajectory_balance_rmse": 731.8292541359951} +{"canonical_length": 273, "canonical_nll": 0.7574252732944139, "generation_rouge_l_f1": 0.33548387096774196, "generation_token_f1": 0.546236559139785, "prompt_id": "75be9357ba7c860a", "reading_canonical_log_prob": -206.777099609375, "reading_distances": [24, 18, 17, 16], "reading_log_probs": [-207.9896240234375, -206.99765014648438, -201.46286010742188, -200.668212890625], "reading_logprob_std": 3.747364199398036, "tokenization_path_logprob_std": 144.30345733477859, "trajectory_balance_rmse": 5007.3293379127135} +{"canonical_length": 347, "canonical_nll": 1.0870453672381575, "generation_rouge_l_f1": 0.275974025974026, "generation_token_f1": 0.538961038961039, "prompt_id": "8d16eaea1c875841", "reading_canonical_log_prob": -377.2047424316406, "reading_distances": [41, 42, 47, 53], "reading_log_probs": [-381.760009765625, -385.26080322265625, -383.13641357421875, -382.8310546875], "reading_logprob_std": 1.4664905065597618, "tokenization_path_logprob_std": 117.88782868166746, "trajectory_balance_rmse": 7849.2206979210605} +{"canonical_length": 355, "canonical_nll": 0.9918620364766725, "generation_rouge_l_f1": 0.24025974025974023, "generation_token_f1": 0.45454545454545453, "prompt_id": "f4746ccf1632be92", "reading_canonical_log_prob": -352.11102294921875, "reading_distances": [23, 26, 23, 20], "reading_log_probs": [-346.18109130859375, -346.289794921875, -347.2056884765625, -345.1099853515625], "reading_logprob_std": 0.8578882511333498, "tokenization_path_logprob_std": 323.60261523844673, "trajectory_balance_rmse": 7637.7636532576535} +{"canonical_length": 450, "canonical_nll": 1.04010009765625, "generation_rouge_l_f1": 0.1471004243281471, "generation_token_f1": 0.29985855728429983, "prompt_id": "06190df6c1e1573b", "reading_canonical_log_prob": -468.0450439453125, "reading_distances": [16, 15, 17, 12], "reading_log_probs": [-465.74322509765625, -468.04449462890625, -465.0039978027344, -468.76055908203125], "reading_logprob_std": 1.7985115308309916, "tokenization_path_logprob_std": 305.1596536176013, "trajectory_balance_rmse": 7836.887318678927} +{"canonical_length": 102, "canonical_nll": 0.9370375988530177, "generation_rouge_l_f1": 0.33179723502304154, "generation_token_f1": 0.5069124423963134, "prompt_id": "56fa8783ee42ce19", "reading_canonical_log_prob": -95.57783508300781, "reading_distances": [13, 19, 16, 19], "reading_log_probs": [-92.68904113769531, -95.40811920166016, -91.97200775146484, -93.73207092285156], "reading_logprob_std": 1.491901907688066, "tokenization_path_logprob_std": 35.07686839546865, "trajectory_balance_rmse": 2016.6966771845798} +{"canonical_length": 202, "canonical_nll": 2.120256329526996, "generation_rouge_l_f1": 0.28187919463087246, "generation_token_f1": 0.46085011185682323, "prompt_id": "11f8f65e51ee2cf5", "reading_canonical_log_prob": -428.2917785644531, "reading_distances": [22, 23, 23, 20], "reading_log_probs": [-420.1524658203125, -423.30914306640625, -420.44049072265625, -420.57147216796875], "reading_logprob_std": 1.4709492472612848, "tokenization_path_logprob_std": 212.67928469221008, "trajectory_balance_rmse": 3706.7116414396487} +{"canonical_length": 11, "canonical_nll": 2.598480224609375, "generation_rouge_l_f1": 0.3636363636363636, "generation_token_f1": 0.3636363636363636, "prompt_id": "8980941d8fb125ec", "reading_canonical_log_prob": -28.583282470703125, "reading_distances": [15, 18, 16, 21], "reading_log_probs": [-27.59136390686035, -28.858524322509766, -27.44101333618164, -26.51776123046875], "reading_logprob_std": 0.9627143579472338, "tokenization_path_logprob_std": 38.03919550629052, "trajectory_balance_rmse": 229.3683212929659} +{"canonical_length": 69, "canonical_nll": 0.5458819071451823, "generation_rouge_l_f1": 0.3508771929824561, "generation_token_f1": 0.4035087719298246, "prompt_id": "2d0e3ff402d06b28", "reading_canonical_log_prob": -37.66585159301758, "reading_distances": [16, 14, 13, 14], "reading_log_probs": [-40.735843658447266, -39.499507904052734, -41.87228775024414, -38.8023796081543], "reading_logprob_std": 1.3570409771955272, "tokenization_path_logprob_std": 152.50467707242254, "trajectory_balance_rmse": 1032.3054797179918} +{"canonical_length": 53, "canonical_nll": 0.533127874698279, "generation_rouge_l_f1": 0.22916666666666666, "generation_token_f1": 0.34375, "prompt_id": "54b6e30181a1bec1", "reading_canonical_log_prob": -28.25577735900879, "reading_distances": [19, 24, 19, 16], "reading_log_probs": [-26.61971664428711, -25.486570358276367, -26.841840744018555, -26.574058532714844], "reading_logprob_std": 0.6073535451062669, "tokenization_path_logprob_std": 26.43713161561751, "trajectory_balance_rmse": 456.8153546693416} +{"canonical_length": 9, "canonical_nll": 0.3241233825683594, "generation_rouge_l_f1": 0.75, "generation_token_f1": 0.75, "prompt_id": "2996ca79ae58108b", "reading_canonical_log_prob": -2.9171104431152344, "reading_distances": [30, 28, 26, 36], "reading_log_probs": [-6.191239356994629, -5.06817626953125, -5.671997547149658, -5.3410797119140625], "reading_logprob_std": 0.483238831176633, "tokenization_path_logprob_std": 42.50872589196099, "trajectory_balance_rmse": 150.37127823399828} +{"canonical_length": 11, "canonical_nll": 1.6711847131902522, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.23529411764705882, "prompt_id": "e6cc80b8ba91ddd7", "reading_canonical_log_prob": -18.383031845092773, "reading_distances": [21, 22, 27, 25], "reading_log_probs": [-21.607337951660156, -21.23921012878418, -21.463687896728516, -21.479976654052734], "reading_logprob_std": 0.1530246287344644, "tokenization_path_logprob_std": 32.458156152642516, "trajectory_balance_rmse": 248.8576454508956} +{"canonical_length": 218, "canonical_nll": 1.7449984769208715, "generation_rouge_l_f1": 0.19624217118997914, "generation_token_f1": 0.42588726513569936, "prompt_id": "1826061f4f79f3ac", "reading_canonical_log_prob": -380.40966796875, "reading_distances": [14, 16, 13, 20], "reading_log_probs": [-376.83831787109375, -377.45233154296875, -379.4849853515625, -377.95501708984375], "reading_logprob_std": 1.1311519140928665, "tokenization_path_logprob_std": 129.6268865480385, "trajectory_balance_rmse": 4680.778921708573} +{"canonical_length": 182, "canonical_nll": 1.796819330571772, "generation_rouge_l_f1": 0.22477064220183487, "generation_token_f1": 0.44036697247706413, "prompt_id": "d48fbaa3dac1814d", "reading_canonical_log_prob": -327.0211181640625, "reading_distances": [29, 25, 30, 24], "reading_log_probs": [-326.0454406738281, -325.7747802734375, -323.83209228515625, -328.2964782714844], "reading_logprob_std": 1.828092768801382, "tokenization_path_logprob_std": 10.178400866414734, "trajectory_balance_rmse": 3786.206892260544} +{"canonical_length": 105, "canonical_nll": 0.5302913847423735, "generation_rouge_l_f1": 0.32558139534883723, "generation_token_f1": 0.43853820598006643, "prompt_id": "7fbf1f51feeeeb73", "reading_canonical_log_prob": -55.68059539794922, "reading_distances": [17, 20, 17, 20], "reading_log_probs": [-56.873722076416016, -56.39137268066406, -55.88173294067383, -57.73883819580078], "reading_logprob_std": 0.7900087126477512, "tokenization_path_logprob_std": 67.8963392628304, "trajectory_balance_rmse": 923.8223159007575} +{"canonical_length": 18, "canonical_nll": 0.30847030215793186, "generation_rouge_l_f1": 0.26923076923076916, "generation_token_f1": 0.26923076923076916, "prompt_id": "8cc0717e876412c6", "reading_canonical_log_prob": -5.552465438842773, "reading_distances": [20, 21, 29, 26], "reading_log_probs": [-6.462574005126953, -7.06609582901001, -6.653608322143555, -6.822485446929932], "reading_logprob_std": 0.2563006686452535, "tokenization_path_logprob_std": 91.95194932159961, "trajectory_balance_rmse": 387.2826861172966} +{"canonical_length": 152, "canonical_nll": 1.8689735814144737, "generation_rouge_l_f1": 0.26239067055393583, "generation_token_f1": 0.45481049562682213, "prompt_id": "3f19d2fd29921f16", "reading_canonical_log_prob": -284.083984375, "reading_distances": [19, 17, 16, 12], "reading_log_probs": [-293.4758605957031, -296.92535400390625, -296.61419677734375, -295.7314453125], "reading_logprob_std": 1.558239073137269, "tokenization_path_logprob_std": 294.4559822462775, "trajectory_balance_rmse": 3171.2223211227943} +{"canonical_length": 257, "canonical_nll": 1.9279244864497203, "generation_rouge_l_f1": 0.21473684210526317, "generation_token_f1": 0.4378947368421053, "prompt_id": "774504b35575be36", "reading_canonical_log_prob": -495.4765930175781, "reading_distances": [31, 24, 28, 29], "reading_log_probs": [-497.06365966796875, -493.57354736328125, -496.855712890625, -496.038818359375], "reading_logprob_std": 1.6018716599026122, "tokenization_path_logprob_std": 326.56901746835183, "trajectory_balance_rmse": 5916.995254066985} +{"canonical_length": 9, "canonical_nll": 0.3011060290866428, "generation_rouge_l_f1": 0.25454545454545446, "generation_token_f1": 0.2909090909090909, "prompt_id": "18b528d0286abbc1", "reading_canonical_log_prob": -2.709954261779785, "reading_distances": [34, 30, 33, 31], "reading_log_probs": [-1.9700525999069214, -1.8320499658584595, -1.9349405765533447, -2.6852498054504395], "reading_logprob_std": 0.3908628283493086, "tokenization_path_logprob_std": 25.48114007586195, "trajectory_balance_rmse": 213.8575380094} +{"canonical_length": 4, "canonical_nll": 0.9109907150268555, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.5714285714285715, "prompt_id": "ad4fd52fcf4632b3", "reading_canonical_log_prob": -3.643962860107422, "reading_distances": [22, 29, 20, 25], "reading_log_probs": [-4.121651649475098, -3.4151225090026855, -4.548590660095215, -4.020552158355713], "reading_logprob_std": 0.46742211663252103, "tokenization_path_logprob_std": 36.68397000349939, "trajectory_balance_rmse": 102.62022628480419} +{"canonical_length": 349, "canonical_nll": 1.389245762866001, "generation_rouge_l_f1": 0.21088435374149656, "generation_token_f1": 0.42857142857142855, "prompt_id": "ba9050b431bd2ba0", "reading_canonical_log_prob": -484.8467712402344, "reading_distances": [18, 15, 21, 19], "reading_log_probs": [-480.16961669921875, -482.73040771484375, -478.0311279296875, -479.84088134765625], "reading_logprob_std": 1.9353434736970991, "tokenization_path_logprob_std": 118.6008788002155, "trajectory_balance_rmse": 6557.045698146198} +{"canonical_length": 35, "canonical_nll": 0.664465822492327, "generation_rouge_l_f1": 0.09090909090909091, "generation_token_f1": 0.12337662337662338, "prompt_id": "da7f7e74060fc4e6", "reading_canonical_log_prob": -23.256303787231445, "reading_distances": [27, 33, 26, 26], "reading_log_probs": [-22.03156280517578, -21.948284149169922, -23.0767765045166, -22.212261199951172], "reading_logprob_std": 0.5182200717321735, "tokenization_path_logprob_std": 76.80161035361556, "trajectory_balance_rmse": 742.8511988859152} +{"canonical_length": 366, "canonical_nll": 1.2029314718611253, "generation_rouge_l_f1": 0.21840873634945399, "generation_token_f1": 0.4492979719188767, "prompt_id": "378c4d815b07ba59", "reading_canonical_log_prob": -440.2729187011719, "reading_distances": [29, 25, 29, 34], "reading_log_probs": [-442.3946533203125, -439.8016357421875, -441.49761962890625, -441.99945068359375], "reading_logprob_std": 1.141752344577535, "tokenization_path_logprob_std": 363.9607553868021, "trajectory_balance_rmse": 7320.366183058312} +{"canonical_length": 158, "canonical_nll": 0.5548084597044354, "generation_rouge_l_f1": 0.24489795918367344, "generation_token_f1": 0.2693877551020408, "prompt_id": "e2790ed28464c480", "reading_canonical_log_prob": -87.65973663330078, "reading_distances": [14, 24, 22, 19], "reading_log_probs": [-87.91102600097656, -86.42166137695312, -85.59164428710938, -85.60816955566406], "reading_logprob_std": 1.089795727146032, "tokenization_path_logprob_std": 147.96324119724568, "trajectory_balance_rmse": 2036.3278740983196} +{"canonical_length": 253, "canonical_nll": 1.2261631177819294, "generation_rouge_l_f1": 0.25742574257425743, "generation_token_f1": 0.499009900990099, "prompt_id": "2481a184df55ff40", "reading_canonical_log_prob": -310.2192687988281, "reading_distances": [47, 39, 37, 41], "reading_log_probs": [-328.3554382324219, -321.4063720703125, -322.1323547363281, -317.5189208984375], "reading_logprob_std": 4.484877329018538, "tokenization_path_logprob_std": 256.3122088693747, "trajectory_balance_rmse": 4926.287690547454} +{"canonical_length": 342, "canonical_nll": 1.562209993775128, "generation_rouge_l_f1": 0.1891891891891892, "generation_token_f1": 0.48648648648648646, "prompt_id": "130ff8d17ea30994", "reading_canonical_log_prob": -534.2758178710938, "reading_distances": [18, 20, 17, 21], "reading_log_probs": [-529.5007934570312, -529.08056640625, -527.6178588867188, -529.2593994140625], "reading_logprob_std": 0.8488450138057962, "tokenization_path_logprob_std": 325.4897753035695, "trajectory_balance_rmse": 7657.690162719559} +{"canonical_length": 16, "canonical_nll": 0.7570456266403198, "generation_rouge_l_f1": 0.7777777777777778, "generation_token_f1": 0.7777777777777778, "prompt_id": "16bf564574acdadf", "reading_canonical_log_prob": -12.112730026245117, "reading_distances": [27, 28, 22, 31], "reading_log_probs": [-14.178277969360352, -14.948406219482422, -11.88417911529541, -14.284039497375488], "reading_logprob_std": 1.3372039646099003, "tokenization_path_logprob_std": 26.25306926570001, "trajectory_balance_rmse": 219.66352416769988} +{"canonical_length": 50, "canonical_nll": 0.763231201171875, "generation_rouge_l_f1": 0.35658914728682173, "generation_token_f1": 0.41860465116279066, "prompt_id": "effedc7b02b1c679", "reading_canonical_log_prob": -38.16156005859375, "reading_distances": [29, 29, 27, 26], "reading_log_probs": [-38.986793518066406, -40.344635009765625, -40.104522705078125, -38.72234344482422], "reading_logprob_std": 0.8043055618220215, "tokenization_path_logprob_std": 69.07872470988521, "trajectory_balance_rmse": 831.2651038174215} +{"canonical_length": 176, "canonical_nll": 1.4042973951859907, "generation_rouge_l_f1": 0.34375, "generation_token_f1": 0.5416666666666666, "prompt_id": "33cc463f90c8e3a1", "reading_canonical_log_prob": -247.15634155273438, "reading_distances": [25, 17, 25, 28], "reading_log_probs": [-243.81295776367188, -243.96051025390625, -246.59213256835938, -245.12722778320312], "reading_logprob_std": 1.2879409227294016, "tokenization_path_logprob_std": 73.00511218904887, "trajectory_balance_rmse": 3682.084381334865} +{"canonical_length": 283, "canonical_nll": 1.205348793272416, "generation_rouge_l_f1": 0.24275362318840582, "generation_token_f1": 0.5, "prompt_id": "e32f1e4a8a806d4a", "reading_canonical_log_prob": -341.11370849609375, "reading_distances": [15, 14, 15, 15], "reading_log_probs": [-341.2640380859375, -338.9508972167969, -340.1932067871094, -340.4023132324219], "reading_logprob_std": 0.9545302989157856, "tokenization_path_logprob_std": 358.0437693303116, "trajectory_balance_rmse": 6166.530459880657} +{"canonical_length": 193, "canonical_nll": 1.2326541564625162, "generation_rouge_l_f1": 0.19913419913419914, "generation_token_f1": 0.4329004329004329, "prompt_id": "f16b4d6e1a4f11c6", "reading_canonical_log_prob": -237.90225219726562, "reading_distances": [11, 8, 9, 13], "reading_log_probs": [-237.05596923828125, -237.0238494873047, -236.6767578125, -236.73037719726562], "reading_logprob_std": 0.19585641321741803, "tokenization_path_logprob_std": 174.04096233159038, "trajectory_balance_rmse": 4284.234939329452} +{"canonical_length": 11, "canonical_nll": 0.8110313415527344, "generation_rouge_l_f1": 0.19999999999999998, "generation_token_f1": 0.19999999999999998, "prompt_id": "7cf86e055cecbc26", "reading_canonical_log_prob": -8.921344757080078, "reading_distances": [9, 16, 13, 16], "reading_log_probs": [-10.147351264953613, -10.421610832214355, -10.50709342956543, -10.268771171569824], "reading_logprob_std": 0.1599062912896903, "tokenization_path_logprob_std": 44.547950270366215, "trajectory_balance_rmse": 214.3938806135602} +{"canonical_length": 251, "canonical_nll": 1.4556571078965388, "generation_rouge_l_f1": 0.19444444444444445, "generation_token_f1": 0.3492063492063492, "prompt_id": "279ec9f263f853e8", "reading_canonical_log_prob": -365.36993408203125, "reading_distances": [29, 20, 27, 24], "reading_log_probs": [-363.62744140625, -363.7901306152344, -361.5589904785156, -361.47607421875], "reading_logprob_std": 1.2673152796541314, "tokenization_path_logprob_std": 175.44003980306874, "trajectory_balance_rmse": 5260.062873796798} +{"canonical_length": 102, "canonical_nll": 0.9638950123506433, "generation_rouge_l_f1": 0.27299703264094954, "generation_token_f1": 0.39762611275964393, "prompt_id": "cf98e6698c25bbf0", "reading_canonical_log_prob": -98.31729125976562, "reading_distances": [110, 109, 100, 109], "reading_log_probs": [-90.2217788696289, -91.65375518798828, -92.24874877929688, -90.728759765625], "reading_logprob_std": 0.9099539057337358, "tokenization_path_logprob_std": 143.1021339122695, "trajectory_balance_rmse": 1948.8537639166996} +{"canonical_length": 125, "canonical_nll": 1.4804044189453125, "generation_rouge_l_f1": 0.21374045801526717, "generation_token_f1": 0.42748091603053434, "prompt_id": "588a316f7d854f31", "reading_canonical_log_prob": -185.05055236816406, "reading_distances": [34, 25, 33, 35], "reading_log_probs": [-183.60089111328125, -183.65939331054688, -185.54043579101562, -184.41281127929688], "reading_logprob_std": 0.9037867977011983, "tokenization_path_logprob_std": 155.21766331429686, "trajectory_balance_rmse": 3177.7812086358} +{"canonical_length": 116, "canonical_nll": 1.66598497588059, "generation_rouge_l_f1": 0.29596412556053814, "generation_token_f1": 0.46636771300448426, "prompt_id": "6bdc8fa7262e4a79", "reading_canonical_log_prob": -193.25425720214844, "reading_distances": [18, 15, 18, 11], "reading_log_probs": [-194.02093505859375, -195.17242431640625, -193.70953369140625, -195.9130859375], "reading_logprob_std": 1.022554170929689, "tokenization_path_logprob_std": 171.06653571208693, "trajectory_balance_rmse": 2607.670975324181} +{"canonical_length": 7, "canonical_nll": 1.7849723270961217, "generation_rouge_l_f1": 0.125, "generation_token_f1": 0.125, "prompt_id": "73284d596fde1232", "reading_canonical_log_prob": -12.494806289672852, "reading_distances": [26, 25, 31, 27], "reading_log_probs": [-12.297134399414062, -11.982185363769531, -12.744224548339844, -12.471930503845215], "reading_logprob_std": 0.3194179918434705, "tokenization_path_logprob_std": 28.511045336564948, "trajectory_balance_rmse": 186.0255590487174} +{"canonical_length": 47, "canonical_nll": 0.9748502690741356, "generation_rouge_l_f1": 0.21481481481481482, "generation_token_f1": 0.2740740740740741, "prompt_id": "08b3c472b8ed1840", "reading_canonical_log_prob": -45.817962646484375, "reading_distances": [37, 51, 48, 51], "reading_log_probs": [-46.26726531982422, -46.35626983642578, -45.30498123168945, -43.56801986694336], "reading_logprob_std": 1.2947464002182894, "tokenization_path_logprob_std": 66.60429885710577, "trajectory_balance_rmse": 687.3030020302019} +{"canonical_length": 149, "canonical_nll": 0.7907848486164272, "generation_rouge_l_f1": 0.2945736434108527, "generation_token_f1": 0.4754521963824289, "prompt_id": "51fa8b33eba8ebe6", "reading_canonical_log_prob": -117.82694244384766, "reading_distances": [40, 36, 42, 45], "reading_log_probs": [-116.85363006591797, -119.01384735107422, -118.2828598022461, -118.654296875], "reading_logprob_std": 0.9466265611820835, "tokenization_path_logprob_std": 171.1107448593964, "trajectory_balance_rmse": 2459.0241088606435} +{"canonical_length": 25, "canonical_nll": 1.2222698974609374, "generation_rouge_l_f1": 0.08088235294117647, "generation_token_f1": 0.11029411764705882, "prompt_id": "6e11957de8d37c71", "reading_canonical_log_prob": -30.556747436523438, "reading_distances": [26, 22, 27, 19], "reading_log_probs": [-27.520977020263672, -27.622060775756836, -29.43310546875, -26.68169403076172], "reading_logprob_std": 1.1584935067364723, "tokenization_path_logprob_std": 35.37063865667261, "trajectory_balance_rmse": 522.5868972573605} +{"canonical_length": 124, "canonical_nll": 1.1440220494424143, "generation_rouge_l_f1": 0.25252525252525254, "generation_token_f1": 0.4040404040404041, "prompt_id": "613e08997d787f9d", "reading_canonical_log_prob": -141.85873413085938, "reading_distances": [21, 18, 24, 20], "reading_log_probs": [-141.850830078125, -142.48321533203125, -143.1194610595703, -142.308837890625], "reading_logprob_std": 0.5253120400189178, "tokenization_path_logprob_std": 186.50134253125208, "trajectory_balance_rmse": 2805.754560102874} +{"canonical_length": 7, "canonical_nll": 0.22852298191615514, "generation_rouge_l_f1": 0.9090909090909091, "generation_token_f1": 0.9090909090909091, "prompt_id": "82a7bb73a8e6bdca", "reading_canonical_log_prob": -1.599660873413086, "reading_distances": [33, 34, 31, 40], "reading_log_probs": [-3.999227523803711, -3.0847413539886475, -3.969539165496826, -3.9073784351348877], "reading_logprob_std": 0.4386594635379547, "tokenization_path_logprob_std": 18.99324223703932, "trajectory_balance_rmse": 120.85802598939387} +{"canonical_length": 500, "canonical_nll": 0.76746337890625, "generation_rouge_l_f1": 0.1752021563342318, "generation_token_f1": 0.29649595687331537, "prompt_id": "04c1fbf027769e8a", "reading_canonical_log_prob": -383.731689453125, "reading_distances": [24, 24, 28, 22], "reading_log_probs": [-385.84576416015625, -385.416015625, -390.2679443359375, -385.93609619140625], "reading_logprob_std": 2.278982330362672, "tokenization_path_logprob_std": 209.2842487729655, "trajectory_balance_rmse": 8429.887669126583} +{"canonical_length": 47, "canonical_nll": 1.6398782933011968, "generation_rouge_l_f1": 0.15282392026578073, "generation_token_f1": 0.20598006644518274, "prompt_id": "5950f3bf48416566", "reading_canonical_log_prob": -77.07427978515625, "reading_distances": [32, 35, 36, 48], "reading_log_probs": [-79.29147338867188, -78.1435317993164, -78.400634765625, -78.14067840576172], "reading_logprob_std": 0.5453882274542314, "tokenization_path_logprob_std": 69.93542691066533, "trajectory_balance_rmse": 1185.7292865199186} +{"canonical_length": 310, "canonical_nll": 1.1176283313382056, "generation_rouge_l_f1": 0.31127012522361364, "generation_token_f1": 0.51520572450805, "prompt_id": "f93b23d1ff15b3d1", "reading_canonical_log_prob": -346.46478271484375, "reading_distances": [22, 25, 23, 24], "reading_log_probs": [-344.00006103515625, -343.3915100097656, -344.3099060058594, -343.9224548339844], "reading_logprob_std": 0.38164479574461124, "tokenization_path_logprob_std": 215.39373471710218, "trajectory_balance_rmse": 5916.5163596418} +{"canonical_length": 228, "canonical_nll": 1.4110301502963953, "generation_rouge_l_f1": 0.21868787276341947, "generation_token_f1": 0.4015904572564612, "prompt_id": "cb117bc3f0e8e883", "reading_canonical_log_prob": -321.7148742675781, "reading_distances": [25, 19, 19, 23], "reading_log_probs": [-314.07049560546875, -313.3575439453125, -313.10076904296875, -312.8687744140625], "reading_logprob_std": 0.5205329008658508, "tokenization_path_logprob_std": 277.3915622701937, "trajectory_balance_rmse": 4780.602574017085} +{"canonical_length": 318, "canonical_nll": 1.8252210317167845, "generation_rouge_l_f1": 0.21143847487001732, "generation_token_f1": 0.4679376083188908, "prompt_id": "0ff4ea021b8e685b", "reading_canonical_log_prob": -580.4202880859375, "reading_distances": [16, 20, 14, 14], "reading_log_probs": [-585.3006591796875, -587.779541015625, -585.0988159179688, -583.2642822265625], "reading_logprob_std": 1.854539246470124, "tokenization_path_logprob_std": 192.48219472853518, "trajectory_balance_rmse": 6579.482872090807} +{"canonical_length": 70, "canonical_nll": 1.2476207188197546, "generation_rouge_l_f1": 0.24615384615384614, "generation_token_f1": 0.35384615384615387, "prompt_id": "d8931b93e350f362", "reading_canonical_log_prob": -87.33345031738281, "reading_distances": [14, 16, 14, 11], "reading_log_probs": [-87.06845092773438, -85.87855529785156, -86.633056640625, -87.40277862548828], "reading_logprob_std": 0.6584195576965511, "tokenization_path_logprob_std": 146.857220500928, "trajectory_balance_rmse": 1581.602903357317} +{"canonical_length": 43, "canonical_nll": 2.38672726653343, "generation_rouge_l_f1": 0.273972602739726, "generation_token_f1": 0.4109589041095891, "prompt_id": "cfd710328af17995", "reading_canonical_log_prob": -102.6292724609375, "reading_distances": [32, 31, 27, 30], "reading_log_probs": [-101.7010726928711, -102.16861724853516, -102.41573333740234, -101.59943389892578], "reading_logprob_std": 0.38633416158751094, "tokenization_path_logprob_std": 106.59203412867616, "trajectory_balance_rmse": 935.1902964685362} +{"canonical_length": 237, "canonical_nll": 0.7302322065779931, "generation_rouge_l_f1": 0.3676148796498906, "generation_token_f1": 0.48577680525164113, "prompt_id": "3f921f925896c713", "reading_canonical_log_prob": -173.06503295898438, "reading_distances": [37, 35, 42, 39], "reading_log_probs": [-177.91693115234375, -173.05194091796875, -176.69918823242188, -176.49627685546875], "reading_logprob_std": 2.0891848984881225, "tokenization_path_logprob_std": 123.19981636008374, "trajectory_balance_rmse": 2480.5688955346395} +{"canonical_length": 242, "canonical_nll": 1.0082615623789386, "generation_rouge_l_f1": 0.4703389830508475, "generation_token_f1": 0.6016949152542372, "prompt_id": "74c9b8835b365ad8", "reading_canonical_log_prob": -243.99929809570312, "reading_distances": [21, 15, 17, 19], "reading_log_probs": [-244.85482788085938, -243.08416748046875, -243.95050048828125, -245.16641235351562], "reading_logprob_std": 0.9405162335056697, "tokenization_path_logprob_std": 66.457753782426, "trajectory_balance_rmse": 4707.6493864852355} +{"canonical_length": 243, "canonical_nll": 1.2490406428835519, "generation_rouge_l_f1": 0.23577235772357721, "generation_token_f1": 0.5406504065040649, "prompt_id": "d4cbdb2804902b19", "reading_canonical_log_prob": -303.5168762207031, "reading_distances": [26, 16, 23, 24], "reading_log_probs": [-296.6678466796875, -298.0566101074219, -302.2359619140625, -299.4747009277344], "reading_logprob_std": 2.378962085031567, "tokenization_path_logprob_std": 84.19488945419013, "trajectory_balance_rmse": 4552.348444441037} +{"canonical_length": 140, "canonical_nll": 1.1344250270298548, "generation_rouge_l_f1": 0.26451612903225813, "generation_token_f1": 0.42580645161290326, "prompt_id": "ad65facd3056e4a9", "reading_canonical_log_prob": -158.8195037841797, "reading_distances": [18, 13, 19, 17], "reading_log_probs": [-167.57675170898438, -158.76071166992188, -162.99588012695312, -166.41510009765625], "reading_logprob_std": 3.9609948261941015, "tokenization_path_logprob_std": 121.54684876810973, "trajectory_balance_rmse": 1994.7973077047948} +{"canonical_length": 245, "canonical_nll": 1.2431430115991708, "generation_rouge_l_f1": 0.23382045929018788, "generation_token_f1": 0.45929018789144055, "prompt_id": "4eba5e8b52e3bb5e", "reading_canonical_log_prob": -304.5700378417969, "reading_distances": [19, 24, 22, 25], "reading_log_probs": [-303.7947998046875, -303.98046875, -305.1629333496094, -306.3393859863281], "reading_logprob_std": 1.1806754794540335, "tokenization_path_logprob_std": 226.79927827593505, "trajectory_balance_rmse": 4903.105120721221} +{"canonical_length": 118, "canonical_nll": 1.7444421800516419, "generation_rouge_l_f1": 0.24489795918367344, "generation_token_f1": 0.40816326530612246, "prompt_id": "4873d3597b3d71b4", "reading_canonical_log_prob": -205.84417724609375, "reading_distances": [21, 22, 20, 20], "reading_log_probs": [-208.22653198242188, -208.14695739746094, -207.6150360107422, -205.504638671875], "reading_logprob_std": 1.2750009205578576, "tokenization_path_logprob_std": 125.7054828863438, "trajectory_balance_rmse": 3018.6284968668965} +{"canonical_length": 38, "canonical_nll": 1.3443063434801603, "generation_rouge_l_f1": 0.1395348837209302, "generation_token_f1": 0.1395348837209302, "prompt_id": "2089d8bccf6a73c9", "reading_canonical_log_prob": -51.083641052246094, "reading_distances": [12, 10, 9, 11], "reading_log_probs": [-57.45539855957031, -52.03897476196289, -53.4576416015625, -54.65771484375], "reading_logprob_std": 2.299589322551354, "tokenization_path_logprob_std": 121.54513845856944, "trajectory_balance_rmse": 764.7021158180619} +{"canonical_length": 277, "canonical_nll": 1.5700346468157718, "generation_rouge_l_f1": 0.24603174603174602, "generation_token_f1": 0.41666666666666663, "prompt_id": "9b806d8ec328a266", "reading_canonical_log_prob": -434.89959716796875, "reading_distances": [31, 23, 30, 30], "reading_log_probs": [-442.5446472167969, -440.377685546875, -442.28179931640625, -443.0289306640625], "reading_logprob_std": 1.1623359211922326, "tokenization_path_logprob_std": 272.5460549639842, "trajectory_balance_rmse": 6316.603253368838} +{"canonical_length": 400, "canonical_nll": 0.8718416595458984, "generation_rouge_l_f1": 0.0981132075471698, "generation_token_f1": 0.1169811320754717, "prompt_id": "ce03043d2a4264da", "reading_canonical_log_prob": -348.7366638183594, "reading_distances": [20, 23, 24, 18], "reading_log_probs": [-348.66064453125, -350.301513671875, -350.16094970703125, -347.5862731933594], "reading_logprob_std": 1.2948258944496567, "tokenization_path_logprob_std": 233.4753763397314, "trajectory_balance_rmse": 6846.3098613285} +{"canonical_length": 12, "canonical_nll": 0.9119040171305338, "generation_rouge_l_f1": 0.8571428571428572, "generation_token_f1": 0.8571428571428572, "prompt_id": "59741a0906fe97c9", "reading_canonical_log_prob": -10.942848205566406, "reading_distances": [27, 27, 30, 23], "reading_log_probs": [-11.362863540649414, -11.641809463500977, -10.8627290725708, -10.824501991271973], "reading_logprob_std": 0.39730315265568655, "tokenization_path_logprob_std": 43.61541901043229, "trajectory_balance_rmse": 234.09660715528716} +{"canonical_length": 103, "canonical_nll": 0.5547386095361803, "generation_rouge_l_f1": 0.18348623853211007, "generation_token_f1": 0.24464831804281345, "prompt_id": "995c665a4ec8066d", "reading_canonical_log_prob": -57.13807678222656, "reading_distances": [31, 27, 34, 26], "reading_log_probs": [-59.11677551269531, -59.37531280517578, -58.270992279052734, -62.27018737792969], "reading_logprob_std": 1.7397217511475362, "tokenization_path_logprob_std": 164.5587117237138, "trajectory_balance_rmse": 1432.5868082354248} +{"canonical_length": 63, "canonical_nll": 0.9774222601027716, "generation_rouge_l_f1": 0.6440677966101694, "generation_token_f1": 0.711864406779661, "prompt_id": "e7b51e88b3b34da7", "reading_canonical_log_prob": -61.57760238647461, "reading_distances": [29, 30, 26, 28], "reading_log_probs": [-63.321998596191406, -64.20639038085938, -63.168575286865234, -63.161495208740234], "reading_logprob_std": 0.50003042370459, "tokenization_path_logprob_std": 79.09601462885834, "trajectory_balance_rmse": 1096.2342672595419} +{"canonical_length": 90, "canonical_nll": 0.861476559109158, "generation_rouge_l_f1": 0.47668393782383417, "generation_token_f1": 0.621761658031088, "prompt_id": "6b580b925c517c44", "reading_canonical_log_prob": -77.53289031982422, "reading_distances": [26, 23, 27, 20], "reading_log_probs": [-78.1744384765625, -77.7001724243164, -77.8858871459961, -77.38272094726562], "reading_logprob_std": 0.33209542134701814, "tokenization_path_logprob_std": 183.506498845106, "trajectory_balance_rmse": 1794.8312692173877} +{"canonical_length": 169, "canonical_nll": 0.7145143768491125, "generation_rouge_l_f1": 0.27383863080684595, "generation_token_f1": 0.5036674816625918, "prompt_id": "58b4edff73083c67", "reading_canonical_log_prob": -120.7529296875, "reading_distances": [54, 51, 52, 62], "reading_log_probs": [-125.30089569091797, -126.35462951660156, -123.15853881835938, -120.99913787841797], "reading_logprob_std": 2.3763499160066552, "tokenization_path_logprob_std": 45.67015196022786, "trajectory_balance_rmse": 2632.7179488629235} +{"canonical_length": 6, "canonical_nll": 1.0196982224782307, "generation_rouge_l_f1": 0.10526315789473684, "generation_token_f1": 0.10526315789473684, "prompt_id": "dc845859825dc504", "reading_canonical_log_prob": -6.118189334869385, "reading_distances": [33, 29, 30, 35], "reading_log_probs": [-8.253782272338867, -9.86422061920166, -8.517308235168457, -8.547863960266113], "reading_logprob_std": 0.724416035934449, "tokenization_path_logprob_std": 16.604917334973056, "trajectory_balance_rmse": 116.9661803436332} +{"canonical_length": 55, "canonical_nll": 0.6456975069912997, "generation_rouge_l_f1": 0.08450704225352113, "generation_token_f1": 0.09859154929577464, "prompt_id": "f074eb196b10e4a0", "reading_canonical_log_prob": -35.513362884521484, "reading_distances": [30, 31, 27, 29], "reading_log_probs": [-39.746826171875, -37.262474060058594, -40.743324279785156, -38.260887145996094], "reading_logprob_std": 1.5451178420199776, "tokenization_path_logprob_std": 50.150353464721114, "trajectory_balance_rmse": 403.0704035944849} +{"canonical_length": 39, "canonical_nll": 2.228496649326422, "generation_rouge_l_f1": 0.11447811447811447, "generation_token_f1": 0.14814814814814814, "prompt_id": "b1076398ceba44c5", "reading_canonical_log_prob": -86.91136932373047, "reading_distances": [9, 12, 15, 13], "reading_log_probs": [-90.88914489746094, -91.89073944091797, -89.21388244628906, -91.91675567626953], "reading_logprob_std": 1.2694304409870043, "tokenization_path_logprob_std": 49.47668639110611, "trajectory_balance_rmse": 937.9482730145502} +{"canonical_length": 278, "canonical_nll": 1.7125583346799123, "generation_rouge_l_f1": 0.22423146473779385, "generation_token_f1": 0.41229656419529837, "prompt_id": "43d587270196a951", "reading_canonical_log_prob": -476.0912170410156, "reading_distances": [23, 24, 29, 22], "reading_log_probs": [-480.5403747558594, -478.1227722167969, -480.3588562011719, -478.48223876953125], "reading_logprob_std": 1.2504884223523323, "tokenization_path_logprob_std": 271.62888902921355, "trajectory_balance_rmse": 6209.3142859874215} +{"canonical_length": 77, "canonical_nll": 1.3625456574675325, "generation_rouge_l_f1": 0.3181818181818182, "generation_token_f1": 0.4431818181818181, "prompt_id": "9ec5781f60b3ed7d", "reading_canonical_log_prob": -104.916015625, "reading_distances": [14, 16, 9, 14], "reading_log_probs": [-103.15609741210938, -103.54182434082031, -102.67892456054688, -104.05486297607422], "reading_logprob_std": 0.5834716451374721, "tokenization_path_logprob_std": 210.4701631650414, "trajectory_balance_rmse": 1659.8392113705813} +{"canonical_length": 50, "canonical_nll": 1.1943716430664062, "generation_rouge_l_f1": 0.12861736334405147, "generation_token_f1": 0.18006430868167203, "prompt_id": "9c285d8b80679a4e", "reading_canonical_log_prob": -59.71858215332031, "reading_distances": [21, 24, 26, 20], "reading_log_probs": [-59.685882568359375, -58.460533142089844, -59.307334899902344, -58.29188919067383], "reading_logprob_std": 0.6686205375347212, "tokenization_path_logprob_std": 17.83039024841186, "trajectory_balance_rmse": 1098.182074362923} +{"canonical_length": 260, "canonical_nll": 1.425127469576322, "generation_rouge_l_f1": 0.26347305389221554, "generation_token_f1": 0.4950099800399202, "prompt_id": "46fcd75d760af9c5", "reading_canonical_log_prob": -370.53314208984375, "reading_distances": [24, 20, 34, 25], "reading_log_probs": [-372.80047607421875, -372.1297607421875, -372.09954833984375, -370.34783935546875], "reading_logprob_std": 1.0488575770506918, "tokenization_path_logprob_std": 132.3210500613385, "trajectory_balance_rmse": 5282.607983277252} +{"canonical_length": 328, "canonical_nll": 1.0461799807664824, "generation_rouge_l_f1": 0.18402777777777773, "generation_token_f1": 0.3715277777777778, "prompt_id": "5feba3faef17f1e8", "reading_canonical_log_prob": -343.14703369140625, "reading_distances": [27, 24, 28, 26], "reading_log_probs": [-347.0394287109375, -345.6710205078125, -345.1341552734375, -347.30926513671875], "reading_logprob_std": 1.0519264961672836, "tokenization_path_logprob_std": 361.16542806796775, "trajectory_balance_rmse": 7127.935548640731} +{"canonical_length": 232, "canonical_nll": 1.1563336602572738, "generation_rouge_l_f1": 0.2301255230125523, "generation_token_f1": 0.48117154811715485, "prompt_id": "49ec1dbe6be4038c", "reading_canonical_log_prob": -268.2694091796875, "reading_distances": [17, 24, 23, 16], "reading_log_probs": [-267.7467956542969, -267.8120422363281, -267.66229248046875, -268.5076904296875], "reading_logprob_std": 0.38852368577710394, "tokenization_path_logprob_std": 206.52090746255388, "trajectory_balance_rmse": 5248.984689126972} +{"canonical_length": 368, "canonical_nll": 1.1570902285368547, "generation_rouge_l_f1": 0.20602218700475436, "generation_token_f1": 0.4310618066561014, "prompt_id": "af5752cd92c29a4d", "reading_canonical_log_prob": -425.8092041015625, "reading_distances": [17, 20, 19, 15], "reading_log_probs": [-420.100830078125, -415.81121826171875, -419.2786865234375, -418.8201904296875], "reading_logprob_std": 1.870922138740187, "tokenization_path_logprob_std": 351.1787124124027, "trajectory_balance_rmse": 7100.449432873703} +{"canonical_length": 152, "canonical_nll": 1.5411243438720703, "generation_rouge_l_f1": 0.15094339622641512, "generation_token_f1": 0.2641509433962264, "prompt_id": "a82619f7741e0303", "reading_canonical_log_prob": -234.2509002685547, "reading_distances": [26, 16, 20, 19], "reading_log_probs": [-234.4703826904297, -236.3258819580078, -234.5677947998047, -234.57186889648438], "reading_logprob_std": 0.8958289910818422, "tokenization_path_logprob_std": 168.8678225077434, "trajectory_balance_rmse": 3797.2232763418333} +{"canonical_length": 419, "canonical_nll": 1.4199518827379176, "generation_rouge_l_f1": 0.26832641770401106, "generation_token_f1": 0.5338865836791148, "prompt_id": "9294062be51005dc", "reading_canonical_log_prob": -594.9598388671875, "reading_distances": [19, 22, 20, 26], "reading_log_probs": [-595.558837890625, -593.1666259765625, -593.5357666015625, -591.684326171875], "reading_logprob_std": 1.5965756459222888, "tokenization_path_logprob_std": 171.91596828480021, "trajectory_balance_rmse": 7265.873787399642} +{"canonical_length": 9, "canonical_nll": 0.10965875784556071, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "3e7c9dafc0fa7462", "reading_canonical_log_prob": -0.9869288206100464, "reading_distances": [23, 32, 26, 29], "reading_log_probs": [-0.7580252885818481, -0.6230512857437134, -0.7520253658294678, -0.6985834240913391], "reading_logprob_std": 0.06257180819091267, "tokenization_path_logprob_std": 23.04981679544062, "trajectory_balance_rmse": 198.9211723360297} +{"canonical_length": 12, "canonical_nll": 0.8442509174346924, "generation_rouge_l_f1": 0.75, "generation_token_f1": 0.75, "prompt_id": "c34fdb5fae4b8ff6", "reading_canonical_log_prob": -10.131011009216309, "reading_distances": [54, 44, 42, 59], "reading_log_probs": [-9.516709327697754, -8.995537757873535, -9.206143379211426, -9.4434232711792], "reading_logprob_std": 0.23711807317816885, "tokenization_path_logprob_std": 94.90342327983998, "trajectory_balance_rmse": 309.67812608784266} +{"canonical_length": 122, "canonical_nll": 1.5334817855084528, "generation_rouge_l_f1": 0.19020172910662828, "generation_token_f1": 0.3688760806916426, "prompt_id": "d51b2c6c349d91e4", "reading_canonical_log_prob": -187.08477783203125, "reading_distances": [20, 25, 30, 31], "reading_log_probs": [-186.465576171875, -184.92152404785156, -188.2984161376953, -188.69146728515625], "reading_logprob_std": 1.7432811963283235, "tokenization_path_logprob_std": 32.30516924139491, "trajectory_balance_rmse": 2563.545137571752} +{"canonical_length": 311, "canonical_nll": 1.2179278898085812, "generation_rouge_l_f1": 0.22535211267605632, "generation_token_f1": 0.41197183098591544, "prompt_id": "fc10a987783bea33", "reading_canonical_log_prob": -378.77557373046875, "reading_distances": [27, 25, 26, 25], "reading_log_probs": [-377.7646789550781, -379.98486328125, -377.33062744140625, -377.73114013671875], "reading_logprob_std": 1.204276765526735, "tokenization_path_logprob_std": 395.303431285025, "trajectory_balance_rmse": 7133.117557763526} +{"canonical_length": 31, "canonical_nll": 1.2096471478862147, "generation_rouge_l_f1": 0.15450643776824036, "generation_token_f1": 0.18025751072961374, "prompt_id": "7386d29fb448f53c", "reading_canonical_log_prob": -37.499061584472656, "reading_distances": [15, 17, 12, 17], "reading_log_probs": [-35.80206298828125, -35.93602752685547, -36.04838562011719, -37.76602554321289], "reading_logprob_std": 0.9241020073621619, "tokenization_path_logprob_std": 37.97128703684617, "trajectory_balance_rmse": 377.1805132210472} +{"canonical_length": 253, "canonical_nll": 0.804674050553514, "generation_rouge_l_f1": 0.2840690978886756, "generation_token_f1": 0.45297504798464494, "prompt_id": "d5d2ea5bf0257ac3", "reading_canonical_log_prob": -203.58253479003906, "reading_distances": [34, 33, 30, 34], "reading_log_probs": [-210.85699462890625, -208.874755859375, -208.0635986328125, -210.23416137695312], "reading_logprob_std": 1.269434093859726, "tokenization_path_logprob_std": 189.4178596315803, "trajectory_balance_rmse": 5684.063125647418} +{"canonical_length": 379, "canonical_nll": 1.1455335793205805, "generation_rouge_l_f1": 0.19090909090909092, "generation_token_f1": 0.4151515151515151, "prompt_id": "0a54bd903d89ace0", "reading_canonical_log_prob": -434.1572265625, "reading_distances": [21, 18, 22, 20], "reading_log_probs": [-432.7702941894531, -433.1107177734375, -434.595703125, -433.6072998046875], "reading_logprob_std": 0.7946330772837533, "tokenization_path_logprob_std": 238.95249290884084, "trajectory_balance_rmse": 7234.102616802063} +{"canonical_length": 128, "canonical_nll": 0.9470404982566833, "generation_rouge_l_f1": 0.2455242966751918, "generation_token_f1": 0.42966751918158563, "prompt_id": "49a85c182167a715", "reading_canonical_log_prob": -121.22118377685547, "reading_distances": [15, 10, 18, 18], "reading_log_probs": [-119.58869171142578, -120.02135467529297, -119.70310974121094, -119.6391830444336], "reading_logprob_std": 0.19456355358738509, "tokenization_path_logprob_std": 70.65800031483904, "trajectory_balance_rmse": 2952.2704575905186} +{"canonical_length": 127, "canonical_nll": 1.1865264411986343, "generation_rouge_l_f1": 0.19121447028423771, "generation_token_f1": 0.34625322997416014, "prompt_id": "50a15e61cf0ab437", "reading_canonical_log_prob": -150.68885803222656, "reading_distances": [35, 28, 41, 28], "reading_log_probs": [-162.87698364257812, -163.90542602539062, -166.79782104492188, -163.85586547851562], "reading_logprob_std": 1.6934279108422743, "tokenization_path_logprob_std": 109.66056437976891, "trajectory_balance_rmse": 2970.5404245165037} +{"canonical_length": 14, "canonical_nll": 0.6810308865138462, "generation_rouge_l_f1": 0.5517241379310345, "generation_token_f1": 0.6206896551724138, "prompt_id": "394139a0ae5c1d75", "reading_canonical_log_prob": -9.534432411193848, "reading_distances": [20, 29, 26, 29], "reading_log_probs": [-7.9141716957092285, -7.753642559051514, -8.237651824951172, -7.879474639892578], "reading_logprob_std": 0.20615577875214003, "tokenization_path_logprob_std": 45.20779501678558, "trajectory_balance_rmse": 242.11462981658335} +{"canonical_length": 58, "canonical_nll": 0.34860397207325905, "generation_rouge_l_f1": 0.17391304347826086, "generation_token_f1": 0.18972332015810275, "prompt_id": "ee3925e13ca1cdfe", "reading_canonical_log_prob": -20.219030380249023, "reading_distances": [21, 20, 14, 22], "reading_log_probs": [-21.430187225341797, -21.1337890625, -21.23797035217285, -21.12539291381836], "reading_logprob_std": 0.14180288127682952, "tokenization_path_logprob_std": 17.28647363181913, "trajectory_balance_rmse": 895.5471612011427} +{"canonical_length": 293, "canonical_nll": 1.1886381077685046, "generation_rouge_l_f1": 0.19965576592082618, "generation_token_f1": 0.40619621342512907, "prompt_id": "e5cf7a10fe204155", "reading_canonical_log_prob": -348.2709655761719, "reading_distances": [28, 24, 27, 30], "reading_log_probs": [-355.0102233886719, -351.08740234375, -355.4532775878906, -350.96697998046875], "reading_logprob_std": 2.434729171236553, "tokenization_path_logprob_std": 136.12727809746275, "trajectory_balance_rmse": 5712.2448645911145} +{"canonical_length": 302, "canonical_nll": 1.1714271393832782, "generation_rouge_l_f1": 0.24956063268892797, "generation_token_f1": 0.49209138840070293, "prompt_id": "a6e94611dc8387e2", "reading_canonical_log_prob": -353.77099609375, "reading_distances": [32, 34, 29, 28], "reading_log_probs": [-354.58984375, -354.755859375, -354.8782043457031, -356.406005859375], "reading_logprob_std": 0.8406984697934928, "tokenization_path_logprob_std": 198.20436822528646, "trajectory_balance_rmse": 6589.440264047962} +{"canonical_length": 195, "canonical_nll": 1.807507793719952, "generation_rouge_l_f1": 0.26229508196721313, "generation_token_f1": 0.4964871194379392, "prompt_id": "747395c5d51f280d", "reading_canonical_log_prob": -352.4640197753906, "reading_distances": [21, 17, 16, 15], "reading_log_probs": [-357.0328674316406, -359.7287292480469, -355.6176452636719, -358.76593017578125], "reading_logprob_std": 1.826055006951491, "tokenization_path_logprob_std": 156.320807299839, "trajectory_balance_rmse": 4053.7248684216934} +{"canonical_length": 35, "canonical_nll": 0.8779841831752232, "generation_rouge_l_f1": 0.15767634854771787, "generation_token_f1": 0.16597510373443983, "prompt_id": "c31ebbb64961b272", "reading_canonical_log_prob": -30.729446411132812, "reading_distances": [16, 17, 15, 20], "reading_log_probs": [-30.2487850189209, -30.75258445739746, -29.80858039855957, -31.358903884887695], "reading_logprob_std": 0.6672223331398648, "tokenization_path_logprob_std": 21.52119322597253, "trajectory_balance_rmse": 388.2551219715132} +{"canonical_length": 111, "canonical_nll": 1.3274239205025338, "generation_rouge_l_f1": 0.266304347826087, "generation_token_f1": 0.40760869565217395, "prompt_id": "0a9b1baeb01418e9", "reading_canonical_log_prob": -147.34405517578125, "reading_distances": [25, 28, 28, 22], "reading_log_probs": [-147.40130615234375, -150.01181030273438, -148.99758911132812, -146.29019165039062], "reading_logprob_std": 1.6534438220187175, "tokenization_path_logprob_std": 94.02274731109475, "trajectory_balance_rmse": 2368.36702889578} +{"canonical_length": 86, "canonical_nll": 1.1337791265443313, "generation_rouge_l_f1": 0.1901840490797546, "generation_token_f1": 0.31901840490797545, "prompt_id": "ffff4b859d3e455e", "reading_canonical_log_prob": -97.5050048828125, "reading_distances": [18, 19, 17, 19], "reading_log_probs": [-96.76396179199219, -96.01725769042969, -96.00858306884766, -95.81048583984375], "reading_logprob_std": 0.4202531758147346, "tokenization_path_logprob_std": 53.843673551574916, "trajectory_balance_rmse": 2220.3809024639163} +{"canonical_length": 12, "canonical_nll": 0.9338038762410482, "generation_rouge_l_f1": 0.07905138339920949, "generation_token_f1": 0.07905138339920949, "prompt_id": "a9db5a2fccb43cd0", "reading_canonical_log_prob": -11.205646514892578, "reading_distances": [21, 28, 23, 23], "reading_log_probs": [-13.175929069519043, -12.529105186462402, -11.544240951538086, -12.00660514831543], "reading_logprob_std": 0.7014776180980348, "tokenization_path_logprob_std": 94.16334176893587, "trajectory_balance_rmse": 345.7054579066935} +{"canonical_length": 115, "canonical_nll": 1.888032863451087, "generation_rouge_l_f1": 0.0631578947368421, "generation_token_f1": 0.08421052631578949, "prompt_id": "6d2c7ff3221f2284", "reading_canonical_log_prob": -217.123779296875, "reading_distances": [11, 13, 15, 13], "reading_log_probs": [-216.72023010253906, -216.40353393554688, -216.14718627929688, -216.79119873046875], "reading_logprob_std": 0.2978341796823757, "tokenization_path_logprob_std": 138.65400989265515, "trajectory_balance_rmse": 1712.1877148171777} +{"canonical_length": 16, "canonical_nll": 0.9777133464813232, "generation_rouge_l_f1": 0.4615384615384615, "generation_token_f1": 0.5384615384615385, "prompt_id": "686af343746a5e07", "reading_canonical_log_prob": -15.643413543701172, "reading_distances": [22, 25, 24, 26], "reading_log_probs": [-15.488640785217285, -14.960823059082031, -14.90290641784668, -14.34949016571045], "reading_logprob_std": 0.4657155987512665, "tokenization_path_logprob_std": 70.31168895256097, "trajectory_balance_rmse": 429.12785468928917} +{"canonical_length": 8, "canonical_nll": 3.39987850189209, "generation_rouge_l_f1": 0.0242914979757085, "generation_token_f1": 0.0242914979757085, "prompt_id": "90063df346a3b4e5", "reading_canonical_log_prob": -27.19902801513672, "reading_distances": [19, 18, 22, 26], "reading_log_probs": [-26.30414581298828, -25.88678741455078, -26.30374526977539, -26.383771896362305], "reading_logprob_std": 0.22505188207250557, "tokenization_path_logprob_std": 26.577071947822823, "trajectory_balance_rmse": 255.0888965721798} +{"canonical_length": 276, "canonical_nll": 0.8314038705134738, "generation_rouge_l_f1": 0.2760084925690021, "generation_token_f1": 0.4968152866242038, "prompt_id": "73f47d65b9b8ef25", "reading_canonical_log_prob": -229.46746826171875, "reading_distances": [25, 20, 25, 29], "reading_log_probs": [-239.46762084960938, -236.9219512939453, -241.38829040527344, -235.41912841796875], "reading_logprob_std": 2.65200007042424, "tokenization_path_logprob_std": 309.3062543809177, "trajectory_balance_rmse": 5129.549346389417} +{"canonical_length": 73, "canonical_nll": 0.7594096040072507, "generation_rouge_l_f1": 0.13513513513513511, "generation_token_f1": 0.13513513513513511, "prompt_id": "96bcd59d4201d57e", "reading_canonical_log_prob": -55.4369010925293, "reading_distances": [19, 20, 18, 20], "reading_log_probs": [-56.28223419189453, -57.78461456298828, -56.17936325073242, -55.93069839477539], "reading_logprob_std": 0.839988976687989, "tokenization_path_logprob_std": 93.5028552042117, "trajectory_balance_rmse": 1301.8379724004587} +{"canonical_length": 159, "canonical_nll": 1.295420232808815, "generation_rouge_l_f1": 0.19018404907975459, "generation_token_f1": 0.4539877300613497, "prompt_id": "3d7e3fa66ce5f5da", "reading_canonical_log_prob": -205.97181701660156, "reading_distances": [24, 22, 25, 26], "reading_log_probs": [-208.29425048828125, -209.22279357910156, -208.800537109375, -205.72357177734375], "reading_logprob_std": 1.571026120293179, "tokenization_path_logprob_std": 107.99785774688804, "trajectory_balance_rmse": 3893.7657225299663} +{"canonical_length": 40, "canonical_nll": 0.8552804946899414, "generation_rouge_l_f1": 0.1276595744680851, "generation_token_f1": 0.1276595744680851, "prompt_id": "d99d42ddbcbf4292", "reading_canonical_log_prob": -34.211219787597656, "reading_distances": [18, 15, 21, 17], "reading_log_probs": [-33.55882263183594, -33.00991439819336, -33.94755935668945, -34.79880905151367], "reading_logprob_std": 0.7524366530925337, "tokenization_path_logprob_std": 91.97065251673348, "trajectory_balance_rmse": 703.0713098263194} +{"canonical_length": 15, "canonical_nll": 0.2553767204284668, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "4bb936b473e4256b", "reading_canonical_log_prob": -3.830650806427002, "reading_distances": [17, 17, 23, 18], "reading_log_probs": [-3.6684441566467285, -3.773542881011963, -3.7973270416259766, -4.075847625732422], "reading_logprob_std": 0.17396400184962638, "tokenization_path_logprob_std": 44.20375824713775, "trajectory_balance_rmse": 218.21836124207326} +{"canonical_length": 378, "canonical_nll": 1.3691816380415014, "generation_rouge_l_f1": 0.24561403508771928, "generation_token_f1": 0.45933014354066987, "prompt_id": "d413361106deda8e", "reading_canonical_log_prob": -517.5506591796875, "reading_distances": [24, 24, 35, 20], "reading_log_probs": [-518.3969116210938, -517.5999755859375, -516.819580078125, -517.4918823242188], "reading_logprob_std": 0.6464551725063403, "tokenization_path_logprob_std": 115.79639774818337, "trajectory_balance_rmse": 8180.65350627641} +{"canonical_length": 370, "canonical_nll": 0.6173069309543919, "generation_rouge_l_f1": 0.22260869565217392, "generation_token_f1": 0.38260869565217387, "prompt_id": "2807bad41faf5ef6", "reading_canonical_log_prob": -228.403564453125, "reading_distances": [20, 12, 12, 15], "reading_log_probs": [-229.48294067382812, -228.6659393310547, -230.21221923828125, -229.26547241210938], "reading_logprob_std": 0.638577969864645, "tokenization_path_logprob_std": 145.056117646452, "trajectory_balance_rmse": 5727.837074433934} +{"canonical_length": 76, "canonical_nll": 1.4671808543958162, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.32894736842105265, "prompt_id": "f47b46ea69a09ffc", "reading_canonical_log_prob": -111.50574493408203, "reading_distances": [26, 30, 23, 20], "reading_log_probs": [-109.88687133789062, -109.69242095947266, -110.18812561035156, -111.65365600585938], "reading_logprob_std": 0.8892899085575752, "tokenization_path_logprob_std": 84.33336995317687, "trajectory_balance_rmse": 1820.3084278547647} +{"canonical_length": 148, "canonical_nll": 1.3126195959142737, "generation_rouge_l_f1": 0.22062350119904078, "generation_token_f1": 0.38848920863309355, "prompt_id": "abdba20a44568a50", "reading_canonical_log_prob": -194.2677001953125, "reading_distances": [31, 31, 31, 29], "reading_log_probs": [-195.30508422851562, -194.7488555908203, -201.40365600585938, -199.85848999023438], "reading_logprob_std": 3.304261671528422, "tokenization_path_logprob_std": 178.74975415263742, "trajectory_balance_rmse": 3458.1475248335523} +{"canonical_length": 19, "canonical_nll": 0.44361405623586553, "generation_rouge_l_f1": 0.07171314741035857, "generation_token_f1": 0.07171314741035857, "prompt_id": "f10790becafaebfa", "reading_canonical_log_prob": -8.428667068481445, "reading_distances": [31, 29, 23, 29], "reading_log_probs": [-10.856468200683594, -11.386817932128906, -8.395136833190918, -8.813575744628906], "reading_logprob_std": 1.4792919139780647, "tokenization_path_logprob_std": 79.51909457405687, "trajectory_balance_rmse": 280.3268027338847} +{"canonical_length": 7, "canonical_nll": 0.14031061104365758, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "3eb075dbec50f70d", "reading_canonical_log_prob": -0.982174277305603, "reading_distances": [65, 69, 68, 66], "reading_log_probs": [-1.8029026985168457, -1.4106706380844116, -1.89762544631958, -1.3137661218643188], "reading_logprob_std": 0.28715271199634645, "tokenization_path_logprob_std": 54.706424876379785, "trajectory_balance_rmse": 166.5452376771586} +{"canonical_length": 5, "canonical_nll": 1.6029743194580077, "generation_rouge_l_f1": 0.02909090909090909, "generation_token_f1": 0.02909090909090909, "prompt_id": "56ce6de59318b6ea", "reading_canonical_log_prob": -8.014871597290039, "reading_distances": [33, 24, 33, 24], "reading_log_probs": [-11.437002182006836, -10.410707473754883, -10.683293342590332, -11.684301376342773], "reading_logprob_std": 0.6042127317533741, "tokenization_path_logprob_std": 42.03812751982711, "trajectory_balance_rmse": 93.88558240765607} +{"canonical_length": 128, "canonical_nll": 1.188525676727295, "generation_rouge_l_f1": 0.20603015075376885, "generation_token_f1": 0.3919597989949749, "prompt_id": "7b4da759d302cb91", "reading_canonical_log_prob": -152.13128662109375, "reading_distances": [51, 42, 40, 46], "reading_log_probs": [-150.95947265625, -151.0792694091797, -148.75177001953125, -154.52976989746094], "reading_logprob_std": 2.3864865074446495, "tokenization_path_logprob_std": 49.080913828961876, "trajectory_balance_rmse": 2613.8257204787037} +{"canonical_length": 179, "canonical_nll": 1.8901997997774092, "generation_rouge_l_f1": 0.26809651474530827, "generation_token_f1": 0.5308310991957105, "prompt_id": "e1ce0d3f6081406f", "reading_canonical_log_prob": -338.34576416015625, "reading_distances": [15, 18, 20, 21], "reading_log_probs": [-335.556640625, -336.0940246582031, -334.71484375, -336.22039794921875], "reading_logprob_std": 0.6845181393003638, "tokenization_path_logprob_std": 227.7015240013809, "trajectory_balance_rmse": 4383.536565074934} +{"canonical_length": 193, "canonical_nll": 0.9583542581666936, "generation_rouge_l_f1": 0.13774104683195593, "generation_token_f1": 0.22038567493112945, "prompt_id": "50afdbd8ae3129b4", "reading_canonical_log_prob": -184.96237182617188, "reading_distances": [20, 20, 21, 18], "reading_log_probs": [-182.5688018798828, -183.07427978515625, -183.03912353515625, -182.40380859375], "reading_logprob_std": 0.33644288913591736, "tokenization_path_logprob_std": 68.00614052498787, "trajectory_balance_rmse": 2740.0653783784555} +{"canonical_length": 308, "canonical_nll": 0.9997557602919541, "generation_rouge_l_f1": 0.2127659574468085, "generation_token_f1": 0.3971631205673759, "prompt_id": "517e8812395e8d56", "reading_canonical_log_prob": -307.9247741699219, "reading_distances": [13, 15, 16, 16], "reading_log_probs": [-304.265625, -306.04833984375, -302.11749267578125, -305.262939453125], "reading_logprob_std": 1.701700851979084, "tokenization_path_logprob_std": 245.45453309864914, "trajectory_balance_rmse": 5150.098664947425} +{"canonical_length": 9, "canonical_nll": 1.2406654357910156, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.4, "prompt_id": "7febc4a2c9567c65", "reading_canonical_log_prob": -11.16598892211914, "reading_distances": [33, 34, 34, 39], "reading_log_probs": [-14.984333038330078, -13.4361572265625, -13.152185440063477, -16.306835174560547], "reading_logprob_std": 1.4656020341575957, "tokenization_path_logprob_std": 53.165839999821664, "trajectory_balance_rmse": 122.97158814534632} +{"canonical_length": 13, "canonical_nll": 0.97162232032189, "generation_rouge_l_f1": 0.4761904761904762, "generation_token_f1": 0.761904761904762, "prompt_id": "a3b4630e36098b06", "reading_canonical_log_prob": -12.63109016418457, "reading_distances": [39, 44, 46, 44], "reading_log_probs": [-12.292699813842773, -12.465139389038086, -14.309663772583008, -12.545681953430176], "reading_logprob_std": 0.9434982841078373, "tokenization_path_logprob_std": 75.26985543249056, "trajectory_balance_rmse": 379.5621531479517} +{"canonical_length": 67, "canonical_nll": 1.1351552934788947, "generation_rouge_l_f1": 0.2042042042042042, "generation_token_f1": 0.2942942942942943, "prompt_id": "0d478580ddc19eb6", "reading_canonical_log_prob": -76.05540466308594, "reading_distances": [26, 22, 27, 24], "reading_log_probs": [-74.94329071044922, -75.30939483642578, -75.75358581542969, -74.65553283691406], "reading_logprob_std": 0.4746908174323221, "tokenization_path_logprob_std": 141.80991236792534, "trajectory_balance_rmse": 1534.8205591139174} +{"canonical_length": 5, "canonical_nll": 2.7764169692993166, "generation_rouge_l_f1": 0.5714285714285715, "generation_token_f1": 0.5714285714285715, "prompt_id": "7c79a39b53212c05", "reading_canonical_log_prob": -13.882084846496582, "reading_distances": [33, 32, 33, 33], "reading_log_probs": [-14.216672897338867, -13.943645477294922, -13.835407257080078, -14.086967468261719], "reading_logprob_std": 0.16640067210641502, "tokenization_path_logprob_std": 22.311230915654047, "trajectory_balance_rmse": 108.19520650656216} +{"canonical_length": 89, "canonical_nll": 1.020350852709138, "generation_rouge_l_f1": 0.25925925925925924, "generation_token_f1": 0.41975308641975306, "prompt_id": "1791e9720c6a20af", "reading_canonical_log_prob": -90.81122589111328, "reading_distances": [25, 25, 25, 20], "reading_log_probs": [-92.8751220703125, -92.75652313232422, -90.41499328613281, -93.14276123046875], "reading_logprob_std": 1.2652598517729037, "tokenization_path_logprob_std": 143.31721868737628, "trajectory_balance_rmse": 2352.2236458336706} +{"canonical_length": 14, "canonical_nll": 1.5976005281720842, "generation_rouge_l_f1": 0.13333333333333333, "generation_token_f1": 0.13333333333333333, "prompt_id": "a58d338f47d878b7", "reading_canonical_log_prob": -22.36640739440918, "reading_distances": [9, 9, 9, 8], "reading_log_probs": [-22.304275512695312, -21.176013946533203, -21.29090118408203, -22.387611389160156], "reading_logprob_std": 0.6449022712868616, "tokenization_path_logprob_std": 56.04494056688055, "trajectory_balance_rmse": 287.3485049775588} +{"canonical_length": 5, "canonical_nll": 1.9778274536132812, "generation_rouge_l_f1": 0.39999999999999997, "generation_token_f1": 0.39999999999999997, "prompt_id": "c979ade41611b06d", "reading_canonical_log_prob": -9.889137268066406, "reading_distances": [30, 29, 36, 38], "reading_log_probs": [-11.297682762145996, -9.650337219238281, -10.28773021697998, -11.099953651428223], "reading_logprob_std": 0.7604939635944199, "tokenization_path_logprob_std": 22.717915192004135, "trajectory_balance_rmse": 107.03876534201154} +{"canonical_length": 197, "canonical_nll": 1.0739488940553616, "generation_rouge_l_f1": 0.2606635071090047, "generation_token_f1": 0.4075829383886256, "prompt_id": "58bae05f46b8fa96", "reading_canonical_log_prob": -211.56793212890625, "reading_distances": [21, 21, 21, 18], "reading_log_probs": [-210.0673370361328, -209.57525634765625, -209.3268585205078, -209.74020385742188], "reading_logprob_std": 0.31054139307369844, "tokenization_path_logprob_std": 105.59516477046309, "trajectory_balance_rmse": 4170.0665085379505} +{"canonical_length": 91, "canonical_nll": 1.1564902840079843, "generation_rouge_l_f1": 0.35175879396984927, "generation_token_f1": 0.6030150753768844, "prompt_id": "005445d3abeec9e6", "reading_canonical_log_prob": -105.24061584472656, "reading_distances": [18, 23, 20, 21], "reading_log_probs": [-106.02078247070312, -105.6156005859375, -104.79763793945312, -105.0782699584961], "reading_logprob_std": 0.5465898953266399, "tokenization_path_logprob_std": 165.73488576721383, "trajectory_balance_rmse": 2138.68032606701} +{"canonical_length": 48, "canonical_nll": 1.201575517654419, "generation_rouge_l_f1": 0.1042654028436019, "generation_token_f1": 0.11374407582938388, "prompt_id": "3ba7853ae081fbb0", "reading_canonical_log_prob": -57.67562484741211, "reading_distances": [23, 12, 18, 19], "reading_log_probs": [-60.69709777832031, -58.0443229675293, -59.254783630371094, -59.38795852661133], "reading_logprob_std": 1.084728728124096, "tokenization_path_logprob_std": 46.65786811724992, "trajectory_balance_rmse": 932.6514786511143} +{"canonical_length": 60, "canonical_nll": 1.0229727427164714, "generation_rouge_l_f1": 0.4125, "generation_token_f1": 0.4874999999999999, "prompt_id": "19c4e3cd0eb041a5", "reading_canonical_log_prob": -61.37836456298828, "reading_distances": [36, 31, 31, 26], "reading_log_probs": [-63.735267639160156, -61.35710906982422, -62.213932037353516, -60.79130554199219], "reading_logprob_std": 1.2817682577494554, "tokenization_path_logprob_std": 168.34057704331568, "trajectory_balance_rmse": 1599.9325922089529} +{"canonical_length": 139, "canonical_nll": 1.3279934889978642, "generation_rouge_l_f1": 0.24806201550387594, "generation_token_f1": 0.4883720930232558, "prompt_id": "9a7d94fcb2412368", "reading_canonical_log_prob": -184.59109497070312, "reading_distances": [22, 32, 29, 23], "reading_log_probs": [-185.56695556640625, -185.39328002929688, -185.57388305664062, -184.76942443847656], "reading_logprob_std": 0.38026679256695317, "tokenization_path_logprob_std": 179.84914703615996, "trajectory_balance_rmse": 3383.882760837477} +{"canonical_length": 27, "canonical_nll": 1.2459763420952692, "generation_rouge_l_f1": 0.14545454545454545, "generation_token_f1": 0.1818181818181818, "prompt_id": "2a25293cb727ba1d", "reading_canonical_log_prob": -33.641361236572266, "reading_distances": [31, 27, 32, 29], "reading_log_probs": [-33.0792236328125, -32.038124084472656, -34.54222869873047, -32.75164031982422], "reading_logprob_std": 1.0534639291141348, "tokenization_path_logprob_std": 60.78013996208352, "trajectory_balance_rmse": 748.4303709375904} +{"canonical_length": 33, "canonical_nll": 1.8488179987127131, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.39999999999999997, "prompt_id": "cc6712ec80ac0d9e", "reading_canonical_log_prob": -61.01099395751953, "reading_distances": [29, 23, 29, 25], "reading_log_probs": [-60.905975341796875, -61.024559020996094, -61.028385162353516, -60.95022964477539], "reading_logprob_std": 0.0596185631942748, "tokenization_path_logprob_std": 80.83736511539448, "trajectory_balance_rmse": 714.9604703502888} +{"canonical_length": 48, "canonical_nll": 0.5925509929656982, "generation_rouge_l_f1": 0.4050632911392405, "generation_token_f1": 0.5822784810126582, "prompt_id": "aa6e8f58031f9e55", "reading_canonical_log_prob": -28.442447662353516, "reading_distances": [30, 38, 36, 39], "reading_log_probs": [-27.91506004333496, -30.196517944335938, -27.894498825073242, -30.006336212158203], "reading_logprob_std": 1.27063721541002, "tokenization_path_logprob_std": 105.16158294640438, "trajectory_balance_rmse": 1018.2359726792948} +{"canonical_length": 259, "canonical_nll": 1.5465118526031612, "generation_rouge_l_f1": 0.25918762088974856, "generation_token_f1": 0.5183752417794971, "prompt_id": "ef58deb0251c5bd3", "reading_canonical_log_prob": -400.54656982421875, "reading_distances": [24, 25, 18, 23], "reading_log_probs": [-403.3681335449219, -402.04510498046875, -403.7904968261719, -402.50994873046875], "reading_logprob_std": 0.7941217545101334, "tokenization_path_logprob_std": 204.14290332050817, "trajectory_balance_rmse": 5467.781295981453} +{"canonical_length": 136, "canonical_nll": 1.4881877899169922, "generation_rouge_l_f1": 0.1442786069651741, "generation_token_f1": 0.31840796019900497, "prompt_id": "209efb2fe475a408", "reading_canonical_log_prob": -202.39353942871094, "reading_distances": [25, 28, 23, 24], "reading_log_probs": [-201.84500122070312, -201.90452575683594, -203.498291015625, -204.34129333496094], "reading_logprob_std": 1.2300732185670324, "tokenization_path_logprob_std": 102.24073314924634, "trajectory_balance_rmse": 3549.751630717675} +{"canonical_length": 16, "canonical_nll": 0.9015076756477356, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.2222222222222222, "prompt_id": "aa1f86a1c5954cf7", "reading_canonical_log_prob": -14.42412281036377, "reading_distances": [28, 29, 27, 28], "reading_log_probs": [-16.042272567749023, -18.59197425842285, -16.133586883544922, -16.080297470092773], "reading_logprob_std": 1.2538537741856062, "tokenization_path_logprob_std": 16.36198908537132, "trajectory_balance_rmse": 154.59952266742897} +{"canonical_length": 8, "canonical_nll": 0.029944026842713356, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "878c5da5498115bc", "reading_canonical_log_prob": -0.23955221474170685, "reading_distances": [32, 33, 32, 30], "reading_log_probs": [-0.5222495198249817, -0.46769392490386963, -0.5670494437217712, -0.44409462809562683], "reading_logprob_std": 0.055255366130554, "tokenization_path_logprob_std": 62.82695921595781, "trajectory_balance_rmse": 148.189349419005} +{"canonical_length": 30, "canonical_nll": 0.1634758472442627, "generation_rouge_l_f1": 0.21212121212121213, "generation_token_f1": 0.33333333333333337, "prompt_id": "3e49b0b70cea783a", "reading_canonical_log_prob": -4.904275417327881, "reading_distances": [41, 34, 38, 39], "reading_log_probs": [-4.641483783721924, -3.737514019012451, -4.564767360687256, -3.9486610889434814], "reading_logprob_std": 0.4482900982257286, "tokenization_path_logprob_std": 97.81113600638534, "trajectory_balance_rmse": 591.3214764223189} +{"canonical_length": 406, "canonical_nll": 1.2239203241658327, "generation_rouge_l_f1": 0.18210862619808307, "generation_token_f1": 0.3897763578274761, "prompt_id": "50289ab6cb9cf49e", "reading_canonical_log_prob": -496.9116516113281, "reading_distances": [25, 21, 20, 23], "reading_log_probs": [-496.0755615234375, -496.37701416015625, -495.49517822265625, -496.3677673339844], "reading_logprob_std": 0.413545098850269, "tokenization_path_logprob_std": 151.18706346543175, "trajectory_balance_rmse": 7164.7762399888725} +{"canonical_length": 168, "canonical_nll": 0.2868897574288504, "generation_rouge_l_f1": 0.28211586901763225, "generation_token_f1": 0.4937027707808564, "prompt_id": "97290896fed51229", "reading_canonical_log_prob": -48.197479248046875, "reading_distances": [30, 30, 34, 39], "reading_log_probs": [-49.99075698852539, -49.39812088012695, -50.8140869140625, -49.89265060424805], "reading_logprob_std": 0.58717394613532, "tokenization_path_logprob_std": 54.96695763580788, "trajectory_balance_rmse": 1533.4012412462214} +{"canonical_length": 36, "canonical_nll": 1.0408379236857097, "generation_rouge_l_f1": 0.1048951048951049, "generation_token_f1": 0.14685314685314682, "prompt_id": "11c3ac95688ce9b6", "reading_canonical_log_prob": -37.47016525268555, "reading_distances": [23, 21, 17, 18], "reading_log_probs": [-38.191463470458984, -40.033355712890625, -38.669185638427734, -39.197113037109375], "reading_logprob_std": 0.7890440258386187, "tokenization_path_logprob_std": 37.94888626611262, "trajectory_balance_rmse": 668.4502882672537} +{"canonical_length": 12, "canonical_nll": 0.569157878557841, "generation_rouge_l_f1": 0.08333333333333333, "generation_token_f1": 0.08333333333333333, "prompt_id": "b6d0ecb07aebae2f", "reading_canonical_log_prob": -6.829894542694092, "reading_distances": [26, 27, 26, 27], "reading_log_probs": [-5.8783721923828125, -6.157473087310791, -6.229376316070557, -5.810559272766113], "reading_logprob_std": 0.20547237877077193, "tokenization_path_logprob_std": 48.9566405960287, "trajectory_balance_rmse": 245.38142097501458} +{"canonical_length": 27, "canonical_nll": 1.1758640430591725, "generation_rouge_l_f1": 0.21621621621621623, "generation_token_f1": 0.2702702702702703, "prompt_id": "a7179a571daab3f1", "reading_canonical_log_prob": -31.748329162597656, "reading_distances": [28, 19, 18, 19], "reading_log_probs": [-32.03996658325195, -31.600135803222656, -32.509498596191406, -31.78348731994629], "reading_logprob_std": 0.3944768560910573, "tokenization_path_logprob_std": 49.10335349513169, "trajectory_balance_rmse": 427.5326709248083} +{"canonical_length": 226, "canonical_nll": 0.9710305813139519, "generation_rouge_l_f1": 0.09999999999999999, "generation_token_f1": 0.10833333333333332, "prompt_id": "a32be7c1cc7860a7", "reading_canonical_log_prob": -219.45291137695312, "reading_distances": [38, 29, 34, 38], "reading_log_probs": [-228.09255981445312, -225.63265991210938, -227.04718017578125, -227.28590393066406], "reading_logprob_std": 1.0241131214259893, "tokenization_path_logprob_std": 180.98627267974342, "trajectory_balance_rmse": 3515.769506599248} +{"canonical_length": 317, "canonical_nll": 1.1597601953741128, "generation_rouge_l_f1": 0.16695059625212946, "generation_token_f1": 0.3747870528109029, "prompt_id": "7ebfd1446f72fed6", "reading_canonical_log_prob": -367.64398193359375, "reading_distances": [25, 29, 19, 24], "reading_log_probs": [-367.27166748046875, -369.23138427734375, -368.0714416503906, -368.4842529296875], "reading_logprob_std": 0.8177498302900851, "tokenization_path_logprob_std": 233.1598027089058, "trajectory_balance_rmse": 7449.601087944525} +{"canonical_length": 7, "canonical_nll": 1.7566142763410295, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.4444444444444444, "prompt_id": "05c61d7a9c151e3a", "reading_canonical_log_prob": -12.296299934387207, "reading_distances": [48, 57, 44, 50], "reading_log_probs": [-11.082387924194336, -13.782154083251953, -12.58355712890625, -10.59069538116455], "reading_logprob_std": 1.4542337358940252, "tokenization_path_logprob_std": 53.087546428508254, "trajectory_balance_rmse": 205.8870221091891} +{"canonical_length": 20, "canonical_nll": 1.6731128692626953, "generation_rouge_l_f1": 0.5333333333333333, "generation_token_f1": 0.6222222222222222, "prompt_id": "3b1c1940eb0db061", "reading_canonical_log_prob": -33.462257385253906, "reading_distances": [19, 25, 18, 21], "reading_log_probs": [-32.956993103027344, -31.751056671142578, -31.863332748413086, -31.598960876464844], "reading_logprob_std": 0.6191565978778153, "tokenization_path_logprob_std": 51.17552553743377, "trajectory_balance_rmse": 531.8086062751914} +{"canonical_length": 191, "canonical_nll": 1.4629066027895943, "generation_rouge_l_f1": 0.23076923076923078, "generation_token_f1": 0.47058823529411764, "prompt_id": "13ae7fb963a43898", "reading_canonical_log_prob": -279.4151611328125, "reading_distances": [42, 40, 46, 36], "reading_log_probs": [-282.5049133300781, -284.61865234375, -283.449462890625, -284.23602294921875], "reading_logprob_std": 0.9349194689188536, "tokenization_path_logprob_std": 51.07162068184312, "trajectory_balance_rmse": 3241.191994208054} +{"canonical_length": 346, "canonical_nll": 1.7418498662165824, "generation_rouge_l_f1": 0.19387755102040818, "generation_token_f1": 0.38775510204081637, "prompt_id": "6aeff1f7dd4bd317", "reading_canonical_log_prob": -602.6800537109375, "reading_distances": [22, 24, 20, 26], "reading_log_probs": [-601.7655029296875, -602.2252197265625, -606.3515014648438, -606.5467529296875], "reading_logprob_std": 2.5794548358809117, "tokenization_path_logprob_std": 467.4580700716866, "trajectory_balance_rmse": 6783.91812068726} +{"canonical_length": 35, "canonical_nll": 2.7437711443219865, "generation_rouge_l_f1": 0.36, "generation_token_f1": 0.39999999999999997, "prompt_id": "a5fa29779141ee07", "reading_canonical_log_prob": -96.03199005126953, "reading_distances": [15, 16, 16, 13], "reading_log_probs": [-100.54917907714844, -99.64032745361328, -97.92080688476562, -95.13642120361328], "reading_logprob_std": 2.3809491379436616, "tokenization_path_logprob_std": 68.4121122952299, "trajectory_balance_rmse": 758.4713762837562} +{"canonical_length": 44, "canonical_nll": 1.6000789295543323, "generation_rouge_l_f1": 0.10847457627118644, "generation_token_f1": 0.17627118644067796, "prompt_id": "635690e839a377b9", "reading_canonical_log_prob": -70.40347290039062, "reading_distances": [32, 30, 23, 35], "reading_log_probs": [-70.13699340820312, -72.07859802246094, -70.26663208007812, -70.79602813720703], "reading_logprob_std": 0.8864455101511259, "tokenization_path_logprob_std": 90.81254579333182, "trajectory_balance_rmse": 1372.5376934293097} +{"canonical_length": 124, "canonical_nll": 1.4634630756993448, "generation_rouge_l_f1": 0.21551724137931036, "generation_token_f1": 0.35344827586206895, "prompt_id": "908852670070fd4b", "reading_canonical_log_prob": -181.46942138671875, "reading_distances": [21, 20, 23, 20], "reading_log_probs": [-181.16456604003906, -176.45835876464844, -177.4668731689453, -178.4903564453125], "reading_logprob_std": 2.024153122075718, "tokenization_path_logprob_std": 100.30565433247683, "trajectory_balance_rmse": 2926.109278690032} +{"canonical_length": 14, "canonical_nll": 2.7929464067731584, "generation_rouge_l_f1": 0.2727272727272727, "generation_token_f1": 0.2727272727272727, "prompt_id": "410cbaa296c6e2c6", "reading_canonical_log_prob": -39.10124969482422, "reading_distances": [25, 24, 23, 21], "reading_log_probs": [-39.19443893432617, -38.837379455566406, -38.71991729736328, -39.441436767578125], "reading_logprob_std": 0.33077473775629695, "tokenization_path_logprob_std": 36.3381820550752, "trajectory_balance_rmse": 375.9172734416584} +{"canonical_length": 24, "canonical_nll": 1.1961379845937092, "generation_rouge_l_f1": 0.5106382978723404, "generation_token_f1": 0.5531914893617023, "prompt_id": "b9eb5c3d26d069d5", "reading_canonical_log_prob": -28.707311630249023, "reading_distances": [15, 12, 16, 17], "reading_log_probs": [-30.672237396240234, -29.04948616027832, -32.767364501953125, -28.29533576965332], "reading_logprob_std": 1.980381774285574, "tokenization_path_logprob_std": 75.48766235948933, "trajectory_balance_rmse": 507.77725417795267} +{"canonical_length": 120, "canonical_nll": 0.7457335789998373, "generation_rouge_l_f1": 0.3384615384615385, "generation_token_f1": 0.40769230769230774, "prompt_id": "fd96f68afab8f1c0", "reading_canonical_log_prob": -89.48802947998047, "reading_distances": [32, 25, 24, 30], "reading_log_probs": [-91.38858032226562, -91.6458740234375, -90.0040054321289, -90.13968658447266], "reading_logprob_std": 0.8428980608037561, "tokenization_path_logprob_std": 67.91455743804346, "trajectory_balance_rmse": 1518.113274185282} +{"canonical_length": 92, "canonical_nll": 2.350745491359545, "generation_rouge_l_f1": 0.27419354838709675, "generation_token_f1": 0.4677419354838709, "prompt_id": "930a5444ff77055e", "reading_canonical_log_prob": -216.26858520507812, "reading_distances": [46, 50, 45, 43], "reading_log_probs": [-221.1793975830078, -227.1768341064453, -227.52789306640625, -228.8612060546875], "reading_logprob_std": 3.415905150042906, "tokenization_path_logprob_std": 233.58966859690275, "trajectory_balance_rmse": 2327.1299869602262} +{"canonical_length": 88, "canonical_nll": 0.7697162628173828, "generation_rouge_l_f1": 0.24848484848484848, "generation_token_f1": 0.4, "prompt_id": "6881f74e519464d9", "reading_canonical_log_prob": -67.73503112792969, "reading_distances": [14, 19, 18, 17], "reading_log_probs": [-71.54078674316406, -70.14570617675781, -70.21460723876953, -71.36425018310547], "reading_logprob_std": 0.7386611070540272, "tokenization_path_logprob_std": 216.34124239080347, "trajectory_balance_rmse": 1836.7167684960157} +{"canonical_length": 242, "canonical_nll": 1.3203886678396177, "generation_rouge_l_f1": 0.20948616600790515, "generation_token_f1": 0.47035573122529645, "prompt_id": "1afba92bdf32d325", "reading_canonical_log_prob": -319.5340576171875, "reading_distances": [32, 28, 26, 33], "reading_log_probs": [-316.9312438964844, -314.7392578125, -317.5038757324219, -317.0981750488281], "reading_logprob_std": 1.2427379730927672, "tokenization_path_logprob_std": 338.6284822821152, "trajectory_balance_rmse": 5247.638202643151} +{"canonical_length": 114, "canonical_nll": 1.244216383549205, "generation_rouge_l_f1": 0.3344051446945338, "generation_token_f1": 0.5016077170418006, "prompt_id": "19d150b4a3d84266", "reading_canonical_log_prob": -141.84066772460938, "reading_distances": [35, 36, 33, 35], "reading_log_probs": [-144.80029296875, -142.34765625, -144.4137725830078, -141.96697998046875], "reading_logprob_std": 1.431579329966967, "tokenization_path_logprob_std": 64.7115097595893, "trajectory_balance_rmse": 2431.6022452239677} +{"canonical_length": 6, "canonical_nll": 2.2365800539652505, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.5333333333333333, "prompt_id": "dee72943ee9496d3", "reading_canonical_log_prob": -13.419480323791504, "reading_distances": [38, 45, 40, 41], "reading_log_probs": [-11.064242362976074, -9.696393966674805, -11.191794395446777, -10.425379753112793], "reading_logprob_std": 0.6862007914182643, "tokenization_path_logprob_std": 25.07711514540403, "trajectory_balance_rmse": 145.7220525042113} +{"canonical_length": 40, "canonical_nll": 0.6510015487670898, "generation_rouge_l_f1": 0.17204301075268819, "generation_token_f1": 0.22222222222222218, "prompt_id": "f1a8da244ce8319b", "reading_canonical_log_prob": -26.040061950683594, "reading_distances": [21, 23, 19, 25], "reading_log_probs": [-27.132007598876953, -27.311119079589844, -28.525482177734375, -27.232288360595703], "reading_logprob_std": 0.6542903802781257, "tokenization_path_logprob_std": 22.279386542193233, "trajectory_balance_rmse": 665.9036354440638} +{"canonical_length": 286, "canonical_nll": 1.458584338634998, "generation_rouge_l_f1": 0.3065134099616858, "generation_token_f1": 0.5440613026819923, "prompt_id": "201696a6ff512973", "reading_canonical_log_prob": -417.1551208496094, "reading_distances": [23, 21, 20, 17], "reading_log_probs": [-413.7655029296875, -415.00250244140625, -412.6708679199219, -417.8145751953125], "reading_logprob_std": 2.215953088003681, "tokenization_path_logprob_std": 190.36912307567232, "trajectory_balance_rmse": 4596.104231880706} +{"canonical_length": 202, "canonical_nll": 1.9728191111347464, "generation_rouge_l_f1": 0.27539503386004516, "generation_token_f1": 0.46049661399548536, "prompt_id": "8d88867a9f02da1b", "reading_canonical_log_prob": -398.50946044921875, "reading_distances": [25, 29, 29, 32], "reading_log_probs": [-401.4798583984375, -401.0140380859375, -403.6352844238281, -404.44866943359375], "reading_logprob_std": 1.6584608554634268, "tokenization_path_logprob_std": 89.2677686683895, "trajectory_balance_rmse": 3562.1543981673076} +{"canonical_length": 71, "canonical_nll": 0.748474013637489, "generation_rouge_l_f1": 0.27046263345195726, "generation_token_f1": 0.34163701067615665, "prompt_id": "de7efec5b0e0f3a3", "reading_canonical_log_prob": -53.14165496826172, "reading_distances": [21, 25, 25, 22], "reading_log_probs": [-52.81938171386719, -52.723453521728516, -52.74799728393555, -51.45088195800781], "reading_logprob_std": 0.6576244170295377, "tokenization_path_logprob_std": 34.05974924875222, "trajectory_balance_rmse": 897.0024293907938} +{"canonical_length": 8, "canonical_nll": 2.4605088233947754, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.375, "prompt_id": "5d9fef90efd251af", "reading_canonical_log_prob": -19.684070587158203, "reading_distances": [11, 13, 11, 15], "reading_log_probs": [-19.486066818237305, -19.632373809814453, -20.56992530822754, -20.23824691772461], "reading_logprob_std": 0.5097405756824533, "tokenization_path_logprob_std": 24.026607270509984, "trajectory_balance_rmse": 193.13853187700022} +{"canonical_length": 35, "canonical_nll": 1.7628290448869979, "generation_rouge_l_f1": 0.23333333333333334, "generation_token_f1": 0.26666666666666666, "prompt_id": "207d314006c9124a", "reading_canonical_log_prob": -61.69901657104492, "reading_distances": [27, 21, 21, 30], "reading_log_probs": [-72.76319122314453, -73.28870391845703, -73.90650177001953, -74.502685546875], "reading_logprob_std": 0.7538801726809758, "tokenization_path_logprob_std": 70.18148973337081, "trajectory_balance_rmse": 924.8431283722554} +{"canonical_length": 243, "canonical_nll": 0.6068815380457497, "generation_rouge_l_f1": 0.3961165048543689, "generation_token_f1": 0.5553398058252427, "prompt_id": "182ccb8ddd147837", "reading_canonical_log_prob": -147.4722137451172, "reading_distances": [16, 6, 14, 12], "reading_log_probs": [-148.6901397705078, -146.65843200683594, -145.41400146484375, -149.63717651367188], "reading_logprob_std": 1.91517037822573, "tokenization_path_logprob_std": 140.3077209922708, "trajectory_balance_rmse": 1837.4118715894112} +{"canonical_length": 8, "canonical_nll": 0.9617423415184021, "generation_rouge_l_f1": 0.03305785123966942, "generation_token_f1": 0.03305785123966942, "prompt_id": "e6557b0f23801243", "reading_canonical_log_prob": -7.693938732147217, "reading_distances": [8, 6, 8, 9], "reading_log_probs": [-7.1461005210876465, -6.579564571380615, -7.04179048538208, -7.037508010864258], "reading_logprob_std": 0.2528206261470365, "tokenization_path_logprob_std": 29.851760771253243, "trajectory_balance_rmse": 222.52642980428544} +{"canonical_length": 56, "canonical_nll": 0.7175111770629883, "generation_rouge_l_f1": 0.14107883817427386, "generation_token_f1": 0.20746887966804978, "prompt_id": "f461f97cdf966620", "reading_canonical_log_prob": -40.180625915527344, "reading_distances": [24, 22, 20, 22], "reading_log_probs": [-42.46866226196289, -42.70215606689453, -43.619659423828125, -42.94697570800781], "reading_logprob_std": 0.4968525958822966, "tokenization_path_logprob_std": 24.70716189255056, "trajectory_balance_rmse": 964.0564791200034} +{"canonical_length": 319, "canonical_nll": 1.0560534247036637, "generation_rouge_l_f1": 0.26736111111111105, "generation_token_f1": 0.53125, "prompt_id": "e632ea14cc0c360e", "reading_canonical_log_prob": -336.88104248046875, "reading_distances": [12, 15, 14, 15], "reading_log_probs": [-332.4310302734375, -333.66021728515625, -332.6773681640625, -333.21990966796875], "reading_logprob_std": 0.5513716412502991, "tokenization_path_logprob_std": 203.26123909533067, "trajectory_balance_rmse": 5979.004224665075} +{"canonical_length": 123, "canonical_nll": 1.3138338414634145, "generation_rouge_l_f1": 0.2545454545454546, "generation_token_f1": 0.44155844155844154, "prompt_id": "fbb4ad27a6d06f7c", "reading_canonical_log_prob": -161.6015625, "reading_distances": [43, 40, 44, 45], "reading_log_probs": [-160.15841674804688, -159.267333984375, -160.48336791992188, -158.77574157714844], "reading_logprob_std": 0.7878135216012564, "tokenization_path_logprob_std": 218.17415303786254, "trajectory_balance_rmse": 2595.4103443629465} +{"canonical_length": 285, "canonical_nll": 1.0154583847313596, "generation_rouge_l_f1": 0.33001988071570576, "generation_token_f1": 0.5725646123260437, "prompt_id": "db0aefd45e9cb0fd", "reading_canonical_log_prob": -289.4056396484375, "reading_distances": [23, 26, 25, 26], "reading_log_probs": [-288.27386474609375, -290.65716552734375, -287.8052062988281, -289.389404296875], "reading_logprob_std": 1.2713105154088151, "tokenization_path_logprob_std": 149.3508165318951, "trajectory_balance_rmse": 5079.355222609161} +{"canonical_length": 32, "canonical_nll": 0.18985475599765778, "generation_rouge_l_f1": 0.4220183486238532, "generation_token_f1": 0.4220183486238532, "prompt_id": "5f8ab685b6a4b0df", "reading_canonical_log_prob": -6.075352191925049, "reading_distances": [27, 24, 32, 28], "reading_log_probs": [-5.913116455078125, -7.8035478591918945, -7.129713535308838, -8.103788375854492], "reading_logprob_std": 0.9723640295964381, "tokenization_path_logprob_std": 95.79856808290543, "trajectory_balance_rmse": 401.71941184624734} +{"canonical_length": 12, "canonical_nll": 1.1035297711690266, "generation_rouge_l_f1": 0.3157894736842105, "generation_token_f1": 0.3157894736842105, "prompt_id": "6c7141219130f618", "reading_canonical_log_prob": -13.24235725402832, "reading_distances": [22, 20, 26, 20], "reading_log_probs": [-15.788032531738281, -13.904097557067871, -14.12342643737793, -16.430910110473633], "reading_logprob_std": 1.2413293868134174, "tokenization_path_logprob_std": 21.167810628855236, "trajectory_balance_rmse": 327.0812153855792} +{"canonical_length": 8, "canonical_nll": 0.10443764925003052, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1d180306e638dc8c", "reading_canonical_log_prob": -0.8355011940002441, "reading_distances": [24, 26, 26, 24], "reading_log_probs": [-1.3282864093780518, -1.9078400135040283, -2.0259056091308594, -1.656043529510498], "reading_logprob_std": 0.30877696506809904, "tokenization_path_logprob_std": 19.509159889636102, "trajectory_balance_rmse": 130.94302368663847} +{"canonical_length": 162, "canonical_nll": 1.2497601921175734, "generation_rouge_l_f1": 0.2865497076023392, "generation_token_f1": 0.52046783625731, "prompt_id": "042d56a64e3fa50f", "reading_canonical_log_prob": -202.46115112304688, "reading_distances": [29, 27, 31, 21], "reading_log_probs": [-206.68043518066406, -210.55181884765625, -203.949951171875, -208.72215270996094], "reading_logprob_std": 2.8331079428901753, "tokenization_path_logprob_std": 93.14213242953763, "trajectory_balance_rmse": 3567.976325655089} +{"canonical_length": 278, "canonical_nll": 1.4646032319652091, "generation_rouge_l_f1": 0.2713754646840148, "generation_token_f1": 0.48698884758364314, "prompt_id": "519601f060a2cff2", "reading_canonical_log_prob": -407.1596984863281, "reading_distances": [22, 21, 30, 27], "reading_log_probs": [-404.4485168457031, -403.9124755859375, -402.2189025878906, -403.24737548828125], "reading_logprob_std": 0.9604497504543946, "tokenization_path_logprob_std": 123.28059114820819, "trajectory_balance_rmse": 6119.382933215652} +{"canonical_length": 41, "canonical_nll": 1.5845694193025914, "generation_rouge_l_f1": 0.10731707317073172, "generation_token_f1": 0.13658536585365855, "prompt_id": "acf7f177082b10c2", "reading_canonical_log_prob": -64.96734619140625, "reading_distances": [10, 11, 15, 17], "reading_log_probs": [-64.76002502441406, -64.68196105957031, -63.870262145996094, -63.50278091430664], "reading_logprob_std": 0.6166305019852636, "tokenization_path_logprob_std": 151.90928267135124, "trajectory_balance_rmse": 935.0976178630489} +{"canonical_length": 10, "canonical_nll": 1.2963465690612792, "generation_rouge_l_f1": 0.4800000000000001, "generation_token_f1": 0.56, "prompt_id": "78f8a9aafe333ebf", "reading_canonical_log_prob": -12.963465690612793, "reading_distances": [23, 24, 29, 25], "reading_log_probs": [-14.085536003112793, -14.182517051696777, -13.727832794189453, -13.113789558410645], "reading_logprob_std": 0.4837021627153537, "tokenization_path_logprob_std": 49.57989670134406, "trajectory_balance_rmse": 301.48374719178133} +{"canonical_length": 303, "canonical_nll": 1.0189092151402641, "generation_rouge_l_f1": 0.2328042328042328, "generation_token_f1": 0.4620811287477954, "prompt_id": "c2a3bc6c2ee0dd9e", "reading_canonical_log_prob": -308.7294921875, "reading_distances": [30, 28, 31, 32], "reading_log_probs": [-309.93072509765625, -310.818359375, -310.02801513671875, -310.60394287109375], "reading_logprob_std": 0.4332910242014557, "tokenization_path_logprob_std": 55.15574837043174, "trajectory_balance_rmse": 6493.507238438834} +{"canonical_length": 335, "canonical_nll": 1.2427359054337686, "generation_rouge_l_f1": 0.24707846410684475, "generation_token_f1": 0.48414023372287146, "prompt_id": "994646dcabc2bac8", "reading_canonical_log_prob": -416.3165283203125, "reading_distances": [20, 18, 23, 20], "reading_log_probs": [-415.95867919921875, -413.2906494140625, -413.5234375, -412.02520751953125], "reading_logprob_std": 1.6437054449921729, "tokenization_path_logprob_std": 303.44955252188925, "trajectory_balance_rmse": 7519.450931445464} +{"canonical_length": 10, "canonical_nll": 0.8846105575561524, "generation_rouge_l_f1": 0.7777777777777778, "generation_token_f1": 0.7777777777777778, "prompt_id": "bb2e66b420114e36", "reading_canonical_log_prob": -8.846105575561523, "reading_distances": [25, 24, 29, 26], "reading_log_probs": [-8.829644203186035, -8.915371894836426, -8.829793930053711, -8.859821319580078], "reading_logprob_std": 0.04038462627982119, "tokenization_path_logprob_std": 21.59905563836988, "trajectory_balance_rmse": 234.97868234006657} +{"canonical_length": 282, "canonical_nll": 0.9716747094553413, "generation_rouge_l_f1": 0.2791586998087954, "generation_token_f1": 0.5583173996175907, "prompt_id": "57dc4fcf5ed48b59", "reading_canonical_log_prob": -274.01226806640625, "reading_distances": [20, 17, 18, 18], "reading_log_probs": [-276.6977233886719, -277.751708984375, -277.85888671875, -276.6301574707031], "reading_logprob_std": 0.6609898825442302, "tokenization_path_logprob_std": 271.24022402586917, "trajectory_balance_rmse": 6220.096676026079} +{"canonical_length": 15, "canonical_nll": 0.7767427444458008, "generation_rouge_l_f1": 0.02952029520295203, "generation_token_f1": 0.02952029520295203, "prompt_id": "3b7a47e13ee4a0c9", "reading_canonical_log_prob": -11.651141166687012, "reading_distances": [16, 17, 24, 21], "reading_log_probs": [-11.703739166259766, -11.603255271911621, -11.479791641235352, -10.9529390335083], "reading_logprob_std": 0.33412553711481346, "tokenization_path_logprob_std": 32.59615498874646, "trajectory_balance_rmse": 257.32916482485956} +{"canonical_length": 38, "canonical_nll": 1.301848662526984, "generation_rouge_l_f1": 0.22950819672131148, "generation_token_f1": 0.36065573770491804, "prompt_id": "4ae552c8baec5eb5", "reading_canonical_log_prob": -49.47024917602539, "reading_distances": [40, 48, 42, 38], "reading_log_probs": [-57.117462158203125, -53.755287170410156, -56.965457916259766, -58.520957946777344], "reading_logprob_std": 2.015223744356609, "tokenization_path_logprob_std": 86.9509055038306, "trajectory_balance_rmse": 820.2213920677021} +{"canonical_length": 27, "canonical_nll": 1.3106296680591725, "generation_rouge_l_f1": 0.17647058823529413, "generation_token_f1": 0.17647058823529413, "prompt_id": "17ccb345d2ac2714", "reading_canonical_log_prob": -35.387001037597656, "reading_distances": [33, 36, 31, 36], "reading_log_probs": [-30.974319458007812, -30.225759506225586, -32.117469787597656, -34.057830810546875], "reading_logprob_std": 1.6684203134532891, "tokenization_path_logprob_std": 72.88770271309811, "trajectory_balance_rmse": 675.8496955182145} +{"canonical_length": 3, "canonical_nll": 4.129148483276367, "generation_rouge_l_f1": 0.015810276679841896, "generation_token_f1": 0.015810276679841896, "prompt_id": "5e428837a52a1f42", "reading_canonical_log_prob": -12.387445449829102, "reading_distances": [38, 45, 47, 50], "reading_log_probs": [-11.423620223999023, -13.51614761352539, -11.797697067260742, -11.467002868652344], "reading_logprob_std": 0.9908713140813302, "tokenization_path_logprob_std": 32.41231566502491, "trajectory_balance_rmse": 65.23927951251008} +{"canonical_length": 22, "canonical_nll": 1.164869568564675, "generation_rouge_l_f1": 0.3214285714285714, "generation_token_f1": 0.42857142857142855, "prompt_id": "6a0a62922b0dc5d9", "reading_canonical_log_prob": -25.62713050842285, "reading_distances": [25, 21, 22, 14], "reading_log_probs": [-27.66661834716797, -28.120447158813477, -26.47233772277832, -29.266212463378906], "reading_logprob_std": 1.155629231528174, "tokenization_path_logprob_std": 69.15081454468852, "trajectory_balance_rmse": 646.3697549004999} +{"canonical_length": 138, "canonical_nll": 1.502722145854563, "generation_rouge_l_f1": 0.2931937172774869, "generation_token_f1": 0.4607329842931937, "prompt_id": "b8e84fe41fdcb18b", "reading_canonical_log_prob": -207.3756561279297, "reading_distances": [64, 67, 61, 50], "reading_log_probs": [-201.0255889892578, -200.5728759765625, -201.72720336914062, -201.83956909179688], "reading_logprob_std": 0.5992623695231464, "tokenization_path_logprob_std": 159.80160490904785, "trajectory_balance_rmse": 2417.932082546681} +{"canonical_length": 16, "canonical_nll": 0.6988978385925293, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.380952380952381, "prompt_id": "b63558947632bbd1", "reading_canonical_log_prob": -11.182365417480469, "reading_distances": [27, 27, 22, 28], "reading_log_probs": [-10.62000846862793, -12.421710014343262, -10.645591735839844, -11.411308288574219], "reading_logprob_std": 0.848271182771352, "tokenization_path_logprob_std": 52.52354999787344, "trajectory_balance_rmse": 295.9149621798561} +{"canonical_length": 145, "canonical_nll": 1.1971194563240841, "generation_rouge_l_f1": 0.3592233009708738, "generation_token_f1": 0.5048543689320388, "prompt_id": "f6960da157e7277f", "reading_canonical_log_prob": -173.5823211669922, "reading_distances": [18, 13, 15, 21], "reading_log_probs": [-171.0002899169922, -171.59246826171875, -171.85247802734375, -171.64852905273438], "reading_logprob_std": 0.36622588876827067, "tokenization_path_logprob_std": 153.9881412325349, "trajectory_balance_rmse": 3289.1152682491274} +{"canonical_length": 147, "canonical_nll": 1.3440009915098852, "generation_rouge_l_f1": 0.24078624078624078, "generation_token_f1": 0.40786240786240785, "prompt_id": "92be2f9693f01c35", "reading_canonical_log_prob": -197.56814575195312, "reading_distances": [70, 65, 62, 72], "reading_log_probs": [-192.6441192626953, -192.3946533203125, -191.93707275390625, -192.80459594726562], "reading_logprob_std": 0.37836602774878886, "tokenization_path_logprob_std": 117.16964331369934, "trajectory_balance_rmse": 2600.050335512569} +{"canonical_length": 21, "canonical_nll": 0.32431929452078684, "generation_rouge_l_f1": 0.14814814814814814, "generation_token_f1": 0.14814814814814814, "prompt_id": "45f3d6045c73150a", "reading_canonical_log_prob": -6.810705184936523, "reading_distances": [33, 34, 22, 17], "reading_log_probs": [-6.117176532745361, -6.8377180099487305, -6.90336275100708, -6.858846187591553], "reading_logprob_std": 0.37573040239558714, "tokenization_path_logprob_std": 90.74244469012616, "trajectory_balance_rmse": 366.73300775775056} +{"canonical_length": 98, "canonical_nll": 1.2920188903808594, "generation_rouge_l_f1": 0.242603550295858, "generation_token_f1": 0.3905325443786983, "prompt_id": "949e9bf0a5c54d25", "reading_canonical_log_prob": -126.61785125732422, "reading_distances": [17, 15, 17, 14], "reading_log_probs": [-123.42654418945312, -124.15220642089844, -122.5907974243164, -122.45266723632812], "reading_logprob_std": 0.7915704902671661, "tokenization_path_logprob_std": 140.72117472516223, "trajectory_balance_rmse": 2214.0164066622674} +{"canonical_length": 14, "canonical_nll": 1.0293284143720354, "generation_rouge_l_f1": 0.1797752808988764, "generation_token_f1": 0.1797752808988764, "prompt_id": "5602e1c51a5b2686", "reading_canonical_log_prob": -14.410597801208496, "reading_distances": [19, 17, 19, 20], "reading_log_probs": [-13.834470748901367, -14.317708015441895, -14.65577220916748, -14.161664962768555], "reading_logprob_std": 0.34130743745044234, "tokenization_path_logprob_std": 65.22132846670873, "trajectory_balance_rmse": 378.2235018565276} +{"canonical_length": 179, "canonical_nll": 1.2134982487342878, "generation_rouge_l_f1": 0.24390243902439024, "generation_token_f1": 0.5055432372505543, "prompt_id": "94a9111b40eefe1b", "reading_canonical_log_prob": -217.2161865234375, "reading_distances": [37, 36, 37, 38], "reading_log_probs": [-218.00128173828125, -218.34326171875, -216.8790283203125, -215.70431518554688], "reading_logprob_std": 1.1951439099107093, "tokenization_path_logprob_std": 139.4257512521885, "trajectory_balance_rmse": 4023.3953156690127} +{"canonical_length": 137, "canonical_nll": 1.545525878015226, "generation_rouge_l_f1": 0.2394014962593516, "generation_token_f1": 0.4588528678304239, "prompt_id": "726f150c14796002", "reading_canonical_log_prob": -211.73704528808594, "reading_distances": [17, 16, 13, 22], "reading_log_probs": [-215.53302001953125, -215.43106079101562, -209.69442749023438, -216.48060607910156], "reading_logprob_std": 3.09650615650268, "tokenization_path_logprob_std": 173.83388228766847, "trajectory_balance_rmse": 3597.481009185884} +{"canonical_length": 57, "canonical_nll": 0.2633675357751679, "generation_rouge_l_f1": 0.25675675675675674, "generation_token_f1": 0.277027027027027, "prompt_id": "39559806a2fa846b", "reading_canonical_log_prob": -15.01194953918457, "reading_distances": [27, 21, 21, 22], "reading_log_probs": [-16.454984664916992, -15.886778831481934, -17.919631958007812, -17.23458480834961], "reading_logprob_std": 0.8894843229136172, "tokenization_path_logprob_std": 18.836029746371494, "trajectory_balance_rmse": 668.9367495484605} +{"canonical_length": 13, "canonical_nll": 1.1923173757699819, "generation_rouge_l_f1": 0.0711743772241993, "generation_token_f1": 0.0711743772241993, "prompt_id": "3dc6a3c96d3d9987", "reading_canonical_log_prob": -15.500125885009766, "reading_distances": [24, 34, 30, 30], "reading_log_probs": [-17.645572662353516, -18.61302947998047, -16.239376068115234, -17.3433780670166], "reading_logprob_std": 0.9776565847623371, "tokenization_path_logprob_std": 32.87721054291838, "trajectory_balance_rmse": 229.8460710337049} +{"canonical_length": 14, "canonical_nll": 0.6351594924926758, "generation_rouge_l_f1": 0.5333333333333333, "generation_token_f1": 0.5333333333333333, "prompt_id": "8e74067a85468916", "reading_canonical_log_prob": -8.892232894897461, "reading_distances": [17, 17, 17, 23], "reading_log_probs": [-9.689396858215332, -9.96517562866211, -9.850996017456055, -10.200201988220215], "reading_logprob_std": 0.21473031199145887, "tokenization_path_logprob_std": 46.27203168422624, "trajectory_balance_rmse": 363.5534934563379} +{"canonical_length": 27, "canonical_nll": 1.3704136036060475, "generation_rouge_l_f1": 0.08695652173913042, "generation_token_f1": 0.11594202898550723, "prompt_id": "3090724d0b15fbad", "reading_canonical_log_prob": -37.00116729736328, "reading_distances": [25, 17, 19, 26], "reading_log_probs": [-38.47765350341797, -36.939903259277344, -37.49382400512695, -36.64579772949219], "reading_logprob_std": 0.8062653973220756, "tokenization_path_logprob_std": 51.39276097138929, "trajectory_balance_rmse": 535.2247594185532} +{"canonical_length": 273, "canonical_nll": 1.5162223843864469, "generation_rouge_l_f1": 0.26217228464419473, "generation_token_f1": 0.48689138576779023, "prompt_id": "d410f93c21deda57", "reading_canonical_log_prob": -413.9287109375, "reading_distances": [22, 17, 18, 13], "reading_log_probs": [-414.3331298828125, -413.658935546875, -412.04241943359375, -412.7655029296875], "reading_logprob_std": 1.0038900403434685, "tokenization_path_logprob_std": 353.33578800415836, "trajectory_balance_rmse": 6620.979762650726} +{"canonical_length": 354, "canonical_nll": 1.1559518932622705, "generation_rouge_l_f1": 0.22972972972972971, "generation_token_f1": 0.4358108108108108, "prompt_id": "710012fa888db2a9", "reading_canonical_log_prob": -409.20697021484375, "reading_distances": [33, 26, 36, 26], "reading_log_probs": [-411.3159484863281, -409.50372314453125, -408.93707275390625, -411.1409912109375], "reading_logprob_std": 1.1843710159011531, "tokenization_path_logprob_std": 206.7733952691864, "trajectory_balance_rmse": 7134.8014132306525} +{"canonical_length": 191, "canonical_nll": 1.3912676266974804, "generation_rouge_l_f1": 0.23946784922394682, "generation_token_f1": 0.48780487804878053, "prompt_id": "b61ed23f42bef95d", "reading_canonical_log_prob": -265.73211669921875, "reading_distances": [20, 19, 23, 25], "reading_log_probs": [-265.22271728515625, -266.6978454589844, -265.6806335449219, -266.91552734375], "reading_logprob_std": 0.8092364971639101, "tokenization_path_logprob_std": 145.23180164768058, "trajectory_balance_rmse": 4576.42842511207} +{"canonical_length": 255, "canonical_nll": 1.1080396465226716, "generation_rouge_l_f1": 0.2907662082514735, "generation_token_f1": 0.5461689587426325, "prompt_id": "0cd98f075a94d815", "reading_canonical_log_prob": -282.55010986328125, "reading_distances": [17, 29, 27, 20], "reading_log_probs": [-279.56951904296875, -278.73919677734375, -278.5932312011719, -280.3450622558594], "reading_logprob_std": 0.8120479367303729, "tokenization_path_logprob_std": 347.6916136911146, "trajectory_balance_rmse": 5427.847659402235} +{"canonical_length": 315, "canonical_nll": 1.3162622845362104, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.4338624338624339, "prompt_id": "3f6acc98eba7e06b", "reading_canonical_log_prob": -414.62261962890625, "reading_distances": [22, 23, 25, 22], "reading_log_probs": [-413.44268798828125, -414.816162109375, -415.09857177734375, -413.622802734375], "reading_logprob_std": 0.8337955231251324, "tokenization_path_logprob_std": 153.55631721127963, "trajectory_balance_rmse": 7671.205827357909} +{"canonical_length": 44, "canonical_nll": 0.9125357541170988, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.42236024844720493, "prompt_id": "43a708261a941980", "reading_canonical_log_prob": -40.151573181152344, "reading_distances": [15, 12, 13, 12], "reading_log_probs": [-39.07910919189453, -39.43896484375, -40.10272979736328, -39.567298889160156], "reading_logprob_std": 0.4242016239306334, "tokenization_path_logprob_std": 107.31418500848203, "trajectory_balance_rmse": 1079.6137566493153} +{"canonical_length": 186, "canonical_nll": 1.604940434937836, "generation_rouge_l_f1": 0.22760290556900725, "generation_token_f1": 0.4939467312348669, "prompt_id": "6a69846ab5b6aa8a", "reading_canonical_log_prob": -298.5189208984375, "reading_distances": [19, 25, 20, 24], "reading_log_probs": [-294.12042236328125, -294.1175231933594, -296.64697265625, -295.6737060546875], "reading_logprob_std": 1.2437585033210634, "tokenization_path_logprob_std": 235.61662295212398, "trajectory_balance_rmse": 4189.637879933712} +{"canonical_length": 99, "canonical_nll": 1.0295672175860164, "generation_rouge_l_f1": 0.2727272727272727, "generation_token_f1": 0.4380165289256198, "prompt_id": "692adc91823ed2ff", "reading_canonical_log_prob": -101.92715454101562, "reading_distances": [29, 20, 22, 31], "reading_log_probs": [-100.068603515625, -101.37971496582031, -99.91561126708984, -100.07674407958984], "reading_logprob_std": 0.6837264734521264, "tokenization_path_logprob_std": 56.99143957331886, "trajectory_balance_rmse": 2426.373505968139} +{"canonical_length": 11, "canonical_nll": 1.4260387420654297, "generation_rouge_l_f1": 0.10256410256410256, "generation_token_f1": 0.10256410256410256, "prompt_id": "9477d8c348a14e9c", "reading_canonical_log_prob": -15.686426162719727, "reading_distances": [24, 27, 24, 25], "reading_log_probs": [-16.031700134277344, -16.02961540222168, -15.874802589416504, -15.656577110290527], "reading_logprob_std": 0.17703234233971696, "tokenization_path_logprob_std": 46.85282393777463, "trajectory_balance_rmse": 230.92253859717832} +{"canonical_length": 53, "canonical_nll": 0.6600339637612397, "generation_rouge_l_f1": 0.5567010309278351, "generation_token_f1": 0.6597938144329897, "prompt_id": "7e54ccf509187c23", "reading_canonical_log_prob": -34.9818000793457, "reading_distances": [14, 14, 14, 14], "reading_log_probs": [-34.46582794189453, -34.52418518066406, -34.48430252075195, -34.585018157958984], "reading_logprob_std": 0.05274790853081444, "tokenization_path_logprob_std": 98.29596268164755, "trajectory_balance_rmse": 890.115995054901} +{"canonical_length": 65, "canonical_nll": 1.4369557307316707, "generation_rouge_l_f1": 0.16842105263157894, "generation_token_f1": 0.2456140350877193, "prompt_id": "ba0c8c0d4edf2258", "reading_canonical_log_prob": -93.4021224975586, "reading_distances": [25, 25, 22, 20], "reading_log_probs": [-94.07797241210938, -94.32392883300781, -93.3436279296875, -93.66796875], "reading_logprob_std": 0.43438940307621665, "tokenization_path_logprob_std": 143.61173465006135, "trajectory_balance_rmse": 1638.4765086346415} +{"canonical_length": 181, "canonical_nll": 1.538821057061464, "generation_rouge_l_f1": 0.26213592233009714, "generation_token_f1": 0.39805825242718446, "prompt_id": "1f441ed2a6f2f793", "reading_canonical_log_prob": -278.526611328125, "reading_distances": [21, 19, 24, 17], "reading_log_probs": [-276.78912353515625, -278.8203125, -281.08721923828125, -279.6622314453125], "reading_logprob_std": 1.796579811479255, "tokenization_path_logprob_std": 245.04201804207608, "trajectory_balance_rmse": 3581.960892269208} +{"canonical_length": 234, "canonical_nll": 1.3672469701522436, "generation_rouge_l_f1": 0.23529411764705885, "generation_token_f1": 0.5294117647058824, "prompt_id": "a1958dd73d8cbad2", "reading_canonical_log_prob": -319.935791015625, "reading_distances": [16, 12, 15, 15], "reading_log_probs": [-318.20123291015625, -318.6697998046875, -319.1831970214844, -318.08306884765625], "reading_logprob_std": 0.5013218454959504, "tokenization_path_logprob_std": 168.8525107779274, "trajectory_balance_rmse": 5443.493599701951} +{"canonical_length": 248, "canonical_nll": 0.9464220846852949, "generation_rouge_l_f1": 0.27118644067796605, "generation_token_f1": 0.5, "prompt_id": "3e4f398cc2a68a1a", "reading_canonical_log_prob": -234.71267700195312, "reading_distances": [14, 16, 12, 12], "reading_log_probs": [-237.75869750976562, -238.0087432861328, -235.607421875, -238.03700256347656], "reading_logprob_std": 1.1703976844814858, "tokenization_path_logprob_std": 235.07889737925362, "trajectory_balance_rmse": 4797.039540542921} +{"canonical_length": 7, "canonical_nll": 0.6401353563581195, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "96fa3fcdd459d2ed", "reading_canonical_log_prob": -4.480947494506836, "reading_distances": [14, 15, 19, 19], "reading_log_probs": [-5.573293209075928, -5.859443187713623, -5.54518985748291, -5.488106727600098], "reading_logprob_std": 0.16578921133960575, "tokenization_path_logprob_std": 7.819237564274728, "trajectory_balance_rmse": 109.75893552944225} +{"canonical_length": 17, "canonical_nll": 0.5365151798023897, "generation_rouge_l_f1": 0.45454545454545453, "generation_token_f1": 0.45454545454545453, "prompt_id": "27d776f87907a327", "reading_canonical_log_prob": -9.120758056640625, "reading_distances": [23, 20, 24, 22], "reading_log_probs": [-9.485784530639648, -9.798202514648438, -9.270157814025879, -9.357982635498047], "reading_logprob_std": 0.23107919803480545, "tokenization_path_logprob_std": 75.91911288675156, "trajectory_balance_rmse": 361.2407966578621} +{"canonical_length": 37, "canonical_nll": 0.7574756725414379, "generation_rouge_l_f1": 0.1769230769230769, "generation_token_f1": 0.23076923076923075, "prompt_id": "38bff471c2090786", "reading_canonical_log_prob": -28.026599884033203, "reading_distances": [40, 34, 37, 46], "reading_log_probs": [-27.229591369628906, -29.170257568359375, -26.12171745300293, -28.178966522216797], "reading_logprob_std": 1.3039495004366441, "tokenization_path_logprob_std": 42.60348278662538, "trajectory_balance_rmse": 528.2408157326302} +{"canonical_length": 60, "canonical_nll": 1.414673105875651, "generation_rouge_l_f1": 0.31351351351351353, "generation_token_f1": 0.4216216216216216, "prompt_id": "68df2af9448f2f1d", "reading_canonical_log_prob": -84.88038635253906, "reading_distances": [31, 30, 28, 24], "reading_log_probs": [-84.09217834472656, -82.74409484863281, -83.55164337158203, -84.87176513671875], "reading_logprob_std": 0.8962467890407485, "tokenization_path_logprob_std": 118.51495874714693, "trajectory_balance_rmse": 1738.4680464055425} +{"canonical_length": 14, "canonical_nll": 0.6675855772835868, "generation_rouge_l_f1": 0.06716417910447761, "generation_token_f1": 0.08208955223880599, "prompt_id": "6da3ca8bcb458aa8", "reading_canonical_log_prob": -9.346198081970215, "reading_distances": [15, 21, 15, 19], "reading_log_probs": [-10.095718383789062, -10.889922142028809, -9.442215919494629, -11.113344192504883], "reading_logprob_std": 0.7654969949463613, "tokenization_path_logprob_std": 49.643273657317735, "trajectory_balance_rmse": 291.0099534162267} +{"canonical_length": 29, "canonical_nll": 0.9597707288018589, "generation_rouge_l_f1": 0.2145922746781116, "generation_token_f1": 0.22317596566523604, "prompt_id": "eba9c966fdb6009a", "reading_canonical_log_prob": -27.833351135253906, "reading_distances": [35, 30, 32, 34], "reading_log_probs": [-28.257118225097656, -27.864707946777344, -28.185426712036133, -28.971702575683594], "reading_logprob_std": 0.4669300145854989, "tokenization_path_logprob_std": 114.21768639875746, "trajectory_balance_rmse": 609.1051868802981} +{"canonical_length": 237, "canonical_nll": 1.2825541435917722, "generation_rouge_l_f1": 0.2844638949671773, "generation_token_f1": 0.5076586433260394, "prompt_id": "16c0d54d78758bbb", "reading_canonical_log_prob": -303.96533203125, "reading_distances": [20, 34, 21, 33], "reading_log_probs": [-302.56927490234375, -303.0857238769531, -305.70977783203125, -304.805419921875], "reading_logprob_std": 1.4660235224030493, "tokenization_path_logprob_std": 97.1639628836044, "trajectory_balance_rmse": 5174.210630354158} +{"canonical_length": 40, "canonical_nll": 1.5241122245788574, "generation_rouge_l_f1": 0.07582938388625593, "generation_token_f1": 0.08530805687203792, "prompt_id": "51ab3680c9c02305", "reading_canonical_log_prob": -60.9644889831543, "reading_distances": [23, 18, 18, 19], "reading_log_probs": [-62.26685333251953, -61.76115417480469, -61.52101135253906, -62.31227493286133], "reading_logprob_std": 0.38746781990931567, "tokenization_path_logprob_std": 61.88446881924021, "trajectory_balance_rmse": 676.1256805192419} +{"canonical_length": 133, "canonical_nll": 1.2787503120594454, "generation_rouge_l_f1": 0.2388059701492537, "generation_token_f1": 0.41791044776119407, "prompt_id": "a142f5697eaf1704", "reading_canonical_log_prob": -170.07379150390625, "reading_distances": [23, 19, 18, 26], "reading_log_probs": [-177.5304718017578, -175.60447692871094, -173.94775390625, -178.09841918945312], "reading_logprob_std": 1.894298231343214, "tokenization_path_logprob_std": 185.44581134801527, "trajectory_balance_rmse": 3420.3404588973103} +{"canonical_length": 248, "canonical_nll": 1.0769429360666583, "generation_rouge_l_f1": 0.2392156862745098, "generation_token_f1": 0.49019607843137253, "prompt_id": "879233d0df49ade7", "reading_canonical_log_prob": -267.08184814453125, "reading_distances": [21, 27, 28, 22], "reading_log_probs": [-266.608154296875, -267.5699768066406, -266.9626770019531, -266.5535888671875], "reading_logprob_std": 0.46752690413404746, "tokenization_path_logprob_std": 105.81598779884595, "trajectory_balance_rmse": 4720.6796770208775} +{"canonical_length": 189, "canonical_nll": 1.106958621393436, "generation_rouge_l_f1": 0.28978622327790976, "generation_token_f1": 0.5463182897862233, "prompt_id": "f6b26f3fe5bcfb2a", "reading_canonical_log_prob": -209.21517944335938, "reading_distances": [8, 13, 18, 16], "reading_log_probs": [-201.6190185546875, -201.60110473632812, -199.07760620117188, -202.24224853515625], "reading_logprob_std": 1.4036139246543393, "tokenization_path_logprob_std": 102.63187607625983, "trajectory_balance_rmse": 3044.6520834027533} +{"canonical_length": 12, "canonical_nll": 0.9566566944122314, "generation_rouge_l_f1": 0.05925925925925926, "generation_token_f1": 0.06666666666666667, "prompt_id": "61b8f081f2257b08", "reading_canonical_log_prob": -11.479880332946777, "reading_distances": [43, 37, 38, 42], "reading_log_probs": [-14.43034839630127, -18.583209991455078, -14.92666244506836, -18.407880783081055], "reading_logprob_std": 2.214221314567132, "tokenization_path_logprob_std": 28.49751695070364, "trajectory_balance_rmse": 282.11305847799065} +{"canonical_length": 7, "canonical_nll": 0.04574885964393616, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "756f69ca5eb428f4", "reading_canonical_log_prob": -0.3202420175075531, "reading_distances": [39, 46, 37, 36], "reading_log_probs": [-2.1363401412963867, -1.6800830364227295, -1.2979657649993896, -1.9098215103149414], "reading_logprob_std": 0.3577142317715321, "tokenization_path_logprob_std": 17.793816273080346, "trajectory_balance_rmse": 197.71327139799766} +{"canonical_length": 59, "canonical_nll": 1.0021579548463984, "generation_rouge_l_f1": 0.29411764705882354, "generation_token_f1": 0.411764705882353, "prompt_id": "f3fd9530a70f7802", "reading_canonical_log_prob": -59.1273193359375, "reading_distances": [25, 30, 31, 37], "reading_log_probs": [-62.35430908203125, -65.97880554199219, -66.73426818847656, -65.45059204101562], "reading_logprob_std": 1.923658456537674, "tokenization_path_logprob_std": 61.998094739930416, "trajectory_balance_rmse": 1587.520372228747} +{"canonical_length": 84, "canonical_nll": 0.8174148741222563, "generation_rouge_l_f1": 0.30567685589519655, "generation_token_f1": 0.43668122270742354, "prompt_id": "d431ad8fbed7b76c", "reading_canonical_log_prob": -68.66284942626953, "reading_distances": [27, 28, 21, 22], "reading_log_probs": [-67.36770629882812, -67.41525268554688, -67.38320922851562, -68.31301879882812], "reading_logprob_std": 0.4625719166351781, "tokenization_path_logprob_std": 74.52825692392794, "trajectory_balance_rmse": 1366.480249223806} +{"canonical_length": 376, "canonical_nll": 1.4242540724734043, "generation_rouge_l_f1": 0.161993769470405, "generation_token_f1": 0.29283489096573206, "prompt_id": "455b2b67c74aee94", "reading_canonical_log_prob": -535.51953125, "reading_distances": [27, 19, 18, 24], "reading_log_probs": [-536.446044921875, -534.806396484375, -532.2506103515625, -537.45263671875], "reading_logprob_std": 2.2711790942430685, "tokenization_path_logprob_std": 142.4186499889829, "trajectory_balance_rmse": 7449.064591491557} +{"canonical_length": 307, "canonical_nll": 1.6554521691139048, "generation_rouge_l_f1": 0.1979522184300341, "generation_token_f1": 0.38566552901023887, "prompt_id": "b3e34a08eb2e7c34", "reading_canonical_log_prob": -508.22381591796875, "reading_distances": [25, 26, 28, 27], "reading_log_probs": [-515.6603393554688, -518.5628051757812, -517.5003662109375, -511.79180908203125], "reading_logprob_std": 2.9768316811252715, "tokenization_path_logprob_std": 165.19006842997808, "trajectory_balance_rmse": 6690.038033825986} +{"canonical_length": 165, "canonical_nll": 1.3302497632575758, "generation_rouge_l_f1": 0.2600472813238771, "generation_token_f1": 0.4822695035460993, "prompt_id": "7461cdb891fcaab7", "reading_canonical_log_prob": -219.4912109375, "reading_distances": [18, 18, 17, 23], "reading_log_probs": [-216.97445678710938, -216.31057739257812, -217.7942657470703, -218.6161651611328], "reading_logprob_std": 1.0000255707253645, "tokenization_path_logprob_std": 156.83667656483271, "trajectory_balance_rmse": 3704.424003221428} +{"canonical_length": 119, "canonical_nll": 1.2376955176601891, "generation_rouge_l_f1": 0.20460358056265984, "generation_token_f1": 0.30179028132992325, "prompt_id": "eb07969998e95372", "reading_canonical_log_prob": -147.2857666015625, "reading_distances": [18, 19, 18, 17], "reading_log_probs": [-142.61968994140625, -154.3134765625, -152.90151977539062, -150.32574462890625], "reading_logprob_std": 5.21515575566011, "tokenization_path_logprob_std": 54.00623533812821, "trajectory_balance_rmse": 2136.142652449279} +{"canonical_length": 134, "canonical_nll": 1.5178904746895405, "generation_rouge_l_f1": 0.2888888888888889, "generation_token_f1": 0.461111111111111, "prompt_id": "b32326c583d0044a", "reading_canonical_log_prob": -203.39732360839844, "reading_distances": [23, 33, 27, 21], "reading_log_probs": [-197.87353515625, -198.70059204101562, -198.31735229492188, -199.30189514160156], "reading_logprob_std": 0.605459311162404, "tokenization_path_logprob_std": 103.89236900604436, "trajectory_balance_rmse": 2730.7532790475466} +{"canonical_length": 324, "canonical_nll": 1.070399343231578, "generation_rouge_l_f1": 0.3236245954692557, "generation_token_f1": 0.5728155339805825, "prompt_id": "e86ca4a6dca6aa44", "reading_canonical_log_prob": -346.80938720703125, "reading_distances": [38, 31, 34, 32], "reading_log_probs": [-340.786865234375, -341.44482421875, -339.2135314941406, -339.15045166015625], "reading_logprob_std": 1.1486561058887217, "tokenization_path_logprob_std": 392.07263571753464, "trajectory_balance_rmse": 6525.764192898991} +{"canonical_length": 17, "canonical_nll": 0.045534428428201115, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "39176769401db5cc", "reading_canonical_log_prob": -0.774085283279419, "reading_distances": [14, 22, 19, 18], "reading_log_probs": [-0.8980345129966736, -0.9719318151473999, -0.8958382606506348, -1.0496245622634888], "reading_logprob_std": 0.07298504206365111, "tokenization_path_logprob_std": 76.9953043462331, "trajectory_balance_rmse": 335.0953594659645} +{"canonical_length": 46, "canonical_nll": 1.3564103997271995, "generation_rouge_l_f1": 0.1238390092879257, "generation_token_f1": 0.19814241486068113, "prompt_id": "6614284ebae06e44", "reading_canonical_log_prob": -62.39487838745117, "reading_distances": [15, 22, 21, 18], "reading_log_probs": [-63.79412078857422, -65.5857925415039, -64.85209655761719, -63.74403762817383], "reading_logprob_std": 0.88929158462181, "tokenization_path_logprob_std": 187.72946583204063, "trajectory_balance_rmse": 1021.425678738948} +{"canonical_length": 169, "canonical_nll": 1.9819856000369822, "generation_rouge_l_f1": 0.21604938271604937, "generation_token_f1": 0.43209876543209874, "prompt_id": "130451267d979653", "reading_canonical_log_prob": -334.95556640625, "reading_distances": [29, 24, 16, 21], "reading_log_probs": [-332.9437561035156, -334.0497741699219, -335.4052734375, -335.7472839355469], "reading_logprob_std": 1.290284229128985, "tokenization_path_logprob_std": 75.19072482705157, "trajectory_balance_rmse": 3677.926647726831} +{"canonical_length": 89, "canonical_nll": 1.727983796194698, "generation_rouge_l_f1": 0.24489795918367346, "generation_token_f1": 0.40816326530612246, "prompt_id": "13cc5e7d950f14d1", "reading_canonical_log_prob": -153.79055786132812, "reading_distances": [24, 22, 20, 26], "reading_log_probs": [-149.6562957763672, -150.21998596191406, -150.54446411132812, -150.18492126464844], "reading_logprob_std": 0.3676299299125166, "tokenization_path_logprob_std": 138.4638382565623, "trajectory_balance_rmse": 2020.6244348223593} +{"canonical_length": 50, "canonical_nll": 1.2181208038330078, "generation_rouge_l_f1": 0.24999999999999994, "generation_token_f1": 0.40476190476190477, "prompt_id": "aa5825f0f273154e", "reading_canonical_log_prob": -60.90604019165039, "reading_distances": [35, 32, 29, 29], "reading_log_probs": [-62.7775764465332, -63.46116638183594, -63.93566131591797, -63.98167037963867], "reading_logprob_std": 0.559500123043561, "tokenization_path_logprob_std": 23.282957511088284, "trajectory_balance_rmse": 1365.268169688715} +{"canonical_length": 217, "canonical_nll": 1.392833516345046, "generation_rouge_l_f1": 0.275711159737418, "generation_token_f1": 0.5207877461706784, "prompt_id": "046e2cee68c73a19", "reading_canonical_log_prob": -302.244873046875, "reading_distances": [23, 31, 24, 29], "reading_log_probs": [-300.52716064453125, -301.5296630859375, -302.0565490722656, -303.57037353515625], "reading_logprob_std": 1.2694801196475773, "tokenization_path_logprob_std": 200.42182882090387, "trajectory_balance_rmse": 4298.837895331111} +{"canonical_length": 114, "canonical_nll": 1.813457823636239, "generation_rouge_l_f1": 0.23333333333333334, "generation_token_f1": 0.33333333333333337, "prompt_id": "f13ae64aa84fa6f5", "reading_canonical_log_prob": -206.73419189453125, "reading_distances": [25, 29, 25, 26], "reading_log_probs": [-207.97689819335938, -207.89920043945312, -205.723388671875, -207.72193908691406], "reading_logprob_std": 1.0766122966211513, "tokenization_path_logprob_std": 105.1309840410262, "trajectory_balance_rmse": 2371.8069049681358} +{"canonical_length": 180, "canonical_nll": 1.2153906928168403, "generation_rouge_l_f1": 0.23348017621145375, "generation_token_f1": 0.41850220264317184, "prompt_id": "55942a6d7a51de21", "reading_canonical_log_prob": -218.77032470703125, "reading_distances": [17, 19, 19, 15], "reading_log_probs": [-218.95501708984375, -219.0145721435547, -218.8612518310547, -219.89974975585938], "reading_logprob_std": 0.48221563311011345, "tokenization_path_logprob_std": 228.12124357344499, "trajectory_balance_rmse": 4408.667611107218} +{"canonical_length": 158, "canonical_nll": 1.194859275334998, "generation_rouge_l_f1": 0.26697892271662765, "generation_token_f1": 0.4262295081967213, "prompt_id": "ab7cc228a99beb55", "reading_canonical_log_prob": -188.7877655029297, "reading_distances": [34, 32, 33, 32], "reading_log_probs": [-182.3800048828125, -184.6732940673828, -183.02951049804688, -183.4993896484375], "reading_logprob_std": 0.9675965716407601, "tokenization_path_logprob_std": 93.03043420113801, "trajectory_balance_rmse": 3180.981965450851} +{"canonical_length": 11, "canonical_nll": 1.196750207380815, "generation_rouge_l_f1": 0.12121212121212122, "generation_token_f1": 0.12121212121212122, "prompt_id": "c2611a4dd26cfbb1", "reading_canonical_log_prob": -13.164252281188965, "reading_distances": [14, 11, 14, 16], "reading_log_probs": [-13.304189682006836, -13.434394836425781, -13.290770530700684, -13.710390090942383], "reading_logprob_std": 0.1947250465670722, "tokenization_path_logprob_std": 44.04884846501426, "trajectory_balance_rmse": 219.9585986556523} +{"canonical_length": 138, "canonical_nll": 1.3672635728034421, "generation_rouge_l_f1": 0.3783783783783784, "generation_token_f1": 0.555984555984556, "prompt_id": "6f0ecfb4c606f967", "reading_canonical_log_prob": -188.682373046875, "reading_distances": [28, 22, 23, 24], "reading_log_probs": [-188.41751098632812, -188.53604125976562, -188.4929962158203, -189.36334228515625], "reading_logprob_std": 0.44329508929010863, "tokenization_path_logprob_std": 132.15970211572846, "trajectory_balance_rmse": 3153.8948549499473} +{"canonical_length": 202, "canonical_nll": 0.6900657427192914, "generation_rouge_l_f1": 0.5287356321839081, "generation_token_f1": 0.6168582375478928, "prompt_id": "5d860ca9ad882bc0", "reading_canonical_log_prob": -139.39328002929688, "reading_distances": [13, 10, 13, 6], "reading_log_probs": [-141.5067901611328, -138.78988647460938, -139.43838500976562, -139.63929748535156], "reading_logprob_std": 1.1665408138402014, "tokenization_path_logprob_std": 226.2466141594489, "trajectory_balance_rmse": 3895.103361738161} +{"canonical_length": 253, "canonical_nll": 1.5748237941576086, "generation_rouge_l_f1": 0.2581967213114754, "generation_token_f1": 0.4877049180327869, "prompt_id": "bef65ba5cd28353b", "reading_canonical_log_prob": -398.430419921875, "reading_distances": [38, 36, 31, 36], "reading_log_probs": [-393.9472351074219, -390.07196044921875, -392.235107421875, -392.89373779296875], "reading_logprob_std": 1.636419303653233, "tokenization_path_logprob_std": 229.18156532813035, "trajectory_balance_rmse": 5224.605313919788} +{"canonical_length": 49, "canonical_nll": 0.7294379253776706, "generation_rouge_l_f1": 0.2588235294117647, "generation_token_f1": 0.39999999999999997, "prompt_id": "ba40c7a321bb21e7", "reading_canonical_log_prob": -35.74245834350586, "reading_distances": [12, 16, 15, 19], "reading_log_probs": [-35.56303024291992, -35.80370330810547, -35.875205993652344, -36.34469223022461], "reading_logprob_std": 0.3271813539250958, "tokenization_path_logprob_std": 86.38180590539875, "trajectory_balance_rmse": 1059.5776750063642} +{"canonical_length": 15, "canonical_nll": 0.6550783157348633, "generation_rouge_l_f1": 0.30136986301369867, "generation_token_f1": 0.3835616438356165, "prompt_id": "a165c584ac95ad42", "reading_canonical_log_prob": -9.82617473602295, "reading_distances": [26, 34, 24, 22], "reading_log_probs": [-11.900867462158203, -12.818907737731934, -12.295246124267578, -10.676610946655273], "reading_logprob_std": 0.9119929347831005, "tokenization_path_logprob_std": 36.74567875745301, "trajectory_balance_rmse": 366.72047811007246} +{"canonical_length": 242, "canonical_nll": 1.5864050999160642, "generation_rouge_l_f1": 0.18661257606490875, "generation_token_f1": 0.40973630831643, "prompt_id": "9c4dbe029e6ba3bb", "reading_canonical_log_prob": -383.9100341796875, "reading_distances": [16, 16, 19, 19], "reading_log_probs": [-386.49969482421875, -387.1658935546875, -385.4879455566406, -385.63067626953125], "reading_logprob_std": 0.7860992425953877, "tokenization_path_logprob_std": 206.2415979787022, "trajectory_balance_rmse": 5203.303245873239} +{"canonical_length": 62, "canonical_nll": 0.6842442174111644, "generation_rouge_l_f1": 0.26785714285714285, "generation_token_f1": 0.4821428571428571, "prompt_id": "dcfe7ed6b29a3c6d", "reading_canonical_log_prob": -42.42314147949219, "reading_distances": [24, 17, 20, 19], "reading_log_probs": [-45.025360107421875, -45.304283142089844, -44.20014953613281, -44.83326721191406], "reading_logprob_std": 0.4688172391250374, "tokenization_path_logprob_std": 73.29502855703424, "trajectory_balance_rmse": 1165.3758923932069} +{"canonical_length": 148, "canonical_nll": 1.6217478159311656, "generation_rouge_l_f1": 0.2386634844868735, "generation_token_f1": 0.477326968973747, "prompt_id": "c0cc49ea5faf89df", "reading_canonical_log_prob": -240.0186767578125, "reading_distances": [28, 23, 21, 26], "reading_log_probs": [-242.14401245117188, -244.70338439941406, -242.9661865234375, -242.0367431640625], "reading_logprob_std": 1.232563675541514, "tokenization_path_logprob_std": 111.62073770025185, "trajectory_balance_rmse": 3668.056764460223} +{"canonical_length": 14, "canonical_nll": 1.878906522478376, "generation_rouge_l_f1": 0.24390243902439024, "generation_token_f1": 0.24390243902439024, "prompt_id": "adf9198c8ba81554", "reading_canonical_log_prob": -26.304691314697266, "reading_distances": [18, 15, 10, 13], "reading_log_probs": [-25.84716033935547, -25.678842544555664, -25.95305633544922, -25.638076782226562], "reading_logprob_std": 0.1470056533321832, "tokenization_path_logprob_std": 74.71170903745275, "trajectory_balance_rmse": 387.0609475088027} +{"canonical_length": 280, "canonical_nll": 1.082346670968192, "generation_rouge_l_f1": 0.19012797074954296, "generation_token_f1": 0.4095063985374771, "prompt_id": "7eceb236d5391917", "reading_canonical_log_prob": -303.05706787109375, "reading_distances": [21, 27, 22, 29], "reading_log_probs": [-299.76422119140625, -300.88262939453125, -298.42755126953125, -301.77197265625], "reading_logprob_std": 1.4454550547147387, "tokenization_path_logprob_std": 154.8406803741702, "trajectory_balance_rmse": 5781.020131598813} +{"canonical_length": 7, "canonical_nll": 1.6545636313302177, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "8279f5e0aa9b5caf", "reading_canonical_log_prob": -11.581945419311523, "reading_distances": [22, 28, 24, 21], "reading_log_probs": [-11.924067497253418, -12.057735443115234, -12.077420234680176, -11.991823196411133], "reading_logprob_std": 0.06954242400657189, "tokenization_path_logprob_std": 10.79367747439796, "trajectory_balance_rmse": 172.42507161789683} +{"canonical_length": 323, "canonical_nll": 0.7622261637873694, "generation_rouge_l_f1": 0.17122040072859743, "generation_token_f1": 0.3387978142076502, "prompt_id": "f4edb7d5c1554a64", "reading_canonical_log_prob": -246.1990509033203, "reading_distances": [21, 21, 22, 13], "reading_log_probs": [-248.8551788330078, -248.62615966796875, -248.23353576660156, -248.71800231933594], "reading_logprob_std": 0.2669275365341999, "tokenization_path_logprob_std": 79.14662260857077, "trajectory_balance_rmse": 6769.432663942546} +{"canonical_length": 18, "canonical_nll": 0.611156251695421, "generation_rouge_l_f1": 0.08547008547008547, "generation_token_f1": 0.10256410256410256, "prompt_id": "39c97bd4effbfbe9", "reading_canonical_log_prob": -11.000812530517578, "reading_distances": [25, 27, 28, 26], "reading_log_probs": [-10.556442260742188, -9.961331367492676, -10.958608627319336, -9.97331428527832], "reading_logprob_std": 0.4848920379134886, "tokenization_path_logprob_std": 66.07569616613463, "trajectory_balance_rmse": 362.7052237200712} +{"canonical_length": 32, "canonical_nll": 0.5286539793014526, "generation_rouge_l_f1": 0.4473684210526316, "generation_token_f1": 0.5789473684210527, "prompt_id": "e245da9a65e5ed0d", "reading_canonical_log_prob": -16.916927337646484, "reading_distances": [22, 20, 24, 17], "reading_log_probs": [-18.790401458740234, -27.556276321411133, -22.905597686767578, -24.103452682495117], "reading_logprob_std": 3.616968975588031, "tokenization_path_logprob_std": 55.86518737313324, "trajectory_balance_rmse": 372.84555506701696} +{"canonical_length": 244, "canonical_nll": 1.2723723864946208, "generation_rouge_l_f1": 0.3347457627118644, "generation_token_f1": 0.4915254237288135, "prompt_id": "1dc2c6d9af2ea2a5", "reading_canonical_log_prob": -310.4588623046875, "reading_distances": [15, 15, 18, 18], "reading_log_probs": [-315.0556335449219, -305.33343505859375, -306.7266845703125, -308.5186767578125], "reading_logprob_std": 4.300411640666413, "tokenization_path_logprob_std": 158.13421872221735, "trajectory_balance_rmse": 4767.000681272276} +{"canonical_length": 143, "canonical_nll": 1.2876266799606644, "generation_rouge_l_f1": 0.24747474747474746, "generation_token_f1": 0.4393939393939394, "prompt_id": "8214cf7581d9c6a8", "reading_canonical_log_prob": -184.130615234375, "reading_distances": [13, 10, 10, 7], "reading_log_probs": [-183.0820770263672, -183.2392578125, -182.31207275390625, -182.38026428222656], "reading_logprob_std": 0.47542486345034934, "tokenization_path_logprob_std": 73.00615518820722, "trajectory_balance_rmse": 3422.239379436238} +{"canonical_length": 79, "canonical_nll": 0.6336692134036294, "generation_rouge_l_f1": 0.29906542056074764, "generation_token_f1": 0.3987538940809969, "prompt_id": "1669f97fd38169a6", "reading_canonical_log_prob": -50.05986785888672, "reading_distances": [15, 27, 30, 22], "reading_log_probs": [-52.38369369506836, -51.40287399291992, -53.1629638671875, -50.68923568725586], "reading_logprob_std": 1.08654591613805, "tokenization_path_logprob_std": 89.2016294752636, "trajectory_balance_rmse": 1279.5182487529253} +{"canonical_length": 354, "canonical_nll": 0.8890809312378619, "generation_rouge_l_f1": 0.2099447513812155, "generation_token_f1": 0.3167587476979742, "prompt_id": "85c649ae8170e371", "reading_canonical_log_prob": -314.7346496582031, "reading_distances": [22, 24, 24, 26], "reading_log_probs": [-313.49481201171875, -311.76885986328125, -312.83160400390625, -313.5796813964844], "reading_logprob_std": 0.8363650524553914, "tokenization_path_logprob_std": 326.7860536100962, "trajectory_balance_rmse": 5800.474167817494} +{"canonical_length": 364, "canonical_nll": 1.0273847475156679, "generation_rouge_l_f1": 0.30016863406408095, "generation_token_f1": 0.475548060708263, "prompt_id": "ff47696a7202ba2a", "reading_canonical_log_prob": -373.9680480957031, "reading_distances": [34, 28, 33, 29], "reading_log_probs": [-373.4309997558594, -374.46258544921875, -375.1084289550781, -374.6240234375], "reading_logprob_std": 0.7058722753723408, "tokenization_path_logprob_std": 170.50504255732938, "trajectory_balance_rmse": 7239.4443500203915} +{"canonical_length": 41, "canonical_nll": 1.0584355796255716, "generation_rouge_l_f1": 0.34285714285714286, "generation_token_f1": 0.4, "prompt_id": "3a5ef38cf12c45a8", "reading_canonical_log_prob": -43.39585876464844, "reading_distances": [24, 17, 17, 21], "reading_log_probs": [-45.395790100097656, -44.76527786254883, -44.06354522705078, -45.23575973510742], "reading_logprob_std": 0.5976272903606094, "tokenization_path_logprob_std": 47.21775522709222, "trajectory_balance_rmse": 834.6638668365637} +{"canonical_length": 230, "canonical_nll": 1.130615632430367, "generation_rouge_l_f1": 0.24000000000000002, "generation_token_f1": 0.5136842105263159, "prompt_id": "13b43b11a6bf37ad", "reading_canonical_log_prob": -260.0415954589844, "reading_distances": [20, 19, 22, 18], "reading_log_probs": [-253.06292724609375, -254.44866943359375, -253.7211151123047, -253.29830932617188], "reading_logprob_std": 0.6082985698533752, "tokenization_path_logprob_std": 294.5832556313143, "trajectory_balance_rmse": 4619.164303917498} +{"canonical_length": 111, "canonical_nll": 1.1990676742416244, "generation_rouge_l_f1": 0.34703196347031967, "generation_token_f1": 0.4931506849315069, "prompt_id": "d6b3da36a5892656", "reading_canonical_log_prob": -133.0965118408203, "reading_distances": [32, 35, 30, 40], "reading_log_probs": [-131.47830200195312, -130.20474243164062, -130.76702880859375, -131.21803283691406], "reading_logprob_std": 0.5584152891355263, "tokenization_path_logprob_std": 180.89368398943913, "trajectory_balance_rmse": 2448.3120449596886} +{"canonical_length": 184, "canonical_nll": 1.2377038209334663, "generation_rouge_l_f1": 0.26185101580135434, "generation_token_f1": 0.4153498871331828, "prompt_id": "f1036fae25a980bc", "reading_canonical_log_prob": -227.7375030517578, "reading_distances": [17, 8, 10, 14], "reading_log_probs": [-224.983642578125, -227.2390899658203, -225.40882873535156, -225.8299560546875], "reading_logprob_std": 0.9788145661088603, "tokenization_path_logprob_std": 280.19035998082524, "trajectory_balance_rmse": 4765.327955684193} +{"canonical_length": 199, "canonical_nll": 1.0819798761875785, "generation_rouge_l_f1": 0.3458980044345898, "generation_token_f1": 0.5365853658536586, "prompt_id": "272e476421ef865b", "reading_canonical_log_prob": -215.31399536132812, "reading_distances": [25, 28, 25, 30], "reading_log_probs": [-212.51339721679688, -212.1631317138672, -212.19619750976562, -210.86752319335938], "reading_logprob_std": 0.7289989198699187, "tokenization_path_logprob_std": 162.41788727453914, "trajectory_balance_rmse": 3562.238346459199} +{"canonical_length": 202, "canonical_nll": 0.9411618072207611, "generation_rouge_l_f1": 0.2453222453222453, "generation_token_f1": 0.49480249480249483, "prompt_id": "548a4dcfac376298", "reading_canonical_log_prob": -190.11468505859375, "reading_distances": [23, 23, 22, 25], "reading_log_probs": [-191.0971221923828, -190.37908935546875, -189.91070556640625, -190.67156982421875], "reading_logprob_std": 0.49900676353163786, "tokenization_path_logprob_std": 111.98163572919074, "trajectory_balance_rmse": 4277.173977238176} +{"canonical_length": 415, "canonical_nll": 0.9419332113610692, "generation_rouge_l_f1": 0.19639934533551553, "generation_token_f1": 0.4124386252045827, "prompt_id": "8ea8da1e0ef16ea4", "reading_canonical_log_prob": -390.90228271484375, "reading_distances": [32, 28, 28, 26], "reading_log_probs": [-398.452880859375, -395.5301513671875, -394.6744384765625, -392.9893493652344], "reading_logprob_std": 2.2857638327561007, "tokenization_path_logprob_std": 105.83712845181014, "trajectory_balance_rmse": 4394.751898435028} +{"canonical_length": 70, "canonical_nll": 0.9506768362862723, "generation_rouge_l_f1": 0.14409221902017288, "generation_token_f1": 0.20172910662824206, "prompt_id": "68d1893ed02225a9", "reading_canonical_log_prob": -66.54737854003906, "reading_distances": [12, 15, 21, 12], "reading_log_probs": [-65.35298919677734, -64.71218872070312, -65.00274658203125, -65.76177978515625], "reading_logprob_std": 0.4530086056277807, "tokenization_path_logprob_std": 63.83489447566249, "trajectory_balance_rmse": 1539.1760806849866} +{"canonical_length": 207, "canonical_nll": 1.2070938331493433, "generation_rouge_l_f1": 0.3014861995753716, "generation_token_f1": 0.5138004246284501, "prompt_id": "06716de5ff9ed39f", "reading_canonical_log_prob": -249.86842346191406, "reading_distances": [27, 29, 30, 30], "reading_log_probs": [-255.50042724609375, -253.10513305664062, -253.42225646972656, -257.10748291015625], "reading_logprob_std": 1.8783663439159872, "tokenization_path_logprob_std": 272.82697517774346, "trajectory_balance_rmse": 4977.452704084967} +{"canonical_length": 24, "canonical_nll": 0.5930384000142416, "generation_rouge_l_f1": 0.15833333333333335, "generation_token_f1": 0.19166666666666665, "prompt_id": "8d76cf27c40ca6eb", "reading_canonical_log_prob": -14.232921600341797, "reading_distances": [33, 27, 25, 27], "reading_log_probs": [-13.572534561157227, -14.268010139465332, -13.135797500610352, -14.173359870910645], "reading_logprob_std": 0.5325110638993025, "tokenization_path_logprob_std": 93.42975330247296, "trajectory_balance_rmse": 443.09582242942247} +{"canonical_length": 37, "canonical_nll": 1.3197612246951542, "generation_rouge_l_f1": 0.11038961038961038, "generation_token_f1": 0.14935064935064937, "prompt_id": "f7669b85c591c347", "reading_canonical_log_prob": -48.8311653137207, "reading_distances": [13, 25, 26, 18], "reading_log_probs": [-50.07414627075195, -48.8554573059082, -48.97455978393555, -49.83656311035156], "reading_logprob_std": 0.6103648313142993, "tokenization_path_logprob_std": 107.51453682047534, "trajectory_balance_rmse": 921.3599494634138} +{"canonical_length": 292, "canonical_nll": 1.1726427470168022, "generation_rouge_l_f1": 0.2939068100358423, "generation_token_f1": 0.5698924731182796, "prompt_id": "9875c8d91c22a365", "reading_canonical_log_prob": -342.41168212890625, "reading_distances": [29, 21, 25, 22], "reading_log_probs": [-340.2279968261719, -343.30792236328125, -340.1816711425781, -340.40911865234375], "reading_logprob_std": 1.5206669183345762, "tokenization_path_logprob_std": 192.73982410286672, "trajectory_balance_rmse": 6587.23399525228} +{"canonical_length": 293, "canonical_nll": 1.3611663714203819, "generation_rouge_l_f1": 0.23445825932504438, "generation_token_f1": 0.4724689165186501, "prompt_id": "967c6a5c34b145df", "reading_canonical_log_prob": -398.8217468261719, "reading_distances": [37, 32, 28, 37], "reading_log_probs": [-396.0498352050781, -397.42767333984375, -394.8431091308594, -397.1957702636719], "reading_logprob_std": 1.1880150986440814, "tokenization_path_logprob_std": 180.05810242101143, "trajectory_balance_rmse": 6622.31882537261} +{"canonical_length": 9, "canonical_nll": 0.282691134346856, "generation_rouge_l_f1": 0.875, "generation_token_f1": 1.0, "prompt_id": "abcadc4b6374d8f6", "reading_canonical_log_prob": -2.544220209121704, "reading_distances": [43, 51, 41, 50], "reading_log_probs": [-2.515141010284424, -2.5483107566833496, -2.9172022342681885, -2.623260021209717], "reading_logprob_std": 0.18315374223504977, "tokenization_path_logprob_std": 48.37041903738015, "trajectory_balance_rmse": 197.94775043053033} +{"canonical_length": 261, "canonical_nll": 1.2787204318576388, "generation_rouge_l_f1": 0.18475750577367206, "generation_token_f1": 0.2586605080831409, "prompt_id": "ba38639bb894f700", "reading_canonical_log_prob": -333.74603271484375, "reading_distances": [21, 30, 29, 25], "reading_log_probs": [-334.3077697753906, -332.6167907714844, -331.74468994140625, -333.013916015625], "reading_logprob_std": 1.0658349201260888, "tokenization_path_logprob_std": 246.17236263429024, "trajectory_balance_rmse": 4688.990070936298} +{"canonical_length": 258, "canonical_nll": 1.037570687227471, "generation_rouge_l_f1": 0.18534482758620693, "generation_token_f1": 0.33189655172413796, "prompt_id": "49e9fe2e1aacf453", "reading_canonical_log_prob": -267.6932373046875, "reading_distances": [15, 21, 17, 17], "reading_log_probs": [-267.62451171875, -269.27984619140625, -265.9716796875, -267.1037902832031], "reading_logprob_std": 1.375499790604142, "tokenization_path_logprob_std": 135.68563138874936, "trajectory_balance_rmse": 4218.365256008759} +{"canonical_length": 255, "canonical_nll": 1.0065716911764706, "generation_rouge_l_f1": 0.19844357976653695, "generation_token_f1": 0.37743190661478604, "prompt_id": "6193ec67d5074a6a", "reading_canonical_log_prob": -256.67578125, "reading_distances": [21, 21, 17, 15], "reading_log_probs": [-256.3648376464844, -257.90985107421875, -255.97268676757812, -257.6296081542969], "reading_logprob_std": 0.9450318939941275, "tokenization_path_logprob_std": 100.67479440909493, "trajectory_balance_rmse": 4581.537999597852} +{"canonical_length": 289, "canonical_nll": 0.8778642172631921, "generation_rouge_l_f1": 0.3244206773618538, "generation_token_f1": 0.5062388591800357, "prompt_id": "821b32f3ecbd67a9", "reading_canonical_log_prob": -253.7027587890625, "reading_distances": [17, 18, 17, 17], "reading_log_probs": [-252.08392333984375, -251.81947326660156, -252.57049560546875, -252.05181884765625], "reading_logprob_std": 0.31553680595401346, "tokenization_path_logprob_std": 209.1365165518757, "trajectory_balance_rmse": 5731.190459852555} +{"canonical_length": 22, "canonical_nll": 0.44111459905450995, "generation_rouge_l_f1": 0.4923076923076923, "generation_token_f1": 0.5538461538461539, "prompt_id": "ebadf5b1dd406fa9", "reading_canonical_log_prob": -9.704521179199219, "reading_distances": [30, 36, 37, 43], "reading_log_probs": [-10.220832824707031, -9.367488861083984, -9.62793254852295, -10.314712524414062], "reading_logprob_std": 0.4587365233036272, "tokenization_path_logprob_std": 83.35580919888703, "trajectory_balance_rmse": 533.5639429049913} +{"canonical_length": 8, "canonical_nll": 1.0006468296051025, "generation_rouge_l_f1": 0.059259259259259255, "generation_token_f1": 0.059259259259259255, "prompt_id": "6edf6d71173f0e03", "reading_canonical_log_prob": -8.00517463684082, "reading_distances": [30, 32, 33, 27], "reading_log_probs": [-17.7645206451416, -18.44696044921875, -18.22195816040039, -19.15361976623535], "reading_logprob_std": 0.5789758416711118, "tokenization_path_logprob_std": 26.364666768743632, "trajectory_balance_rmse": 167.8097065228511} +{"canonical_length": 16, "canonical_nll": 1.3502293825149536, "generation_rouge_l_f1": 0.3076923076923077, "generation_token_f1": 0.46153846153846156, "prompt_id": "938d726d90109110", "reading_canonical_log_prob": -21.603670120239258, "reading_distances": [25, 24, 32, 23], "reading_log_probs": [-21.449405670166016, -23.011354446411133, -22.30550765991211, -23.481399536132812], "reading_logprob_std": 0.8852248982006321, "tokenization_path_logprob_std": 67.95852438084941, "trajectory_balance_rmse": 498.70263794413614} +{"canonical_length": 6, "canonical_nll": 0.23918883005777994, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "b4288f8c25b7da9a", "reading_canonical_log_prob": -1.4351329803466797, "reading_distances": [42, 33, 33, 37], "reading_log_probs": [-2.770320415496826, -1.8075505495071411, -2.6101090908050537, -2.239583730697632], "reading_logprob_std": 0.42839981679418265, "tokenization_path_logprob_std": 15.42454903551837, "trajectory_balance_rmse": 159.79501100608198} +{"canonical_length": 266, "canonical_nll": 1.2731450589975917, "generation_rouge_l_f1": 0.2829827915869981, "generation_token_f1": 0.5468451242829828, "prompt_id": "3c7a013fe65f77bc", "reading_canonical_log_prob": -338.6565856933594, "reading_distances": [27, 25, 25, 26], "reading_log_probs": [-333.0740966796875, -332.0958557128906, -333.1029968261719, -332.1750183105469], "reading_logprob_std": 0.5513526235415445, "tokenization_path_logprob_std": 51.97267149540695, "trajectory_balance_rmse": 5484.339166085644} +{"canonical_length": 28, "canonical_nll": 1.6156219754900252, "generation_rouge_l_f1": 0.11194029850746269, "generation_token_f1": 0.13432835820895522, "prompt_id": "66c2a007594b76e9", "reading_canonical_log_prob": -45.2374153137207, "reading_distances": [61, 62, 51, 74], "reading_log_probs": [-47.21540832519531, -44.821449279785156, -45.59288787841797, -46.620323181152344], "reading_logprob_std": 1.064754547840242, "tokenization_path_logprob_std": 71.00533025075791, "trajectory_balance_rmse": 603.9253597398244} +{"canonical_length": 321, "canonical_nll": 1.5551831967362733, "generation_rouge_l_f1": 0.20103986135181978, "generation_token_f1": 0.415944540727903, "prompt_id": "dd27e1dd582c59a2", "reading_canonical_log_prob": -499.21380615234375, "reading_distances": [22, 25, 24, 24], "reading_log_probs": [-505.2488098144531, -505.1122131347656, -505.69586181640625, -504.99407958984375], "reading_logprob_std": 0.3069339294019096, "tokenization_path_logprob_std": 171.47827824621714, "trajectory_balance_rmse": 6751.3275219411125} +{"canonical_length": 21, "canonical_nll": 0.4228297642299107, "generation_rouge_l_f1": 0.14285714285714285, "generation_token_f1": 0.2857142857142857, "prompt_id": "3c6903de8714c50f", "reading_canonical_log_prob": -8.879425048828125, "reading_distances": [18, 15, 16, 12], "reading_log_probs": [-15.1636962890625, -19.732044219970703, -17.934425354003906, -16.88898277282715], "reading_logprob_std": 1.9133465009039485, "tokenization_path_logprob_std": 65.68352410072158, "trajectory_balance_rmse": 590.9443401370664} +{"canonical_length": 45, "canonical_nll": 1.3689893086751301, "generation_rouge_l_f1": 0.07717041800643087, "generation_token_f1": 0.12218649517684886, "prompt_id": "0692a96622cd2aed", "reading_canonical_log_prob": -61.60451889038086, "reading_distances": [21, 27, 23, 25], "reading_log_probs": [-63.596954345703125, -63.46385192871094, -61.351932525634766, -62.774330139160156], "reading_logprob_std": 1.0284862924666, "tokenization_path_logprob_std": 48.257828586155846, "trajectory_balance_rmse": 1310.8767606357724} +{"canonical_length": 322, "canonical_nll": 0.9593758908858211, "generation_rouge_l_f1": 0.2326388888888889, "generation_token_f1": 0.45833333333333337, "prompt_id": "e5a0acf44a48d7ab", "reading_canonical_log_prob": -308.9190368652344, "reading_distances": [25, 33, 23, 26], "reading_log_probs": [-311.7083740234375, -309.4376220703125, -310.181396484375, -311.374755859375], "reading_logprob_std": 1.053924025189389, "tokenization_path_logprob_std": 162.4008248635208, "trajectory_balance_rmse": 5402.6076589164} +{"canonical_length": 191, "canonical_nll": 0.9514161754028959, "generation_rouge_l_f1": 0.17575757575757578, "generation_token_f1": 0.40606060606060607, "prompt_id": "4e165cac00833611", "reading_canonical_log_prob": -181.72048950195312, "reading_distances": [18, 19, 18, 15], "reading_log_probs": [-177.4768829345703, -177.49813842773438, -175.76296997070312, -176.21945190429688], "reading_logprob_std": 0.8838036544198486, "tokenization_path_logprob_std": 35.56319505082702, "trajectory_balance_rmse": 2961.616513638619} +{"canonical_length": 296, "canonical_nll": 1.2419410911766258, "generation_rouge_l_f1": 0.20701754385964913, "generation_token_f1": 0.4701754385964912, "prompt_id": "eadf46baa67f8e23", "reading_canonical_log_prob": -367.61456298828125, "reading_distances": [25, 30, 24, 31], "reading_log_probs": [-370.5550537109375, -372.2698059082031, -370.2533264160156, -370.89208984375], "reading_logprob_std": 0.8905671239265319, "tokenization_path_logprob_std": 344.519151321479, "trajectory_balance_rmse": 6969.003567322371} +{"canonical_length": 97, "canonical_nll": 1.007236913307426, "generation_rouge_l_f1": 0.29965156794425085, "generation_token_f1": 0.42508710801393734, "prompt_id": "228080ce5449beea", "reading_canonical_log_prob": -97.70198059082031, "reading_distances": [17, 17, 23, 21], "reading_log_probs": [-96.38429260253906, -94.98582458496094, -95.91560363769531, -95.53662109375], "reading_logprob_std": 0.5919897787955629, "tokenization_path_logprob_std": 83.9104798391754, "trajectory_balance_rmse": 1172.775175716004} +{"canonical_length": 345, "canonical_nll": 1.0864547066066577, "generation_rouge_l_f1": 0.20270270270270271, "generation_token_f1": 0.4155405405405405, "prompt_id": "0584d61b9b2b3717", "reading_canonical_log_prob": -374.8268737792969, "reading_distances": [39, 24, 40, 33], "reading_log_probs": [-384.31787109375, -381.55059814453125, -385.392333984375, -385.7952880859375], "reading_logprob_std": 1.913411656731544, "tokenization_path_logprob_std": 191.6948690962116, "trajectory_balance_rmse": 6293.5198748537405} +{"canonical_length": 16, "canonical_nll": 1.5553994178771973, "generation_rouge_l_f1": 0.5185185185185186, "generation_token_f1": 0.5185185185185186, "prompt_id": "f605a0e57df216ae", "reading_canonical_log_prob": -24.886390686035156, "reading_distances": [34, 36, 38, 30], "reading_log_probs": [-26.42900848388672, -26.345434188842773, -28.653362274169922, -27.426095962524414], "reading_logprob_std": 1.0781721570569476, "tokenization_path_logprob_std": 52.87705640991752, "trajectory_balance_rmse": 291.3147846194167} +{"canonical_length": 337, "canonical_nll": 1.1077427170750649, "generation_rouge_l_f1": 0.22673434856175972, "generation_token_f1": 0.47038917089678506, "prompt_id": "ab315c4f854a9130", "reading_canonical_log_prob": -373.3092956542969, "reading_distances": [27, 25, 31, 21], "reading_log_probs": [-373.94366455078125, -374.02142333984375, -374.592041015625, -373.00628662109375], "reading_logprob_std": 0.6567495680717176, "tokenization_path_logprob_std": 200.25989551894722, "trajectory_balance_rmse": 7752.39578911746} +{"canonical_length": 283, "canonical_nll": 1.1397315790290967, "generation_rouge_l_f1": 0.22344322344322345, "generation_token_f1": 0.4578754578754579, "prompt_id": "93f170181468154b", "reading_canonical_log_prob": -322.5440368652344, "reading_distances": [27, 25, 27, 26], "reading_log_probs": [-319.088134765625, -322.138427734375, -322.43829345703125, -317.7754211425781], "reading_logprob_std": 2.293454303402111, "tokenization_path_logprob_std": 62.74276927481467, "trajectory_balance_rmse": 5890.722554358485} +{"canonical_length": 32, "canonical_nll": 0.8535487651824951, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.23529411764705882, "prompt_id": "ce3af5b67c725855", "reading_canonical_log_prob": -27.313560485839844, "reading_distances": [15, 15, 15, 17], "reading_log_probs": [-31.7949161529541, -28.774036407470703, -31.72905158996582, -32.2695426940918], "reading_logprob_std": 1.596823213266092, "tokenization_path_logprob_std": 97.96086702076768, "trajectory_balance_rmse": 860.0298361144888} +{"canonical_length": 64, "canonical_nll": 1.5658202171325684, "generation_rouge_l_f1": 0.2105263157894737, "generation_token_f1": 0.2982456140350877, "prompt_id": "5d5d479145243991", "reading_canonical_log_prob": -100.21249389648438, "reading_distances": [23, 19, 19, 21], "reading_log_probs": [-108.5742416381836, -107.55641174316406, -113.07633972167969, -111.36782836914062], "reading_logprob_std": 2.5335170596511167, "tokenization_path_logprob_std": 92.53750218153775, "trajectory_balance_rmse": 1378.9431159270234} +{"canonical_length": 259, "canonical_nll": 1.9085857140971887, "generation_rouge_l_f1": 0.20754716981132076, "generation_token_f1": 0.4641509433962264, "prompt_id": "a2496cc91dc1f05e", "reading_canonical_log_prob": -494.3236999511719, "reading_distances": [14, 15, 15, 16], "reading_log_probs": [-494.61614990234375, -492.6492919921875, -494.0259094238281, -494.08258056640625], "reading_logprob_std": 0.8393555765168552, "tokenization_path_logprob_std": 92.1403235053795, "trajectory_balance_rmse": 6466.347560902792} +{"canonical_length": 201, "canonical_nll": 1.0795198506976835, "generation_rouge_l_f1": 0.19069767441860466, "generation_token_f1": 0.3534883720930232, "prompt_id": "419c50c5b36f610d", "reading_canonical_log_prob": -216.98348999023438, "reading_distances": [36, 29, 46, 37], "reading_log_probs": [-216.70046997070312, -214.8464813232422, -216.93948364257812, -216.01123046875], "reading_logprob_std": 0.9384564350898491, "tokenization_path_logprob_std": 212.6403727685858, "trajectory_balance_rmse": 4944.315985207996} +{"canonical_length": 120, "canonical_nll": 0.606718381245931, "generation_rouge_l_f1": 0.23097112860892388, "generation_token_f1": 0.3779527559055118, "prompt_id": "07d185247528da20", "reading_canonical_log_prob": -72.80620574951172, "reading_distances": [20, 20, 15, 20], "reading_log_probs": [-78.01546478271484, -77.80101013183594, -81.73532104492188, -81.24406433105469], "reading_logprob_std": 2.079301449020986, "tokenization_path_logprob_std": 96.02941181669614, "trajectory_balance_rmse": 1918.461722294876} +{"canonical_length": 187, "canonical_nll": 1.3564384582845923, "generation_rouge_l_f1": 0.2546583850931677, "generation_token_f1": 0.48447204968944096, "prompt_id": "f75ec5897fcc8e0d", "reading_canonical_log_prob": -253.65399169921875, "reading_distances": [29, 29, 22, 32], "reading_log_probs": [-248.03530883789062, -248.2773895263672, -248.0213623046875, -248.23638916015625], "reading_logprob_std": 0.13313476204187166, "tokenization_path_logprob_std": 74.18453957556979, "trajectory_balance_rmse": 4006.007806149114} +{"canonical_length": 362, "canonical_nll": 1.537109375, "generation_rouge_l_f1": 0.3103448275862069, "generation_token_f1": 0.4420062695924765, "prompt_id": "cd40c2a793f29f44", "reading_canonical_log_prob": -556.43359375, "reading_distances": [25, 21, 17, 17], "reading_log_probs": [-556.2593994140625, -559.0138549804688, -559.49169921875, -556.5173950195312], "reading_logprob_std": 1.6685443293792386, "tokenization_path_logprob_std": 256.3529858900021, "trajectory_balance_rmse": 7982.6303390563935} +{"canonical_length": 131, "canonical_nll": 0.9639425496108659, "generation_rouge_l_f1": 0.19076923076923077, "generation_token_f1": 0.3446153846153846, "prompt_id": "79a9080b044df25c", "reading_canonical_log_prob": -126.27647399902344, "reading_distances": [32, 28, 28, 29], "reading_log_probs": [-125.9842300415039, -125.6384506225586, -126.96647644042969, -127.39305114746094], "reading_logprob_std": 0.8212485341923692, "tokenization_path_logprob_std": 104.12839257585514, "trajectory_balance_rmse": 2673.000961462308} +{"canonical_length": 59, "canonical_nll": 1.8823462017511918, "generation_rouge_l_f1": 0.13166144200626959, "generation_token_f1": 0.2068965517241379, "prompt_id": "56223c1e8305751f", "reading_canonical_log_prob": -111.05842590332031, "reading_distances": [20, 19, 18, 21], "reading_log_probs": [-110.65141296386719, -110.08001708984375, -109.20925903320312, -109.27253723144531], "reading_logprob_std": 0.6905219605088163, "tokenization_path_logprob_std": 98.94090889823758, "trajectory_balance_rmse": 1417.9643098945983} +{"canonical_length": 24, "canonical_nll": 0.8137693405151367, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.34285714285714286, "prompt_id": "5528bdba3c5f6117", "reading_canonical_log_prob": -19.53046417236328, "reading_distances": [47, 41, 48, 44], "reading_log_probs": [-19.127620697021484, -19.494033813476562, -18.885149002075195, -17.82988739013672], "reading_logprob_std": 0.7147764179599323, "tokenization_path_logprob_std": 76.78132657393009, "trajectory_balance_rmse": 653.2554760601084} +{"canonical_length": 351, "canonical_nll": 1.3060545309995995, "generation_rouge_l_f1": 0.284789644012945, "generation_token_f1": 0.5242718446601942, "prompt_id": "daf735e510f876d3", "reading_canonical_log_prob": -458.4251403808594, "reading_distances": [19, 21, 16, 18], "reading_log_probs": [-459.68170166015625, -459.32940673828125, -459.346435546875, -456.9859924316406], "reading_logprob_std": 1.2438828862526066, "tokenization_path_logprob_std": 567.0942594756056, "trajectory_balance_rmse": 7953.419627007412} +{"canonical_length": 73, "canonical_nll": 1.2171351811657214, "generation_rouge_l_f1": 0.2011834319526627, "generation_token_f1": 0.28994082840236685, "prompt_id": "d276e797ca2de2bf", "reading_canonical_log_prob": -88.85086822509766, "reading_distances": [53, 53, 48, 42], "reading_log_probs": [-85.74519348144531, -85.34709167480469, -86.02994537353516, -86.09745025634766], "reading_logprob_std": 0.3412645122594866, "tokenization_path_logprob_std": 144.4270469070739, "trajectory_balance_rmse": 1298.9561225883576} +{"canonical_length": 107, "canonical_nll": 1.5938740667895737, "generation_rouge_l_f1": 0.2209302325581395, "generation_token_f1": 0.3662790697674418, "prompt_id": "60ad70d4930701a2", "reading_canonical_log_prob": -170.54452514648438, "reading_distances": [12, 13, 10, 12], "reading_log_probs": [-166.79832458496094, -169.40142822265625, -171.0762939453125, -168.2601318359375], "reading_logprob_std": 1.8086033399016894, "tokenization_path_logprob_std": 91.11780121890504, "trajectory_balance_rmse": 2363.0222366773755} +{"canonical_length": 11, "canonical_nll": 0.04532590779391202, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "91d144e62a7c6f3d", "reading_canonical_log_prob": -0.4985849857330322, "reading_distances": [35, 35, 25, 26], "reading_log_probs": [-0.5829036831855774, -0.590377151966095, -0.4248327612876892, -0.48574328422546387], "reading_logprob_std": 0.0798674399548776, "tokenization_path_logprob_std": 43.23038876832368, "trajectory_balance_rmse": 301.8076207407363} +{"canonical_length": 31, "canonical_nll": 1.0667569560389365, "generation_rouge_l_f1": 0.2012578616352201, "generation_token_f1": 0.25157232704402516, "prompt_id": "663423620e0283d5", "reading_canonical_log_prob": -33.06946563720703, "reading_distances": [15, 15, 18, 18], "reading_log_probs": [-33.57304382324219, -34.656333923339844, -34.06163024902344, -33.975799560546875], "reading_logprob_std": 0.4470843233879336, "tokenization_path_logprob_std": 48.798844174752105, "trajectory_balance_rmse": 498.86579516957556} +{"canonical_length": 51, "canonical_nll": 0.08912045347924326, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "eece35ce9b12a6f2", "reading_canonical_log_prob": -4.545143127441406, "reading_distances": [24, 22, 23, 23], "reading_log_probs": [-4.9041666984558105, -4.945049285888672, -4.88663387298584, -4.7186665534973145], "reading_logprob_std": 0.09969269954445625, "tokenization_path_logprob_std": 75.63468463295757, "trajectory_balance_rmse": 1072.5944388720197} +{"canonical_length": 156, "canonical_nll": 1.4880605844350963, "generation_rouge_l_f1": 0.23796033994334276, "generation_token_f1": 0.424929178470255, "prompt_id": "25055edbaab1f2e6", "reading_canonical_log_prob": -232.137451171875, "reading_distances": [9, 5, 8, 8], "reading_log_probs": [-232.6441192626953, -228.10845947265625, -228.75030517578125, -229.2993927001953], "reading_logprob_std": 2.021818047798904, "tokenization_path_logprob_std": 178.26833447152865, "trajectory_balance_rmse": 3311.8574243368207} +{"canonical_length": 297, "canonical_nll": 0.9560351644702231, "generation_rouge_l_f1": 0.37586206896551727, "generation_token_f1": 0.5896551724137931, "prompt_id": "51338c6674b54120", "reading_canonical_log_prob": -283.94244384765625, "reading_distances": [28, 22, 22, 29], "reading_log_probs": [-280.4610900878906, -278.73126220703125, -280.75921630859375, -281.353759765625], "reading_logprob_std": 1.1262628504479084, "tokenization_path_logprob_std": 199.08932436604368, "trajectory_balance_rmse": 5396.526054171748} +{"canonical_length": 130, "canonical_nll": 2.1048358623798076, "generation_rouge_l_f1": 0.14814814814814814, "generation_token_f1": 0.2857142857142857, "prompt_id": "156614ecc1fc9430", "reading_canonical_log_prob": -273.628662109375, "reading_distances": [67, 59, 68, 64], "reading_log_probs": [-276.17327880859375, -272.9148864746094, -274.10577392578125, -276.7738037109375], "reading_logprob_std": 1.795368237382573, "tokenization_path_logprob_std": 96.97716885752095, "trajectory_balance_rmse": 3382.324388843573} +{"canonical_length": 234, "canonical_nll": 1.5171941773504274, "generation_rouge_l_f1": 0.2677824267782427, "generation_token_f1": 0.5146443514644351, "prompt_id": "39459679cbe7ca40", "reading_canonical_log_prob": -355.0234375, "reading_distances": [25, 26, 27, 29], "reading_log_probs": [-357.5332336425781, -354.65960693359375, -356.5848083496094, -355.92431640625], "reading_logprob_std": 1.2072004743062432, "tokenization_path_logprob_std": 87.89537668124595, "trajectory_balance_rmse": 4280.3991262710715} +{"canonical_length": 81, "canonical_nll": 1.123798417456356, "generation_rouge_l_f1": 0.31155778894472363, "generation_token_f1": 0.4824120603015076, "prompt_id": "0fc229d90ecbfd68", "reading_canonical_log_prob": -91.02767181396484, "reading_distances": [24, 26, 32, 30], "reading_log_probs": [-91.7421646118164, -91.37423706054688, -91.23896026611328, -90.58145141601562], "reading_logprob_std": 0.48433521420840053, "tokenization_path_logprob_std": 97.22056384128074, "trajectory_balance_rmse": 1851.0704090400907} +{"canonical_length": 7, "canonical_nll": 2.245678084237235, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.25, "prompt_id": "fc7a567baaf9e2df", "reading_canonical_log_prob": -15.719746589660645, "reading_distances": [19, 24, 24, 21], "reading_log_probs": [-15.929374694824219, -16.13247299194336, -16.358802795410156, -15.086027145385742], "reading_logprob_std": 0.555512109530439, "tokenization_path_logprob_std": 48.38054360022746, "trajectory_balance_rmse": 120.58313528743558} +{"canonical_length": 3, "canonical_nll": 1.6848023732503254, "generation_rouge_l_f1": 0.05882352941176471, "generation_token_f1": 0.05882352941176471, "prompt_id": "14871a1dde668d87", "reading_canonical_log_prob": -5.054407119750977, "reading_distances": [30, 26, 20, 26], "reading_log_probs": [-4.5701117515563965, -4.504090309143066, -4.456147193908691, -4.646451950073242], "reading_logprob_std": 0.08264172498064722, "tokenization_path_logprob_std": 24.52409260094443, "trajectory_balance_rmse": 41.70082247576132} +{"canonical_length": 3, "canonical_nll": 1.8749135335286458, "generation_rouge_l_f1": 0.0076045627376425855, "generation_token_f1": 0.0076045627376425855, "prompt_id": "408cd1b200daeec5", "reading_canonical_log_prob": -5.6247406005859375, "reading_distances": [31, 29, 23, 30], "reading_log_probs": [-5.508069038391113, -4.915828227996826, -5.5214362144470215, -5.322650909423828], "reading_logprob_std": 0.28241356587913174, "tokenization_path_logprob_std": 21.453109937047028, "trajectory_balance_rmse": 57.159632954847126} +{"canonical_length": 129, "canonical_nll": 1.5767617632252302, "generation_rouge_l_f1": 0.11428571428571431, "generation_token_f1": 0.12857142857142856, "prompt_id": "6db695863766c033", "reading_canonical_log_prob": -203.4022674560547, "reading_distances": [9, 13, 11, 12], "reading_log_probs": [-199.83154296875, -200.97021484375, -200.9489288330078, -200.9207000732422], "reading_logprob_std": 0.5579045375670363, "tokenization_path_logprob_std": 93.92870473441435, "trajectory_balance_rmse": 2563.153591383067} +{"canonical_length": 6, "canonical_nll": 0.39832321802775067, "generation_rouge_l_f1": 0.888888888888889, "generation_token_f1": 0.888888888888889, "prompt_id": "66a509bfd96df65b", "reading_canonical_log_prob": -2.389939308166504, "reading_distances": [16, 17, 13, 19], "reading_log_probs": [-2.41169810295105, -2.3813223838806152, -2.398186445236206, -2.3939805030822754], "reading_logprob_std": 0.012521571536385499, "tokenization_path_logprob_std": 28.55277891576118, "trajectory_balance_rmse": 104.32616330914547} +{"canonical_length": 3, "canonical_nll": 0.9496790568033854, "generation_rouge_l_f1": 0.02247191011235955, "generation_token_f1": 0.02247191011235955, "prompt_id": "2c889c2667571076", "reading_canonical_log_prob": -2.8490371704101562, "reading_distances": [13, 18, 20, 19], "reading_log_probs": [-3.0269203186035156, -4.058485984802246, -3.06827712059021, -3.0089492797851562], "reading_logprob_std": 0.5124875404683528, "tokenization_path_logprob_std": 53.599188110496954, "trajectory_balance_rmse": 96.43854467182499} +{"canonical_length": 15, "canonical_nll": 0.7033049265543619, "generation_rouge_l_f1": 0.0948905109489051, "generation_token_f1": 0.0948905109489051, "prompt_id": "79afdb8eb4cf0c62", "reading_canonical_log_prob": -10.54957389831543, "reading_distances": [23, 24, 26, 24], "reading_log_probs": [-13.363221168518066, -12.579143524169922, -11.885316848754883, -12.619470596313477], "reading_logprob_std": 0.6037485195092513, "tokenization_path_logprob_std": 78.12662356873935, "trajectory_balance_rmse": 302.33019832119015} +{"canonical_length": 145, "canonical_nll": 1.3110752500336746, "generation_rouge_l_f1": 0.27676240208877284, "generation_token_f1": 0.4751958224543081, "prompt_id": "51fd58caa03ae57f", "reading_canonical_log_prob": -190.1059112548828, "reading_distances": [29, 30, 29, 22], "reading_log_probs": [-194.91221618652344, -194.48196411132812, -193.50515747070312, -191.85369873046875], "reading_logprob_std": 1.3573474807830315, "tokenization_path_logprob_std": 93.38102122876286, "trajectory_balance_rmse": 3216.410108331784} +{"canonical_length": 19, "canonical_nll": 1.2548827121132298, "generation_rouge_l_f1": 0.21487603305785125, "generation_token_f1": 0.2644628099173554, "prompt_id": "8671c943f4179a67", "reading_canonical_log_prob": -23.842771530151367, "reading_distances": [16, 13, 15, 16], "reading_log_probs": [-25.987287521362305, -25.832996368408203, -25.121288299560547, -26.48445701599121], "reading_logprob_std": 0.5634784984435279, "tokenization_path_logprob_std": 71.91488962054537, "trajectory_balance_rmse": 473.4888331102084} +{"canonical_length": 289, "canonical_nll": 1.2038521420171928, "generation_rouge_l_f1": 0.2589928057553957, "generation_token_f1": 0.4604316546762589, "prompt_id": "9cb312e17f963300", "reading_canonical_log_prob": -347.91326904296875, "reading_distances": [21, 13, 21, 18], "reading_log_probs": [-352.66375732421875, -351.9927062988281, -352.7725830078125, -349.38909912109375], "reading_logprob_std": 1.581678435613915, "tokenization_path_logprob_std": 122.23187498570472, "trajectory_balance_rmse": 6633.067168496811} +{"canonical_length": 28, "canonical_nll": 1.822824205671038, "generation_rouge_l_f1": 0.2631578947368421, "generation_token_f1": 0.3684210526315789, "prompt_id": "9227ddd5a86a7e6a", "reading_canonical_log_prob": -51.03907775878906, "reading_distances": [25, 25, 27, 24], "reading_log_probs": [-55.6065559387207, -54.82276153564453, -54.819122314453125, -55.98735046386719], "reading_logprob_std": 0.5845530681679896, "tokenization_path_logprob_std": 29.24938334135159, "trajectory_balance_rmse": 689.3266291236752} +{"canonical_length": 223, "canonical_nll": 1.6019456803531391, "generation_rouge_l_f1": 0.26359832635983266, "generation_token_f1": 0.49372384937238495, "prompt_id": "3216cdb4c067a3d8", "reading_canonical_log_prob": -357.23388671875, "reading_distances": [46, 47, 49, 44], "reading_log_probs": [-356.3343505859375, -356.65447998046875, -358.9302978515625, -358.0061950683594], "reading_logprob_std": 1.207508993192802, "tokenization_path_logprob_std": 188.65909476007621, "trajectory_balance_rmse": 5171.315355534234} +{"canonical_length": 108, "canonical_nll": 0.805842364275897, "generation_rouge_l_f1": 0.3220338983050847, "generation_token_f1": 0.4830508474576271, "prompt_id": "1ef389c9aee61602", "reading_canonical_log_prob": -87.03097534179688, "reading_distances": [14, 9, 14, 15], "reading_log_probs": [-88.65896606445312, -86.95802307128906, -87.53179931640625, -87.19412231445312], "reading_logprob_std": 0.7532411602281405, "tokenization_path_logprob_std": 69.91376465681834, "trajectory_balance_rmse": 1587.355309603232} +{"canonical_length": 6, "canonical_nll": 2.8807589213053384, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "a730d1195f266d12", "reading_canonical_log_prob": -17.28455352783203, "reading_distances": [32, 30, 25, 29], "reading_log_probs": [-21.850780487060547, -18.42804718017578, -17.522947311401367, -18.41429901123047], "reading_logprob_std": 1.9119917812216762, "tokenization_path_logprob_std": 38.95363520849906, "trajectory_balance_rmse": 125.97685868965898} +{"canonical_length": 26, "canonical_nll": 1.29210692185622, "generation_rouge_l_f1": 0.08835341365461848, "generation_token_f1": 0.11244979919678713, "prompt_id": "7f409abdb6fbebf8", "reading_canonical_log_prob": -33.59477996826172, "reading_distances": [15, 11, 11, 11], "reading_log_probs": [-37.15867614746094, -38.072879791259766, -35.855751037597656, -38.25042724609375], "reading_logprob_std": 1.0957098064034339, "tokenization_path_logprob_std": 67.7905366646785, "trajectory_balance_rmse": 465.38097386732505} +{"canonical_length": 168, "canonical_nll": 1.383310045514788, "generation_rouge_l_f1": 0.26044226044226043, "generation_token_f1": 0.49631449631449637, "prompt_id": "1c78ad010cb27d62", "reading_canonical_log_prob": -232.39608764648438, "reading_distances": [39, 33, 35, 41], "reading_log_probs": [-232.43600463867188, -232.213134765625, -233.20338439941406, -233.3011474609375], "reading_logprob_std": 0.5447429331808243, "tokenization_path_logprob_std": 26.031601732479636, "trajectory_balance_rmse": 3793.5680389476615} +{"canonical_length": 15, "canonical_nll": 0.5600173950195313, "generation_rouge_l_f1": 0.07954545454545454, "generation_token_f1": 0.09090909090909091, "prompt_id": "779d0398df55911b", "reading_canonical_log_prob": -8.400260925292969, "reading_distances": [6, 5, 8, 7], "reading_log_probs": [-11.791425704956055, -8.124237060546875, -7.864350318908691, -8.849211692810059], "reading_logprob_std": 1.8048522145492103, "tokenization_path_logprob_std": 59.64287503902799, "trajectory_balance_rmse": 217.25837475895335} +{"canonical_length": 11, "canonical_nll": 0.4202992699363015, "generation_rouge_l_f1": 0.5454545454545454, "generation_token_f1": 0.5454545454545454, "prompt_id": "3e344435d6b1501d", "reading_canonical_log_prob": -4.623291969299316, "reading_distances": [41, 47, 43, 38], "reading_log_probs": [-5.578853607177734, -5.93688440322876, -5.603094577789307, -5.838541030883789], "reading_logprob_std": 0.17624162683043612, "tokenization_path_logprob_std": 33.70998109385471, "trajectory_balance_rmse": 156.66842260340715} +{"canonical_length": 376, "canonical_nll": 0.48820735038595, "generation_rouge_l_f1": 0.26824457593688367, "generation_token_f1": 0.45759368836291914, "prompt_id": "e2124bce16d0a6b8", "reading_canonical_log_prob": -183.5659637451172, "reading_distances": [13, 17, 9, 12], "reading_log_probs": [-182.57191467285156, -181.65879821777344, -183.24244689941406, -183.38844299316406], "reading_logprob_std": 0.7890417543109961, "tokenization_path_logprob_std": 193.7239044624914, "trajectory_balance_rmse": 5220.584437716102} +{"canonical_length": 240, "canonical_nll": 2.105642064412435, "generation_rouge_l_f1": 0.30991735537190085, "generation_token_f1": 0.5165289256198348, "prompt_id": "6a9d3b5c30c1cdf4", "reading_canonical_log_prob": -505.3540954589844, "reading_distances": [11, 18, 14, 13], "reading_log_probs": [-507.35552978515625, -512.914794921875, -508.2596435546875, -510.1816711425781], "reading_logprob_std": 2.458734692101496, "tokenization_path_logprob_std": 166.19669962803695, "trajectory_balance_rmse": 4532.526915021255} +{"canonical_length": 271, "canonical_nll": 1.0767484432656826, "generation_rouge_l_f1": 0.26305609284332687, "generation_token_f1": 0.5377176015473888, "prompt_id": "63c15bf241bee263", "reading_canonical_log_prob": -291.798828125, "reading_distances": [33, 24, 32, 30], "reading_log_probs": [-290.7236022949219, -290.50244140625, -292.57720947265625, -293.3027648925781], "reading_logprob_std": 1.3786997569574235, "tokenization_path_logprob_std": 39.553043789170026, "trajectory_balance_rmse": 6131.586981686497} +{"canonical_length": 25, "canonical_nll": 1.298721923828125, "generation_rouge_l_f1": 0.2777777777777778, "generation_token_f1": 0.4444444444444445, "prompt_id": "9932e0e2da85e27d", "reading_canonical_log_prob": -32.468048095703125, "reading_distances": [24, 18, 24, 22], "reading_log_probs": [-31.52056121826172, -31.431713104248047, -31.273977279663086, -31.189884185791016], "reading_logprob_std": 0.14957673352352271, "tokenization_path_logprob_std": 26.40419680540599, "trajectory_balance_rmse": 545.5832063243062} +{"canonical_length": 114, "canonical_nll": 0.8219611184638843, "generation_rouge_l_f1": 0.29347826086956524, "generation_token_f1": 0.4891304347826088, "prompt_id": "f5ce2237f02c3721", "reading_canonical_log_prob": -93.70356750488281, "reading_distances": [24, 22, 21, 23], "reading_log_probs": [-94.60708618164062, -93.75623321533203, -93.46829223632812, -93.01624298095703], "reading_logprob_std": 0.6699769021490852, "tokenization_path_logprob_std": 154.17961069178756, "trajectory_balance_rmse": 2562.4629392516795} +{"canonical_length": 297, "canonical_nll": 1.4608878704032513, "generation_rouge_l_f1": 0.23595505617977527, "generation_token_f1": 0.453183520599251, "prompt_id": "3c5a7e9edbe46ab0", "reading_canonical_log_prob": -433.8836975097656, "reading_distances": [13, 21, 18, 16], "reading_log_probs": [-432.853759765625, -438.9137268066406, -436.6324462890625, -436.06536865234375], "reading_logprob_std": 2.499250988512229, "tokenization_path_logprob_std": 299.0509637415297, "trajectory_balance_rmse": 6093.888529835024} +{"canonical_length": 129, "canonical_nll": 1.1595475544301115, "generation_rouge_l_f1": 0.2921348314606741, "generation_token_f1": 0.5018726591760299, "prompt_id": "6097f6fb796f153f", "reading_canonical_log_prob": -149.58163452148438, "reading_distances": [26, 25, 32, 21], "reading_log_probs": [-147.74249267578125, -147.9511260986328, -147.33822631835938, -145.90118408203125], "reading_logprob_std": 0.9237773600535539, "tokenization_path_logprob_std": 191.10890546069479, "trajectory_balance_rmse": 2618.14699366868} +{"canonical_length": 10, "canonical_nll": 0.40540170669555664, "generation_rouge_l_f1": 0.10769230769230768, "generation_token_f1": 0.10769230769230768, "prompt_id": "b2680f937aebb340", "reading_canonical_log_prob": -4.054017066955566, "reading_distances": [7, 10, 5, 6], "reading_log_probs": [-4.006986141204834, -4.909751892089844, -3.6354727745056152, -4.2198486328125], "reading_logprob_std": 0.5353789579498213, "tokenization_path_logprob_std": 16.82551075916569, "trajectory_balance_rmse": 48.974065773969336} +{"canonical_length": 8, "canonical_nll": 0.9823455214500427, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "a6d7ae5badbbf976", "reading_canonical_log_prob": -7.858764171600342, "reading_distances": [29, 35, 25, 32], "reading_log_probs": [-9.58810806274414, -11.427614212036133, -11.388418197631836, -10.618824005126953], "reading_logprob_std": 0.8629034301134703, "tokenization_path_logprob_std": 16.594327006671048, "trajectory_balance_rmse": 87.91969965983384} +{"canonical_length": 31, "canonical_nll": 2.1707684916834675, "generation_rouge_l_f1": 0.06666666666666667, "generation_token_f1": 0.1, "prompt_id": "5eee9bc9190ed48e", "reading_canonical_log_prob": -67.2938232421875, "reading_distances": [26, 31, 23, 22], "reading_log_probs": [-76.7789535522461, -80.55557250976562, -73.11416625976562, -77.86541748046875], "reading_logprob_std": 3.0830148899274783, "tokenization_path_logprob_std": 143.03000073822037, "trajectory_balance_rmse": 799.375528018939} +{"canonical_length": 14, "canonical_nll": 0.8950517518179757, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.26666666666666666, "prompt_id": "89b79b80175b4a32", "reading_canonical_log_prob": -12.53072452545166, "reading_distances": [28, 41, 37, 37], "reading_log_probs": [-13.080567359924316, -13.741639137268066, -13.675737380981445, -13.752391815185547], "reading_logprob_std": 0.3231261249238737, "tokenization_path_logprob_std": 64.65655624029817, "trajectory_balance_rmse": 285.20405514707335} +{"canonical_length": 10, "canonical_nll": 0.4224852561950684, "generation_rouge_l_f1": 0.8888888888888888, "generation_token_f1": 0.8888888888888888, "prompt_id": "5a773e9fe3bdab08", "reading_canonical_log_prob": -4.224852561950684, "reading_distances": [19, 16, 20, 21], "reading_log_probs": [-3.065910577774048, -3.2584619522094727, -2.860952138900757, -3.070187568664551], "reading_logprob_std": 0.16236354433744643, "tokenization_path_logprob_std": 40.2267677693877, "trajectory_balance_rmse": 227.95746283184602} +{"canonical_length": 37, "canonical_nll": 1.6261190465978674, "generation_rouge_l_f1": 0.360655737704918, "generation_token_f1": 0.4918032786885245, "prompt_id": "bc2e832a6508e078", "reading_canonical_log_prob": -60.166404724121094, "reading_distances": [22, 12, 14, 17], "reading_log_probs": [-58.998897552490234, -58.088966369628906, -58.018131256103516, -58.2159309387207], "reading_logprob_std": 0.4530614687421513, "tokenization_path_logprob_std": 52.91443060952837, "trajectory_balance_rmse": 914.8905434828498} +{"canonical_length": 245, "canonical_nll": 0.7767197589485013, "generation_rouge_l_f1": 0.24053452115812915, "generation_token_f1": 0.31180400890868604, "prompt_id": "66a81182d7c90c5b", "reading_canonical_log_prob": -190.2963409423828, "reading_distances": [16, 10, 19, 16], "reading_log_probs": [-187.4478302001953, -187.49545288085938, -187.1167449951172, -187.0511474609375], "reading_logprob_std": 0.226269112221499, "tokenization_path_logprob_std": 192.9905782425385, "trajectory_balance_rmse": 3827.5129908788767} +{"canonical_length": 92, "canonical_nll": 1.474192246146824, "generation_rouge_l_f1": 0.23104693140794222, "generation_token_f1": 0.3537906137184116, "prompt_id": "aed52e93b5146883", "reading_canonical_log_prob": -135.6256866455078, "reading_distances": [24, 29, 23, 28], "reading_log_probs": [-136.96798706054688, -137.87640380859375, -136.24449157714844, -138.0652618408203], "reading_logprob_std": 0.8449205034894889, "tokenization_path_logprob_std": 165.64233678142193, "trajectory_balance_rmse": 1627.7226544687721} +{"canonical_length": 131, "canonical_nll": 1.0051942781637644, "generation_rouge_l_f1": 0.39426523297491045, "generation_token_f1": 0.6021505376344086, "prompt_id": "b1543c03f1b0c7a8", "reading_canonical_log_prob": -131.68045043945312, "reading_distances": [17, 20, 19, 19], "reading_log_probs": [-133.21083068847656, -133.818603515625, -131.72100830078125, -130.5833282470703], "reading_logprob_std": 1.462131870738828, "tokenization_path_logprob_std": 162.12659926771224, "trajectory_balance_rmse": 2778.9009514628124} +{"canonical_length": 501, "canonical_nll": 0.3674742494990488, "generation_rouge_l_f1": 0.11877394636015327, "generation_token_f1": 0.1724137931034483, "prompt_id": "c52f77a538102872", "reading_canonical_log_prob": -184.10459899902344, "reading_distances": [32, 29, 23, 31], "reading_log_probs": [-179.56825256347656, -182.77755737304688, -178.8671112060547, -182.62725830078125], "reading_logprob_std": 2.0330936581707593, "tokenization_path_logprob_std": 58.73630680541382, "trajectory_balance_rmse": 2906.3031252752317} +{"canonical_length": 67, "canonical_nll": 0.6357298893715019, "generation_rouge_l_f1": 0.21276595744680848, "generation_token_f1": 0.24113475177304966, "prompt_id": "fdb23a8f949c30eb", "reading_canonical_log_prob": -42.593902587890625, "reading_distances": [19, 20, 23, 21], "reading_log_probs": [-42.83153533935547, -42.8802490234375, -42.76683044433594, -42.57638931274414], "reading_logprob_std": 0.13326713892964343, "tokenization_path_logprob_std": 48.655749281743134, "trajectory_balance_rmse": 916.8778932892039} +{"canonical_length": 268, "canonical_nll": 1.258349632149312, "generation_rouge_l_f1": 0.2896825396825397, "generation_token_f1": 0.48015873015873023, "prompt_id": "8aa2f18f3606aa8c", "reading_canonical_log_prob": -337.2377014160156, "reading_distances": [22, 27, 24, 22], "reading_log_probs": [-341.71868896484375, -341.1161193847656, -341.9170837402344, -340.56732177734375], "reading_logprob_std": 0.6118739695462404, "tokenization_path_logprob_std": 396.6157351653902, "trajectory_balance_rmse": 5973.566115254468} +{"canonical_length": 123, "canonical_nll": 0.8465830485026041, "generation_rouge_l_f1": 0.17595307917888567, "generation_token_f1": 0.26392961876832843, "prompt_id": "75e528b4a6af3ee2", "reading_canonical_log_prob": -104.12971496582031, "reading_distances": [21, 18, 15, 19], "reading_log_probs": [-103.77888488769531, -102.08326721191406, -104.0050048828125, -103.23673248291016], "reading_logprob_std": 0.8580005376952415, "tokenization_path_logprob_std": 158.90093642838525, "trajectory_balance_rmse": 1540.9162680706029} +{"canonical_length": 323, "canonical_nll": 1.553100774900832, "generation_rouge_l_f1": 0.16055846422338568, "generation_token_f1": 0.32460732984293195, "prompt_id": "af23751b53e953f7", "reading_canonical_log_prob": -501.65155029296875, "reading_distances": [33, 36, 31, 28], "reading_log_probs": [-492.3343505859375, -492.0703125, -495.5277099609375, -495.8411560058594], "reading_logprob_std": 2.0173425348209837, "tokenization_path_logprob_std": 156.67425985788944, "trajectory_balance_rmse": 6615.425629882424} +{"canonical_length": 250, "canonical_nll": 1.67936865234375, "generation_rouge_l_f1": 0.23770491803278687, "generation_token_f1": 0.47540983606557374, "prompt_id": "b8fd0dbaa9eea228", "reading_canonical_log_prob": -419.8421630859375, "reading_distances": [15, 20, 19, 21], "reading_log_probs": [-417.6737976074219, -416.8128967285156, -416.964599609375, -416.4197998046875], "reading_logprob_std": 0.5236863243527875, "tokenization_path_logprob_std": 48.01318917166772, "trajectory_balance_rmse": 5489.310229578736} +{"canonical_length": 13, "canonical_nll": 2.238401853121244, "generation_rouge_l_f1": 0.28571428571428575, "generation_token_f1": 0.28571428571428575, "prompt_id": "44c43dead4708e04", "reading_canonical_log_prob": -29.099224090576172, "reading_distances": [34, 24, 29, 22], "reading_log_probs": [-30.821788787841797, -29.876663208007812, -30.067096710205078, -30.57366371154785], "reading_logprob_std": 0.43808991606876224, "tokenization_path_logprob_std": 35.23046749625592, "trajectory_balance_rmse": 399.45074779869145} +{"canonical_length": 138, "canonical_nll": 1.6932361989781477, "generation_rouge_l_f1": 0.21296296296296297, "generation_token_f1": 0.3888888888888889, "prompt_id": "36e5fedd319115db", "reading_canonical_log_prob": -233.66659545898438, "reading_distances": [30, 27, 35, 26], "reading_log_probs": [-225.07501220703125, -225.73382568359375, -228.5883331298828, -224.8309783935547], "reading_logprob_std": 1.7300788086728114, "tokenization_path_logprob_std": 82.9320446383179, "trajectory_balance_rmse": 3086.176751753272} +{"canonical_length": 13, "canonical_nll": 0.6030029883751502, "generation_rouge_l_f1": 0.8695652173913043, "generation_token_f1": 0.8695652173913043, "prompt_id": "15b19e1f65328c81", "reading_canonical_log_prob": -7.839038848876953, "reading_distances": [24, 21, 16, 24], "reading_log_probs": [-7.70454216003418, -7.587217330932617, -8.015172004699707, -7.30666971206665], "reading_logprob_std": 0.29331242851964073, "tokenization_path_logprob_std": 40.78151875726747, "trajectory_balance_rmse": 249.22935334154056} +{"canonical_length": 20, "canonical_nll": 1.0530811309814454, "generation_rouge_l_f1": 0.08633093525179855, "generation_token_f1": 0.1223021582733813, "prompt_id": "7e1ececb041b4ecb", "reading_canonical_log_prob": -21.061622619628906, "reading_distances": [21, 24, 26, 23], "reading_log_probs": [-23.679718017578125, -24.50615119934082, -23.52962875366211, -21.306434631347656], "reading_logprob_std": 1.3684652879021277, "tokenization_path_logprob_std": 95.05559187208553, "trajectory_balance_rmse": 498.20508648481854} +{"canonical_length": 359, "canonical_nll": 1.4923588744777159, "generation_rouge_l_f1": 0.28980891719745216, "generation_token_f1": 0.4585987261146497, "prompt_id": "4db416527bef041c", "reading_canonical_log_prob": -535.7568359375, "reading_distances": [18, 21, 22, 15], "reading_log_probs": [-532.680908203125, -532.4166259765625, -531.348388671875, -532.772705078125], "reading_logprob_std": 0.6551386973705011, "tokenization_path_logprob_std": 307.7321527465888, "trajectory_balance_rmse": 8124.962382123528} +{"canonical_length": 78, "canonical_nll": 0.7116216023763021, "generation_rouge_l_f1": 0.46478873239436613, "generation_token_f1": 0.6478873239436619, "prompt_id": "ba11bfd32f28c7ee", "reading_canonical_log_prob": -55.50648498535156, "reading_distances": [16, 14, 14, 18], "reading_log_probs": [-56.56306457519531, -57.58688735961914, -56.49299621582031, -56.827003479003906], "reading_logprob_std": 0.5006968986817538, "tokenization_path_logprob_std": 112.08431490514266, "trajectory_balance_rmse": 1669.794085791152} +{"canonical_length": 292, "canonical_nll": 1.4845747229171127, "generation_rouge_l_f1": 0.19168173598553345, "generation_token_f1": 0.42314647377938513, "prompt_id": "2bb0a1d6597bcbfc", "reading_canonical_log_prob": -433.4958190917969, "reading_distances": [50, 64, 52, 54], "reading_log_probs": [-441.4607849121094, -441.076416015625, -441.59136962890625, -444.3539733886719], "reading_logprob_std": 1.504849673241929, "tokenization_path_logprob_std": 215.63346941043397, "trajectory_balance_rmse": 6714.168834612766} +{"canonical_length": 2, "canonical_nll": 5.022811412811279, "generation_rouge_l_f1": 0.0078125, "generation_token_f1": 0.0078125, "prompt_id": "bad903d11c6658df", "reading_canonical_log_prob": -10.045622825622559, "reading_distances": [28, 34, 34, 30], "reading_log_probs": [-11.341814994812012, -11.378705024719238, -11.080408096313477, -10.33979606628418], "reading_logprob_std": 0.48223053221931544, "tokenization_path_logprob_std": 24.732117332762407, "trajectory_balance_rmse": 45.58544647750043} +{"canonical_length": 99, "canonical_nll": 0.9700160171046401, "generation_rouge_l_f1": 0.2156862745098039, "generation_token_f1": 0.33333333333333337, "prompt_id": "7c8a24a51c291f25", "reading_canonical_log_prob": -96.03158569335938, "reading_distances": [10, 13, 11, 10], "reading_log_probs": [-95.3028564453125, -95.24079895019531, -96.62167358398438, -97.74383544921875], "reading_logprob_std": 1.1948766322978674, "tokenization_path_logprob_std": 89.12922073760514, "trajectory_balance_rmse": 2055.8306273243984} +{"canonical_length": 108, "canonical_nll": 1.7094140229401764, "generation_rouge_l_f1": 0.21983914209115282, "generation_token_f1": 0.3914209115281502, "prompt_id": "cef16e435e5a8a53", "reading_canonical_log_prob": -184.61671447753906, "reading_distances": [20, 21, 18, 23], "reading_log_probs": [-185.92044067382812, -186.03851318359375, -186.27996826171875, -186.60768127441406], "reading_logprob_std": 0.303472889530618, "tokenization_path_logprob_std": 139.47211417772294, "trajectory_balance_rmse": 2777.147369442682} +{"canonical_length": 38, "canonical_nll": 2.9239859330026725, "generation_rouge_l_f1": 0.18823529411764708, "generation_token_f1": 0.23529411764705882, "prompt_id": "289239d155fcb98c", "reading_canonical_log_prob": -111.11146545410156, "reading_distances": [28, 17, 22, 19], "reading_log_probs": [-112.1539306640625, -111.33457946777344, -112.69694519042969, -112.89786529541016], "reading_logprob_std": 0.6988032462818332, "tokenization_path_logprob_std": 65.57795336743388, "trajectory_balance_rmse": 844.5770485008709} +{"canonical_length": 8, "canonical_nll": 1.0308445692062378, "generation_rouge_l_f1": 0.6, "generation_token_f1": 0.6, "prompt_id": "c036d970930b8362", "reading_canonical_log_prob": -8.246756553649902, "reading_distances": [24, 36, 31, 26], "reading_log_probs": [-6.416956901550293, -8.400917053222656, -7.138289928436279, -6.45182991027832], "reading_logprob_std": 0.927453082793802, "tokenization_path_logprob_std": 28.274265714860213, "trajectory_balance_rmse": 102.37647777778633} +{"canonical_length": 99, "canonical_nll": 1.0779903681591303, "generation_rouge_l_f1": 0.21666666666666667, "generation_token_f1": 0.33888888888888885, "prompt_id": "1d5889fbec769861", "reading_canonical_log_prob": -106.7210464477539, "reading_distances": [22, 16, 23, 24], "reading_log_probs": [-104.15177917480469, -104.98670959472656, -108.46324920654297, -104.98628234863281], "reading_logprob_std": 1.9182870956031504, "tokenization_path_logprob_std": 146.31785890739792, "trajectory_balance_rmse": 1910.7268780265895} +{"canonical_length": 185, "canonical_nll": 0.9173011573585304, "generation_rouge_l_f1": 0.18181818181818185, "generation_token_f1": 0.31449631449631454, "prompt_id": "c20bcd26ca03079c", "reading_canonical_log_prob": -169.70071411132812, "reading_distances": [38, 36, 38, 37], "reading_log_probs": [-164.73231506347656, -164.1512908935547, -164.59994506835938, -165.41311645507812], "reading_logprob_std": 0.5222821281731631, "tokenization_path_logprob_std": 155.4082717230081, "trajectory_balance_rmse": 3430.7060923581116} +{"canonical_length": 16, "canonical_nll": 0.9146069884300232, "generation_rouge_l_f1": 0.11398963730569948, "generation_token_f1": 0.12435233160621761, "prompt_id": "93426790937338d2", "reading_canonical_log_prob": -14.633711814880371, "reading_distances": [15, 17, 15, 16], "reading_log_probs": [-14.092694282531738, -12.490560531616211, -13.819975852966309, -13.692123413085938], "reading_logprob_std": 0.7088200065511128, "tokenization_path_logprob_std": 33.25198064769416, "trajectory_balance_rmse": 233.50349142184862} +{"canonical_length": 203, "canonical_nll": 1.449615478515625, "generation_rouge_l_f1": 0.18867924528301883, "generation_token_f1": 0.4276729559748428, "prompt_id": "f380c029d22e1217", "reading_canonical_log_prob": -294.2719421386719, "reading_distances": [19, 12, 12, 11], "reading_log_probs": [-293.4353942871094, -292.652099609375, -292.2770080566406, -292.13671875], "reading_logprob_std": 0.582234170708933, "tokenization_path_logprob_std": 145.67124444407108, "trajectory_balance_rmse": 4156.586288719923} +{"canonical_length": 191, "canonical_nll": 0.761893706796057, "generation_rouge_l_f1": 0.32592592592592595, "generation_token_f1": 0.437037037037037, "prompt_id": "b8396f86fe7c1b16", "reading_canonical_log_prob": -145.52169799804688, "reading_distances": [41, 36, 35, 34], "reading_log_probs": [-145.7591552734375, -147.0733642578125, -146.74314880371094, -145.58889770507812], "reading_logprob_std": 0.7285462095173529, "tokenization_path_logprob_std": 197.24507220801644, "trajectory_balance_rmse": 3021.6638773784894} +{"canonical_length": 149, "canonical_nll": 1.7084912805749266, "generation_rouge_l_f1": 0.16058394160583941, "generation_token_f1": 0.2822384428223844, "prompt_id": "6cb0fdf8b6463b4b", "reading_canonical_log_prob": -254.56520080566406, "reading_distances": [45, 51, 38, 49], "reading_log_probs": [-249.9072265625, -249.72938537597656, -249.87554931640625, -249.67193603515625], "reading_logprob_std": 0.11332668865785812, "tokenization_path_logprob_std": 124.54787366852169, "trajectory_balance_rmse": 3462.027220252949} +{"canonical_length": 118, "canonical_nll": 1.5649467080326405, "generation_rouge_l_f1": 0.23140495867768596, "generation_token_f1": 0.3471074380165289, "prompt_id": "7c85ca0add25adb8", "reading_canonical_log_prob": -184.66371154785156, "reading_distances": [33, 33, 38, 35], "reading_log_probs": [-177.72544860839844, -179.91067504882812, -178.01498413085938, -179.3507080078125], "reading_logprob_std": 1.048486168604494, "tokenization_path_logprob_std": 147.1365852567521, "trajectory_balance_rmse": 2843.97542961372} +{"canonical_length": 114, "canonical_nll": 1.561451560572574, "generation_rouge_l_f1": 0.21951219512195125, "generation_token_f1": 0.43089430894308944, "prompt_id": "9a1872631342b4a7", "reading_canonical_log_prob": -178.00547790527344, "reading_distances": [30, 29, 23, 28], "reading_log_probs": [-177.57431030273438, -177.46221923828125, -175.09024047851562, -178.16966247558594], "reading_logprob_std": 1.358528938315001, "tokenization_path_logprob_std": 216.90722028040528, "trajectory_balance_rmse": 2785.6550600792025} +{"canonical_length": 13, "canonical_nll": 2.055712626530574, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.2222222222222222, "prompt_id": "d49306af1b19f211", "reading_canonical_log_prob": -26.72426414489746, "reading_distances": [22, 22, 20, 22], "reading_log_probs": [-24.514266967773438, -24.634984970092773, -23.863544464111328, -24.76572036743164], "reading_logprob_std": 0.40076749119387634, "tokenization_path_logprob_std": 87.53086982567451, "trajectory_balance_rmse": 258.8650535929572} +{"canonical_length": 253, "canonical_nll": 1.6073753145843626, "generation_rouge_l_f1": 0.15073529411764705, "generation_token_f1": 0.32352941176470584, "prompt_id": "807f35218626237b", "reading_canonical_log_prob": -406.66595458984375, "reading_distances": [17, 15, 21, 21], "reading_log_probs": [-410.20751953125, -410.29327392578125, -409.56756591796875, -411.08251953125], "reading_logprob_std": 0.6209646146709962, "tokenization_path_logprob_std": 182.52120762800314, "trajectory_balance_rmse": 5559.485601730511} +{"canonical_length": 4, "canonical_nll": 2.7035861015319824, "generation_rouge_l_f1": 0.02222222222222222, "generation_token_f1": 0.02222222222222222, "prompt_id": "1a44d44c0ed30f8a", "reading_canonical_log_prob": -10.81434440612793, "reading_distances": [20, 20, 28, 22], "reading_log_probs": [-12.440252304077148, -12.790826797485352, -11.866523742675781, -11.094198226928711], "reading_logprob_std": 0.7412433171164391, "tokenization_path_logprob_std": 39.724407551013705, "trajectory_balance_rmse": 148.79105034090063} +{"canonical_length": 119, "canonical_nll": 0.2780339216985622, "generation_rouge_l_f1": 0.5289672544080605, "generation_token_f1": 0.6045340050377833, "prompt_id": "c78f29d81d80f8fc", "reading_canonical_log_prob": -33.086036682128906, "reading_distances": [21, 27, 20, 22], "reading_log_probs": [-31.76095199584961, -31.371618270874023, -30.97896957397461, -32.509578704833984], "reading_logprob_std": 0.6529044383135811, "tokenization_path_logprob_std": 92.49851044303797, "trajectory_balance_rmse": 1663.525493184312} +{"canonical_length": 279, "canonical_nll": 0.9341257840501792, "generation_rouge_l_f1": 0.3005780346820809, "generation_token_f1": 0.4431599229287091, "prompt_id": "dbd0e34044b57777", "reading_canonical_log_prob": -260.62109375, "reading_distances": [25, 18, 19, 23], "reading_log_probs": [-266.47943115234375, -264.6715087890625, -274.6369934082031, -273.42987060546875], "reading_logprob_std": 4.96319742944441, "tokenization_path_logprob_std": 264.8140656787576, "trajectory_balance_rmse": 4897.152835432877} +{"canonical_length": 11, "canonical_nll": 0.098567626693032, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "1c94de058e5cb066", "reading_canonical_log_prob": -1.084243893623352, "reading_distances": [20, 22, 23, 12], "reading_log_probs": [-1.1013141870498657, -1.1181721687316895, -1.0726304054260254, -1.0798015594482422], "reading_logprob_std": 0.020751556834643865, "tokenization_path_logprob_std": 33.492329702237136, "trajectory_balance_rmse": 219.89177057181476} +{"canonical_length": 40, "canonical_nll": 0.5853424549102784, "generation_rouge_l_f1": 0.09154929577464789, "generation_token_f1": 0.15492957746478872, "prompt_id": "91069c7b26de61af", "reading_canonical_log_prob": -23.413698196411133, "reading_distances": [45, 43, 46, 36], "reading_log_probs": [-21.100452423095703, -21.838090896606445, -20.85187530517578, -21.035167694091797], "reading_logprob_std": 0.43407623742356594, "tokenization_path_logprob_std": 70.71170390273464, "trajectory_balance_rmse": 710.5255395027569} +{"canonical_length": 10, "canonical_nll": 0.6863163471221924, "generation_rouge_l_f1": 0.823529411764706, "generation_token_f1": 0.823529411764706, "prompt_id": "c9ec0273bafaec5d", "reading_canonical_log_prob": -6.863163471221924, "reading_distances": [41, 38, 42, 49], "reading_log_probs": [-7.473551273345947, -6.877530574798584, -7.320590019226074, -8.052690505981445], "reading_logprob_std": 0.48539620646419696, "tokenization_path_logprob_std": 63.06685733539012, "trajectory_balance_rmse": 196.89938746949076} +{"canonical_length": 12, "canonical_nll": 0.11220042904218037, "generation_rouge_l_f1": 0.9565217391304348, "generation_token_f1": 0.9565217391304348, "prompt_id": "1cb768d05dd2b1b9", "reading_canonical_log_prob": -1.3464051485061646, "reading_distances": [41, 43, 37, 40], "reading_log_probs": [-1.7278848886489868, -1.6121056079864502, -1.675844669342041, -1.6835076808929443], "reading_logprob_std": 0.04769871463007656, "tokenization_path_logprob_std": 36.2758831927308, "trajectory_balance_rmse": 187.42431490477188} +{"canonical_length": 117, "canonical_nll": 0.9043556278587407, "generation_rouge_l_f1": 0.3076923076923077, "generation_token_f1": 0.423076923076923, "prompt_id": "31716783a47d0f33", "reading_canonical_log_prob": -105.80960845947266, "reading_distances": [29, 30, 29, 26], "reading_log_probs": [-105.16139221191406, -105.8160400390625, -105.49168395996094, -104.9051513671875], "reading_logprob_std": 0.39604937696265485, "tokenization_path_logprob_std": 103.20812571274101, "trajectory_balance_rmse": 2326.4652353768456} +{"canonical_length": 38, "canonical_nll": 0.32818811818173055, "generation_rouge_l_f1": 0.5306122448979592, "generation_token_f1": 0.5714285714285714, "prompt_id": "64444cea846a2353", "reading_canonical_log_prob": -12.471148490905762, "reading_distances": [621, 595, 616, 591], "reading_log_probs": [-21.859291076660156, -21.765411376953125, -22.021116256713867, -23.531333923339844], "reading_logprob_std": 0.8314321426515029, "tokenization_path_logprob_std": 51.654523449411414, "trajectory_balance_rmse": 917.1408640458503} +{"canonical_length": 17, "canonical_nll": 1.254073647891774, "generation_rouge_l_f1": 0.45161290322580644, "generation_token_f1": 0.45161290322580644, "prompt_id": "bcb6a50b95a69177", "reading_canonical_log_prob": -21.319252014160156, "reading_distances": [31, 30, 30, 36], "reading_log_probs": [-22.375600814819336, -23.090742111206055, -22.53647232055664, -22.60173988342285], "reading_logprob_std": 0.3080913863167207, "tokenization_path_logprob_std": 49.27562741354721, "trajectory_balance_rmse": 441.35829565389304} +{"canonical_length": 47, "canonical_nll": 0.77936497140438, "generation_rouge_l_f1": 0.2613065326633166, "generation_token_f1": 0.29145728643216084, "prompt_id": "5a3dfb14391da9f8", "reading_canonical_log_prob": -36.63015365600586, "reading_distances": [25, 28, 24, 23], "reading_log_probs": [-36.00477981567383, -35.303466796875, -34.85393524169922, -35.884498596191406], "reading_logprob_std": 0.5348277101149422, "tokenization_path_logprob_std": 82.13919866813465, "trajectory_balance_rmse": 799.4825092828178} +{"canonical_length": 255, "canonical_nll": 0.16970751893286612, "generation_rouge_l_f1": 0.935201401050788, "generation_token_f1": 0.956217162872154, "prompt_id": "1b64247b2a4cbf46", "reading_canonical_log_prob": -43.27541732788086, "reading_distances": [40, 50, 41, 46], "reading_log_probs": [-46.16736602783203, -45.75627136230469, -44.360755920410156, -45.23506164550781], "reading_logprob_std": 0.779185674489344, "tokenization_path_logprob_std": 104.34757625064248, "trajectory_balance_rmse": 3187.4048477002143} +{"canonical_length": 8, "canonical_nll": 0.069785937666893, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "ffe7a66de245047f", "reading_canonical_log_prob": -0.558287501335144, "reading_distances": [23, 21, 28, 26], "reading_log_probs": [-0.9766830205917358, -0.7899319529533386, -0.8189982175827026, -0.9100141525268555], "reading_logprob_std": 0.08550520410127467, "tokenization_path_logprob_std": 23.946105170089634, "trajectory_balance_rmse": 168.86303609775885} +{"canonical_length": 195, "canonical_nll": 1.436400115184295, "generation_rouge_l_f1": 0.27388535031847133, "generation_token_f1": 0.4140127388535032, "prompt_id": "9b0fbce06c332eb0", "reading_canonical_log_prob": -280.0980224609375, "reading_distances": [22, 24, 21, 17], "reading_log_probs": [-277.07293701171875, -279.20172119140625, -277.4855651855469, -278.44775390625], "reading_logprob_std": 0.9588000671322842, "tokenization_path_logprob_std": 284.0967775027297, "trajectory_balance_rmse": 4766.188158440812} +{"canonical_length": 103, "canonical_nll": 1.5972269298960862, "generation_rouge_l_f1": 0.28054298642533937, "generation_token_f1": 0.5158371040723981, "prompt_id": "f63c128779f1525d", "reading_canonical_log_prob": -164.51437377929688, "reading_distances": [33, 28, 31, 28], "reading_log_probs": [-166.35572814941406, -163.87901306152344, -166.43528747558594, -165.82498168945312], "reading_logprob_std": 1.1942900628682915, "tokenization_path_logprob_std": 137.8156164279004, "trajectory_balance_rmse": 2341.0230407571576} +{"canonical_length": 45, "canonical_nll": 0.7559434678819444, "generation_rouge_l_f1": 0.3669724770642202, "generation_token_f1": 0.49541284403669716, "prompt_id": "e0f5d30e662ee161", "reading_canonical_log_prob": -34.0174560546875, "reading_distances": [36, 27, 29, 30], "reading_log_probs": [-36.55665588378906, -37.499786376953125, -37.48533630371094, -37.373077392578125], "reading_logprob_std": 0.451603860697472, "tokenization_path_logprob_std": 29.428664752199676, "trajectory_balance_rmse": 1126.2671990641697} +{"canonical_length": 10, "canonical_nll": 2.4702875137329103, "generation_rouge_l_f1": 0.5714285714285714, "generation_token_f1": 0.5714285714285714, "prompt_id": "10787d0a825f897d", "reading_canonical_log_prob": -24.7028751373291, "reading_distances": [35, 37, 43, 32], "reading_log_probs": [-24.558271408081055, -25.926288604736328, -26.57447052001953, -23.93548583984375], "reading_logprob_std": 1.2135370945963884, "tokenization_path_logprob_std": 40.90144569339682, "trajectory_balance_rmse": 217.34239995024384} +{"canonical_length": 138, "canonical_nll": 0.89587518443232, "generation_rouge_l_f1": 0.2535211267605634, "generation_token_f1": 0.3755868544600939, "prompt_id": "55c070ce04e00179", "reading_canonical_log_prob": -123.63077545166016, "reading_distances": [20, 14, 18, 20], "reading_log_probs": [-125.46971130371094, -123.26112365722656, -128.6287841796875, -125.52986145019531], "reading_logprob_std": 2.2064959326393883, "tokenization_path_logprob_std": 151.2160365738142, "trajectory_balance_rmse": 3099.7920444200718} +{"canonical_length": 57, "canonical_nll": 1.5331985741330867, "generation_rouge_l_f1": 0.4418604651162791, "generation_token_f1": 0.4883720930232559, "prompt_id": "9718a6939e1171c6", "reading_canonical_log_prob": -87.39231872558594, "reading_distances": [19, 17, 25, 27], "reading_log_probs": [-86.302490234375, -86.89408874511719, -86.3526840209961, -86.18965911865234], "reading_logprob_std": 0.3137350750140931, "tokenization_path_logprob_std": 96.1236124358843, "trajectory_balance_rmse": 1443.4515658162156} +{"canonical_length": 180, "canonical_nll": 1.3714408026801215, "generation_rouge_l_f1": 0.19501133786848074, "generation_token_f1": 0.4126984126984127, "prompt_id": "abf26ab3207cd57f", "reading_canonical_log_prob": -246.85934448242188, "reading_distances": [23, 22, 24, 26], "reading_log_probs": [-248.34051513671875, -248.95875549316406, -250.0813446044922, -249.7906494140625], "reading_logprob_std": 0.79332398272356, "tokenization_path_logprob_std": 268.64227299123837, "trajectory_balance_rmse": 3233.2966565210945} +{"canonical_length": 8, "canonical_nll": 0.17075198888778687, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "9706792c2f395e2c", "reading_canonical_log_prob": -1.366015911102295, "reading_distances": [26, 27, 19, 24], "reading_log_probs": [-2.2515482902526855, -1.9843841791152954, -2.4293906688690186, -1.1222610473632812], "reading_logprob_std": 0.5793796372347368, "tokenization_path_logprob_std": 40.03438679833323, "trajectory_balance_rmse": 167.3898225479705} +{"canonical_length": 198, "canonical_nll": 1.2881031999684343, "generation_rouge_l_f1": 0.20689655172413793, "generation_token_f1": 0.40948275862068967, "prompt_id": "4d13949d8c9ff709", "reading_canonical_log_prob": -255.04443359375, "reading_distances": [31, 26, 23, 26], "reading_log_probs": [-256.00579833984375, -256.05914306640625, -253.68057250976562, -252.28648376464844], "reading_logprob_std": 1.850154526413372, "tokenization_path_logprob_std": 285.8262701318614, "trajectory_balance_rmse": 4887.013961449993} +{"canonical_length": 259, "canonical_nll": 1.412681078818774, "generation_rouge_l_f1": 0.22521008403361342, "generation_token_f1": 0.4638655462184874, "prompt_id": "cf2b1f1029bc9513", "reading_canonical_log_prob": -365.8843994140625, "reading_distances": [29, 30, 35, 24], "reading_log_probs": [-365.97052001953125, -370.8983154296875, -365.90728759765625, -365.89300537109375], "reading_logprob_std": 2.4875836007563223, "tokenization_path_logprob_std": 50.86104388470267, "trajectory_balance_rmse": 4515.232135305601} +{"canonical_length": 159, "canonical_nll": 1.780654619324882, "generation_rouge_l_f1": 0.20304568527918782, "generation_token_f1": 0.3908629441624366, "prompt_id": "0c1e711645beb310", "reading_canonical_log_prob": -283.12408447265625, "reading_distances": [20, 22, 27, 25], "reading_log_probs": [-283.57672119140625, -287.0618591308594, -286.1009826660156, -286.08843994140625], "reading_logprob_std": 1.491581916046994, "tokenization_path_logprob_std": 47.748079310136326, "trajectory_balance_rmse": 3174.5423285810507} +{"canonical_length": 36, "canonical_nll": 1.1251078711615667, "generation_rouge_l_f1": 0.35135135135135137, "generation_token_f1": 0.37837837837837834, "prompt_id": "602cd10603b86b5c", "reading_canonical_log_prob": -40.503883361816406, "reading_distances": [11, 13, 14, 14], "reading_log_probs": [-44.945499420166016, -45.88692092895508, -43.4472770690918, -43.25326919555664], "reading_logprob_std": 1.25566086432422, "tokenization_path_logprob_std": 74.51609489297493, "trajectory_balance_rmse": 754.047970881406} +{"canonical_length": 8, "canonical_nll": 2.755354404449463, "generation_rouge_l_f1": 0.36363636363636365, "generation_token_f1": 0.36363636363636365, "prompt_id": "b36657ba534be993", "reading_canonical_log_prob": -22.042835235595703, "reading_distances": [31, 35, 29, 34], "reading_log_probs": [-20.038070678710938, -21.012142181396484, -21.58926010131836, -21.9508113861084], "reading_logprob_std": 0.8345889822636919, "tokenization_path_logprob_std": 11.343923695191366, "trajectory_balance_rmse": 139.0903578075428} +{"canonical_length": 33, "canonical_nll": 1.7704578746448865, "generation_rouge_l_f1": 0.32, "generation_token_f1": 0.36000000000000004, "prompt_id": "f11461f699664b03", "reading_canonical_log_prob": -58.42510986328125, "reading_distances": [21, 20, 20, 18], "reading_log_probs": [-61.60820007324219, -64.94236755371094, -62.60210037231445, -61.218605041503906], "reading_logprob_std": 1.671167554413203, "tokenization_path_logprob_std": 80.36510372494075, "trajectory_balance_rmse": 784.2990838239955} +{"canonical_length": 323, "canonical_nll": 1.3667251067264898, "generation_rouge_l_f1": 0.2847341337907376, "generation_token_f1": 0.5042881646655232, "prompt_id": "167bb9079aab9f14", "reading_canonical_log_prob": -441.45220947265625, "reading_distances": [19, 26, 28, 23], "reading_log_probs": [-440.52838134765625, -440.3110046386719, -443.7518005371094, -441.8071594238281], "reading_logprob_std": 1.5793495994647968, "tokenization_path_logprob_std": 233.0572381203494, "trajectory_balance_rmse": 8065.215917472543} +{"canonical_length": 330, "canonical_nll": 1.0874781753077651, "generation_rouge_l_f1": 0.1908127208480565, "generation_token_f1": 0.37455830388692585, "prompt_id": "e95dc10521ada79e", "reading_canonical_log_prob": -358.8677978515625, "reading_distances": [39, 43, 31, 31], "reading_log_probs": [-363.37872314453125, -371.1759338378906, -367.1161804199219, -367.7108154296875], "reading_logprob_std": 3.1934090880219537, "tokenization_path_logprob_std": 242.61679336265252, "trajectory_balance_rmse": 6305.803577293064} +{"canonical_length": 126, "canonical_nll": 1.2026793464781746, "generation_rouge_l_f1": 0.23076923076923075, "generation_token_f1": 0.41958041958041964, "prompt_id": "78584ee870db48f1", "reading_canonical_log_prob": -151.53759765625, "reading_distances": [32, 35, 34, 38], "reading_log_probs": [-153.5211181640625, -153.18496704101562, -154.1492919921875, -154.68836975097656], "reading_logprob_std": 0.6677588981568533, "tokenization_path_logprob_std": 173.24120700062704, "trajectory_balance_rmse": 2805.9588898999627} +{"canonical_length": 318, "canonical_nll": 1.3349340666764937, "generation_rouge_l_f1": 0.24489795918367344, "generation_token_f1": 0.5, "prompt_id": "16eff2d1d4178803", "reading_canonical_log_prob": -424.509033203125, "reading_distances": [17, 25, 25, 22], "reading_log_probs": [-422.3733215332031, -425.0767822265625, -424.5492248535156, -425.7083740234375], "reading_logprob_std": 1.44875469140371, "tokenization_path_logprob_std": 216.46548035828138, "trajectory_balance_rmse": 7777.29463520016} +{"canonical_length": 104, "canonical_nll": 1.174095447246845, "generation_rouge_l_f1": 0.20385674931129477, "generation_token_f1": 0.3911845730027548, "prompt_id": "22619d053543f675", "reading_canonical_log_prob": -122.10592651367188, "reading_distances": [43, 39, 43, 38], "reading_log_probs": [-117.75988006591797, -118.3377685546875, -118.22809600830078, -116.79200744628906], "reading_logprob_std": 0.7043741022821892, "tokenization_path_logprob_std": 113.90219139598622, "trajectory_balance_rmse": 2071.1375263813306} +{"canonical_length": 180, "canonical_nll": 1.0176872253417968, "generation_rouge_l_f1": 0.23129251700680276, "generation_token_f1": 0.36054421768707484, "prompt_id": "3c84043f06521168", "reading_canonical_log_prob": -183.18370056152344, "reading_distances": [20, 28, 25, 24], "reading_log_probs": [-183.01663208007812, -183.89999389648438, -182.97784423828125, -183.53256225585938], "reading_logprob_std": 0.4416913159775543, "tokenization_path_logprob_std": 77.48915019779417, "trajectory_balance_rmse": 3765.380889007504} +{"canonical_length": 238, "canonical_nll": 1.4238327411042542, "generation_rouge_l_f1": 0.21960784313725493, "generation_token_f1": 0.42745098039215684, "prompt_id": "4c550a924c0f2e5b", "reading_canonical_log_prob": -338.8721923828125, "reading_distances": [31, 22, 30, 26], "reading_log_probs": [-335.2757568359375, -333.6578369140625, -335.26446533203125, -336.59478759765625], "reading_logprob_std": 1.2018848407024767, "tokenization_path_logprob_std": 169.76830258294981, "trajectory_balance_rmse": 4635.8904194407005} +{"canonical_length": 72, "canonical_nll": 1.4149816301133897, "generation_rouge_l_f1": 0.16806722689075632, "generation_token_f1": 0.2857142857142857, "prompt_id": "9e6e21f46355104c", "reading_canonical_log_prob": -101.87867736816406, "reading_distances": [19, 13, 20, 21], "reading_log_probs": [-100.03520202636719, -99.69328308105469, -99.53297424316406, -99.33387756347656], "reading_logprob_std": 0.2965788331213685, "tokenization_path_logprob_std": 80.21486198156344, "trajectory_balance_rmse": 1438.166365701459} +{"canonical_length": 27, "canonical_nll": 1.7037982234248408, "generation_rouge_l_f1": 0.32653061224489793, "generation_token_f1": 0.5306122448979592, "prompt_id": "4a624dc7ffd3988d", "reading_canonical_log_prob": -46.0025520324707, "reading_distances": [30, 26, 25, 36], "reading_log_probs": [-48.47758102416992, -48.46235275268555, -48.80754470825195, -47.31887435913086], "reading_logprob_std": 0.6515717407250603, "tokenization_path_logprob_std": 81.34256177025483, "trajectory_balance_rmse": 717.0086148931798} +{"canonical_length": 7, "canonical_nll": 0.2504089730126517, "generation_rouge_l_f1": 0.9090909090909091, "generation_token_f1": 0.9090909090909091, "prompt_id": "0cd953773dcd678b", "reading_canonical_log_prob": -1.752862811088562, "reading_distances": [27, 29, 36, 31], "reading_log_probs": [-1.7004612684249878, -1.6535528898239136, -1.6649720668792725, -1.5657856464385986], "reading_logprob_std": 0.05720492052165601, "tokenization_path_logprob_std": 14.449995185302186, "trajectory_balance_rmse": 160.31294852824738} +{"canonical_length": 165, "canonical_nll": 0.5393556074662642, "generation_rouge_l_f1": 0.1907514450867052, "generation_token_f1": 0.5202312138728324, "prompt_id": "a86300073620913b", "reading_canonical_log_prob": -88.9936752319336, "reading_distances": [135, 135, 154, 137], "reading_log_probs": [-87.6500473022461, -88.56778717041016, -88.88835144042969, -86.88751220703125], "reading_logprob_std": 0.9076777336226545, "tokenization_path_logprob_std": 64.62613070874286, "trajectory_balance_rmse": 2900.827899563693} +{"canonical_length": 294, "canonical_nll": 1.3565314675674958, "generation_rouge_l_f1": 0.17739130434782607, "generation_token_f1": 0.3443478260869565, "prompt_id": "2837da3fee02c3ea", "reading_canonical_log_prob": -398.82025146484375, "reading_distances": [32, 24, 25, 27], "reading_log_probs": [-395.24407958984375, -396.30224609375, -397.3415222167969, -397.5747375488281], "reading_logprob_std": 1.0686700145275219, "tokenization_path_logprob_std": 237.73720601164106, "trajectory_balance_rmse": 6092.589653090549} +{"canonical_length": 23, "canonical_nll": 1.193168474280316, "generation_rouge_l_f1": 0.09285714285714285, "generation_token_f1": 0.1142857142857143, "prompt_id": "8041588a81157d9c", "reading_canonical_log_prob": -27.442874908447266, "reading_distances": [54, 53, 49, 37], "reading_log_probs": [-29.667095184326172, -26.736886978149414, -27.13154411315918, -28.975038528442383], "reading_logprob_std": 1.4159104861203868, "tokenization_path_logprob_std": 91.0822716421827, "trajectory_balance_rmse": 557.4249608901048} +{"canonical_length": 308, "canonical_nll": 0.8606423712396002, "generation_rouge_l_f1": 0.20512820512820512, "generation_token_f1": 0.38264299802761337, "prompt_id": "bc95295408e3badf", "reading_canonical_log_prob": -265.0778503417969, "reading_distances": [30, 32, 28, 32], "reading_log_probs": [-268.833251953125, -273.46826171875, -270.8211364746094, -266.35687255859375], "reading_logprob_std": 3.014910660105989, "tokenization_path_logprob_std": 153.38953309403792, "trajectory_balance_rmse": 5286.286509508412} +{"canonical_length": 126, "canonical_nll": 1.3404129270523313, "generation_rouge_l_f1": 0.32460732984293195, "generation_token_f1": 0.4450261780104712, "prompt_id": "2a5a99cace11afef", "reading_canonical_log_prob": -168.89202880859375, "reading_distances": [27, 28, 22, 22], "reading_log_probs": [-168.41983032226562, -168.76419067382812, -170.47789001464844, -169.79043579101562], "reading_logprob_std": 0.9440732175426616, "tokenization_path_logprob_std": 195.53680879731445, "trajectory_balance_rmse": 2925.373956311614} +{"canonical_length": 14, "canonical_nll": 1.2155554635184151, "generation_rouge_l_f1": 0.19607843137254902, "generation_token_f1": 0.2352941176470588, "prompt_id": "90db691d0627dd6d", "reading_canonical_log_prob": -17.017776489257812, "reading_distances": [16, 21, 16, 22], "reading_log_probs": [-17.83428192138672, -19.0533390045166, -18.668397903442383, -18.553308486938477], "reading_logprob_std": 0.5091090738581033, "tokenization_path_logprob_std": 12.528258032220489, "trajectory_balance_rmse": 350.0476405490206} +{"canonical_length": 14, "canonical_nll": 0.8770405905587333, "generation_rouge_l_f1": 0.45454545454545453, "generation_token_f1": 0.45454545454545453, "prompt_id": "07c0c6eb902aa919", "reading_canonical_log_prob": -12.278568267822266, "reading_distances": [26, 26, 23, 27], "reading_log_probs": [-11.785439491271973, -12.170975685119629, -12.04699993133545, -11.606565475463867], "reading_logprob_std": 0.25445355114980683, "tokenization_path_logprob_std": 43.68371307656633, "trajectory_balance_rmse": 310.7011481265211} +{"canonical_length": 6, "canonical_nll": 1.1541812419891357, "generation_rouge_l_f1": 0.6, "generation_token_f1": 0.6, "prompt_id": "285daef948799d4d", "reading_canonical_log_prob": -6.9250874519348145, "reading_distances": [43, 43, 52, 50], "reading_log_probs": [-7.470198631286621, -7.477071762084961, -8.087685585021973, -7.520806312561035], "reading_logprob_std": 0.3000017467417546, "tokenization_path_logprob_std": 25.39458177054891, "trajectory_balance_rmse": 135.76907827306366} +{"canonical_length": 68, "canonical_nll": 0.6738255444694968, "generation_rouge_l_f1": 0.1095890410958904, "generation_token_f1": 0.1095890410958904, "prompt_id": "924060b32d38a1e4", "reading_canonical_log_prob": -45.82013702392578, "reading_distances": [34, 30, 33, 36], "reading_log_probs": [-45.2783317565918, -45.49442672729492, -44.87141418457031, -44.52779769897461], "reading_logprob_std": 0.42974497421757407, "tokenization_path_logprob_std": 76.39309763795741, "trajectory_balance_rmse": 1255.1142243392474} +{"canonical_length": 147, "canonical_nll": 1.2292564547791773, "generation_rouge_l_f1": 0.3493150684931507, "generation_token_f1": 0.5684931506849313, "prompt_id": "ac4b8108300a4ae5", "reading_canonical_log_prob": -180.70069885253906, "reading_distances": [32, 32, 30, 29], "reading_log_probs": [-176.96017456054688, -177.1175537109375, -177.89218139648438, -177.16622924804688], "reading_logprob_std": 0.4148581438679014, "tokenization_path_logprob_std": 78.71403305248703, "trajectory_balance_rmse": 2376.088327129754} +{"canonical_length": 53, "canonical_nll": 1.6621003060970667, "generation_rouge_l_f1": 0.19791666666666666, "generation_token_f1": 0.3229166666666667, "prompt_id": "10e2361fe555104f", "reading_canonical_log_prob": -88.09131622314453, "reading_distances": [25, 24, 18, 28], "reading_log_probs": [-93.5503158569336, -92.50089263916016, -89.18696594238281, -89.21859741210938], "reading_logprob_std": 2.248341405552354, "tokenization_path_logprob_std": 88.05896785064185, "trajectory_balance_rmse": 1102.5697151747993} +{"canonical_length": 27, "canonical_nll": 1.375862545437283, "generation_rouge_l_f1": 0.18446601941747576, "generation_token_f1": 0.1941747572815534, "prompt_id": "e53f2608a52ca6d9", "reading_canonical_log_prob": -37.14828872680664, "reading_distances": [24, 33, 32, 36], "reading_log_probs": [-40.07561492919922, -38.309844970703125, -38.167213439941406, -39.07657241821289], "reading_logprob_std": 0.8752701589119893, "tokenization_path_logprob_std": 36.92614355442217, "trajectory_balance_rmse": 439.9870366900285} +{"canonical_length": 18, "canonical_nll": 2.449670579698351, "generation_rouge_l_f1": 0.0746268656716418, "generation_token_f1": 0.08208955223880599, "prompt_id": "3f4c8b5b47d9a555", "reading_canonical_log_prob": -44.09407043457031, "reading_distances": [17, 19, 15, 16], "reading_log_probs": [-47.11442947387695, -45.8165168762207, -47.4915657043457, -45.02678680419922], "reading_logprob_std": 1.1434471986647605, "tokenization_path_logprob_std": 52.69664985837098, "trajectory_balance_rmse": 432.4273663368634} +{"canonical_length": 11, "canonical_nll": 0.9052762118252841, "generation_rouge_l_f1": 0.5454545454545455, "generation_token_f1": 0.5454545454545455, "prompt_id": "94a2d875be1c7a06", "reading_canonical_log_prob": -9.958038330078125, "reading_distances": [17, 16, 13, 17], "reading_log_probs": [-10.878299713134766, -9.9348783493042, -9.994353294372559, -10.065743446350098], "reading_logprob_std": 0.4432279628081343, "tokenization_path_logprob_std": 35.83256116277635, "trajectory_balance_rmse": 221.62406579006145} +{"canonical_length": 18, "canonical_nll": 0.721736854977078, "generation_rouge_l_f1": 0.30952380952380953, "generation_token_f1": 0.30952380952380953, "prompt_id": "a86c032862f0a2ee", "reading_canonical_log_prob": -12.991263389587402, "reading_distances": [17, 17, 15, 14], "reading_log_probs": [-12.125428199768066, -12.412904739379883, -12.76806640625, -12.132640838623047], "reading_logprob_std": 0.30333314511333714, "tokenization_path_logprob_std": 41.04777739784057, "trajectory_balance_rmse": 314.22770917031386} +{"canonical_length": 80, "canonical_nll": 1.3083341598510743, "generation_rouge_l_f1": 0.16519174041297935, "generation_token_f1": 0.26548672566371684, "prompt_id": "26b8508f97bd840b", "reading_canonical_log_prob": -104.66673278808594, "reading_distances": [33, 29, 38, 28], "reading_log_probs": [-102.8161392211914, -105.36550903320312, -104.288818359375, -104.41815948486328], "reading_logprob_std": 1.0530907806802114, "tokenization_path_logprob_std": 79.70278477840752, "trajectory_balance_rmse": 2020.8202088679432} +{"canonical_length": 282, "canonical_nll": 0.8796754120089484, "generation_rouge_l_f1": 0.32443531827515404, "generation_token_f1": 0.4681724845995893, "prompt_id": "d2afb79833a3e8c4", "reading_canonical_log_prob": -248.06846618652344, "reading_distances": [21, 26, 22, 22], "reading_log_probs": [-248.14703369140625, -250.5404815673828, -249.2048797607422, -248.00851440429688], "reading_logprob_std": 1.1723480214679276, "tokenization_path_logprob_std": 105.34920359568329, "trajectory_balance_rmse": 6242.491895984638} +{"canonical_length": 206, "canonical_nll": 1.005771229568037, "generation_rouge_l_f1": 0.31034482758620685, "generation_token_f1": 0.5603448275862069, "prompt_id": "8d6230df8e48bdfc", "reading_canonical_log_prob": -207.18887329101562, "reading_distances": [14, 14, 12, 17], "reading_log_probs": [-202.2585906982422, -203.79568481445312, -204.04254150390625, -204.078125], "reading_logprob_std": 0.8659205478914138, "tokenization_path_logprob_std": 88.50954520603979, "trajectory_balance_rmse": 3997.0419181742764} +{"canonical_length": 14, "canonical_nll": 3.3043929508754184, "generation_rouge_l_f1": 0.1764705882352941, "generation_token_f1": 0.1764705882352941, "prompt_id": "74ebf94d0f403056", "reading_canonical_log_prob": -46.26150131225586, "reading_distances": [22, 24, 30, 26], "reading_log_probs": [-45.3068733215332, -45.600433349609375, -44.88698196411133, -44.76463317871094], "reading_logprob_std": 0.3850383795850172, "tokenization_path_logprob_std": 15.056692169496744, "trajectory_balance_rmse": 286.79773471314314} +{"canonical_length": 129, "canonical_nll": 1.0986180268516836, "generation_rouge_l_f1": 0.3244680851063829, "generation_token_f1": 0.48404255319148937, "prompt_id": "c307d4b735027f71", "reading_canonical_log_prob": -141.7217254638672, "reading_distances": [18, 17, 16, 15], "reading_log_probs": [-141.86656188964844, -141.8678436279297, -141.06036376953125, -140.81048583984375], "reading_logprob_std": 0.5475492267211111, "tokenization_path_logprob_std": 201.90835126594166, "trajectory_balance_rmse": 2666.6282060096887} +{"canonical_length": 73, "canonical_nll": 0.949903618799497, "generation_rouge_l_f1": 0.32394366197183094, "generation_token_f1": 0.5492957746478873, "prompt_id": "1044800b3405ca7b", "reading_canonical_log_prob": -69.34296417236328, "reading_distances": [12, 13, 13, 12], "reading_log_probs": [-69.95616912841797, -70.51248931884766, -68.94477844238281, -70.48056030273438], "reading_logprob_std": 0.7317076006746345, "tokenization_path_logprob_std": 90.6805308045444, "trajectory_balance_rmse": 1480.364660953609} +{"canonical_length": 205, "canonical_nll": 1.4151571134241616, "generation_rouge_l_f1": 0.2141327623126338, "generation_token_f1": 0.4668094218415418, "prompt_id": "c87b3b9596ed8821", "reading_canonical_log_prob": -290.1072082519531, "reading_distances": [30, 29, 32, 27], "reading_log_probs": [-288.3570556640625, -290.5868225097656, -288.9515380859375, -288.2189636230469], "reading_logprob_std": 1.0863515064124203, "tokenization_path_logprob_std": 60.72404974427151, "trajectory_balance_rmse": 4670.566185412497} +{"canonical_length": 11, "canonical_nll": 1.430615251714533, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "f2f04e5d3550740c", "reading_canonical_log_prob": -15.736767768859863, "reading_distances": [19, 11, 13, 17], "reading_log_probs": [-18.706817626953125, -18.16828155517578, -18.854148864746094, -18.518224716186523], "reading_logprob_std": 0.29622824759624145, "tokenization_path_logprob_std": 55.330551776837844, "trajectory_balance_rmse": 217.93437318194475} +{"canonical_length": 149, "canonical_nll": 1.4566976048002307, "generation_rouge_l_f1": 0.19900497512437812, "generation_token_f1": 0.31343283582089554, "prompt_id": "e64143ec52b6e7c2", "reading_canonical_log_prob": -217.04794311523438, "reading_distances": [20, 18, 16, 15], "reading_log_probs": [-216.95571899414062, -216.03903198242188, -216.98538208007812, -218.04470825195312], "reading_logprob_std": 0.8199380222343198, "tokenization_path_logprob_std": 89.14400763795483, "trajectory_balance_rmse": 2961.0166815189195} +{"canonical_length": 48, "canonical_nll": 1.2514936129252117, "generation_rouge_l_f1": 0.2782608695652174, "generation_token_f1": 0.33043478260869563, "prompt_id": "c95d33d4bb11096d", "reading_canonical_log_prob": -60.071693420410156, "reading_distances": [18, 26, 25, 24], "reading_log_probs": [-60.93735885620117, -61.04751205444336, -61.71349334716797, -61.474700927734375], "reading_logprob_std": 0.3635815145067889, "tokenization_path_logprob_std": 139.5232990418915, "trajectory_balance_rmse": 1282.7748539684455} +{"canonical_length": 211, "canonical_nll": 1.271536732171949, "generation_rouge_l_f1": 0.22705314009661837, "generation_token_f1": 0.46859903381642515, "prompt_id": "0accd9d43b5400d9", "reading_canonical_log_prob": -268.29425048828125, "reading_distances": [23, 20, 20, 30], "reading_log_probs": [-267.0465087890625, -266.84326171875, -266.0078430175781, -266.6127624511719], "reading_logprob_std": 0.44956031128931156, "tokenization_path_logprob_std": 151.67367915624274, "trajectory_balance_rmse": 4536.128433712007} +{"canonical_length": 28, "canonical_nll": 0.9094762120928083, "generation_rouge_l_f1": 0.11143695014662756, "generation_token_f1": 0.11730205278592376, "prompt_id": "c5287ae056861247", "reading_canonical_log_prob": -25.465333938598633, "reading_distances": [19, 16, 23, 19], "reading_log_probs": [-23.28387451171875, -22.500391006469727, -22.483531951904297, -22.989429473876953], "reading_logprob_std": 0.39120195651992723, "tokenization_path_logprob_std": 31.79627805047168, "trajectory_balance_rmse": 618.4667219060349} +{"canonical_length": 9, "canonical_nll": 0.9110553529527452, "generation_rouge_l_f1": 0.06097560975609756, "generation_token_f1": 0.08536585365853658, "prompt_id": "9ba45d22569ee03c", "reading_canonical_log_prob": -8.199498176574707, "reading_distances": [41, 44, 43, 46], "reading_log_probs": [-8.834897994995117, -7.946135520935059, -7.188365936279297, -9.193718910217285], "reading_logprob_std": 0.9028583596741248, "tokenization_path_logprob_std": 38.160095874743234, "trajectory_balance_rmse": 239.79065629808412} +{"canonical_length": 9, "canonical_nll": 1.2486481136745877, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.375, "prompt_id": "61752b6441afd01d", "reading_canonical_log_prob": -11.237833023071289, "reading_distances": [20, 20, 16, 23], "reading_log_probs": [-11.987711906433105, -12.115636825561523, -12.156991004943848, -11.889286041259766], "reading_logprob_std": 0.12224246478114861, "tokenization_path_logprob_std": 42.43802400129487, "trajectory_balance_rmse": 246.23859307213309} +{"canonical_length": 147, "canonical_nll": 1.1298968256736288, "generation_rouge_l_f1": 0.23280423280423282, "generation_token_f1": 0.4603174603174603, "prompt_id": "1c1ecaad77fec061", "reading_canonical_log_prob": -166.09483337402344, "reading_distances": [14, 14, 17, 16], "reading_log_probs": [-166.48194885253906, -165.58865356445312, -169.22164916992188, -167.42623901367188], "reading_logprob_std": 1.554415570530653, "tokenization_path_logprob_std": 109.64029681993063, "trajectory_balance_rmse": 2855.8040164086933} +{"canonical_length": 58, "canonical_nll": 1.1579568797144397, "generation_rouge_l_f1": 0.5441176470588235, "generation_token_f1": 0.6617647058823529, "prompt_id": "697abdbcde745481", "reading_canonical_log_prob": -67.1614990234375, "reading_distances": [21, 25, 17, 22], "reading_log_probs": [-63.46318435668945, -66.9068603515625, -64.90226745605469, -64.31074523925781], "reading_logprob_std": 1.4650463263501556, "tokenization_path_logprob_std": 42.878926020022256, "trajectory_balance_rmse": 1065.4632251626074} +{"canonical_length": 338, "canonical_nll": 1.2910331410063793, "generation_rouge_l_f1": 0.12195121951219512, "generation_token_f1": 0.2146341463414634, "prompt_id": "4034832cbc1502c1", "reading_canonical_log_prob": -436.36920166015625, "reading_distances": [29, 20, 22, 17], "reading_log_probs": [-435.84808349609375, -436.6382141113281, -434.4914855957031, -436.57269287109375], "reading_logprob_std": 0.9972413229617924, "tokenization_path_logprob_std": 131.85641977875724, "trajectory_balance_rmse": 7315.677491748336} +{"canonical_length": 13, "canonical_nll": 1.5862811161921575, "generation_rouge_l_f1": 0.10389610389610389, "generation_token_f1": 0.11688311688311691, "prompt_id": "228b1bd7f7bc8ed3", "reading_canonical_log_prob": -20.621654510498047, "reading_distances": [14, 16, 14, 15], "reading_log_probs": [-16.859760284423828, -16.23470687866211, -17.03447723388672, -16.576385498046875], "reading_logprob_std": 0.34973132312815797, "tokenization_path_logprob_std": 31.49242015588562, "trajectory_balance_rmse": 234.73094730994995} +{"canonical_length": 115, "canonical_nll": 1.2281882908033288, "generation_rouge_l_f1": 0.32592592592592595, "generation_token_f1": 0.5333333333333333, "prompt_id": "28e1d60f144a6d02", "reading_canonical_log_prob": -141.2416534423828, "reading_distances": [20, 24, 27, 23], "reading_log_probs": [-138.4510498046875, -137.54290771484375, -137.15463256835938, -136.77438354492188], "reading_logprob_std": 0.7189463216858648, "tokenization_path_logprob_std": 69.2735526401691, "trajectory_balance_rmse": 2295.1265647181503} +{"canonical_length": 131, "canonical_nll": 1.3812192960549856, "generation_rouge_l_f1": 0.3082191780821918, "generation_token_f1": 0.48630136986301375, "prompt_id": "f38548affd15660d", "reading_canonical_log_prob": -180.93972778320312, "reading_distances": [27, 26, 30, 26], "reading_log_probs": [-179.75106811523438, -178.9605712890625, -179.1868133544922, -178.89700317382812], "reading_logprob_std": 0.3885777388554499, "tokenization_path_logprob_std": 153.37060127418565, "trajectory_balance_rmse": 2832.769560441689} +{"canonical_length": 39, "canonical_nll": 1.2911787766676683, "generation_rouge_l_f1": 0.17283950617283952, "generation_token_f1": 0.28395061728395066, "prompt_id": "b15afd8e011cefe5", "reading_canonical_log_prob": -50.35597229003906, "reading_distances": [20, 19, 30, 25], "reading_log_probs": [-50.871280670166016, -50.35527420043945, -50.370445251464844, -50.452392578125], "reading_logprob_std": 0.2430609545462558, "tokenization_path_logprob_std": 100.13666142489387, "trajectory_balance_rmse": 824.0685447862461} +{"canonical_length": 286, "canonical_nll": 0.9919779317362325, "generation_rouge_l_f1": 0.139917695473251, "generation_token_f1": 0.2674897119341564, "prompt_id": "60d1476fd22fd625", "reading_canonical_log_prob": -283.7056884765625, "reading_distances": [17, 24, 20, 22], "reading_log_probs": [-281.5307312011719, -285.0550842285156, -277.8558349609375, -280.7249450683594], "reading_logprob_std": 2.9634763637382298, "tokenization_path_logprob_std": 84.34472939345592, "trajectory_balance_rmse": 4877.927471071142} +{"canonical_length": 58, "canonical_nll": 1.126910505623653, "generation_rouge_l_f1": 0.1507537688442211, "generation_token_f1": 0.2512562814070351, "prompt_id": "f1aff688c2307612", "reading_canonical_log_prob": -65.36080932617188, "reading_distances": [35, 51, 43, 44], "reading_log_probs": [-65.05221557617188, -66.46353912353516, -65.98287963867188, -65.66181945800781], "reading_logprob_std": 0.5920623733618834, "tokenization_path_logprob_std": 80.85108431786107, "trajectory_balance_rmse": 1308.106196886667} +{"canonical_length": 123, "canonical_nll": 1.3906835540523375, "generation_rouge_l_f1": 0.18652849740932642, "generation_token_f1": 0.33160621761658027, "prompt_id": "a3f93481aca6425b", "reading_canonical_log_prob": -171.0540771484375, "reading_distances": [24, 22, 23, 28], "reading_log_probs": [-170.0248565673828, -169.49559020996094, -169.7301025390625, -169.62803649902344], "reading_logprob_std": 0.22498601983951746, "tokenization_path_logprob_std": 162.84349984944248, "trajectory_balance_rmse": 2526.407402389063} +{"canonical_length": 163, "canonical_nll": 1.204262201040069, "generation_rouge_l_f1": 0.19672131147540983, "generation_token_f1": 0.4168618266978923, "prompt_id": "39b8bde022351616", "reading_canonical_log_prob": -196.29473876953125, "reading_distances": [17, 22, 26, 16], "reading_log_probs": [-194.41812133789062, -196.49464416503906, -195.83523559570312, -194.7285919189453], "reading_logprob_std": 0.9658749176100627, "tokenization_path_logprob_std": 148.8500731063716, "trajectory_balance_rmse": 3541.2657050933367} +{"canonical_length": 56, "canonical_nll": 1.3476896286010742, "generation_rouge_l_f1": 0.12734082397003746, "generation_token_f1": 0.18726591760299624, "prompt_id": "969cc5c017329874", "reading_canonical_log_prob": -75.47061920166016, "reading_distances": [32, 31, 33, 33], "reading_log_probs": [-79.53971862792969, -81.77262878417969, -76.61058044433594, -78.61438751220703], "reading_logprob_std": 2.14200937008379, "tokenization_path_logprob_std": 89.62559447502578, "trajectory_balance_rmse": 925.2015328323271} +{"canonical_length": 262, "canonical_nll": 1.5763626681029341, "generation_rouge_l_f1": 0.36293436293436293, "generation_token_f1": 0.5598455598455598, "prompt_id": "a4cec9116b07e32b", "reading_canonical_log_prob": -413.00701904296875, "reading_distances": [12, 20, 13, 16], "reading_log_probs": [-407.5838317871094, -409.38397216796875, -406.8766784667969, -409.852294921875], "reading_logprob_std": 1.4214616126987747, "tokenization_path_logprob_std": 116.45164742746518, "trajectory_balance_rmse": 5106.207300396033} +{"canonical_length": 412, "canonical_nll": 0.8832561900314776, "generation_rouge_l_f1": 0.19061583577712612, "generation_token_f1": 0.3958944281524927, "prompt_id": "63a1199c25b071e9", "reading_canonical_log_prob": -363.90155029296875, "reading_distances": [209, 221, 217, 245], "reading_log_probs": [-367.697998046875, -363.72967529296875, -363.5972595214844, -364.66595458984375], "reading_logprob_std": 1.9103488221079148, "tokenization_path_logprob_std": 142.01864532170038, "trajectory_balance_rmse": 7281.710735602679} +{"canonical_length": 34, "canonical_nll": 1.5818245831657858, "generation_rouge_l_f1": 0.10862619808306709, "generation_token_f1": 0.13418530351437702, "prompt_id": "85280303ae668b99", "reading_canonical_log_prob": -53.78203582763672, "reading_distances": [10, 14, 10, 15], "reading_log_probs": [-52.22100830078125, -54.03550720214844, -53.549774169921875, -53.73322677612305], "reading_logprob_std": 0.8013423430145686, "tokenization_path_logprob_std": 48.31622955388801, "trajectory_balance_rmse": 741.1528250489625} +{"canonical_length": 18, "canonical_nll": 0.9876987669203017, "generation_rouge_l_f1": 0.4000000000000001, "generation_token_f1": 0.4000000000000001, "prompt_id": "963484b861b43113", "reading_canonical_log_prob": -17.77857780456543, "reading_distances": [15, 15, 17, 12], "reading_log_probs": [-17.153661727905273, -17.90960693359375, -17.300935745239258, -18.18850326538086], "reading_logprob_std": 0.49160279359945036, "tokenization_path_logprob_std": 23.77180451627364, "trajectory_balance_rmse": 346.85340174438} +{"canonical_length": 242, "canonical_nll": 1.1490361237328899, "generation_rouge_l_f1": 0.2173913043478261, "generation_token_f1": 0.5296442687747036, "prompt_id": "edfd3e88db49ba9c", "reading_canonical_log_prob": -278.0667419433594, "reading_distances": [24, 28, 23, 31], "reading_log_probs": [-281.64215087890625, -279.7629699707031, -279.8834228515625, -283.63165283203125], "reading_logprob_std": 1.8168885295548038, "tokenization_path_logprob_std": 129.99817730163815, "trajectory_balance_rmse": 5588.846403555272} +{"canonical_length": 3, "canonical_nll": 4.626187960306804, "generation_rouge_l_f1": 0.015936254980079678, "generation_token_f1": 0.015936254980079678, "prompt_id": "315bed08881fe880", "reading_canonical_log_prob": -13.87856388092041, "reading_distances": [31, 32, 31, 29], "reading_log_probs": [-14.929605484008789, -14.479272842407227, -14.303628921508789, -14.569074630737305], "reading_logprob_std": 0.2636293505624119, "tokenization_path_logprob_std": 21.08613875914061, "trajectory_balance_rmse": 85.73249222983634} +{"canonical_length": 40, "canonical_nll": 1.3332687377929688, "generation_rouge_l_f1": 0.24242424242424246, "generation_token_f1": 0.33333333333333337, "prompt_id": "ec3686b4d6045b07", "reading_canonical_log_prob": -53.33074951171875, "reading_distances": [20, 30, 28, 25], "reading_log_probs": [-54.38011932373047, -54.102996826171875, -54.14835739135742, -54.309661865234375], "reading_logprob_std": 0.13110488102581203, "tokenization_path_logprob_std": 119.78296176343363, "trajectory_balance_rmse": 1075.7424119061902} +{"canonical_length": 266, "canonical_nll": 1.4310812125528665, "generation_rouge_l_f1": 0.19548872180451124, "generation_token_f1": 0.3947368421052631, "prompt_id": "a1b8dba044d5aaa9", "reading_canonical_log_prob": -380.6676025390625, "reading_distances": [26, 27, 19, 23], "reading_log_probs": [-379.2150573730469, -380.3525390625, -380.87860107421875, -379.32147216796875], "reading_logprob_std": 0.808138651469084, "tokenization_path_logprob_std": 147.0374520818318, "trajectory_balance_rmse": 6175.790014265214} +{"canonical_length": 154, "canonical_nll": 1.3431005106343852, "generation_rouge_l_f1": 0.24401913875598086, "generation_token_f1": 0.4258373205741627, "prompt_id": "e45ec8d7efaf0407", "reading_canonical_log_prob": -206.8374786376953, "reading_distances": [22, 15, 16, 15], "reading_log_probs": [-213.3935546875, -215.94607543945312, -215.25074768066406, -212.1969451904297], "reading_logprob_std": 1.714194909598836, "tokenization_path_logprob_std": 67.2977819512275, "trajectory_balance_rmse": 3825.9172397835555} +{"canonical_length": 3, "canonical_nll": 2.6051510175069175, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "3521244aca15fc91", "reading_canonical_log_prob": -7.815453052520752, "reading_distances": [49, 49, 46, 48], "reading_log_probs": [-7.52752161026001, -7.763688087463379, -8.03310775756836, -7.718008518218994], "reading_logprob_std": 0.20849427085112393, "tokenization_path_logprob_std": 17.977430148081922, "trajectory_balance_rmse": 68.48108538876718} +{"canonical_length": 23, "canonical_nll": 0.7814996968144956, "generation_rouge_l_f1": 0.1111111111111111, "generation_token_f1": 0.15151515151515152, "prompt_id": "99de541f6c007900", "reading_canonical_log_prob": -17.9744930267334, "reading_distances": [26, 23, 26, 26], "reading_log_probs": [-17.517166137695312, -16.86669158935547, -17.934696197509766, -18.163185119628906], "reading_logprob_std": 0.5692484459440039, "tokenization_path_logprob_std": 59.776463336463806, "trajectory_balance_rmse": 278.36795371838826} +{"canonical_length": 176, "canonical_nll": 1.2595517418601296, "generation_rouge_l_f1": 0.22072072072072071, "generation_token_f1": 0.42342342342342343, "prompt_id": "124ab50eaad8157c", "reading_canonical_log_prob": -221.6811065673828, "reading_distances": [18, 19, 23, 25], "reading_log_probs": [-222.6846160888672, -223.06390380859375, -223.0381317138672, -223.9207763671875], "reading_logprob_std": 0.5252682704837593, "tokenization_path_logprob_std": 228.42163901348707, "trajectory_balance_rmse": 4178.874999978802} +{"canonical_length": 15, "canonical_nll": 0.4526746114095052, "generation_rouge_l_f1": 0.8888888888888888, "generation_token_f1": 0.8888888888888888, "prompt_id": "7efa869db692612b", "reading_canonical_log_prob": -6.790119171142578, "reading_distances": [48, 52, 54, 51], "reading_log_probs": [-13.168814659118652, -11.27535629272461, -15.58546257019043, -15.0487060546875], "reading_logprob_std": 1.9592229640978311, "tokenization_path_logprob_std": 41.932847085275, "trajectory_balance_rmse": 231.4068582165871} +{"canonical_length": 248, "canonical_nll": 1.4153881688271799, "generation_rouge_l_f1": 0.22574257425742572, "generation_token_f1": 0.4673267326732673, "prompt_id": "b9153e23659ffc43", "reading_canonical_log_prob": -351.0162658691406, "reading_distances": [30, 27, 29, 36], "reading_log_probs": [-350.09307861328125, -348.77593994140625, -348.8460693359375, -349.1169128417969], "reading_logprob_std": 0.6080928443086095, "tokenization_path_logprob_std": 176.5322170655385, "trajectory_balance_rmse": 6047.262971223007} +{"canonical_length": 48, "canonical_nll": 0.9265921910603842, "generation_rouge_l_f1": 0.2119205298013245, "generation_token_f1": 0.24503311258278146, "prompt_id": "2f3de773aa866413", "reading_canonical_log_prob": -44.47642517089844, "reading_distances": [23, 33, 25, 29], "reading_log_probs": [-43.22085189819336, -47.06412887573242, -45.89308166503906, -46.01605987548828], "reading_logprob_std": 1.638335062607309, "tokenization_path_logprob_std": 83.70972068511624, "trajectory_balance_rmse": 846.9067333547579} +{"canonical_length": 301, "canonical_nll": 1.2940437595709613, "generation_rouge_l_f1": 0.2101449275362319, "generation_token_f1": 0.38405797101449274, "prompt_id": "54066b815b8806cf", "reading_canonical_log_prob": -389.5071716308594, "reading_distances": [32, 37, 45, 40], "reading_log_probs": [-389.64874267578125, -389.520751953125, -391.19097900390625, -390.96124267578125], "reading_logprob_std": 0.8677068183126462, "tokenization_path_logprob_std": 288.5012978374656, "trajectory_balance_rmse": 6139.977736355498} +{"canonical_length": 394, "canonical_nll": 1.2946654189056552, "generation_rouge_l_f1": 0.20972644376899693, "generation_token_f1": 0.3829787234042553, "prompt_id": "38de2822dc803e85", "reading_canonical_log_prob": -510.0981750488281, "reading_distances": [26, 31, 38, 36], "reading_log_probs": [-507.09130859375, -509.4534912109375, -508.6618957519531, -508.228759765625], "reading_logprob_std": 0.9855050808819443, "tokenization_path_logprob_std": 306.24385670163934, "trajectory_balance_rmse": 7850.8074572679925} +{"canonical_length": 329, "canonical_nll": 0.939455351206307, "generation_rouge_l_f1": 0.18815331010452963, "generation_token_f1": 0.4006968641114982, "prompt_id": "93d5783f4b9e5595", "reading_canonical_log_prob": -309.080810546875, "reading_distances": [32, 31, 30, 32], "reading_log_probs": [-306.5494689941406, -308.0525817871094, -306.4867858886719, -308.86273193359375], "reading_logprob_std": 1.1678913452221305, "tokenization_path_logprob_std": 279.0134328714143, "trajectory_balance_rmse": 7716.872054788229} +{"canonical_length": 25, "canonical_nll": 1.0111183166503905, "generation_rouge_l_f1": 0.04964539007092199, "generation_token_f1": 0.08510638297872342, "prompt_id": "fffceba44eca0753", "reading_canonical_log_prob": -25.277957916259766, "reading_distances": [27, 31, 30, 28], "reading_log_probs": [-23.71304702758789, -22.800857543945312, -22.7327880859375, -23.700620651245117], "reading_logprob_std": 0.5434502918409274, "tokenization_path_logprob_std": 76.10771062538717, "trajectory_balance_rmse": 406.0262341493473} +{"canonical_length": 49, "canonical_nll": 1.875201789700255, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.3636363636363637, "prompt_id": "4d1f490ef0029856", "reading_canonical_log_prob": -91.8848876953125, "reading_distances": [21, 21, 25, 25], "reading_log_probs": [-90.67964172363281, -92.58154296875, -91.64584350585938, -90.6765365600586], "reading_logprob_std": 0.9126391551147917, "tokenization_path_logprob_std": 121.5421300436001, "trajectory_balance_rmse": 1250.8563936522696} +{"canonical_length": 58, "canonical_nll": 1.9066114754512393, "generation_rouge_l_f1": 0.30927835051546393, "generation_token_f1": 0.37113402061855666, "prompt_id": "04c2c0f3f6e1a037", "reading_canonical_log_prob": -110.58346557617188, "reading_distances": [11, 14, 10, 10], "reading_log_probs": [-111.36534118652344, -110.73951721191406, -109.30169677734375, -112.79357147216797], "reading_logprob_std": 1.4482685361623961, "tokenization_path_logprob_std": 93.57747216316618, "trajectory_balance_rmse": 1366.959050281528} +{"canonical_length": 68, "canonical_nll": 0.41837215423583984, "generation_rouge_l_f1": 0.23622047244094488, "generation_token_f1": 0.4094488188976378, "prompt_id": "42b3ed92d904a820", "reading_canonical_log_prob": -28.44930648803711, "reading_distances": [10, 11, 11, 14], "reading_log_probs": [-29.062070846557617, -31.325225830078125, -29.35773468017578, -29.34163475036621], "reading_logprob_std": 1.0445636155213651, "tokenization_path_logprob_std": 60.73954305114628, "trajectory_balance_rmse": 1352.5341959827986} +{"canonical_length": 4, "canonical_nll": 0.7562856078147888, "generation_rouge_l_f1": 0.01568627450980392, "generation_token_f1": 0.01568627450980392, "prompt_id": "a0febc02441ecdc9", "reading_canonical_log_prob": -3.0251424312591553, "reading_distances": [48, 47, 48, 41], "reading_log_probs": [-6.257871150970459, -6.133947849273682, -5.4182538986206055, -5.092516899108887], "reading_logprob_std": 0.5613425495551005, "tokenization_path_logprob_std": 28.723821205641386, "trajectory_balance_rmse": 118.25832241447748} +{"canonical_length": 84, "canonical_nll": 1.9110136486235119, "generation_rouge_l_f1": 0.1483679525222552, "generation_token_f1": 0.24332344213649854, "prompt_id": "95d007f697f6007d", "reading_canonical_log_prob": -160.525146484375, "reading_distances": [28, 23, 21, 24], "reading_log_probs": [-167.54763793945312, -166.05572509765625, -167.16131591796875, -164.30970764160156], "reading_logprob_std": 1.4509140517331394, "tokenization_path_logprob_std": 115.04334500959799, "trajectory_balance_rmse": 1709.6742219941652} +{"canonical_length": 178, "canonical_nll": 1.013051236613413, "generation_rouge_l_f1": 0.2936708860759494, "generation_token_f1": 0.46582278481012657, "prompt_id": "42731119470564d3", "reading_canonical_log_prob": -180.3231201171875, "reading_distances": [57, 55, 51, 60], "reading_log_probs": [-189.45455932617188, -189.49740600585938, -190.339111328125, -188.1495819091797], "reading_logprob_std": 0.903989807396762, "tokenization_path_logprob_std": 149.45922201952186, "trajectory_balance_rmse": 2635.2290080354637} +{"canonical_length": 302, "canonical_nll": 1.2318331484763039, "generation_rouge_l_f1": 0.2555366269165248, "generation_token_f1": 0.5144804088586031, "prompt_id": "f73070b2f6b0ff25", "reading_canonical_log_prob": -372.01361083984375, "reading_distances": [25, 27, 22, 27], "reading_log_probs": [-372.2209167480469, -372.848388671875, -373.36614990234375, -373.1601257324219], "reading_logprob_std": 0.49959117907469897, "tokenization_path_logprob_std": 102.86444185013839, "trajectory_balance_rmse": 6511.612900399439} +{"canonical_length": 143, "canonical_nll": 1.1723014991600196, "generation_rouge_l_f1": 0.2164179104477612, "generation_token_f1": 0.34328358208955223, "prompt_id": "08c2f6d9de2bf434", "reading_canonical_log_prob": -167.6391143798828, "reading_distances": [21, 22, 20, 18], "reading_log_probs": [-172.55726623535156, -168.43067932128906, -169.7734832763672, -170.42376708984375], "reading_logprob_std": 1.7206678954922228, "tokenization_path_logprob_std": 165.06667819838282, "trajectory_balance_rmse": 2976.260049944354} +{"canonical_length": 118, "canonical_nll": 1.0996673066737288, "generation_rouge_l_f1": 0.2325581395348837, "generation_token_f1": 0.3772609819121447, "prompt_id": "5f825922a6155ea5", "reading_canonical_log_prob": -129.7607421875, "reading_distances": [37, 29, 34, 36], "reading_log_probs": [-129.78868103027344, -128.3882598876953, -129.38218688964844, -131.21060180664062], "reading_logprob_std": 1.170643749486222, "tokenization_path_logprob_std": 165.88706850773497, "trajectory_balance_rmse": 2425.526946169968} +{"canonical_length": 55, "canonical_nll": 1.2990217729048295, "generation_rouge_l_f1": 0.19230769230769235, "generation_token_f1": 0.3269230769230769, "prompt_id": "3828b93ba3b0437a", "reading_canonical_log_prob": -71.44619750976562, "reading_distances": [22, 30, 26, 26], "reading_log_probs": [-69.29820251464844, -68.94575500488281, -71.01516723632812, -69.89729309082031], "reading_logprob_std": 0.9068549561961584, "tokenization_path_logprob_std": 121.40014319079577, "trajectory_balance_rmse": 1128.5191419574214} +{"canonical_length": 109, "canonical_nll": 1.8627949285944667, "generation_rouge_l_f1": 0.2396694214876033, "generation_token_f1": 0.4545454545454545, "prompt_id": "bf57ac1eadcdbbef", "reading_canonical_log_prob": -203.04464721679688, "reading_distances": [36, 45, 38, 37], "reading_log_probs": [-203.306396484375, -202.345458984375, -201.94720458984375, -201.44760131835938], "reading_logprob_std": 0.7874156745504627, "tokenization_path_logprob_std": 126.39682074341452, "trajectory_balance_rmse": 2509.09975459869} +{"canonical_length": 15, "canonical_nll": 2.6369616190592446, "generation_rouge_l_f1": 0.2, "generation_token_f1": 0.2, "prompt_id": "fd15860ce78ef664", "reading_canonical_log_prob": -39.55442428588867, "reading_distances": [34, 33, 22, 31], "reading_log_probs": [-39.74559783935547, -40.114017486572266, -40.447776794433594, -40.18810272216797], "reading_logprob_std": 0.2899588209624512, "tokenization_path_logprob_std": 47.74478103369188, "trajectory_balance_rmse": 290.5491745372374} +{"canonical_length": 25, "canonical_nll": 1.529100341796875, "generation_rouge_l_f1": 0.24242424242424243, "generation_token_f1": 0.24242424242424243, "prompt_id": "b2df929cec29e4f9", "reading_canonical_log_prob": -38.227508544921875, "reading_distances": [31, 39, 30, 29], "reading_log_probs": [-47.28986358642578, -46.11351776123047, -45.9378662109375, -46.100528717041016], "reading_logprob_std": 0.6247456221400084, "tokenization_path_logprob_std": 34.58566673250156, "trajectory_balance_rmse": 613.8236163194183} +{"canonical_length": 280, "canonical_nll": 0.7670741489955357, "generation_rouge_l_f1": 0.4063116370808678, "generation_token_f1": 0.6232741617357002, "prompt_id": "190309fcf6be3c92", "reading_canonical_log_prob": -214.78076171875, "reading_distances": [24, 29, 28, 27], "reading_log_probs": [-215.5707244873047, -215.51834106445312, -219.91616821289062, -218.24740600585938], "reading_logprob_std": 2.1529765351825225, "tokenization_path_logprob_std": 178.0252107304337, "trajectory_balance_rmse": 3850.0882922047967} +{"canonical_length": 20, "canonical_nll": 0.8513469696044922, "generation_rouge_l_f1": 0.26373626373626374, "generation_token_f1": 0.2857142857142857, "prompt_id": "48be81c0aa95d468", "reading_canonical_log_prob": -17.026939392089844, "reading_distances": [14, 23, 18, 23], "reading_log_probs": [-17.075702667236328, -16.040517807006836, -15.702773094177246, -15.7970609664917], "reading_logprob_std": 0.6307205425420995, "tokenization_path_logprob_std": 58.21311758678996, "trajectory_balance_rmse": 534.8282290662305} +{"canonical_length": 86, "canonical_nll": 1.8043968732966933, "generation_rouge_l_f1": 0.25870646766169153, "generation_token_f1": 0.40796019900497504, "prompt_id": "87fbf3fe9c551bf6", "reading_canonical_log_prob": -155.17813110351562, "reading_distances": [32, 30, 29, 27], "reading_log_probs": [-155.05389404296875, -155.3004608154297, -157.0125274658203, -154.58181762695312], "reading_logprob_std": 1.0597117048770257, "tokenization_path_logprob_std": 80.27842824041672, "trajectory_balance_rmse": 1840.7144575837565} +{"canonical_length": 122, "canonical_nll": 1.240175090852331, "generation_rouge_l_f1": 0.1712707182320442, "generation_token_f1": 0.27624309392265195, "prompt_id": "31fa13f44ebbb510", "reading_canonical_log_prob": -151.30136108398438, "reading_distances": [22, 22, 21, 22], "reading_log_probs": [-153.04275512695312, -147.04603576660156, -148.7825927734375, -151.7359619140625], "reading_logprob_std": 2.7317692931780932, "tokenization_path_logprob_std": 98.47147075647115, "trajectory_balance_rmse": 2211.264145622307} +{"canonical_length": 113, "canonical_nll": 1.7860993241841814, "generation_rouge_l_f1": 0.2651933701657458, "generation_token_f1": 0.4530386740331491, "prompt_id": "1e48963a206d1f06", "reading_canonical_log_prob": -201.8292236328125, "reading_distances": [13, 13, 15, 13], "reading_log_probs": [-192.47630310058594, -194.35549926757812, -189.8839111328125, -191.7224884033203], "reading_logprob_std": 1.8513131372157705, "tokenization_path_logprob_std": 245.41874720245616, "trajectory_balance_rmse": 2439.1151672171136} +{"canonical_length": 7, "canonical_nll": 2.394629887172154, "generation_rouge_l_f1": 0.7499999999999999, "generation_token_f1": 0.7499999999999999, "prompt_id": "b92fcbf5058fa547", "reading_canonical_log_prob": -16.762409210205078, "reading_distances": [21, 17, 20, 17], "reading_log_probs": [-16.93624496459961, -16.62836265563965, -16.622989654541016, -16.312965393066406], "reading_logprob_std": 0.2544630252812468, "tokenization_path_logprob_std": 25.669043833938662, "trajectory_balance_rmse": 125.4299215057616} +{"canonical_length": 14, "canonical_nll": 1.0377512659345354, "generation_rouge_l_f1": 0.64, "generation_token_f1": 0.8, "prompt_id": "5f82c19500742c50", "reading_canonical_log_prob": -14.528517723083496, "reading_distances": [28, 27, 25, 23], "reading_log_probs": [-16.168167114257812, -16.012962341308594, -15.569380760192871, -15.747198104858398], "reading_logprob_std": 0.2675292480552838, "tokenization_path_logprob_std": 56.581999699739804, "trajectory_balance_rmse": 410.58447571314645} +{"canonical_length": 107, "canonical_nll": 1.336607175452687, "generation_rouge_l_f1": 0.3121951219512195, "generation_token_f1": 0.5560975609756097, "prompt_id": "cb0faecefbcafa87", "reading_canonical_log_prob": -143.0169677734375, "reading_distances": [26, 27, 26, 28], "reading_log_probs": [-143.97547912597656, -145.9649658203125, -146.83078002929688, -144.0806884765625], "reading_logprob_std": 1.4137718010096845, "tokenization_path_logprob_std": 137.85250738684314, "trajectory_balance_rmse": 2409.0541153405097} +{"canonical_length": 111, "canonical_nll": 1.5703837076822917, "generation_rouge_l_f1": 0.09999999999999999, "generation_token_f1": 0.11666666666666667, "prompt_id": "935eaf327e8c6a5b", "reading_canonical_log_prob": -174.31259155273438, "reading_distances": [36, 33, 34, 33], "reading_log_probs": [-175.1980743408203, -174.21603393554688, -176.30511474609375, -179.18890380859375], "reading_logprob_std": 2.1510965115027862, "tokenization_path_logprob_std": 99.9496507310432, "trajectory_balance_rmse": 2220.730070213146} +{"canonical_length": 72, "canonical_nll": 1.2695833841959636, "generation_rouge_l_f1": 0.1337386018237082, "generation_token_f1": 0.2188449848024316, "prompt_id": "116cdf21f0043438", "reading_canonical_log_prob": -91.41000366210938, "reading_distances": [12, 18, 18, 16], "reading_log_probs": [-92.42716217041016, -91.49087524414062, -90.35655212402344, -91.20259094238281], "reading_logprob_std": 0.8538741284412944, "tokenization_path_logprob_std": 227.20028934796082, "trajectory_balance_rmse": 1742.294356827168} +{"canonical_length": 31, "canonical_nll": 1.4187206145255797, "generation_rouge_l_f1": 0.3018867924528302, "generation_token_f1": 0.339622641509434, "prompt_id": "da79267701ae23c7", "reading_canonical_log_prob": -43.98033905029297, "reading_distances": [33, 35, 40, 31], "reading_log_probs": [-44.15851593017578, -45.177825927734375, -44.84222412109375, -45.55148696899414], "reading_logprob_std": 0.5917574820133581, "tokenization_path_logprob_std": 62.19045461726365, "trajectory_balance_rmse": 892.1047475600258} +{"canonical_length": 269, "canonical_nll": 0.7474990901450685, "generation_rouge_l_f1": 0.250453720508167, "generation_token_f1": 0.3883847549909257, "prompt_id": "3348c8386d3ef99b", "reading_canonical_log_prob": -201.07725524902344, "reading_distances": [15, 17, 11, 12], "reading_log_probs": [-197.2611083984375, -197.41148376464844, -198.06011962890625, -197.911376953125], "reading_logprob_std": 0.3847755794268198, "tokenization_path_logprob_std": 145.3416312448838, "trajectory_balance_rmse": 4611.8198747784345} +{"canonical_length": 10, "canonical_nll": 0.27609341144561766, "generation_rouge_l_f1": 0.8888888888888888, "generation_token_f1": 0.8888888888888888, "prompt_id": "d04f01367f709567", "reading_canonical_log_prob": -2.7609341144561768, "reading_distances": [20, 27, 14, 25], "reading_log_probs": [-2.884014844894409, -2.72570538520813, -2.852170705795288, -2.9331066608428955], "reading_logprob_std": 0.08852746084306974, "tokenization_path_logprob_std": 42.50001081555976, "trajectory_balance_rmse": 179.95487503279546} +{"canonical_length": 281, "canonical_nll": 0.8563879698620996, "generation_rouge_l_f1": 0.22568093385214008, "generation_token_f1": 0.33073929961089493, "prompt_id": "f04ba139f8e40a51", "reading_canonical_log_prob": -240.64501953125, "reading_distances": [14, 12, 15, 9], "reading_log_probs": [-244.92721557617188, -252.43545532226562, -245.3077392578125, -243.7435302734375], "reading_logprob_std": 3.9446189121989237, "tokenization_path_logprob_std": 162.4733653859967, "trajectory_balance_rmse": 4181.1518792139195} +{"canonical_length": 501, "canonical_nll": 0.7982560270084831, "generation_rouge_l_f1": 0.4803370786516854, "generation_token_f1": 0.5870786516853932, "prompt_id": "16085c364de7d67f", "reading_canonical_log_prob": -399.92626953125, "reading_distances": [14, 13, 17, 15], "reading_log_probs": [-400.40447998046875, -399.2894287109375, -400.0686340332031, -400.23248291015625], "reading_logprob_std": 0.49236448538016314, "tokenization_path_logprob_std": 87.9665446893182, "trajectory_balance_rmse": 6906.173150833747} +{"canonical_length": 405, "canonical_nll": 1.1208959508825231, "generation_rouge_l_f1": 0.2132921174652241, "generation_token_f1": 0.3863987635239567, "prompt_id": "122921a791ee9365", "reading_canonical_log_prob": -453.9628601074219, "reading_distances": [24, 22, 23, 21], "reading_log_probs": [-458.98828125, -458.90716552734375, -454.2850646972656, -459.22283935546875], "reading_logprob_std": 2.3809474275237354, "tokenization_path_logprob_std": 171.2237975565998, "trajectory_balance_rmse": 7424.50954106793} +{"canonical_length": 67, "canonical_nll": 1.6613028227393307, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.2888888888888889, "prompt_id": "5090b423053cf048", "reading_canonical_log_prob": -111.30728912353516, "reading_distances": [33, 38, 36, 33], "reading_log_probs": [-108.58267974853516, -105.27680969238281, -107.78693389892578, -107.39796447753906], "reading_logprob_std": 1.4117618107735908, "tokenization_path_logprob_std": 115.81142408947628, "trajectory_balance_rmse": 1532.7009811445646} +{"canonical_length": 349, "canonical_nll": 1.064802372011551, "generation_rouge_l_f1": 0.24600638977635783, "generation_token_f1": 0.5143769968051118, "prompt_id": "4811a788bad2e4f1", "reading_canonical_log_prob": -371.61602783203125, "reading_distances": [31, 24, 29, 22], "reading_log_probs": [-370.99566650390625, -374.397216796875, -372.217529296875, -373.1866455078125], "reading_logprob_std": 1.4439410407473867, "tokenization_path_logprob_std": 71.05518939131274, "trajectory_balance_rmse": 7770.2604654413435} +{"canonical_length": 14, "canonical_nll": 1.2556791305541992, "generation_rouge_l_f1": 0.4827586206896552, "generation_token_f1": 0.4827586206896552, "prompt_id": "b2fd6c1ae0104b61", "reading_canonical_log_prob": -17.57950782775879, "reading_distances": [24, 18, 27, 20], "reading_log_probs": [-17.623855590820312, -18.15519905090332, -18.018495559692383, -17.735034942626953], "reading_logprob_std": 0.24596799752754492, "tokenization_path_logprob_std": 46.13084215188907, "trajectory_balance_rmse": 339.7211781878738} +{"canonical_length": 12, "canonical_nll": 0.7027338345845541, "generation_rouge_l_f1": 0.11200000000000002, "generation_token_f1": 0.128, "prompt_id": "3b02f9d0f5b11977", "reading_canonical_log_prob": -8.432806015014648, "reading_distances": [26, 20, 18, 19], "reading_log_probs": [-8.006949424743652, -7.841660499572754, -7.866694927215576, -7.39384651184082], "reading_logprob_std": 0.2657750691796134, "tokenization_path_logprob_std": 12.445643415624405, "trajectory_balance_rmse": 224.22756400891333} +{"canonical_length": 6, "canonical_nll": 0.44956378142038983, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "842d7e43c22f686e", "reading_canonical_log_prob": -2.697382688522339, "reading_distances": [23, 27, 22, 26], "reading_log_probs": [-3.0779476165771484, -2.842862606048584, -2.9925405979156494, -2.9205987453460693], "reading_logprob_std": 0.10039091386630157, "tokenization_path_logprob_std": 30.879929425171635, "trajectory_balance_rmse": 177.09660758012413} +{"canonical_length": 74, "canonical_nll": 1.7810149321684967, "generation_rouge_l_f1": 0.21364985163204747, "generation_token_f1": 0.33827893175074186, "prompt_id": "f726d0133e785ebc", "reading_canonical_log_prob": -131.79510498046875, "reading_distances": [22, 22, 20, 22], "reading_log_probs": [-132.45054626464844, -131.36778259277344, -131.23658752441406, -130.37310791015625], "reading_logprob_std": 0.8521546232579662, "tokenization_path_logprob_std": 93.00404254474299, "trajectory_balance_rmse": 1346.6524501349263} +{"canonical_length": 48, "canonical_nll": 1.3966913223266602, "generation_rouge_l_f1": 0.2814070351758794, "generation_token_f1": 0.36180904522613067, "prompt_id": "cb06f10a99b5f8fb", "reading_canonical_log_prob": -67.04118347167969, "reading_distances": [21, 19, 24, 17], "reading_log_probs": [-65.70957946777344, -65.82240295410156, -66.22718048095703, -65.79505920410156], "reading_logprob_std": 0.23080801149821722, "tokenization_path_logprob_std": 29.053784986696108, "trajectory_balance_rmse": 1063.9896750061864} +{"canonical_length": 171, "canonical_nll": 0.8997269122921235, "generation_rouge_l_f1": 0.25510204081632654, "generation_token_f1": 0.40816326530612246, "prompt_id": "443ce871037b41c0", "reading_canonical_log_prob": -153.85330200195312, "reading_distances": [30, 36, 30, 28], "reading_log_probs": [-156.68995666503906, -157.38558959960938, -157.5720977783203, -156.8378143310547], "reading_logprob_std": 0.42406320669743097, "tokenization_path_logprob_std": 7.015477965526784, "trajectory_balance_rmse": 2364.3062304604364} +{"canonical_length": 307, "canonical_nll": 1.3094726960123169, "generation_rouge_l_f1": 0.2755102040816326, "generation_token_f1": 0.5306122448979592, "prompt_id": "d1c3cdcce82ccde0", "reading_canonical_log_prob": -402.00811767578125, "reading_distances": [44, 41, 34, 36], "reading_log_probs": [-403.61822509765625, -406.5935974121094, -404.902099609375, -406.7818603515625], "reading_logprob_std": 1.4983288020911198, "tokenization_path_logprob_std": 235.54041944215805, "trajectory_balance_rmse": 6921.700584699324} +{"canonical_length": 61, "canonical_nll": 0.7142162010317943, "generation_rouge_l_f1": 0.19999999999999998, "generation_token_f1": 0.32499999999999996, "prompt_id": "66bf9508c6f9f94b", "reading_canonical_log_prob": -43.56718826293945, "reading_distances": [41, 46, 45, 47], "reading_log_probs": [-43.98893356323242, -43.14229965209961, -43.081783294677734, -43.36254119873047], "reading_logprob_std": 0.414635538012128, "tokenization_path_logprob_std": 138.08342689006534, "trajectory_balance_rmse": 1282.553598726258} +{"canonical_length": 265, "canonical_nll": 0.8306321629938089, "generation_rouge_l_f1": 0.18490566037735848, "generation_token_f1": 0.43018867924528303, "prompt_id": "c9e81615c39cdb45", "reading_canonical_log_prob": -220.11752319335938, "reading_distances": [19, 27, 23, 24], "reading_log_probs": [-220.70608520507812, -220.8304901123047, -220.60879516601562, -219.4749755859375], "reading_logprob_std": 0.6266769467832233, "tokenization_path_logprob_std": 297.6173143609457, "trajectory_balance_rmse": 5800.6620297687095} +{"canonical_length": 41, "canonical_nll": 1.2734603881835938, "generation_rouge_l_f1": 0.06557377049180328, "generation_token_f1": 0.11803278688524589, "prompt_id": "768e651293911174", "reading_canonical_log_prob": -52.211875915527344, "reading_distances": [55, 54, 52, 61], "reading_log_probs": [-56.90073013305664, -56.02985763549805, -56.07576370239258, -55.86668395996094], "reading_logprob_std": 0.4637426873531009, "tokenization_path_logprob_std": 80.28782958326921, "trajectory_balance_rmse": 822.4894310539182} +{"canonical_length": 367, "canonical_nll": 1.1376968092749489, "generation_rouge_l_f1": 0.1910631741140216, "generation_token_f1": 0.32973805855161786, "prompt_id": "bd6b470fc22dea74", "reading_canonical_log_prob": -417.53472900390625, "reading_distances": [48, 35, 46, 46], "reading_log_probs": [-421.876220703125, -421.016845703125, -422.0648498535156, -424.43231201171875], "reading_logprob_std": 1.4627687502293207, "tokenization_path_logprob_std": 175.1885082391874, "trajectory_balance_rmse": 7718.087168236956} +{"canonical_length": 303, "canonical_nll": 1.1735794520614171, "generation_rouge_l_f1": 0.26712328767123283, "generation_token_f1": 0.4417808219178082, "prompt_id": "ca8e3b0a9ea7294b", "reading_canonical_log_prob": -355.5945739746094, "reading_distances": [24, 17, 25, 21], "reading_log_probs": [-355.6749267578125, -355.04461669921875, -355.80596923828125, -357.1378173828125], "reading_logprob_std": 0.8798427015187782, "tokenization_path_logprob_std": 140.36748425151254, "trajectory_balance_rmse": 6686.9654557121885} +{"canonical_length": 246, "canonical_nll": 1.4495919080284554, "generation_rouge_l_f1": 0.2737642585551331, "generation_token_f1": 0.5437262357414449, "prompt_id": "5cd1f02fb9a1877d", "reading_canonical_log_prob": -356.599609375, "reading_distances": [25, 17, 30, 29], "reading_log_probs": [-353.7880859375, -353.2986755371094, -357.237060546875, -353.29229736328125], "reading_logprob_std": 1.9029106750654041, "tokenization_path_logprob_std": 167.0989153861875, "trajectory_balance_rmse": 5058.59039467536} +{"canonical_length": 210, "canonical_nll": 0.9163587297712054, "generation_rouge_l_f1": 0.21074380165289258, "generation_token_f1": 0.4752066115702479, "prompt_id": "754987e5b5772585", "reading_canonical_log_prob": -192.43533325195312, "reading_distances": [27, 24, 27, 33], "reading_log_probs": [-192.30325317382812, -195.2393035888672, -192.12713623046875, -193.13575744628906], "reading_logprob_std": 1.4280604964628856, "tokenization_path_logprob_std": 229.47287148094793, "trajectory_balance_rmse": 5056.010810455437} +{"canonical_length": 221, "canonical_nll": 1.2213990914875565, "generation_rouge_l_f1": 0.25672877846790887, "generation_token_f1": 0.48861283643892334, "prompt_id": "5aac80ddf699bdd3", "reading_canonical_log_prob": -269.92919921875, "reading_distances": [20, 20, 20, 20], "reading_log_probs": [-272.8553466796875, -270.9046630859375, -273.60491943359375, -272.1617431640625], "reading_logprob_std": 1.1475518885881273, "tokenization_path_logprob_std": 248.27331227264852, "trajectory_balance_rmse": 4286.689653826932} +{"canonical_length": 25, "canonical_nll": 1.2626068115234375, "generation_rouge_l_f1": 0.1461187214611872, "generation_token_f1": 0.1735159817351598, "prompt_id": "f552b8f246a9b742", "reading_canonical_log_prob": -31.565170288085938, "reading_distances": [23, 22, 27, 26], "reading_log_probs": [-29.767276763916016, -28.98063850402832, -32.42805099487305, -31.261638641357422], "reading_logprob_std": 1.5378493903217518, "tokenization_path_logprob_std": 64.48918594024413, "trajectory_balance_rmse": 473.930898300558} +{"canonical_length": 212, "canonical_nll": 1.2281425044221699, "generation_rouge_l_f1": 0.2566844919786096, "generation_token_f1": 0.4705882352941177, "prompt_id": "1dcab7a046130541", "reading_canonical_log_prob": -260.3662109375, "reading_distances": [20, 18, 19, 26], "reading_log_probs": [-258.8084716796875, -259.7038879394531, -259.254638671875, -258.8397216796875], "reading_logprob_std": 0.4205726418639448, "tokenization_path_logprob_std": 122.32687464248943, "trajectory_balance_rmse": 4453.779180802161} +{"canonical_length": 21, "canonical_nll": 1.4607285090855189, "generation_rouge_l_f1": 0.4571428571428571, "generation_token_f1": 0.4571428571428571, "prompt_id": "721929550e55e7e5", "reading_canonical_log_prob": -30.6752986907959, "reading_distances": [24, 25, 26, 27], "reading_log_probs": [-29.905338287353516, -31.243492126464844, -29.669342041015625, -29.679967880249023], "reading_logprob_std": 0.7538684243230794, "tokenization_path_logprob_std": 26.392545093252323, "trajectory_balance_rmse": 358.5766669179379} +{"canonical_length": 106, "canonical_nll": 2.20602071510171, "generation_rouge_l_f1": 0.2603174603174603, "generation_token_f1": 0.380952380952381, "prompt_id": "a21c76c0694212ca", "reading_canonical_log_prob": -233.83819580078125, "reading_distances": [34, 36, 35, 33], "reading_log_probs": [-235.95059204101562, -236.1703643798828, -237.45413208007812, -234.99398803710938], "reading_logprob_std": 1.0127624462471068, "tokenization_path_logprob_std": 93.32070544875883, "trajectory_balance_rmse": 1941.6907314309742} +{"canonical_length": 10, "canonical_nll": 1.4245681762695312, "generation_rouge_l_f1": 0.0446096654275093, "generation_token_f1": 0.0446096654275093, "prompt_id": "7cb87ccdae4ac547", "reading_canonical_log_prob": -14.245681762695312, "reading_distances": [19, 12, 18, 15], "reading_log_probs": [-13.979202270507812, -15.100025177001953, -14.436493873596191, -14.474018096923828], "reading_logprob_std": 0.4604136747075493, "tokenization_path_logprob_std": 29.48473664804821, "trajectory_balance_rmse": 266.40785548185687} +{"canonical_length": 263, "canonical_nll": 0.8771423803989425, "generation_rouge_l_f1": 0.34583333333333327, "generation_token_f1": 0.6041666666666666, "prompt_id": "02274883d489f1d7", "reading_canonical_log_prob": -230.68844604492188, "reading_distances": [52, 51, 65, 54], "reading_log_probs": [-225.48196411132812, -228.05221557617188, -227.0081787109375, -228.60182189941406], "reading_logprob_std": 1.3723706652227345, "tokenization_path_logprob_std": 219.1440719416359, "trajectory_balance_rmse": 4516.999882412386} +{"canonical_length": 219, "canonical_nll": 1.1891490640161244, "generation_rouge_l_f1": 0.2236024844720497, "generation_token_f1": 0.4596273291925466, "prompt_id": "dd480c5be4f865e4", "reading_canonical_log_prob": -260.42364501953125, "reading_distances": [25, 18, 27, 22], "reading_log_probs": [-264.3691101074219, -263.61334228515625, -260.150634765625, -262.84613037109375], "reading_logprob_std": 1.8378166514867926, "tokenization_path_logprob_std": 249.56489989140806, "trajectory_balance_rmse": 5055.043320927036} +{"canonical_length": 17, "canonical_nll": 0.5222193774055032, "generation_rouge_l_f1": 0.07174887892376682, "generation_token_f1": 0.07174887892376682, "prompt_id": "48b6b6146fd695d3", "reading_canonical_log_prob": -8.877729415893555, "reading_distances": [17, 19, 18, 17], "reading_log_probs": [-13.369529724121094, -16.121337890625, -17.53411865234375, -10.376243591308594], "reading_logprob_std": 3.1637694339889015, "tokenization_path_logprob_std": 26.19736669933955, "trajectory_balance_rmse": 256.16482730385724} +{"canonical_length": 230, "canonical_nll": 2.216341499660326, "generation_rouge_l_f1": 0.2708803611738149, "generation_token_f1": 0.4604966139954853, "prompt_id": "619b20aa15b49441", "reading_canonical_log_prob": -509.758544921875, "reading_distances": [26, 22, 19, 24], "reading_log_probs": [-507.9873352050781, -507.512939453125, -507.2484130859375, -507.4876708984375], "reading_logprob_std": 0.3093782267970167, "tokenization_path_logprob_std": 91.60674708407358, "trajectory_balance_rmse": 6122.088979644191} +{"canonical_length": 264, "canonical_nll": 1.1879332571318655, "generation_rouge_l_f1": 0.24535315985130113, "generation_token_f1": 0.4944237918215613, "prompt_id": "49784b3e9fd7ae19", "reading_canonical_log_prob": -313.6143798828125, "reading_distances": [38, 31, 38, 35], "reading_log_probs": [-310.2970275878906, -312.4705505371094, -312.2850341796875, -313.6502685546875], "reading_logprob_std": 1.39076157253752, "tokenization_path_logprob_std": 252.56278044172856, "trajectory_balance_rmse": 6143.176029197931} +{"canonical_length": 128, "canonical_nll": 1.140223741531372, "generation_rouge_l_f1": 0.3103448275862069, "generation_token_f1": 0.47413793103448276, "prompt_id": "211ba51e9f80e5d0", "reading_canonical_log_prob": -145.94863891601562, "reading_distances": [11, 9, 14, 12], "reading_log_probs": [-146.5888214111328, -146.54443359375, -146.95819091796875, -146.40725708007812], "reading_logprob_std": 0.23539157628022947, "tokenization_path_logprob_std": 23.365235991056597, "trajectory_balance_rmse": 2663.5841700671926} +{"canonical_length": 19, "canonical_nll": 0.6200102755897924, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "c295bb5aa04e8e4c", "reading_canonical_log_prob": -11.780195236206055, "reading_distances": [23, 23, 25, 18], "reading_log_probs": [-12.356668472290039, -14.212398529052734, -12.547639846801758, -12.755993843078613], "reading_logprob_std": 0.8453604915671935, "tokenization_path_logprob_std": 95.94341733400697, "trajectory_balance_rmse": 602.3876171321572} +{"canonical_length": 163, "canonical_nll": 0.5815738607769364, "generation_rouge_l_f1": 0.24870466321243526, "generation_token_f1": 0.3937823834196892, "prompt_id": "55b91ee2bfd58fff", "reading_canonical_log_prob": -94.79653930664062, "reading_distances": [39, 24, 36, 32], "reading_log_probs": [-96.12945556640625, -96.06095886230469, -96.92732238769531, -97.6467056274414], "reading_logprob_std": 0.7486666936179667, "tokenization_path_logprob_std": 153.3734180247863, "trajectory_balance_rmse": 2343.122356089937} +{"canonical_length": 79, "canonical_nll": 0.9834332043611551, "generation_rouge_l_f1": 0.45312500000000006, "generation_token_f1": 0.546875, "prompt_id": "aacdb99cf2030827", "reading_canonical_log_prob": -77.69122314453125, "reading_distances": [13, 29, 28, 24], "reading_log_probs": [-77.84149169921875, -78.78514862060547, -78.79613494873047, -78.32675170898438], "reading_logprob_std": 0.45349370171490566, "tokenization_path_logprob_std": 86.59908150984272, "trajectory_balance_rmse": 1865.926318463405} +{"canonical_length": 7, "canonical_nll": 0.13896214962005615, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "732d6c9c54c4fcf0", "reading_canonical_log_prob": -0.9727350473403931, "reading_distances": [19, 17, 23, 18], "reading_log_probs": [-1.4036383628845215, -1.11483633518219, -1.2070965766906738, -1.117742896080017], "reading_logprob_std": 0.13548554146806577, "tokenization_path_logprob_std": 43.75285391693249, "trajectory_balance_rmse": 169.31644529848157} +{"canonical_length": 234, "canonical_nll": 0.6121381449903178, "generation_rouge_l_f1": 0.3391304347826087, "generation_token_f1": 0.5956521739130434, "prompt_id": "1e288eaee9c72d47", "reading_canonical_log_prob": -143.24032592773438, "reading_distances": [34, 34, 31, 32], "reading_log_probs": [-145.81617736816406, -144.81149291992188, -145.37033081054688, -145.61764526367188], "reading_logprob_std": 0.4350221084467276, "tokenization_path_logprob_std": 347.03750194942006, "trajectory_balance_rmse": 4087.2110418333914} +{"canonical_length": 85, "canonical_nll": 1.1727152207318474, "generation_rouge_l_f1": 0.5921052631578947, "generation_token_f1": 0.6842105263157894, "prompt_id": "34251c7523ed14e6", "reading_canonical_log_prob": -99.68079376220703, "reading_distances": [26, 25, 27, 25], "reading_log_probs": [-100.41748809814453, -99.56485748291016, -99.09978485107422, -100.13250732421875], "reading_logprob_std": 0.5880454529371125, "tokenization_path_logprob_std": 112.74582629530077, "trajectory_balance_rmse": 1872.0504360850682} +{"canonical_length": 62, "canonical_nll": 2.3622438984532512, "generation_rouge_l_f1": 0.21476510067114096, "generation_token_f1": 0.4161073825503356, "prompt_id": "915708e38c0c1bff", "reading_canonical_log_prob": -146.45912170410156, "reading_distances": [13, 14, 11, 14], "reading_log_probs": [-151.61148071289062, -158.00341796875, -149.37820434570312, -150.15492248535156], "reading_logprob_std": 3.9217601230173744, "tokenization_path_logprob_std": 71.16046186155607, "trajectory_balance_rmse": 1086.6831325666562} +{"canonical_length": 145, "canonical_nll": 1.6431690084523167, "generation_rouge_l_f1": 0.21243523316062174, "generation_token_f1": 0.41450777202072536, "prompt_id": "fc76d6d236be83db", "reading_canonical_log_prob": -238.25950622558594, "reading_distances": [23, 27, 22, 25], "reading_log_probs": [-239.54037475585938, -239.4301300048828, -239.77244567871094, -238.54464721679688], "reading_logprob_std": 0.5374502304106794, "tokenization_path_logprob_std": 212.82577136203042, "trajectory_balance_rmse": 2611.1255908743447} +{"canonical_length": 391, "canonical_nll": 0.705644456321931, "generation_rouge_l_f1": 0.3106180665610142, "generation_token_f1": 0.4469096671949287, "prompt_id": "cf72c2330315d4cd", "reading_canonical_log_prob": -275.906982421875, "reading_distances": [23, 19, 28, 22], "reading_log_probs": [-273.8889465332031, -274.75872802734375, -272.906494140625, -274.0752868652344], "reading_logprob_std": 0.7648743849399915, "tokenization_path_logprob_std": 106.29824183926031, "trajectory_balance_rmse": 4295.447932249394} +{"canonical_length": 247, "canonical_nll": 1.1754187456509362, "generation_rouge_l_f1": 0.3125, "generation_token_f1": 0.53125, "prompt_id": "12725fe24fa84c62", "reading_canonical_log_prob": -290.32843017578125, "reading_distances": [34, 37, 32, 31], "reading_log_probs": [-280.2603759765625, -280.99822998046875, -283.5926513671875, -282.4167175292969], "reading_logprob_std": 1.483920661379695, "tokenization_path_logprob_std": 356.19773129237444, "trajectory_balance_rmse": 6046.688201878476} +{"canonical_length": 164, "canonical_nll": 1.6251823611375762, "generation_rouge_l_f1": 0.3269230769230769, "generation_token_f1": 0.5384615384615384, "prompt_id": "767d0dc7cec8c49a", "reading_canonical_log_prob": -266.5299072265625, "reading_distances": [13, 12, 17, 13], "reading_log_probs": [-273.41473388671875, -272.355224609375, -275.1376647949219, -274.76934814453125], "reading_logprob_std": 1.279050871896307, "tokenization_path_logprob_std": 156.58937224520946, "trajectory_balance_rmse": 3622.764014926963} +{"canonical_length": 211, "canonical_nll": 1.2893665187166765, "generation_rouge_l_f1": 0.28893905191873587, "generation_token_f1": 0.41986455981941306, "prompt_id": "0305770999fedf01", "reading_canonical_log_prob": -272.05633544921875, "reading_distances": [33, 26, 28, 37], "reading_log_probs": [-265.3141174316406, -264.5231018066406, -265.2801513671875, -266.87945556640625], "reading_logprob_std": 0.9899681766183464, "tokenization_path_logprob_std": 119.7917319097203, "trajectory_balance_rmse": 2805.329593711226} +{"canonical_length": 28, "canonical_nll": 0.4739522252764021, "generation_rouge_l_f1": 0.6, "generation_token_f1": 0.72, "prompt_id": "47deaa7fb6aebf76", "reading_canonical_log_prob": -13.270662307739258, "reading_distances": [17, 18, 14, 15], "reading_log_probs": [-11.79254150390625, -11.227853775024414, -12.865056037902832, -12.585139274597168], "reading_logprob_std": 0.7471266879514721, "tokenization_path_logprob_std": 39.432220051174816, "trajectory_balance_rmse": 809.3163617693652} +{"canonical_length": 55, "canonical_nll": 1.234315352006392, "generation_rouge_l_f1": 0.2962962962962963, "generation_token_f1": 0.3333333333333333, "prompt_id": "83b035c8880eabe5", "reading_canonical_log_prob": -67.88734436035156, "reading_distances": [38, 38, 29, 28], "reading_log_probs": [-69.58152770996094, -68.60125732421875, -67.8480453491211, -70.19718933105469], "reading_logprob_std": 1.0399413065938912, "tokenization_path_logprob_std": 125.83073213240776, "trajectory_balance_rmse": 1557.373328323696} +{"canonical_length": 235, "canonical_nll": 1.4445977393617022, "generation_rouge_l_f1": 0.2980392156862745, "generation_token_f1": 0.48627450980392156, "prompt_id": "3f8f300bcd0a09f9", "reading_canonical_log_prob": -339.48046875, "reading_distances": [30, 24, 30, 33], "reading_log_probs": [-334.0218200683594, -335.523681640625, -333.8543395996094, -332.8082275390625], "reading_logprob_std": 1.1184508999669487, "tokenization_path_logprob_std": 283.7071403324945, "trajectory_balance_rmse": 5196.7983404910065} +{"canonical_length": 32, "canonical_nll": 1.0435372591018677, "generation_rouge_l_f1": 0.20547945205479454, "generation_token_f1": 0.273972602739726, "prompt_id": "de8521c6b6a04f6f", "reading_canonical_log_prob": -33.393192291259766, "reading_distances": [20, 21, 24, 20], "reading_log_probs": [-32.98687744140625, -30.392562866210938, -31.380470275878906, -31.59834861755371], "reading_logprob_std": 1.069126422645565, "tokenization_path_logprob_std": 102.32755220811032, "trajectory_balance_rmse": 676.1392731443917} +{"canonical_length": 12, "canonical_nll": 0.31348711252212524, "generation_rouge_l_f1": 0.1221374045801527, "generation_token_f1": 0.1221374045801527, "prompt_id": "89d6bb48c1695f7e", "reading_canonical_log_prob": -3.761845350265503, "reading_distances": [23, 26, 34, 26], "reading_log_probs": [-4.2504425048828125, -4.673972129821777, -4.993436813354492, -6.067981719970703], "reading_logprob_std": 0.7764681717983912, "tokenization_path_logprob_std": 63.856673858609305, "trajectory_balance_rmse": 273.90569395754255} +{"canonical_length": 325, "canonical_nll": 1.585799278846154, "generation_rouge_l_f1": 0.22866344605475042, "generation_token_f1": 0.3896940418679549, "prompt_id": "b1d79661f636c046", "reading_canonical_log_prob": -515.384765625, "reading_distances": [16, 19, 22, 14], "reading_log_probs": [-510.5003967285156, -509.9328918457031, -510.24853515625, -514.0081787109375], "reading_logprob_std": 1.904655346590553, "tokenization_path_logprob_std": 427.8079528497277, "trajectory_balance_rmse": 7340.5387618716695} +{"canonical_length": 15, "canonical_nll": 1.1603349049886067, "generation_rouge_l_f1": 0.0625, "generation_token_f1": 0.0763888888888889, "prompt_id": "db698be75c6d5259", "reading_canonical_log_prob": -17.4050235748291, "reading_distances": [22, 19, 26, 22], "reading_log_probs": [-18.42074203491211, -18.181941986083984, -18.367740631103516, -17.791057586669922], "reading_logprob_std": 0.28522099624934827, "tokenization_path_logprob_std": 43.36843210170258, "trajectory_balance_rmse": 378.621357184464} +{"canonical_length": 33, "canonical_nll": 0.9341249176950166, "generation_rouge_l_f1": 0.12000000000000001, "generation_token_f1": 0.13333333333333333, "prompt_id": "4c3924ee604d1034", "reading_canonical_log_prob": -30.826122283935547, "reading_distances": [31, 23, 26, 19], "reading_log_probs": [-29.933692932128906, -29.797374725341797, -29.952739715576172, -29.43832015991211], "reading_logprob_std": 0.23840189409746995, "tokenization_path_logprob_std": 79.5216258434645, "trajectory_balance_rmse": 819.0753515641742} +{"canonical_length": 63, "canonical_nll": 1.3488428025018602, "generation_rouge_l_f1": 0.13664596273291926, "generation_token_f1": 0.2484472049689441, "prompt_id": "1fd45a1cb9cbeccc", "reading_canonical_log_prob": -84.97709655761719, "reading_distances": [16, 16, 20, 19], "reading_log_probs": [-82.0712890625, -82.46510314941406, -85.66651153564453, -84.01699829101562], "reading_logprob_std": 1.6392271222170982, "tokenization_path_logprob_std": 105.95088740868442, "trajectory_balance_rmse": 1321.0286029001459} +{"canonical_length": 205, "canonical_nll": 1.0993242961604421, "generation_rouge_l_f1": 0.43645083932853723, "generation_token_f1": 0.6187050359712231, "prompt_id": "326b7dac8c34a0d5", "reading_canonical_log_prob": -225.36148071289062, "reading_distances": [21, 21, 21, 12], "reading_log_probs": [-225.7592315673828, -225.60189819335938, -224.6602020263672, -225.5148468017578], "reading_logprob_std": 0.49304588563390317, "tokenization_path_logprob_std": 95.19949752714415, "trajectory_balance_rmse": 4377.411682639178} +{"canonical_length": 322, "canonical_nll": 1.4500838569972827, "generation_rouge_l_f1": 0.1785063752276867, "generation_token_f1": 0.31693989071038253, "prompt_id": "1e2cec78b7ee8d07", "reading_canonical_log_prob": -466.927001953125, "reading_distances": [15, 19, 18, 17], "reading_log_probs": [-471.1279602050781, -477.2107238769531, -466.85504150390625, -470.5960388183594], "reading_logprob_std": 4.286899279772017, "tokenization_path_logprob_std": 297.4766201933876, "trajectory_balance_rmse": 6553.247900410006} +{"canonical_length": 360, "canonical_nll": 1.0957943386501736, "generation_rouge_l_f1": 0.27504244482173174, "generation_token_f1": 0.5127334465195246, "prompt_id": "e3633ae650b9ffe5", "reading_canonical_log_prob": -394.4859619140625, "reading_distances": [22, 15, 20, 21], "reading_log_probs": [-390.4874267578125, -390.5625, -389.5838623046875, -391.475830078125], "reading_logprob_std": 0.7730055768594013, "tokenization_path_logprob_std": 224.9631473445306, "trajectory_balance_rmse": 6121.035246988458} +{"canonical_length": 199, "canonical_nll": 1.0647820707541615, "generation_rouge_l_f1": 0.2539682539682539, "generation_token_f1": 0.49886621315192753, "prompt_id": "e7d8d0860929d3e3", "reading_canonical_log_prob": -211.89163208007812, "reading_distances": [24, 30, 28, 25], "reading_log_probs": [-212.23739624023438, -212.34976196289062, -211.97752380371094, -212.3643798828125], "reading_logprob_std": 0.17905239946387608, "tokenization_path_logprob_std": 211.0261147168623, "trajectory_balance_rmse": 3659.3020857250035} +{"canonical_length": 58, "canonical_nll": 1.8902507650441136, "generation_rouge_l_f1": 0.15217391304347824, "generation_token_f1": 0.1956521739130435, "prompt_id": "8c4c85aa12a6cb37", "reading_canonical_log_prob": -109.6345443725586, "reading_distances": [30, 24, 30, 29], "reading_log_probs": [-108.7899398803711, -107.12741088867188, -106.94270324707031, -107.51378631591797], "reading_logprob_std": 0.8323857530659143, "tokenization_path_logprob_std": 95.03698435362229, "trajectory_balance_rmse": 1376.063432997693} +{"canonical_length": 35, "canonical_nll": 2.1395045689174106, "generation_rouge_l_f1": 0.24242424242424246, "generation_token_f1": 0.3636363636363636, "prompt_id": "5f2bc0710dc7dd88", "reading_canonical_log_prob": -74.88265991210938, "reading_distances": [13, 16, 14, 25], "reading_log_probs": [-76.25203704833984, -74.91458129882812, -75.77996826171875, -75.29907989501953], "reading_logprob_std": 0.580786216011615, "tokenization_path_logprob_std": 49.5226428315994, "trajectory_balance_rmse": 756.1756628689269} +{"canonical_length": 24, "canonical_nll": 1.0870805581410725, "generation_rouge_l_f1": 0.26666666666666666, "generation_token_f1": 0.28888888888888886, "prompt_id": "e06153205819c799", "reading_canonical_log_prob": -26.089933395385742, "reading_distances": [70, 71, 64, 77], "reading_log_probs": [-24.37866973876953, -21.167081832885742, -22.493404388427734, -23.9260196685791], "reading_logprob_std": 1.4576428016934504, "tokenization_path_logprob_std": 82.1358582263011, "trajectory_balance_rmse": 529.3589157667897} +{"canonical_length": 81, "canonical_nll": 0.789891278302228, "generation_rouge_l_f1": 0.1609907120743034, "generation_token_f1": 0.19814241486068113, "prompt_id": "788acaefe94c61a6", "reading_canonical_log_prob": -63.98119354248047, "reading_distances": [19, 19, 26, 26], "reading_log_probs": [-63.60624694824219, -61.768394470214844, -61.628578186035156, -62.79840087890625], "reading_logprob_std": 0.9305212546276426, "tokenization_path_logprob_std": 20.24688213485841, "trajectory_balance_rmse": 990.6987187111259} +{"canonical_length": 21, "canonical_nll": 1.991160619826544, "generation_rouge_l_f1": 0.021739130434782608, "generation_token_f1": 0.028985507246376812, "prompt_id": "24304bef6f934812", "reading_canonical_log_prob": -41.81437301635742, "reading_distances": [17, 17, 14, 14], "reading_log_probs": [-45.1628532409668, -44.599388122558594, -42.932979583740234, -42.27525329589844], "reading_logprob_std": 1.3613469705305983, "tokenization_path_logprob_std": 59.05698993512186, "trajectory_balance_rmse": 366.50844532533796} +{"canonical_length": 315, "canonical_nll": 1.4662618001302083, "generation_rouge_l_f1": 0.2534722222222222, "generation_token_f1": 0.5104166666666667, "prompt_id": "969a799cc7257588", "reading_canonical_log_prob": -461.8724670410156, "reading_distances": [20, 15, 16, 18], "reading_log_probs": [-456.6146545410156, -455.1985778808594, -456.2050476074219, -455.623046875], "reading_logprob_std": 0.6250477848933802, "tokenization_path_logprob_std": 279.8602879013647, "trajectory_balance_rmse": 6649.299361298303} +{"canonical_length": 152, "canonical_nll": 1.2160228930021588, "generation_rouge_l_f1": 0.14588235294117646, "generation_token_f1": 0.2776470588235294, "prompt_id": "7f301b40ae5af8a5", "reading_canonical_log_prob": -184.83547973632812, "reading_distances": [23, 32, 31, 24], "reading_log_probs": [-191.72915649414062, -184.2848663330078, -183.72756958007812, -185.55548095703125], "reading_logprob_std": 3.6835605508296436, "tokenization_path_logprob_std": 224.3490663867623, "trajectory_balance_rmse": 2873.6965714729517} +{"canonical_length": 354, "canonical_nll": 1.4175681421312236, "generation_rouge_l_f1": 0.2358803986710964, "generation_token_f1": 0.4053156146179402, "prompt_id": "d838a6ece3beb4b3", "reading_canonical_log_prob": -501.8191223144531, "reading_distances": [15, 19, 22, 17], "reading_log_probs": [-499.06201171875, -498.359130859375, -500.0452575683594, -501.05029296875], "reading_logprob_std": 1.172943205717846, "tokenization_path_logprob_std": 197.5023746146527, "trajectory_balance_rmse": 8153.461752403382} +{"canonical_length": 15, "canonical_nll": 1.9992457071940104, "generation_rouge_l_f1": 0.23076923076923075, "generation_token_f1": 0.23076923076923075, "prompt_id": "3ad7d2823c1e6bd8", "reading_canonical_log_prob": -29.988685607910156, "reading_distances": [40, 44, 31, 41], "reading_log_probs": [-32.50925064086914, -31.978483200073242, -33.31809616088867, -32.431365966796875], "reading_logprob_std": 0.5573723543118824, "tokenization_path_logprob_std": 56.676099042952025, "trajectory_balance_rmse": 502.88277096496654} +{"canonical_length": 8, "canonical_nll": 0.4210316836833954, "generation_rouge_l_f1": 0.7692307692307692, "generation_token_f1": 0.7692307692307692, "prompt_id": "7a4144ac9a54046a", "reading_canonical_log_prob": -3.368253469467163, "reading_distances": [32, 28, 27, 30], "reading_log_probs": [-2.931877613067627, -2.951077699661255, -2.9453868865966797, -2.9857070446014404], "reading_logprob_std": 0.022923865878145155, "tokenization_path_logprob_std": 35.63893350405611, "trajectory_balance_rmse": 140.59425679479767} +{"canonical_length": 34, "canonical_nll": 0.3801690830903895, "generation_rouge_l_f1": 0.19534883720930232, "generation_token_f1": 0.25116279069767444, "prompt_id": "c898f115e1329204", "reading_canonical_log_prob": -12.925748825073242, "reading_distances": [10, 7, 6, 7], "reading_log_probs": [-13.821107864379883, -12.597553253173828, -13.39953327178955, -13.308515548706055], "reading_logprob_std": 0.5078127499098293, "tokenization_path_logprob_std": 65.0138184755996, "trajectory_balance_rmse": 574.6314546170682} +{"canonical_length": 104, "canonical_nll": 1.6035682971660907, "generation_rouge_l_f1": 0.1844380403458213, "generation_token_f1": 0.33429394812680113, "prompt_id": "5328de30202d5ac5", "reading_canonical_log_prob": -166.77110290527344, "reading_distances": [14, 12, 11, 14], "reading_log_probs": [-166.05117797851562, -166.34962463378906, -166.63233947753906, -166.9857940673828], "reading_logprob_std": 0.3989460809179672, "tokenization_path_logprob_std": 50.95088094078247, "trajectory_balance_rmse": 2646.6844688011515} +{"canonical_length": 35, "canonical_nll": 1.3744481767926897, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.2916666666666667, "prompt_id": "f9e71d1877230558", "reading_canonical_log_prob": -48.10568618774414, "reading_distances": [29, 22, 28, 26], "reading_log_probs": [-42.76726531982422, -45.01190948486328, -43.213714599609375, -43.235687255859375], "reading_logprob_std": 0.993567500936529, "tokenization_path_logprob_std": 116.11266509728112, "trajectory_balance_rmse": 830.1523123628126} +{"canonical_length": 211, "canonical_nll": 1.429405031611004, "generation_rouge_l_f1": 0.2510121457489879, "generation_token_f1": 0.4858299595141701, "prompt_id": "4df4b024265f29c3", "reading_canonical_log_prob": -301.6044616699219, "reading_distances": [27, 28, 31, 30], "reading_log_probs": [-303.0189208984375, -304.40435791015625, -303.23992919921875, -304.3837890625], "reading_logprob_std": 0.7357466827935758, "tokenization_path_logprob_std": 127.38693180382052, "trajectory_balance_rmse": 4788.577534284297} +{"canonical_length": 21, "canonical_nll": 0.2719938187372117, "generation_rouge_l_f1": 0.34108527131782945, "generation_token_f1": 0.34108527131782945, "prompt_id": "6e5cc5a69c7269eb", "reading_canonical_log_prob": -5.711870193481445, "reading_distances": [19, 26, 29, 24], "reading_log_probs": [-5.287653923034668, -9.060351371765137, -7.355820655822754, -13.379783630371094], "reading_logprob_std": 3.4380732587651415, "tokenization_path_logprob_std": 74.89634694940061, "trajectory_balance_rmse": 502.74379805793484} +{"canonical_length": 81, "canonical_nll": 1.4100236304012346, "generation_rouge_l_f1": 0.18443804034582134, "generation_token_f1": 0.27665706051873196, "prompt_id": "15d75a153e79b8a7", "reading_canonical_log_prob": -114.2119140625, "reading_distances": [37, 41, 41, 42], "reading_log_probs": [-111.80067443847656, -111.46053314208984, -113.06336212158203, -111.7164535522461], "reading_logprob_std": 0.7168151860972971, "tokenization_path_logprob_std": 121.39432561523728, "trajectory_balance_rmse": 2230.9951693204785} +{"canonical_length": 4, "canonical_nll": 1.340318202972412, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "9acb0ff71abeda7d", "reading_canonical_log_prob": -5.361272811889648, "reading_distances": [21, 16, 14, 15], "reading_log_probs": [-14.320337295532227, -13.056140899658203, -8.067463874816895, -7.537396430969238], "reading_logprob_std": 3.4439477805139, "tokenization_path_logprob_std": 17.46351087454058, "trajectory_balance_rmse": 60.377778640727946} +{"canonical_length": 114, "canonical_nll": 0.6296315778765762, "generation_rouge_l_f1": 0.2674094707520892, "generation_token_f1": 0.456824512534819, "prompt_id": "1b111d554cbb9115", "reading_canonical_log_prob": -71.77799987792969, "reading_distances": [9, 12, 11, 10], "reading_log_probs": [-72.7623291015625, -72.66732788085938, -71.76150512695312, -71.98634338378906], "reading_logprob_std": 0.495617012988001, "tokenization_path_logprob_std": 38.71214674213503, "trajectory_balance_rmse": 1596.467068300357} +{"canonical_length": 135, "canonical_nll": 1.3312375669126157, "generation_rouge_l_f1": 0.2057416267942584, "generation_token_f1": 0.3827751196172249, "prompt_id": "4c468581b163e3ed", "reading_canonical_log_prob": -179.71707153320312, "reading_distances": [19, 23, 22, 28], "reading_log_probs": [-180.33413696289062, -178.23147583007812, -177.58535766601562, -179.100830078125], "reading_logprob_std": 1.189114497457825, "tokenization_path_logprob_std": 266.83300620607656, "trajectory_balance_rmse": 2894.9357349772336} +{"canonical_length": 38, "canonical_nll": 0.5103757757889597, "generation_rouge_l_f1": 0.26190476190476186, "generation_token_f1": 0.2976190476190476, "prompt_id": "5a9339c4c26875b0", "reading_canonical_log_prob": -19.39427947998047, "reading_distances": [20, 17, 15, 19], "reading_log_probs": [-28.449176788330078, -29.655921936035156, -24.577388763427734, -27.957523345947266], "reading_logprob_std": 2.175432260125269, "tokenization_path_logprob_std": 54.88276204054833, "trajectory_balance_rmse": 471.58067535940927} +{"canonical_length": 245, "canonical_nll": 1.2421735491071428, "generation_rouge_l_f1": 0.2783300198807157, "generation_token_f1": 0.5447316103379721, "prompt_id": "dadb86ce514e9684", "reading_canonical_log_prob": -304.33251953125, "reading_distances": [14, 16, 17, 17], "reading_log_probs": [-307.1573486328125, -308.50103759765625, -309.1609191894531, -309.195068359375], "reading_logprob_std": 0.9526447953335265, "tokenization_path_logprob_std": 57.47878509572041, "trajectory_balance_rmse": 5541.179096007137} +{"canonical_length": 153, "canonical_nll": 1.5902527453852635, "generation_rouge_l_f1": 0.20846905537459282, "generation_token_f1": 0.4039087947882737, "prompt_id": "8be6dabc44180a81", "reading_canonical_log_prob": -243.3086700439453, "reading_distances": [21, 22, 22, 24], "reading_log_probs": [-246.42901611328125, -246.08934020996094, -248.16738891601562, -250.2127685546875], "reading_logprob_std": 1.892058576074269, "tokenization_path_logprob_std": 129.25353522501433, "trajectory_balance_rmse": 3628.723973359234} +{"canonical_length": 8, "canonical_nll": 1.7116957902908325, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "5786bd96f280e551", "reading_canonical_log_prob": -13.69356632232666, "reading_distances": [14, 23, 19, 22], "reading_log_probs": [-16.463024139404297, -16.24063491821289, -17.3209171295166, -15.692421913146973], "reading_logprob_std": 0.6769298932201239, "tokenization_path_logprob_std": 28.616651567600506, "trajectory_balance_rmse": 100.5569952825572} +{"canonical_length": 103, "canonical_nll": 1.4280524392729823, "generation_rouge_l_f1": 0.2345679012345679, "generation_token_f1": 0.34567901234567905, "prompt_id": "edcbad1e6d6e267f", "reading_canonical_log_prob": -147.0894012451172, "reading_distances": [20, 28, 20, 24], "reading_log_probs": [-144.89324951171875, -146.42599487304688, -145.99337768554688, -150.31301879882812], "reading_logprob_std": 2.3609479283592028, "tokenization_path_logprob_std": 187.47118904049603, "trajectory_balance_rmse": 2308.3026233116043} +{"canonical_length": 15, "canonical_nll": 2.406737518310547, "generation_rouge_l_f1": 0.2465753424657534, "generation_token_f1": 0.273972602739726, "prompt_id": "f125519b3e50ff0b", "reading_canonical_log_prob": -36.1010627746582, "reading_distances": [14, 14, 15, 14], "reading_log_probs": [-35.97430419921875, -34.27562713623047, -33.7144889831543, -33.60589599609375], "reading_logprob_std": 1.0945636213979681, "tokenization_path_logprob_std": 49.32386868700731, "trajectory_balance_rmse": 335.48675520353055} +{"canonical_length": 335, "canonical_nll": 0.6764401108471315, "generation_rouge_l_f1": 0.23172905525846701, "generation_token_f1": 0.4206773618538324, "prompt_id": "805fd7652d88dc11", "reading_canonical_log_prob": -226.60743713378906, "reading_distances": [24, 26, 23, 25], "reading_log_probs": [-232.43844604492188, -232.65194702148438, -231.4080810546875, -233.1064910888672], "reading_logprob_std": 0.718327244226961, "tokenization_path_logprob_std": 198.32223464948999, "trajectory_balance_rmse": 4984.416252915225} +{"canonical_length": 58, "canonical_nll": 1.4999443580364358, "generation_rouge_l_f1": 0.3146067415730337, "generation_token_f1": 0.449438202247191, "prompt_id": "7efa21552d311157", "reading_canonical_log_prob": -86.99677276611328, "reading_distances": [38, 36, 33, 36], "reading_log_probs": [-84.75601959228516, -84.10061645507812, -85.296142578125, -84.745849609375], "reading_logprob_std": 0.48903139183431044, "tokenization_path_logprob_std": 64.77105576212603, "trajectory_balance_rmse": 1492.4700419842195} +{"canonical_length": 337, "canonical_nll": 1.0344649408266877, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.45107794361525705, "prompt_id": "961872748d95ea69", "reading_canonical_log_prob": -348.61468505859375, "reading_distances": [27, 32, 27, 29], "reading_log_probs": [-350.9202575683594, -349.4285888671875, -350.54345703125, -351.265625], "reading_logprob_std": 0.7971559954638239, "tokenization_path_logprob_std": 281.10209534000523, "trajectory_balance_rmse": 7988.897684670213} +{"canonical_length": 62, "canonical_nll": 0.8076175566642515, "generation_rouge_l_f1": 0.22304832713754644, "generation_token_f1": 0.2899628252788104, "prompt_id": "e754a1aad6ea85da", "reading_canonical_log_prob": -50.072288513183594, "reading_distances": [6, 6, 9, 7], "reading_log_probs": [-49.19367218017578, -50.03994369506836, -50.2943115234375, -48.88958740234375], "reading_logprob_std": 0.6696592691851002, "tokenization_path_logprob_std": 130.59937697721, "trajectory_balance_rmse": 1004.0913683988871} +{"canonical_length": 200, "canonical_nll": 1.8344602966308594, "generation_rouge_l_f1": 0.19067796610169493, "generation_token_f1": 0.3516949152542373, "prompt_id": "4b3ba24fb950898f", "reading_canonical_log_prob": -366.8920593261719, "reading_distances": [42, 39, 42, 44], "reading_log_probs": [-364.2806701660156, -363.24029541015625, -363.2620544433594, -362.12689208984375], "reading_logprob_std": 0.8797467250350409, "tokenization_path_logprob_std": 174.05880584694742, "trajectory_balance_rmse": 4907.93194185812} +{"canonical_length": 33, "canonical_nll": 1.0653553587017637, "generation_rouge_l_f1": 0.15384615384615385, "generation_token_f1": 0.1794871794871795, "prompt_id": "7b02e9bcfb31548c", "reading_canonical_log_prob": -35.1567268371582, "reading_distances": [11, 16, 15, 12], "reading_log_probs": [-37.2099609375, -37.50407028198242, -37.90493392944336, -38.36505126953125], "reading_logprob_std": 0.5014486790473881, "tokenization_path_logprob_std": 63.542412054010136, "trajectory_balance_rmse": 752.1867726579172} +{"canonical_length": 64, "canonical_nll": 0.8017109036445618, "generation_rouge_l_f1": 0.217687074829932, "generation_token_f1": 0.2857142857142857, "prompt_id": "2e9002691e2c9828", "reading_canonical_log_prob": -51.30949783325195, "reading_distances": [33, 45, 41, 34], "reading_log_probs": [-51.40324401855469, -52.08136749267578, -50.778839111328125, -51.602317810058594], "reading_logprob_std": 0.5395638098409534, "tokenization_path_logprob_std": 108.64934313213274, "trajectory_balance_rmse": 1192.5115455624457} +{"canonical_length": 110, "canonical_nll": 1.3457778930664062, "generation_rouge_l_f1": 0.24210526315789474, "generation_token_f1": 0.4052631578947368, "prompt_id": "660c03c6bae9ac45", "reading_canonical_log_prob": -148.0355682373047, "reading_distances": [31, 30, 29, 28], "reading_log_probs": [-144.28663635253906, -143.29931640625, -142.3663787841797, -141.62136840820312], "reading_logprob_std": 1.1549446441071998, "tokenization_path_logprob_std": 162.593816292403, "trajectory_balance_rmse": 2501.72324293801} +{"canonical_length": 129, "canonical_nll": 1.05588501922844, "generation_rouge_l_f1": 0.25287356321839083, "generation_token_f1": 0.41954022988505746, "prompt_id": "dc5df7204109382e", "reading_canonical_log_prob": -136.20916748046875, "reading_distances": [34, 21, 31, 29], "reading_log_probs": [-141.34130859375, -141.6905059814453, -143.42137145996094, -141.08511352539062], "reading_logprob_std": 1.0541478676883345, "tokenization_path_logprob_std": 165.88227108344955, "trajectory_balance_rmse": 2727.648187347446} +{"canonical_length": 330, "canonical_nll": 1.2516781893643465, "generation_rouge_l_f1": 0.3135313531353135, "generation_token_f1": 0.5379537953795379, "prompt_id": "e7afe4dd1ceeee4d", "reading_canonical_log_prob": -413.0538024902344, "reading_distances": [24, 17, 18, 16], "reading_log_probs": [-417.7541809082031, -418.27386474609375, -419.8087463378906, -419.5685119628906], "reading_logprob_std": 0.9946855886813184, "tokenization_path_logprob_std": 173.53110155511075, "trajectory_balance_rmse": 6118.843975764275} +{"canonical_length": 142, "canonical_nll": 0.8647716146119884, "generation_rouge_l_f1": 0.20281690140845068, "generation_token_f1": 0.28169014084507044, "prompt_id": "64cb8edf4fa19a8d", "reading_canonical_log_prob": -122.79756927490234, "reading_distances": [14, 17, 19, 20], "reading_log_probs": [-131.28062438964844, -129.76173400878906, -131.5875244140625, -130.18679809570312], "reading_logprob_std": 0.8695734527291654, "tokenization_path_logprob_std": 148.20292684492875, "trajectory_balance_rmse": 1912.533436476917} +{"canonical_length": 67, "canonical_nll": 1.51964648801889, "generation_rouge_l_f1": 0.16909620991253646, "generation_token_f1": 0.2448979591836735, "prompt_id": "d242cdc5b38c5c82", "reading_canonical_log_prob": -101.81631469726562, "reading_distances": [24, 28, 31, 27], "reading_log_probs": [-101.48153686523438, -101.63040161132812, -99.8546371459961, -100.63128662109375], "reading_logprob_std": 0.8239459677857525, "tokenization_path_logprob_std": 103.77946315991487, "trajectory_balance_rmse": 1487.8615475727797} +{"canonical_length": 71, "canonical_nll": 3.285308837890625, "generation_rouge_l_f1": 0.16883116883116883, "generation_token_f1": 0.22077922077922077, "prompt_id": "406fdefc259c7cdd", "reading_canonical_log_prob": -233.25692749023438, "reading_distances": [32, 31, 32, 35], "reading_log_probs": [-231.73606872558594, -234.4407196044922, -232.7172088623047, -231.83377075195312], "reading_logprob_std": 1.2528119213598754, "tokenization_path_logprob_std": 125.54406317319867, "trajectory_balance_rmse": 1302.5396880734781} +{"canonical_length": 112, "canonical_nll": 1.1545516422816686, "generation_rouge_l_f1": 0.23437499999999997, "generation_token_f1": 0.34375, "prompt_id": "8b97cb84ee36bc33", "reading_canonical_log_prob": -129.30978393554688, "reading_distances": [27, 21, 23, 24], "reading_log_probs": [-129.47964477539062, -128.11891174316406, -128.33758544921875, -130.0004425048828], "reading_logprob_std": 0.9027824557591498, "tokenization_path_logprob_std": 129.699068526538, "trajectory_balance_rmse": 2522.0560431129184} +{"canonical_length": 3, "canonical_nll": 2.6352081298828125, "generation_rouge_l_f1": 0.01556420233463035, "generation_token_f1": 0.01556420233463035, "prompt_id": "3c3cb4aabb9bd15f", "reading_canonical_log_prob": -7.9056243896484375, "reading_distances": [24, 22, 21, 22], "reading_log_probs": [-7.749499320983887, -7.533482074737549, -7.518275260925293, -7.966504096984863], "reading_logprob_std": 0.21131877536198257, "tokenization_path_logprob_std": 16.300843036660844, "trajectory_balance_rmse": 44.77113432441018} +{"canonical_length": 76, "canonical_nll": 1.880769830001028, "generation_rouge_l_f1": 0.19178082191780824, "generation_token_f1": 0.3698630136986301, "prompt_id": "4d20c1324c854a9e", "reading_canonical_log_prob": -142.93850708007812, "reading_distances": [11, 17, 12, 15], "reading_log_probs": [-142.07553100585938, -140.44442749023438, -141.81695556640625, -140.8068389892578], "reading_logprob_std": 0.7838181274747196, "tokenization_path_logprob_std": 195.9645352740281, "trajectory_balance_rmse": 1337.090864905036} +{"canonical_length": 194, "canonical_nll": 0.8892064045384988, "generation_rouge_l_f1": 0.32211538461538464, "generation_token_f1": 0.548076923076923, "prompt_id": "45005e42ada68a08", "reading_canonical_log_prob": -172.50604248046875, "reading_distances": [25, 26, 20, 23], "reading_log_probs": [-170.47811889648438, -165.4114990234375, -168.7102508544922, -168.07992553710938], "reading_logprob_std": 2.1005336514064035, "tokenization_path_logprob_std": 97.87830108625197, "trajectory_balance_rmse": 2529.4616732061722} +{"canonical_length": 3, "canonical_nll": 2.1699318885803223, "generation_rouge_l_f1": 0.017094017094017096, "generation_token_f1": 0.017094017094017096, "prompt_id": "1b11c060139de682", "reading_canonical_log_prob": -6.509795665740967, "reading_distances": [17, 18, 19, 19], "reading_log_probs": [-8.674261093139648, -8.819050788879395, -9.550426483154297, -7.782259941101074], "reading_logprob_std": 0.72574987028013, "tokenization_path_logprob_std": 23.193744005191537, "trajectory_balance_rmse": 40.44757989275386} +{"canonical_length": 180, "canonical_nll": 1.3829270256890192, "generation_rouge_l_f1": 0.24651162790697675, "generation_token_f1": 0.4279069767441861, "prompt_id": "fb27819d14f6e258", "reading_canonical_log_prob": -248.92686462402344, "reading_distances": [18, 20, 19, 19], "reading_log_probs": [-250.90505981445312, -250.24124145507812, -250.21401977539062, -251.2249298095703], "reading_logprob_std": 0.5009017584500994, "tokenization_path_logprob_std": 196.9444668195934, "trajectory_balance_rmse": 4169.032416312104} +{"canonical_length": 92, "canonical_nll": 1.8146435281504756, "generation_rouge_l_f1": 0.17834394904458598, "generation_token_f1": 0.33121019108280253, "prompt_id": "b1a0ce36dec4c1a2", "reading_canonical_log_prob": -166.94720458984375, "reading_distances": [31, 34, 28, 28], "reading_log_probs": [-172.90867614746094, -176.7659912109375, -173.70892333984375, -171.53085327148438], "reading_logprob_std": 2.2157390298660755, "tokenization_path_logprob_std": 160.287256853117, "trajectory_balance_rmse": 1887.56474409049} +{"canonical_length": 205, "canonical_nll": 1.286507657679116, "generation_rouge_l_f1": 0.2021505376344086, "generation_token_f1": 0.3698924731182796, "prompt_id": "fe99f791079b2601", "reading_canonical_log_prob": -263.73406982421875, "reading_distances": [21, 31, 27, 25], "reading_log_probs": [-262.7356872558594, -260.9892272949219, -260.80267333984375, -261.9243469238281], "reading_logprob_std": 0.8950019113842654, "tokenization_path_logprob_std": 182.25493079362857, "trajectory_balance_rmse": 3750.1652994677293} +{"canonical_length": 265, "canonical_nll": 1.8430858684035967, "generation_rouge_l_f1": 0.22053231939163498, "generation_token_f1": 0.4486692015209125, "prompt_id": "7fb35b5d3ec740ba", "reading_canonical_log_prob": -488.4177551269531, "reading_distances": [22, 26, 27, 27], "reading_log_probs": [-486.410400390625, -487.059814453125, -486.85223388671875, -484.026123046875], "reading_logprob_std": 1.4004466440434198, "tokenization_path_logprob_std": 128.1287316676748, "trajectory_balance_rmse": 6685.679580926229} +{"canonical_length": 11, "canonical_nll": 0.41971739855679596, "generation_rouge_l_f1": 0.88, "generation_token_f1": 0.88, "prompt_id": "cb7972227cbb9dc7", "reading_canonical_log_prob": -4.616891384124756, "reading_distances": [29, 19, 21, 21], "reading_log_probs": [-5.179606914520264, -5.219077110290527, -5.251780033111572, -4.342415809631348], "reading_logprob_std": 0.4381974069684745, "tokenization_path_logprob_std": 81.10669958335968, "trajectory_balance_rmse": 262.4261522492235} +{"canonical_length": 109, "canonical_nll": 1.926689497921445, "generation_rouge_l_f1": 0.1370716510903427, "generation_token_f1": 0.2429906542056075, "prompt_id": "ce54506448843c44", "reading_canonical_log_prob": -210.0091552734375, "reading_distances": [22, 21, 20, 20], "reading_log_probs": [-206.70291137695312, -208.7923126220703, -206.37966918945312, -206.13424682617188], "reading_logprob_std": 1.215862406239116, "tokenization_path_logprob_std": 52.510989754911385, "trajectory_balance_rmse": 2039.4316773257485} +{"canonical_length": 90, "canonical_nll": 0.6730166965060764, "generation_rouge_l_f1": 0.2540983606557377, "generation_token_f1": 0.3770491803278688, "prompt_id": "110898cad92a62bf", "reading_canonical_log_prob": -60.571502685546875, "reading_distances": [12, 16, 16, 17], "reading_log_probs": [-61.60014343261719, -63.32349395751953, -63.19310760498047, -63.764373779296875], "reading_logprob_std": 0.9455635682711687, "tokenization_path_logprob_std": 109.64607139376376, "trajectory_balance_rmse": 1158.5267528362406} +{"canonical_length": 6, "canonical_nll": 0.6216837565104166, "generation_rouge_l_f1": 0.09345794392523364, "generation_token_f1": 0.09345794392523364, "prompt_id": "bb43b5a4398228cc", "reading_canonical_log_prob": -3.7301025390625, "reading_distances": [34, 27, 22, 30], "reading_log_probs": [-5.785811901092529, -6.632392406463623, -4.851685523986816, -5.591262340545654], "reading_logprob_std": 0.7319484758185926, "tokenization_path_logprob_std": 22.45980922522088, "trajectory_balance_rmse": 111.0706195792583} +{"canonical_length": 344, "canonical_nll": 1.2918548583984375, "generation_rouge_l_f1": 0.21869488536155204, "generation_token_f1": 0.41269841269841273, "prompt_id": "6dc0804e4c40522b", "reading_canonical_log_prob": -444.3980712890625, "reading_distances": [16, 21, 19, 19], "reading_log_probs": [-450.3147277832031, -448.972412109375, -450.6934814453125, -450.2779846191406], "reading_logprob_std": 0.7519891740571691, "tokenization_path_logprob_std": 202.3288265305175, "trajectory_balance_rmse": 5675.827682966315} +{"canonical_length": 325, "canonical_nll": 1.0926975661057692, "generation_rouge_l_f1": 0.2367066895368782, "generation_token_f1": 0.4391080617495712, "prompt_id": "8ed2c6a41c30f4da", "reading_canonical_log_prob": -355.126708984375, "reading_distances": [30, 35, 29, 23], "reading_log_probs": [-358.166015625, -358.7875061035156, -358.9949951171875, -358.00701904296875], "reading_logprob_std": 0.47671145991355607, "tokenization_path_logprob_std": 63.451068840789034, "trajectory_balance_rmse": 7634.989121440628} +{"canonical_length": 116, "canonical_nll": 0.7412006115091259, "generation_rouge_l_f1": 0.31176470588235294, "generation_token_f1": 0.4235294117647059, "prompt_id": "b2cdbde556d30889", "reading_canonical_log_prob": -85.9792709350586, "reading_distances": [15, 15, 13, 12], "reading_log_probs": [-85.84851837158203, -85.81919860839844, -84.59565734863281, -85.52594757080078], "reading_logprob_std": 0.5861641973967594, "tokenization_path_logprob_std": 139.99136771547083, "trajectory_balance_rmse": 1509.9632786262587} +{"canonical_length": 15, "canonical_nll": 1.5296780904134115, "generation_rouge_l_f1": 0.4347826086956522, "generation_token_f1": 0.5217391304347826, "prompt_id": "eb54d1c2e52f2aa1", "reading_canonical_log_prob": -22.945171356201172, "reading_distances": [21, 23, 21, 19], "reading_log_probs": [-23.929590225219727, -23.411344528198242, -23.494298934936523, -23.835603713989258], "reading_logprob_std": 0.253353642283628, "tokenization_path_logprob_std": 38.01692850971864, "trajectory_balance_rmse": 328.1212783315276} +{"canonical_length": 61, "canonical_nll": 1.4373826824250768, "generation_rouge_l_f1": 0.11578947368421053, "generation_token_f1": 0.23157894736842105, "prompt_id": "2d9f537706729121", "reading_canonical_log_prob": -87.68034362792969, "reading_distances": [11, 15, 10, 10], "reading_log_probs": [-85.520263671875, -84.91487121582031, -86.26189422607422, -85.73710632324219], "reading_logprob_std": 0.5574854665677481, "tokenization_path_logprob_std": 64.89951885858068, "trajectory_balance_rmse": 1281.1417916105343} +{"canonical_length": 2, "canonical_nll": 4.503090858459473, "generation_rouge_l_f1": 0.0078125, "generation_token_f1": 0.0078125, "prompt_id": "e6a02b338700cb67", "reading_canonical_log_prob": -9.006181716918945, "reading_distances": [22, 19, 25, 21], "reading_log_probs": [-8.757791519165039, -9.369111061096191, -9.259462356567383, -9.274588584899902], "reading_logprob_std": 0.27593032911312637, "tokenization_path_logprob_std": 24.834702014223172, "trajectory_balance_rmse": 47.418264886139205} +{"canonical_length": 169, "canonical_nll": 0.8421167678381565, "generation_rouge_l_f1": 0.28490028490028496, "generation_token_f1": 0.47293447293447294, "prompt_id": "56062aaca5480bb3", "reading_canonical_log_prob": -142.31773376464844, "reading_distances": [27, 24, 27, 23], "reading_log_probs": [-142.75787353515625, -143.0045166015625, -143.20452880859375, -141.77011108398438], "reading_logprob_std": 0.6362209043054702, "tokenization_path_logprob_std": 115.2796957513867, "trajectory_balance_rmse": 2176.229173024556} +{"canonical_length": 3, "canonical_nll": 0.5439906120300293, "generation_rouge_l_f1": 0.021660649819494584, "generation_token_f1": 0.021660649819494584, "prompt_id": "a6378b47a0666677", "reading_canonical_log_prob": -1.631971836090088, "reading_distances": [29, 31, 27, 30], "reading_log_probs": [-2.8243541717529297, -4.260770320892334, -3.2240607738494873, -4.124848365783691], "reading_logprob_std": 0.6963596782894559, "tokenization_path_logprob_std": 24.339472001040342, "trajectory_balance_rmse": 64.44873299462957} +{"canonical_length": 199, "canonical_nll": 1.0287653477347676, "generation_rouge_l_f1": 0.2723214285714286, "generation_token_f1": 0.53125, "prompt_id": "41d1448457af2a4f", "reading_canonical_log_prob": -204.72430419921875, "reading_distances": [13, 18, 21, 16], "reading_log_probs": [-205.38876342773438, -202.64306640625, -205.003173828125, -203.02320861816406], "reading_logprob_std": 1.381974987922489, "tokenization_path_logprob_std": 88.06969317965152, "trajectory_balance_rmse": 4556.074742120464} +{"canonical_length": 94, "canonical_nll": 1.8458501937541556, "generation_rouge_l_f1": 0.18579234972677597, "generation_token_f1": 0.2677595628415301, "prompt_id": "5749ffa6095e5a40", "reading_canonical_log_prob": -173.50991821289062, "reading_distances": [31, 23, 22, 19], "reading_log_probs": [-170.50430297851562, -170.203857421875, -171.71502685546875, -173.2130889892578], "reading_logprob_std": 1.3686045000466713, "tokenization_path_logprob_std": 111.52242161169515, "trajectory_balance_rmse": 2244.7725097742555} +{"canonical_length": 58, "canonical_nll": 1.1813365015490302, "generation_rouge_l_f1": 0.48484848484848486, "generation_token_f1": 0.5656565656565656, "prompt_id": "e503fe8212044208", "reading_canonical_log_prob": -68.51751708984375, "reading_distances": [31, 31, 34, 34], "reading_log_probs": [-69.21698760986328, -68.60000610351562, -69.77059173583984, -67.80377960205078], "reading_logprob_std": 0.8444375798264286, "tokenization_path_logprob_std": 91.84281519093066, "trajectory_balance_rmse": 1462.5636657708883} +{"canonical_length": 161, "canonical_nll": 0.9792929702664014, "generation_rouge_l_f1": 0.325, "generation_token_f1": 0.55, "prompt_id": "cf2e46f2d41b486e", "reading_canonical_log_prob": -157.66616821289062, "reading_distances": [25, 21, 22, 22], "reading_log_probs": [-158.97885131835938, -160.51083374023438, -158.00091552734375, -158.73716735839844], "reading_logprob_std": 1.054726991412635, "tokenization_path_logprob_std": 135.27211651560518, "trajectory_balance_rmse": 3145.9379511966886} +{"canonical_length": 62, "canonical_nll": 0.7014582234044229, "generation_rouge_l_f1": 0.18497109826589594, "generation_token_f1": 0.24277456647398846, "prompt_id": "45e020ec1f1a052d", "reading_canonical_log_prob": -43.49040985107422, "reading_distances": [12, 12, 12, 18], "reading_log_probs": [-44.1637077331543, -45.569252014160156, -44.80534362792969, -45.68318176269531], "reading_logprob_std": 0.7108204458455865, "tokenization_path_logprob_std": 36.049460689603826, "trajectory_balance_rmse": 1075.9500050209833} +{"canonical_length": 284, "canonical_nll": 1.481077274806063, "generation_rouge_l_f1": 0.20817843866171004, "generation_token_f1": 0.39776951672862454, "prompt_id": "238ad0e8d97d1bdf", "reading_canonical_log_prob": -420.6259460449219, "reading_distances": [13, 19, 17, 26], "reading_log_probs": [-420.13861083984375, -417.31878662109375, -422.11083984375, -420.8629150390625], "reading_logprob_std": 2.029935325149592, "tokenization_path_logprob_std": 295.3298449080645, "trajectory_balance_rmse": 5654.5658240841} +{"canonical_length": 30, "canonical_nll": 0.7256017049153646, "generation_rouge_l_f1": 0.06106870229007634, "generation_token_f1": 0.0916030534351145, "prompt_id": "aca807bc21ebbe28", "reading_canonical_log_prob": -21.768051147460938, "reading_distances": [9, 11, 9, 11], "reading_log_probs": [-19.924192428588867, -20.33416175842285, -17.36450958251953, -21.241657257080078], "reading_logprob_std": 1.6615861125858586, "tokenization_path_logprob_std": 18.512084420435475, "trajectory_balance_rmse": 402.1972859613312} +{"canonical_length": 39, "canonical_nll": 0.8106153439252805, "generation_rouge_l_f1": 0.24, "generation_token_f1": 0.2866666666666667, "prompt_id": "1c53f5181f02d61c", "reading_canonical_log_prob": -31.613998413085938, "reading_distances": [21, 21, 21, 22], "reading_log_probs": [-31.688697814941406, -30.51422119140625, -31.405364990234375, -31.917835235595703], "reading_logprob_std": 0.615025215875164, "tokenization_path_logprob_std": 40.74292557994715, "trajectory_balance_rmse": 481.1836108211318} +{"canonical_length": 10, "canonical_nll": 2.7842439651489257, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.22222222222222224, "prompt_id": "1a83cc65795284e0", "reading_canonical_log_prob": -27.842439651489258, "reading_distances": [21, 14, 22, 17], "reading_log_probs": [-28.47153091430664, -29.44921112060547, -30.426546096801758, -31.468830108642578], "reading_logprob_std": 1.2871851033624182, "tokenization_path_logprob_std": 58.32239094618375, "trajectory_balance_rmse": 133.8518086733224} +{"canonical_length": 30, "canonical_nll": 1.6765352884928386, "generation_rouge_l_f1": 0.4266666666666666, "generation_token_f1": 0.48, "prompt_id": "a5ddf34551bf57f2", "reading_canonical_log_prob": -50.296058654785156, "reading_distances": [24, 20, 18, 17], "reading_log_probs": [-47.93980407714844, -48.430877685546875, -48.274749755859375, -48.333595275878906], "reading_logprob_std": 0.2132518068770369, "tokenization_path_logprob_std": 89.22523819456609, "trajectory_balance_rmse": 838.983626504051} +{"canonical_length": 33, "canonical_nll": 0.9876854636452415, "generation_rouge_l_f1": 0.088135593220339, "generation_token_f1": 0.1152542372881356, "prompt_id": "626308b96b881737", "reading_canonical_log_prob": -32.59362030029297, "reading_distances": [22, 18, 14, 14], "reading_log_probs": [-32.06796646118164, -31.305908203125, -32.47102737426758, -31.3969783782959], "reading_logprob_std": 0.5562368800866284, "tokenization_path_logprob_std": 52.37503586515931, "trajectory_balance_rmse": 722.7819715328632} +{"canonical_length": 79, "canonical_nll": 0.9026301903060719, "generation_rouge_l_f1": 0.12820512820512822, "generation_token_f1": 0.17094017094017094, "prompt_id": "b884d1bbe7793313", "reading_canonical_log_prob": -71.30778503417969, "reading_distances": [19, 23, 24, 21], "reading_log_probs": [-71.34602355957031, -71.02279663085938, -70.69744873046875, -71.8524398803711], "reading_logprob_std": 0.4924214133970276, "tokenization_path_logprob_std": 43.179957310996755, "trajectory_balance_rmse": 997.1310469809611} +{"canonical_length": 50, "canonical_nll": 1.6844540405273438, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.46666666666666673, "prompt_id": "d8395ff4ce8fbfae", "reading_canonical_log_prob": -84.22270202636719, "reading_distances": [38, 31, 29, 29], "reading_log_probs": [-88.18439483642578, -86.5652084350586, -85.7428207397461, -86.05931091308594], "reading_logprob_std": 1.085182404296903, "tokenization_path_logprob_std": 128.85214248334648, "trajectory_balance_rmse": 1153.8675796155999} +{"canonical_length": 228, "canonical_nll": 1.4216599046138294, "generation_rouge_l_f1": 0.2510121457489879, "generation_token_f1": 0.4817813765182186, "prompt_id": "cabaeed82d3c2f7b", "reading_canonical_log_prob": -324.1384582519531, "reading_distances": [24, 21, 27, 23], "reading_log_probs": [-322.5471496582031, -322.3363952636719, -321.7958984375, -323.3512878417969], "reading_logprob_std": 0.6452915357269206, "tokenization_path_logprob_std": 207.99811036995973, "trajectory_balance_rmse": 5444.60754395059} +{"canonical_length": 60, "canonical_nll": 0.03827650149663289, "generation_rouge_l_f1": 0.6486486486486486, "generation_token_f1": 0.954954954954955, "prompt_id": "ef7598410b16ef36", "reading_canonical_log_prob": -2.2965900897979736, "reading_distances": [32, 36, 31, 32], "reading_log_probs": [-1.8401378393173218, -4.74761962890625, -2.7339916229248047, -2.9788196086883545], "reading_logprob_std": 1.2176587205154294, "tokenization_path_logprob_std": 123.58828277194506, "trajectory_balance_rmse": 1110.2154573505304} +{"canonical_length": 239, "canonical_nll": 1.2471905951719404, "generation_rouge_l_f1": 0.26970954356846477, "generation_token_f1": 0.4896265560165975, "prompt_id": "39157986b3a30c4b", "reading_canonical_log_prob": -298.07855224609375, "reading_distances": [42, 44, 40, 40], "reading_log_probs": [-294.1251220703125, -296.3516540527344, -298.734375, -297.43695068359375], "reading_logprob_std": 1.9516959238190774, "tokenization_path_logprob_std": 104.35232641253634, "trajectory_balance_rmse": 4541.169991857424} +{"canonical_length": 113, "canonical_nll": 1.6596764758624862, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "952e6be856268bb7", "reading_canonical_log_prob": -187.54344177246094, "reading_distances": [25, 27, 29, 33], "reading_log_probs": [-186.29116821289062, -188.00765991210938, -186.42770385742188, -186.11380004882812], "reading_logprob_std": 0.8745449825899163, "tokenization_path_logprob_std": 70.10907075689978, "trajectory_balance_rmse": 2357.169093630192} +{"canonical_length": 324, "canonical_nll": 0.9153328412844811, "generation_rouge_l_f1": 0.23853211009174313, "generation_token_f1": 0.44036697247706413, "prompt_id": "b30cec3b9127ceb8", "reading_canonical_log_prob": -296.5678405761719, "reading_distances": [24, 19, 21, 20], "reading_log_probs": [-296.9700927734375, -295.2458190917969, -295.75421142578125, -296.3161926269531], "reading_logprob_std": 0.7415669065587144, "tokenization_path_logprob_std": 239.28115017633812, "trajectory_balance_rmse": 6827.824613121965} +{"canonical_length": 211, "canonical_nll": 1.1787994529398698, "generation_rouge_l_f1": 0.22573363431151242, "generation_token_f1": 0.4063205417607224, "prompt_id": "ce18c132bb5bd7f2", "reading_canonical_log_prob": -248.7266845703125, "reading_distances": [25, 28, 28, 24], "reading_log_probs": [-246.58120727539062, -246.10812377929688, -246.15960693359375, -247.56149291992188], "reading_logprob_std": 0.6734693435310772, "tokenization_path_logprob_std": 364.80477282536077, "trajectory_balance_rmse": 4378.551317900042} +{"canonical_length": 53, "canonical_nll": 1.382719651708063, "generation_rouge_l_f1": 0.10876132930513595, "generation_token_f1": 0.1691842900302115, "prompt_id": "f1a703b446eae2b2", "reading_canonical_log_prob": -73.28414154052734, "reading_distances": [54, 59, 45, 56], "reading_log_probs": [-74.84776306152344, -72.9175033569336, -73.7155532836914, -74.27596282958984], "reading_logprob_std": 0.8231601553870901, "tokenization_path_logprob_std": 115.9431511291882, "trajectory_balance_rmse": 1016.812647814877} +{"canonical_length": 56, "canonical_nll": 1.7122482572283064, "generation_rouge_l_f1": 0.23622047244094488, "generation_token_f1": 0.4566929133858268, "prompt_id": "7188e613a859c6bb", "reading_canonical_log_prob": -95.88590240478516, "reading_distances": [26, 25, 28, 24], "reading_log_probs": [-94.43159484863281, -91.74493408203125, -93.01683044433594, -97.5911865234375], "reading_logprob_std": 2.5153502978290208, "tokenization_path_logprob_std": 92.51998484059891, "trajectory_balance_rmse": 1073.8048478542717} +{"canonical_length": 198, "canonical_nll": 1.5939844304865056, "generation_rouge_l_f1": 0.25968109339407747, "generation_token_f1": 0.4738041002277904, "prompt_id": "85df6700a3ed9560", "reading_canonical_log_prob": -315.6089172363281, "reading_distances": [40, 30, 32, 41], "reading_log_probs": [-317.0246276855469, -313.5686950683594, -314.3338928222656, -318.39251708984375], "reading_logprob_std": 2.261670901942487, "tokenization_path_logprob_std": 94.22448280094605, "trajectory_balance_rmse": 4061.9778885324995} +{"canonical_length": 94, "canonical_nll": 1.225603793529754, "generation_rouge_l_f1": 0.18732782369146006, "generation_token_f1": 0.3360881542699724, "prompt_id": "0baa5b0629968fb8", "reading_canonical_log_prob": -115.20675659179688, "reading_distances": [26, 24, 29, 21], "reading_log_probs": [-117.79397583007812, -116.58235168457031, -116.4205093383789, -117.67292785644531], "reading_logprob_std": 0.7160771235114174, "tokenization_path_logprob_std": 44.157243572293446, "trajectory_balance_rmse": 2338.8218776879316} +{"canonical_length": 331, "canonical_nll": 1.1664727432850264, "generation_rouge_l_f1": 0.25631768953068595, "generation_token_f1": 0.4584837545126354, "prompt_id": "3d704d3c66ac7a28", "reading_canonical_log_prob": -386.10247802734375, "reading_distances": [25, 32, 31, 32], "reading_log_probs": [-383.1753845214844, -382.35345458984375, -381.41510009765625, -382.8498840332031], "reading_logprob_std": 0.76733722052932, "tokenization_path_logprob_std": 164.80016640881928, "trajectory_balance_rmse": 6033.015425836189} +{"canonical_length": 14, "canonical_nll": 0.5969912665230888, "generation_rouge_l_f1": 0.5454545454545454, "generation_token_f1": 0.5454545454545454, "prompt_id": "d990b55cc5500b31", "reading_canonical_log_prob": -8.357877731323242, "reading_distances": [59, 60, 60, 60], "reading_log_probs": [-8.96672248840332, -7.717412948608398, -8.964180946350098, -8.87835693359375], "reading_logprob_std": 0.6108857227124894, "tokenization_path_logprob_std": 46.887645290487136, "trajectory_balance_rmse": 333.5455372587167} +{"canonical_length": 125, "canonical_nll": 1.6934378662109375, "generation_rouge_l_f1": 0.17877094972067037, "generation_token_f1": 0.3128491620111732, "prompt_id": "96948513433b0e29", "reading_canonical_log_prob": -211.6797332763672, "reading_distances": [14, 19, 14, 18], "reading_log_probs": [-211.207763671875, -211.6348876953125, -211.08724975585938, -210.6031951904297], "reading_logprob_std": 0.4243688591878837, "tokenization_path_logprob_std": 161.06891410014478, "trajectory_balance_rmse": 2943.061887077468} +{"canonical_length": 88, "canonical_nll": 1.6827812194824219, "generation_rouge_l_f1": 0.2938388625592417, "generation_token_f1": 0.4170616113744076, "prompt_id": "5b5a0a9c17c96287", "reading_canonical_log_prob": -148.08474731445312, "reading_distances": [43, 45, 35, 38], "reading_log_probs": [-155.23587036132812, -155.82838439941406, -156.60630798339844, -155.00057983398438], "reading_logprob_std": 0.7160843865808447, "tokenization_path_logprob_std": 131.67033926252037, "trajectory_balance_rmse": 2400.3994306471895} +{"canonical_length": 310, "canonical_nll": 1.3003084244266634, "generation_rouge_l_f1": 0.26987060998151574, "generation_token_f1": 0.5249537892791127, "prompt_id": "9ae64366f18d52ad", "reading_canonical_log_prob": -403.0956115722656, "reading_distances": [20, 27, 15, 21], "reading_log_probs": [-403.22711181640625, -399.9850158691406, -402.35943603515625, -402.29132080078125], "reading_logprob_std": 1.3874828967478587, "tokenization_path_logprob_std": 189.03087678168146, "trajectory_balance_rmse": 7078.038013047847} +{"canonical_length": 192, "canonical_nll": 1.3627996444702148, "generation_rouge_l_f1": 0.17011494252873563, "generation_token_f1": 0.4183908045977012, "prompt_id": "eaabe2e9ef0c1d39", "reading_canonical_log_prob": -261.65753173828125, "reading_distances": [28, 29, 31, 35], "reading_log_probs": [-272.95086669921875, -271.841796875, -269.56414794921875, -272.155029296875], "reading_logprob_std": 1.4529201134962264, "tokenization_path_logprob_std": 296.83752800117566, "trajectory_balance_rmse": 4214.364960511915} +{"canonical_length": 208, "canonical_nll": 1.3753984891451323, "generation_rouge_l_f1": 0.3578431372549019, "generation_token_f1": 0.553921568627451, "prompt_id": "74426ee6c2f5003e", "reading_canonical_log_prob": -286.0828857421875, "reading_distances": [20, 21, 28, 22], "reading_log_probs": [-282.70068359375, -283.8288879394531, -283.36236572265625, -284.3526306152344], "reading_logprob_std": 0.701912338653052, "tokenization_path_logprob_std": 133.9853401304282, "trajectory_balance_rmse": 4774.499372874049} +{"canonical_length": 341, "canonical_nll": 1.282621590622709, "generation_rouge_l_f1": 0.30375939849624056, "generation_token_f1": 0.5172932330827067, "prompt_id": "6909a2a1a6b2d043", "reading_canonical_log_prob": -437.37396240234375, "reading_distances": [36, 44, 34, 33], "reading_log_probs": [-439.7613525390625, -443.3399658203125, -442.7459411621094, -442.99420166015625], "reading_logprob_std": 1.6507474854645077, "tokenization_path_logprob_std": 307.65381967279836, "trajectory_balance_rmse": 5644.743306109596} +{"canonical_length": 19, "canonical_nll": 1.7732837074681331, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "fcaeb274bce92488", "reading_canonical_log_prob": -33.69239044189453, "reading_distances": [24, 30, 32, 34], "reading_log_probs": [-37.785247802734375, -41.148895263671875, -34.88779830932617, -41.13389587402344], "reading_logprob_std": 3.0157651952530315, "tokenization_path_logprob_std": 27.396449738086027, "trajectory_balance_rmse": 337.44967580762125} +{"canonical_length": 20, "canonical_nll": 0.6020589828491211, "generation_rouge_l_f1": 0.1774193548387097, "generation_token_f1": 0.1935483870967742, "prompt_id": "59d62ad713859ec4", "reading_canonical_log_prob": -12.041179656982422, "reading_distances": [13, 15, 14, 16], "reading_log_probs": [-13.988241195678711, -13.181041717529297, -12.75259017944336, -13.811911582946777], "reading_logprob_std": 0.5710535517266064, "tokenization_path_logprob_std": 44.44050788243532, "trajectory_balance_rmse": 192.71887094199354} +{"canonical_length": 227, "canonical_nll": 1.3446467059299285, "generation_rouge_l_f1": 0.19631901840490798, "generation_token_f1": 0.4212678936605317, "prompt_id": "0205b5c76a68ce56", "reading_canonical_log_prob": -305.23480224609375, "reading_distances": [32, 25, 25, 26], "reading_log_probs": [-310.227294921875, -306.1378173828125, -312.6470947265625, -307.3209228515625], "reading_logprob_std": 2.9320753909480675, "tokenization_path_logprob_std": 245.70729326538327, "trajectory_balance_rmse": 4170.999334529597} +{"canonical_length": 251, "canonical_nll": 1.460833667283989, "generation_rouge_l_f1": 0.23274161735700197, "generation_token_f1": 0.5207100591715976, "prompt_id": "28e138f0fd885003", "reading_canonical_log_prob": -366.66925048828125, "reading_distances": [23, 25, 29, 28], "reading_log_probs": [-369.5150146484375, -366.41802978515625, -367.6085510253906, -364.58123779296875], "reading_logprob_std": 2.072113331483971, "tokenization_path_logprob_std": 195.24494419413676, "trajectory_balance_rmse": 4794.465225865701} +{"canonical_length": 25, "canonical_nll": 0.4681684112548828, "generation_rouge_l_f1": 0.8800000000000001, "generation_token_f1": 0.8800000000000001, "prompt_id": "3cdea1f3e20549fa", "reading_canonical_log_prob": -11.70421028137207, "reading_distances": [31, 36, 33, 37], "reading_log_probs": [-10.45977783203125, -9.632486343383789, -10.506257057189941, -9.143510818481445], "reading_logprob_std": 0.6632483202574656, "tokenization_path_logprob_std": 38.99480355242837, "trajectory_balance_rmse": 466.6182183723648} +{"canonical_length": 143, "canonical_nll": 2.2415028818837412, "generation_rouge_l_f1": 0.162291169451074, "generation_token_f1": 0.2959427207637232, "prompt_id": "b7ff5b60544bcb0d", "reading_canonical_log_prob": -320.534912109375, "reading_distances": [17, 16, 16, 18], "reading_log_probs": [-330.17987060546875, -334.8846435546875, -336.02801513671875, -328.53961181640625], "reading_logprob_std": 3.6132746283582207, "tokenization_path_logprob_std": 143.82101018106712, "trajectory_balance_rmse": 3010.284504707401} +{"canonical_length": 297, "canonical_nll": 1.3307022454361321, "generation_rouge_l_f1": 0.2179732313575526, "generation_token_f1": 0.4168260038240917, "prompt_id": "2c96188d7d0e9844", "reading_canonical_log_prob": -395.21856689453125, "reading_distances": [20, 19, 18, 15], "reading_log_probs": [-395.8830871582031, -393.97412109375, -392.9064636230469, -394.8565368652344], "reading_logprob_std": 1.2675297605869127, "tokenization_path_logprob_std": 193.72625874584517, "trajectory_balance_rmse": 5277.001441542966} +{"canonical_length": 101, "canonical_nll": 1.0542641819113552, "generation_rouge_l_f1": 0.34441087613293053, "generation_token_f1": 0.4531722054380664, "prompt_id": "f320a0730139386c", "reading_canonical_log_prob": -106.48068237304688, "reading_distances": [35, 41, 38, 37], "reading_log_probs": [-109.07771301269531, -108.98846435546875, -108.40170288085938, -109.14970397949219], "reading_logprob_std": 0.34155656052582967, "tokenization_path_logprob_std": 175.15596992985758, "trajectory_balance_rmse": 1652.4218281481305} +{"canonical_length": 109, "canonical_nll": 1.9284046418076262, "generation_rouge_l_f1": 0.20118343195266272, "generation_token_f1": 0.41420118343195267, "prompt_id": "2a4224ce9ad0c001", "reading_canonical_log_prob": -210.19610595703125, "reading_distances": [17, 13, 19, 14], "reading_log_probs": [-215.00437927246094, -221.03518676757812, -224.0552978515625, -214.05125427246094], "reading_logprob_std": 4.806033407974768, "tokenization_path_logprob_std": 128.12411045130077, "trajectory_balance_rmse": 2227.357021506014} +{"canonical_length": 20, "canonical_nll": 0.5270429611206054, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5833333333333334, "prompt_id": "af538126b4c7deeb", "reading_canonical_log_prob": -10.54085922241211, "reading_distances": [27, 37, 35, 32], "reading_log_probs": [-12.213939666748047, -13.45311164855957, -13.480108261108398, -13.313056945800781], "reading_logprob_std": 0.6051886194790447, "tokenization_path_logprob_std": 33.998901413732305, "trajectory_balance_rmse": 355.9444954783673} +{"canonical_length": 48, "canonical_nll": 1.5930792490641277, "generation_rouge_l_f1": 0.18378378378378377, "generation_token_f1": 0.3567567567567568, "prompt_id": "6c3b3c80c0d299e0", "reading_canonical_log_prob": -76.46780395507812, "reading_distances": [29, 23, 19, 32], "reading_log_probs": [-71.83950805664062, -73.70252227783203, -71.8785171508789, -71.29730224609375], "reading_logprob_std": 1.0494530205681876, "tokenization_path_logprob_std": 53.43048414714685, "trajectory_balance_rmse": 1157.8964629637155} +{"canonical_length": 131, "canonical_nll": 1.3417036915553435, "generation_rouge_l_f1": 0.24615384615384617, "generation_token_f1": 0.46153846153846156, "prompt_id": "783057aa1290e5ce", "reading_canonical_log_prob": -175.76318359375, "reading_distances": [35, 27, 24, 27], "reading_log_probs": [-176.58668518066406, -175.15200805664062, -177.69345092773438, -171.62310791015625], "reading_logprob_std": 2.640730022981906, "tokenization_path_logprob_std": 243.86065879607364, "trajectory_balance_rmse": 2933.966882736896} +{"canonical_length": 358, "canonical_nll": 1.5373313520207752, "generation_rouge_l_f1": 0.21247892074198987, "generation_token_f1": 0.42158516020236086, "prompt_id": "609ea42a956a4ffa", "reading_canonical_log_prob": -550.3646240234375, "reading_distances": [13, 18, 13, 14], "reading_log_probs": [-544.0768432617188, -545.3800659179688, -545.5286865234375, -545.37451171875], "reading_logprob_std": 0.6792195506998331, "tokenization_path_logprob_std": 187.09397333498285, "trajectory_balance_rmse": 6755.510670890577} +{"canonical_length": 185, "canonical_nll": 1.0237441604201858, "generation_rouge_l_f1": 0.4092071611253197, "generation_token_f1": 0.6291560102301791, "prompt_id": "0fd242df3fc4b686", "reading_canonical_log_prob": -189.39266967773438, "reading_distances": [21, 23, 25, 23], "reading_log_probs": [-188.49977111816406, -188.6080322265625, -187.02474975585938, -189.36831665039062], "reading_logprob_std": 0.9797491772851806, "tokenization_path_logprob_std": 95.82559261078649, "trajectory_balance_rmse": 3963.382082582223} +{"canonical_length": 78, "canonical_nll": 0.7187477014003656, "generation_rouge_l_f1": 0.30303030303030304, "generation_token_f1": 0.41414141414141414, "prompt_id": "c7ad7826ab1191ed", "reading_canonical_log_prob": -56.062320709228516, "reading_distances": [33, 30, 29, 24], "reading_log_probs": [-54.23053741455078, -54.23108673095703, -55.5008544921875, -55.489898681640625], "reading_logprob_std": 0.7301103971865576, "tokenization_path_logprob_std": 101.72727625517508, "trajectory_balance_rmse": 1540.2389610304122} +{"canonical_length": 9, "canonical_nll": 1.6047488318549261, "generation_rouge_l_f1": 0.10526315789473685, "generation_token_f1": 0.10526315789473685, "prompt_id": "f4e51cd8f0bc2528", "reading_canonical_log_prob": -14.442739486694336, "reading_distances": [30, 27, 27, 24], "reading_log_probs": [-17.612979888916016, -17.878841400146484, -18.409849166870117, -18.330326080322266], "reading_logprob_std": 0.3777565500947928, "tokenization_path_logprob_std": 52.439088919670795, "trajectory_balance_rmse": 194.02860827140566} +{"canonical_length": 110, "canonical_nll": 1.2816144076260654, "generation_rouge_l_f1": 0.45783132530120485, "generation_token_f1": 0.5060240963855421, "prompt_id": "67c67cc47e6c2592", "reading_canonical_log_prob": -140.9775848388672, "reading_distances": [20, 31, 25, 28], "reading_log_probs": [-141.49484252929688, -141.80233764648438, -143.4459686279297, -143.4958953857422], "reading_logprob_std": 1.059788191678758, "tokenization_path_logprob_std": 74.53062661810452, "trajectory_balance_rmse": 1880.8212993810262} +{"canonical_length": 18, "canonical_nll": 1.0016746520996094, "generation_rouge_l_f1": 0.7333333333333334, "generation_token_f1": 0.7333333333333334, "prompt_id": "82aeadf42ab79a24", "reading_canonical_log_prob": -18.03014373779297, "reading_distances": [41, 46, 37, 40], "reading_log_probs": [-20.01367950439453, -19.328659057617188, -18.92024040222168, -18.155956268310547], "reading_logprob_std": 0.7768615048182063, "tokenization_path_logprob_std": 24.142045923157823, "trajectory_balance_rmse": 431.36456284579134} +{"canonical_length": 12, "canonical_nll": 0.16939171155293783, "generation_rouge_l_f1": 0.5555555555555556, "generation_token_f1": 0.888888888888889, "prompt_id": "736b52db67801c89", "reading_canonical_log_prob": -2.032700538635254, "reading_distances": [24, 21, 27, 28], "reading_log_probs": [-4.052457809448242, -3.1037979125976562, -1.9710789918899536, -2.192972183227539], "reading_logprob_std": 0.9509500699812491, "tokenization_path_logprob_std": 34.15508699643304, "trajectory_balance_rmse": 227.5359855214191} +{"canonical_length": 29, "canonical_nll": 0.7775309332485857, "generation_rouge_l_f1": 0.4666666666666667, "generation_token_f1": 0.5, "prompt_id": "3c9993234c0b5914", "reading_canonical_log_prob": -22.548397064208984, "reading_distances": [21, 21, 19, 29], "reading_log_probs": [-22.56182098388672, -23.492910385131836, -22.903974533081055, -22.874019622802734], "reading_logprob_std": 0.3886121700632417, "tokenization_path_logprob_std": 56.85980986263004, "trajectory_balance_rmse": 731.2609412966019} +{"canonical_length": 221, "canonical_nll": 1.0703486792102659, "generation_rouge_l_f1": 0.21181262729124237, "generation_token_f1": 0.48065173116089616, "prompt_id": "1560e8f9b6710dda", "reading_canonical_log_prob": -236.54705810546875, "reading_distances": [14, 14, 16, 10], "reading_log_probs": [-236.626953125, -235.46115112304688, -237.396728515625, -234.59219360351562], "reading_logprob_std": 1.2402572937597425, "tokenization_path_logprob_std": 179.88724802607604, "trajectory_balance_rmse": 4661.680903226516} +{"canonical_length": 184, "canonical_nll": 1.7252251998237942, "generation_rouge_l_f1": 0.21315192743764172, "generation_token_f1": 0.3083900226757369, "prompt_id": "1073dfb1d10b700d", "reading_canonical_log_prob": -317.4414367675781, "reading_distances": [15, 18, 21, 20], "reading_log_probs": [-314.91094970703125, -314.9851989746094, -315.8543701171875, -315.67047119140625], "reading_logprob_std": 0.4770834317644943, "tokenization_path_logprob_std": 113.12551816141946, "trajectory_balance_rmse": 4315.304604816223} +{"canonical_length": 67, "canonical_nll": 1.370326711170709, "generation_rouge_l_f1": 0.20289855072463767, "generation_token_f1": 0.4927536231884058, "prompt_id": "2045b4674c0f9cd0", "reading_canonical_log_prob": -91.8118896484375, "reading_distances": [29, 22, 24, 21], "reading_log_probs": [-91.211669921875, -90.9241714477539, -91.17271423339844, -91.87109375], "reading_logprob_std": 0.40467909255166695, "tokenization_path_logprob_std": 171.97580371889612, "trajectory_balance_rmse": 1605.8547912926022} +{"canonical_length": 18, "canonical_nll": 0.7177205615573459, "generation_rouge_l_f1": 0.4878048780487805, "generation_token_f1": 0.5365853658536585, "prompt_id": "e0eb2633a5eb27ba", "reading_canonical_log_prob": -12.918970108032227, "reading_distances": [17, 23, 24, 27], "reading_log_probs": [-13.89398193359375, -13.249222755432129, -13.81756591796875, -13.716706275939941], "reading_logprob_std": 0.2893538669465305, "tokenization_path_logprob_std": 44.51327904183628, "trajectory_balance_rmse": 510.9011050362652} +{"canonical_length": 10, "canonical_nll": 1.718442153930664, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.5333333333333333, "prompt_id": "d46e9e081fc93a1e", "reading_canonical_log_prob": -17.18442153930664, "reading_distances": [15, 16, 10, 18], "reading_log_probs": [-15.778467178344727, -16.4310245513916, -16.539154052734375, -16.475202560424805], "reading_logprob_std": 0.3544537249463601, "tokenization_path_logprob_std": 45.33284361161793, "trajectory_balance_rmse": 304.80815123998303} +{"canonical_length": 104, "canonical_nll": 1.6285394521859975, "generation_rouge_l_f1": 0.11510791366906474, "generation_token_f1": 0.1870503597122302, "prompt_id": "d0a49199c6983c98", "reading_canonical_log_prob": -169.36810302734375, "reading_distances": [26, 23, 34, 26], "reading_log_probs": [-165.36834716796875, -165.54473876953125, -165.02362060546875, -166.05557250976562], "reading_logprob_std": 0.43008420244009, "tokenization_path_logprob_std": 154.54079602319223, "trajectory_balance_rmse": 1792.1580931713581} +{"canonical_length": 2, "canonical_nll": 2.916135311126709, "generation_rouge_l_f1": 0.01234567901234568, "generation_token_f1": 0.01234567901234568, "prompt_id": "22a76db8343e67a9", "reading_canonical_log_prob": -5.832270622253418, "reading_distances": [28, 20, 26, 22], "reading_log_probs": [-6.395717620849609, -7.142444133758545, -5.934515476226807, -5.965531349182129], "reading_logprob_std": 0.56277182026244, "tokenization_path_logprob_std": 16.453483872576403, "trajectory_balance_rmse": 35.20164374369103} +{"canonical_length": 193, "canonical_nll": 1.2816438032555457, "generation_rouge_l_f1": 0.2878048780487804, "generation_token_f1": 0.5170731707317072, "prompt_id": "ab11d19cfada82a0", "reading_canonical_log_prob": -247.3572540283203, "reading_distances": [23, 24, 17, 23], "reading_log_probs": [-244.7706756591797, -246.5201416015625, -247.8230743408203, -247.10057067871094], "reading_logprob_std": 1.3026486727582525, "tokenization_path_logprob_std": 189.7550076888555, "trajectory_balance_rmse": 4495.904720075302} +{"canonical_length": 246, "canonical_nll": 1.2504823266006098, "generation_rouge_l_f1": 0.2411873840445269, "generation_token_f1": 0.4712430426716141, "prompt_id": "65ef693d02b228c6", "reading_canonical_log_prob": -307.61865234375, "reading_distances": [76, 81, 86, 85], "reading_log_probs": [-308.93414306640625, -307.4502258300781, -307.0738525390625, -306.2431640625], "reading_logprob_std": 1.1251947061099155, "tokenization_path_logprob_std": 163.01414859273413, "trajectory_balance_rmse": 5423.262716212356} +{"canonical_length": 24, "canonical_nll": 0.5884541273117065, "generation_rouge_l_f1": 0.48, "generation_token_f1": 0.48, "prompt_id": "4fb4dab7cec52a74", "reading_canonical_log_prob": -14.122899055480957, "reading_distances": [19, 27, 21, 27], "reading_log_probs": [-18.647918701171875, -17.53217124938965, -16.35317611694336, -17.371408462524414], "reading_logprob_std": 0.9395426901318495, "tokenization_path_logprob_std": 90.80261735491395, "trajectory_balance_rmse": 431.3791481264582} +{"canonical_length": 36, "canonical_nll": 1.0919818878173828, "generation_rouge_l_f1": 0.16901408450704228, "generation_token_f1": 0.2042253521126761, "prompt_id": "d445e9b3b0bb59b8", "reading_canonical_log_prob": -39.31134796142578, "reading_distances": [49, 43, 38, 42], "reading_log_probs": [-44.73255920410156, -42.23947525024414, -42.53138732910156, -42.0964469909668], "reading_logprob_std": 1.235062605261443, "tokenization_path_logprob_std": 103.81385567070784, "trajectory_balance_rmse": 865.3299680519992} +{"canonical_length": 17, "canonical_nll": 0.9222063176772174, "generation_rouge_l_f1": 0.09737827715355804, "generation_token_f1": 0.09737827715355804, "prompt_id": "9384aa5dbac32997", "reading_canonical_log_prob": -15.677507400512695, "reading_distances": [27, 24, 26, 25], "reading_log_probs": [-12.587363243103027, -16.554828643798828, -12.391334533691406, -14.482848167419434], "reading_logprob_std": 1.944528384090631, "tokenization_path_logprob_std": 62.64978881639483, "trajectory_balance_rmse": 301.6169253306817} +{"canonical_length": 227, "canonical_nll": 1.4699596791540475, "generation_rouge_l_f1": 0.15860735009671179, "generation_token_f1": 0.32882011605415856, "prompt_id": "7e391a560a4818b8", "reading_canonical_log_prob": -333.68084716796875, "reading_distances": [12, 17, 13, 11], "reading_log_probs": [-334.77154541015625, -332.2646484375, -331.2200927734375, -334.3936462402344], "reading_logprob_std": 1.7013554848754318, "tokenization_path_logprob_std": 146.40321502708565, "trajectory_balance_rmse": 4610.689851387718} +{"canonical_length": 30, "canonical_nll": 3.9581642150878906, "generation_rouge_l_f1": 0.007751937984496124, "generation_token_f1": 0.007751937984496124, "prompt_id": "62ca6c153160a733", "reading_canonical_log_prob": -118.74492645263672, "reading_distances": [47, 42, 48, 45], "reading_log_probs": [-121.29187774658203, -123.32278442382812, -122.35034942626953, -120.58462524414062], "reading_logprob_std": 1.2009051355724487, "tokenization_path_logprob_std": 0.0, "trajectory_balance_rmse": 118.7826919555664} +{"canonical_length": 85, "canonical_nll": 0.47511906343347887, "generation_rouge_l_f1": 0.4352331606217617, "generation_token_f1": 0.6632124352331606, "prompt_id": "5fd334d7d6294d39", "reading_canonical_log_prob": -40.3851203918457, "reading_distances": [25, 29, 20, 25], "reading_log_probs": [-46.85642623901367, -46.013370513916016, -42.819549560546875, -45.01506805419922], "reading_logprob_std": 1.7420083662442147, "tokenization_path_logprob_std": 106.12648927477935, "trajectory_balance_rmse": 1598.151710496629} +{"canonical_length": 161, "canonical_nll": 1.636929908894604, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.38461538461538464, "prompt_id": "ec23e8fead4fc5cf", "reading_canonical_log_prob": -263.54571533203125, "reading_distances": [25, 31, 28, 26], "reading_log_probs": [-264.67169189453125, -266.36187744140625, -266.24041748046875, -265.2524719238281], "reading_logprob_std": 0.8101698451919334, "tokenization_path_logprob_std": 207.47525602783716, "trajectory_balance_rmse": 2851.665845555486} +{"canonical_length": 296, "canonical_nll": 1.2145255836280617, "generation_rouge_l_f1": 0.28260869565217395, "generation_token_f1": 0.5, "prompt_id": "6d37aa3b89a989f6", "reading_canonical_log_prob": -359.49957275390625, "reading_distances": [16, 20, 24, 16], "reading_log_probs": [-364.0448913574219, -361.4652404785156, -360.72113037109375, -362.79150390625], "reading_logprob_std": 1.468334855111461, "tokenization_path_logprob_std": 118.02156040344225, "trajectory_balance_rmse": 6020.328843936819} +{"canonical_length": 19, "canonical_nll": 3.3567067196494653, "generation_rouge_l_f1": 0.15384615384615383, "generation_token_f1": 0.15384615384615383, "prompt_id": "b2ecf966e39675a0", "reading_canonical_log_prob": -63.777427673339844, "reading_distances": [23, 29, 29, 24], "reading_log_probs": [-65.0726089477539, -63.70307159423828, -63.01255798339844, -65.63742065429688], "reading_logprob_std": 1.2092309600267017, "tokenization_path_logprob_std": 53.230229948202314, "trajectory_balance_rmse": 428.7157683675241} +{"canonical_length": 17, "canonical_nll": 1.3624974419088924, "generation_rouge_l_f1": 0.13333333333333333, "generation_token_f1": 0.13333333333333333, "prompt_id": "efd422672564486e", "reading_canonical_log_prob": -23.162456512451172, "reading_distances": [14, 19, 18, 17], "reading_log_probs": [-27.376516342163086, -26.232704162597656, -23.835979461669922, -27.887248992919922], "reading_logprob_std": 1.80275595148145, "tokenization_path_logprob_std": 41.16634070609234, "trajectory_balance_rmse": 245.16102112789684} +{"canonical_length": 196, "canonical_nll": 1.6110196794782365, "generation_rouge_l_f1": 0.2175925925925926, "generation_token_f1": 0.44907407407407407, "prompt_id": "743badce8c66694d", "reading_canonical_log_prob": -315.7598571777344, "reading_distances": [18, 18, 22, 15], "reading_log_probs": [-318.87811279296875, -321.1621398925781, -317.46551513671875, -321.2814636230469], "reading_logprob_std": 1.8535785823469848, "tokenization_path_logprob_std": 335.2785195226042, "trajectory_balance_rmse": 4512.222700042905} +{"canonical_length": 14, "canonical_nll": 0.12168938773018974, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "685c7dde1bc34a6f", "reading_canonical_log_prob": -1.7036514282226562, "reading_distances": [33, 28, 26, 35], "reading_log_probs": [-2.087247371673584, -1.8782241344451904, -1.9594167470932007, -2.1965975761413574], "reading_logprob_std": 0.14029655611514355, "tokenization_path_logprob_std": 27.890253509374567, "trajectory_balance_rmse": 240.52562253471817} +{"canonical_length": 214, "canonical_nll": 1.2763629093348423, "generation_rouge_l_f1": 0.2544642857142857, "generation_token_f1": 0.47767857142857134, "prompt_id": "d0bf7e97f3a575eb", "reading_canonical_log_prob": -273.14166259765625, "reading_distances": [28, 25, 35, 29], "reading_log_probs": [-280.85809326171875, -280.6146240234375, -281.4754638671875, -286.1951904296875], "reading_logprob_std": 2.631295868812358, "tokenization_path_logprob_std": 283.23366342327745, "trajectory_balance_rmse": 4594.2744540461745} +{"canonical_length": 120, "canonical_nll": 0.6908111572265625, "generation_rouge_l_f1": 0.16402116402116404, "generation_token_f1": 0.35978835978835977, "prompt_id": "181490950025f621", "reading_canonical_log_prob": -82.8973388671875, "reading_distances": [31, 28, 22, 22], "reading_log_probs": [-83.8851547241211, -79.64147186279297, -80.4603271484375, -80.62862396240234], "reading_logprob_std": 1.871198319635825, "tokenization_path_logprob_std": 159.26648866107462, "trajectory_balance_rmse": 2254.52203848929} +{"canonical_length": 38, "canonical_nll": 1.7289165697599713, "generation_rouge_l_f1": 0.10921501706484642, "generation_token_f1": 0.13651877133105803, "prompt_id": "70f652d8f02698c0", "reading_canonical_log_prob": -65.6988296508789, "reading_distances": [28, 29, 27, 30], "reading_log_probs": [-65.74485778808594, -66.45382690429688, -65.34432983398438, -66.17317962646484], "reading_logprob_std": 0.4867629530095949, "tokenization_path_logprob_std": 35.23621484380458, "trajectory_balance_rmse": 1041.2531852555737} +{"canonical_length": 278, "canonical_nll": 1.136248032823741, "generation_rouge_l_f1": 0.22099447513812154, "generation_token_f1": 0.43830570902394106, "prompt_id": "fe3deab131dcdb79", "reading_canonical_log_prob": -315.876953125, "reading_distances": [25, 32, 23, 30], "reading_log_probs": [-315.3106384277344, -313.3836669921875, -315.82220458984375, -312.60601806640625], "reading_logprob_std": 1.5325618359202893, "tokenization_path_logprob_std": 171.17402242511653, "trajectory_balance_rmse": 6552.700679596649} +{"canonical_length": 16, "canonical_nll": 1.2084088325500488, "generation_rouge_l_f1": 0.19999999999999998, "generation_token_f1": 0.26666666666666666, "prompt_id": "16c98792a8b466f0", "reading_canonical_log_prob": -19.33454132080078, "reading_distances": [33, 33, 34, 31], "reading_log_probs": [-25.112070083618164, -27.632673263549805, -26.81594467163086, -25.162837982177734], "reading_logprob_std": 1.2503032971564645, "tokenization_path_logprob_std": 46.98238902155503, "trajectory_balance_rmse": 331.24508915637307} +{"canonical_length": 53, "canonical_nll": 1.2598337137474205, "generation_rouge_l_f1": 0.14285714285714288, "generation_token_f1": 0.22448979591836732, "prompt_id": "52f4380d6ba7596b", "reading_canonical_log_prob": -66.77118682861328, "reading_distances": [13, 13, 14, 16], "reading_log_probs": [-67.30288696289062, -67.1505355834961, -66.00150299072266, -66.09304809570312], "reading_logprob_std": 0.6848027369686378, "tokenization_path_logprob_std": 120.9476626154935, "trajectory_balance_rmse": 1007.3827410125213} +{"canonical_length": 78, "canonical_nll": 0.8278409517728366, "generation_rouge_l_f1": 0.125, "generation_token_f1": 0.1953125, "prompt_id": "9b002f163aa12acd", "reading_canonical_log_prob": -64.57159423828125, "reading_distances": [18, 19, 21, 20], "reading_log_probs": [-65.34799194335938, -65.435791015625, -65.76025390625, -66.32453918457031], "reading_logprob_std": 0.44204830693594616, "tokenization_path_logprob_std": 110.80067557911457, "trajectory_balance_rmse": 1379.004785437} +{"canonical_length": 22, "canonical_nll": 2.7533702850341797, "generation_rouge_l_f1": 0.17391304347826086, "generation_token_f1": 0.34782608695652173, "prompt_id": "6e455bb516dd4dc1", "reading_canonical_log_prob": -60.57414627075195, "reading_distances": [12, 14, 16, 15], "reading_log_probs": [-62.035667419433594, -62.38982391357422, -61.96596145629883, -62.37848663330078], "reading_logprob_std": 0.2231919381957795, "tokenization_path_logprob_std": 34.75853221750436, "trajectory_balance_rmse": 516.9335704266139} +{"canonical_length": 130, "canonical_nll": 1.4671863262469953, "generation_rouge_l_f1": 0.27358490566037735, "generation_token_f1": 0.4716981132075472, "prompt_id": "7c910d294599c2d9", "reading_canonical_log_prob": -190.73422241210938, "reading_distances": [14, 14, 13, 6], "reading_log_probs": [-193.93356323242188, -194.938232421875, -194.71368408203125, -194.45767211914062], "reading_logprob_std": 0.43200512458409845, "tokenization_path_logprob_std": 153.48052717481232, "trajectory_balance_rmse": 2589.315025746273} +{"canonical_length": 22, "canonical_nll": 1.265551133589311, "generation_rouge_l_f1": 0.06870229007633588, "generation_token_f1": 0.09160305343511452, "prompt_id": "e55841df337a493f", "reading_canonical_log_prob": -27.842124938964844, "reading_distances": [19, 21, 21, 24], "reading_log_probs": [-28.86298179626465, -28.01894187927246, -27.87594985961914, -30.529325485229492], "reading_logprob_std": 1.2188181260450912, "tokenization_path_logprob_std": 30.374843069395236, "trajectory_balance_rmse": 523.9985612989898} +{"canonical_length": 440, "canonical_nll": 2.0465739857066763, "generation_rouge_l_f1": 0.2503681885125184, "generation_token_f1": 0.46539027982326947, "prompt_id": "58ceb01ab8ce3495", "reading_canonical_log_prob": -900.4925537109375, "reading_distances": [80, 81, 74, 88], "reading_log_probs": [-898.39501953125, -904.270263671875, -899.8695068359375, -913.7316284179688], "reading_logprob_std": 6.909815946719724, "tokenization_path_logprob_std": 128.2175455689952, "trajectory_balance_rmse": 7921.814735372733} +{"canonical_length": 32, "canonical_nll": 1.0908325910568237, "generation_rouge_l_f1": 0.28776978417266186, "generation_token_f1": 0.3453237410071942, "prompt_id": "a8059315f9c843b5", "reading_canonical_log_prob": -34.90664291381836, "reading_distances": [23, 31, 21, 20], "reading_log_probs": [-34.941131591796875, -34.57523727416992, -33.91490936279297, -34.52643585205078], "reading_logprob_std": 0.425378502820849, "tokenization_path_logprob_std": 73.08901321627403, "trajectory_balance_rmse": 849.3696174395974} +{"canonical_length": 12, "canonical_nll": 1.709822654724121, "generation_rouge_l_f1": 0.15384615384615385, "generation_token_f1": 0.17094017094017094, "prompt_id": "3715d1d4c18fe319", "reading_canonical_log_prob": -20.517871856689453, "reading_distances": [13, 9, 11, 19], "reading_log_probs": [-21.710227966308594, -20.524986267089844, -21.004026412963867, -21.601112365722656], "reading_logprob_std": 0.552227550494679, "tokenization_path_logprob_std": 64.45905514431988, "trajectory_balance_rmse": 264.7339217636866} +{"canonical_length": 264, "canonical_nll": 1.8564414978027344, "generation_rouge_l_f1": 0.2554027504911591, "generation_token_f1": 0.4950884086444008, "prompt_id": "996c63537529bc50", "reading_canonical_log_prob": -490.1005554199219, "reading_distances": [31, 19, 25, 22], "reading_log_probs": [-484.4481201171875, -496.23663330078125, -495.58221435546875, -492.15228271484375], "reading_logprob_std": 5.409604374335901, "tokenization_path_logprob_std": 46.36136935334647, "trajectory_balance_rmse": 6193.462363470775} +{"canonical_length": 15, "canonical_nll": 0.5065308888753255, "generation_rouge_l_f1": 0.8148148148148148, "generation_token_f1": 0.888888888888889, "prompt_id": "a9d8f6ac4429b707", "reading_canonical_log_prob": -7.597963333129883, "reading_distances": [23, 23, 20, 23], "reading_log_probs": [-8.215347290039062, -8.228415489196777, -8.4659423828125, -8.159802436828613], "reading_logprob_std": 0.1356780354806162, "tokenization_path_logprob_std": 68.4049946116178, "trajectory_balance_rmse": 223.20869364983471} +{"canonical_length": 17, "canonical_nll": 0.9221078087301815, "generation_rouge_l_f1": 0.5185185185185185, "generation_token_f1": 0.5185185185185185, "prompt_id": "b1822d4ae9e3de29", "reading_canonical_log_prob": -15.675832748413086, "reading_distances": [21, 22, 25, 16], "reading_log_probs": [-17.034465789794922, -16.93427276611328, -16.910608291625977, -17.164926528930664], "reading_logprob_std": 0.11577213891351601, "tokenization_path_logprob_std": 58.97111946597559, "trajectory_balance_rmse": 446.245431396109} +{"canonical_length": 14, "canonical_nll": 0.8373917170933315, "generation_rouge_l_f1": 0.16326530612244897, "generation_token_f1": 0.16326530612244897, "prompt_id": "92e5aa90e47873e8", "reading_canonical_log_prob": -11.72348403930664, "reading_distances": [27, 17, 21, 21], "reading_log_probs": [-11.540109634399414, -11.773839950561523, -11.574920654296875, -11.858101844787598], "reading_logprob_std": 0.15379132332477774, "tokenization_path_logprob_std": 52.097170145729045, "trajectory_balance_rmse": 312.4475633901251} +{"canonical_length": 176, "canonical_nll": 1.699462890625, "generation_rouge_l_f1": 0.20069204152249134, "generation_token_f1": 0.3598615916955017, "prompt_id": "d3105cc3dd2ecc61", "reading_canonical_log_prob": -299.10546875, "reading_distances": [31, 31, 33, 35], "reading_log_probs": [-298.68072509765625, -295.3873291015625, -297.3862609863281, -299.307861328125], "reading_logprob_std": 1.731431100448678, "tokenization_path_logprob_std": 83.63454782941587, "trajectory_balance_rmse": 3780.453312211766} +{"canonical_length": 46, "canonical_nll": 0.7682019109311311, "generation_rouge_l_f1": 0.13937282229965156, "generation_token_f1": 0.16724738675958187, "prompt_id": "f2effbcb126018c5", "reading_canonical_log_prob": -35.33728790283203, "reading_distances": [23, 26, 29, 23], "reading_log_probs": [-37.35614013671875, -34.5216064453125, -35.662376403808594, -35.050506591796875], "reading_logprob_std": 1.230678640835939, "tokenization_path_logprob_std": 42.8673182546295, "trajectory_balance_rmse": 473.8540899983429} +{"canonical_length": 226, "canonical_nll": 1.3548035748237002, "generation_rouge_l_f1": 0.16173570019723865, "generation_token_f1": 0.378698224852071, "prompt_id": "e73472036e75a1c6", "reading_canonical_log_prob": -306.18560791015625, "reading_distances": [41, 37, 54, 44], "reading_log_probs": [-302.8804931640625, -304.1025390625, -302.3192138671875, -305.501220703125], "reading_logprob_std": 1.4123966053308052, "tokenization_path_logprob_std": 183.2268583130276, "trajectory_balance_rmse": 4179.828948122547} +{"canonical_length": 8, "canonical_nll": 0.3916812241077423, "generation_rouge_l_f1": 0.7692307692307692, "generation_token_f1": 0.7692307692307692, "prompt_id": "dea67577bfdb2da1", "reading_canonical_log_prob": -3.1334497928619385, "reading_distances": [22, 26, 30, 27], "reading_log_probs": [-3.9673569202423096, -4.1520562171936035, -4.449762344360352, -3.677459716796875], "reading_logprob_std": 0.3241901881305157, "tokenization_path_logprob_std": 18.60815928321628, "trajectory_balance_rmse": 139.66619638686808} +{"canonical_length": 107, "canonical_nll": 1.1705834219388873, "generation_rouge_l_f1": 0.16377171215880895, "generation_token_f1": 0.29280397022332505, "prompt_id": "90fd3570044a7e38", "reading_canonical_log_prob": -125.25242614746094, "reading_distances": [24, 25, 24, 18], "reading_log_probs": [-133.9582061767578, -133.59811401367188, -132.87330627441406, -138.2232208251953], "reading_logprob_std": 2.4158430049183885, "tokenization_path_logprob_std": 77.10387000273623, "trajectory_balance_rmse": 2101.346029407826} +{"canonical_length": 222, "canonical_nll": 0.6329674935555673, "generation_rouge_l_f1": 0.23370786516853934, "generation_token_f1": 0.4584269662921348, "prompt_id": "1c7402f9ee727b52", "reading_canonical_log_prob": -140.51878356933594, "reading_distances": [78, 90, 86, 86], "reading_log_probs": [-132.85621643066406, -133.51986694335938, -131.459716796875, -133.20559692382812], "reading_logprob_std": 0.9084690257104221, "tokenization_path_logprob_std": 87.60501625149533, "trajectory_balance_rmse": 2522.932100924096} +{"canonical_length": 19, "canonical_nll": 1.3614195773476048, "generation_rouge_l_f1": 0.29411764705882354, "generation_token_f1": 0.4117647058823529, "prompt_id": "219202257fbe0a6c", "reading_canonical_log_prob": -25.866971969604492, "reading_distances": [21, 22, 20, 19], "reading_log_probs": [-27.144378662109375, -26.92904281616211, -27.088804244995117, -26.838104248046875], "reading_logprob_std": 0.14139368352234472, "tokenization_path_logprob_std": 45.23876065110469, "trajectory_balance_rmse": 422.2839548142955} +{"canonical_length": 47, "canonical_nll": 1.3919206984499668, "generation_rouge_l_f1": 0.35955056179775274, "generation_token_f1": 0.40449438202247195, "prompt_id": "1eb21676f35636bf", "reading_canonical_log_prob": -65.42027282714844, "reading_distances": [20, 29, 28, 23], "reading_log_probs": [-68.68782806396484, -75.19590759277344, -72.24949645996094, -74.71197509765625], "reading_logprob_std": 2.97643884903259, "tokenization_path_logprob_std": 22.404129249477318, "trajectory_balance_rmse": 1104.8188353726398} +{"canonical_length": 15, "canonical_nll": 0.15404055913289388, "generation_rouge_l_f1": 0.923076923076923, "generation_token_f1": 0.923076923076923, "prompt_id": "5f8f43706c38c575", "reading_canonical_log_prob": -2.310608386993408, "reading_distances": [29, 31, 29, 33], "reading_log_probs": [-2.911881923675537, -2.9751338958740234, -3.038835048675537, -2.978485345840454], "reading_logprob_std": 0.051853221931190845, "tokenization_path_logprob_std": 42.38106221185393, "trajectory_balance_rmse": 302.8205557425446} +{"canonical_length": 18, "canonical_nll": 0.8047979672749838, "generation_rouge_l_f1": 0.7741935483870968, "generation_token_f1": 0.8387096774193549, "prompt_id": "46d2c294deb9298b", "reading_canonical_log_prob": -14.486363410949707, "reading_distances": [73, 64, 59, 69], "reading_log_probs": [-14.530344009399414, -16.069576263427734, -16.015714645385742, -12.330707550048828], "reading_logprob_std": 1.7553556152971077, "tokenization_path_logprob_std": 42.50719765820369, "trajectory_balance_rmse": 434.28083126994375} +{"canonical_length": 228, "canonical_nll": 1.2541814435992324, "generation_rouge_l_f1": 0.2127659574468085, "generation_token_f1": 0.3404255319148936, "prompt_id": "7cb804cc64427944", "reading_canonical_log_prob": -285.953369140625, "reading_distances": [24, 19, 19, 19], "reading_log_probs": [-283.0941162109375, -284.4356384277344, -281.37030029296875, -282.4829406738281], "reading_logprob_std": 1.2777602060520064, "tokenization_path_logprob_std": 140.81814144554352, "trajectory_balance_rmse": 4815.657303834849} +{"canonical_length": 171, "canonical_nll": 1.4634122792740314, "generation_rouge_l_f1": 0.2441860465116279, "generation_token_f1": 0.5116279069767442, "prompt_id": "d5a4beae5409050e", "reading_canonical_log_prob": -250.24349975585938, "reading_distances": [29, 18, 28, 22], "reading_log_probs": [-250.88018798828125, -251.87738037109375, -254.15655517578125, -255.3896484375], "reading_logprob_std": 2.0638837154631897, "tokenization_path_logprob_std": 324.6831257677846, "trajectory_balance_rmse": 4040.0500428880523} +{"canonical_length": 168, "canonical_nll": 1.7773978823707217, "generation_rouge_l_f1": 0.2037914691943128, "generation_token_f1": 0.3601895734597156, "prompt_id": "ad2f9057c92483ba", "reading_canonical_log_prob": -298.60284423828125, "reading_distances": [13, 14, 14, 11], "reading_log_probs": [-294.1968688964844, -291.68414306640625, -294.2728271484375, -292.12030029296875], "reading_logprob_std": 1.3588165250395674, "tokenization_path_logprob_std": 155.01482724071414, "trajectory_balance_rmse": 3693.052997938293} +{"canonical_length": 159, "canonical_nll": 1.2820881777589426, "generation_rouge_l_f1": 0.29901960784313725, "generation_token_f1": 0.4950980392156863, "prompt_id": "240077b10fcaa781", "reading_canonical_log_prob": -203.85202026367188, "reading_distances": [27, 25, 26, 24], "reading_log_probs": [-202.60888671875, -203.6920623779297, -203.93621826171875, -203.47100830078125], "reading_logprob_std": 0.5775832886672027, "tokenization_path_logprob_std": 136.0946288946527, "trajectory_balance_rmse": 3636.827719512242} +{"canonical_length": 327, "canonical_nll": 0.8262465357415902, "generation_rouge_l_f1": 0.19723183391003457, "generation_token_f1": 0.42560553633217996, "prompt_id": "dcea5e8dbdd12a1a", "reading_canonical_log_prob": -270.1826171875, "reading_distances": [68, 68, 70, 69], "reading_log_probs": [-270.023193359375, -272.2053527832031, -267.9727783203125, -271.89599609375], "reading_logprob_std": 1.9552373578915596, "tokenization_path_logprob_std": 184.7420479549218, "trajectory_balance_rmse": 5664.834052421721} +{"canonical_length": 4, "canonical_nll": 2.057771682739258, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "09f4cd01fd363213", "reading_canonical_log_prob": -8.231086730957031, "reading_distances": [30, 33, 33, 28], "reading_log_probs": [-12.817010879516602, -12.336835861206055, -12.756393432617188, -11.918725967407227], "reading_logprob_std": 0.417701002282498, "tokenization_path_logprob_std": 17.161589348854722, "trajectory_balance_rmse": 93.77039776109125} +{"canonical_length": 139, "canonical_nll": 1.088539837075652, "generation_rouge_l_f1": 0.24157303370786518, "generation_token_f1": 0.41573033707865165, "prompt_id": "0eb3ed1212b155f2", "reading_canonical_log_prob": -151.30703735351562, "reading_distances": [26, 19, 22, 22], "reading_log_probs": [-152.59820556640625, -152.79080200195312, -153.32498168945312, -152.71775817871094], "reading_logprob_std": 0.3213245345149231, "tokenization_path_logprob_std": 170.3572533870091, "trajectory_balance_rmse": 2696.8520301633935} +{"canonical_length": 3, "canonical_nll": 2.4078493118286133, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "423d33be2ee3658e", "reading_canonical_log_prob": -7.22354793548584, "reading_distances": [22, 21, 21, 23], "reading_log_probs": [-8.816905975341797, -8.536005020141602, -8.249473571777344, -8.263687133789062], "reading_logprob_std": 0.2682342405079558, "tokenization_path_logprob_std": 9.10794152787239, "trajectory_balance_rmse": 30.664816338735566} +{"canonical_length": 33, "canonical_nll": 1.0960924553148674, "generation_rouge_l_f1": 0.07633587786259542, "generation_token_f1": 0.0916030534351145, "prompt_id": "82dd428e952ac035", "reading_canonical_log_prob": -36.171051025390625, "reading_distances": [36, 31, 30, 32], "reading_log_probs": [-37.10548400878906, -37.22064971923828, -36.531768798828125, -37.2000617980957], "reading_logprob_std": 0.32569842378931096, "tokenization_path_logprob_std": 41.64889360387235, "trajectory_balance_rmse": 749.244144881398} +{"canonical_length": 293, "canonical_nll": 1.1936274714030504, "generation_rouge_l_f1": 0.22980251346499103, "generation_token_f1": 0.481149012567325, "prompt_id": "a3746b2776781f66", "reading_canonical_log_prob": -349.73284912109375, "reading_distances": [21, 25, 23, 27], "reading_log_probs": [-352.67913818359375, -353.1080322265625, -354.5940856933594, -355.20068359375], "reading_logprob_std": 1.195988816289013, "tokenization_path_logprob_std": 219.1983933416794, "trajectory_balance_rmse": 6632.45222491066} +{"canonical_length": 64, "canonical_nll": 1.7935295104980469, "generation_rouge_l_f1": 0.11076923076923076, "generation_token_f1": 0.17846153846153845, "prompt_id": "d4e9c302f1022491", "reading_canonical_log_prob": -114.785888671875, "reading_distances": [30, 25, 32, 28], "reading_log_probs": [-112.72350311279297, -116.08455657958984, -114.87984466552734, -115.66033172607422], "reading_logprob_std": 1.4947661237467118, "tokenization_path_logprob_std": 149.21235106170946, "trajectory_balance_rmse": 1409.8095451968916} +{"canonical_length": 99, "canonical_nll": 0.9978038132792771, "generation_rouge_l_f1": 0.1734417344173442, "generation_token_f1": 0.29810298102981025, "prompt_id": "ed00a510b9a4ea44", "reading_canonical_log_prob": -98.78257751464844, "reading_distances": [30, 33, 35, 28], "reading_log_probs": [-99.62674713134766, -101.06329345703125, -100.81901550292969, -100.2355728149414], "reading_logprob_std": 0.6416798578134556, "tokenization_path_logprob_std": 222.72571555834563, "trajectory_balance_rmse": 2327.928206131379} +{"canonical_length": 275, "canonical_nll": 1.3463433283025568, "generation_rouge_l_f1": 0.23486238532110093, "generation_token_f1": 0.42568807339449544, "prompt_id": "00a5d180fc30eed4", "reading_canonical_log_prob": -370.2444152832031, "reading_distances": [30, 25, 32, 26], "reading_log_probs": [-373.51751708984375, -372.6017150878906, -375.1291809082031, -375.36553955078125], "reading_logprob_std": 1.3207956922742168, "tokenization_path_logprob_std": 186.75068230187287, "trajectory_balance_rmse": 6003.058859255139} +{"canonical_length": 115, "canonical_nll": 1.2665139903192935, "generation_rouge_l_f1": 0.2981818181818182, "generation_token_f1": 0.3927272727272727, "prompt_id": "300b5dc6095f893e", "reading_canonical_log_prob": -145.64910888671875, "reading_distances": [6, 10, 5, 6], "reading_log_probs": [-147.38150024414062, -144.58670043945312, -145.74496459960938, -145.9594268798828], "reading_logprob_std": 1.14685785707629, "tokenization_path_logprob_std": 136.63872128665966, "trajectory_balance_rmse": 2409.9668826187763} +{"canonical_length": 6, "canonical_nll": 2.171886444091797, "generation_rouge_l_f1": 0.30769230769230765, "generation_token_f1": 0.30769230769230765, "prompt_id": "2c288dbc9a356333", "reading_canonical_log_prob": -13.031318664550781, "reading_distances": [19, 21, 19, 22], "reading_log_probs": [-13.76438045501709, -13.26124095916748, -14.197920799255371, -13.422563552856445], "reading_logprob_std": 0.4145799418863309, "tokenization_path_logprob_std": 34.25598392273279, "trajectory_balance_rmse": 144.32441845998017} +{"canonical_length": 233, "canonical_nll": 1.0093485950912018, "generation_rouge_l_f1": 0.3134657836644592, "generation_token_f1": 0.5695364238410596, "prompt_id": "810a6641955f3c0f", "reading_canonical_log_prob": -235.17822265625, "reading_distances": [42, 39, 33, 35], "reading_log_probs": [-232.49423217773438, -232.28604125976562, -234.9157257080078, -229.53128051757812], "reading_logprob_std": 2.2019357765511143, "tokenization_path_logprob_std": 84.26406060368845, "trajectory_balance_rmse": 2939.5073661784572} +{"canonical_length": 167, "canonical_nll": 2.0818011803541356, "generation_rouge_l_f1": 0.28771929824561404, "generation_token_f1": 0.512280701754386, "prompt_id": "eb867c21f2710dc4", "reading_canonical_log_prob": -347.6607971191406, "reading_distances": [23, 31, 25, 26], "reading_log_probs": [-347.1059265136719, -344.99017333984375, -344.5908203125, -345.3497009277344], "reading_logprob_std": 1.1087207989209942, "tokenization_path_logprob_std": 138.0879035716218, "trajectory_balance_rmse": 3899.9548949094583} +{"canonical_length": 48, "canonical_nll": 1.8833068211873372, "generation_rouge_l_f1": 0.13194444444444442, "generation_token_f1": 0.2013888888888889, "prompt_id": "ee8008c4016a6c6e", "reading_canonical_log_prob": -90.39872741699219, "reading_distances": [24, 23, 32, 28], "reading_log_probs": [-89.22744750976562, -91.91016387939453, -90.2990951538086, -91.64144897460938], "reading_logprob_std": 1.2464102716721717, "tokenization_path_logprob_std": 44.281827613389844, "trajectory_balance_rmse": 1265.4286733019887} +{"canonical_length": 232, "canonical_nll": 1.1950580991547683, "generation_rouge_l_f1": 0.23841059602649006, "generation_token_f1": 0.4768211920529801, "prompt_id": "5adec8000ecf22fe", "reading_canonical_log_prob": -277.25347900390625, "reading_distances": [22, 18, 20, 23], "reading_log_probs": [-282.6348876953125, -284.8027648925781, -284.180908203125, -282.0703430175781], "reading_logprob_std": 1.2817924329429577, "tokenization_path_logprob_std": 126.45484586973203, "trajectory_balance_rmse": 5003.005700489801} +{"canonical_length": 283, "canonical_nll": 1.4330396888112853, "generation_rouge_l_f1": 0.2076923076923077, "generation_token_f1": 0.46923076923076923, "prompt_id": "893df9640ff6dc30", "reading_canonical_log_prob": -405.55023193359375, "reading_distances": [22, 23, 24, 19], "reading_log_probs": [-403.395751953125, -403.36260986328125, -404.640869140625, -404.5642395019531], "reading_logprob_std": 0.7071368933161934, "tokenization_path_logprob_std": 33.59659053740137, "trajectory_balance_rmse": 5825.840703362219} +{"canonical_length": 13, "canonical_nll": 0.6066227692824143, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.5555555555555556, "prompt_id": "5f407731bdd7e46c", "reading_canonical_log_prob": -7.886096000671387, "reading_distances": [25, 31, 31, 29], "reading_log_probs": [-8.637343406677246, -8.597146987915039, -8.980306625366211, -8.18209171295166], "reading_logprob_std": 0.3269458239451944, "tokenization_path_logprob_std": 42.54252824426213, "trajectory_balance_rmse": 145.77894112464656} +{"canonical_length": 226, "canonical_nll": 1.0900163228532909, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.4708333333333333, "prompt_id": "c371f7349e2c9b37", "reading_canonical_log_prob": -246.34368896484375, "reading_distances": [24, 30, 21, 19], "reading_log_probs": [-247.3643798828125, -247.0595703125, -245.71054077148438, -245.20680236816406], "reading_logprob_std": 1.0404172059443495, "tokenization_path_logprob_std": 233.2400884700911, "trajectory_balance_rmse": 5084.250569587199} +{"canonical_length": 9, "canonical_nll": 1.5009980731540256, "generation_rouge_l_f1": 0.35294117647058826, "generation_token_f1": 0.35294117647058826, "prompt_id": "59b66b10bf9d5bcd", "reading_canonical_log_prob": -13.50898265838623, "reading_distances": [64, 50, 67, 58], "reading_log_probs": [-17.677927017211914, -16.124387741088867, -18.502025604248047, -15.392074584960938], "reading_logprob_std": 1.4194771283519243, "tokenization_path_logprob_std": 14.494663589489026, "trajectory_balance_rmse": 189.82011769028736} +{"canonical_length": 387, "canonical_nll": 1.3325330946180556, "generation_rouge_l_f1": 0.19202363367799113, "generation_token_f1": 0.3663220088626292, "prompt_id": "d9970f787452110a", "reading_canonical_log_prob": -515.6903076171875, "reading_distances": [32, 28, 31, 31], "reading_log_probs": [-513.3318481445312, -515.7570190429688, -514.232666015625, -516.9249267578125], "reading_logprob_std": 1.5952813004560442, "tokenization_path_logprob_std": 367.0707838116541, "trajectory_balance_rmse": 7688.290351426898} +{"canonical_length": 199, "canonical_nll": 1.0990277007596576, "generation_rouge_l_f1": 0.28205128205128205, "generation_token_f1": 0.49230769230769234, "prompt_id": "4f2d283e14485973", "reading_canonical_log_prob": -218.70651245117188, "reading_distances": [20, 21, 24, 20], "reading_log_probs": [-218.72482299804688, -217.89273071289062, -217.281005859375, -217.087890625], "reading_logprob_std": 0.7368741102397951, "tokenization_path_logprob_std": 180.4251929830084, "trajectory_balance_rmse": 4243.740852270617} +{"canonical_length": 366, "canonical_nll": 1.321778677851776, "generation_rouge_l_f1": 0.18336162988115448, "generation_token_f1": 0.3870967741935484, "prompt_id": "74c803f89db0a539", "reading_canonical_log_prob": -483.77099609375, "reading_distances": [12, 15, 17, 12], "reading_log_probs": [-482.77288818359375, -479.59710693359375, -479.6339111328125, -481.7896728515625], "reading_logprob_std": 1.590636253832369, "tokenization_path_logprob_std": 133.95971488220448, "trajectory_balance_rmse": 7350.014240220255} +{"canonical_length": 132, "canonical_nll": 1.1636032335685962, "generation_rouge_l_f1": 0.30985915492957744, "generation_token_f1": 0.5492957746478874, "prompt_id": "456727f2988746cd", "reading_canonical_log_prob": -153.5956268310547, "reading_distances": [26, 29, 30, 31], "reading_log_probs": [-156.5611572265625, -157.52427673339844, -156.78749084472656, -157.76832580566406], "reading_logprob_std": 0.577388711177762, "tokenization_path_logprob_std": 191.11017826778192, "trajectory_balance_rmse": 3486.3142056450474} +{"canonical_length": 12, "canonical_nll": 0.1318132976690928, "generation_rouge_l_f1": 0.9090909090909091, "generation_token_f1": 0.9090909090909091, "prompt_id": "4105efabcbebe823", "reading_canonical_log_prob": -1.5817595720291138, "reading_distances": [29, 18, 24, 23], "reading_log_probs": [-5.176367282867432, -2.3988890647888184, -5.08861780166626, -2.5807480812072754], "reading_logprob_std": 1.5279738266195084, "tokenization_path_logprob_std": 23.818638161442752, "trajectory_balance_rmse": 265.12824994388694} +{"canonical_length": 57, "canonical_nll": 1.4406130606668037, "generation_rouge_l_f1": 0.09523809523809525, "generation_token_f1": 0.09523809523809525, "prompt_id": "61331d091d23a9af", "reading_canonical_log_prob": -82.11494445800781, "reading_distances": [16, 16, 15, 16], "reading_log_probs": [-84.61519622802734, -86.11763000488281, -85.98291015625, -83.70588684082031], "reading_logprob_std": 1.1537721435571495, "tokenization_path_logprob_std": 125.76772470164906, "trajectory_balance_rmse": 1253.1741976639034} +{"canonical_length": 32, "canonical_nll": 2.60089373588562, "generation_rouge_l_f1": 0.39999999999999997, "generation_token_f1": 0.39999999999999997, "prompt_id": "1a1747f5e15bef6b", "reading_canonical_log_prob": -83.22859954833984, "reading_distances": [21, 38, 30, 33], "reading_log_probs": [-81.03681945800781, -80.85885620117188, -81.30390167236328, -81.03889465332031], "reading_logprob_std": 0.18342032154459828, "tokenization_path_logprob_std": 34.65157205950523, "trajectory_balance_rmse": 911.4066475295649} +{"canonical_length": 9, "canonical_nll": 0.2255400021870931, "generation_rouge_l_f1": 0.9333333333333333, "generation_token_f1": 0.9333333333333333, "prompt_id": "8d4efdacc1d27322", "reading_canonical_log_prob": -2.029860019683838, "reading_distances": [33, 36, 36, 33], "reading_log_probs": [-3.1480824947357178, -3.7379837036132812, -2.7332868576049805, -2.208902359008789], "reading_logprob_std": 0.647082091477223, "tokenization_path_logprob_std": 18.430426908475617, "trajectory_balance_rmse": 254.32551588944148} +{"canonical_length": 263, "canonical_nll": 1.5964645559796815, "generation_rouge_l_f1": 0.18540433925049307, "generation_token_f1": 0.39447731755424065, "prompt_id": "123b7e61f56166d9", "reading_canonical_log_prob": -419.87017822265625, "reading_distances": [34, 36, 32, 39], "reading_log_probs": [-422.59332275390625, -422.17156982421875, -422.8707580566406, -427.412109375], "reading_logprob_std": 2.4503663186171374, "tokenization_path_logprob_std": 172.71037201134487, "trajectory_balance_rmse": 4914.15508039376} +{"canonical_length": 63, "canonical_nll": 1.2566806853763641, "generation_rouge_l_f1": 0.5692307692307692, "generation_token_f1": 0.5846153846153845, "prompt_id": "0407e3355bf3ebea", "reading_canonical_log_prob": -79.17088317871094, "reading_distances": [18, 15, 20, 19], "reading_log_probs": [-81.59867858886719, -85.11212158203125, -84.02816772460938, -84.11715698242188], "reading_logprob_std": 1.4933815697159247, "tokenization_path_logprob_std": 28.69938144736805, "trajectory_balance_rmse": 1175.1763178625272} +{"canonical_length": 29, "canonical_nll": 0.9030856428475216, "generation_rouge_l_f1": 0.4931506849315069, "generation_token_f1": 0.6027397260273972, "prompt_id": "f217efc1d8ea8fee", "reading_canonical_log_prob": -26.189483642578125, "reading_distances": [10, 17, 13, 15], "reading_log_probs": [-27.573415756225586, -27.355947494506836, -27.113094329833984, -26.652912139892578], "reading_logprob_std": 0.3949176270123515, "tokenization_path_logprob_std": 29.7319062245119, "trajectory_balance_rmse": 604.5746004876938} +{"canonical_length": 125, "canonical_nll": 1.557568359375, "generation_rouge_l_f1": 0.22395833333333331, "generation_token_f1": 0.39583333333333337, "prompt_id": "65c5ecab43401060", "reading_canonical_log_prob": -194.696044921875, "reading_distances": [31, 17, 21, 32], "reading_log_probs": [-197.106201171875, -196.95010375976562, -198.1203155517578, -197.03143310546875], "reading_logprob_std": 0.5492464325277098, "tokenization_path_logprob_std": 198.29045626598608, "trajectory_balance_rmse": 2794.9654028548657} +{"canonical_length": 41, "canonical_nll": 1.0475666697432355, "generation_rouge_l_f1": 0.4814814814814815, "generation_token_f1": 0.4814814814814815, "prompt_id": "5cd588db0ae30114", "reading_canonical_log_prob": -42.950233459472656, "reading_distances": [26, 32, 31, 25], "reading_log_probs": [-45.52919006347656, -45.96818542480469, -45.55657196044922, -46.759586334228516], "reading_logprob_std": 0.5737541797687195, "tokenization_path_logprob_std": 18.356393657174035, "trajectory_balance_rmse": 825.9988049317369} +{"canonical_length": 145, "canonical_nll": 1.8172855771821121, "generation_rouge_l_f1": 0.21836228287841192, "generation_token_f1": 0.4317617866004963, "prompt_id": "f88ab53f7308d0a9", "reading_canonical_log_prob": -263.50640869140625, "reading_distances": [26, 29, 27, 18], "reading_log_probs": [-275.84490966796875, -278.45257568359375, -272.5187072753906, -271.2583312988281], "reading_logprob_std": 3.259048170064451, "tokenization_path_logprob_std": 189.43864608269243, "trajectory_balance_rmse": 3515.6969890011424} +{"canonical_length": 55, "canonical_nll": 1.2667669122869318, "generation_rouge_l_f1": 0.30656934306569344, "generation_token_f1": 0.3795620437956205, "prompt_id": "8c737ded95e1b468", "reading_canonical_log_prob": -69.67218017578125, "reading_distances": [12, 20, 18, 16], "reading_log_probs": [-82.3282470703125, -80.76917266845703, -79.87104034423828, -80.54058837890625], "reading_logprob_std": 1.0396909998532184, "tokenization_path_logprob_std": 108.7531262464245, "trajectory_balance_rmse": 1125.8047527008202} +{"canonical_length": 22, "canonical_nll": 0.29206950014287775, "generation_rouge_l_f1": 0.5074626865671642, "generation_token_f1": 0.5970149253731344, "prompt_id": "cd208c8b818d2286", "reading_canonical_log_prob": -6.4255290031433105, "reading_distances": [24, 27, 28, 30], "reading_log_probs": [-4.563438892364502, -5.859652042388916, -5.09006404876709, -5.201903343200684], "reading_logprob_std": 0.5324899451025574, "tokenization_path_logprob_std": 43.44387784803009, "trajectory_balance_rmse": 328.34294511494045} +{"canonical_length": 165, "canonical_nll": 0.9678187514796401, "generation_rouge_l_f1": 0.40455840455840464, "generation_token_f1": 0.621082621082621, "prompt_id": "a88030476f29d6bf", "reading_canonical_log_prob": -159.69009399414062, "reading_distances": [14, 12, 16, 17], "reading_log_probs": [-161.6131591796875, -164.46206665039062, -161.9443817138672, -161.71192932128906], "reading_logprob_std": 1.3598947437203452, "tokenization_path_logprob_std": 155.40381300907762, "trajectory_balance_rmse": 4009.3484272077626} +{"canonical_length": 377, "canonical_nll": 1.263330758092258, "generation_rouge_l_f1": 0.228125, "generation_token_f1": 0.49062500000000003, "prompt_id": "842add8204336566", "reading_canonical_log_prob": -476.27569580078125, "reading_distances": [35, 31, 24, 24], "reading_log_probs": [-481.64288330078125, -481.099853515625, -478.3043212890625, -481.3873291015625], "reading_logprob_std": 1.5521159056964884, "tokenization_path_logprob_std": 82.49195198868196, "trajectory_balance_rmse": 7830.883578401098} +{"canonical_length": 112, "canonical_nll": 1.2127843584333147, "generation_rouge_l_f1": 0.24404761904761907, "generation_token_f1": 0.42857142857142855, "prompt_id": "95ed700077e4b7ce", "reading_canonical_log_prob": -135.83184814453125, "reading_distances": [12, 13, 12, 14], "reading_log_probs": [-132.2274169921875, -132.44732666015625, -131.5888214111328, -133.20223999023438], "reading_logprob_std": 0.6656130861625706, "tokenization_path_logprob_std": 188.29773118569824, "trajectory_balance_rmse": 1636.4832802945791} +{"canonical_length": 11, "canonical_nll": 0.88103571805087, "generation_rouge_l_f1": 0.45454545454545453, "generation_token_f1": 0.45454545454545453, "prompt_id": "e5cef34a838082a8", "reading_canonical_log_prob": -9.69139289855957, "reading_distances": [35, 36, 30, 38], "reading_log_probs": [-10.664528846740723, -10.4421968460083, -10.231606483459473, -10.354371070861816], "reading_logprob_std": 0.18261616134565087, "tokenization_path_logprob_std": 35.73183113056643, "trajectory_balance_rmse": 359.7226179785971} +{"canonical_length": 10, "canonical_nll": 0.35053443908691406, "generation_rouge_l_f1": 0.8888888888888888, "generation_token_f1": 0.8888888888888888, "prompt_id": "47116aef05367776", "reading_canonical_log_prob": -3.5053443908691406, "reading_distances": [32, 37, 32, 38], "reading_log_probs": [-5.794540882110596, -5.2147345542907715, -6.028613090515137, -5.729606628417969], "reading_logprob_std": 0.34303522931915337, "tokenization_path_logprob_std": 29.197779780392516, "trajectory_balance_rmse": 143.37994009000238} +{"canonical_length": 375, "canonical_nll": 1.39995654296875, "generation_rouge_l_f1": 0.21428571428571427, "generation_token_f1": 0.4254658385093167, "prompt_id": "7dea199ced8c5f4e", "reading_canonical_log_prob": -524.9837036132812, "reading_distances": [32, 26, 28, 27], "reading_log_probs": [-522.8887939453125, -530.2958984375, -526.5645751953125, -529.7894287109375], "reading_logprob_std": 3.422648704383785, "tokenization_path_logprob_std": 65.74203977925234, "trajectory_balance_rmse": 7136.5828239956345} +{"canonical_length": 377, "canonical_nll": 1.4046740949312002, "generation_rouge_l_f1": 0.2496050552922591, "generation_token_f1": 0.4581358609794629, "prompt_id": "c009e5e48dec12c7", "reading_canonical_log_prob": -529.5621337890625, "reading_distances": [27, 29, 32, 26], "reading_log_probs": [-523.821533203125, -523.24951171875, -520.0885009765625, -522.9759521484375], "reading_logprob_std": 1.6678814786999459, "tokenization_path_logprob_std": 288.9559613579595, "trajectory_balance_rmse": 6909.173554302699} +{"canonical_length": 206, "canonical_nll": 1.114674836686514, "generation_rouge_l_f1": 0.31308411214953275, "generation_token_f1": 0.5186915887850467, "prompt_id": "12ce96bbad2517ca", "reading_canonical_log_prob": -229.62301635742188, "reading_distances": [21, 23, 23, 20], "reading_log_probs": [-230.8557891845703, -231.3118438720703, -232.14779663085938, -232.79840087890625], "reading_logprob_std": 0.8652053682654653, "tokenization_path_logprob_std": 80.1877129164902, "trajectory_balance_rmse": 4536.433170898222} +{"canonical_length": 193, "canonical_nll": 1.358192878683614, "generation_rouge_l_f1": 0.33995584988962474, "generation_token_f1": 0.5695364238410597, "prompt_id": "cda8b27fc8b243fa", "reading_canonical_log_prob": -262.1312255859375, "reading_distances": [22, 12, 18, 13], "reading_log_probs": [-259.8150634765625, -260.2635498046875, -259.7496337890625, -259.5455017089844], "reading_logprob_std": 0.3026903288341923, "tokenization_path_logprob_std": 207.60676615348132, "trajectory_balance_rmse": 4001.1763133799905} +{"canonical_length": 247, "canonical_nll": 0.9705039576480263, "generation_rouge_l_f1": 0.25572519083969464, "generation_token_f1": 0.4732824427480916, "prompt_id": "1ba77d2f5f7df1e4", "reading_canonical_log_prob": -239.7144775390625, "reading_distances": [17, 17, 17, 19], "reading_log_probs": [-236.6970672607422, -238.7820281982422, -239.06674194335938, -238.695068359375], "reading_logprob_std": 1.087090681043595, "tokenization_path_logprob_std": 132.12476650808077, "trajectory_balance_rmse": 5818.065550392408} +{"canonical_length": 10, "canonical_nll": 0.4000696659088135, "generation_rouge_l_f1": 0.5333333333333333, "generation_token_f1": 0.6666666666666666, "prompt_id": "78c9a1c087a98da8", "reading_canonical_log_prob": -4.000696659088135, "reading_distances": [16, 21, 21, 21], "reading_log_probs": [-4.986835479736328, -5.531436443328857, -5.829042434692383, -6.314491271972656], "reading_logprob_std": 0.5557260589801671, "tokenization_path_logprob_std": 50.14254303198095, "trajectory_balance_rmse": 188.74859595721233} +{"canonical_length": 128, "canonical_nll": 1.05869460105896, "generation_rouge_l_f1": 0.1837837837837838, "generation_token_f1": 0.372972972972973, "prompt_id": "6de4ae8779beb5f7", "reading_canonical_log_prob": -135.51290893554688, "reading_distances": [38, 40, 33, 35], "reading_log_probs": [-135.466064453125, -134.3790283203125, -136.13401794433594, -132.2999725341797], "reading_logprob_std": 1.6771581422427568, "tokenization_path_logprob_std": 181.6470134776444, "trajectory_balance_rmse": 2725.099759945311} +{"canonical_length": 292, "canonical_nll": 0.9066303200917701, "generation_rouge_l_f1": 0.3007246376811594, "generation_token_f1": 0.5615942028985508, "prompt_id": "bcb2851b6b713edf", "reading_canonical_log_prob": -264.7360534667969, "reading_distances": [46, 47, 43, 50], "reading_log_probs": [-263.31591796875, -263.9115905761719, -265.70953369140625, -265.52685546875], "reading_logprob_std": 1.1848881158057774, "tokenization_path_logprob_std": 199.24241380731402, "trajectory_balance_rmse": 5362.885385814434} +{"canonical_length": 207, "canonical_nll": 1.609183933423913, "generation_rouge_l_f1": 0.2982456140350877, "generation_token_f1": 0.5263157894736842, "prompt_id": "790529293961a686", "reading_canonical_log_prob": -333.10107421875, "reading_distances": [30, 26, 26, 35], "reading_log_probs": [-331.1263427734375, -332.62567138671875, -331.0154113769531, -328.1472473144531], "reading_logprob_std": 1.871071136599441, "tokenization_path_logprob_std": 146.93671914134794, "trajectory_balance_rmse": 4734.0425074362975} +{"canonical_length": 130, "canonical_nll": 1.759496600811298, "generation_rouge_l_f1": 0.20472440944881892, "generation_token_f1": 0.33070866141732286, "prompt_id": "ecb22237f8778fd5", "reading_canonical_log_prob": -228.73455810546875, "reading_distances": [19, 21, 20, 14], "reading_log_probs": [-230.37266540527344, -229.32574462890625, -230.7551727294922, -229.5767364501953], "reading_logprob_std": 0.6690061982811895, "tokenization_path_logprob_std": 165.16262100867243, "trajectory_balance_rmse": 2629.8622206118607} +{"canonical_length": 37, "canonical_nll": 1.2326850891113281, "generation_rouge_l_f1": 0.05387205387205387, "generation_token_f1": 0.08754208754208753, "prompt_id": "db3b67bebed71ebb", "reading_canonical_log_prob": -45.60934829711914, "reading_distances": [29, 33, 32, 29], "reading_log_probs": [-44.50115203857422, -44.99432373046875, -44.53448486328125, -45.65480422973633], "reading_logprob_std": 0.5383651587294584, "tokenization_path_logprob_std": 35.53522679633727, "trajectory_balance_rmse": 840.2479241772357} +{"canonical_length": 10, "canonical_nll": 0.38550610542297364, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.8571428571428571, "prompt_id": "a75ceffa04e21449", "reading_canonical_log_prob": -3.8550610542297363, "reading_distances": [38, 37, 35, 32], "reading_log_probs": [-4.772831916809082, -4.85196590423584, -4.434940338134766, -4.764355182647705], "reading_logprob_std": 0.18497857151386524, "tokenization_path_logprob_std": 30.529950251221234, "trajectory_balance_rmse": 150.38289053467847} +{"canonical_length": 7, "canonical_nll": 0.04848673513957432, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "7af908712adcd78e", "reading_canonical_log_prob": -0.33940714597702026, "reading_distances": [31, 36, 38, 38], "reading_log_probs": [-0.7108153104782104, -2.833611011505127, -2.722865581512451, -0.7165200710296631], "reading_logprob_std": 1.19283979553355, "tokenization_path_logprob_std": 67.31554540070213, "trajectory_balance_rmse": 189.60719209677913} +{"canonical_length": 45, "canonical_nll": 1.097307840983073, "generation_rouge_l_f1": 0.05263157894736842, "generation_token_f1": 0.05263157894736842, "prompt_id": "3c24680236bfa43e", "reading_canonical_log_prob": -49.37885284423828, "reading_distances": [26, 20, 23, 28], "reading_log_probs": [-46.43682861328125, -46.22419738769531, -46.90016174316406, -47.53586196899414], "reading_logprob_std": 0.5808951788066159, "tokenization_path_logprob_std": 140.6499341121692, "trajectory_balance_rmse": 820.0894241290281} +{"canonical_length": 22, "canonical_nll": 0.45793424953113904, "generation_rouge_l_f1": 0.10702341137123746, "generation_token_f1": 0.11371237458193979, "prompt_id": "8be7f076246178b6", "reading_canonical_log_prob": -10.074553489685059, "reading_distances": [28, 32, 38, 35], "reading_log_probs": [-10.535174369812012, -10.485359191894531, -10.991539001464844, -10.449426651000977], "reading_logprob_std": 0.2532287911311684, "tokenization_path_logprob_std": 64.44067164296743, "trajectory_balance_rmse": 500.6037623793623} +{"canonical_length": 20, "canonical_nll": 1.309000015258789, "generation_rouge_l_f1": 0.5833333333333334, "generation_token_f1": 0.5833333333333334, "prompt_id": "ba65e8996b196d2c", "reading_canonical_log_prob": -26.18000030517578, "reading_distances": [39, 40, 26, 40], "reading_log_probs": [-26.27469825744629, -25.66522216796875, -24.8404483795166, -25.692508697509766], "reading_logprob_std": 0.5898079943221602, "tokenization_path_logprob_std": 72.52144265485727, "trajectory_balance_rmse": 350.69390475263} +{"canonical_length": 8, "canonical_nll": 1.223426342010498, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.4444444444444444, "prompt_id": "e026b81849f6617b", "reading_canonical_log_prob": -9.787410736083984, "reading_distances": [29, 25, 43, 28], "reading_log_probs": [-11.003327369689941, -10.8558988571167, -10.925796508789062, -10.419809341430664], "reading_logprob_std": 0.26129840077121796, "tokenization_path_logprob_std": 64.6323716761505, "trajectory_balance_rmse": 225.17111826939643} +{"canonical_length": 84, "canonical_nll": 2.5596702212379094, "generation_rouge_l_f1": 0.4077669902912621, "generation_token_f1": 0.4077669902912621, "prompt_id": "1c6873b5ff06ab16", "reading_canonical_log_prob": -215.01229858398438, "reading_distances": [22, 21, 23, 21], "reading_log_probs": [-212.42013549804688, -215.20909118652344, -211.76547241210938, -210.9908905029297], "reading_logprob_std": 1.8371468651380496, "tokenization_path_logprob_std": 68.3317350395582, "trajectory_balance_rmse": 2004.709859999286} +{"canonical_length": 212, "canonical_nll": 0.861411544511903, "generation_rouge_l_f1": 0.27368421052631575, "generation_token_f1": 0.3957894736842105, "prompt_id": "9cabb6227a1b5412", "reading_canonical_log_prob": -182.61924743652344, "reading_distances": [16, 14, 17, 20], "reading_log_probs": [-182.84341430664062, -180.82977294921875, -184.96865844726562, -181.41845703125], "reading_logprob_std": 1.8412579399903124, "tokenization_path_logprob_std": 114.13593625247661, "trajectory_balance_rmse": 3643.3947498507177} +{"canonical_length": 123, "canonical_nll": 1.4874490877477133, "generation_rouge_l_f1": 0.23076923076923075, "generation_token_f1": 0.4717948717948718, "prompt_id": "3aaeb505baa59e0c", "reading_canonical_log_prob": -182.95623779296875, "reading_distances": [16, 18, 18, 19], "reading_log_probs": [-182.36184692382812, -181.68295288085938, -182.38291931152344, -183.06344604492188], "reading_logprob_std": 0.563649825450976, "tokenization_path_logprob_std": 174.85801762247556, "trajectory_balance_rmse": 2771.0254043179903} +{"canonical_length": 305, "canonical_nll": 0.9890316822489754, "generation_rouge_l_f1": 0.28057553956834536, "generation_token_f1": 0.4676258992805756, "prompt_id": "5964386ba6fbd7d7", "reading_canonical_log_prob": -301.6546630859375, "reading_distances": [30, 32, 37, 40], "reading_log_probs": [-300.7191467285156, -296.5392150878906, -297.76202392578125, -297.2835388183594], "reading_logprob_std": 1.832530872201627, "tokenization_path_logprob_std": 96.11978666587505, "trajectory_balance_rmse": 6088.129953693103} +{"canonical_length": 125, "canonical_nll": 1.0272264404296876, "generation_rouge_l_f1": 0.25069637883008355, "generation_token_f1": 0.38997214484679665, "prompt_id": "a562f2a21ae863a6", "reading_canonical_log_prob": -128.40330505371094, "reading_distances": [10, 8, 8, 7], "reading_log_probs": [-128.87823486328125, -127.39289093017578, -126.63007354736328, -126.70210266113281], "reading_logprob_std": 1.0432439091854055, "tokenization_path_logprob_std": 138.60514050143195, "trajectory_balance_rmse": 2609.9760849187956} +{"canonical_length": 166, "canonical_nll": 1.1048253070877259, "generation_rouge_l_f1": 0.17518248175182483, "generation_token_f1": 0.3357664233576642, "prompt_id": "f84f62888219bcaf", "reading_canonical_log_prob": -183.4010009765625, "reading_distances": [87, 91, 90, 94], "reading_log_probs": [-181.02273559570312, -181.93441772460938, -180.7560272216797, -182.56301879882812], "reading_logprob_std": 0.8328531310380958, "tokenization_path_logprob_std": 62.06712655752997, "trajectory_balance_rmse": 2583.009379787582} +{"canonical_length": 86, "canonical_nll": 0.6658691583677779, "generation_rouge_l_f1": 0.06976744186046512, "generation_token_f1": 0.06976744186046512, "prompt_id": "6b3af0ffe0ba6c9d", "reading_canonical_log_prob": -57.264747619628906, "reading_distances": [32, 26, 25, 21], "reading_log_probs": [-57.863243103027344, -57.416133880615234, -59.07623291015625, -57.89335632324219], "reading_logprob_std": 0.7103417488842729, "tokenization_path_logprob_std": 144.47221557659913, "trajectory_balance_rmse": 1946.1568208214821} +{"canonical_length": 61, "canonical_nll": 0.9910483438460553, "generation_rouge_l_f1": 0.4337349397590362, "generation_token_f1": 0.5783132530120482, "prompt_id": "53e2d36ebe126c1f", "reading_canonical_log_prob": -60.453948974609375, "reading_distances": [22, 21, 23, 24], "reading_log_probs": [-63.14604187011719, -64.36095428466797, -65.88374328613281, -63.445919036865234], "reading_logprob_std": 1.230179880188298, "tokenization_path_logprob_std": 71.51711404842246, "trajectory_balance_rmse": 1241.4327272559578} +{"canonical_length": 133, "canonical_nll": 0.828750036712876, "generation_rouge_l_f1": 0.3272727272727273, "generation_token_f1": 0.5163636363636364, "prompt_id": "bce0ef5bdff3466e", "reading_canonical_log_prob": -110.2237548828125, "reading_distances": [23, 21, 18, 19], "reading_log_probs": [-111.91091918945312, -111.84899139404297, -110.65680694580078, -110.26516723632812], "reading_logprob_std": 0.8350806570807656, "tokenization_path_logprob_std": 98.43387497085492, "trajectory_balance_rmse": 2632.213432627389} +{"canonical_length": 74, "canonical_nll": 2.2996566360061235, "generation_rouge_l_f1": 0.2584269662921348, "generation_token_f1": 0.3595505617977528, "prompt_id": "2c030e502077ba6b", "reading_canonical_log_prob": -170.17459106445312, "reading_distances": [32, 25, 29, 23], "reading_log_probs": [-172.43638610839844, -171.47830200195312, -171.84976196289062, -171.02349853515625], "reading_logprob_std": 0.597623368417182, "tokenization_path_logprob_std": 28.111886458032753, "trajectory_balance_rmse": 1335.6643239770476} +{"canonical_length": 152, "canonical_nll": 1.4111338163677014, "generation_rouge_l_f1": 0.18999999999999997, "generation_token_f1": 0.33999999999999997, "prompt_id": "b73bca8ae43941f1", "reading_canonical_log_prob": -214.49234008789062, "reading_distances": [18, 23, 21, 12], "reading_log_probs": [-218.78477478027344, -220.0049591064453, -218.49240112304688, -217.75213623046875], "reading_logprob_std": 0.9377148284823251, "tokenization_path_logprob_std": 77.5633937701651, "trajectory_balance_rmse": 3298.806377347088} +{"canonical_length": 139, "canonical_nll": 1.3882707554659397, "generation_rouge_l_f1": 0.17560975609756102, "generation_token_f1": 0.31219512195121957, "prompt_id": "98eb458a641b4b4a", "reading_canonical_log_prob": -192.96963500976562, "reading_distances": [37, 38, 30, 23], "reading_log_probs": [-193.0074462890625, -193.0692901611328, -191.66104125976562, -191.25833129882812], "reading_logprob_std": 0.926505383069954, "tokenization_path_logprob_std": 132.46253634565684, "trajectory_balance_rmse": 3503.146782066274} +{"canonical_length": 79, "canonical_nll": 1.8493206169031844, "generation_rouge_l_f1": 0.20754716981132074, "generation_token_f1": 0.33962264150943394, "prompt_id": "3cfac8c1932346e3", "reading_canonical_log_prob": -146.09632873535156, "reading_distances": [22, 28, 26, 21], "reading_log_probs": [-144.725830078125, -145.63914489746094, -146.450927734375, -144.7505340576172], "reading_logprob_std": 0.824149835910449, "tokenization_path_logprob_std": 83.04857646593332, "trajectory_balance_rmse": 1709.2040177484694} +{"canonical_length": 22, "canonical_nll": 1.1969670382412998, "generation_rouge_l_f1": 0.05128205128205128, "generation_token_f1": 0.05128205128205128, "prompt_id": "7f70df4099ae17f5", "reading_canonical_log_prob": -26.333274841308594, "reading_distances": [28, 29, 27, 31], "reading_log_probs": [-26.78912925720215, -28.61693572998047, -28.962623596191406, -27.235721588134766], "reading_logprob_std": 1.0517383231863726, "tokenization_path_logprob_std": 61.99457933449234, "trajectory_balance_rmse": 489.6042434225869} +{"canonical_length": 87, "canonical_nll": 0.5939362624595905, "generation_rouge_l_f1": 0.4117647058823529, "generation_token_f1": 0.5661764705882353, "prompt_id": "0f424e2f0c95df2d", "reading_canonical_log_prob": -51.672454833984375, "reading_distances": [21, 22, 21, 18], "reading_log_probs": [-50.02240753173828, -50.60034942626953, -50.83778381347656, -51.315574645996094], "reading_logprob_std": 0.5375362609842078, "tokenization_path_logprob_std": 117.28512999799796, "trajectory_balance_rmse": 1668.2381838501149} +{"canonical_length": 222, "canonical_nll": 1.118125159461219, "generation_rouge_l_f1": 0.27125506072874495, "generation_token_f1": 0.51417004048583, "prompt_id": "7d4605b87d9b1016", "reading_canonical_log_prob": -248.22378540039062, "reading_distances": [26, 19, 24, 24], "reading_log_probs": [-249.89617919921875, -249.55465698242188, -250.45602416992188, -251.43844604492188], "reading_logprob_std": 0.8233532867850687, "tokenization_path_logprob_std": 273.18330333775464, "trajectory_balance_rmse": 5353.74151150851} +{"canonical_length": 68, "canonical_nll": 2.1617144416360294, "generation_rouge_l_f1": 0.19428571428571428, "generation_token_f1": 0.29714285714285715, "prompt_id": "7bba6cec3efa18f3", "reading_canonical_log_prob": -146.99658203125, "reading_distances": [16, 18, 20, 17], "reading_log_probs": [-146.24423217773438, -147.40829467773438, -147.65985107421875, -145.80511474609375], "reading_logprob_std": 0.8956076076913091, "tokenization_path_logprob_std": 132.27856162025805, "trajectory_balance_rmse": 1575.3204277337504} +{"canonical_length": 10, "canonical_nll": 0.6574690341949463, "generation_rouge_l_f1": 0.5263157894736842, "generation_token_f1": 0.5263157894736842, "prompt_id": "23ac4432bd5b4331", "reading_canonical_log_prob": -6.574690341949463, "reading_distances": [23, 18, 24, 23], "reading_log_probs": [-9.607294082641602, -7.200238227844238, -7.9644365310668945, -8.998923301696777], "reading_logprob_std": 1.0705311668413438, "tokenization_path_logprob_std": 47.21908511085319, "trajectory_balance_rmse": 236.75842923431077} +{"canonical_length": 35, "canonical_nll": 0.7416265760149274, "generation_rouge_l_f1": 0.14754098360655737, "generation_token_f1": 0.21311475409836064, "prompt_id": "f25ec758946024eb", "reading_canonical_log_prob": -25.95693016052246, "reading_distances": [37, 39, 42, 43], "reading_log_probs": [-26.806781768798828, -28.3355712890625, -27.718706130981445, -27.253280639648438], "reading_logprob_std": 0.6542592020959167, "tokenization_path_logprob_std": 52.132439376538315, "trajectory_balance_rmse": 424.98950361895726} +{"canonical_length": 225, "canonical_nll": 1.5074031575520834, "generation_rouge_l_f1": 0.2751540041067762, "generation_token_f1": 0.48049281314168374, "prompt_id": "e70426e74522ced9", "reading_canonical_log_prob": -339.16571044921875, "reading_distances": [29, 30, 30, 31], "reading_log_probs": [-347.11993408203125, -348.91668701171875, -345.88427734375, -345.133056640625], "reading_logprob_std": 1.6527388974221577, "tokenization_path_logprob_std": 275.8579076230442, "trajectory_balance_rmse": 4469.023432696317} +{"canonical_length": 17, "canonical_nll": 0.3494947938358082, "generation_rouge_l_f1": 0.5185185185185185, "generation_token_f1": 0.5185185185185185, "prompt_id": "fa1698c1de8ba5af", "reading_canonical_log_prob": -5.94141149520874, "reading_distances": [22, 23, 21, 22], "reading_log_probs": [-8.038515090942383, -7.755838394165039, -7.329394817352295, -7.292163848876953], "reading_logprob_std": 0.35800747280569123, "tokenization_path_logprob_std": 38.47262969267204, "trajectory_balance_rmse": 336.2572022949444} +{"canonical_length": 41, "canonical_nll": 1.2746574587938262, "generation_rouge_l_f1": 0.31081081081081074, "generation_token_f1": 0.32432432432432434, "prompt_id": "2ce41f448f87f273", "reading_canonical_log_prob": -52.260955810546875, "reading_distances": [38, 37, 28, 41], "reading_log_probs": [-49.3395881652832, -50.5054931640625, -49.477081298828125, -50.216590881347656], "reading_logprob_std": 0.5653422641850014, "tokenization_path_logprob_std": 66.05373421711849, "trajectory_balance_rmse": 809.255905501528} +{"canonical_length": 17, "canonical_nll": 0.5635455636417165, "generation_rouge_l_f1": 0.12230215827338127, "generation_token_f1": 0.12230215827338127, "prompt_id": "f7f7943a60f6d582", "reading_canonical_log_prob": -9.58027458190918, "reading_distances": [33, 25, 33, 33], "reading_log_probs": [-9.900559425354004, -11.743240356445312, -12.110119819641113, -10.77114486694336], "reading_logprob_std": 0.996158172936319, "tokenization_path_logprob_std": 28.25245570042247, "trajectory_balance_rmse": 351.74325354751153} +{"canonical_length": 45, "canonical_nll": 2.3584662543402777, "generation_rouge_l_f1": 0.1271186440677966, "generation_token_f1": 0.1694915254237288, "prompt_id": "d6cc1411da921bc9", "reading_canonical_log_prob": -106.1309814453125, "reading_distances": [17, 21, 19, 19], "reading_log_probs": [-104.71891021728516, -105.3153076171875, -106.63478088378906, -106.91181945800781], "reading_logprob_std": 1.0488768333392007, "tokenization_path_logprob_std": 28.637569452731704, "trajectory_balance_rmse": 1101.751765388743} +{"canonical_length": 19, "canonical_nll": 0.3488219411749589, "generation_rouge_l_f1": 0.21739130434782608, "generation_token_f1": 0.23188405797101452, "prompt_id": "9f836d8d503579f9", "reading_canonical_log_prob": -6.627616882324219, "reading_distances": [34, 42, 46, 38], "reading_log_probs": [-8.109380722045898, -8.58411979675293, -7.861033916473389, -7.79533052444458], "reading_logprob_std": 0.35765899653133887, "tokenization_path_logprob_std": 64.7983234136344, "trajectory_balance_rmse": 341.3666811399251} +{"canonical_length": 193, "canonical_nll": 1.4539281024834034, "generation_rouge_l_f1": 0.26754385964912275, "generation_token_f1": 0.4342105263157895, "prompt_id": "d3c6858742c763af", "reading_canonical_log_prob": -280.6081237792969, "reading_distances": [35, 34, 29, 42], "reading_log_probs": [-288.2548828125, -288.173095703125, -287.9568176269531, -290.0053405761719], "reading_logprob_std": 0.9469238847891104, "tokenization_path_logprob_std": 71.93954469010419, "trajectory_balance_rmse": 4829.729918069677} +{"canonical_length": 19, "canonical_nll": 0.7619344811690482, "generation_rouge_l_f1": 0.5217391304347826, "generation_token_f1": 0.5217391304347826, "prompt_id": "923c3d05ff40a1cb", "reading_canonical_log_prob": -14.476755142211914, "reading_distances": [31, 34, 27, 31], "reading_log_probs": [-16.604496002197266, -16.060869216918945, -15.586343765258789, -16.792343139648438], "reading_logprob_std": 0.5463601987424165, "tokenization_path_logprob_std": 19.174817427138866, "trajectory_balance_rmse": 567.5845202542031} +{"canonical_length": 200, "canonical_nll": 1.4241213989257813, "generation_rouge_l_f1": 0.3435897435897436, "generation_token_f1": 0.5692307692307693, "prompt_id": "180d045694c4d4f7", "reading_canonical_log_prob": -284.82427978515625, "reading_distances": [12, 11, 15, 8], "reading_log_probs": [-283.604736328125, -282.86297607421875, -283.20733642578125, -282.47515869140625], "reading_logprob_std": 0.48210918566987987, "tokenization_path_logprob_std": 244.81744372546945, "trajectory_balance_rmse": 4237.562127704442} +{"canonical_length": 314, "canonical_nll": 0.9739049437699044, "generation_rouge_l_f1": 0.2252559726962457, "generation_token_f1": 0.4675767918088737, "prompt_id": "1350704a8f43bb33", "reading_canonical_log_prob": -305.80615234375, "reading_distances": [33, 45, 36, 44], "reading_log_probs": [-308.88836669921875, -310.1300048828125, -309.2131652832031, -312.03668212890625], "reading_logprob_std": 1.4144289157959795, "tokenization_path_logprob_std": 387.4794786767062, "trajectory_balance_rmse": 6797.773976625852} +{"canonical_length": 79, "canonical_nll": 0.621457256848299, "generation_rouge_l_f1": 0.29197080291970806, "generation_token_f1": 0.48175182481751827, "prompt_id": "3269989441292641", "reading_canonical_log_prob": -49.095123291015625, "reading_distances": [29, 31, 32, 34], "reading_log_probs": [-49.2197265625, -50.55651092529297, -51.486976623535156, -50.46711730957031], "reading_logprob_std": 0.9308271233230551, "tokenization_path_logprob_std": 57.8937521642013, "trajectory_balance_rmse": 1895.8153581320532} +{"canonical_length": 204, "canonical_nll": 1.111798828723384, "generation_rouge_l_f1": 0.24786324786324784, "generation_token_f1": 0.5427350427350427, "prompt_id": "0f5981f2ed369cdc", "reading_canonical_log_prob": -226.8069610595703, "reading_distances": [28, 42, 38, 33], "reading_log_probs": [-229.98191833496094, -233.17657470703125, -231.00277709960938, -231.36865234375], "reading_logprob_std": 1.3322555175360087, "tokenization_path_logprob_std": 282.80605404977484, "trajectory_balance_rmse": 5242.721817457495} +{"canonical_length": 53, "canonical_nll": 0.9805329520747347, "generation_rouge_l_f1": 0.20535714285714288, "generation_token_f1": 0.3214285714285714, "prompt_id": "801510c180dde0a4", "reading_canonical_log_prob": -51.96824645996094, "reading_distances": [34, 36, 39, 36], "reading_log_probs": [-51.36524963378906, -50.54463577270508, -51.184410095214844, -52.51556396484375], "reading_logprob_std": 0.8213389908599732, "tokenization_path_logprob_std": 86.73384024020343, "trajectory_balance_rmse": 1268.4480924568416} +{"canonical_length": 18, "canonical_nll": 1.975206798977322, "generation_rouge_l_f1": 0.17435897435897435, "generation_token_f1": 0.18461538461538463, "prompt_id": "a62ba8086bae83ba", "reading_canonical_log_prob": -35.5537223815918, "reading_distances": [21, 20, 17, 18], "reading_log_probs": [-34.70103454589844, -34.84967041015625, -34.35673904418945, -36.689414978027344], "reading_logprob_std": 1.0473501155449965, "tokenization_path_logprob_std": 39.50899465178051, "trajectory_balance_rmse": 346.52099889085923} +{"canonical_length": 278, "canonical_nll": 1.1725826126208407, "generation_rouge_l_f1": 0.17319587628865982, "generation_token_f1": 0.40824742268041236, "prompt_id": "6197952fa284f196", "reading_canonical_log_prob": -325.97796630859375, "reading_distances": [20, 24, 23, 23], "reading_log_probs": [-322.85101318359375, -323.5076599121094, -324.0028381347656, -323.5562438964844], "reading_logprob_std": 0.47453874903367543, "tokenization_path_logprob_std": 95.48062940985443, "trajectory_balance_rmse": 5896.130315274239} +{"canonical_length": 13, "canonical_nll": 0.9020236822275015, "generation_rouge_l_f1": 0.5217391304347826, "generation_token_f1": 0.6086956521739131, "prompt_id": "d2ae9ca225f37e77", "reading_canonical_log_prob": -11.72630786895752, "reading_distances": [24, 28, 33, 31], "reading_log_probs": [-11.854166984558105, -11.399702072143555, -11.458812713623047, -15.580522537231445], "reading_logprob_std": 2.0149403374203954, "tokenization_path_logprob_std": 19.984063506547383, "trajectory_balance_rmse": 278.7830600543322} +{"canonical_length": 14, "canonical_nll": 2.456315721784319, "generation_rouge_l_f1": 0.4, "generation_token_f1": 0.4, "prompt_id": "3d0956e900302c45", "reading_canonical_log_prob": -34.38842010498047, "reading_distances": [19, 18, 20, 17], "reading_log_probs": [-41.880165100097656, -41.2913703918457, -38.152191162109375, -33.73951721191406], "reading_logprob_std": 3.7290664717477355, "tokenization_path_logprob_std": 37.53972475252191, "trajectory_balance_rmse": 289.1091546643347} +{"canonical_length": 13, "canonical_nll": 0.33706305577204776, "generation_rouge_l_f1": 0.05090909090909091, "generation_token_f1": 0.06545454545454546, "prompt_id": "360b4e609ec3ff2b", "reading_canonical_log_prob": -4.381819725036621, "reading_distances": [22, 16, 20, 23], "reading_log_probs": [-4.336426258087158, -4.4769768714904785, -4.475541114807129, -4.328832626342773], "reading_logprob_std": 0.08298455724132751, "tokenization_path_logprob_std": 20.593873998032603, "trajectory_balance_rmse": 115.01143502520121} +{"canonical_length": 86, "canonical_nll": 1.679521072742551, "generation_rouge_l_f1": 0.2872340425531915, "generation_token_f1": 0.48936170212765956, "prompt_id": "1ad837078d8b4fb2", "reading_canonical_log_prob": -144.43881225585938, "reading_distances": [31, 36, 32, 30], "reading_log_probs": [-143.14532470703125, -142.7490997314453, -143.62417602539062, -141.74264526367188], "reading_logprob_std": 0.799618115124113, "tokenization_path_logprob_std": 180.43115673360305, "trajectory_balance_rmse": 1888.1188605814646} +{"canonical_length": 20, "canonical_nll": 0.711705493927002, "generation_rouge_l_f1": 0.12363636363636364, "generation_token_f1": 0.12363636363636364, "prompt_id": "24f992c3a95d7b8a", "reading_canonical_log_prob": -14.234109878540039, "reading_distances": [38, 41, 32, 31], "reading_log_probs": [-13.555185317993164, -13.418112754821777, -13.580957412719727, -13.451007843017578], "reading_logprob_std": 0.07894797592525764, "tokenization_path_logprob_std": 37.705902083130596, "trajectory_balance_rmse": 550.2349733660681} +{"canonical_length": 29, "canonical_nll": 0.7895263803416285, "generation_rouge_l_f1": 0.2692307692307692, "generation_token_f1": 0.3846153846153846, "prompt_id": "429af0c846ac90d6", "reading_canonical_log_prob": -22.896265029907227, "reading_distances": [20, 20, 14, 22], "reading_log_probs": [-25.60879898071289, -25.78135871887207, -24.686418533325195, -25.309764862060547], "reading_logprob_std": 0.48130373105508156, "tokenization_path_logprob_std": 68.52775930789512, "trajectory_balance_rmse": 682.0390971145681} +{"canonical_length": 132, "canonical_nll": 1.160652969822739, "generation_rouge_l_f1": 0.2179176755447942, "generation_token_f1": 0.4358353510895884, "prompt_id": "a4d2b99ec03649f1", "reading_canonical_log_prob": -153.20619201660156, "reading_distances": [19, 21, 19, 21], "reading_log_probs": [-153.82003784179688, -155.6219024658203, -154.11227416992188, -152.16526794433594], "reading_logprob_std": 1.4168190503639695, "tokenization_path_logprob_std": 63.909812113958395, "trajectory_balance_rmse": 2799.2597246434148} +{"canonical_length": 27, "canonical_nll": 0.5842949196144387, "generation_rouge_l_f1": 0.2042553191489362, "generation_token_f1": 0.2042553191489362, "prompt_id": "363a6f01446a3b83", "reading_canonical_log_prob": -15.775962829589844, "reading_distances": [45, 40, 44, 33], "reading_log_probs": [-15.856888771057129, -15.180269241333008, -15.645794868469238, -15.664252281188965], "reading_logprob_std": 0.2873409428732854, "tokenization_path_logprob_std": 43.43290740971329, "trajectory_balance_rmse": 722.5904130268009} +{"canonical_length": 56, "canonical_nll": 1.9802792412894112, "generation_rouge_l_f1": 0.09655172413793103, "generation_token_f1": 0.2068965517241379, "prompt_id": "15d4cbd9fbd2e327", "reading_canonical_log_prob": -110.89563751220703, "reading_distances": [28, 29, 30, 28], "reading_log_probs": [-110.4952392578125, -110.70565795898438, -112.00231170654297, -110.12364196777344], "reading_logprob_std": 0.8166538171067886, "tokenization_path_logprob_std": 42.84939307328882, "trajectory_balance_rmse": 1333.746331948303} +{"canonical_length": 56, "canonical_nll": 1.0119959967476981, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.39097744360902253, "prompt_id": "c741581c2c318b9d", "reading_canonical_log_prob": -56.671775817871094, "reading_distances": [19, 28, 30, 29], "reading_log_probs": [-57.10033416748047, -57.800750732421875, -57.49092102050781, -59.548553466796875], "reading_logprob_std": 1.0809555697176159, "tokenization_path_logprob_std": 82.76304210912362, "trajectory_balance_rmse": 1100.1762134341222} +{"canonical_length": 10, "canonical_nll": 0.8215114593505859, "generation_rouge_l_f1": 0.11155378486055778, "generation_token_f1": 0.12749003984063745, "prompt_id": "1551065a718dd716", "reading_canonical_log_prob": -8.21511459350586, "reading_distances": [19, 24, 17, 29], "reading_log_probs": [-10.02885913848877, -10.299196243286133, -9.45287799835205, -9.971839904785156], "reading_logprob_std": 0.35365070119037373, "tokenization_path_logprob_std": 62.09038857674922, "trajectory_balance_rmse": 185.80962004629666} +{"canonical_length": 22, "canonical_nll": 1.5927600860595703, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.6111111111111112, "prompt_id": "0281e958164e52e0", "reading_canonical_log_prob": -35.04072189331055, "reading_distances": [20, 23, 25, 22], "reading_log_probs": [-36.360286712646484, -37.09598159790039, -35.26951217651367, -35.802513122558594], "reading_logprob_std": 0.7818401735176119, "tokenization_path_logprob_std": 46.561379395494264, "trajectory_balance_rmse": 442.0625880653953} +{"canonical_length": 338, "canonical_nll": 1.263502222546459, "generation_rouge_l_f1": 0.28058727569331154, "generation_token_f1": 0.4828711256117455, "prompt_id": "059afc2f6690f050", "reading_canonical_log_prob": -427.0637512207031, "reading_distances": [28, 20, 24, 34], "reading_log_probs": [-422.346923828125, -421.6510009765625, -422.21929931640625, -422.9683837890625], "reading_logprob_std": 0.5405550495617933, "tokenization_path_logprob_std": 353.94978761758995, "trajectory_balance_rmse": 6510.454003278822} +{"canonical_length": 11, "canonical_nll": 1.0450155084783381, "generation_rouge_l_f1": 0.049999999999999996, "generation_token_f1": 0.06428571428571428, "prompt_id": "3463b16713cbef14", "reading_canonical_log_prob": -11.495170593261719, "reading_distances": [37, 40, 33, 39], "reading_log_probs": [-10.858577728271484, -9.901597023010254, -10.26191520690918, -10.825037002563477], "reading_logprob_std": 0.46301727479905475, "tokenization_path_logprob_std": 47.12089242568431, "trajectory_balance_rmse": 295.8698671542311} +{"canonical_length": 184, "canonical_nll": 1.9362179300059443, "generation_rouge_l_f1": 0.23389021479713604, "generation_token_f1": 0.3675417661097852, "prompt_id": "75052e7415fb5476", "reading_canonical_log_prob": -356.26409912109375, "reading_distances": [25, 20, 30, 28], "reading_log_probs": [-355.0051574707031, -356.33953857421875, -354.95526123046875, -354.7082214355469], "reading_logprob_std": 0.7365279811839358, "tokenization_path_logprob_std": 236.79597573100665, "trajectory_balance_rmse": 4068.6587679629947} +{"canonical_length": 155, "canonical_nll": 2.2564019972278224, "generation_rouge_l_f1": 0.26022304832713755, "generation_token_f1": 0.4386617100371747, "prompt_id": "3e8a52f189b6e093", "reading_canonical_log_prob": -349.7423095703125, "reading_distances": [27, 25, 28, 31], "reading_log_probs": [-352.25494384765625, -355.28973388671875, -355.423828125, -357.05224609375], "reading_logprob_std": 2.000876112489806, "tokenization_path_logprob_std": 115.68469099316297, "trajectory_balance_rmse": 3486.052498221255} +{"canonical_length": 278, "canonical_nll": 1.440628381084195, "generation_rouge_l_f1": 0.20512820512820512, "generation_token_f1": 0.4542124542124542, "prompt_id": "39b34b632ecf238e", "reading_canonical_log_prob": -400.49468994140625, "reading_distances": [27, 21, 17, 23], "reading_log_probs": [-401.73138427734375, -400.216552734375, -401.1932678222656, -402.005859375], "reading_logprob_std": 0.7892779480167045, "tokenization_path_logprob_std": 82.95994810410139, "trajectory_balance_rmse": 6138.0373622464895} +{"canonical_length": 144, "canonical_nll": 1.9652173784044054, "generation_rouge_l_f1": 0.14536340852130328, "generation_token_f1": 0.3208020050125313, "prompt_id": "76cab7f49b66aca5", "reading_canonical_log_prob": -282.9913024902344, "reading_distances": [17, 14, 15, 11], "reading_log_probs": [-273.94464111328125, -280.3742980957031, -274.6285095214844, -275.0369567871094], "reading_logprob_std": 2.953380071989208, "tokenization_path_logprob_std": 185.66383486733184, "trajectory_balance_rmse": 2807.831880040646} +{"canonical_length": 40, "canonical_nll": 1.4780935287475585, "generation_rouge_l_f1": 0.3716814159292036, "generation_token_f1": 0.4247787610619469, "prompt_id": "b6ca714d2fc1e411", "reading_canonical_log_prob": -59.123741149902344, "reading_distances": [14, 16, 13, 18], "reading_log_probs": [-60.04133605957031, -60.189300537109375, -60.01139831542969, -60.64225769042969], "reading_logprob_std": 0.29136167327171575, "tokenization_path_logprob_std": 81.87461387030255, "trajectory_balance_rmse": 829.7228524107114} +{"canonical_length": 39, "canonical_nll": 1.3481631156725762, "generation_rouge_l_f1": 0.30357142857142855, "generation_token_f1": 0.46428571428571436, "prompt_id": "56f7c951f93e4273", "reading_canonical_log_prob": -52.57836151123047, "reading_distances": [34, 27, 29, 26], "reading_log_probs": [-51.95175552368164, -50.49736785888672, -51.647315979003906, -51.131797790527344], "reading_logprob_std": 0.6371092671724035, "tokenization_path_logprob_std": 99.21789323297145, "trajectory_balance_rmse": 811.5883646242888} +{"canonical_length": 17, "canonical_nll": 0.6354425654691809, "generation_rouge_l_f1": 0.23255813953488372, "generation_token_f1": 0.2790697674418605, "prompt_id": "3f9be20b25197a77", "reading_canonical_log_prob": -10.802523612976074, "reading_distances": [19, 14, 11, 10], "reading_log_probs": [-11.328486442565918, -9.354823112487793, -10.15070629119873, -10.474576950073242], "reading_logprob_std": 0.8166927891115109, "tokenization_path_logprob_std": 61.06123116348258, "trajectory_balance_rmse": 258.17455108364516} +{"canonical_length": 231, "canonical_nll": 1.3381869493624865, "generation_rouge_l_f1": 0.20042643923240935, "generation_token_f1": 0.40938166311300644, "prompt_id": "053c343603667759", "reading_canonical_log_prob": -309.1211853027344, "reading_distances": [22, 20, 18, 20], "reading_log_probs": [-307.1770935058594, -302.85638427734375, -305.48773193359375, -304.2864990234375], "reading_logprob_std": 1.8323524588291156, "tokenization_path_logprob_std": 134.8124227088084, "trajectory_balance_rmse": 4948.258297940576} +{"canonical_length": 53, "canonical_nll": 1.094731888681088, "generation_rouge_l_f1": 0.5599999999999999, "generation_token_f1": 0.5599999999999999, "prompt_id": "e23c57e239d810b9", "reading_canonical_log_prob": -58.020790100097656, "reading_distances": [29, 21, 23, 29], "reading_log_probs": [-58.39463424682617, -58.74266815185547, -57.74717330932617, -60.359100341796875], "reading_logprob_std": 1.111511213470958, "tokenization_path_logprob_std": 91.52676524550279, "trajectory_balance_rmse": 1052.7257973954138} +{"canonical_length": 66, "canonical_nll": 1.1104322491270122, "generation_rouge_l_f1": 0.1933085501858736, "generation_token_f1": 0.2899628252788104, "prompt_id": "e9b590e25ca62203", "reading_canonical_log_prob": -73.28852844238281, "reading_distances": [21, 22, 30, 20], "reading_log_probs": [-73.4744644165039, -72.284423828125, -73.41946411132812, -72.73710632324219], "reading_logprob_std": 0.5716765278618441, "tokenization_path_logprob_std": 134.5706612633623, "trajectory_balance_rmse": 1392.8035827144845} +{"canonical_length": 16, "canonical_nll": 1.657745361328125, "generation_rouge_l_f1": 0.0625, "generation_token_f1": 0.08333333333333336, "prompt_id": "d046e49475ca84ec", "reading_canonical_log_prob": -26.52392578125, "reading_distances": [37, 44, 42, 37], "reading_log_probs": [-27.66199493408203, -27.814983367919922, -27.89020538330078, -26.381366729736328], "reading_logprob_std": 0.7102231554912556, "tokenization_path_logprob_std": 98.8867849069284, "trajectory_balance_rmse": 511.4256986197} +{"canonical_length": 56, "canonical_nll": 0.9317019326346261, "generation_rouge_l_f1": 0.4077669902912621, "generation_token_f1": 0.5631067961165049, "prompt_id": "750881d0bf75959b", "reading_canonical_log_prob": -52.17530822753906, "reading_distances": [48, 44, 52, 48], "reading_log_probs": [-55.11922836303711, -55.60280990600586, -56.691219329833984, -55.84312438964844], "reading_logprob_std": 0.6576906135758461, "tokenization_path_logprob_std": 89.08565067261367, "trajectory_balance_rmse": 1361.7586827821947} +{"canonical_length": 2, "canonical_nll": 0.5499050617218018, "generation_rouge_l_f1": 0.013333333333333332, "generation_token_f1": 0.013333333333333332, "prompt_id": "73f8da9f02677521", "reading_canonical_log_prob": -1.0998101234436035, "reading_distances": [18, 14, 15, 17], "reading_log_probs": [-1.687907338142395, -1.5385138988494873, -1.4680452346801758, -1.0736626386642456], "reading_logprob_std": 0.2621298110106067, "tokenization_path_logprob_std": 8.78861344722444, "trajectory_balance_rmse": 50.032866892372155} +{"canonical_length": 156, "canonical_nll": 2.464812841170873, "generation_rouge_l_f1": 0.2422680412371134, "generation_token_f1": 0.3814432989690722, "prompt_id": "9c25897ff633e0cd", "reading_canonical_log_prob": -384.51080322265625, "reading_distances": [29, 34, 28, 35], "reading_log_probs": [-384.2742919921875, -386.5460205078125, -383.1744079589844, -389.1219787597656], "reading_logprob_std": 2.6338737766718783, "tokenization_path_logprob_std": 239.1450332827666, "trajectory_balance_rmse": 2911.57344307209} +{"canonical_length": 263, "canonical_nll": 1.0064707708902685, "generation_rouge_l_f1": 0.38240917782026773, "generation_token_f1": 0.5736137667304015, "prompt_id": "5d0b3ece32e43007", "reading_canonical_log_prob": -264.7018127441406, "reading_distances": [13, 21, 23, 25], "reading_log_probs": [-263.21197509765625, -260.6978454589844, -261.1274719238281, -263.498779296875], "reading_logprob_std": 1.4259840430219501, "tokenization_path_logprob_std": 101.92258957914459, "trajectory_balance_rmse": 6353.322647537094} +{"canonical_length": 21, "canonical_nll": 0.5106473196120489, "generation_rouge_l_f1": 0.4827586206896552, "generation_token_f1": 0.4827586206896552, "prompt_id": "16cea9158da9fd5a", "reading_canonical_log_prob": -10.723593711853027, "reading_distances": [31, 29, 34, 38], "reading_log_probs": [-12.023866653442383, -10.783964157104492, -11.572608947753906, -11.024673461914062], "reading_logprob_std": 0.5567400368250593, "tokenization_path_logprob_std": 48.11043069977041, "trajectory_balance_rmse": 467.1925953190543} +{"canonical_length": 18, "canonical_nll": 0.8619933658175998, "generation_rouge_l_f1": 0.3478260869565218, "generation_token_f1": 0.3478260869565218, "prompt_id": "d643de98c824ec5b", "reading_canonical_log_prob": -15.515880584716797, "reading_distances": [27, 23, 18, 22], "reading_log_probs": [-16.749263763427734, -15.740224838256836, -15.865373611450195, -16.500293731689453], "reading_logprob_std": 0.4880145086006665, "tokenization_path_logprob_std": 52.004490138968094, "trajectory_balance_rmse": 324.9771730236701} +{"canonical_length": 97, "canonical_nll": 1.7106776286646264, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.47126436781609193, "prompt_id": "cdeaf78291feca5e", "reading_canonical_log_prob": -165.93572998046875, "reading_distances": [37, 29, 32, 39], "reading_log_probs": [-177.12527465820312, -176.98460388183594, -177.89202880859375, -177.61468505859375], "reading_logprob_std": 0.42274544913236517, "tokenization_path_logprob_std": 68.5067567774424, "trajectory_balance_rmse": 1920.7754841852618} +{"canonical_length": 308, "canonical_nll": 1.7562517438616072, "generation_rouge_l_f1": 0.12903225806451615, "generation_token_f1": 0.18064516129032257, "prompt_id": "26a8450ba4613a47", "reading_canonical_log_prob": -540.925537109375, "reading_distances": [22, 17, 20, 17], "reading_log_probs": [-544.1469116210938, -545.0009765625, -542.746337890625, -543.297607421875], "reading_logprob_std": 0.9874677795360136, "tokenization_path_logprob_std": 173.36362399664202, "trajectory_balance_rmse": 5542.355414552324} +{"canonical_length": 104, "canonical_nll": 0.5345119329599234, "generation_rouge_l_f1": 0.5257731958762887, "generation_token_f1": 0.6391752577319587, "prompt_id": "c15330337653f483", "reading_canonical_log_prob": -55.58924102783203, "reading_distances": [281, 304, 302, 287], "reading_log_probs": [-56.91282653808594, -57.957923889160156, -61.926605224609375, -58.990570068359375], "reading_logprob_std": 2.1599443565507626, "tokenization_path_logprob_std": 149.7566426378672, "trajectory_balance_rmse": 1859.631586267861} +{"canonical_length": 11, "canonical_nll": 1.9353042949329724, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "8609ed9bb93a15ed", "reading_canonical_log_prob": -21.288347244262695, "reading_distances": [17, 14, 15, 19], "reading_log_probs": [-24.475156784057617, -24.313026428222656, -22.937328338623047, -25.782039642333984], "reading_logprob_std": 1.1634027979577115, "tokenization_path_logprob_std": 31.38889982339144, "trajectory_balance_rmse": 162.9751402976581} +{"canonical_length": 367, "canonical_nll": 1.3467467097560457, "generation_rouge_l_f1": 0.21974522292993628, "generation_token_f1": 0.4745222929936306, "prompt_id": "4cb8d591dfd52931", "reading_canonical_log_prob": -494.25604248046875, "reading_distances": [22, 25, 30, 31], "reading_log_probs": [-495.56005859375, -495.9986877441406, -497.20843505859375, -497.75787353515625], "reading_logprob_std": 1.0246965758114641, "tokenization_path_logprob_std": 176.61755138038058, "trajectory_balance_rmse": 8012.477519635513} +{"canonical_length": 277, "canonical_nll": 0.4210614917080325, "generation_rouge_l_f1": 0.4988235294117648, "generation_token_f1": 0.6164705882352941, "prompt_id": "7e37b68aba5b5c9c", "reading_canonical_log_prob": -116.634033203125, "reading_distances": [35, 28, 36, 36], "reading_log_probs": [-117.48504638671875, -117.92820739746094, -117.30381774902344, -119.66618347167969], "reading_logprob_std": 1.079266323294392, "tokenization_path_logprob_std": 118.39806899379738, "trajectory_balance_rmse": 4475.876776155538} +{"canonical_length": 87, "canonical_nll": 1.3846940665409482, "generation_rouge_l_f1": 0.15774647887323945, "generation_token_f1": 0.30422535211267604, "prompt_id": "fef695db887a3a23", "reading_canonical_log_prob": -120.4683837890625, "reading_distances": [17, 19, 19, 15], "reading_log_probs": [-124.83766174316406, -120.50511169433594, -125.34380340576172, -122.81999969482422], "reading_logprob_std": 2.2030160869796913, "tokenization_path_logprob_std": 30.06880317096995, "trajectory_balance_rmse": 1810.8027982350507} +{"canonical_length": 174, "canonical_nll": 0.6045607205094963, "generation_rouge_l_f1": 0.4468664850136239, "generation_token_f1": 0.7029972752043596, "prompt_id": "471b9b9cabf2decb", "reading_canonical_log_prob": -105.19356536865234, "reading_distances": [22, 24, 22, 22], "reading_log_probs": [-103.51014709472656, -103.57856750488281, -102.95481872558594, -103.87032318115234], "reading_logprob_std": 0.3824406365484608, "tokenization_path_logprob_std": 86.2872456267364, "trajectory_balance_rmse": 2804.2205466252667} +{"canonical_length": 10, "canonical_nll": 1.3327089309692384, "generation_rouge_l_f1": 0.06498194945848376, "generation_token_f1": 0.06498194945848376, "prompt_id": "71f95c366e84f3b3", "reading_canonical_log_prob": -13.327089309692383, "reading_distances": [29, 41, 33, 31], "reading_log_probs": [-11.361428260803223, -10.837533950805664, -10.740819931030273, -11.097070693969727], "reading_logprob_std": 0.2788566535787667, "tokenization_path_logprob_std": 37.85913612909798, "trajectory_balance_rmse": 286.35819306013855} +{"canonical_length": 7, "canonical_nll": 0.39148671286446707, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "78998e3018151a20", "reading_canonical_log_prob": -2.7404069900512695, "reading_distances": [15, 19, 16, 24], "reading_log_probs": [-4.476529121398926, -5.438956260681152, -4.531179904937744, -4.654647350311279], "reading_logprob_std": 0.4486480988331016, "tokenization_path_logprob_std": 25.139213981297402, "trajectory_balance_rmse": 111.30954410068233} +{"canonical_length": 179, "canonical_nll": 1.4969279539651712, "generation_rouge_l_f1": 0.14376321353065538, "generation_token_f1": 0.30866807610993663, "prompt_id": "3025be82a097d5b7", "reading_canonical_log_prob": -267.9501037597656, "reading_distances": [24, 21, 30, 26], "reading_log_probs": [-271.920654296875, -271.40374755859375, -277.1248474121094, -273.4136047363281], "reading_logprob_std": 2.583981926838298, "tokenization_path_logprob_std": 58.32453848757318, "trajectory_balance_rmse": 3665.897334476665} +{"canonical_length": 259, "canonical_nll": 1.7596952813932794, "generation_rouge_l_f1": 0.12154696132596686, "generation_token_f1": 0.19521178637200737, "prompt_id": "317ccf6fb834b830", "reading_canonical_log_prob": -455.7610778808594, "reading_distances": [20, 24, 29, 20], "reading_log_probs": [-450.84228515625, -449.6602783203125, -449.5027160644531, -451.500244140625], "reading_logprob_std": 0.958510738789822, "tokenization_path_logprob_std": 198.24788845941325, "trajectory_balance_rmse": 5676.997760812165} +{"canonical_length": 35, "canonical_nll": 1.057232448032924, "generation_rouge_l_f1": 0.29850746268656714, "generation_token_f1": 0.3582089552238806, "prompt_id": "43805108665c2c05", "reading_canonical_log_prob": -37.003135681152344, "reading_distances": [31, 17, 23, 25], "reading_log_probs": [-34.44585037231445, -35.12565612792969, -35.79206848144531, -35.12287139892578], "reading_logprob_std": 0.5496009497868912, "tokenization_path_logprob_std": 57.66118251757369, "trajectory_balance_rmse": 769.8466638734669} +{"canonical_length": 118, "canonical_nll": 1.4709612312963454, "generation_rouge_l_f1": 0.3164556962025316, "generation_token_f1": 0.48734177215189867, "prompt_id": "5056b26f2553e0fc", "reading_canonical_log_prob": -173.57342529296875, "reading_distances": [26, 26, 33, 14], "reading_log_probs": [-172.939697265625, -175.22381591796875, -174.59368896484375, -173.71249389648438], "reading_logprob_std": 1.0003232881555448, "tokenization_path_logprob_std": 103.8490665086534, "trajectory_balance_rmse": 2610.2218805854313} +{"canonical_length": 372, "canonical_nll": 0.9480719617618028, "generation_rouge_l_f1": 0.22857142857142856, "generation_token_f1": 0.4, "prompt_id": "1ea4408820560cde", "reading_canonical_log_prob": -352.6827697753906, "reading_distances": [12, 13, 13, 14], "reading_log_probs": [-360.8534851074219, -356.2761535644531, -355.0948181152344, -355.98931884765625], "reading_logprob_std": 2.582832459117524, "tokenization_path_logprob_std": 331.97919155157064, "trajectory_balance_rmse": 6644.25118169416} +{"canonical_length": 9, "canonical_nll": 0.7318431006537544, "generation_rouge_l_f1": 0.8571428571428571, "generation_token_f1": 0.8571428571428571, "prompt_id": "2b79f5bffeeb5daf", "reading_canonical_log_prob": -6.586587905883789, "reading_distances": [35, 37, 32, 32], "reading_log_probs": [-7.180337429046631, -7.037948131561279, -7.773674488067627, -6.50204610824585], "reading_logprob_std": 0.5226475613116006, "tokenization_path_logprob_std": 48.396370389048506, "trajectory_balance_rmse": 184.94112226398337} +{"canonical_length": 36, "canonical_nll": 0.9788318210177951, "generation_rouge_l_f1": 0.14388489208633093, "generation_token_f1": 0.1870503597122302, "prompt_id": "91748c9b4543a8c6", "reading_canonical_log_prob": -35.237945556640625, "reading_distances": [10, 13, 13, 11], "reading_log_probs": [-33.38871765136719, -33.33372497558594, -34.47254180908203, -33.55072784423828], "reading_logprob_std": 0.5321091141915327, "tokenization_path_logprob_std": 80.4889770551651, "trajectory_balance_rmse": 545.3963717403352} +{"canonical_length": 187, "canonical_nll": 1.1610255113897476, "generation_rouge_l_f1": 0.38421052631578945, "generation_token_f1": 0.5947368421052632, "prompt_id": "9a3c9e43bea33f6c", "reading_canonical_log_prob": -217.1117706298828, "reading_distances": [21, 17, 25, 16], "reading_log_probs": [-216.51504516601562, -215.68658447265625, -216.370849609375, -215.17950439453125], "reading_logprob_std": 0.6215210532224259, "tokenization_path_logprob_std": 37.83240024962413, "trajectory_balance_rmse": 4394.4463566653085} +{"canonical_length": 376, "canonical_nll": 1.2512320660530252, "generation_rouge_l_f1": 0.2371794871794872, "generation_token_f1": 0.45512820512820507, "prompt_id": "7473e8ab94e776ba", "reading_canonical_log_prob": -470.4632568359375, "reading_distances": [22, 19, 22, 14], "reading_log_probs": [-472.42974853515625, -472.8045654296875, -474.71478271484375, -470.8689880371094], "reading_logprob_std": 1.580700500344942, "tokenization_path_logprob_std": 97.22572736022546, "trajectory_balance_rmse": 7555.138819239052} +{"canonical_length": 87, "canonical_nll": 0.5000590181898796, "generation_rouge_l_f1": 0.27184466019417475, "generation_token_f1": 0.38834951456310673, "prompt_id": "81563152acabd240", "reading_canonical_log_prob": -43.50513458251953, "reading_distances": [40, 35, 40, 34], "reading_log_probs": [-45.12342071533203, -42.604400634765625, -44.79692077636719, -42.22627639770508], "reading_logprob_std": 1.4833476948310544, "tokenization_path_logprob_std": 114.51331464993645, "trajectory_balance_rmse": 1080.387178194924} +{"canonical_length": 19, "canonical_nll": 2.552821912263569, "generation_rouge_l_f1": 0.05555555555555555, "generation_token_f1": 0.05555555555555555, "prompt_id": "49b74500a1071228", "reading_canonical_log_prob": -48.50361633300781, "reading_distances": [22, 20, 18, 28], "reading_log_probs": [-47.12595748901367, -47.65103530883789, -49.11857604980469, -47.140167236328125], "reading_logprob_std": 0.9387580807352123, "tokenization_path_logprob_std": 56.240479735659186, "trajectory_balance_rmse": 507.42798887793293} +{"canonical_length": 83, "canonical_nll": 1.1004267313394203, "generation_rouge_l_f1": 0.23270440251572327, "generation_token_f1": 0.3018867924528301, "prompt_id": "c1a613d92b944071", "reading_canonical_log_prob": -91.33541870117188, "reading_distances": [36, 36, 31, 34], "reading_log_probs": [-92.7776870727539, -93.64320373535156, -92.8425521850586, -93.88211059570312], "reading_logprob_std": 0.5591570274472255, "tokenization_path_logprob_std": 63.933089935776316, "trajectory_balance_rmse": 1726.3706277656315} +{"canonical_length": 190, "canonical_nll": 1.317216732627467, "generation_rouge_l_f1": 0.28140703517587945, "generation_token_f1": 0.5427135678391959, "prompt_id": "00050e95a0ce36d0", "reading_canonical_log_prob": -250.27117919921875, "reading_distances": [34, 29, 22, 38], "reading_log_probs": [-251.49267578125, -251.87840270996094, -250.0809783935547, -251.81924438476562], "reading_logprob_std": 0.8418295733900865, "tokenization_path_logprob_std": 140.5722279410812, "trajectory_balance_rmse": 3822.3728943940814} +{"canonical_length": 309, "canonical_nll": 1.3127337705741808, "generation_rouge_l_f1": 0.18794326241134748, "generation_token_f1": 0.40070921985815605, "prompt_id": "3e9a72eac16389eb", "reading_canonical_log_prob": -405.6347351074219, "reading_distances": [18, 19, 20, 22], "reading_log_probs": [-405.8262939453125, -406.65252685546875, -406.2584228515625, -405.571044921875], "reading_logprob_std": 0.4771406007155169, "tokenization_path_logprob_std": 313.9648785097252, "trajectory_balance_rmse": 7078.702065181023} +{"canonical_length": 335, "canonical_nll": 1.1509536059934702, "generation_rouge_l_f1": 0.2572463768115942, "generation_token_f1": 0.5144927536231884, "prompt_id": "bef02b456531b6b6", "reading_canonical_log_prob": -385.5694580078125, "reading_distances": [18, 20, 20, 21], "reading_log_probs": [-384.10772705078125, -383.01409912109375, -383.1563720703125, -382.47601318359375], "reading_logprob_std": 0.6792308624944272, "tokenization_path_logprob_std": 135.3250591545362, "trajectory_balance_rmse": 6368.138858358627} +{"canonical_length": 110, "canonical_nll": 1.5511116721413352, "generation_rouge_l_f1": 0.20441988950276244, "generation_token_f1": 0.3259668508287293, "prompt_id": "9d2a25bad82737d8", "reading_canonical_log_prob": -170.62228393554688, "reading_distances": [16, 19, 17, 18], "reading_log_probs": [-171.77386474609375, -172.826171875, -173.03553771972656, -169.7713623046875], "reading_logprob_std": 1.492752539263645, "tokenization_path_logprob_std": 60.60729532405938, "trajectory_balance_rmse": 2808.355016393266} +{"canonical_length": 10, "canonical_nll": 2.455540084838867, "generation_rouge_l_f1": 0.05303030303030303, "generation_token_f1": 0.06060606060606061, "prompt_id": "77bd9a6bbbf6a3af", "reading_canonical_log_prob": -24.555400848388672, "reading_distances": [21, 22, 17, 19], "reading_log_probs": [-25.990921020507812, -25.76656150817871, -25.74981117248535, -26.451080322265625], "reading_logprob_std": 0.3267061286026719, "tokenization_path_logprob_std": 39.34087440715075, "trajectory_balance_rmse": 182.30855821560561} +{"canonical_length": 18, "canonical_nll": 0.7748158772786459, "generation_rouge_l_f1": 0.06993006993006994, "generation_token_f1": 0.07692307692307691, "prompt_id": "45f03bee685e0c93", "reading_canonical_log_prob": -13.946685791015625, "reading_distances": [23, 23, 27, 23], "reading_log_probs": [-14.244157791137695, -15.379406929016113, -13.777816772460938, -15.036612510681152], "reading_logprob_std": 0.7303774587304316, "tokenization_path_logprob_std": 78.15086459669288, "trajectory_balance_rmse": 443.10240180250526} +{"canonical_length": 300, "canonical_nll": 0.9872585042317709, "generation_rouge_l_f1": 0.2544802867383512, "generation_token_f1": 0.5125448028673835, "prompt_id": "1c2335f122ec1aa9", "reading_canonical_log_prob": -296.17755126953125, "reading_distances": [29, 22, 25, 22], "reading_log_probs": [-300.764404296875, -300.9473571777344, -303.85791015625, -298.30316162109375], "reading_logprob_std": 2.2726505421169527, "tokenization_path_logprob_std": 143.31237495229473, "trajectory_balance_rmse": 5361.0916923310215} +{"canonical_length": 24, "canonical_nll": 0.5922455787658691, "generation_rouge_l_f1": 0.4999999999999999, "generation_token_f1": 0.4999999999999999, "prompt_id": "0b6723f81e5854b3", "reading_canonical_log_prob": -14.21389389038086, "reading_distances": [20, 28, 25, 33], "reading_log_probs": [-13.49437141418457, -13.799700736999512, -13.807609558105469, -13.795587539672852], "reading_logprob_std": 0.1533784212339262, "tokenization_path_logprob_std": 73.8056113283538, "trajectory_balance_rmse": 631.4021437907786} +{"canonical_length": 43, "canonical_nll": 0.27840232849121094, "generation_rouge_l_f1": 0.5063291139240506, "generation_token_f1": 0.9367088607594937, "prompt_id": "6d26cacb241c8ba0", "reading_canonical_log_prob": -11.97130012512207, "reading_distances": [32, 37, 42, 38], "reading_log_probs": [-11.491569519042969, -12.303497314453125, -11.367713928222656, -11.620222091674805], "reading_logprob_std": 0.4180743938383393, "tokenization_path_logprob_std": 23.79532755093373, "trajectory_balance_rmse": 880.5594729371909} +{"canonical_length": 10, "canonical_nll": 1.2122369766235352, "generation_rouge_l_f1": 0.4444444444444445, "generation_token_f1": 0.888888888888889, "prompt_id": "d7683fe01587262b", "reading_canonical_log_prob": -12.122369766235352, "reading_distances": [40, 33, 45, 37], "reading_log_probs": [-11.80408000946045, -12.142169952392578, -12.271820068359375, -12.019219398498535], "reading_logprob_std": 0.1989773057858405, "tokenization_path_logprob_std": 23.574698572691258, "trajectory_balance_rmse": 226.37401920299075} +{"canonical_length": 71, "canonical_nll": 0.7598797435491857, "generation_rouge_l_f1": 0.5443786982248521, "generation_token_f1": 0.6272189349112426, "prompt_id": "e455ade71af9293b", "reading_canonical_log_prob": -53.95146179199219, "reading_distances": [52, 34, 39, 53], "reading_log_probs": [-49.63652420043945, -55.05358123779297, -52.586429595947266, -51.95478820800781], "reading_logprob_std": 2.2269022412580095, "tokenization_path_logprob_std": 110.28203934356897, "trajectory_balance_rmse": 1560.8345770628794} +{"canonical_length": 334, "canonical_nll": 1.6853297798933384, "generation_rouge_l_f1": 0.04142011834319527, "generation_token_f1": 0.04142011834319527, "prompt_id": "83deb2f6c2a5fe75", "reading_canonical_log_prob": -562.900146484375, "reading_distances": [43, 35, 31, 37], "reading_log_probs": [-562.4219360351562, -560.404296875, -563.9390869140625, -561.1216430664062], "reading_logprob_std": 1.554849735178593, "tokenization_path_logprob_std": 111.31584935554827, "trajectory_balance_rmse": 6176.719187265919} +{"canonical_length": 251, "canonical_nll": 1.2927902647223606, "generation_rouge_l_f1": 0.20881670533642693, "generation_token_f1": 0.33874709976798145, "prompt_id": "cc738e3b9180c6b5", "reading_canonical_log_prob": -324.4903564453125, "reading_distances": [11, 11, 12, 15], "reading_log_probs": [-314.57415771484375, -313.0662841796875, -318.4918518066406, -315.6062316894531], "reading_logprob_std": 2.289506991539778, "tokenization_path_logprob_std": 143.13731053727665, "trajectory_balance_rmse": 4838.380897986391} +{"canonical_length": 15, "canonical_nll": 0.08792632420857747, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "d829b51f56a6d0c5", "reading_canonical_log_prob": -1.318894863128662, "reading_distances": [18, 16, 15, 19], "reading_log_probs": [-1.9443597793579102, -3.5865445137023926, -2.174409866333008, -2.5940394401550293], "reading_logprob_std": 0.726123093813366, "tokenization_path_logprob_std": 53.925345924597906, "trajectory_balance_rmse": 431.4607717641375} +{"canonical_length": 158, "canonical_nll": 1.3817871673197686, "generation_rouge_l_f1": 0.187683284457478, "generation_token_f1": 0.5043988269794722, "prompt_id": "f53fbba7ee710f29", "reading_canonical_log_prob": -218.32237243652344, "reading_distances": [22, 23, 22, 29], "reading_log_probs": [-229.6526641845703, -223.22369384765625, -224.11865234375, -228.18576049804688], "reading_logprob_std": 3.1101068840715875, "tokenization_path_logprob_std": 190.63192538303352, "trajectory_balance_rmse": 3820.359789666488} +{"canonical_length": 268, "canonical_nll": 1.2478247115861123, "generation_rouge_l_f1": 0.20758483033932135, "generation_token_f1": 0.4351297405189621, "prompt_id": "2ce5516156798d8a", "reading_canonical_log_prob": -334.4170227050781, "reading_distances": [31, 29, 26, 30], "reading_log_probs": [-333.228271484375, -331.8336486816406, -332.5972900390625, -332.6900634765625], "reading_logprob_std": 0.5743096972555112, "tokenization_path_logprob_std": 263.00138662550484, "trajectory_balance_rmse": 5317.602697291226} +{"canonical_length": 28, "canonical_nll": 0.19822190489087785, "generation_rouge_l_f1": 0.8771929824561403, "generation_token_f1": 0.8771929824561403, "prompt_id": "83deea31656f880e", "reading_canonical_log_prob": -5.55021333694458, "reading_distances": [33, 26, 28, 23], "reading_log_probs": [-6.575000286102295, -5.3946075439453125, -6.03763484954834, -5.654641151428223], "reading_logprob_std": 0.5129112447985111, "tokenization_path_logprob_std": 69.34084511909276, "trajectory_balance_rmse": 549.3838809343549} +{"canonical_length": 16, "canonical_nll": 0.2877300977706909, "generation_rouge_l_f1": 0.8666666666666666, "generation_token_f1": 0.8666666666666666, "prompt_id": "2a011ac439c0c89f", "reading_canonical_log_prob": -4.603681564331055, "reading_distances": [50, 50, 40, 47], "reading_log_probs": [-5.708817005157471, -5.844925880432129, -6.397146224975586, -5.479832172393799], "reading_logprob_std": 0.38992120892780463, "tokenization_path_logprob_std": 44.030461243742955, "trajectory_balance_rmse": 371.20094976915425} +{"canonical_length": 27, "canonical_nll": 1.1332490002667461, "generation_rouge_l_f1": 0.08839779005524863, "generation_token_f1": 0.143646408839779, "prompt_id": "a62ae1af2ae45a72", "reading_canonical_log_prob": -30.59772300720215, "reading_distances": [27, 23, 29, 22], "reading_log_probs": [-31.245594024658203, -32.113792419433594, -34.27391815185547, -33.0088005065918], "reading_logprob_std": 1.2942538996881405, "tokenization_path_logprob_std": 41.28158925832495, "trajectory_balance_rmse": 453.67666222185585} +{"canonical_length": 99, "canonical_nll": 1.6374174175840435, "generation_rouge_l_f1": 0.2247838616714697, "generation_token_f1": 0.3400576368876081, "prompt_id": "53aea8a2764798f8", "reading_canonical_log_prob": -162.1043243408203, "reading_distances": [34, 31, 28, 33], "reading_log_probs": [-165.49478149414062, -164.38677978515625, -161.12730407714844, -163.9211883544922], "reading_logprob_std": 1.8579888044780015, "tokenization_path_logprob_std": 150.03523601347547, "trajectory_balance_rmse": 2139.137688771133} +{"canonical_length": 159, "canonical_nll": 0.5562636177494841, "generation_rouge_l_f1": 0.35625, "generation_token_f1": 0.5562499999999999, "prompt_id": "7dff62aaf925a1c6", "reading_canonical_log_prob": -88.44591522216797, "reading_distances": [30, 32, 34, 32], "reading_log_probs": [-92.62215423583984, -91.84927368164062, -92.77510070800781, -90.97205352783203], "reading_logprob_std": 0.8277086512332062, "tokenization_path_logprob_std": 60.748569778625026, "trajectory_balance_rmse": 2298.8526194680903} +{"canonical_length": 217, "canonical_nll": 1.278441679642497, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.42767295597484273, "prompt_id": "5470eb105173aff8", "reading_canonical_log_prob": -277.4218444824219, "reading_distances": [25, 38, 33, 35], "reading_log_probs": [-283.84332275390625, -280.3414001464844, -277.9201965332031, -281.4227294921875], "reading_logprob_std": 2.4580716549944914, "tokenization_path_logprob_std": 150.55748708044575, "trajectory_balance_rmse": 4949.469683207771} +{"canonical_length": 15, "canonical_nll": 0.596145757039388, "generation_rouge_l_f1": 0.5000000000000001, "generation_token_f1": 0.5000000000000001, "prompt_id": "861fa387c936a923", "reading_canonical_log_prob": -8.94218635559082, "reading_distances": [9, 9, 11, 10], "reading_log_probs": [-7.600919723510742, -8.638494491577148, -8.234281539916992, -8.158036231994629], "reading_logprob_std": 0.4270176606716453, "tokenization_path_logprob_std": 61.787139984910795, "trajectory_balance_rmse": 180.69419919022818} +{"canonical_length": 11, "canonical_nll": 1.4350913654674182, "generation_rouge_l_f1": 0.10937499999999999, "generation_token_f1": 0.125, "prompt_id": "6c2751afcc2de989", "reading_canonical_log_prob": -15.786005020141602, "reading_distances": [11, 9, 9, 14], "reading_log_probs": [-18.85114288330078, -16.229141235351562, -17.995929718017578, -17.862903594970703], "reading_logprob_std": 1.0951147592270767, "tokenization_path_logprob_std": 40.047637738259304, "trajectory_balance_rmse": 200.5684397961855} +{"canonical_length": 69, "canonical_nll": 0.8046763323355413, "generation_rouge_l_f1": 0.4225352112676056, "generation_token_f1": 0.5774647887323943, "prompt_id": "2dc95060fe386398", "reading_canonical_log_prob": -55.522666931152344, "reading_distances": [21, 18, 24, 20], "reading_log_probs": [-53.98952865600586, -55.181392669677734, -54.084190368652344, -54.0894889831543], "reading_logprob_std": 0.5653632578016652, "tokenization_path_logprob_std": 140.81496887164698, "trajectory_balance_rmse": 1655.361505334801} +{"canonical_length": 367, "canonical_nll": 1.2698924716876703, "generation_rouge_l_f1": 0.19397993311036787, "generation_token_f1": 0.3779264214046823, "prompt_id": "346df7037bb677e7", "reading_canonical_log_prob": -466.050537109375, "reading_distances": [27, 25, 27, 28], "reading_log_probs": [-469.3719482421875, -468.4825439453125, -468.3191223144531, -468.07562255859375], "reading_logprob_std": 0.5650609473661291, "tokenization_path_logprob_std": 220.42414189303759, "trajectory_balance_rmse": 6929.075378937401} +{"canonical_length": 168, "canonical_nll": 1.703438713437035, "generation_rouge_l_f1": 0.14382022471910114, "generation_token_f1": 0.34606741573033706, "prompt_id": "cccaf997f15db0c8", "reading_canonical_log_prob": -286.1777038574219, "reading_distances": [19, 12, 16, 12], "reading_log_probs": [-287.0606384277344, -286.5128479003906, -286.8876647949219, -287.0673828125], "reading_logprob_std": 0.2598613516919772, "tokenization_path_logprob_std": 105.17718381862116, "trajectory_balance_rmse": 3976.360619237105} +{"canonical_length": 15, "canonical_nll": 0.39997501373291017, "generation_rouge_l_f1": 0.07843137254901962, "generation_token_f1": 0.07843137254901962, "prompt_id": "c7d4acf2c03f4f98", "reading_canonical_log_prob": -5.999625205993652, "reading_distances": [15, 14, 19, 20], "reading_log_probs": [-9.864358901977539, -6.73574686050415, -6.888645648956299, -6.880923271179199], "reading_logprob_std": 1.5162586822565596, "tokenization_path_logprob_std": 27.470961210400407, "trajectory_balance_rmse": 229.67120352030014} +{"canonical_length": 31, "canonical_nll": 2.0474745227444555, "generation_rouge_l_f1": 0.29787234042553196, "generation_token_f1": 0.3829787234042554, "prompt_id": "f4a99dd0e66e3bbe", "reading_canonical_log_prob": -63.471710205078125, "reading_distances": [22, 18, 23, 19], "reading_log_probs": [-60.79366683959961, -60.78919219970703, -60.616363525390625, -61.20506286621094], "reading_logprob_std": 0.25001470859229574, "tokenization_path_logprob_std": 31.57132487957919, "trajectory_balance_rmse": 765.4645690704467} +{"canonical_length": 51, "canonical_nll": 1.2948847751991421, "generation_rouge_l_f1": 0.3142857142857143, "generation_token_f1": 0.4142857142857143, "prompt_id": "b701439d534ac688", "reading_canonical_log_prob": -66.03912353515625, "reading_distances": [21, 24, 23, 21], "reading_log_probs": [-62.72926330566406, -63.770347595214844, -62.482513427734375, -63.1043586730957], "reading_logprob_std": 0.5608169733649343, "tokenization_path_logprob_std": 48.83535243166864, "trajectory_balance_rmse": 1191.9020808784894} +{"canonical_length": 448, "canonical_nll": 1.2677873883928572, "generation_rouge_l_f1": 0.19512195121951217, "generation_token_f1": 0.3242467718794835, "prompt_id": "7fdb8e03f79c4b21", "reading_canonical_log_prob": -567.96875, "reading_distances": [15, 14, 13, 13], "reading_log_probs": [-570.640380859375, -567.68359375, -567.7823486328125, -567.6864013671875], "reading_logprob_std": 1.4621872698653318, "tokenization_path_logprob_std": 264.88749022496484, "trajectory_balance_rmse": 7872.530475945663} +{"canonical_length": 392, "canonical_nll": 1.2658440726143974, "generation_rouge_l_f1": 0.22364217252396168, "generation_token_f1": 0.4984025559105431, "prompt_id": "79079ac51debbea1", "reading_canonical_log_prob": -496.21087646484375, "reading_distances": [26, 23, 23, 25], "reading_log_probs": [-492.95928955078125, -492.5911865234375, -497.0499572753906, -489.7967529296875], "reading_logprob_std": 2.9884384043696555, "tokenization_path_logprob_std": 320.6931577930703, "trajectory_balance_rmse": 7289.323063641232} +{"canonical_length": 5, "canonical_nll": 2.186310577392578, "generation_rouge_l_f1": 0.030303030303030304, "generation_token_f1": 0.030303030303030304, "prompt_id": "e0c4de6979cf5d26", "reading_canonical_log_prob": -10.93155288696289, "reading_distances": [19, 19, 26, 24], "reading_log_probs": [-8.5929536819458, -9.484336853027344, -10.013348579406738, -9.774001121520996], "reading_logprob_std": 0.6210209790895967, "tokenization_path_logprob_std": 13.10760845676967, "trajectory_balance_rmse": 114.43460102786219} +{"canonical_length": 199, "canonical_nll": 1.485579140821294, "generation_rouge_l_f1": 0.19444444444444445, "generation_token_f1": 0.32407407407407407, "prompt_id": "a5453cc0769ffb94", "reading_canonical_log_prob": -295.6302490234375, "reading_distances": [22, 13, 22, 20], "reading_log_probs": [-321.04058837890625, -299.335693359375, -321.49810791015625, -324.20556640625], "reading_logprob_std": 11.5410225335552, "tokenization_path_logprob_std": 91.11080758997582, "trajectory_balance_rmse": 3318.9169576774952} +{"canonical_length": 16, "canonical_nll": 1.0265154838562012, "generation_rouge_l_f1": 0.4444444444444445, "generation_token_f1": 0.5185185185185186, "prompt_id": "c1820a57ac4bcbc9", "reading_canonical_log_prob": -16.42424774169922, "reading_distances": [17, 13, 13, 14], "reading_log_probs": [-15.598358154296875, -15.860116958618164, -15.688837051391602, -15.766414642333984], "reading_logprob_std": 0.11146086717750617, "tokenization_path_logprob_std": 51.09920584923711, "trajectory_balance_rmse": 419.99473878787234} +{"canonical_length": 87, "canonical_nll": 0.4266950234599497, "generation_rouge_l_f1": 0.308411214953271, "generation_token_f1": 0.3551401869158879, "prompt_id": "326b224fa8e4dc3a", "reading_canonical_log_prob": -37.122467041015625, "reading_distances": [25, 28, 21, 30], "reading_log_probs": [-38.372962951660156, -39.508941650390625, -36.00933837890625, -37.51044464111328], "reading_logprob_std": 1.4752299234705721, "tokenization_path_logprob_std": 136.76025935069055, "trajectory_balance_rmse": 1162.82970364049} +{"canonical_length": 210, "canonical_nll": 0.921806408110119, "generation_rouge_l_f1": 0.14187643020594964, "generation_token_f1": 0.2562929061784897, "prompt_id": "a8f7e780841cbd23", "reading_canonical_log_prob": -193.579345703125, "reading_distances": [15, 21, 12, 16], "reading_log_probs": [-197.18988037109375, -194.84161376953125, -197.86715698242188, -200.0622100830078], "reading_logprob_std": 2.149614952535218, "tokenization_path_logprob_std": 146.23396139503853, "trajectory_balance_rmse": 3101.9195803930807} +{"canonical_length": 221, "canonical_nll": 1.427251617293552, "generation_rouge_l_f1": 0.2634989200863931, "generation_token_f1": 0.509719222462203, "prompt_id": "675093cc227e317b", "reading_canonical_log_prob": -315.422607421875, "reading_distances": [15, 17, 14, 19], "reading_log_probs": [-312.57244873046875, -314.54632568359375, -314.4603271484375, -313.3690185546875], "reading_logprob_std": 0.9433628191467134, "tokenization_path_logprob_std": 104.11403945289723, "trajectory_balance_rmse": 4796.870437999004} +{"canonical_length": 327, "canonical_nll": 1.1539128388104454, "generation_rouge_l_f1": 0.25838926174496646, "generation_token_f1": 0.5771812080536913, "prompt_id": "df931a7b15ed70cf", "reading_canonical_log_prob": -377.3294982910156, "reading_distances": [34, 35, 42, 34], "reading_log_probs": [-375.11376953125, -375.67608642578125, -375.14892578125, -375.7230529785156], "reading_logprob_std": 0.32893626684608307, "tokenization_path_logprob_std": 201.57185099456083, "trajectory_balance_rmse": 7172.319720666364} +{"canonical_length": 267, "canonical_nll": 1.3241295975245786, "generation_rouge_l_f1": 0.24444444444444446, "generation_token_f1": 0.5444444444444444, "prompt_id": "64f98636b7dde269", "reading_canonical_log_prob": -353.5426025390625, "reading_distances": [27, 26, 32, 24], "reading_log_probs": [-353.2967529296875, -355.4453125, -353.66070556640625, -353.90509033203125], "reading_logprob_std": 0.9458538599174617, "tokenization_path_logprob_std": 34.53954950998981, "trajectory_balance_rmse": 5141.072625557398} +{"canonical_length": 110, "canonical_nll": 1.153987676447088, "generation_rouge_l_f1": 0.2843137254901961, "generation_token_f1": 0.4215686274509804, "prompt_id": "3485bbb40ac08228", "reading_canonical_log_prob": -126.93864440917969, "reading_distances": [16, 15, 10, 14], "reading_log_probs": [-127.26556396484375, -130.3046875, -130.8638153076172, -130.14132690429688], "reading_logprob_std": 1.6154204084916297, "tokenization_path_logprob_std": 57.757271881820685, "trajectory_balance_rmse": 2428.157270744196} +{"canonical_length": 20, "canonical_nll": 1.1862266540527344, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.18181818181818182, "prompt_id": "5de8b78dfeffe43c", "reading_canonical_log_prob": -23.724533081054688, "reading_distances": [13, 19, 12, 11], "reading_log_probs": [-24.74658966064453, -24.06095314025879, -23.758590698242188, -24.824228286743164], "reading_logprob_std": 0.5213654036650961, "tokenization_path_logprob_std": 33.53860641033366, "trajectory_balance_rmse": 274.4001950750702} +{"canonical_length": 14, "canonical_nll": 2.8355211530412947, "generation_rouge_l_f1": 0.07567567567567567, "generation_token_f1": 0.08648648648648649, "prompt_id": "75d59497e8843334", "reading_canonical_log_prob": -39.697296142578125, "reading_distances": [7, 7, 11, 10], "reading_log_probs": [-38.07136917114258, -37.98216247558594, -34.85546875, -34.75726318359375], "reading_logprob_std": 1.8600874818210276, "tokenization_path_logprob_std": 48.35669221741314, "trajectory_balance_rmse": 244.65106621230944} +{"canonical_length": 53, "canonical_nll": 2.477832722214033, "generation_rouge_l_f1": 0.13793103448275862, "generation_token_f1": 0.15673981191222572, "prompt_id": "9fe6000f50208baa", "reading_canonical_log_prob": -131.32513427734375, "reading_distances": [24, 22, 20, 23], "reading_log_probs": [-132.59149169921875, -133.708251953125, -133.399658203125, -133.86915588378906], "reading_logprob_std": 0.5682033512796912, "tokenization_path_logprob_std": 49.73336775945298, "trajectory_balance_rmse": 1353.1987394169416} +{"canonical_length": 13, "canonical_nll": 0.10075572820810172, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "2321fcc6e5fd7b96", "reading_canonical_log_prob": -1.3098244667053223, "reading_distances": [31, 35, 36, 45], "reading_log_probs": [-1.3524338006973267, -1.3815076351165771, -1.2733240127563477, -1.3056495189666748], "reading_logprob_std": 0.0481278710810278, "tokenization_path_logprob_std": 60.594967822768744, "trajectory_balance_rmse": 188.6596253942017} +{"canonical_length": 6, "canonical_nll": 1.76884921391805, "generation_rouge_l_f1": 0.4444444444444444, "generation_token_f1": 0.4444444444444444, "prompt_id": "9ecc7ea82de4a5d6", "reading_canonical_log_prob": -10.6130952835083, "reading_distances": [47, 46, 47, 40], "reading_log_probs": [-10.566946029663086, -11.015260696411133, -9.894817352294922, -10.683594703674316], "reading_logprob_std": 0.4702760927256967, "tokenization_path_logprob_std": 10.970047873428756, "trajectory_balance_rmse": 74.93485115204605} +{"canonical_length": 268, "canonical_nll": 1.0764973199189598, "generation_rouge_l_f1": 0.20238095238095236, "generation_token_f1": 0.39682539682539686, "prompt_id": "aceb2f4e30eb8855", "reading_canonical_log_prob": -288.50128173828125, "reading_distances": [25, 27, 32, 27], "reading_log_probs": [-296.50079345703125, -295.51104736328125, -297.48126220703125, -298.20721435546875], "reading_logprob_std": 1.1736999220612652, "tokenization_path_logprob_std": 224.70599715429123, "trajectory_balance_rmse": 4735.869531116175} +{"canonical_length": 118, "canonical_nll": 1.3247416868048199, "generation_rouge_l_f1": 0.25076452599388377, "generation_token_f1": 0.4342507645259938, "prompt_id": "c575c5fadea559fc", "reading_canonical_log_prob": -156.31951904296875, "reading_distances": [53, 46, 38, 46], "reading_log_probs": [-152.39791870117188, -151.89675903320312, -154.35165405273438, -153.810546875], "reading_logprob_std": 1.1563466764037906, "tokenization_path_logprob_std": 65.97161560255002, "trajectory_balance_rmse": 2648.909881858953} +{"canonical_length": 61, "canonical_nll": 2.394657822905994, "generation_rouge_l_f1": 0.2686567164179105, "generation_token_f1": 0.3432835820895523, "prompt_id": "3adbad55ce76ffec", "reading_canonical_log_prob": -146.07412719726562, "reading_distances": [25, 21, 24, 19], "reading_log_probs": [-141.5847625732422, -143.49349975585938, -140.58392333984375, -142.70494079589844], "reading_logprob_std": 1.2742949441680156, "tokenization_path_logprob_std": 82.26604632130564, "trajectory_balance_rmse": 1189.913402550869} +{"canonical_length": 12, "canonical_nll": 2.023510456085205, "generation_rouge_l_f1": 0.5384615384615384, "generation_token_f1": 0.5384615384615384, "prompt_id": "70c54675265e6b63", "reading_canonical_log_prob": -24.28212547302246, "reading_distances": [27, 34, 22, 28], "reading_log_probs": [-23.497726440429688, -24.0764102935791, -25.395822525024414, -23.55051040649414], "reading_logprob_std": 0.8833188381440257, "tokenization_path_logprob_std": 40.763924966340696, "trajectory_balance_rmse": 329.82502534160733} +{"canonical_length": 11, "canonical_nll": 1.276305372064764, "generation_rouge_l_f1": 0.6153846153846153, "generation_token_f1": 0.6153846153846153, "prompt_id": "5e3e21467d43fd88", "reading_canonical_log_prob": -14.039359092712402, "reading_distances": [16, 20, 12, 17], "reading_log_probs": [-15.373889923095703, -16.76906394958496, -15.357696533203125, -15.06024169921875], "reading_logprob_std": 0.7662494613396301, "tokenization_path_logprob_std": 25.700829184681375, "trajectory_balance_rmse": 246.4700915233345} +{"canonical_length": 17, "canonical_nll": 0.3162778966567096, "generation_rouge_l_f1": 0.4210526315789474, "generation_token_f1": 0.4210526315789474, "prompt_id": "46d8b242e85d5c4e", "reading_canonical_log_prob": -5.3767242431640625, "reading_distances": [12, 8, 11, 10], "reading_log_probs": [-5.201502799987793, -5.396244525909424, -5.397093772888184, -5.211949348449707], "reading_logprob_std": 0.10974712996468224, "tokenization_path_logprob_std": 13.927785059174715, "trajectory_balance_rmse": 131.55897860144495} +{"canonical_length": 48, "canonical_nll": 1.3380786577860515, "generation_rouge_l_f1": 0.13114754098360654, "generation_token_f1": 0.18360655737704917, "prompt_id": "6509666af0922805", "reading_canonical_log_prob": -64.22777557373047, "reading_distances": [30, 28, 32, 23], "reading_log_probs": [-62.54646301269531, -63.74774932861328, -62.50995635986328, -63.002777099609375], "reading_logprob_std": 0.5760954370752336, "tokenization_path_logprob_std": 71.16076210026203, "trajectory_balance_rmse": 943.2972636568368} +{"canonical_length": 175, "canonical_nll": 1.2133504813058036, "generation_rouge_l_f1": 0.2760416666666667, "generation_token_f1": 0.4479166666666667, "prompt_id": "42f6dd40e9d218d5", "reading_canonical_log_prob": -212.33633422851562, "reading_distances": [8, 15, 12, 13], "reading_log_probs": [-212.2258758544922, -212.60299682617188, -211.8959503173828, -211.97486877441406], "reading_logprob_std": 0.3181670655781, "tokenization_path_logprob_std": 144.64673571049534, "trajectory_balance_rmse": 3262.1185109004614} +{"canonical_length": 85, "canonical_nll": 1.122826789407169, "generation_rouge_l_f1": 0.1648936170212766, "generation_token_f1": 0.29787234042553196, "prompt_id": "68880c3a241cb55b", "reading_canonical_log_prob": -95.44027709960938, "reading_distances": [35, 41, 42, 37], "reading_log_probs": [-96.02993774414062, -95.21261596679688, -95.41224670410156, -94.71785736083984], "reading_logprob_std": 0.5429799804772125, "tokenization_path_logprob_std": 160.51149103183286, "trajectory_balance_rmse": 1712.9507920363799} +{"canonical_length": 14, "canonical_nll": 1.4122591018676758, "generation_rouge_l_f1": 0.17142857142857143, "generation_token_f1": 0.34285714285714286, "prompt_id": "a0d2e6c5b9eebb8b", "reading_canonical_log_prob": -19.77162742614746, "reading_distances": [26, 28, 25, 32], "reading_log_probs": [-19.64903450012207, -20.024478912353516, -19.995691299438477, -19.34605598449707], "reading_logprob_std": 0.32094104627850745, "tokenization_path_logprob_std": 37.64590626789672, "trajectory_balance_rmse": 362.6373038014614} +{"canonical_length": 256, "canonical_nll": 0.5549212694168091, "generation_rouge_l_f1": 0.22350674373795765, "generation_token_f1": 0.44315992292870904, "prompt_id": "c51ea1a2a74eb3ff", "reading_canonical_log_prob": -142.05984497070312, "reading_distances": [28, 24, 27, 20], "reading_log_probs": [-143.62591552734375, -141.70518493652344, -142.79322814941406, -143.2604522705078], "reading_logprob_std": 0.8335231728723134, "tokenization_path_logprob_std": 183.87745011145134, "trajectory_balance_rmse": 3367.1594111244935} +{"canonical_length": 61, "canonical_nll": 0.8667129766745646, "generation_rouge_l_f1": 0.10967741935483873, "generation_token_f1": 0.18064516129032257, "prompt_id": "9ce480763a8f4a02", "reading_canonical_log_prob": -52.86949157714844, "reading_distances": [30, 32, 30, 29], "reading_log_probs": [-54.16447830200195, -53.98280715942383, -52.97605514526367, -55.031166076660156], "reading_logprob_std": 0.8432372481479334, "tokenization_path_logprob_std": 58.17477892903987, "trajectory_balance_rmse": 955.5274014409051} +{"canonical_length": 17, "canonical_nll": 0.7004403507008272, "generation_rouge_l_f1": 0.11683848797250859, "generation_token_f1": 0.12371134020618557, "prompt_id": "690ac1ef381d5161", "reading_canonical_log_prob": -11.907485961914062, "reading_distances": [32, 37, 39, 41], "reading_log_probs": [-10.61812973022461, -10.859827041625977, -10.561050415039062, -11.830181121826172], "reading_logprob_std": 0.5896532511364734, "tokenization_path_logprob_std": 23.589933868148844, "trajectory_balance_rmse": 314.57148883489873} +{"canonical_length": 49, "canonical_nll": 1.2012458334163743, "generation_rouge_l_f1": 0.33103448275862063, "generation_token_f1": 0.4689655172413793, "prompt_id": "f544199aa0f474b9", "reading_canonical_log_prob": -58.861045837402344, "reading_distances": [20, 18, 19, 19], "reading_log_probs": [-59.53773880004883, -59.02239990234375, -59.380653381347656, -59.12110137939453], "reading_logprob_std": 0.23616581869775247, "tokenization_path_logprob_std": 107.13316532349437, "trajectory_balance_rmse": 1110.558909773138} +{"canonical_length": 236, "canonical_nll": 0.5354343996209613, "generation_rouge_l_f1": 0.14761904761904762, "generation_token_f1": 0.4, "prompt_id": "672d37d3ef4c74d7", "reading_canonical_log_prob": -126.36251831054688, "reading_distances": [20, 24, 25, 26], "reading_log_probs": [-129.03875732421875, -131.31912231445312, -132.6329345703125, -132.0918731689453], "reading_logprob_std": 1.582605575550417, "tokenization_path_logprob_std": 114.11754275439222, "trajectory_balance_rmse": 4135.174590902362} +{"canonical_length": 76, "canonical_nll": 1.0749070016961348, "generation_rouge_l_f1": 0.2516556291390728, "generation_token_f1": 0.41059602649006627, "prompt_id": "e905bcd075fa835d", "reading_canonical_log_prob": -81.69293212890625, "reading_distances": [19, 17, 18, 19], "reading_log_probs": [-81.76875305175781, -82.54554748535156, -82.55133056640625, -80.54204559326172], "reading_logprob_std": 0.9474495468609394, "tokenization_path_logprob_std": 81.79394764095598, "trajectory_balance_rmse": 1740.0967077881908} +{"canonical_length": 10, "canonical_nll": 0.8830163955688477, "generation_rouge_l_f1": 0.6250000000000001, "generation_token_f1": 0.8750000000000001, "prompt_id": "1d805df4e7b3cce0", "reading_canonical_log_prob": -8.830163955688477, "reading_distances": [34, 41, 36, 34], "reading_log_probs": [-9.620016098022461, -9.194384574890137, -9.135674476623535, -9.016599655151367], "reading_logprob_std": 0.2628519180172412, "tokenization_path_logprob_std": 31.156515923606243, "trajectory_balance_rmse": 226.0483155111613} +{"canonical_length": 140, "canonical_nll": 1.413069806780134, "generation_rouge_l_f1": 0.17475728155339806, "generation_token_f1": 0.3592233009708738, "prompt_id": "6a438344f95ac9cf", "reading_canonical_log_prob": -197.82977294921875, "reading_distances": [25, 26, 18, 25], "reading_log_probs": [-199.8575439453125, -199.70570373535156, -201.71914672851562, -199.03854370117188], "reading_logprob_std": 1.1490601072117443, "tokenization_path_logprob_std": 194.15074600532066, "trajectory_balance_rmse": 3335.7879755821205} +{"canonical_length": 103, "canonical_nll": 1.7665120837757888, "generation_rouge_l_f1": 0.18591549295774645, "generation_token_f1": 0.34929577464788736, "prompt_id": "208ec2e52d9d1ce6", "reading_canonical_log_prob": -181.95074462890625, "reading_distances": [32, 29, 36, 30], "reading_log_probs": [-180.39596557617188, -180.13800048828125, -180.36012268066406, -179.59701538085938], "reading_logprob_std": 0.3686109082459771, "tokenization_path_logprob_std": 131.90270639318624, "trajectory_balance_rmse": 2131.879000549968} +{"canonical_length": 14, "canonical_nll": 0.33737826347351074, "generation_rouge_l_f1": 0.8, "generation_token_f1": 0.8, "prompt_id": "6a81e01870d01f92", "reading_canonical_log_prob": -4.72329568862915, "reading_distances": [25, 28, 31, 24], "reading_log_probs": [-5.993273735046387, -6.1184611320495605, -6.386274814605713, -5.760799407958984], "reading_logprob_std": 0.26061331627233797, "tokenization_path_logprob_std": 41.59239109016975, "trajectory_balance_rmse": 368.2635537336241} +{"canonical_length": 274, "canonical_nll": 1.1530564579650433, "generation_rouge_l_f1": 0.22014925373134328, "generation_token_f1": 0.4253731343283582, "prompt_id": "7857d9054d4c3033", "reading_canonical_log_prob": -315.9374694824219, "reading_distances": [29, 31, 23, 30], "reading_log_probs": [-316.0498352050781, -315.50384521484375, -315.95782470703125, -314.93121337890625], "reading_logprob_std": 0.5120075853169899, "tokenization_path_logprob_std": 204.40576802352274, "trajectory_balance_rmse": 4568.2532407289} +{"canonical_length": 158, "canonical_nll": 1.2105925354776503, "generation_rouge_l_f1": 0.26732673267326734, "generation_token_f1": 0.4108910891089109, "prompt_id": "77515334ad84a428", "reading_canonical_log_prob": -191.27362060546875, "reading_distances": [19, 18, 20, 16], "reading_log_probs": [-191.00875854492188, -190.9105682373047, -191.0258331298828, -190.08767700195312], "reading_logprob_std": 0.44989785764944573, "tokenization_path_logprob_std": 112.80935353012762, "trajectory_balance_rmse": 2770.691767446187} +{"canonical_length": 43, "canonical_nll": 0.906076475631359, "generation_rouge_l_f1": 0.1488673139158576, "generation_token_f1": 0.1553398058252427, "prompt_id": "78695f86d2d683db", "reading_canonical_log_prob": -38.96128845214844, "reading_distances": [40, 37, 35, 31], "reading_log_probs": [-38.84104919433594, -38.37690353393555, -38.73469161987305, -39.17340850830078], "reading_logprob_std": 0.3281400663985352, "tokenization_path_logprob_std": 119.24261859466885, "trajectory_balance_rmse": 1230.6571098552702} +{"canonical_length": 8, "canonical_nll": 0.24777886271476746, "generation_rouge_l_f1": 0.08695652173913042, "generation_token_f1": 0.08695652173913042, "prompt_id": "73a081a89d8a40bd", "reading_canonical_log_prob": -1.9822309017181396, "reading_distances": [28, 28, 32, 25], "reading_log_probs": [-1.6482690572738647, -1.1780370473861694, -1.6259578466415405, -1.3008863925933838], "reading_logprob_std": 0.23517495317738232, "tokenization_path_logprob_std": 33.58863688678214, "trajectory_balance_rmse": 115.64172697601785} +{"canonical_length": 84, "canonical_nll": 0.9493413652692523, "generation_rouge_l_f1": 0.20598006644518274, "generation_token_f1": 0.27906976744186046, "prompt_id": "7879ca5705298aac", "reading_canonical_log_prob": -79.74467468261719, "reading_distances": [46, 46, 41, 35], "reading_log_probs": [-83.6374282836914, -86.85541534423828, -86.97303009033203, -82.45841217041016], "reading_logprob_std": 2.284020655038704, "tokenization_path_logprob_std": 135.66252329143845, "trajectory_balance_rmse": 1309.3656835681895} +{"canonical_length": 355, "canonical_nll": 1.253795527068662, "generation_rouge_l_f1": 0.18914728682170542, "generation_token_f1": 0.3286821705426357, "prompt_id": "2e88e8dc9b895643", "reading_canonical_log_prob": -445.097412109375, "reading_distances": [23, 20, 20, 21], "reading_log_probs": [-446.20770263671875, -445.0967102050781, -443.9027404785156, -445.517333984375], "reading_logprob_std": 0.9675210536413436, "tokenization_path_logprob_std": 138.12053149605958, "trajectory_balance_rmse": 7954.968695602292} +{"canonical_length": 343, "canonical_nll": 1.3418723237062682, "generation_rouge_l_f1": 0.2057335581787521, "generation_token_f1": 0.39460370994940974, "prompt_id": "a81b52bde10fe226", "reading_canonical_log_prob": -460.26220703125, "reading_distances": [19, 20, 20, 19], "reading_log_probs": [-463.7555236816406, -462.58087158203125, -462.015380859375, -461.82647705078125], "reading_logprob_std": 0.868605252503528, "tokenization_path_logprob_std": 80.77441469550801, "trajectory_balance_rmse": 7055.436612054284} +{"canonical_length": 269, "canonical_nll": 1.1512716640770213, "generation_rouge_l_f1": 0.274074074074074, "generation_token_f1": 0.5259259259259259, "prompt_id": "ff000a10ec04c16c", "reading_canonical_log_prob": -309.69207763671875, "reading_distances": [33, 25, 32, 26], "reading_log_probs": [-309.23858642578125, -310.74102783203125, -311.733154296875, -311.10986328125], "reading_logprob_std": 1.0602956471948164, "tokenization_path_logprob_std": 301.72284464257905, "trajectory_balance_rmse": 5853.858028934221} +{"canonical_length": 405, "canonical_nll": 1.195536144868827, "generation_rouge_l_f1": 0.21417322834645672, "generation_token_f1": 0.415748031496063, "prompt_id": "e7f10884569482cf", "reading_canonical_log_prob": -484.192138671875, "reading_distances": [31, 38, 41, 31], "reading_log_probs": [-479.56982421875, -479.9764404296875, -479.1252746582031, -481.0335693359375], "reading_logprob_std": 0.8159403621443528, "tokenization_path_logprob_std": 188.24078063249135, "trajectory_balance_rmse": 7051.2312371161215} +{"canonical_length": 10, "canonical_nll": 1.793858528137207, "generation_rouge_l_f1": 0.8750000000000001, "generation_token_f1": 0.8750000000000001, "prompt_id": "7d1f1626ed9a7e46", "reading_canonical_log_prob": -17.93858528137207, "reading_distances": [34, 38, 35, 40], "reading_log_probs": [-17.657581329345703, -13.289433479309082, -12.453105926513672, -16.800983428955078], "reading_logprob_std": 2.563128059976658, "tokenization_path_logprob_std": 26.166386569770744, "trajectory_balance_rmse": 264.7148042976287} +{"canonical_length": 41, "canonical_nll": 1.6535586845584032, "generation_rouge_l_f1": 0.1492537313432836, "generation_token_f1": 0.2388059701492537, "prompt_id": "8de4f5d47e6be004", "reading_canonical_log_prob": -67.79590606689453, "reading_distances": [13, 16, 15, 15], "reading_log_probs": [-67.35586547851562, -67.9501953125, -67.79486083984375, -66.1952896118164], "reading_logprob_std": 0.793480316465392, "tokenization_path_logprob_std": 89.6844419182308, "trajectory_balance_rmse": 932.7279518658956} +{"canonical_length": 187, "canonical_nll": 1.1907967144155247, "generation_rouge_l_f1": 0.33333333333333337, "generation_token_f1": 0.5666666666666667, "prompt_id": "37dc889754565fdd", "reading_canonical_log_prob": -222.67898559570312, "reading_distances": [32, 40, 40, 36], "reading_log_probs": [-222.01483154296875, -222.4973907470703, -222.6243133544922, -222.33447265625], "reading_logprob_std": 0.2634959184202422, "tokenization_path_logprob_std": 138.33491434243248, "trajectory_balance_rmse": 3970.054266257463} +{"canonical_length": 81, "canonical_nll": 1.4312572714723186, "generation_rouge_l_f1": 0.32407407407407407, "generation_token_f1": 0.45370370370370366, "prompt_id": "fccb0a4a57193f97", "reading_canonical_log_prob": -115.93183898925781, "reading_distances": [24, 26, 24, 21], "reading_log_probs": [-112.88484954833984, -113.40678405761719, -111.33692932128906, -111.86985778808594], "reading_logprob_std": 0.9411488844699569, "tokenization_path_logprob_std": 169.53568519861403, "trajectory_balance_rmse": 1971.101106885059} +{"canonical_length": 8, "canonical_nll": 0.33173710107803345, "generation_rouge_l_f1": 0.20000000000000004, "generation_token_f1": 0.20000000000000004, "prompt_id": "a29eafa220e503a1", "reading_canonical_log_prob": -2.6538968086242676, "reading_distances": [35, 27, 28, 37], "reading_log_probs": [-7.68294095993042, -3.7484357357025146, -3.2100110054016113, -3.959388017654419], "reading_logprob_std": 2.0463025235263417, "tokenization_path_logprob_std": 42.823751880835836, "trajectory_balance_rmse": 188.38691971250134} +{"canonical_length": 20, "canonical_nll": 1.3349299430847168, "generation_rouge_l_f1": 0.5625000000000001, "generation_token_f1": 0.6250000000000001, "prompt_id": "276e3276fb9b91f2", "reading_canonical_log_prob": -26.698598861694336, "reading_distances": [40, 49, 46, 48], "reading_log_probs": [-27.811281204223633, -27.81852912902832, -27.590709686279297, -30.804662704467773], "reading_logprob_std": 1.5358880863344777, "tokenization_path_logprob_std": 32.220362561223666, "trajectory_balance_rmse": 467.84843199091705} +{"canonical_length": 229, "canonical_nll": 1.1110782540000683, "generation_rouge_l_f1": 0.21825396825396826, "generation_token_f1": 0.40079365079365087, "prompt_id": "14d153fd0e1b36b1", "reading_canonical_log_prob": -254.43692016601562, "reading_distances": [15, 21, 16, 22], "reading_log_probs": [-252.568115234375, -252.19427490234375, -253.09341430664062, -251.95135498046875], "reading_logprob_std": 0.497313631865757, "tokenization_path_logprob_std": 131.04778094721624, "trajectory_balance_rmse": 4384.278116321009} +{"canonical_length": 315, "canonical_nll": 0.5081312391493056, "generation_rouge_l_f1": 0.266199649737303, "generation_token_f1": 0.38528896672504376, "prompt_id": "0ee97cbd046603bc", "reading_canonical_log_prob": -160.06134033203125, "reading_distances": [25, 28, 24, 29], "reading_log_probs": [-159.4280242919922, -158.98138427734375, -159.891845703125, -159.25091552734375], "reading_logprob_std": 0.38279277590436783, "tokenization_path_logprob_std": 206.67351894172342, "trajectory_balance_rmse": 5050.573195564666} +{"canonical_length": 282, "canonical_nll": 0.8461900535204732, "generation_rouge_l_f1": 0.2867256637168142, "generation_token_f1": 0.5238938053097346, "prompt_id": "740c83c01085ccfc", "reading_canonical_log_prob": -238.62559509277344, "reading_distances": [16, 17, 28, 24], "reading_log_probs": [-242.06158447265625, -241.6029815673828, -243.708740234375, -241.59991455078125], "reading_logprob_std": 1.0007477445574042, "tokenization_path_logprob_std": 301.0809526035359, "trajectory_balance_rmse": 6734.941208126635} +{"canonical_length": 128, "canonical_nll": 1.383910059928894, "generation_rouge_l_f1": 0.26436781609195403, "generation_token_f1": 0.4597701149425287, "prompt_id": "055891d31fa97386", "reading_canonical_log_prob": -177.14048767089844, "reading_distances": [26, 24, 21, 22], "reading_log_probs": [-173.952880859375, -175.68243408203125, -175.57699584960938, -175.16064453125], "reading_logprob_std": 0.7929127166690514, "tokenization_path_logprob_std": 337.555772956543, "trajectory_balance_rmse": 2672.815966001249} +{"canonical_length": 197, "canonical_nll": 1.1342542619269511, "generation_rouge_l_f1": 0.19327731092436973, "generation_token_f1": 0.4621848739495798, "prompt_id": "54415ac83a3ddec9", "reading_canonical_log_prob": -223.44808959960938, "reading_distances": [20, 19, 19, 19], "reading_log_probs": [-219.68356323242188, -218.6822509765625, -220.4840850830078, -220.5302734375], "reading_logprob_std": 0.867194046227685, "tokenization_path_logprob_std": 132.31850169448938, "trajectory_balance_rmse": 3601.1589391961593} +{"canonical_length": 71, "canonical_nll": 1.4876498370103433, "generation_rouge_l_f1": 0.2082018927444795, "generation_token_f1": 0.32807570977917977, "prompt_id": "6bb4c6466fd9ddad", "reading_canonical_log_prob": -105.62313842773438, "reading_distances": [11, 17, 13, 18], "reading_log_probs": [-105.9159927368164, -106.17892456054688, -106.68437194824219, -107.85165405273438], "reading_logprob_std": 0.8574367330233174, "tokenization_path_logprob_std": 117.82468974300795, "trajectory_balance_rmse": 1461.9858729907444} +{"canonical_length": 96, "canonical_nll": 0.9294789632161459, "generation_rouge_l_f1": 0.3790849673202614, "generation_token_f1": 0.5359477124183006, "prompt_id": "5ca3d0261fdb6645", "reading_canonical_log_prob": -89.22998046875, "reading_distances": [20, 27, 25, 23], "reading_log_probs": [-95.87471008300781, -95.85897064208984, -94.08922576904297, -94.92034149169922], "reading_logprob_std": 0.8564848858160709, "tokenization_path_logprob_std": 192.08370017184666, "trajectory_balance_rmse": 2046.5423138483184} +{"canonical_length": 120, "canonical_nll": 1.3736916859944661, "generation_rouge_l_f1": 0.24215246636771298, "generation_token_f1": 0.41255605381165916, "prompt_id": "2749992f70103af2", "reading_canonical_log_prob": -164.84300231933594, "reading_distances": [38, 44, 37, 44], "reading_log_probs": [-164.24050903320312, -171.43099975585938, -164.0920867919922, -165.15597534179688], "reading_logprob_std": 3.49917488239687, "tokenization_path_logprob_std": 164.39014551390832, "trajectory_balance_rmse": 2838.4855500715116} +{"canonical_length": 18, "canonical_nll": 0.032307697667015925, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "67a8cb2f3ba290b7", "reading_canonical_log_prob": -0.5815385580062866, "reading_distances": [39, 50, 47, 39], "reading_log_probs": [-0.6537017822265625, -0.7928690314292908, -0.7085583209991455, -0.6845636963844299], "reading_logprob_std": 0.0596820075798018, "tokenization_path_logprob_std": 34.629248328248636, "trajectory_balance_rmse": 396.3733714940122} +{"canonical_length": 144, "canonical_nll": 1.006561067369249, "generation_rouge_l_f1": 0.2173913043478261, "generation_token_f1": 0.2857142857142857, "prompt_id": "14aa2b70fa0eb52a", "reading_canonical_log_prob": -144.94479370117188, "reading_distances": [32, 26, 28, 30], "reading_log_probs": [-147.5772705078125, -146.0459747314453, -142.52224731445312, -144.354736328125], "reading_logprob_std": 2.177877617672983, "tokenization_path_logprob_std": 226.2103131725501, "trajectory_balance_rmse": 2532.4820911660963} +{"canonical_length": 46, "canonical_nll": 1.7128470047660496, "generation_rouge_l_f1": 0.09240924092409242, "generation_token_f1": 0.13201320132013203, "prompt_id": "632380e80642e576", "reading_canonical_log_prob": -78.79096221923828, "reading_distances": [26, 32, 28, 25], "reading_log_probs": [-81.16499328613281, -82.4292984008789, -82.09194946289062, -81.30755615234375], "reading_logprob_std": 0.6100151768170805, "tokenization_path_logprob_std": 92.07886892699301, "trajectory_balance_rmse": 993.5851608638675} +{"canonical_length": 186, "canonical_nll": 1.2710000315020162, "generation_rouge_l_f1": 0.2517482517482518, "generation_token_f1": 0.4615384615384615, "prompt_id": "e0e28697fbbba739", "reading_canonical_log_prob": -236.406005859375, "reading_distances": [36, 28, 34, 28], "reading_log_probs": [-239.1299591064453, -238.45631408691406, -240.81967163085938, -237.08982849121094], "reading_logprob_std": 1.550148663969778, "tokenization_path_logprob_std": 311.43075750595995, "trajectory_balance_rmse": 4197.478421569773} +{"canonical_length": 41, "canonical_nll": 1.1228616295791254, "generation_rouge_l_f1": 0.21761658031088082, "generation_token_f1": 0.29015544041450775, "prompt_id": "dd2bdffacb538a3a", "reading_canonical_log_prob": -46.03732681274414, "reading_distances": [145, 138, 159, 126], "reading_log_probs": [-51.44599151611328, -46.494667053222656, -50.1565055847168, -49.09259796142578], "reading_logprob_std": 2.10172657182855, "tokenization_path_logprob_std": 72.46576615506255, "trajectory_balance_rmse": 1022.2399987387765} +{"canonical_length": 11, "canonical_nll": 0.8529382185502485, "generation_rouge_l_f1": 0.06986899563318777, "generation_token_f1": 0.06986899563318777, "prompt_id": "fbdd1075d5b5f58e", "reading_canonical_log_prob": -9.382320404052734, "reading_distances": [25, 19, 24, 15], "reading_log_probs": [-8.711532592773438, -8.536040306091309, -9.076618194580078, -8.795395851135254], "reading_logprob_std": 0.22540590131196606, "tokenization_path_logprob_std": 38.26365330657089, "trajectory_balance_rmse": 192.48794240767103} +{"canonical_length": 25, "canonical_nll": 1.4759715270996094, "generation_rouge_l_f1": 0.30769230769230765, "generation_token_f1": 0.3846153846153846, "prompt_id": "71ff4b9aa3b9c919", "reading_canonical_log_prob": -36.899288177490234, "reading_distances": [30, 22, 22, 23], "reading_log_probs": [-36.31425857543945, -36.580177307128906, -34.889652252197266, -36.5655632019043], "reading_logprob_std": 0.8077817366167388, "tokenization_path_logprob_std": 100.75997085719403, "trajectory_balance_rmse": 543.6195859256892} +{"canonical_length": 227, "canonical_nll": 0.5237842845496627, "generation_rouge_l_f1": 0.31315240083507306, "generation_token_f1": 0.48851774530271397, "prompt_id": "84c9e255e3abf202", "reading_canonical_log_prob": -118.89903259277344, "reading_distances": [29, 23, 23, 24], "reading_log_probs": [-122.79146575927734, -124.23934173583984, -125.17918395996094, -122.10591888427734], "reading_logprob_std": 1.3888626120688112, "tokenization_path_logprob_std": 73.02668918201657, "trajectory_balance_rmse": 4313.9331739080635} +{"canonical_length": 20, "canonical_nll": 0.3186829090118408, "generation_rouge_l_f1": 0.46153846153846156, "generation_token_f1": 0.46153846153846156, "prompt_id": "7aac721bb785059a", "reading_canonical_log_prob": -6.373658180236816, "reading_distances": [19, 22, 28, 23], "reading_log_probs": [-6.403043746948242, -6.7244367599487305, -6.011399745941162, -6.133668422698975], "reading_logprob_std": 0.31644117391738347, "tokenization_path_logprob_std": 44.61127043220368, "trajectory_balance_rmse": 393.92950222977436} +{"canonical_length": 16, "canonical_nll": 0.6540147662162781, "generation_rouge_l_f1": 0.8387096774193549, "generation_token_f1": 0.8387096774193549, "prompt_id": "e8e5a8f77a9d33c4", "reading_canonical_log_prob": -10.46423625946045, "reading_distances": [43, 41, 42, 43], "reading_log_probs": [-10.11688232421875, -8.18280029296875, -9.128328323364258, -9.286474227905273], "reading_logprob_std": 0.7929175798213779, "tokenization_path_logprob_std": 44.451751083322286, "trajectory_balance_rmse": 294.58195524063336} +{"canonical_length": 139, "canonical_nll": 1.3256963277034621, "generation_rouge_l_f1": 0.135, "generation_token_f1": 0.24, "prompt_id": "b0bb5fcadef34afc", "reading_canonical_log_prob": -184.27178955078125, "reading_distances": [16, 10, 16, 17], "reading_log_probs": [-190.52169799804688, -186.85104370117188, -188.37063598632812, -188.54994201660156], "reading_logprob_std": 1.5059924537095248, "tokenization_path_logprob_std": 119.51901926882952, "trajectory_balance_rmse": 2560.1646924783254} +{"canonical_length": 214, "canonical_nll": 1.4921612605870327, "generation_rouge_l_f1": 0.2017291066282421, "generation_token_f1": 0.38040345821325644, "prompt_id": "6db457998b4e4d8a", "reading_canonical_log_prob": -319.322509765625, "reading_distances": [39, 38, 39, 35], "reading_log_probs": [-314.8750305175781, -309.5017395019531, -310.9292907714844, -309.24053955078125], "reading_logprob_std": 2.600426498103639, "tokenization_path_logprob_std": 206.42333534021958, "trajectory_balance_rmse": 3951.782658145172} +{"canonical_length": 124, "canonical_nll": 1.4859520696824597, "generation_rouge_l_f1": 0.16551724137931031, "generation_token_f1": 0.2758620689655173, "prompt_id": "b7fe2d1d82e7517d", "reading_canonical_log_prob": -184.258056640625, "reading_distances": [21, 28, 32, 26], "reading_log_probs": [-183.51602172851562, -185.38058471679688, -184.447509765625, -183.24234008789062], "reading_logprob_std": 0.9710097686876693, "tokenization_path_logprob_std": 41.95986045571869, "trajectory_balance_rmse": 2343.539250587481} +{"canonical_length": 3, "canonical_nll": 1.0550432999928792, "generation_rouge_l_f1": 0.0076045627376425855, "generation_token_f1": 0.0076045627376425855, "prompt_id": "fe8dff1c5ccd0457", "reading_canonical_log_prob": -3.1651298999786377, "reading_distances": [57, 77, 54, 55], "reading_log_probs": [-3.5782272815704346, -6.103477478027344, -4.232165336608887, -3.4826178550720215], "reading_logprob_std": 1.2160797814392632, "tokenization_path_logprob_std": 17.11727937793052, "trajectory_balance_rmse": 68.15990833673534} +{"canonical_length": 36, "canonical_nll": 1.374233881632487, "generation_rouge_l_f1": 0.10909090909090911, "generation_token_f1": 0.16727272727272727, "prompt_id": "007cda46c19b41d2", "reading_canonical_log_prob": -49.47241973876953, "reading_distances": [22, 16, 22, 16], "reading_log_probs": [-50.10575866699219, -53.46506118774414, -52.413177490234375, -51.89038848876953], "reading_logprob_std": 1.403963449625002, "tokenization_path_logprob_std": 150.87397261973373, "trajectory_balance_rmse": 889.2841904021487} +{"canonical_length": 121, "canonical_nll": 1.7307900673101757, "generation_rouge_l_f1": 0.2487046632124352, "generation_token_f1": 0.39378238341968913, "prompt_id": "0f24b7ddfdbc9dca", "reading_canonical_log_prob": -209.42559814453125, "reading_distances": [20, 21, 20, 22], "reading_log_probs": [-214.68734741210938, -217.2478790283203, -217.5260772705078, -214.80624389648438], "reading_logprob_std": 1.529305825693441, "tokenization_path_logprob_std": 83.96774900047616, "trajectory_balance_rmse": 2025.6340040685413} +{"canonical_length": 4, "canonical_nll": 0.8253794312477112, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.2222222222222222, "prompt_id": "0131c4e6a39cb194", "reading_canonical_log_prob": -3.3015177249908447, "reading_distances": [17, 27, 22, 21], "reading_log_probs": [-3.1061439514160156, -3.4635071754455566, -3.341745615005493, -3.240565776824951], "reading_logprob_std": 0.15167178088313463, "tokenization_path_logprob_std": 6.462283724567541, "trajectory_balance_rmse": 56.62794350711432} +{"canonical_length": 182, "canonical_nll": 1.0372909713577438, "generation_rouge_l_f1": 0.09999999999999999, "generation_token_f1": 0.2181818181818182, "prompt_id": "21b0719467934ebb", "reading_canonical_log_prob": -188.78695678710938, "reading_distances": [46, 42, 48, 46], "reading_log_probs": [-180.21688842773438, -179.26266479492188, -180.52651977539062, -179.5760040283203], "reading_logprob_std": 0.5785137709793269, "tokenization_path_logprob_std": 202.18838676446603, "trajectory_balance_rmse": 3003.8956161289793} +{"canonical_length": 246, "canonical_nll": 1.374127767919525, "generation_rouge_l_f1": 0.24206349206349206, "generation_token_f1": 0.4642857142857143, "prompt_id": "5d16c846337cb10b", "reading_canonical_log_prob": -338.0354309082031, "reading_distances": [17, 19, 21, 15], "reading_log_probs": [-336.99835205078125, -336.1363830566406, -334.9102783203125, -336.5899658203125], "reading_logprob_std": 0.9037072723681364, "tokenization_path_logprob_std": 131.3524529569415, "trajectory_balance_rmse": 5242.506559901419} +{"canonical_length": 116, "canonical_nll": 0.5277818482497643, "generation_rouge_l_f1": 0.34394904458598724, "generation_token_f1": 0.47133757961783435, "prompt_id": "d66394bc1b63c0cb", "reading_canonical_log_prob": -61.222694396972656, "reading_distances": [22, 14, 17, 19], "reading_log_probs": [-62.64634704589844, -62.8958740234375, -63.43498611450195, -62.4798469543457], "reading_logprob_std": 0.41712795376515405, "tokenization_path_logprob_std": 40.38451932804418, "trajectory_balance_rmse": 1194.6239266180503} +{"canonical_length": 75, "canonical_nll": 0.942134501139323, "generation_rouge_l_f1": 0.17301038062283738, "generation_token_f1": 0.3044982698961938, "prompt_id": "4b9d4e7ec5d5b8b5", "reading_canonical_log_prob": -70.66008758544922, "reading_distances": [39, 32, 36, 40], "reading_log_probs": [-66.9906234741211, -66.46339416503906, -67.83158874511719, -66.02198791503906], "reading_logprob_std": 0.7780798603190028, "tokenization_path_logprob_std": 130.4718723314256, "trajectory_balance_rmse": 1456.0589348364078} +{"canonical_length": 363, "canonical_nll": 1.3069955704954372, "generation_rouge_l_f1": 0.28205128205128205, "generation_token_f1": 0.532051282051282, "prompt_id": "cddcde885956c90e", "reading_canonical_log_prob": -474.43939208984375, "reading_distances": [21, 21, 25, 24], "reading_log_probs": [-467.8624572753906, -468.2728576660156, -467.5155029296875, -467.1064147949219], "reading_logprob_std": 0.4968177315903413, "tokenization_path_logprob_std": 138.02641998855637, "trajectory_balance_rmse": 7805.291805942497} +{"canonical_length": 204, "canonical_nll": 1.6790197035845589, "generation_rouge_l_f1": 0.19313304721030042, "generation_token_f1": 0.36909871244635195, "prompt_id": "e708ffc93eb420d5", "reading_canonical_log_prob": -342.52001953125, "reading_distances": [40, 42, 33, 39], "reading_log_probs": [-334.7813720703125, -336.7624206542969, -335.57110595703125, -337.7404479980469], "reading_logprob_std": 1.3033986514109008, "tokenization_path_logprob_std": 13.606629787851759, "trajectory_balance_rmse": 4266.552008006529} +{"canonical_length": 73, "canonical_nll": 1.696274848833476, "generation_rouge_l_f1": 0.18, "generation_token_f1": 0.25333333333333335, "prompt_id": "276028c9e0375416", "reading_canonical_log_prob": -123.82806396484375, "reading_distances": [30, 25, 29, 25], "reading_log_probs": [-126.35247802734375, -127.31460571289062, -127.69268798828125, -126.97342681884766], "reading_logprob_std": 0.5689255735537793, "tokenization_path_logprob_std": 141.03333887028452, "trajectory_balance_rmse": 1607.3319544907063} +{"canonical_length": 100, "canonical_nll": 0.94643798828125, "generation_rouge_l_f1": 0.48275862068965514, "generation_token_f1": 0.7011494252873562, "prompt_id": "328ee04612e4bcf9", "reading_canonical_log_prob": -94.643798828125, "reading_distances": [28, 27, 25, 29], "reading_log_probs": [-101.58056640625, -98.47746276855469, -98.55479431152344, -101.41432189941406], "reading_logprob_std": 1.7228901575432911, "tokenization_path_logprob_std": 110.50644962955457, "trajectory_balance_rmse": 1749.7132175394142} +{"canonical_length": 90, "canonical_nll": 1.0611573113335504, "generation_rouge_l_f1": 0.15028901734104047, "generation_token_f1": 0.2485549132947977, "prompt_id": "bdf95ab3a5ffa3ff", "reading_canonical_log_prob": -95.50415802001953, "reading_distances": [25, 34, 30, 30], "reading_log_probs": [-93.34233093261719, -92.56382751464844, -91.07252502441406, -92.07637786865234], "reading_logprob_std": 0.9500015087931469, "tokenization_path_logprob_std": 80.97793677822776, "trajectory_balance_rmse": 1119.7168964009832} +{"canonical_length": 132, "canonical_nll": 1.6972814617734966, "generation_rouge_l_f1": 0.242914979757085, "generation_token_f1": 0.4534412955465587, "prompt_id": "80019145cee29a1a", "reading_canonical_log_prob": -224.04115295410156, "reading_distances": [18, 18, 19, 24], "reading_log_probs": [-221.21295166015625, -216.32691955566406, -219.26358032226562, -216.98008728027344], "reading_logprob_std": 2.233372714920074, "tokenization_path_logprob_std": 124.36315074136216, "trajectory_balance_rmse": 2719.531320907496} +{"canonical_length": 250, "canonical_nll": 1.372605712890625, "generation_rouge_l_f1": 0.2834331337325349, "generation_token_f1": 0.4550898203592814, "prompt_id": "77663077eb493ddd", "reading_canonical_log_prob": -343.15142822265625, "reading_distances": [31, 30, 29, 30], "reading_log_probs": [-341.34033203125, -340.5293273925781, -342.1888732910156, -341.9749755859375], "reading_logprob_std": 0.7455572161812912, "tokenization_path_logprob_std": 118.34384370524288, "trajectory_balance_rmse": 5047.2442636778105} +{"canonical_length": 369, "canonical_nll": 1.1641873822302675, "generation_rouge_l_f1": 0.3064516129032258, "generation_token_f1": 0.5935483870967742, "prompt_id": "4825c22b71594c0c", "reading_canonical_log_prob": -429.58514404296875, "reading_distances": [32, 32, 36, 32], "reading_log_probs": [-434.4554443359375, -437.27313232421875, -438.94189453125, -432.34014892578125], "reading_logprob_std": 2.9332037828222606, "tokenization_path_logprob_std": 159.33826763634224, "trajectory_balance_rmse": 7499.36966857105} +{"canonical_length": 198, "canonical_nll": 1.1000085695825441, "generation_rouge_l_f1": 0.2669683257918552, "generation_token_f1": 0.493212669683258, "prompt_id": "c520430df28e5bdb", "reading_canonical_log_prob": -217.80169677734375, "reading_distances": [17, 21, 22, 21], "reading_log_probs": [-215.71371459960938, -216.44908142089844, -216.37722778320312, -217.24502563476562], "reading_logprob_std": 0.6270094922900553, "tokenization_path_logprob_std": 182.97849360294836, "trajectory_balance_rmse": 4026.1976635240712} +{"canonical_length": 290, "canonical_nll": 1.3359288708917025, "generation_rouge_l_f1": 0.2423146473779385, "generation_token_f1": 0.5352622061482821, "prompt_id": "c3e62d8cbe21b9a4", "reading_canonical_log_prob": -387.41937255859375, "reading_distances": [25, 23, 28, 26], "reading_log_probs": [-386.40625, -387.1230773925781, -387.40899658203125, -384.4593505859375], "reading_logprob_std": 1.328763878340439, "tokenization_path_logprob_std": 131.12521268367036, "trajectory_balance_rmse": 6212.920471551909} +{"canonical_length": 38, "canonical_nll": 0.7079125454551295, "generation_rouge_l_f1": 0.199288256227758, "generation_token_f1": 0.24199288256227763, "prompt_id": "ef7fdf70b61d5a5a", "reading_canonical_log_prob": -26.900676727294922, "reading_distances": [34, 27, 27, 31], "reading_log_probs": [-30.118928909301758, -28.871763229370117, -31.9217586517334, -31.12663459777832], "reading_logprob_std": 1.3178336798721833, "tokenization_path_logprob_std": 84.3031193606693, "trajectory_balance_rmse": 662.917585510955} +{"canonical_length": 262, "canonical_nll": 0.9491500272095659, "generation_rouge_l_f1": 0.2595419847328244, "generation_token_f1": 0.5610687022900763, "prompt_id": "f0bfbb195bb03bbd", "reading_canonical_log_prob": -248.67730712890625, "reading_distances": [38, 26, 27, 32], "reading_log_probs": [-248.25885009765625, -248.96353149414062, -249.82162475585938, -248.69265747070312], "reading_logprob_std": 0.6589950210431733, "tokenization_path_logprob_std": 143.91755382033946, "trajectory_balance_rmse": 5766.339051077253} +{"canonical_length": 12, "canonical_nll": 0.4950527747472127, "generation_rouge_l_f1": 0.7000000000000001, "generation_token_f1": 0.7000000000000001, "prompt_id": "064f3342c7039feb", "reading_canonical_log_prob": -5.940633296966553, "reading_distances": [32, 31, 33, 30], "reading_log_probs": [-6.468686103820801, -6.455379486083984, -6.54801082611084, -6.870715141296387], "reading_logprob_std": 0.19436202717711926, "tokenization_path_logprob_std": 47.55223549699323, "trajectory_balance_rmse": 281.378675490576} +{"canonical_length": 174, "canonical_nll": 1.457019148201778, "generation_rouge_l_f1": 0.20956719817767655, "generation_token_f1": 0.4100227790432802, "prompt_id": "5319439232a42bef", "reading_canonical_log_prob": -253.52133178710938, "reading_distances": [27, 27, 25, 31], "reading_log_probs": [-252.46795654296875, -250.5182647705078, -251.32037353515625, -251.2772979736328], "reading_logprob_std": 0.8040148202369343, "tokenization_path_logprob_std": 163.5557084488935, "trajectory_balance_rmse": 4153.001656788848} +{"canonical_length": 274, "canonical_nll": 1.179617665979984, "generation_rouge_l_f1": 0.2527472527472527, "generation_token_f1": 0.5347985347985348, "prompt_id": "7f189f92ecaccbe7", "reading_canonical_log_prob": -323.2152404785156, "reading_distances": [21, 13, 20, 19], "reading_log_probs": [-326.64990234375, -324.1112060546875, -325.2889404296875, -326.30792236328125], "reading_logprob_std": 1.1425516569707264, "tokenization_path_logprob_std": 88.28984423493472, "trajectory_balance_rmse": 5691.058253285187} +{"canonical_length": 225, "canonical_nll": 1.054514431423611, "generation_rouge_l_f1": 0.4978354978354978, "generation_token_f1": 0.6666666666666666, "prompt_id": "89ab6e8e7235950e", "reading_canonical_log_prob": -237.2657470703125, "reading_distances": [30, 18, 26, 22], "reading_log_probs": [-234.55770874023438, -231.99578857421875, -234.41879272460938, -234.583984375], "reading_logprob_std": 1.2642658575489458, "tokenization_path_logprob_std": 164.0292972836186, "trajectory_balance_rmse": 4933.6813786486955} +{"canonical_length": 188, "canonical_nll": 1.2836604016892454, "generation_rouge_l_f1": 0.27586206896551724, "generation_token_f1": 0.5103448275862069, "prompt_id": "3d38b4ece7a08884", "reading_canonical_log_prob": -241.32815551757812, "reading_distances": [26, 21, 21, 29], "reading_log_probs": [-246.11441040039062, -241.59849548339844, -243.62496948242188, -244.19012451171875], "reading_logprob_std": 1.8582298453815627, "tokenization_path_logprob_std": 70.85192840994294, "trajectory_balance_rmse": 3993.3541252011382} +{"canonical_length": 169, "canonical_nll": 1.7601972049509986, "generation_rouge_l_f1": 0.16628175519630486, "generation_token_f1": 0.31870669745958424, "prompt_id": "49135037e7b65f9f", "reading_canonical_log_prob": -297.47332763671875, "reading_distances": [10, 15, 17, 16], "reading_log_probs": [-295.99761962890625, -301.65374755859375, -300.760986328125, -298.1061706542969], "reading_logprob_std": 2.574843096569594, "tokenization_path_logprob_std": 174.4609413685641, "trajectory_balance_rmse": 4338.941884164515} +{"canonical_length": 250, "canonical_nll": 1.3928369140625, "generation_rouge_l_f1": 0.2901960784313726, "generation_token_f1": 0.5294117647058825, "prompt_id": "daaf9a26e9fd85ac", "reading_canonical_log_prob": -348.209228515625, "reading_distances": [37, 30, 33, 34], "reading_log_probs": [-345.0783386230469, -344.63970947265625, -344.91748046875, -344.31304931640625], "reading_logprob_std": 0.3357999407026041, "tokenization_path_logprob_std": 64.94427663722774, "trajectory_balance_rmse": 6403.909740721896} +{"canonical_length": 32, "canonical_nll": 1.4294737577438354, "generation_rouge_l_f1": 0.11074918566775245, "generation_token_f1": 0.15635179153094464, "prompt_id": "06a9af54aedbd3a8", "reading_canonical_log_prob": -45.743160247802734, "reading_distances": [43, 44, 45, 39], "reading_log_probs": [-50.924461364746094, -51.83375549316406, -49.10915756225586, -49.24909973144531], "reading_logprob_std": 1.3245264377649533, "tokenization_path_logprob_std": 54.385268106360904, "trajectory_balance_rmse": 740.8666720039778} +{"canonical_length": 48, "canonical_nll": 0.635998527208964, "generation_rouge_l_f1": 0.15596330275229356, "generation_token_f1": 0.2385321100917431, "prompt_id": "4148986bb9e9364b", "reading_canonical_log_prob": -30.527929306030273, "reading_distances": [11, 9, 6, 10], "reading_log_probs": [-30.783905029296875, -30.463619232177734, -29.03377914428711, -30.26595687866211], "reading_logprob_std": 0.7657006020626549, "tokenization_path_logprob_std": 85.81890890384679, "trajectory_balance_rmse": 988.2881078531991} +{"canonical_length": 19, "canonical_nll": 0.9748587357370477, "generation_rouge_l_f1": 0.12318840579710144, "generation_token_f1": 0.14492753623188406, "prompt_id": "1c5e2235e187365f", "reading_canonical_log_prob": -18.522315979003906, "reading_distances": [23, 24, 19, 15], "reading_log_probs": [-17.703100204467773, -17.170299530029297, -18.475589752197266, -18.0892276763916], "reading_logprob_std": 0.5573148601772555, "tokenization_path_logprob_std": 61.18236583236273, "trajectory_balance_rmse": 335.362997217949} +{"canonical_length": 369, "canonical_nll": 1.1127167988598832, "generation_rouge_l_f1": 0.24959481361426256, "generation_token_f1": 0.4505672609400324, "prompt_id": "6125f7b585f4ca3f", "reading_canonical_log_prob": -410.5924987792969, "reading_distances": [32, 19, 29, 26], "reading_log_probs": [-416.474365234375, -416.84765625, -419.6036376953125, -417.560791015625], "reading_logprob_std": 1.39610263390784, "tokenization_path_logprob_std": 346.73662501956505, "trajectory_balance_rmse": 7838.893230909157} +{"canonical_length": 21, "canonical_nll": 0.9882870628720238, "generation_rouge_l_f1": 0.04395604395604396, "generation_token_f1": 0.04395604395604396, "prompt_id": "bac6bf1b8815878e", "reading_canonical_log_prob": -20.7540283203125, "reading_distances": [19, 20, 17, 16], "reading_log_probs": [-18.782512664794922, -18.820951461791992, -18.87398338317871, -18.83310317993164], "reading_logprob_std": 0.03767743977419971, "tokenization_path_logprob_std": 75.85414305827912, "trajectory_balance_rmse": 435.7737327752578} +{"canonical_length": 81, "canonical_nll": 2.5625762939453125, "generation_rouge_l_f1": 0.25316455696202533, "generation_token_f1": 0.27848101265822783, "prompt_id": "3e2b40881e16b7da", "reading_canonical_log_prob": -207.5686798095703, "reading_distances": [22, 14, 19, 18], "reading_log_probs": [-206.71701049804688, -206.87808227539062, -206.63494873046875, -206.7848663330078], "reading_logprob_std": 0.10310233008384612, "tokenization_path_logprob_std": 75.67960190672183, "trajectory_balance_rmse": 1578.0464259007256} +{"canonical_length": 32, "canonical_nll": 1.2067759037017822, "generation_rouge_l_f1": 0.14093959731543626, "generation_token_f1": 0.21476510067114093, "prompt_id": "e01d08fb030556a3", "reading_canonical_log_prob": -38.61682891845703, "reading_distances": [32, 27, 24, 30], "reading_log_probs": [-40.377017974853516, -40.1644401550293, -40.08639144897461, -40.42451095581055], "reading_logprob_std": 0.1632895978276619, "tokenization_path_logprob_std": 118.18006124664926, "trajectory_balance_rmse": 755.0510689026797} +{"canonical_length": 64, "canonical_nll": 1.3212883472442627, "generation_rouge_l_f1": 0.1212121212121212, "generation_token_f1": 0.15151515151515152, "prompt_id": "68a9acf034dc69e5", "reading_canonical_log_prob": -84.56245422363281, "reading_distances": [28, 30, 23, 21], "reading_log_probs": [-83.65023803710938, -85.21611785888672, -84.38937377929688, -83.22726440429688], "reading_logprob_std": 0.8740113458510718, "tokenization_path_logprob_std": 78.92885590353266, "trajectory_balance_rmse": 1505.767083896746} +{"canonical_length": 196, "canonical_nll": 1.020385975740394, "generation_rouge_l_f1": 0.3023758099352052, "generation_token_f1": 0.5485961123110151, "prompt_id": "12fd41da356aa8b2", "reading_canonical_log_prob": -199.9956512451172, "reading_distances": [32, 31, 33, 31], "reading_log_probs": [-201.57675170898438, -201.73321533203125, -199.7049560546875, -205.5599365234375], "reading_logprob_std": 2.4568321093975913, "tokenization_path_logprob_std": 119.19625931016829, "trajectory_balance_rmse": 3882.908860056541} +{"canonical_length": 404, "canonical_nll": 0.931619927434638, "generation_rouge_l_f1": 0.22085889570552147, "generation_token_f1": 0.3742331288343558, "prompt_id": "7a350261a78f28df", "reading_canonical_log_prob": -376.37445068359375, "reading_distances": [20, 20, 20, 25], "reading_log_probs": [-373.3504943847656, -374.34771728515625, -375.4652099609375, -376.644287109375], "reading_logprob_std": 1.4209384150905777, "tokenization_path_logprob_std": 212.95317345752838, "trajectory_balance_rmse": 6741.412447395004} +{"canonical_length": 316, "canonical_nll": 1.4097900390625, "generation_rouge_l_f1": 0.13913043478260867, "generation_token_f1": 0.22608695652173913, "prompt_id": "732a9ef0a365d23a", "reading_canonical_log_prob": -445.49365234375, "reading_distances": [20, 24, 26, 24], "reading_log_probs": [-442.3637390136719, -443.77996826171875, -441.1238708496094, -445.3658447265625], "reading_logprob_std": 1.8284742513856056, "tokenization_path_logprob_std": 217.48227149575925, "trajectory_balance_rmse": 8126.502382861726} +{"canonical_length": 244, "canonical_nll": 1.6215057373046875, "generation_rouge_l_f1": 0.27180527383367137, "generation_token_f1": 0.486815415821501, "prompt_id": "26e0cf383cca3d0a", "reading_canonical_log_prob": -395.64739990234375, "reading_distances": [27, 33, 27, 30], "reading_log_probs": [-394.2096862792969, -391.8919677734375, -392.238525390625, -394.1196594238281], "reading_logprob_std": 1.2208868592906874, "tokenization_path_logprob_std": 37.923332340386544, "trajectory_balance_rmse": 5876.1791142628035} +{"canonical_length": 123, "canonical_nll": 1.7154547218384781, "generation_rouge_l_f1": 0.20266666666666663, "generation_token_f1": 0.3786666666666667, "prompt_id": "92f09a8894ee3f5f", "reading_canonical_log_prob": -211.0009307861328, "reading_distances": [14, 20, 14, 18], "reading_log_probs": [-221.01947021484375, -221.65151977539062, -221.193115234375, -220.98410034179688], "reading_logprob_std": 0.30688772602216313, "tokenization_path_logprob_std": 72.21457584108074, "trajectory_balance_rmse": 2915.3175208678567} +{"canonical_length": 114, "canonical_nll": 1.1969123974181057, "generation_rouge_l_f1": 0.42152466367713004, "generation_token_f1": 0.5739910313901345, "prompt_id": "eb8e055ee1d3ca60", "reading_canonical_log_prob": -136.44801330566406, "reading_distances": [21, 21, 17, 21], "reading_log_probs": [-140.7734375, -140.88229370117188, -138.58132934570312, -140.9171905517578], "reading_logprob_std": 1.1398009731903456, "tokenization_path_logprob_std": 180.60696769266417, "trajectory_balance_rmse": 2670.0447969914926} +{"canonical_length": 221, "canonical_nll": 1.6104552670301895, "generation_rouge_l_f1": 0.23132530120481928, "generation_token_f1": 0.4385542168674698, "prompt_id": "5a05098dc8854389", "reading_canonical_log_prob": -355.9106140136719, "reading_distances": [22, 20, 19, 25], "reading_log_probs": [-362.3785400390625, -358.667236328125, -356.00347900390625, -360.70758056640625], "reading_logprob_std": 2.747642371188086, "tokenization_path_logprob_std": 174.95541083414975, "trajectory_balance_rmse": 4703.240052170219} +{"canonical_length": 6, "canonical_nll": 0.8682984511057535, "generation_rouge_l_f1": 0.8000000000000002, "generation_token_f1": 0.8000000000000002, "prompt_id": "0744be99f0e84455", "reading_canonical_log_prob": -5.2097907066345215, "reading_distances": [20, 24, 17, 15], "reading_log_probs": [-5.285199165344238, -4.910661697387695, -5.437718868255615, -5.509424209594727], "reading_logprob_std": 0.26696830469067173, "tokenization_path_logprob_std": 46.11117370713604, "trajectory_balance_rmse": 157.3116235089807} +{"canonical_length": 3, "canonical_nll": 3.148922602335612, "generation_rouge_l_f1": 0.015748031496062992, "generation_token_f1": 0.015748031496062992, "prompt_id": "1e2019bd6514f4f1", "reading_canonical_log_prob": -9.446767807006836, "reading_distances": [45, 55, 50, 54], "reading_log_probs": [-8.8090238571167, -9.368242263793945, -9.429305076599121, -9.15910816192627], "reading_logprob_std": 0.27995612014112226, "tokenization_path_logprob_std": 16.540031014953147, "trajectory_balance_rmse": 39.84482214496705} +{"canonical_length": 17, "canonical_nll": 0.9500353196088005, "generation_rouge_l_f1": 0.5384615384615385, "generation_token_f1": 0.5384615384615385, "prompt_id": "982c9409d78c0796", "reading_canonical_log_prob": -16.15060043334961, "reading_distances": [16, 16, 19, 17], "reading_log_probs": [-23.660579681396484, -21.54863166809082, -19.0662899017334, -18.667604446411133], "reading_logprob_std": 2.329501257010229, "tokenization_path_logprob_std": 58.445187815886314, "trajectory_balance_rmse": 436.2414567184898} +{"canonical_length": 185, "canonical_nll": 1.4712715767525337, "generation_rouge_l_f1": 0.2162162162162162, "generation_token_f1": 0.4144144144144144, "prompt_id": "f2e5fb29aaf06273", "reading_canonical_log_prob": -272.18524169921875, "reading_distances": [35, 39, 36, 37], "reading_log_probs": [-266.0307922363281, -270.4259033203125, -272.07635498046875, -271.8238220214844], "reading_logprob_std": 2.801293818615827, "tokenization_path_logprob_std": 236.62437307438805, "trajectory_balance_rmse": 3828.659951443487} +{"canonical_length": 309, "canonical_nll": 1.0198542030112256, "generation_rouge_l_f1": 0.2814569536423841, "generation_token_f1": 0.4503311258278146, "prompt_id": "3aa4d1f8b5499bdf", "reading_canonical_log_prob": -315.13494873046875, "reading_distances": [65, 56, 54, 65], "reading_log_probs": [-312.4190673828125, -312.0469970703125, -309.8651123046875, -316.0318603515625], "reading_logprob_std": 2.555744083887824, "tokenization_path_logprob_std": 243.77554385336776, "trajectory_balance_rmse": 5217.99848830447} +{"canonical_length": 253, "canonical_nll": 1.5950260690078433, "generation_rouge_l_f1": 0.20553359683794467, "generation_token_f1": 0.4426877470355731, "prompt_id": "aa727754a09ad816", "reading_canonical_log_prob": -403.5415954589844, "reading_distances": [24, 22, 21, 24], "reading_log_probs": [-405.63818359375, -405.5276794433594, -406.5260009765625, -409.6609191894531], "reading_logprob_std": 1.9341419593985787, "tokenization_path_logprob_std": 65.14262025390839, "trajectory_balance_rmse": 5565.5601905737285} +{"canonical_length": 218, "canonical_nll": 1.2742564350093177, "generation_rouge_l_f1": 0.28512396694214875, "generation_token_f1": 0.49173553719008267, "prompt_id": "4d9af47029652710", "reading_canonical_log_prob": -277.78790283203125, "reading_distances": [22, 24, 22, 19], "reading_log_probs": [-290.92578125, -283.57000732421875, -285.6549987792969, -284.6715087890625], "reading_logprob_std": 3.2600131559295993, "tokenization_path_logprob_std": 157.80951666822747, "trajectory_balance_rmse": 4989.432618406828} +{"canonical_length": 286, "canonical_nll": 1.1939862658093858, "generation_rouge_l_f1": 0.19421487603305784, "generation_token_f1": 0.4214876033057851, "prompt_id": "4d68d14019e3f5eb", "reading_canonical_log_prob": -341.4800720214844, "reading_distances": [25, 24, 24, 22], "reading_log_probs": [-334.90570068359375, -339.487548828125, -337.5904846191406, -338.09942626953125], "reading_logprob_std": 1.9188980946664755, "tokenization_path_logprob_std": 357.67974762558856, "trajectory_balance_rmse": 5416.605831995352} +{"canonical_length": 104, "canonical_nll": 1.3178800436166616, "generation_rouge_l_f1": 0.1764705882352941, "generation_token_f1": 0.2941176470588235, "prompt_id": "9752f3aa2c81e0d1", "reading_canonical_log_prob": -137.0595245361328, "reading_distances": [23, 32, 17, 28], "reading_log_probs": [-131.585693359375, -132.48684692382812, -137.5747833251953, -131.52122497558594], "reading_logprob_std": 2.8889231529143378, "tokenization_path_logprob_std": 91.32610826821794, "trajectory_balance_rmse": 2379.3973568826395} +{"canonical_length": 129, "canonical_nll": 0.3131689618724261, "generation_rouge_l_f1": 0.5650969529085872, "generation_token_f1": 0.6038781163434902, "prompt_id": "46cccd71028358c4", "reading_canonical_log_prob": -40.39879608154297, "reading_distances": [20, 25, 24, 25], "reading_log_probs": [-40.177425384521484, -39.220909118652344, -40.03211212158203, -40.80492401123047], "reading_logprob_std": 0.6515485110126386, "tokenization_path_logprob_std": 61.19546257533747, "trajectory_balance_rmse": 1301.7275090464357} +{"canonical_length": 3, "canonical_nll": 0.8825006484985352, "generation_rouge_l_f1": 0.011627906976744186, "generation_token_f1": 0.011627906976744186, "prompt_id": "f385579e0d438649", "reading_canonical_log_prob": -2.6475019454956055, "reading_distances": [15, 11, 14, 12], "reading_log_probs": [-1.9564892053604126, -2.1100211143493652, -2.225541591644287, -2.0717902183532715], "reading_logprob_std": 0.11094353823454621, "tokenization_path_logprob_std": 15.435951782584375, "trajectory_balance_rmse": 36.67991923372804} +{"canonical_length": 3, "canonical_nll": 3.1389284133911133, "generation_rouge_l_f1": 0.014760147601476014, "generation_token_f1": 0.014760147601476014, "prompt_id": "f015c3c34e66dddf", "reading_canonical_log_prob": -9.41678524017334, "reading_distances": [26, 19, 21, 22], "reading_log_probs": [-9.149847984313965, -9.704351425170898, -8.956768989562988, -8.938608169555664], "reading_logprob_std": 0.35764832242585726, "tokenization_path_logprob_std": 17.840226719222557, "trajectory_balance_rmse": 34.46514188832861} +{"canonical_length": 47, "canonical_nll": 0.6774944548911237, "generation_rouge_l_f1": 0.1652892561983471, "generation_token_f1": 0.19008264462809918, "prompt_id": "f91ba1a01be59af5", "reading_canonical_log_prob": -31.842239379882812, "reading_distances": [25, 28, 29, 28], "reading_log_probs": [-32.034027099609375, -32.9599723815918, -32.16444396972656, -31.7476806640625], "reading_logprob_std": 0.5190203770734876, "tokenization_path_logprob_std": 110.75114717885006, "trajectory_balance_rmse": 784.7913564153811} +{"canonical_length": 179, "canonical_nll": 1.4919387561648918, "generation_rouge_l_f1": 0.26, "generation_token_f1": 0.5, "prompt_id": "0764d55b71d1b589", "reading_canonical_log_prob": -267.0570373535156, "reading_distances": [14, 15, 18, 17], "reading_log_probs": [-267.296630859375, -264.95703125, -263.8807373046875, -265.3001708984375], "reading_logprob_std": 1.4265007502884468, "tokenization_path_logprob_std": 235.15617241521255, "trajectory_balance_rmse": 3870.972528021758} +{"canonical_length": 75, "canonical_nll": 1.4789422607421876, "generation_rouge_l_f1": 0.12034383954154727, "generation_token_f1": 0.24068767908309455, "prompt_id": "a40186552cfde2fc", "reading_canonical_log_prob": -110.92066955566406, "reading_distances": [12, 12, 6, 5], "reading_log_probs": [-116.04682922363281, -114.09713745117188, -110.87159729003906, -110.574951171875], "reading_logprob_std": 2.6366594620208494, "tokenization_path_logprob_std": 75.08423534549297, "trajectory_balance_rmse": 1518.8665954997527} +{"canonical_length": 263, "canonical_nll": 1.1040004251574382, "generation_rouge_l_f1": 0.30740037950664134, "generation_token_f1": 0.6034155597722961, "prompt_id": "a3ec61d54b8539e4", "reading_canonical_log_prob": -290.35211181640625, "reading_distances": [22, 25, 26, 23], "reading_log_probs": [-289.10498046875, -290.09521484375, -289.2660827636719, -289.37396240234375], "reading_logprob_std": 0.43762354963576633, "tokenization_path_logprob_std": 187.34987043591195, "trajectory_balance_rmse": 5931.5287042698255} +{"canonical_length": 210, "canonical_nll": 1.5696807861328126, "generation_rouge_l_f1": 0.1791044776119403, "generation_token_f1": 0.3667377398720682, "prompt_id": "ca23abf600fc132e", "reading_canonical_log_prob": -329.6329650878906, "reading_distances": [43, 45, 50, 45], "reading_log_probs": [-329.0832824707031, -328.0029296875, -328.6209411621094, -327.74896240234375], "reading_logprob_std": 0.60333214931705, "tokenization_path_logprob_std": 93.59242535398307, "trajectory_balance_rmse": 5116.365793559218} +{"canonical_length": 132, "canonical_nll": 1.568010156804865, "generation_rouge_l_f1": 0.34877384196185285, "generation_token_f1": 0.5177111716621254, "prompt_id": "a83cfaa0169d4436", "reading_canonical_log_prob": -206.9773406982422, "reading_distances": [40, 56, 39, 47], "reading_log_probs": [-210.92967224121094, -212.46145629882812, -209.79876708984375, -214.05978393554688], "reading_logprob_std": 1.853459326432698, "tokenization_path_logprob_std": 90.125520899413, "trajectory_balance_rmse": 1048.6174211896198} +{"canonical_length": 115, "canonical_nll": 1.4065131145974865, "generation_rouge_l_f1": 0.2761506276150628, "generation_token_f1": 0.4853556485355649, "prompt_id": "bd9008593d2eaa20", "reading_canonical_log_prob": -161.74900817871094, "reading_distances": [41, 42, 41, 41], "reading_log_probs": [-164.1239013671875, -164.5350341796875, -165.79891967773438, -165.3609161376953], "reading_logprob_std": 0.7624662142492353, "tokenization_path_logprob_std": 30.987989569427345, "trajectory_balance_rmse": 2869.8110703086913} +{"canonical_length": 7, "canonical_nll": 1.0465387616838728, "generation_rouge_l_f1": 0.04428044280442805, "generation_token_f1": 0.04428044280442805, "prompt_id": "48bde41f90ff68c1", "reading_canonical_log_prob": -7.325771331787109, "reading_distances": [31, 34, 27, 35], "reading_log_probs": [-10.176053047180176, -10.449481010437012, -9.446908950805664, -10.063825607299805], "reading_logprob_std": 0.42362286179385245, "tokenization_path_logprob_std": 35.460776684258214, "trajectory_balance_rmse": 169.68211712787593} +{"canonical_length": 130, "canonical_nll": 0.6682967552771935, "generation_rouge_l_f1": 0.34161490683229817, "generation_token_f1": 0.49689440993788825, "prompt_id": "8dbb7f1da28c2195", "reading_canonical_log_prob": -86.87857818603516, "reading_distances": [11, 9, 18, 15], "reading_log_probs": [-84.61102294921875, -83.34098052978516, -84.41312408447266, -85.08837127685547], "reading_logprob_std": 0.7381767024992427, "tokenization_path_logprob_std": 126.72554002399956, "trajectory_balance_rmse": 1903.307458416883} +{"canonical_length": 260, "canonical_nll": 2.0543172983022835, "generation_rouge_l_f1": 0.23352165725047078, "generation_token_f1": 0.48587570621468923, "prompt_id": "d859a6a8a580026f", "reading_canonical_log_prob": -534.1224975585938, "reading_distances": [38, 35, 33, 31], "reading_log_probs": [-539.9915771484375, -551.1435546875, -545.986328125, -544.58935546875], "reading_logprob_std": 4.591198587186648, "tokenization_path_logprob_std": 239.39450647474806, "trajectory_balance_rmse": 6130.251240393396} +{"canonical_length": 242, "canonical_nll": 1.2241468193117253, "generation_rouge_l_f1": 0.22134387351778656, "generation_token_f1": 0.43873517786561267, "prompt_id": "b4d85be43e32fd7f", "reading_canonical_log_prob": -296.2435302734375, "reading_distances": [37, 43, 34, 38], "reading_log_probs": [-295.3394470214844, -296.56597900390625, -296.575927734375, -298.53314208984375], "reading_logprob_std": 1.3207785287499716, "tokenization_path_logprob_std": 157.38511315175384, "trajectory_balance_rmse": 5729.429021778301} +{"canonical_length": 19, "canonical_nll": 1.1375531648334705, "generation_rouge_l_f1": 0.35294117647058826, "generation_token_f1": 0.35294117647058826, "prompt_id": "64c4e6c526576026", "reading_canonical_log_prob": -21.613510131835938, "reading_distances": [27, 28, 30, 29], "reading_log_probs": [-22.022937774658203, -21.102510452270508, -22.419960021972656, -21.22293472290039], "reading_logprob_std": 0.6342885084727071, "tokenization_path_logprob_std": 18.912016996555998, "trajectory_balance_rmse": 469.1002387546757} +{"canonical_length": 56, "canonical_nll": 1.4726554325648717, "generation_rouge_l_f1": 0.04487179487179487, "generation_token_f1": 0.09615384615384615, "prompt_id": "505a139a155b8b27", "reading_canonical_log_prob": -82.46870422363281, "reading_distances": [26, 29, 18, 32], "reading_log_probs": [-82.64680480957031, -82.8988037109375, -83.193359375, -84.606689453125], "reading_logprob_std": 0.8758097116694954, "tokenization_path_logprob_std": 77.81419429230296, "trajectory_balance_rmse": 1297.2920761651098} +{"canonical_length": 171, "canonical_nll": 1.0874625758120888, "generation_rouge_l_f1": 0.4837758112094395, "generation_token_f1": 0.6253687315634219, "prompt_id": "ca5c2935eb3033fb", "reading_canonical_log_prob": -185.9561004638672, "reading_distances": [18, 24, 21, 20], "reading_log_probs": [-183.3065185546875, -184.80850219726562, -185.18411254882812, -183.071533203125], "reading_logprob_std": 1.0589959119984205, "tokenization_path_logprob_std": 206.00096450255035, "trajectory_balance_rmse": 3672.499008028136} +{"canonical_length": 78, "canonical_nll": 0.814874013264974, "generation_rouge_l_f1": 0.1746987951807229, "generation_token_f1": 0.25903614457831325, "prompt_id": "f8aa35c9d5781c5a", "reading_canonical_log_prob": -63.56017303466797, "reading_distances": [22, 28, 23, 24], "reading_log_probs": [-63.91216278076172, -64.35385131835938, -63.5860481262207, -64.40034484863281], "reading_logprob_std": 0.386708741161569, "tokenization_path_logprob_std": 103.29060898328967, "trajectory_balance_rmse": 1551.8745811613367} +{"canonical_length": 339, "canonical_nll": 1.1842026612048304, "generation_rouge_l_f1": 0.21775544388609716, "generation_token_f1": 0.509212730318258, "prompt_id": "de613d5bcfdf1d4f", "reading_canonical_log_prob": -401.4447021484375, "reading_distances": [30, 37, 37, 36], "reading_log_probs": [-401.0164794921875, -400.360107421875, -401.1829528808594, -399.94134521484375], "reading_logprob_std": 0.5779617511752934, "tokenization_path_logprob_std": 415.7992740130077, "trajectory_balance_rmse": 7615.168402433706} +{"canonical_length": 19, "canonical_nll": 1.9562237388209294, "generation_rouge_l_f1": 0.08, "generation_token_f1": 0.08, "prompt_id": "cb6d19eef2166daa", "reading_canonical_log_prob": -37.168251037597656, "reading_distances": [23, 30, 27, 26], "reading_log_probs": [-34.21559143066406, -34.52787399291992, -34.354331970214844, -34.95412063598633], "reading_logprob_std": 0.3206431299066185, "tokenization_path_logprob_std": 93.4597119189705, "trajectory_balance_rmse": 514.2476766594789} +{"canonical_length": 389, "canonical_nll": 1.119570303085837, "generation_rouge_l_f1": 0.20447284345047925, "generation_token_f1": 0.39616613418530355, "prompt_id": "980fa61309c5ee0c", "reading_canonical_log_prob": -435.5128479003906, "reading_distances": [20, 23, 21, 26], "reading_log_probs": [-433.6981201171875, -437.2823486328125, -437.3597412109375, -437.1358947753906], "reading_logprob_std": 1.7830218557551758, "tokenization_path_logprob_std": 215.34746974518197, "trajectory_balance_rmse": 7665.611213869747} +{"canonical_length": 14, "canonical_nll": 2.4821760995047435, "generation_rouge_l_f1": 0.09523809523809525, "generation_token_f1": 0.09523809523809525, "prompt_id": "0f4edb0d2ff1676b", "reading_canonical_log_prob": -34.750465393066406, "reading_distances": [14, 13, 14, 17], "reading_log_probs": [-38.72965621948242, -37.665374755859375, -39.437034606933594, -38.9106559753418], "reading_logprob_std": 0.7434389986052539, "tokenization_path_logprob_std": 105.07884813227811, "trajectory_balance_rmse": 348.66691161798343} +{"canonical_length": 80, "canonical_nll": 1.107280158996582, "generation_rouge_l_f1": 0.17964071856287425, "generation_token_f1": 0.2754491017964072, "prompt_id": "7bc8145a69124b53", "reading_canonical_log_prob": -88.58241271972656, "reading_distances": [25, 28, 26, 23], "reading_log_probs": [-85.6324462890625, -87.25621032714844, -86.30184936523438, -85.92970275878906], "reading_logprob_std": 0.7060448409858504, "tokenization_path_logprob_std": 64.47790355679012, "trajectory_balance_rmse": 1410.1203603095043} +{"canonical_length": 12, "canonical_nll": 0.5282876491546631, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "5167b2d9f21714f8", "reading_canonical_log_prob": -6.339451789855957, "reading_distances": [27, 20, 22, 22], "reading_log_probs": [-5.9573235511779785, -6.087973594665527, -6.029306411743164, -6.159202575683594], "reading_logprob_std": 0.08582663053560417, "tokenization_path_logprob_std": 25.028970669267384, "trajectory_balance_rmse": 260.07137393977285} +{"canonical_length": 130, "canonical_nll": 1.156507051908053, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.28828828828828834, "prompt_id": "786583b4f8ed0931", "reading_canonical_log_prob": -150.34591674804688, "reading_distances": [23, 23, 21, 20], "reading_log_probs": [-153.83795166015625, -151.06341552734375, -154.66506958007812, -152.97238159179688], "reading_logprob_std": 1.544145997130801, "tokenization_path_logprob_std": 162.6495692349498, "trajectory_balance_rmse": 3523.608457196703} +{"canonical_length": 351, "canonical_nll": 1.1590342290720708, "generation_rouge_l_f1": 0.19765494137353432, "generation_token_f1": 0.37855946398659973, "prompt_id": "99a05025a306c903", "reading_canonical_log_prob": -406.8210144042969, "reading_distances": [18, 15, 13, 19], "reading_log_probs": [-410.0880126953125, -410.3284912109375, -410.4390869140625, -409.27459716796875], "reading_logprob_std": 0.5261248486009329, "tokenization_path_logprob_std": 329.8125033198481, "trajectory_balance_rmse": 6476.7863018323} +{"canonical_length": 32, "canonical_nll": 1.3687644004821777, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "4d4e1a944b4c608a", "reading_canonical_log_prob": -43.80046081542969, "reading_distances": [12, 11, 14, 14], "reading_log_probs": [-42.45423889160156, -42.6044807434082, -42.766265869140625, -41.900333404541016], "reading_logprob_std": 0.3762292820798447, "tokenization_path_logprob_std": 23.85937743867688, "trajectory_balance_rmse": 647.262915086521} +{"canonical_length": 375, "canonical_nll": 1.2885912272135416, "generation_rouge_l_f1": 0.2809667673716012, "generation_token_f1": 0.5045317220543807, "prompt_id": "8de18ae965c9405f", "reading_canonical_log_prob": -483.2217102050781, "reading_distances": [29, 25, 28, 29], "reading_log_probs": [-481.23455810546875, -483.8643798828125, -483.22174072265625, -484.5613708496094], "reading_logprob_std": 1.4325365249626067, "tokenization_path_logprob_std": 396.950050028144, "trajectory_balance_rmse": 7309.691974001254} +{"canonical_length": 154, "canonical_nll": 1.0670441466492493, "generation_rouge_l_f1": 0.35351089588377715, "generation_token_f1": 0.5181598062953996, "prompt_id": "577845af12167b4c", "reading_canonical_log_prob": -164.32479858398438, "reading_distances": [9, 11, 10, 8], "reading_log_probs": [-164.66175842285156, -163.14907836914062, -164.93360900878906, -162.92984008789062], "reading_logprob_std": 1.025075189157335, "tokenization_path_logprob_std": 41.22540387478605, "trajectory_balance_rmse": 2904.3929110217578} +{"canonical_length": 94, "canonical_nll": 1.0623639695187832, "generation_rouge_l_f1": 0.1948424068767908, "generation_token_f1": 0.3782234957020057, "prompt_id": "aea07a40db5e5411", "reading_canonical_log_prob": -99.86221313476562, "reading_distances": [18, 24, 30, 29], "reading_log_probs": [-103.00836944580078, -104.3733901977539, -102.67144775390625, -104.80596160888672], "reading_logprob_std": 1.0347306803190737, "tokenization_path_logprob_std": 109.59164374156003, "trajectory_balance_rmse": 2127.653086535697} +{"canonical_length": 111, "canonical_nll": 1.4447561727987754, "generation_rouge_l_f1": 0.15053763440860216, "generation_token_f1": 0.3602150537634409, "prompt_id": "fdb4e9b8596975db", "reading_canonical_log_prob": -160.36793518066406, "reading_distances": [24, 23, 23, 20], "reading_log_probs": [-159.60711669921875, -158.97523498535156, -160.78990173339844, -158.77537536621094], "reading_logprob_std": 0.907437984123273, "tokenization_path_logprob_std": 120.96343971743632, "trajectory_balance_rmse": 2674.5478876431544} +{"canonical_length": 364, "canonical_nll": 1.6092707036615728, "generation_rouge_l_f1": 0.24333333333333332, "generation_token_f1": 0.48, "prompt_id": "219a3ab27e90cb77", "reading_canonical_log_prob": -585.7745361328125, "reading_distances": [13, 15, 18, 13], "reading_log_probs": [-586.7681884765625, -589.4608764648438, -588.0596313476562, -588.2838134765625], "reading_logprob_std": 1.1035826209782365, "tokenization_path_logprob_std": 394.5063039104158, "trajectory_balance_rmse": 7914.682548953144} +{"canonical_length": 42, "canonical_nll": 1.321574438185919, "generation_rouge_l_f1": 0.14760147601476012, "generation_token_f1": 0.18450184501845016, "prompt_id": "c1bdd5a5ad1d792e", "reading_canonical_log_prob": -55.506126403808594, "reading_distances": [14, 21, 19, 18], "reading_log_probs": [-54.658935546875, -54.84538650512695, -53.4316291809082, -54.61687088012695], "reading_logprob_std": 0.6454030627046541, "tokenization_path_logprob_std": 21.347438915649455, "trajectory_balance_rmse": 575.2795437531323} +{"canonical_length": 262, "canonical_nll": 1.6723588550363788, "generation_rouge_l_f1": 0.2730627306273063, "generation_token_f1": 0.4723247232472325, "prompt_id": "52d1f167e2d1bb05", "reading_canonical_log_prob": -438.15802001953125, "reading_distances": [30, 24, 24, 28], "reading_log_probs": [-442.20806884765625, -441.88165283203125, -442.49188232421875, -443.3760070800781], "reading_logprob_std": 0.6415039916219203, "tokenization_path_logprob_std": 249.9685905117994, "trajectory_balance_rmse": 5885.401571688601} +{"canonical_length": 17, "canonical_nll": 1.7632365507238053, "generation_rouge_l_f1": 0.4615384615384615, "generation_token_f1": 0.4615384615384615, "prompt_id": "c86b63feb26be380", "reading_canonical_log_prob": -29.975021362304688, "reading_distances": [37, 37, 36, 36], "reading_log_probs": [-28.162572860717773, -28.226442337036133, -27.962373733520508, -28.005691528320312], "reading_logprob_std": 0.12553562948695418, "tokenization_path_logprob_std": 47.3958704688159, "trajectory_balance_rmse": 422.95698472721847} +{"canonical_length": 10, "canonical_nll": 1.4963375091552735, "generation_rouge_l_f1": 0.11111111111111112, "generation_token_f1": 0.11111111111111112, "prompt_id": "197710f492b62fe4", "reading_canonical_log_prob": -14.963375091552734, "reading_distances": [14, 7, 9, 16], "reading_log_probs": [-14.73227596282959, -14.936365127563477, -15.503963470458984, -14.476072311401367], "reading_logprob_std": 0.4371681501374481, "tokenization_path_logprob_std": 51.42633984984154, "trajectory_balance_rmse": 236.28371251404644} +{"canonical_length": 25, "canonical_nll": 1.0982602691650392, "generation_rouge_l_f1": 0.23728813559322032, "generation_token_f1": 0.2711864406779661, "prompt_id": "23c24264db31cb91", "reading_canonical_log_prob": -27.456506729125977, "reading_distances": [31, 23, 27, 30], "reading_log_probs": [-27.96982192993164, -29.75811004638672, -28.878019332885742, -29.790393829345703], "reading_logprob_std": 0.863390027717607, "tokenization_path_logprob_std": 87.94328007066248, "trajectory_balance_rmse": 466.3260668805396} +{"canonical_length": 169, "canonical_nll": 0.8549508540588018, "generation_rouge_l_f1": 0.32824427480916035, "generation_token_f1": 0.37404580152671757, "prompt_id": "742a95dae13136e2", "reading_canonical_log_prob": -144.4866943359375, "reading_distances": [19, 18, 18, 16], "reading_log_probs": [-143.5044403076172, -144.16806030273438, -143.31600952148438, -144.27853393554688], "reading_logprob_std": 0.47782203877988544, "tokenization_path_logprob_std": 131.69899481702905, "trajectory_balance_rmse": 2786.372494213491} +{"canonical_length": 19, "canonical_nll": 2.17920604505037, "generation_rouge_l_f1": 0.4444444444444445, "generation_token_f1": 0.5, "prompt_id": "87a7422c0b830ad1", "reading_canonical_log_prob": -41.40491485595703, "reading_distances": [48, 51, 43, 43], "reading_log_probs": [-44.746299743652344, -44.58367156982422, -43.96589660644531, -43.41978454589844], "reading_logprob_std": 0.6075659663685847, "tokenization_path_logprob_std": 56.88383738356834, "trajectory_balance_rmse": 413.1569018212324} +{"canonical_length": 44, "canonical_nll": 0.9685786854137074, "generation_rouge_l_f1": 0.5945945945945946, "generation_token_f1": 0.6216216216216217, "prompt_id": "73b66ff67ed60622", "reading_canonical_log_prob": -42.617462158203125, "reading_distances": [29, 26, 25, 22], "reading_log_probs": [-44.261199951171875, -44.11248016357422, -46.18000411987305, -43.610504150390625], "reading_logprob_std": 1.1275451712700864, "tokenization_path_logprob_std": 37.47969636548652, "trajectory_balance_rmse": 1201.2342314360958} +{"canonical_length": 126, "canonical_nll": 1.1997172037760417, "generation_rouge_l_f1": 0.3879310344827586, "generation_token_f1": 0.5775862068965518, "prompt_id": "f610e308eb24edb1", "reading_canonical_log_prob": -151.16436767578125, "reading_distances": [23, 21, 24, 25], "reading_log_probs": [-149.36199951171875, -150.1986083984375, -148.83497619628906, -149.0705108642578], "reading_logprob_std": 0.5951335315929478, "tokenization_path_logprob_std": 121.72459122806617, "trajectory_balance_rmse": 2156.9886046629226} +{"canonical_length": 251, "canonical_nll": 1.237319034409238, "generation_rouge_l_f1": 0.2693069306930693, "generation_token_f1": 0.5069306930693068, "prompt_id": "da58850d3c7fdcfd", "reading_canonical_log_prob": -310.56707763671875, "reading_distances": [18, 23, 25, 22], "reading_log_probs": [-307.40045166015625, -308.65771484375, -309.34759521484375, -307.929931640625], "reading_logprob_std": 0.8498922045342264, "tokenization_path_logprob_std": 139.45734604090072, "trajectory_balance_rmse": 5503.815367506316} +{"canonical_length": 296, "canonical_nll": 1.303550926414696, "generation_rouge_l_f1": 0.20146520146520147, "generation_token_f1": 0.4065934065934066, "prompt_id": "0a2b0e3667ca975a", "reading_canonical_log_prob": -385.85107421875, "reading_distances": [12, 16, 15, 19], "reading_log_probs": [-390.2305908203125, -390.28338623046875, -391.35406494140625, -391.14923095703125], "reading_logprob_std": 0.5807235508841104, "tokenization_path_logprob_std": 235.66881086536856, "trajectory_balance_rmse": 6196.097012284474} +{"canonical_length": 111, "canonical_nll": 1.0410308837890625, "generation_rouge_l_f1": 0.21538461538461537, "generation_token_f1": 0.40615384615384614, "prompt_id": "b3bdd02a596bc2d5", "reading_canonical_log_prob": -115.55442810058594, "reading_distances": [39, 36, 44, 46], "reading_log_probs": [-117.46170043945312, -115.10661315917969, -116.46025848388672, -116.09030151367188], "reading_logprob_std": 0.9732644255880853, "tokenization_path_logprob_std": 77.92550990033385, "trajectory_balance_rmse": 1554.0958430617297} +{"canonical_length": 14, "canonical_nll": 0.033472148435456414, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "3a6cc5e39c6939e7", "reading_canonical_log_prob": -0.46861007809638977, "reading_distances": [22, 27, 23, 27], "reading_log_probs": [-0.9705091714859009, -1.0009689331054688, -0.8808095455169678, -0.8405895233154297], "reading_logprob_std": 0.07507241984342158, "tokenization_path_logprob_std": 55.443814857248185, "trajectory_balance_rmse": 300.4023849952231} +{"canonical_length": 79, "canonical_nll": 0.7930162164229381, "generation_rouge_l_f1": 0.4773662551440329, "generation_token_f1": 0.6008230452674898, "prompt_id": "9431e7948140e38d", "reading_canonical_log_prob": -62.64828109741211, "reading_distances": [36, 34, 37, 37], "reading_log_probs": [-59.40191650390625, -58.83463668823242, -57.94282913208008, -58.95033645629883], "reading_logprob_std": 0.6109040465562848, "tokenization_path_logprob_std": 149.10917751131603, "trajectory_balance_rmse": 1423.0734866628577} +{"canonical_length": 56, "canonical_nll": 1.0668792043413435, "generation_rouge_l_f1": 0.18012422360248445, "generation_token_f1": 0.24223602484472048, "prompt_id": "8230c7c1426f816d", "reading_canonical_log_prob": -59.745235443115234, "reading_distances": [24, 23, 25, 21], "reading_log_probs": [-59.590553283691406, -57.560096740722656, -58.763877868652344, -57.96745300292969], "reading_logprob_std": 0.8986063854279993, "tokenization_path_logprob_std": 109.01439544459018, "trajectory_balance_rmse": 1321.8394415315854} +{"canonical_length": 101, "canonical_nll": 1.3250656883315284, "generation_rouge_l_f1": 0.09248554913294797, "generation_token_f1": 0.16184971098265896, "prompt_id": "211cd27653ee878e", "reading_canonical_log_prob": -133.83163452148438, "reading_distances": [27, 24, 31, 27], "reading_log_probs": [-136.9608917236328, -136.20237731933594, -138.09674072265625, -137.13243103027344], "reading_logprob_std": 0.778804028394526, "tokenization_path_logprob_std": 78.10662114014316, "trajectory_balance_rmse": 2152.362432341089} +{"canonical_length": 19, "canonical_nll": 2.0880801552220394, "generation_rouge_l_f1": 0.08333333333333334, "generation_token_f1": 0.10606060606060608, "prompt_id": "bd8cff3238489dd7", "reading_canonical_log_prob": -39.67352294921875, "reading_distances": [31, 29, 37, 36], "reading_log_probs": [-40.93223571777344, -42.66557312011719, -39.58805465698242, -39.75044250488281], "reading_logprob_std": 1.4201934340241948, "tokenization_path_logprob_std": 32.394311598896884, "trajectory_balance_rmse": 490.46400083927807} +{"canonical_length": 21, "canonical_nll": 1.3273735954647972, "generation_rouge_l_f1": 0.18666666666666665, "generation_token_f1": 0.21333333333333335, "prompt_id": "9444b92ac8847c28", "reading_canonical_log_prob": -27.874845504760742, "reading_distances": [28, 21, 24, 23], "reading_log_probs": [-29.287193298339844, -28.566051483154297, -28.063325881958008, -28.590717315673828], "reading_logprob_std": 0.5028633779558261, "tokenization_path_logprob_std": 60.8134947782847, "trajectory_balance_rmse": 551.7322163956769} +{"canonical_length": 13, "canonical_nll": 2.134942568265475, "generation_rouge_l_f1": 0.3448275862068966, "generation_token_f1": 0.3448275862068966, "prompt_id": "d790284919634e0e", "reading_canonical_log_prob": -27.754253387451172, "reading_distances": [32, 30, 32, 24], "reading_log_probs": [-26.02541732788086, -28.14845848083496, -26.958955764770508, -23.27528953552246], "reading_logprob_std": 2.07513232096523, "tokenization_path_logprob_std": 25.176058141876208, "trajectory_balance_rmse": 407.65012688642645} +{"canonical_length": 86, "canonical_nll": 1.1927393536235011, "generation_rouge_l_f1": 0.10869565217391304, "generation_token_f1": 0.10869565217391304, "prompt_id": "58457960c61ffd91", "reading_canonical_log_prob": -102.5755844116211, "reading_distances": [32, 30, 29, 29], "reading_log_probs": [-103.973388671875, -103.66355895996094, -104.11907958984375, -103.6941146850586], "reading_logprob_std": 0.22065095372552268, "tokenization_path_logprob_std": 112.57019496634666, "trajectory_balance_rmse": 1963.816877664127} +{"canonical_length": 6, "canonical_nll": 1.5192578633626301, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "02cbb58a2f84ec5b", "reading_canonical_log_prob": -9.115547180175781, "reading_distances": [19, 16, 17, 20], "reading_log_probs": [-7.1798200607299805, -6.724434852600098, -7.619480609893799, -7.532752990722656], "reading_logprob_std": 0.4069444617453152, "tokenization_path_logprob_std": 15.558036967391418, "trajectory_balance_rmse": 170.7461456185328} +{"canonical_length": 50, "canonical_nll": 1.8557330322265626, "generation_rouge_l_f1": 0.24148606811145512, "generation_token_f1": 0.3095975232198142, "prompt_id": "d0f9c73543784d9e", "reading_canonical_log_prob": -92.78665161132812, "reading_distances": [40, 32, 29, 40], "reading_log_probs": [-95.49420928955078, -95.13783264160156, -96.05621337890625, -95.82478332519531], "reading_logprob_std": 0.40010586045039087, "tokenization_path_logprob_std": 44.62213644501208, "trajectory_balance_rmse": 522.5409014227915} +{"canonical_length": 76, "canonical_nll": 1.539385343852796, "generation_rouge_l_f1": 0.14634146341463414, "generation_token_f1": 0.25609756097560976, "prompt_id": "e67d8b619feb7d26", "reading_canonical_log_prob": -116.9932861328125, "reading_distances": [6, 10, 9, 8], "reading_log_probs": [-118.27266693115234, -117.93264770507812, -118.82149505615234, -117.98667907714844], "reading_logprob_std": 0.40707300317926104, "tokenization_path_logprob_std": 73.36470614701993, "trajectory_balance_rmse": 2020.4002820645799} +{"canonical_length": 260, "canonical_nll": 1.2217427180363583, "generation_rouge_l_f1": 0.2669039145907473, "generation_token_f1": 0.40569395017793597, "prompt_id": "aa422d282fda03cd", "reading_canonical_log_prob": -317.6531066894531, "reading_distances": [28, 30, 37, 23], "reading_log_probs": [-318.2695617675781, -319.9256286621094, -320.2231140136719, -320.4269104003906], "reading_logprob_std": 0.9829543454829778, "tokenization_path_logprob_std": 215.90418016901324, "trajectory_balance_rmse": 3926.442827139828} +{"canonical_length": 281, "canonical_nll": 1.3810505171263345, "generation_rouge_l_f1": 0.20502901353965186, "generation_token_f1": 0.3597678916827853, "prompt_id": "050b55490a11894a", "reading_canonical_log_prob": -388.0751953125, "reading_distances": [19, 17, 26, 18], "reading_log_probs": [-391.290771484375, -394.4033203125, -389.7791748046875, -392.12030029296875], "reading_logprob_std": 1.9308196319325692, "tokenization_path_logprob_std": 163.2804367531901, "trajectory_balance_rmse": 6245.360982410098} +{"canonical_length": 251, "canonical_nll": 0.8272713589003362, "generation_rouge_l_f1": 0.27385892116182575, "generation_token_f1": 0.42738589211618255, "prompt_id": "2d149cddaf627a2b", "reading_canonical_log_prob": -207.64511108398438, "reading_distances": [26, 29, 33, 35], "reading_log_probs": [-205.361572265625, -206.32943725585938, -206.89071655273438, -206.14028930664062], "reading_logprob_std": 0.6321522474559141, "tokenization_path_logprob_std": 170.0037899368566, "trajectory_balance_rmse": 5535.060025304954} +{"canonical_length": 157, "canonical_nll": 1.635985939366043, "generation_rouge_l_f1": 0.2725060827250608, "generation_token_f1": 0.4720194647201946, "prompt_id": "f7af4006bdd897fe", "reading_canonical_log_prob": -256.84979248046875, "reading_distances": [29, 22, 22, 32], "reading_log_probs": [-257.834228515625, -257.0542297363281, -257.3716735839844, -258.5814514160156], "reading_logprob_std": 0.6631638988939311, "tokenization_path_logprob_std": 62.537915253942785, "trajectory_balance_rmse": 3714.476235288427} +{"canonical_length": 21, "canonical_nll": 0.432591302054269, "generation_rouge_l_f1": 0.8125000000000001, "generation_token_f1": 0.8125000000000001, "prompt_id": "2d0d3b6d72c67323", "reading_canonical_log_prob": -9.084417343139648, "reading_distances": [16, 31, 24, 23], "reading_log_probs": [-10.58298397064209, -9.63163948059082, -9.897268295288086, -9.149059295654297], "reading_logprob_std": 0.5982377808293702, "tokenization_path_logprob_std": 33.551523536785545, "trajectory_balance_rmse": 400.954809579846} +{"canonical_length": 7, "canonical_nll": 0.6965698514665876, "generation_rouge_l_f1": 0.06557377049180328, "generation_token_f1": 0.06557377049180328, "prompt_id": "974b052b2f839275", "reading_canonical_log_prob": -4.875988960266113, "reading_distances": [15, 25, 19, 23], "reading_log_probs": [-3.6853322982788086, -4.513089179992676, -4.311213970184326, -4.461663722991943], "reading_logprob_std": 0.3814045434893473, "tokenization_path_logprob_std": 32.39515886115327, "trajectory_balance_rmse": 106.66072143530552} +{"canonical_length": 71, "canonical_nll": 1.255708721322073, "generation_rouge_l_f1": 0.028571428571428574, "generation_token_f1": 0.028571428571428574, "prompt_id": "7f8aa3c30a8d7e14", "reading_canonical_log_prob": -89.15531921386719, "reading_distances": [32, 30, 32, 37], "reading_log_probs": [-91.31343078613281, -90.89955139160156, -90.09544372558594, -93.01698303222656], "reading_logprob_std": 1.2322870573165772, "tokenization_path_logprob_std": 68.52145648365635, "trajectory_balance_rmse": 1559.7827953452215} +{"canonical_length": 12, "canonical_nll": 0.417834997177124, "generation_rouge_l_f1": 0.75, "generation_token_f1": 0.75, "prompt_id": "537e5de8a8d8abfa", "reading_canonical_log_prob": -5.014019966125488, "reading_distances": [57, 49, 64, 54], "reading_log_probs": [-7.624604225158691, -7.13963508605957, -5.897461414337158, -8.744928359985352], "reading_logprob_std": 1.1797377340789525, "tokenization_path_logprob_std": 30.431708874929765, "trajectory_balance_rmse": 294.38904548431367} +{"canonical_length": 235, "canonical_nll": 1.4769432554853723, "generation_rouge_l_f1": 0.24145785876993167, "generation_token_f1": 0.39635535307517084, "prompt_id": "219d66b292bdf703", "reading_canonical_log_prob": -347.0816650390625, "reading_distances": [38, 33, 45, 36], "reading_log_probs": [-351.1966552734375, -352.5240478515625, -355.5426025390625, -352.20343017578125], "reading_logprob_std": 1.8714380514103073, "tokenization_path_logprob_std": 268.76446770885565, "trajectory_balance_rmse": 5398.1875504327245} +{"canonical_length": 233, "canonical_nll": 1.5180110030931464, "generation_rouge_l_f1": 0.26424870466321243, "generation_token_f1": 0.4300518134715026, "prompt_id": "b3b45f7b904e5cc3", "reading_canonical_log_prob": -353.6965637207031, "reading_distances": [26, 14, 21, 19], "reading_log_probs": [-351.4208984375, -352.7347412109375, -349.3634338378906, -352.107421875], "reading_logprob_std": 1.4639948526434208, "tokenization_path_logprob_std": 113.00080825156516, "trajectory_balance_rmse": 5467.04713881896} +{"canonical_length": 121, "canonical_nll": 1.3679841096736183, "generation_rouge_l_f1": 0.33766233766233766, "generation_token_f1": 0.5367965367965368, "prompt_id": "dd6de628b051978a", "reading_canonical_log_prob": -165.5260772705078, "reading_distances": [38, 39, 33, 35], "reading_log_probs": [-162.61404418945312, -162.822265625, -163.6858673095703, -163.09486389160156], "reading_logprob_std": 0.4648253505001152, "tokenization_path_logprob_std": 79.01066643733681, "trajectory_balance_rmse": 2573.8035064637966} +{"canonical_length": 105, "canonical_nll": 0.8571244012741815, "generation_rouge_l_f1": 0.23566878980891717, "generation_token_f1": 0.35668789808917195, "prompt_id": "d2e124ed83b7b0d6", "reading_canonical_log_prob": -89.99806213378906, "reading_distances": [29, 23, 31, 31], "reading_log_probs": [-90.17735290527344, -91.3821792602539, -90.63249969482422, -91.06289672851562], "reading_logprob_std": 0.5237807589283507, "tokenization_path_logprob_std": 99.67098013195641, "trajectory_balance_rmse": 1802.6525308784098} +{"canonical_length": 102, "canonical_nll": 0.9317657620299096, "generation_rouge_l_f1": 0.20195439739413681, "generation_token_f1": 0.3061889250814332, "prompt_id": "40e867680ebfa198", "reading_canonical_log_prob": -95.04010772705078, "reading_distances": [36, 38, 40, 29], "reading_log_probs": [-98.87242889404297, -104.9383544921875, -103.62065887451172, -102.74546813964844], "reading_logprob_std": 2.6085213791538444, "tokenization_path_logprob_std": 79.11571776140954, "trajectory_balance_rmse": 1555.7211030991134} +{"canonical_length": 217, "canonical_nll": 1.3773131480414746, "generation_rouge_l_f1": 0.19834710743801653, "generation_token_f1": 0.39669421487603307, "prompt_id": "7763d96632870ad4", "reading_canonical_log_prob": -298.876953125, "reading_distances": [13, 12, 12, 12], "reading_log_probs": [-294.8227844238281, -294.40936279296875, -294.9423522949219, -294.96209716796875], "reading_logprob_std": 0.2573267896163078, "tokenization_path_logprob_std": 106.15835904612985, "trajectory_balance_rmse": 4938.719617166866} +{"canonical_length": 11, "canonical_nll": 1.4402474490079014, "generation_rouge_l_f1": 0.039840637450199196, "generation_token_f1": 0.04780876494023904, "prompt_id": "151a13661f1584a8", "reading_canonical_log_prob": -15.842721939086914, "reading_distances": [15, 12, 13, 15], "reading_log_probs": [-13.991952896118164, -13.318626403808594, -14.581073760986328, -13.707432746887207], "reading_logprob_std": 0.5314740506025577, "tokenization_path_logprob_std": 31.93498398379504, "trajectory_balance_rmse": 229.08374972486078} +{"canonical_length": 166, "canonical_nll": 0.8392213568629989, "generation_rouge_l_f1": 0.3190476190476191, "generation_token_f1": 0.48095238095238085, "prompt_id": "f8789af8051fda52", "reading_canonical_log_prob": -139.3107452392578, "reading_distances": [19, 21, 25, 28], "reading_log_probs": [-142.01666259765625, -140.14341735839844, -140.18405151367188, -139.57070922851562], "reading_logprob_std": 1.0628590958245658, "tokenization_path_logprob_std": 188.25091161355198, "trajectory_balance_rmse": 3710.4557803809607} +{"canonical_length": 199, "canonical_nll": 0.6773711544784469, "generation_rouge_l_f1": 0.17030567685589518, "generation_token_f1": 0.31441048034934493, "prompt_id": "b3a3308a3d93f328", "reading_canonical_log_prob": -134.79685974121094, "reading_distances": [7, 10, 10, 6], "reading_log_probs": [-133.74029541015625, -133.87612915039062, -133.9619140625, -134.5623321533203], "reading_logprob_std": 0.3630938280165038, "tokenization_path_logprob_std": 71.38801052763378, "trajectory_balance_rmse": 3886.234218642251} +{"canonical_length": 33, "canonical_nll": 0.5832396998549952, "generation_rouge_l_f1": 0.11074918566775244, "generation_token_f1": 0.1498371335504886, "prompt_id": "06bad79ff1fc809f", "reading_canonical_log_prob": -19.246910095214844, "reading_distances": [21, 25, 19, 26], "reading_log_probs": [-20.51349639892578, -20.565509796142578, -23.695396423339844, -18.865249633789062], "reading_logprob_std": 2.0178640416093923, "tokenization_path_logprob_std": 48.33567312641554, "trajectory_balance_rmse": 584.9165377026343} +{"canonical_length": 153, "canonical_nll": 1.6186085619957618, "generation_rouge_l_f1": 0.2048780487804878, "generation_token_f1": 0.4487804878048781, "prompt_id": "df98426df5120860", "reading_canonical_log_prob": -247.64710998535156, "reading_distances": [21, 18, 21, 20], "reading_log_probs": [-243.5039520263672, -246.44528198242188, -245.62045288085938, -246.14892578125], "reading_logprob_std": 1.3283554240950983, "tokenization_path_logprob_std": 233.49991158790525, "trajectory_balance_rmse": 3431.1126914446886} +{"canonical_length": 25, "canonical_nll": 1.081060256958008, "generation_rouge_l_f1": 0.10628019323671498, "generation_token_f1": 0.12560386473429952, "prompt_id": "e01eb25b0bbbab91", "reading_canonical_log_prob": -27.026506423950195, "reading_distances": [28, 30, 28, 22], "reading_log_probs": [-25.687538146972656, -26.71664047241211, -26.869487762451172, -26.49335479736328], "reading_logprob_std": 0.5259983965423574, "tokenization_path_logprob_std": 35.94083501981493, "trajectory_balance_rmse": 229.42749571728143} +{"canonical_length": 42, "canonical_nll": 2.0699083237420943, "generation_rouge_l_f1": 0.3181818181818182, "generation_token_f1": 0.42424242424242425, "prompt_id": "050fda8d8b928503", "reading_canonical_log_prob": -86.93614959716797, "reading_distances": [29, 30, 28, 33], "reading_log_probs": [-89.8384017944336, -88.30624389648438, -88.87542724609375, -88.84840393066406], "reading_logprob_std": 0.6372833633856692, "tokenization_path_logprob_std": 67.79665499048755, "trajectory_balance_rmse": 1084.5263027337007} +{"canonical_length": 191, "canonical_nll": 1.1408895921956805, "generation_rouge_l_f1": 0.176, "generation_token_f1": 0.33599999999999997, "prompt_id": "c86e9e8e522d0430", "reading_canonical_log_prob": -217.909912109375, "reading_distances": [30, 21, 19, 24], "reading_log_probs": [-217.869140625, -218.081787109375, -217.72674560546875, -222.57183837890625], "reading_logprob_std": 2.3441842911275614, "tokenization_path_logprob_std": 198.15559461099525, "trajectory_balance_rmse": 3466.4894751791} +{"canonical_length": 71, "canonical_nll": 1.1345868177816902, "generation_rouge_l_f1": 0.22641509433962265, "generation_token_f1": 0.3773584905660377, "prompt_id": "0d01c13272391fbb", "reading_canonical_log_prob": -80.5556640625, "reading_distances": [27, 26, 27, 32], "reading_log_probs": [-77.15011596679688, -77.93998718261719, -77.90447998046875, -77.36407470703125], "reading_logprob_std": 0.394093236189766, "tokenization_path_logprob_std": 25.812898894676913, "trajectory_balance_rmse": 1541.0551551353733} +{"canonical_length": 96, "canonical_nll": 1.4012977282206218, "generation_rouge_l_f1": 0.21671826625386997, "generation_token_f1": 0.35913312693498445, "prompt_id": "f4314cbcdd2435c8", "reading_canonical_log_prob": -134.5245819091797, "reading_distances": [23, 30, 30, 30], "reading_log_probs": [-133.57403564453125, -131.3857421875, -132.93345642089844, -137.2510223388672], "reading_logprob_std": 2.485918797353919, "tokenization_path_logprob_std": 103.91657090453518, "trajectory_balance_rmse": 2224.959988494407} +{"canonical_length": 28, "canonical_nll": 1.1870825631277901, "generation_rouge_l_f1": 0.10869565217391303, "generation_token_f1": 0.13043478260869565, "prompt_id": "6039b41734030698", "reading_canonical_log_prob": -33.238311767578125, "reading_distances": [13, 18, 17, 15], "reading_log_probs": [-35.04169464111328, -37.94511413574219, -37.25874710083008, -36.55667495727539], "reading_logprob_std": 1.2427154435267012, "tokenization_path_logprob_std": 55.11234650102839, "trajectory_balance_rmse": 337.81977610252045} +{"canonical_length": 180, "canonical_nll": 1.124316660563151, "generation_rouge_l_f1": 0.2996941896024465, "generation_token_f1": 0.5137614678899082, "prompt_id": "bc991d771f2ffcd2", "reading_canonical_log_prob": -202.3769989013672, "reading_distances": [28, 34, 27, 26], "reading_log_probs": [-200.06536865234375, -201.464599609375, -201.20285034179688, -201.8300018310547], "reading_logprob_std": 0.7616316133542744, "tokenization_path_logprob_std": 89.69698338220029, "trajectory_balance_rmse": 3409.673784488424} +{"canonical_length": 322, "canonical_nll": 0.880273759735297, "generation_rouge_l_f1": 0.31578947368421056, "generation_token_f1": 0.611842105263158, "prompt_id": "ee15b1eaf2aa94fb", "reading_canonical_log_prob": -283.4481506347656, "reading_distances": [31, 29, 29, 32], "reading_log_probs": [-287.76226806640625, -287.197265625, -287.8251037597656, -288.0432434082031], "reading_logprob_std": 0.360505116146375, "tokenization_path_logprob_std": 157.38845052371747, "trajectory_balance_rmse": 6148.433324134196} +{"canonical_length": 42, "canonical_nll": 1.5982976640973772, "generation_rouge_l_f1": 0.1, "generation_token_f1": 0.17142857142857143, "prompt_id": "e36603aac9ffa742", "reading_canonical_log_prob": -67.12850189208984, "reading_distances": [24, 27, 27, 29], "reading_log_probs": [-66.4591064453125, -65.64511108398438, -66.27494812011719, -67.259521484375], "reading_logprob_std": 0.6651796320255531, "tokenization_path_logprob_std": 123.170943664652, "trajectory_balance_rmse": 1024.2784299286932} +{"canonical_length": 248, "canonical_nll": 1.3969812700825353, "generation_rouge_l_f1": 0.21965317919075145, "generation_token_f1": 0.45472061657032753, "prompt_id": "46c0bc7a74b4f8f8", "reading_canonical_log_prob": -346.45135498046875, "reading_distances": [19, 11, 10, 18], "reading_log_probs": [-345.023193359375, -343.68231201171875, -343.92633056640625, -343.96234130859375], "reading_logprob_std": 0.596219764043386, "tokenization_path_logprob_std": 197.05506222406757, "trajectory_balance_rmse": 5140.675955704219} +{"canonical_length": 193, "canonical_nll": 1.5301583245628239, "generation_rouge_l_f1": 0.2587719298245614, "generation_token_f1": 0.4692982456140351, "prompt_id": "17af0bbd6fc62e57", "reading_canonical_log_prob": -295.320556640625, "reading_distances": [22, 30, 30, 29], "reading_log_probs": [-291.28082275390625, -292.294677734375, -292.1510314941406, -291.635009765625], "reading_logprob_std": 0.4683915836758386, "tokenization_path_logprob_std": 48.51730802030921, "trajectory_balance_rmse": 3545.3823584585884} +{"canonical_length": 27, "canonical_nll": 1.448801111291956, "generation_rouge_l_f1": 0.18823529411764708, "generation_token_f1": 0.2352941176470588, "prompt_id": "917c56964f6f4a40", "reading_canonical_log_prob": -39.11763000488281, "reading_distances": [36, 35, 28, 33], "reading_log_probs": [-40.742698669433594, -40.938018798828125, -40.083465576171875, -40.445343017578125], "reading_logprob_std": 0.37250330270009185, "tokenization_path_logprob_std": 52.54809420625547, "trajectory_balance_rmse": 522.0074315634814} +{"canonical_length": 350, "canonical_nll": 1.383740495954241, "generation_rouge_l_f1": 0.21070811744386875, "generation_token_f1": 0.39723661485319517, "prompt_id": "74f8c2e7c18cb610", "reading_canonical_log_prob": -484.3091735839844, "reading_distances": [43, 41, 32, 39], "reading_log_probs": [-486.0573425292969, -485.967041015625, -487.9342041015625, -487.6408386230469], "reading_logprob_std": 1.0326186753835842, "tokenization_path_logprob_std": 171.37375410791287, "trajectory_balance_rmse": 6310.541915643695} +{"canonical_length": 26, "canonical_nll": 0.9956145653357873, "generation_rouge_l_f1": 0.07692307692307691, "generation_token_f1": 0.07692307692307691, "prompt_id": "c533f2b5441efb8c", "reading_canonical_log_prob": -25.88597869873047, "reading_distances": [25, 25, 23, 27], "reading_log_probs": [-25.45258331298828, -26.94130516052246, -26.214874267578125, -25.25623321533203], "reading_logprob_std": 0.7703946807409799, "tokenization_path_logprob_std": 44.28287958674514, "trajectory_balance_rmse": 265.7844086476906} +{"canonical_length": 91, "canonical_nll": 1.0690978333190246, "generation_rouge_l_f1": 0.22222222222222224, "generation_token_f1": 0.45614035087719296, "prompt_id": "ce28dbb2751f3b46", "reading_canonical_log_prob": -97.28790283203125, "reading_distances": [21, 16, 15, 17], "reading_log_probs": [-97.11412048339844, -97.88862609863281, -96.59442138671875, -98.28206634521484], "reading_logprob_std": 0.7589439046663079, "tokenization_path_logprob_std": 30.81383252978223, "trajectory_balance_rmse": 2164.1614555641045} +{"canonical_length": 13, "canonical_nll": 1.0089403299184947, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.8888888888888888, "prompt_id": "a24d7fb2090fb702", "reading_canonical_log_prob": -13.11622428894043, "reading_distances": [25, 25, 28, 31], "reading_log_probs": [-12.42236328125, -12.998380661010742, -12.597488403320312, -12.679780960083008], "reading_logprob_std": 0.24112948253104244, "tokenization_path_logprob_std": 39.91371318820871, "trajectory_balance_rmse": 212.95453253893558} +{"canonical_length": 133, "canonical_nll": 1.1915734083132636, "generation_rouge_l_f1": 0.5294117647058824, "generation_token_f1": 0.6554621848739496, "prompt_id": "4973bf9719e2751b", "reading_canonical_log_prob": -158.47926330566406, "reading_distances": [18, 18, 23, 24], "reading_log_probs": [-162.61212158203125, -161.4525146484375, -161.46812438964844, -162.82821655273438], "reading_logprob_std": 0.7327326103120284, "tokenization_path_logprob_std": 102.90445266897095, "trajectory_balance_rmse": 2765.3843180489234} +{"canonical_length": 15, "canonical_nll": 2.716638437906901, "generation_rouge_l_f1": 0.25, "generation_token_f1": 0.25, "prompt_id": "c4461f4a56e11ca5", "reading_canonical_log_prob": -40.749576568603516, "reading_distances": [18, 23, 21, 22], "reading_log_probs": [-42.34626388549805, -42.506813049316406, -41.250213623046875, -40.5748405456543], "reading_logprob_std": 0.9189092958248255, "tokenization_path_logprob_std": 7.541888676266082, "trajectory_balance_rmse": 296.5049684220552} +{"canonical_length": 51, "canonical_nll": 0.8440855437634038, "generation_rouge_l_f1": 0.5862068965517242, "generation_token_f1": 0.6379310344827586, "prompt_id": "cdeadd1b86cf8871", "reading_canonical_log_prob": -43.048362731933594, "reading_distances": [13, 10, 11, 10], "reading_log_probs": [-43.07320785522461, -43.10581970214844, -42.25792694091797, -42.80797576904297], "reading_logprob_std": 0.39224582191201324, "tokenization_path_logprob_std": 119.51040744415369, "trajectory_balance_rmse": 747.51252375848} +{"canonical_length": 11, "canonical_nll": 1.018389961936257, "generation_rouge_l_f1": 0.0851063829787234, "generation_token_f1": 0.12765957446808512, "prompt_id": "55ee0fbefdef0ed3", "reading_canonical_log_prob": -11.202289581298828, "reading_distances": [15, 14, 16, 12], "reading_log_probs": [-14.821832656860352, -12.645427703857422, -12.207870483398438, -11.609971046447754], "reading_logprob_std": 1.3996041796237535, "tokenization_path_logprob_std": 63.791289961084054, "trajectory_balance_rmse": 281.69102340836207} +{"canonical_length": 282, "canonical_nll": 1.5363319342863475, "generation_rouge_l_f1": 0.1794392523364486, "generation_token_f1": 0.38878504672897196, "prompt_id": "0b1cafe0e71072e8", "reading_canonical_log_prob": -433.24560546875, "reading_distances": [30, 22, 21, 21], "reading_log_probs": [-427.242431640625, -428.1429748535156, -430.76861572265625, -432.6534118652344], "reading_logprob_std": 2.471742768426577, "tokenization_path_logprob_std": 284.9629442235557, "trajectory_balance_rmse": 5926.035202390745} +{"canonical_length": 98, "canonical_nll": 1.23190011783522, "generation_rouge_l_f1": 0.23076923076923078, "generation_token_f1": 0.423076923076923, "prompt_id": "3a37b4e0a87fe5c3", "reading_canonical_log_prob": -120.72621154785156, "reading_distances": [15, 15, 20, 16], "reading_log_probs": [-126.87223815917969, -123.43990325927734, -124.66850280761719, -124.4512939453125], "reading_logprob_std": 1.4456224412375156, "tokenization_path_logprob_std": 143.4012695138149, "trajectory_balance_rmse": 2283.6236257372393} +{"canonical_length": 261, "canonical_nll": 1.0835754862233598, "generation_rouge_l_f1": 0.16237623762376238, "generation_token_f1": 0.38415841584158417, "prompt_id": "7236a1de81b123d6", "reading_canonical_log_prob": -282.8132019042969, "reading_distances": [24, 22, 23, 24], "reading_log_probs": [-281.410888671875, -282.1896057128906, -282.4313049316406, -281.4893798828125], "reading_logprob_std": 0.5074255254402248, "tokenization_path_logprob_std": 217.87427213440185, "trajectory_balance_rmse": 5382.883401550031} +{"canonical_length": 181, "canonical_nll": 1.388548455844268, "generation_rouge_l_f1": 0.2636363636363636, "generation_token_f1": 0.45, "prompt_id": "564e5eeb9b5a33fe", "reading_canonical_log_prob": -251.3272705078125, "reading_distances": [42, 46, 36, 36], "reading_log_probs": [-248.91253662109375, -248.7040252685547, -250.4956817626953, -249.572509765625], "reading_logprob_std": 0.8063257678010434, "tokenization_path_logprob_std": 130.0696478395246, "trajectory_balance_rmse": 3657.6009402059767} +{"canonical_length": 141, "canonical_nll": 1.074311168481272, "generation_rouge_l_f1": 0.24242424242424238, "generation_token_f1": 0.4462809917355372, "prompt_id": "8686f6b687c7ab8c", "reading_canonical_log_prob": -151.47787475585938, "reading_distances": [22, 20, 25, 24], "reading_log_probs": [-152.65325927734375, -149.87802124023438, -150.9559783935547, -153.56771850585938], "reading_logprob_std": 1.658714064978455, "tokenization_path_logprob_std": 178.69990634207085, "trajectory_balance_rmse": 3029.9816782171556} +{"canonical_length": 101, "canonical_nll": 1.1509016433564743, "generation_rouge_l_f1": 0.39090909090909093, "generation_token_f1": 0.5545454545454545, "prompt_id": "79b5d532fcbe3323", "reading_canonical_log_prob": -116.2410659790039, "reading_distances": [28, 26, 24, 28], "reading_log_probs": [-116.2999496459961, -116.77226257324219, -114.54740905761719, -116.28373718261719], "reading_logprob_std": 0.9788686684155812, "tokenization_path_logprob_std": 119.19114011534097, "trajectory_balance_rmse": 2704.8196626658405} +{"canonical_length": 333, "canonical_nll": 0.9511511633704016, "generation_rouge_l_f1": 0.2581699346405229, "generation_token_f1": 0.5065359477124183, "prompt_id": "f5411f1c63da4002", "reading_canonical_log_prob": -316.73333740234375, "reading_distances": [45, 44, 44, 54], "reading_log_probs": [-319.0907287597656, -319.124267578125, -318.0110778808594, -319.86248779296875], "reading_logprob_std": 0.762357195826959, "tokenization_path_logprob_std": 180.73921822254593, "trajectory_balance_rmse": 7643.819580823251} +{"canonical_length": 105, "canonical_nll": 1.2862608410063243, "generation_rouge_l_f1": 0.19251336898395724, "generation_token_f1": 0.36898395721925137, "prompt_id": "0709a316fd9c3cd1", "reading_canonical_log_prob": -135.05738830566406, "reading_distances": [30, 26, 31, 25], "reading_log_probs": [-132.6116180419922, -134.19725036621094, -134.53672790527344, -133.33538818359375], "reading_logprob_std": 0.868206466392215, "tokenization_path_logprob_std": 29.851998208510256, "trajectory_balance_rmse": 1889.5197052236056} +{"canonical_length": 68, "canonical_nll": 0.6174675436580882, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.41471571906354515, "prompt_id": "28c4e87a8705b13a", "reading_canonical_log_prob": -41.98779296875, "reading_distances": [29, 29, 25, 27], "reading_log_probs": [-42.47949981689453, -44.06097412109375, -43.75668716430664, -42.48664474487305], "reading_logprob_std": 0.8324877791540829, "tokenization_path_logprob_std": 144.52337764477068, "trajectory_balance_rmse": 1356.039108730018} +{"canonical_length": 10, "canonical_nll": 0.3638916015625, "generation_rouge_l_f1": 0.8888888888888888, "generation_token_f1": 0.8888888888888888, "prompt_id": "d86571d08d8aa4d7", "reading_canonical_log_prob": -3.638916015625, "reading_distances": [36, 42, 39, 47], "reading_log_probs": [-3.0473761558532715, -3.307236671447754, -2.989825963973999, -3.0894553661346436], "reading_logprob_std": 0.13865858671152034, "tokenization_path_logprob_std": 26.715809203030297, "trajectory_balance_rmse": 256.0005158384138} +{"canonical_length": 30, "canonical_nll": 0.43417332967122396, "generation_rouge_l_f1": 0.4144144144144144, "generation_token_f1": 0.4504504504504505, "prompt_id": "f8bb815af5d18371", "reading_canonical_log_prob": -13.025199890136719, "reading_distances": [26, 20, 27, 18], "reading_log_probs": [-14.069648742675781, -13.57284927368164, -12.39129638671875, -12.63032054901123], "reading_logprob_std": 0.7893511134626039, "tokenization_path_logprob_std": 116.17636090697704, "trajectory_balance_rmse": 592.1903302109097} +{"canonical_length": 39, "canonical_nll": 1.9959139701647637, "generation_rouge_l_f1": 0.22535211267605634, "generation_token_f1": 0.22535211267605634, "prompt_id": "e8ce7622bc908c26", "reading_canonical_log_prob": -77.84064483642578, "reading_distances": [19, 20, 22, 15], "reading_log_probs": [-80.07231903076172, -79.87445068359375, -79.71598052978516, -77.31563568115234], "reading_logprob_std": 1.2942093628553104, "tokenization_path_logprob_std": 88.38199218793422, "trajectory_balance_rmse": 846.7787114615949} +{"canonical_length": 35, "canonical_nll": 0.6976862226213728, "generation_rouge_l_f1": 0.41666666666666663, "generation_token_f1": 0.4999999999999999, "prompt_id": "300a5e4b8236fdd5", "reading_canonical_log_prob": -24.419017791748047, "reading_distances": [33, 33, 27, 25], "reading_log_probs": [-26.44097900390625, -26.38262939453125, -26.098255157470703, -23.671855926513672], "reading_logprob_std": 1.3261938539538196, "tokenization_path_logprob_std": 129.88399977376085, "trajectory_balance_rmse": 785.4548443007862} +{"canonical_length": 64, "canonical_nll": 0.9237629771232605, "generation_rouge_l_f1": 0.18012422360248448, "generation_token_f1": 0.2546583850931677, "prompt_id": "20d20afcda52df5b", "reading_canonical_log_prob": -59.12083053588867, "reading_distances": [28, 29, 26, 27], "reading_log_probs": [-58.708892822265625, -59.64878845214844, -59.815460205078125, -59.265445709228516], "reading_logprob_std": 0.49116302123826183, "tokenization_path_logprob_std": 164.15153851251188, "trajectory_balance_rmse": 1379.615881910702} +{"canonical_length": 92, "canonical_nll": 1.883028445036515, "generation_rouge_l_f1": 0.3142857142857143, "generation_token_f1": 0.5142857142857143, "prompt_id": "26bff204aae37f53", "reading_canonical_log_prob": -173.23861694335938, "reading_distances": [15, 14, 19, 14], "reading_log_probs": [-170.8629150390625, -169.08499145507812, -169.4573211669922, -168.84521484375], "reading_logprob_std": 0.9027096649588133, "tokenization_path_logprob_std": 110.90909400009562, "trajectory_balance_rmse": 2073.506518489826} +{"canonical_length": 259, "canonical_nll": 1.498043222316904, "generation_rouge_l_f1": 0.20793950850661624, "generation_token_f1": 0.43478260869565216, "prompt_id": "d7bb721e18ff9d29", "reading_canonical_log_prob": -387.9931945800781, "reading_distances": [21, 22, 26, 25], "reading_log_probs": [-395.1220397949219, -395.00439453125, -391.16302490234375, -396.2066650390625], "reading_logprob_std": 2.208014418667186, "tokenization_path_logprob_std": 188.73044685843425, "trajectory_balance_rmse": 6071.154694741302} +{"canonical_length": 110, "canonical_nll": 0.7557094920765269, "generation_rouge_l_f1": 0.36512261580381467, "generation_token_f1": 0.4250681198910082, "prompt_id": "6df1e56bb497e6a2", "reading_canonical_log_prob": -83.12804412841797, "reading_distances": [41, 43, 35, 42], "reading_log_probs": [-87.72574615478516, -87.27970886230469, -86.11087799072266, -86.59524536132812], "reading_logprob_std": 0.7161266262466223, "tokenization_path_logprob_std": 104.31258716381048, "trajectory_balance_rmse": 1959.570859569023} +{"canonical_length": 9, "canonical_nll": 0.37571920288933647, "generation_rouge_l_f1": 0.75, "generation_token_f1": 0.75, "prompt_id": "196c78145c93b0b5", "reading_canonical_log_prob": -3.3814728260040283, "reading_distances": [36, 41, 38, 35], "reading_log_probs": [-3.7934348583221436, -4.138620376586914, -4.513724327087402, -4.4374542236328125], "reading_logprob_std": 0.3276879453371393, "tokenization_path_logprob_std": 43.54372219231054, "trajectory_balance_rmse": 192.05724174119842} +{"canonical_length": 61, "canonical_nll": 0.43842062402944093, "generation_rouge_l_f1": 0.75, "generation_token_f1": 0.7857142857142858, "prompt_id": "31d1718162ea0ff4", "reading_canonical_log_prob": -26.7436580657959, "reading_distances": [25, 22, 22, 19], "reading_log_probs": [-26.456117630004883, -26.749116897583008, -26.796470642089844, -27.03333282470703], "reading_logprob_std": 0.23699347539113588, "tokenization_path_logprob_std": 76.5933797811598, "trajectory_balance_rmse": 970.6574909261973} +{"canonical_length": 117, "canonical_nll": 0.9993038340511485, "generation_rouge_l_f1": 0.21122112211221125, "generation_token_f1": 0.36303630363036304, "prompt_id": "823548660cad1f9e", "reading_canonical_log_prob": -116.91854858398438, "reading_distances": [23, 23, 25, 25], "reading_log_probs": [-118.95951843261719, -116.13577270507812, -120.1902847290039, -115.73208618164062], "reading_logprob_std": 2.167606387120674, "tokenization_path_logprob_std": 117.70511823926361, "trajectory_balance_rmse": 1864.4033401041854} +{"canonical_length": 14, "canonical_nll": 0.7697342463902065, "generation_rouge_l_f1": 0.04098360655737705, "generation_token_f1": 0.05737704918032787, "prompt_id": "912aaeeaa0cce8d9", "reading_canonical_log_prob": -10.77627944946289, "reading_distances": [32, 32, 28, 26], "reading_log_probs": [-9.191853523254395, -10.64920425415039, -9.801517486572266, -9.619504928588867], "reading_logprob_std": 0.6117208754645008, "tokenization_path_logprob_std": 52.37930889838875, "trajectory_balance_rmse": 357.17661427565605} +{"canonical_length": 368, "canonical_nll": 1.5137242856233015, "generation_rouge_l_f1": 0.24290220820189273, "generation_token_f1": 0.4574132492113565, "prompt_id": "4547f9ca1d45d608", "reading_canonical_log_prob": -557.050537109375, "reading_distances": [25, 26, 27, 25], "reading_log_probs": [-560.409912109375, -558.7898559570312, -558.461669921875, -559.1851196289062], "reading_logprob_std": 0.8518443275570613, "tokenization_path_logprob_std": 213.68455782335099, "trajectory_balance_rmse": 7584.973171038353} +{"canonical_length": 228, "canonical_nll": 1.3285724572968065, "generation_rouge_l_f1": 0.2908366533864542, "generation_token_f1": 0.50199203187251, "prompt_id": "99e3f7e86895d3aa", "reading_canonical_log_prob": -302.9145202636719, "reading_distances": [20, 23, 20, 19], "reading_log_probs": [-303.8346252441406, -305.2813720703125, -304.9902648925781, -303.43963623046875], "reading_logprob_std": 0.8881525412079264, "tokenization_path_logprob_std": 35.420988509577136, "trajectory_balance_rmse": 5247.872621302456} +{"canonical_length": 318, "canonical_nll": 0.7689019928938188, "generation_rouge_l_f1": 0.26204238921001927, "generation_token_f1": 0.4816955684007707, "prompt_id": "d3e72bee48c9454b", "reading_canonical_log_prob": -244.51083374023438, "reading_distances": [17, 25, 23, 29], "reading_log_probs": [-244.31375122070312, -242.33436584472656, -243.42794799804688, -242.86541748046875], "reading_logprob_std": 0.846298941868417, "tokenization_path_logprob_std": 96.75995723763683, "trajectory_balance_rmse": 5321.702445821653} +{"canonical_length": 281, "canonical_nll": 1.3940411224908251, "generation_rouge_l_f1": 0.23264540337711068, "generation_token_f1": 0.4878048780487805, "prompt_id": "3126c4e8c381cc6a", "reading_canonical_log_prob": -391.7255554199219, "reading_distances": [25, 25, 24, 26], "reading_log_probs": [-393.06103515625, -397.4903564453125, -394.1541748046875, -394.7774353027344], "reading_logprob_std": 1.884992784006884, "tokenization_path_logprob_std": 248.52617257744052, "trajectory_balance_rmse": 6403.022450276688} +{"canonical_length": 23, "canonical_nll": 0.34140833564426587, "generation_rouge_l_f1": 0.21739130434782605, "generation_token_f1": 0.21739130434782605, "prompt_id": "963018756c1975a3", "reading_canonical_log_prob": -7.852391719818115, "reading_distances": [44, 40, 38, 46], "reading_log_probs": [-6.050078392028809, -6.466460704803467, -6.266556739807129, -6.411492824554443], "reading_logprob_std": 0.18593166962364607, "tokenization_path_logprob_std": 60.446008033843235, "trajectory_balance_rmse": 370.45878085502335} +{"canonical_length": 50, "canonical_nll": 2.10891845703125, "generation_rouge_l_f1": 0.23703703703703702, "generation_token_f1": 0.3259259259259259, "prompt_id": "b9207d2d35a02140", "reading_canonical_log_prob": -105.4459228515625, "reading_distances": [35, 37, 39, 39], "reading_log_probs": [-107.114501953125, -107.57121276855469, -106.79246520996094, -108.3909683227539], "reading_logprob_std": 0.6937409432403887, "tokenization_path_logprob_std": 11.825651353405453, "trajectory_balance_rmse": 1242.8722679101459} +{"canonical_length": 59, "canonical_nll": 0.6769159607968088, "generation_rouge_l_f1": 0.18045112781954886, "generation_token_f1": 0.2556390977443609, "prompt_id": "cc824ceca446b0f2", "reading_canonical_log_prob": -39.93804168701172, "reading_distances": [12, 11, 16, 14], "reading_log_probs": [-44.780548095703125, -42.372798919677734, -43.24102783203125, -46.278045654296875], "reading_logprob_std": 1.7233235435769787, "tokenization_path_logprob_std": 137.63298696425343, "trajectory_balance_rmse": 1295.7567537726995} +{"canonical_length": 27, "canonical_nll": 1.2002299273455586, "generation_rouge_l_f1": 0.7391304347826089, "generation_token_f1": 0.7391304347826089, "prompt_id": "9cebf3474b92652d", "reading_canonical_log_prob": -32.40620803833008, "reading_distances": [21, 26, 27, 25], "reading_log_probs": [-31.20086097717285, -32.06840515136719, -32.22474670410156, -32.916236877441406], "reading_logprob_std": 0.7050360116029519, "tokenization_path_logprob_std": 61.01970499147581, "trajectory_balance_rmse": 632.962890528565} +{"canonical_length": 5, "canonical_nll": 2.5225093841552733, "generation_rouge_l_f1": 0.0, "generation_token_f1": 0.0, "prompt_id": "c004648702d6143d", "reading_canonical_log_prob": -12.612546920776367, "reading_distances": [22, 19, 24, 21], "reading_log_probs": [-16.749805450439453, -17.566394805908203, -16.944637298583984, -18.972270965576172], "reading_logprob_std": 1.0049253310671549, "tokenization_path_logprob_std": 2.308224461879643, "trajectory_balance_rmse": 59.47398876509113} +{"canonical_length": 120, "canonical_nll": 1.2111775716145834, "generation_rouge_l_f1": 0.23826714801444043, "generation_token_f1": 0.47653429602888087, "prompt_id": "ebe4917bfacc9af5", "reading_canonical_log_prob": -145.34130859375, "reading_distances": [28, 27, 18, 32], "reading_log_probs": [-145.56640625, -147.09568786621094, -146.99354553222656, -148.0443115234375], "reading_logprob_std": 1.021839534222939, "tokenization_path_logprob_std": 164.8712474252793, "trajectory_balance_rmse": 2975.5716218057096} +{"canonical_length": 127, "canonical_nll": 1.4465199868510088, "generation_rouge_l_f1": 0.26735218508997427, "generation_token_f1": 0.41645244215938304, "prompt_id": "bd51aa5053398c97", "reading_canonical_log_prob": -183.70803833007812, "reading_distances": [31, 21, 23, 18], "reading_log_probs": [-183.12811279296875, -182.65371704101562, -182.1479949951172, -182.27099609375], "reading_logprob_std": 0.44137008702681413, "tokenization_path_logprob_std": 69.92760000773657, "trajectory_balance_rmse": 3052.3050032953806} +{"canonical_length": 168, "canonical_nll": 1.35940306527274, "generation_rouge_l_f1": 0.15560640732265446, "generation_token_f1": 0.30663615560640733, "prompt_id": "0d7273ba3d245c40", "reading_canonical_log_prob": -228.3797149658203, "reading_distances": [28, 22, 20, 15], "reading_log_probs": [-226.98880004882812, -225.3345947265625, -227.76539611816406, -228.30648803710938], "reading_logprob_std": 1.2945244962602391, "tokenization_path_logprob_std": 246.27200141620096, "trajectory_balance_rmse": 3679.7984813816897} +{"canonical_length": 211, "canonical_nll": 1.6257787044579384, "generation_rouge_l_f1": 0.2320675105485232, "generation_token_f1": 0.3670886075949367, "prompt_id": "16d4d8a6fd8e0475", "reading_canonical_log_prob": -343.039306640625, "reading_distances": [32, 31, 29, 33], "reading_log_probs": [-344.7771301269531, -345.93414306640625, -346.4971923828125, -347.496826171875], "reading_logprob_std": 1.1347654095273079, "tokenization_path_logprob_std": 249.9793900221414, "trajectory_balance_rmse": 4607.091992204139} +{"canonical_length": 131, "canonical_nll": 1.8055461854425094, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.46349206349206346, "prompt_id": "5669cff2cb390e91", "reading_canonical_log_prob": -236.52655029296875, "reading_distances": [26, 24, 21, 20], "reading_log_probs": [-236.17645263671875, -234.939208984375, -239.67376708984375, -237.52609252929688], "reading_logprob_std": 2.0269850563913847, "tokenization_path_logprob_std": 162.30585971664507, "trajectory_balance_rmse": 3039.750107261586} +{"canonical_length": 314, "canonical_nll": 1.0446610177398488, "generation_rouge_l_f1": 0.2222222222222222, "generation_token_f1": 0.4512820512820513, "prompt_id": "ea8af27e7f1b048e", "reading_canonical_log_prob": -328.0235595703125, "reading_distances": [27, 27, 28, 30], "reading_log_probs": [-330.0933532714844, -330.1695556640625, -332.0963134765625, -331.5869140625], "reading_logprob_std": 1.009503528346103, "tokenization_path_logprob_std": 208.88875167478062, "trajectory_balance_rmse": 6709.018892591326} +{"canonical_length": 8, "canonical_nll": 1.0060898065567017, "generation_rouge_l_f1": 0.028469750889679717, "generation_token_f1": 0.028469750889679717, "prompt_id": "b75f2d55eed057da", "reading_canonical_log_prob": -8.048718452453613, "reading_distances": [14, 17, 18, 17], "reading_log_probs": [-7.4438042640686035, -6.966358184814453, -6.716371059417725, -7.522770881652832], "reading_logprob_std": 0.38575887091597255, "tokenization_path_logprob_std": 38.196443677624124, "trajectory_balance_rmse": 154.25339181313686} +{"canonical_length": 22, "canonical_nll": 0.7301451076160778, "generation_rouge_l_f1": 0.06976744186046512, "generation_token_f1": 0.11627906976744186, "prompt_id": "55725f89c2c9fced", "reading_canonical_log_prob": -16.06319236755371, "reading_distances": [35, 29, 30, 32], "reading_log_probs": [-18.20725440979004, -19.309749603271484, -17.222896575927734, -17.769554138183594], "reading_logprob_std": 0.8851567489849247, "tokenization_path_logprob_std": 32.90562695411193, "trajectory_balance_rmse": 488.50981564708604} +{"canonical_length": 6, "canonical_nll": 1.5355720520019531, "generation_rouge_l_f1": 0.03305785123966942, "generation_token_f1": 0.03305785123966942, "prompt_id": "e8e15c8f276bd3e5", "reading_canonical_log_prob": -9.213432312011719, "reading_distances": [21, 24, 21, 25], "reading_log_probs": [-6.4003705978393555, -5.717983245849609, -7.417962074279785, -8.263435363769531], "reading_logprob_std": 1.1201277761229116, "tokenization_path_logprob_std": 18.277366283422253, "trajectory_balance_rmse": 92.12622159227048} +{"canonical_length": 37, "canonical_nll": 1.4572801332216005, "generation_rouge_l_f1": 0.14285714285714285, "generation_token_f1": 0.15584415584415584, "prompt_id": "6e9b2ca73b16e2a8", "reading_canonical_log_prob": -53.91936492919922, "reading_distances": [36, 33, 30, 32], "reading_log_probs": [-55.886871337890625, -55.803977966308594, -56.00773239135742, -57.23149108886719], "reading_logprob_std": 0.6712163033471635, "tokenization_path_logprob_std": 85.75659700144746, "trajectory_balance_rmse": 913.0237245607902} +{"canonical_length": 266, "canonical_nll": 1.4627554757254464, "generation_rouge_l_f1": 0.22997946611909653, "generation_token_f1": 0.4681724845995893, "prompt_id": "fb1406f3dcf9e9ba", "reading_canonical_log_prob": -389.09295654296875, "reading_distances": [24, 25, 22, 33], "reading_log_probs": [-393.15008544921875, -388.91790771484375, -387.5027160644531, -389.36151123046875], "reading_logprob_std": 2.4119852630478142, "tokenization_path_logprob_std": 44.548863954719344, "trajectory_balance_rmse": 5543.545529684412} +{"canonical_length": 3, "canonical_nll": 1.7750802040100098, "generation_rouge_l_f1": 0.014285714285714287, "generation_token_f1": 0.014285714285714287, "prompt_id": "f6b05fa9b7a6a33d", "reading_canonical_log_prob": -5.325240612030029, "reading_distances": [34, 24, 24, 32], "reading_log_probs": [-7.267947673797607, -7.066823959350586, -6.844972610473633, -6.662268161773682], "reading_logprob_std": 0.2633867014358024, "tokenization_path_logprob_std": 27.775470580048644, "trajectory_balance_rmse": 114.0365325417206} +{"canonical_length": 274, "canonical_nll": 1.2253580580662637, "generation_rouge_l_f1": 0.26815642458100564, "generation_token_f1": 0.4916201117318436, "prompt_id": "b4b2d4b733fbf2fd", "reading_canonical_log_prob": -335.74810791015625, "reading_distances": [27, 40, 31, 33], "reading_log_probs": [-346.491455078125, -347.66778564453125, -348.230224609375, -351.9349365234375], "reading_logprob_std": 2.350325649572274, "tokenization_path_logprob_std": 303.91126755195035, "trajectory_balance_rmse": 5337.192232803786} +{"canonical_length": 309, "canonical_nll": 0.6766090763425364, "generation_rouge_l_f1": 0.2533589251439539, "generation_token_f1": 0.42610364683301344, "prompt_id": "8c9f3f4649945cbf", "reading_canonical_log_prob": -209.07220458984375, "reading_distances": [52, 46, 43, 42], "reading_log_probs": [-212.03204345703125, -209.78866577148438, -210.5624237060547, -206.48146057128906], "reading_logprob_std": 2.348617744347897, "tokenization_path_logprob_std": 148.26463141274087, "trajectory_balance_rmse": 2922.443389916957} +{"canonical_length": 8, "canonical_nll": 0.11826930195093155, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "eccb7866ba7ed3da", "reading_canonical_log_prob": -0.9461544156074524, "reading_distances": [20, 26, 26, 26], "reading_log_probs": [-1.0736687183380127, -0.9442846775054932, -0.9117714166641235, -1.3181787729263306], "reading_logprob_std": 0.18456186384953266, "tokenization_path_logprob_std": 41.56998881936445, "trajectory_balance_rmse": 198.1201558341349} +{"canonical_length": 71, "canonical_nll": 2.080932187362456, "generation_rouge_l_f1": 0.15789473684210528, "generation_token_f1": 0.25146198830409355, "prompt_id": "f78297865f8b46fc", "reading_canonical_log_prob": -147.74618530273438, "reading_distances": [26, 17, 21, 20], "reading_log_probs": [-143.86854553222656, -141.50613403320312, -143.47308349609375, -144.536865234375], "reading_logprob_std": 1.3028776780343887, "tokenization_path_logprob_std": 56.55512996218048, "trajectory_balance_rmse": 1850.4121578582522} +{"canonical_length": 147, "canonical_nll": 1.044651680252179, "generation_rouge_l_f1": 0.22113022113022113, "generation_token_f1": 0.36855036855036855, "prompt_id": "65091372e7851e26", "reading_canonical_log_prob": -153.5637969970703, "reading_distances": [27, 26, 19, 24], "reading_log_probs": [-155.82736206054688, -155.32757568359375, -155.5667724609375, -155.041015625], "reading_logprob_std": 0.33563200268519305, "tokenization_path_logprob_std": 148.82264920417998, "trajectory_balance_rmse": 3080.4005490714358} +{"canonical_length": 16, "canonical_nll": 1.0446021556854248, "generation_rouge_l_f1": 0.5217391304347826, "generation_token_f1": 0.5217391304347826, "prompt_id": "aa225c057d9ef9cd", "reading_canonical_log_prob": -16.713634490966797, "reading_distances": [28, 29, 24, 25], "reading_log_probs": [-16.416101455688477, -17.162506103515625, -17.00531005859375, -16.947948455810547], "reading_logprob_std": 0.3241892248075797, "tokenization_path_logprob_std": 21.56574048466105, "trajectory_balance_rmse": 252.65360194031422} +{"canonical_length": 352, "canonical_nll": 0.9384530674327504, "generation_rouge_l_f1": 0.316008316008316, "generation_token_f1": 0.4948024948024948, "prompt_id": "1b2704eec1f82afa", "reading_canonical_log_prob": -330.3354797363281, "reading_distances": [31, 31, 32, 34], "reading_log_probs": [-331.0967102050781, -331.11285400390625, -333.0999755859375, -333.2221984863281], "reading_logprob_std": 1.1882745897167883, "tokenization_path_logprob_std": 440.4071206091425, "trajectory_balance_rmse": 6669.604350428398} +{"canonical_length": 284, "canonical_nll": 1.3686347209231955, "generation_rouge_l_f1": 0.2867924528301886, "generation_token_f1": 0.5433962264150944, "prompt_id": "aa3d50bf234f7b48", "reading_canonical_log_prob": -388.6922607421875, "reading_distances": [36, 33, 34, 30], "reading_log_probs": [-382.8061218261719, -385.148681640625, -383.3226318359375, -384.96099853515625], "reading_logprob_std": 1.1708896540589933, "tokenization_path_logprob_std": 273.254540094792, "trajectory_balance_rmse": 7105.468925150843} +{"canonical_length": 11, "canonical_nll": 0.7278823852539062, "generation_rouge_l_f1": 0.1372549019607843, "generation_token_f1": 0.1372549019607843, "prompt_id": "38ebe3c79fd5a7f3", "reading_canonical_log_prob": -8.006706237792969, "reading_distances": [23, 13, 19, 23], "reading_log_probs": [-7.795566558837891, -7.940579414367676, -7.142337799072266, -7.726130485534668], "reading_logprob_std": 0.3507788749534917, "tokenization_path_logprob_std": 37.006128185605824, "trajectory_balance_rmse": 120.27529432103515} +{"canonical_length": 266, "canonical_nll": 1.295247701773966, "generation_rouge_l_f1": 0.196969696969697, "generation_token_f1": 0.36363636363636365, "prompt_id": "df8d0cc99eb20244", "reading_canonical_log_prob": -344.535888671875, "reading_distances": [24, 18, 23, 28], "reading_log_probs": [-345.28387451171875, -345.8013610839844, -346.16510009765625, -346.48291015625], "reading_logprob_std": 0.5147698356190424, "tokenization_path_logprob_std": 268.227338768567, "trajectory_balance_rmse": 5783.067671383305} +{"canonical_length": 131, "canonical_nll": 1.6491890244811547, "generation_rouge_l_f1": 0.2480211081794195, "generation_token_f1": 0.43271767810026385, "prompt_id": "0f4d389ccaad372e", "reading_canonical_log_prob": -216.04376220703125, "reading_distances": [32, 29, 29, 26], "reading_log_probs": [-212.66912841796875, -211.2410125732422, -210.79373168945312, -211.165283203125], "reading_logprob_std": 0.8247233798527009, "tokenization_path_logprob_std": 138.0576782422868, "trajectory_balance_rmse": 2475.0560995102055} +{"canonical_length": 30, "canonical_nll": 1.4824325561523437, "generation_rouge_l_f1": 0.3111111111111111, "generation_token_f1": 0.4, "prompt_id": "936a9a79e3069ca4", "reading_canonical_log_prob": -44.47297668457031, "reading_distances": [27, 27, 24, 26], "reading_log_probs": [-45.62269973754883, -46.092308044433594, -45.34638977050781, -46.36302185058594], "reading_logprob_std": 0.4571812927507528, "tokenization_path_logprob_std": 117.90394305608308, "trajectory_balance_rmse": 781.6430682533813} +{"canonical_length": 427, "canonical_nll": 1.544063666386124, "generation_rouge_l_f1": 0.18711656441717792, "generation_token_f1": 0.35276073619631904, "prompt_id": "82391a0f7d399b89", "reading_canonical_log_prob": -659.315185546875, "reading_distances": [24, 29, 30, 30], "reading_log_probs": [-661.337158203125, -661.4382934570312, -662.3624267578125, -662.5023193359375], "reading_logprob_std": 0.6072305833481566, "tokenization_path_logprob_std": 274.7422673233811, "trajectory_balance_rmse": 7742.146350087783} +{"canonical_length": 189, "canonical_nll": 0.9785107809399801, "generation_rouge_l_f1": 0.3785488958990536, "generation_token_f1": 0.49842271293375395, "prompt_id": "0bf403e155ca208a", "reading_canonical_log_prob": -184.93853759765625, "reading_distances": [23, 21, 22, 24], "reading_log_probs": [-187.23809814453125, -187.80099487304688, -187.8463134765625, -187.06622314453125], "reading_logprob_std": 0.3944198088855271, "tokenization_path_logprob_std": 77.61234029254895, "trajectory_balance_rmse": 3331.5063530380085} +{"canonical_length": 175, "canonical_nll": 0.7627207728794643, "generation_rouge_l_f1": 0.22131147540983606, "generation_token_f1": 0.36885245901639346, "prompt_id": "3b3aada8838b1247", "reading_canonical_log_prob": -133.47613525390625, "reading_distances": [23, 18, 21, 16], "reading_log_probs": [-126.61104583740234, -128.42955017089844, -127.67539978027344, -127.28056335449219], "reading_logprob_std": 0.7600915707673203, "tokenization_path_logprob_std": 77.47224921581498, "trajectory_balance_rmse": 2893.3916470863524} +{"canonical_length": 62, "canonical_nll": 1.7506265947895665, "generation_rouge_l_f1": 0.3225806451612903, "generation_token_f1": 0.3440860215053763, "prompt_id": "0221b725811d035d", "reading_canonical_log_prob": -108.53884887695312, "reading_distances": [14, 20, 17, 12], "reading_log_probs": [-111.63175964355469, -110.38091278076172, -109.21212005615234, -112.9876480102539], "reading_logprob_std": 1.62464004197883, "tokenization_path_logprob_std": 106.0695576060842, "trajectory_balance_rmse": 1372.831703636014} +{"canonical_length": 109, "canonical_nll": 0.8947961089807913, "generation_rouge_l_f1": 0.23896103896103896, "generation_token_f1": 0.4, "prompt_id": "7ac2b23c73ad8de9", "reading_canonical_log_prob": -97.53277587890625, "reading_distances": [39, 38, 40, 33], "reading_log_probs": [-100.56381225585938, -99.64167022705078, -100.19252014160156, -98.94208526611328], "reading_logprob_std": 0.7056111008931734, "tokenization_path_logprob_std": 118.23542642768636, "trajectory_balance_rmse": 2190.520552188343} +{"canonical_length": 98, "canonical_nll": 1.3168433053152901, "generation_rouge_l_f1": 0.3346938775510204, "generation_token_f1": 0.43265306122448977, "prompt_id": "a2a34e9410c7790e", "reading_canonical_log_prob": -129.05064392089844, "reading_distances": [21, 22, 25, 22], "reading_log_probs": [-128.2227783203125, -129.48309326171875, -128.4640350341797, -129.76295471191406], "reading_logprob_std": 0.7540300780025674, "tokenization_path_logprob_std": 140.44424153260346, "trajectory_balance_rmse": 2290.7303074829224} +{"canonical_length": 139, "canonical_nll": 1.0421229300739097, "generation_rouge_l_f1": 0.09210526315789473, "generation_token_f1": 0.09210526315789473, "prompt_id": "1c2356a54b0c56f0", "reading_canonical_log_prob": -144.85508728027344, "reading_distances": [31, 27, 25, 29], "reading_log_probs": [-144.00341796875, -142.2877960205078, -143.13375854492188, -143.43617248535156], "reading_logprob_std": 0.7157344078093008, "tokenization_path_logprob_std": 147.67363632151952, "trajectory_balance_rmse": 3465.3375386261487} +{"canonical_length": 150, "canonical_nll": 1.4851012166341147, "generation_rouge_l_f1": 0.07389162561576355, "generation_token_f1": 0.11822660098522167, "prompt_id": "55d3c9fe027b6cf4", "reading_canonical_log_prob": -222.7651824951172, "reading_distances": [20, 18, 19, 24], "reading_log_probs": [-224.09439086914062, -223.2109375, -224.76242065429688, -223.51425170898438], "reading_logprob_std": 0.6843695576909904, "tokenization_path_logprob_std": 76.35216743876627, "trajectory_balance_rmse": 3339.3910894091664} +{"canonical_length": 22, "canonical_nll": 2.6123064214533027, "generation_rouge_l_f1": 0.18181818181818185, "generation_token_f1": 0.18181818181818185, "prompt_id": "99817bb6bf526c93", "reading_canonical_log_prob": -57.470741271972656, "reading_distances": [17, 20, 20, 23], "reading_log_probs": [-59.58979034423828, -58.85539245605469, -58.91345977783203, -59.706459045410156], "reading_logprob_std": 0.44411981030010733, "tokenization_path_logprob_std": 45.126276142485395, "trajectory_balance_rmse": 465.16810828074864} +{"canonical_length": 177, "canonical_nll": 1.5364216087901659, "generation_rouge_l_f1": 0.2758620689655173, "generation_token_f1": 0.47214854111405835, "prompt_id": "a99bf80957c76958", "reading_canonical_log_prob": -271.9466247558594, "reading_distances": [22, 28, 30, 26], "reading_log_probs": [-276.1933898925781, -274.1331787109375, -276.46942138671875, -273.35833740234375], "reading_logprob_std": 1.530126317568208, "tokenization_path_logprob_std": 239.4131577423057, "trajectory_balance_rmse": 4041.075418767934} +{"canonical_length": 333, "canonical_nll": 1.6045226134337462, "generation_rouge_l_f1": 0.251207729468599, "generation_token_f1": 0.5249597423510467, "prompt_id": "6b033b3db0a5c2fe", "reading_canonical_log_prob": -534.3060302734375, "reading_distances": [23, 27, 26, 29], "reading_log_probs": [-535.0497436523438, -538.2489624023438, -536.7252197265625, -539.299560546875], "reading_logprob_std": 1.8519346457396704, "tokenization_path_logprob_std": 22.1424519214307, "trajectory_balance_rmse": 7719.645844865462} +{"canonical_length": 190, "canonical_nll": 1.0388924046566612, "generation_rouge_l_f1": 0.2391304347826087, "generation_token_f1": 0.42173913043478256, "prompt_id": "f5608ead26273538", "reading_canonical_log_prob": -197.38955688476562, "reading_distances": [34, 36, 38, 46], "reading_log_probs": [-204.39479064941406, -203.83120727539062, -203.05465698242188, -207.15298461914062], "reading_logprob_std": 1.7831326118279147, "tokenization_path_logprob_std": 100.09710028626408, "trajectory_balance_rmse": 3698.407788262353} +{"canonical_length": 305, "canonical_nll": 1.1291175717213116, "generation_rouge_l_f1": 0.22262118491921004, "generation_token_f1": 0.47037701974865354, "prompt_id": "eb21c6c6320f2679", "reading_canonical_log_prob": -344.380859375, "reading_distances": [32, 25, 34, 31], "reading_log_probs": [-339.0790100097656, -337.49920654296875, -340.70001220703125, -338.1630554199219], "reading_logprob_std": 1.3869751829697023, "tokenization_path_logprob_std": 336.7624392687319, "trajectory_balance_rmse": 6697.834669882271} +{"canonical_length": 222, "canonical_nll": 0.8110759838207348, "generation_rouge_l_f1": 0.32876712328767127, "generation_token_f1": 0.6347031963470319, "prompt_id": "bd9aa33662e77c04", "reading_canonical_log_prob": -180.05886840820312, "reading_distances": [18, 17, 19, 19], "reading_log_probs": [-181.00352478027344, -178.79342651367188, -177.57594299316406, -177.82435607910156], "reading_logprob_std": 1.5605297598625862, "tokenization_path_logprob_std": 149.95212721682321, "trajectory_balance_rmse": 2287.5586641247305} +{"canonical_length": 85, "canonical_nll": 0.7026382446289062, "generation_rouge_l_f1": 0.19631901840490798, "generation_token_f1": 0.29447852760736193, "prompt_id": "4fcfacdaa4123cbd", "reading_canonical_log_prob": -59.72425079345703, "reading_distances": [28, 26, 23, 26], "reading_log_probs": [-57.114200592041016, -54.66069030761719, -55.112239837646484, -55.664398193359375], "reading_logprob_std": 1.066368132216651, "tokenization_path_logprob_std": 150.32482989287215, "trajectory_balance_rmse": 1205.6100768296549} +{"canonical_length": 214, "canonical_nll": 1.6288753046053592, "generation_rouge_l_f1": 0.2431865828092243, "generation_token_f1": 0.5073375262054508, "prompt_id": "42e23f667d7b63a7", "reading_canonical_log_prob": -348.5793151855469, "reading_distances": [17, 18, 22, 22], "reading_log_probs": [-343.149169921875, -342.29620361328125, -345.57763671875, -343.0840148925781], "reading_logprob_std": 1.4211452932516984, "tokenization_path_logprob_std": 212.72782742117002, "trajectory_balance_rmse": 4486.233173414179} +{"canonical_length": 13, "canonical_nll": 1.3063627389761119, "generation_rouge_l_f1": 0.05283018867924529, "generation_token_f1": 0.06037735849056603, "prompt_id": "f4ba682f79ee92c5", "reading_canonical_log_prob": -16.982715606689453, "reading_distances": [12, 12, 19, 18], "reading_log_probs": [-16.431385040283203, -17.70237922668457, -17.21117401123047, -16.85138702392578], "reading_logprob_std": 0.5396615989380227, "tokenization_path_logprob_std": 59.54134263044304, "trajectory_balance_rmse": 327.87482453502486} +{"canonical_length": 112, "canonical_nll": 1.028104373386928, "generation_rouge_l_f1": 0.15897435897435896, "generation_token_f1": 0.27692307692307694, "prompt_id": "8c5355410cf1d176", "reading_canonical_log_prob": -115.14768981933594, "reading_distances": [22, 20, 30, 24], "reading_log_probs": [-114.85954284667969, -116.88895416259766, -115.80046081542969, -115.60279846191406], "reading_logprob_std": 0.8383689761424226, "tokenization_path_logprob_std": 80.96842104868172, "trajectory_balance_rmse": 2256.9537187771093} +{"canonical_length": 255, "canonical_nll": 1.2149921492034315, "generation_rouge_l_f1": 0.2862595419847328, "generation_token_f1": 0.48473282442748095, "prompt_id": "131fb5156c48d9ba", "reading_canonical_log_prob": -309.822998046875, "reading_distances": [22, 25, 26, 26], "reading_log_probs": [-312.126953125, -313.47149658203125, -314.54852294921875, -313.4264221191406], "reading_logprob_std": 0.9908558995396025, "tokenization_path_logprob_std": 228.6547060430338, "trajectory_balance_rmse": 5781.608385246746} +{"canonical_length": 17, "canonical_nll": 1.0061311160816866, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "20a85de31d07820a", "reading_canonical_log_prob": -17.104228973388672, "reading_distances": [17, 18, 17, 23], "reading_log_probs": [-16.596240997314453, -16.487600326538086, -16.88765525817871, -16.58521270751953], "reading_logprob_std": 0.17269682289532812, "tokenization_path_logprob_std": 73.55688605710412, "trajectory_balance_rmse": 271.06379183007556} +{"canonical_length": 236, "canonical_nll": 1.113104998055151, "generation_rouge_l_f1": 0.25416666666666665, "generation_token_f1": 0.525, "prompt_id": "8bdd0bdfad8b9b1b", "reading_canonical_log_prob": -262.6927795410156, "reading_distances": [20, 18, 17, 16], "reading_log_probs": [-262.0198974609375, -263.7833557128906, -263.250244140625, -263.45965576171875], "reading_logprob_std": 0.7707834086817912, "tokenization_path_logprob_std": 179.560965231942, "trajectory_balance_rmse": 5450.040979578683} +{"canonical_length": 314, "canonical_nll": 1.401813677162122, "generation_rouge_l_f1": 0.24913494809688586, "generation_token_f1": 0.5224913494809689, "prompt_id": "f543bc422c7c86fd", "reading_canonical_log_prob": -440.16949462890625, "reading_distances": [23, 23, 27, 26], "reading_log_probs": [-436.49945068359375, -438.09674072265625, -439.42388916015625, -438.9571533203125], "reading_logprob_std": 1.2865798744533474, "tokenization_path_logprob_std": 250.98952347949384, "trajectory_balance_rmse": 7868.8708248236535} +{"canonical_length": 17, "canonical_nll": 1.3209937600528492, "generation_rouge_l_f1": 0.5294117647058824, "generation_token_f1": 0.6470588235294117, "prompt_id": "6b9813758e34b6e5", "reading_canonical_log_prob": -22.456893920898438, "reading_distances": [28, 34, 32, 32], "reading_log_probs": [-19.38747215270996, -20.335636138916016, -20.204811096191406, -19.947511672973633], "reading_logprob_std": 0.41978773838511124, "tokenization_path_logprob_std": 71.42971355686032, "trajectory_balance_rmse": 498.6640287626618} +{"canonical_length": 133, "canonical_nll": 1.2661137401609492, "generation_rouge_l_f1": 0.18639798488664988, "generation_token_f1": 0.38287153652392947, "prompt_id": "89680ea889169bf7", "reading_canonical_log_prob": -168.39312744140625, "reading_distances": [24, 22, 25, 25], "reading_log_probs": [-160.3876953125, -159.8707275390625, -160.9090118408203, -161.4421844482422], "reading_logprob_std": 0.6759412004051277, "tokenization_path_logprob_std": 42.55519374187724, "trajectory_balance_rmse": 3082.098993483271} +{"canonical_length": 15, "canonical_nll": 1.2183774312337239, "generation_rouge_l_f1": 0.35000000000000003, "generation_token_f1": 0.35000000000000003, "prompt_id": "5d244886eb8450ed", "reading_canonical_log_prob": -18.27566146850586, "reading_distances": [33, 28, 32, 31], "reading_log_probs": [-17.97646141052246, -14.970111846923828, -15.177797317504883, -15.633295059204102], "reading_logprob_std": 1.3859867235559453, "tokenization_path_logprob_std": 58.12254156285488, "trajectory_balance_rmse": 464.0093978082025} +{"canonical_length": 245, "canonical_nll": 0.8856919269172513, "generation_rouge_l_f1": 0.24361493123772104, "generation_token_f1": 0.5068762278978389, "prompt_id": "8dfd747741dad44f", "reading_canonical_log_prob": -216.99452209472656, "reading_distances": [21, 20, 17, 20], "reading_log_probs": [-214.94285583496094, -215.2374725341797, -213.6553955078125, -215.593505859375], "reading_logprob_std": 0.8442796598556628, "tokenization_path_logprob_std": 117.80014042465442, "trajectory_balance_rmse": 4904.164716295703} +{"canonical_length": 18, "canonical_nll": 0.7222991519504123, "generation_rouge_l_f1": 0.34615384615384615, "generation_token_f1": 0.34615384615384615, "prompt_id": "73b880653e3d865c", "reading_canonical_log_prob": -13.001384735107422, "reading_distances": [18, 22, 27, 17], "reading_log_probs": [-13.931648254394531, -14.097580909729004, -12.891317367553711, -13.314876556396484], "reading_logprob_std": 0.5580717958614928, "tokenization_path_logprob_std": 28.515568608217077, "trajectory_balance_rmse": 344.7564049675636} +{"canonical_length": 18, "canonical_nll": 1.8399039374457464, "generation_rouge_l_f1": 0.27586206896551724, "generation_token_f1": 0.27586206896551724, "prompt_id": "50d05075f2282d29", "reading_canonical_log_prob": -33.11827087402344, "reading_distances": [22, 24, 29, 30], "reading_log_probs": [-32.956031799316406, -33.58558654785156, -32.85625457763672, -32.823848724365234], "reading_logprob_std": 0.35788578421360395, "tokenization_path_logprob_std": 26.558143745368316, "trajectory_balance_rmse": 426.15508054123336} +{"canonical_length": 12, "canonical_nll": 1.0576226711273193, "generation_rouge_l_f1": 0.07801418439716312, "generation_token_f1": 0.07801418439716312, "prompt_id": "4907b79891256514", "reading_canonical_log_prob": -12.691472053527832, "reading_distances": [36, 32, 29, 40], "reading_log_probs": [-14.420841217041016, -13.655688285827637, -13.242609977722168, -11.922661781311035], "reading_logprob_std": 1.04605872005944, "tokenization_path_logprob_std": 39.65012235905246, "trajectory_balance_rmse": 268.3788231130632} +{"canonical_length": 58, "canonical_nll": 2.474195283034752, "generation_rouge_l_f1": 0.2142857142857143, "generation_token_f1": 0.26530612244897955, "prompt_id": "d9bc5b8ac130964c", "reading_canonical_log_prob": -143.50332641601562, "reading_distances": [13, 12, 10, 14], "reading_log_probs": [-143.64791870117188, -145.39654541015625, -145.92201232910156, -144.5550079345703], "reading_logprob_std": 0.9960342151795657, "tokenization_path_logprob_std": 57.37952121019252, "trajectory_balance_rmse": 1400.8034221952387} +{"canonical_length": 14, "canonical_nll": 1.133443832397461, "generation_rouge_l_f1": 0.15267175572519087, "generation_token_f1": 0.15267175572519087, "prompt_id": "0946ba5ceddc203c", "reading_canonical_log_prob": -15.868213653564453, "reading_distances": [17, 14, 12, 15], "reading_log_probs": [-14.933244705200195, -15.33930492401123, -15.095404624938965, -14.70666790008545], "reading_logprob_std": 0.26666942561056184, "tokenization_path_logprob_std": 22.928302385866246, "trajectory_balance_rmse": 212.48076153373466} +{"canonical_length": 17, "canonical_nll": 1.0525133469525505, "generation_rouge_l_f1": 0.1033210332103321, "generation_token_f1": 0.1180811808118081, "prompt_id": "d08a5a0e92cb2564", "reading_canonical_log_prob": -17.89272689819336, "reading_distances": [27, 16, 21, 20], "reading_log_probs": [-20.63653564453125, -19.90182876586914, -20.02744483947754, -21.802490234375], "reading_logprob_std": 0.8683948545875904, "tokenization_path_logprob_std": 60.575706541873735, "trajectory_balance_rmse": 295.3372508027224} +{"canonical_length": 65, "canonical_nll": 0.5891278780423678, "generation_rouge_l_f1": 0.39603960396039606, "generation_token_f1": 0.5148514851485149, "prompt_id": "02aff4ae4d6fb42e", "reading_canonical_log_prob": -38.293312072753906, "reading_distances": [18, 26, 25, 20], "reading_log_probs": [-36.945030212402344, -36.635337829589844, -37.187835693359375, -36.894012451171875], "reading_logprob_std": 0.2265622122960758, "tokenization_path_logprob_std": 62.74458745723594, "trajectory_balance_rmse": 1123.401565276145} +{"canonical_length": 281, "canonical_nll": 1.1349133257339858, "generation_rouge_l_f1": 0.1933085501858736, "generation_token_f1": 0.4312267657992565, "prompt_id": "a79f040945146e56", "reading_canonical_log_prob": -318.91064453125, "reading_distances": [35, 35, 41, 35], "reading_log_probs": [-319.34100341796875, -320.52618408203125, -319.5830078125, -320.6823425292969], "reading_logprob_std": 0.6698826825542384, "tokenization_path_logprob_std": 281.57861232770085, "trajectory_balance_rmse": 6088.756089460111} +{"canonical_length": 341, "canonical_nll": 1.3483070530150294, "generation_rouge_l_f1": 0.2736842105263158, "generation_token_f1": 0.4491228070175438, "prompt_id": "79e68a828bcf94f4", "reading_canonical_log_prob": -459.772705078125, "reading_distances": [30, 25, 29, 23], "reading_log_probs": [-459.109375, -458.13421630859375, -458.33563232421875, -458.3343811035156], "reading_logprob_std": 0.43116737501886737, "tokenization_path_logprob_std": 165.36944208875815, "trajectory_balance_rmse": 7322.515783928394} +{"canonical_length": 50, "canonical_nll": 0.7883858489990234, "generation_rouge_l_f1": 0.17333333333333337, "generation_token_f1": 0.2533333333333333, "prompt_id": "5debb498a4c42600", "reading_canonical_log_prob": -39.41929244995117, "reading_distances": [33, 32, 30, 30], "reading_log_probs": [-38.66886901855469, -39.194576263427734, -38.99697494506836, -39.35368728637695], "reading_logprob_std": 0.2950429289882606, "tokenization_path_logprob_std": 68.97365216311074, "trajectory_balance_rmse": 1014.13597242015} +{"canonical_length": 34, "canonical_nll": 1.3081263373879826, "generation_rouge_l_f1": 0.3859649122807018, "generation_token_f1": 0.42105263157894735, "prompt_id": "fa77d10032456ab0", "reading_canonical_log_prob": -44.476295471191406, "reading_distances": [27, 28, 32, 27], "reading_log_probs": [-45.301048278808594, -45.2122802734375, -45.65041732788086, -44.30853271484375], "reading_logprob_std": 0.571868551241301, "tokenization_path_logprob_std": 86.71667452800077, "trajectory_balance_rmse": 1017.3971539077556} +{"canonical_length": 8, "canonical_nll": 0.10843963921070099, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "253388f751e77ae9", "reading_canonical_log_prob": -0.8675171136856079, "reading_distances": [48, 51, 45, 55], "reading_log_probs": [-8.231409072875977, -5.85938835144043, -7.774417400360107, -9.193299293518066], "reading_logprob_std": 1.4010734201222905, "tokenization_path_logprob_std": 32.985636577748586, "trajectory_balance_rmse": 153.4023127631528} +{"canonical_length": 51, "canonical_nll": 1.1826764275045956, "generation_rouge_l_f1": 0.2888888888888889, "generation_token_f1": 0.4333333333333334, "prompt_id": "0182330e9a378205", "reading_canonical_log_prob": -60.316497802734375, "reading_distances": [25, 22, 24, 26], "reading_log_probs": [-60.40919876098633, -62.34791564941406, -59.898033142089844, -62.51783752441406], "reading_logprob_std": 1.3341802615273008, "tokenization_path_logprob_std": 114.22814220469915, "trajectory_balance_rmse": 850.6249090804693} +{"canonical_length": 111, "canonical_nll": 1.7463633219401042, "generation_rouge_l_f1": 0.2613065326633166, "generation_token_f1": 0.43216080402010054, "prompt_id": "8709e23cb34a4482", "reading_canonical_log_prob": -193.84632873535156, "reading_distances": [22, 24, 26, 28], "reading_log_probs": [-191.71835327148438, -193.25051879882812, -194.3769073486328, -192.0338897705078], "reading_logprob_std": 1.2163357881907917, "tokenization_path_logprob_std": 77.93032391561566, "trajectory_balance_rmse": 2945.0191997589923} +{"canonical_length": 6, "canonical_nll": 0.40647586186726886, "generation_rouge_l_f1": 0.888888888888889, "generation_token_f1": 0.888888888888889, "prompt_id": "0852cc501290063e", "reading_canonical_log_prob": -2.4388551712036133, "reading_distances": [25, 34, 28, 28], "reading_log_probs": [-2.1217353343963623, -1.7981822490692139, -2.006002187728882, -1.8307846784591675], "reading_logprob_std": 0.15212009107770605, "tokenization_path_logprob_std": 19.407547108547, "trajectory_balance_rmse": 79.2057715786793} +{"canonical_length": 12, "canonical_nll": 1.527944564819336, "generation_rouge_l_f1": 0.7058823529411765, "generation_token_f1": 0.7058823529411765, "prompt_id": "0108b707d4dec974", "reading_canonical_log_prob": -18.33533477783203, "reading_distances": [35, 40, 31, 43], "reading_log_probs": [-19.145265579223633, -19.381301879882812, -19.177610397338867, -19.705503463745117], "reading_logprob_std": 0.25753539988715746, "tokenization_path_logprob_std": 79.43904585440396, "trajectory_balance_rmse": 329.62313111510565} +{"canonical_length": 104, "canonical_nll": 1.6722503075232873, "generation_rouge_l_f1": 0.1793478260869565, "generation_token_f1": 0.31521739130434784, "prompt_id": "5ffba76c8fed8b0d", "reading_canonical_log_prob": -173.91403198242188, "reading_distances": [18, 19, 19, 19], "reading_log_probs": [-176.2208709716797, -177.83880615234375, -176.65077209472656, -175.6761474609375], "reading_logprob_std": 0.919132908504193, "tokenization_path_logprob_std": 17.631851497429533, "trajectory_balance_rmse": 2450.3181014285783} +{"canonical_length": 245, "canonical_nll": 1.0417768205915179, "generation_rouge_l_f1": 0.37675350701402804, "generation_token_f1": 0.6292585170340681, "prompt_id": "66d201ed6635605c", "reading_canonical_log_prob": -255.23532104492188, "reading_distances": [43, 40, 50, 36], "reading_log_probs": [-255.61599731445312, -256.5694885253906, -258.4273681640625, -260.72637939453125], "reading_logprob_std": 2.253624259485701, "tokenization_path_logprob_std": 307.868619418622, "trajectory_balance_rmse": 5335.001147250385} +{"canonical_length": 62, "canonical_nll": 1.0527418813397806, "generation_rouge_l_f1": 0.23529411764705882, "generation_token_f1": 0.23529411764705882, "prompt_id": "d9f443de104a8944", "reading_canonical_log_prob": -65.2699966430664, "reading_distances": [22, 17, 18, 17], "reading_log_probs": [-66.15792846679688, -66.0734634399414, -65.9964370727539, -65.57041931152344], "reading_logprob_std": 0.26122442584609484, "tokenization_path_logprob_std": 27.765215578272418, "trajectory_balance_rmse": 621.1601636690608} +{"canonical_length": 19, "canonical_nll": 1.880444376092208, "generation_rouge_l_f1": 0.3333333333333333, "generation_token_f1": 0.3333333333333333, "prompt_id": "d416f7f8cc539e7b", "reading_canonical_log_prob": -35.72844314575195, "reading_distances": [23, 30, 33, 28], "reading_log_probs": [-31.196090698242188, -32.2354736328125, -33.78999328613281, -32.00131607055664], "reading_logprob_std": 1.0850408102310662, "tokenization_path_logprob_std": 99.61471238718887, "trajectory_balance_rmse": 542.3776816675484} +{"canonical_length": 12, "canonical_nll": 1.6698217391967773, "generation_rouge_l_f1": 0.13333333333333333, "generation_token_f1": 0.13333333333333333, "prompt_id": "111f88e5ddbba7a9", "reading_canonical_log_prob": -20.037860870361328, "reading_distances": [24, 22, 18, 19], "reading_log_probs": [-29.28639793395996, -30.078922271728516, -31.052688598632812, -28.59440040588379], "reading_logprob_std": 1.0575894361361964, "tokenization_path_logprob_std": 32.80165408455897, "trajectory_balance_rmse": 271.27697397491363} +{"canonical_length": 11, "canonical_nll": 0.8701637441461737, "generation_rouge_l_f1": 0.18181818181818182, "generation_token_f1": 0.18181818181818182, "prompt_id": "47566df03369ba07", "reading_canonical_log_prob": -9.57180118560791, "reading_distances": [50, 40, 54, 48], "reading_log_probs": [-11.877717971801758, -13.729880332946777, -10.241931915283203, -14.453170776367188], "reading_logprob_std": 1.896547840995489, "tokenization_path_logprob_std": 34.51777530345129, "trajectory_balance_rmse": 203.90712126040972} +{"canonical_length": 26, "canonical_nll": 0.6286644568810096, "generation_rouge_l_f1": 0.2839506172839506, "generation_token_f1": 0.29629629629629634, "prompt_id": "0ad3db1386b28ee8", "reading_canonical_log_prob": -16.34527587890625, "reading_distances": [30, 31, 33, 37], "reading_log_probs": [-18.155574798583984, -18.36203956604004, -18.496318817138672, -18.113035202026367], "reading_logprob_std": 0.1796951640612014, "tokenization_path_logprob_std": 41.6199297534254, "trajectory_balance_rmse": 651.3746135834093} +{"canonical_length": 104, "canonical_nll": 1.4892410865196815, "generation_rouge_l_f1": 0.19801980198019803, "generation_token_f1": 0.41584158415841577, "prompt_id": "136b8765ce15fa6a", "reading_canonical_log_prob": -154.88107299804688, "reading_distances": [18, 21, 18, 21], "reading_log_probs": [-162.75523376464844, -155.9186553955078, -161.47427368164062, -156.04470825195312], "reading_logprob_std": 3.579708784795262, "tokenization_path_logprob_std": 55.10993166624259, "trajectory_balance_rmse": 2258.6049983405283} +{"canonical_length": 35, "canonical_nll": 0.8207632882254464, "generation_rouge_l_f1": 0.38095238095238093, "generation_token_f1": 0.5333333333333333, "prompt_id": "92e8091e1c9b04af", "reading_canonical_log_prob": -28.726715087890625, "reading_distances": [28, 31, 24, 38], "reading_log_probs": [-27.551185607910156, -28.111181259155273, -28.159236907958984, -28.272411346435547], "reading_logprob_std": 0.32205085245020487, "tokenization_path_logprob_std": 115.30457546777168, "trajectory_balance_rmse": 726.4915291655894} +{"canonical_length": 20, "canonical_nll": 2.088039016723633, "generation_rouge_l_f1": 0.16666666666666666, "generation_token_f1": 0.25, "prompt_id": "3bfad973aaa202e7", "reading_canonical_log_prob": -41.760780334472656, "reading_distances": [21, 15, 21, 17], "reading_log_probs": [-39.80294418334961, -39.4382209777832, -39.44862747192383, -40.15514373779297], "reading_logprob_std": 0.34105958029137046, "tokenization_path_logprob_std": 127.88996621897395, "trajectory_balance_rmse": 529.1473809531403} +{"canonical_length": 292, "canonical_nll": 0.9540346746575342, "generation_rouge_l_f1": 0.32958801498127344, "generation_token_f1": 0.5468164794007491, "prompt_id": "be5dbfe122219a1b", "reading_canonical_log_prob": -278.578125, "reading_distances": [18, 28, 24, 21], "reading_log_probs": [-281.52081298828125, -281.2203369140625, -282.28582763671875, -281.20819091796875], "reading_logprob_std": 0.5057984401941689, "tokenization_path_logprob_std": 165.0195585056613, "trajectory_balance_rmse": 6112.741312623412} +{"canonical_length": 39, "canonical_nll": 1.115949190579928, "generation_rouge_l_f1": 0.16428571428571428, "generation_token_f1": 0.20714285714285716, "prompt_id": "6acd370678d1b813", "reading_canonical_log_prob": -43.52201843261719, "reading_distances": [16, 16, 14, 17], "reading_log_probs": [-38.875675201416016, -35.40040969848633, -39.53148651123047, -38.06023406982422], "reading_logprob_std": 1.8137793897119108, "tokenization_path_logprob_std": 37.85072976205713, "trajectory_balance_rmse": 390.0394973967718} +{"canonical_length": 197, "canonical_nll": 1.3818941842480965, "generation_rouge_l_f1": 0.29081632653061223, "generation_token_f1": 0.5204081632653061, "prompt_id": "2bb939d56ab15ee3", "reading_canonical_log_prob": -272.233154296875, "reading_distances": [22, 20, 21, 20], "reading_log_probs": [-271.10552978515625, -271.0062561035156, -270.1966247558594, -270.9591064453125], "reading_logprob_std": 0.4179820317834921, "tokenization_path_logprob_std": 325.5328893419422, "trajectory_balance_rmse": 4590.104125357642} +{"canonical_length": 14, "canonical_nll": 1.6067022596086775, "generation_rouge_l_f1": 0.4137931034482759, "generation_token_f1": 0.4137931034482759, "prompt_id": "9d34cf3c82e91821", "reading_canonical_log_prob": -22.493831634521484, "reading_distances": [48, 30, 40, 42], "reading_log_probs": [-23.16159439086914, -22.162269592285156, -22.641265869140625, -22.148483276367188], "reading_logprob_std": 0.4802992238703039, "tokenization_path_logprob_std": 53.57522906312776, "trajectory_balance_rmse": 224.22827440644986} +{"canonical_length": 13, "canonical_nll": 0.23183175233694223, "generation_rouge_l_f1": 0.7826086956521738, "generation_token_f1": 0.7826086956521738, "prompt_id": "b8a06c7365e05ce4", "reading_canonical_log_prob": -3.013812780380249, "reading_distances": [30, 24, 29, 22], "reading_log_probs": [-2.8664355278015137, -3.520653009414673, -3.253121852874756, -4.645227432250977], "reading_logprob_std": 0.7646255054452831, "tokenization_path_logprob_std": 29.376520982966206, "trajectory_balance_rmse": 291.91590942655057} +{"canonical_length": 44, "canonical_nll": 1.6481633619828657, "generation_rouge_l_f1": 0.19718309859154928, "generation_token_f1": 0.19718309859154928, "prompt_id": "3c94fb62a787deaa", "reading_canonical_log_prob": -72.5191879272461, "reading_distances": [26, 19, 21, 24], "reading_log_probs": [-81.42095947265625, -75.17581939697266, -79.59194946289062, -80.1427230834961], "reading_logprob_std": 2.715029604178147, "tokenization_path_logprob_std": 65.42215050406497, "trajectory_balance_rmse": 832.7324804411692} +{"canonical_length": 230, "canonical_nll": 1.0207841956097148, "generation_rouge_l_f1": 0.32719836400818003, "generation_token_f1": 0.5562372188139061, "prompt_id": "84f970de38e4a1fd", "reading_canonical_log_prob": -234.78036499023438, "reading_distances": [25, 25, 21, 29], "reading_log_probs": [-236.93359375, -234.27606201171875, -238.1096954345703, -236.15870666503906], "reading_logprob_std": 1.6096995940541658, "tokenization_path_logprob_std": 188.58464696195975, "trajectory_balance_rmse": 4068.9262029361876} +{"canonical_length": 43, "canonical_nll": 1.5860643608625544, "generation_rouge_l_f1": 0.12322274881516589, "generation_token_f1": 0.21800947867298578, "prompt_id": "b9d9b86bb406a853", "reading_canonical_log_prob": -68.20076751708984, "reading_distances": [16, 18, 17, 12], "reading_log_probs": [-72.92666625976562, -73.14115905761719, -74.59864807128906, -71.42335510253906], "reading_logprob_std": 1.2993294701590863, "tokenization_path_logprob_std": 64.09130147838117, "trajectory_balance_rmse": 996.3568490158781} +{"canonical_length": 217, "canonical_nll": 0.8948305894702261, "generation_rouge_l_f1": 0.2948717948717949, "generation_token_f1": 0.5299145299145298, "prompt_id": "2dfd04cd2fcb5555", "reading_canonical_log_prob": -194.17823791503906, "reading_distances": [42, 36, 40, 40], "reading_log_probs": [-192.95155334472656, -188.0241241455078, -188.3852081298828, -188.8133544921875], "reading_logprob_std": 2.2947827124632547, "tokenization_path_logprob_std": 187.9069415315056, "trajectory_balance_rmse": 5043.165829724004} +{"canonical_length": 193, "canonical_nll": 1.0405417328671471, "generation_rouge_l_f1": 0.2863741339491917, "generation_token_f1": 0.4942263279445727, "prompt_id": "daff264becf42549", "reading_canonical_log_prob": -200.82455444335938, "reading_distances": [16, 11, 14, 14], "reading_log_probs": [-199.56649780273438, -199.1583251953125, -198.87213134765625, -199.59674072265625], "reading_logprob_std": 0.34747159655898646, "tokenization_path_logprob_std": 232.45949852020001, "trajectory_balance_rmse": 4251.399916522014} +{"canonical_length": 254, "canonical_nll": 1.2018070971871924, "generation_rouge_l_f1": 0.1976047904191617, "generation_token_f1": 0.31137724550898205, "prompt_id": "f80fc3edc248f324", "reading_canonical_log_prob": -305.2590026855469, "reading_distances": [30, 28, 30, 27], "reading_log_probs": [-307.7042236328125, -307.7628173828125, -308.651611328125, -307.0350341796875], "reading_logprob_std": 0.6634339920549552, "tokenization_path_logprob_std": 166.49920170651572, "trajectory_balance_rmse": 6035.980318119821} +{"canonical_length": 72, "canonical_nll": 1.7487432691786025, "generation_rouge_l_f1": 0.29197080291970806, "generation_token_f1": 0.39416058394160586, "prompt_id": "7da5b774aaa5c557", "reading_canonical_log_prob": -125.90951538085938, "reading_distances": [25, 18, 24, 29], "reading_log_probs": [-121.36181640625, -120.76708221435547, -121.67106628417969, -121.71287536621094], "reading_logprob_std": 0.4364674981513303, "tokenization_path_logprob_std": 47.386523287058196, "trajectory_balance_rmse": 1768.3642229822124} +{"canonical_length": 138, "canonical_nll": 1.5598493935405344, "generation_rouge_l_f1": 0.20151133501259444, "generation_token_f1": 0.36272040302267006, "prompt_id": "0e6878e78767cf2b", "reading_canonical_log_prob": -215.25921630859375, "reading_distances": [16, 20, 24, 22], "reading_log_probs": [-214.3884735107422, -213.35400390625, -215.7691650390625, -217.87179565429688], "reading_logprob_std": 1.953084541075595, "tokenization_path_logprob_std": 73.81211999547195, "trajectory_balance_rmse": 2743.8045255538896} +{"canonical_length": 152, "canonical_nll": 1.251947101793791, "generation_rouge_l_f1": 0.20240963855421687, "generation_token_f1": 0.3759036144578314, "prompt_id": "0ac01e4320c0e1b4", "reading_canonical_log_prob": -190.29595947265625, "reading_distances": [29, 28, 26, 29], "reading_log_probs": [-185.22540283203125, -188.7044219970703, -187.3079833984375, -186.94091796875], "reading_logprob_std": 1.4311543300752727, "tokenization_path_logprob_std": 58.440693085517275, "trajectory_balance_rmse": 3291.764880294389} +{"canonical_length": 259, "canonical_nll": 1.2079108396552245, "generation_rouge_l_f1": 0.28346456692913385, "generation_token_f1": 0.5118110236220472, "prompt_id": "cc034e8024f676e6", "reading_canonical_log_prob": -312.8489074707031, "reading_distances": [23, 15, 30, 24], "reading_log_probs": [-310.5307922363281, -310.48968505859375, -310.77587890625, -312.7872314453125], "reading_logprob_std": 1.1014929464757726, "tokenization_path_logprob_std": 244.9543787249377, "trajectory_balance_rmse": 5548.048739539953} +{"canonical_length": 253, "canonical_nll": 1.385872218919837, "generation_rouge_l_f1": 0.22265625, "generation_token_f1": 0.41406250000000006, "prompt_id": "e0ae35419a54967e", "reading_canonical_log_prob": -350.62567138671875, "reading_distances": [23, 23, 27, 30], "reading_log_probs": [-342.8359680175781, -344.998779296875, -342.4482421875, -348.0872802734375], "reading_logprob_std": 2.5859748761489447, "tokenization_path_logprob_std": 170.91684401870822, "trajectory_balance_rmse": 5003.25345134008} +{"canonical_length": 6, "canonical_nll": 1.137926419576009, "generation_rouge_l_f1": 0.6, "generation_token_f1": 0.6, "prompt_id": "9ad445292250739f", "reading_canonical_log_prob": -6.827558517456055, "reading_distances": [27, 19, 18, 27], "reading_log_probs": [-7.65110445022583, -7.389617443084717, -5.8412322998046875, -6.590659141540527], "reading_logprob_std": 0.8198591555432546, "tokenization_path_logprob_std": 31.581429211387842, "trajectory_balance_rmse": 151.10921345322714} +{"canonical_length": 21, "canonical_nll": 1.1143484569731212, "generation_rouge_l_f1": 0.12682926829268293, "generation_token_f1": 0.14634146341463414, "prompt_id": "1a95335c2e86bbe5", "reading_canonical_log_prob": -23.401317596435547, "reading_distances": [17, 23, 23, 26], "reading_log_probs": [-29.119035720825195, -28.901416778564453, -28.496978759765625, -27.70919418334961], "reading_logprob_std": 0.6209894922658096, "tokenization_path_logprob_std": 10.524876097771255, "trajectory_balance_rmse": 474.2268036667431} +{"canonical_length": 25, "canonical_nll": 0.36661544799804685, "generation_rouge_l_f1": 0.1259259259259259, "generation_token_f1": 0.13333333333333336, "prompt_id": "8fe26f83cb600778", "reading_canonical_log_prob": -9.165386199951172, "reading_distances": [22, 23, 25, 23], "reading_log_probs": [-8.409688949584961, -7.771369457244873, -8.727538108825684, -8.023544311523438], "reading_logprob_std": 0.42141087016031864, "tokenization_path_logprob_std": 38.74484487139711, "trajectory_balance_rmse": 215.46151399995688} +{"canonical_length": 17, "canonical_nll": 1.3037706262925093, "generation_rouge_l_f1": 0.055299539170506916, "generation_token_f1": 0.055299539170506916, "prompt_id": "8bd44913279a5599", "reading_canonical_log_prob": -22.164100646972656, "reading_distances": [14, 12, 11, 13], "reading_log_probs": [-35.601768493652344, -36.50532913208008, -33.44866943359375, -33.05060577392578], "reading_logprob_std": 1.6682681959430128, "tokenization_path_logprob_std": 3.382554395835668, "trajectory_balance_rmse": 255.78773541869657} +{"canonical_length": 355, "canonical_nll": 1.2884119168133803, "generation_rouge_l_f1": 0.274442538593482, "generation_token_f1": 0.5248713550600344, "prompt_id": "e131aed6216cf197", "reading_canonical_log_prob": -457.38623046875, "reading_distances": [45, 39, 43, 45], "reading_log_probs": [-455.636962890625, -453.677978515625, -454.562744140625, -452.94976806640625], "reading_logprob_std": 1.1592876481266734, "tokenization_path_logprob_std": 144.1693254193502, "trajectory_balance_rmse": 7761.51136118236} +{"canonical_length": 46, "canonical_nll": 1.2311978547469429, "generation_rouge_l_f1": 0.10062893081761007, "generation_token_f1": 0.15723270440251572, "prompt_id": "87d0cec84789988b", "reading_canonical_log_prob": -56.635101318359375, "reading_distances": [15, 21, 25, 22], "reading_log_probs": [-56.075801849365234, -54.22735595703125, -53.88413619995117, -55.87786102294922], "reading_logprob_std": 1.1208715563710447, "tokenization_path_logprob_std": 132.0783785913574, "trajectory_balance_rmse": 1070.9954039080342} +{"canonical_length": 138, "canonical_nll": 0.8922261224276777, "generation_rouge_l_f1": 0.2542372881355932, "generation_token_f1": 0.4745762711864407, "prompt_id": "d60af44d5acef8d0", "reading_canonical_log_prob": -123.12720489501953, "reading_distances": [28, 34, 31, 35], "reading_log_probs": [-120.59376525878906, -121.2965316772461, -119.35184478759766, -119.8901138305664], "reading_logprob_std": 0.8456222683603639, "tokenization_path_logprob_std": 147.84004397959154, "trajectory_balance_rmse": 3101.864662294682} +{"canonical_length": 35, "canonical_nll": 1.755157252720424, "generation_rouge_l_f1": 0.21505376344086025, "generation_token_f1": 0.27956989247311825, "prompt_id": "a9f740ec35ab9678", "reading_canonical_log_prob": -61.430503845214844, "reading_distances": [14, 15, 17, 11], "reading_log_probs": [-67.17247772216797, -64.05347442626953, -67.46786499023438, -65.91178894042969], "reading_logprob_std": 1.5528881648927595, "tokenization_path_logprob_std": 108.15673774623285, "trajectory_balance_rmse": 788.2455644887687} +{"canonical_length": 257, "canonical_nll": 1.3145550085876703, "generation_rouge_l_f1": 0.1772151898734177, "generation_token_f1": 0.3942133815551537, "prompt_id": "a25427203b062343", "reading_canonical_log_prob": -337.84063720703125, "reading_distances": [32, 33, 30, 31], "reading_log_probs": [-335.8883361816406, -337.8338317871094, -336.6732482910156, -338.06036376953125], "reading_logprob_std": 1.018213474428885, "tokenization_path_logprob_std": 234.73947813597232, "trajectory_balance_rmse": 5480.376935498411} +{"canonical_length": 28, "canonical_nll": 1.1103965895516532, "generation_rouge_l_f1": 0.2903225806451613, "generation_token_f1": 0.4516129032258065, "prompt_id": "b0bf00b191d5ab96", "reading_canonical_log_prob": -31.09110450744629, "reading_distances": [26, 29, 27, 31], "reading_log_probs": [-31.04715347290039, -31.593351364135742, -31.329069137573242, -31.304473876953125], "reading_logprob_std": 0.22321936024205918, "tokenization_path_logprob_std": 73.32592403693029, "trajectory_balance_rmse": 797.1194413224438} +{"canonical_length": 31, "canonical_nll": 0.8968448023642263, "generation_rouge_l_f1": 0.2016806722689076, "generation_token_f1": 0.2689075630252101, "prompt_id": "77df38b1cb264cc8", "reading_canonical_log_prob": -27.802188873291016, "reading_distances": [8, 10, 12, 10], "reading_log_probs": [-29.10572052001953, -27.838905334472656, -26.634031295776367, -28.408336639404297], "reading_logprob_std": 1.0458069262974636, "tokenization_path_logprob_std": 80.01996040124217, "trajectory_balance_rmse": 739.1068750881352} +{"canonical_length": 54, "canonical_nll": 1.5446270130298756, "generation_rouge_l_f1": 0.2807017543859649, "generation_token_f1": 0.42105263157894735, "prompt_id": "93f9306a97ca7347", "reading_canonical_log_prob": -83.40985870361328, "reading_distances": [13, 14, 13, 13], "reading_log_probs": [-83.02438354492188, -83.56892395019531, -83.57717895507812, -83.54347229003906], "reading_logprob_std": 0.26978575066629146, "tokenization_path_logprob_std": 111.33771071258184, "trajectory_balance_rmse": 1162.6775663697508} +{"canonical_length": 31, "canonical_nll": 1.2693040909305695, "generation_rouge_l_f1": 0.27027027027027023, "generation_token_f1": 0.27027027027027023, "prompt_id": "1cc77327b9384870", "reading_canonical_log_prob": -39.348426818847656, "reading_distances": [26, 23, 26, 19], "reading_log_probs": [-38.363319396972656, -38.983551025390625, -39.70801544189453, -38.967018127441406], "reading_logprob_std": 0.5501171035038667, "tokenization_path_logprob_std": 69.65771517173356, "trajectory_balance_rmse": 688.9741920535535} +{"canonical_length": 182, "canonical_nll": 1.432430393093235, "generation_rouge_l_f1": 0.21809744779582366, "generation_token_f1": 0.4408352668213457, "prompt_id": "0d40240cbf3216d7", "reading_canonical_log_prob": -260.70233154296875, "reading_distances": [27, 22, 27, 25], "reading_log_probs": [-261.1241455078125, -260.953125, -258.7293701171875, -261.6076354980469], "reading_logprob_std": 1.2798385682652835, "tokenization_path_logprob_std": 175.06559405684715, "trajectory_balance_rmse": 3670.4029426020206} +{"canonical_length": 136, "canonical_nll": 1.7673314038444967, "generation_rouge_l_f1": 0.14418604651162792, "generation_token_f1": 0.26046511627906976, "prompt_id": "c887cead4589c177", "reading_canonical_log_prob": -240.35707092285156, "reading_distances": [12, 12, 19, 17], "reading_log_probs": [-246.533447265625, -245.16024780273438, -243.92117309570312, -247.82339477539062], "reading_logprob_std": 1.6889002745048458, "tokenization_path_logprob_std": 232.00455105100374, "trajectory_balance_rmse": 3292.6941406511255} +{"canonical_length": 169, "canonical_nll": 2.1225688866609653, "generation_rouge_l_f1": 0.2716049382716049, "generation_token_f1": 0.4246913580246913, "prompt_id": "5686c4dfdf197ade", "reading_canonical_log_prob": -358.7141418457031, "reading_distances": [13, 11, 11, 10], "reading_log_probs": [-357.4546813964844, -356.8370056152344, -357.44830322265625, -359.01336669921875], "reading_logprob_std": 0.9296379198015565, "tokenization_path_logprob_std": 161.08901694449057, "trajectory_balance_rmse": 3363.8105069870007} +{"canonical_length": 348, "canonical_nll": 1.0871231254489941, "generation_rouge_l_f1": 0.23414634146341462, "generation_token_f1": 0.4617886178861789, "prompt_id": "497e17449fccf12b", "reading_canonical_log_prob": -378.31884765625, "reading_distances": [35, 24, 33, 27], "reading_log_probs": [-377.79486083984375, -378.8181457519531, -377.1076965332031, -377.10260009765625], "reading_logprob_std": 0.8096959885455071, "tokenization_path_logprob_std": 320.784147778736, "trajectory_balance_rmse": 7810.374819900621} +{"canonical_length": 38, "canonical_nll": 1.0286868245978105, "generation_rouge_l_f1": 0.13103448275862067, "generation_token_f1": 0.16551724137931034, "prompt_id": "faaabffbce1c5a62", "reading_canonical_log_prob": -39.0900993347168, "reading_distances": [35, 38, 39, 31], "reading_log_probs": [-40.543800354003906, -40.45766067504883, -39.61211395263672, -40.42186737060547], "reading_logprob_std": 0.4341906518224815, "tokenization_path_logprob_std": 55.49018681670118, "trajectory_balance_rmse": 863.9491474465608} +{"canonical_length": 76, "canonical_nll": 0.40558767318725586, "generation_rouge_l_f1": 0.07801418439716312, "generation_token_f1": 0.10638297872340427, "prompt_id": "bf9932d84bc4ffec", "reading_canonical_log_prob": -30.824663162231445, "reading_distances": [21, 18, 17, 16], "reading_log_probs": [-31.242650985717773, -31.854387283325195, -31.228553771972656, -33.31706237792969], "reading_logprob_std": 0.9819435178448009, "tokenization_path_logprob_std": 14.704783540623575, "trajectory_balance_rmse": 365.97583937089854} +{"canonical_length": 165, "canonical_nll": 1.804565799597538, "generation_rouge_l_f1": 0.19718309859154928, "generation_token_f1": 0.41784037558685444, "prompt_id": "d69cbb797502e917", "reading_canonical_log_prob": -297.75335693359375, "reading_distances": [18, 20, 13, 19], "reading_log_probs": [-300.9651794433594, -298.0176086425781, -298.2973327636719, -297.3612060546875], "reading_logprob_std": 1.5858658748135557, "tokenization_path_logprob_std": 155.8524156468969, "trajectory_balance_rmse": 4233.015318434934} +{"canonical_length": 119, "canonical_nll": 1.4741323775603992, "generation_rouge_l_f1": 0.215633423180593, "generation_token_f1": 0.3557951482479784, "prompt_id": "98ec80f0ace1a8dc", "reading_canonical_log_prob": -175.4217529296875, "reading_distances": [34, 25, 35, 36], "reading_log_probs": [-172.17047119140625, -172.2591094970703, -171.8476104736328, -172.39208984375], "reading_logprob_std": 0.23178359568699533, "tokenization_path_logprob_std": 202.88143986286798, "trajectory_balance_rmse": 2838.3054990345418} +{"canonical_length": 14, "canonical_nll": 1.3277222769601005, "generation_rouge_l_f1": 0.04395604395604395, "generation_token_f1": 0.05128205128205128, "prompt_id": "e2aae622b59df148", "reading_canonical_log_prob": -18.588111877441406, "reading_distances": [33, 23, 26, 32], "reading_log_probs": [-21.35467529296875, -21.38119125366211, -21.58734130859375, -21.21371841430664], "reading_logprob_std": 0.15406829569113356, "tokenization_path_logprob_std": 47.1805961702707, "trajectory_balance_rmse": 261.9310881558975} +{"canonical_length": 10, "canonical_nll": 0.7958105087280274, "generation_rouge_l_f1": 0.5, "generation_token_f1": 0.5, "prompt_id": "aefa9cb1b8d05c05", "reading_canonical_log_prob": -7.958105087280273, "reading_distances": [33, 28, 33, 26], "reading_log_probs": [-9.410355567932129, -9.674020767211914, -9.48410701751709, -9.733522415161133], "reading_logprob_std": 0.15308253323747276, "tokenization_path_logprob_std": 31.48503834639515, "trajectory_balance_rmse": 303.28702710718926} +{"canonical_length": 121, "canonical_nll": 1.0646373654200025, "generation_rouge_l_f1": 0.26178010471204183, "generation_token_f1": 0.41361256544502617, "prompt_id": "7467b21b2ac04b77", "reading_canonical_log_prob": -128.8211212158203, "reading_distances": [11, 5, 12, 9], "reading_log_probs": [-126.98417663574219, -126.49088287353516, -125.4969253540039, -125.14608764648438], "reading_logprob_std": 0.8540754793754166, "tokenization_path_logprob_std": 46.96946272665659, "trajectory_balance_rmse": 2727.2288400506986} +{"canonical_length": 92, "canonical_nll": 1.3601956574813179, "generation_rouge_l_f1": 0.1414141414141414, "generation_token_f1": 0.1616161616161616, "prompt_id": "80fb331e54c070fd", "reading_canonical_log_prob": -125.13800048828125, "reading_distances": [18, 21, 19, 23], "reading_log_probs": [-126.58624267578125, -125.45307922363281, -124.56430053710938, -124.10154724121094], "reading_logprob_std": 1.0945595647713326, "tokenization_path_logprob_std": 89.34033352749708, "trajectory_balance_rmse": 1697.6838832273645} +{"canonical_length": 288, "canonical_nll": 1.2026641633775499, "generation_rouge_l_f1": 0.2794117647058824, "generation_token_f1": 0.5404411764705882, "prompt_id": "1baa92d9506c1060", "reading_canonical_log_prob": -346.3672790527344, "reading_distances": [17, 27, 17, 16], "reading_log_probs": [-344.51312255859375, -344.0140075683594, -345.79620361328125, -344.31256103515625], "reading_logprob_std": 0.7853974527670274, "tokenization_path_logprob_std": 64.50972145261768, "trajectory_balance_rmse": 6694.46130276917} +{"canonical_length": 232, "canonical_nll": 1.4373489906047952, "generation_rouge_l_f1": 0.2688391038696538, "generation_token_f1": 0.505091649694501, "prompt_id": "23fd1d0b807b2a3d", "reading_canonical_log_prob": -333.4649658203125, "reading_distances": [27, 26, 23, 21], "reading_log_probs": [-332.8642578125, -331.13250732421875, -334.49786376953125, -332.137451171875], "reading_logprob_std": 1.4172437993481628, "tokenization_path_logprob_std": 53.6026972710018, "trajectory_balance_rmse": 5100.156841207104} +{"canonical_length": 23, "canonical_nll": 1.0815947159476902, "generation_rouge_l_f1": 0.5106382978723404, "generation_token_f1": 0.5531914893617021, "prompt_id": "e7e3fd3a1ae85dcc", "reading_canonical_log_prob": -24.876678466796875, "reading_distances": [20, 26, 14, 28], "reading_log_probs": [-26.330732345581055, -25.451515197753906, -25.595186233520508, -25.265945434570312], "reading_logprob_std": 0.4664849987834654, "tokenization_path_logprob_std": 59.85394320842439, "trajectory_balance_rmse": 511.7956918095753} +{"canonical_length": 11, "canonical_nll": 0.7823701338334517, "generation_rouge_l_f1": 0.7368421052631577, "generation_token_f1": 0.7368421052631577, "prompt_id": "7a61f927f0ede558", "reading_canonical_log_prob": -8.606071472167969, "reading_distances": [24, 18, 17, 20], "reading_log_probs": [-9.727797508239746, -9.686546325683594, -9.96523380279541, -9.214486122131348], "reading_logprob_std": 0.3143373796366913, "tokenization_path_logprob_std": 30.774041479046247, "trajectory_balance_rmse": 259.6209153917995} +{"canonical_length": 244, "canonical_nll": 1.6493772913198002, "generation_rouge_l_f1": 0.3062015503875969, "generation_token_f1": 0.5658914728682171, "prompt_id": "d3428656e283f9ef", "reading_canonical_log_prob": -402.44805908203125, "reading_distances": [27, 34, 40, 40], "reading_log_probs": [-403.2059326171875, -406.970458984375, -405.4747314453125, -406.0010986328125], "reading_logprob_std": 1.5965061521184372, "tokenization_path_logprob_std": 61.58013424293117, "trajectory_balance_rmse": 6300.197288307756} +{"canonical_length": 18, "canonical_nll": 0.6887931293911405, "generation_rouge_l_f1": 0.0796812749003984, "generation_token_f1": 0.0956175298804781, "prompt_id": "f6a9e161c915b813", "reading_canonical_log_prob": -12.398276329040527, "reading_distances": [12, 15, 12, 19], "reading_log_probs": [-9.61874771118164, -10.597901344299316, -10.87665843963623, -9.328222274780273], "reading_logprob_std": 0.7479377916207418, "tokenization_path_logprob_std": 24.99559557226952, "trajectory_balance_rmse": 356.93815686562743} +{"canonical_length": 4, "canonical_nll": 1.6075596809387207, "generation_rouge_l_f1": 0.6666666666666666, "generation_token_f1": 0.6666666666666666, "prompt_id": "60d17bcb4782ac31", "reading_canonical_log_prob": -6.430238723754883, "reading_distances": [14, 12, 12, 19], "reading_log_probs": [-6.428036689758301, -5.994403839111328, -5.770247459411621, -5.66336727142334], "reading_logprob_std": 0.33871055814025025, "tokenization_path_logprob_std": 27.351812691083822, "trajectory_balance_rmse": 64.18451991363159} +{"canonical_length": 12, "canonical_nll": 1.2303245862325032, "generation_rouge_l_f1": 0.06920415224913495, "generation_token_f1": 0.08304498269896193, "prompt_id": "469967c1d14dd489", "reading_canonical_log_prob": -14.763895034790039, "reading_distances": [26, 23, 30, 26], "reading_log_probs": [-17.543245315551758, -16.548583984375, -17.7437801361084, -16.330156326293945], "reading_logprob_std": 0.705672671549398, "tokenization_path_logprob_std": 36.20334125161454, "trajectory_balance_rmse": 218.03178841420166} +{"canonical_length": 269, "canonical_nll": 1.2409143837854322, "generation_rouge_l_f1": 0.25828970331588136, "generation_token_f1": 0.44677137870855144, "prompt_id": "812c384ebd7a8b2e", "reading_canonical_log_prob": -333.80596923828125, "reading_distances": [30, 30, 26, 37], "reading_log_probs": [-327.1005859375, -325.92120361328125, -325.273193359375, -325.30096435546875], "reading_logprob_std": 0.8550988926289022, "tokenization_path_logprob_std": 104.92843967597368, "trajectory_balance_rmse": 4608.830371991931} +{"canonical_length": 99, "canonical_nll": 1.128062392726089, "generation_rouge_l_f1": 0.1945945945945946, "generation_token_f1": 0.38378378378378375, "prompt_id": "40633ecaff73b957", "reading_canonical_log_prob": -111.67817687988281, "reading_distances": [19, 26, 23, 15], "reading_log_probs": [-114.62617492675781, -113.82861328125, -115.51712799072266, -117.26730346679688], "reading_logprob_std": 1.476037720241703, "tokenization_path_logprob_std": 210.272237080438, "trajectory_balance_rmse": 2561.171677994616} +{"canonical_length": 210, "canonical_nll": 1.2411426362537203, "generation_rouge_l_f1": 0.28947368421052627, "generation_token_f1": 0.4912280701754386, "prompt_id": "cd8cb6a38db69f38", "reading_canonical_log_prob": -260.63995361328125, "reading_distances": [16, 23, 20, 20], "reading_log_probs": [-263.07354736328125, -262.6063537597656, -261.49334716796875, -262.0152587890625], "reading_logprob_std": 0.6889512503774646, "tokenization_path_logprob_std": 122.23641454739263, "trajectory_balance_rmse": 4624.995968924691} +{"canonical_length": 105, "canonical_nll": 1.3394587925502233, "generation_rouge_l_f1": 0.06666666666666667, "generation_token_f1": 0.08888888888888889, "prompt_id": "2de9ac9d5457eaa3", "reading_canonical_log_prob": -140.64317321777344, "reading_distances": [30, 28, 31, 30], "reading_log_probs": [-149.20675659179688, -149.0194091796875, -150.9468536376953, -149.830810546875], "reading_logprob_std": 0.8694560856633171, "tokenization_path_logprob_std": 36.59433935788841, "trajectory_balance_rmse": 1188.330294132759} +{"canonical_length": 99, "canonical_nll": 1.2670719647648359, "generation_rouge_l_f1": 0.18233618233618235, "generation_token_f1": 0.31908831908831914, "prompt_id": "0b44770e70f57065", "reading_canonical_log_prob": -125.44012451171875, "reading_distances": [28, 23, 29, 32], "reading_log_probs": [-122.75703430175781, -125.09931945800781, -121.48928833007812, -124.16984558105469], "reading_logprob_std": 1.5856425858604408, "tokenization_path_logprob_std": 64.32240819449488, "trajectory_balance_rmse": 2214.8335548125237} +{"canonical_length": 111, "canonical_nll": 1.6867901226421733, "generation_rouge_l_f1": 0.18836565096952906, "generation_token_f1": 0.3379501385041552, "prompt_id": "f5df13177d4b7c6c", "reading_canonical_log_prob": -187.23370361328125, "reading_distances": [16, 19, 15, 16], "reading_log_probs": [-187.0572509765625, -185.74542236328125, -189.52340698242188, -186.8809051513672], "reading_logprob_std": 1.5911008368828958, "tokenization_path_logprob_std": 107.38239962830447, "trajectory_balance_rmse": 2426.0422075850825} +{"canonical_length": 129, "canonical_nll": 1.8075883259144865, "generation_rouge_l_f1": 0.22885572139303484, "generation_token_f1": 0.3930348258706468, "prompt_id": "b85464f813712359", "reading_canonical_log_prob": -233.17889404296875, "reading_distances": [41, 46, 44, 43], "reading_log_probs": [-231.89520263671875, -235.20806884765625, -233.65235900878906, -232.83331298828125], "reading_logprob_std": 1.4045541094739271, "tokenization_path_logprob_std": 162.0394344911811, "trajectory_balance_rmse": 3289.6219129396572} +{"canonical_length": 52, "canonical_nll": 1.5155751154972956, "generation_rouge_l_f1": 0.1069182389937107, "generation_token_f1": 0.18867924528301888, "prompt_id": "fa52ad4bc1072fe5", "reading_canonical_log_prob": -78.80990600585938, "reading_distances": [47, 48, 45, 59], "reading_log_probs": [-79.03620147705078, -81.27531433105469, -80.56681823730469, -77.75877380371094], "reading_logprob_std": 1.5743086587636133, "tokenization_path_logprob_std": 33.26026837650255, "trajectory_balance_rmse": 872.9975290399466} +{"canonical_length": 192, "canonical_nll": 1.224640687306722, "generation_rouge_l_f1": 0.507592190889371, "generation_token_f1": 0.6203904555314534, "prompt_id": "63e9efbaa63103ce", "reading_canonical_log_prob": -235.13101196289062, "reading_distances": [16, 24, 25, 23], "reading_log_probs": [-235.599853515625, -235.56594848632812, -235.0928955078125, -234.83895874023438], "reading_logprob_std": 0.3712472138222697, "tokenization_path_logprob_std": 89.73508139561467, "trajectory_balance_rmse": 4477.5473974489805} +{"canonical_length": 265, "canonical_nll": 1.1943995061910377, "generation_rouge_l_f1": 0.2247191011235955, "generation_token_f1": 0.4082397003745319, "prompt_id": "867f08ace4ea14ea", "reading_canonical_log_prob": -316.515869140625, "reading_distances": [22, 25, 33, 21], "reading_log_probs": [-314.30902099609375, -312.7521667480469, -315.9413146972656, -313.46514892578125], "reading_logprob_std": 1.372670848602713, "tokenization_path_logprob_std": 76.63533182689524, "trajectory_balance_rmse": 5560.711112638111} +{"canonical_length": 6, "canonical_nll": 0.20411727825800577, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "126ead873a35c47e", "reading_canonical_log_prob": -1.2247036695480347, "reading_distances": [31, 37, 32, 27], "reading_log_probs": [-3.0350887775421143, -3.0340323448181152, -2.9798152446746826, -3.0267629623413086], "reading_logprob_std": 0.02633442732142293, "tokenization_path_logprob_std": 35.38203581331434, "trajectory_balance_rmse": 110.70686978144076} +{"canonical_length": 51, "canonical_nll": 2.5249137130438113, "generation_rouge_l_f1": 0.1282051282051282, "generation_token_f1": 0.2692307692307692, "prompt_id": "9800df60020cf919", "reading_canonical_log_prob": -128.77059936523438, "reading_distances": [51, 49, 47, 41], "reading_log_probs": [-129.44937133789062, -130.2902069091797, -127.51492309570312, -130.03485107421875], "reading_logprob_std": 1.255299593921627, "tokenization_path_logprob_std": 103.26083388658624, "trajectory_balance_rmse": 1137.9795608271845} +{"canonical_length": 11, "canonical_nll": 1.1855277148160068, "generation_rouge_l_f1": 0.07194244604316548, "generation_token_f1": 0.079136690647482, "prompt_id": "0aba661f60975419", "reading_canonical_log_prob": -13.040804862976074, "reading_distances": [12, 13, 20, 19], "reading_log_probs": [-11.195467948913574, -11.279443740844727, -11.355993270874023, -12.197442054748535], "reading_logprob_std": 0.464882531053686, "tokenization_path_logprob_std": 36.217907895635456, "trajectory_balance_rmse": 226.7325244955861} +{"canonical_length": 245, "canonical_nll": 1.2812582210618622, "generation_rouge_l_f1": 0.32801822323462415, "generation_token_f1": 0.5649202733485192, "prompt_id": "a78682b2f45a9311", "reading_canonical_log_prob": -313.90826416015625, "reading_distances": [15, 17, 17, 15], "reading_log_probs": [-313.11419677734375, -312.9049987792969, -313.27777099609375, -313.8340759277344], "reading_logprob_std": 0.3979494282072487, "tokenization_path_logprob_std": 167.61591790917947, "trajectory_balance_rmse": 4977.727617774461} +{"canonical_length": 41, "canonical_nll": 0.9481013228253621, "generation_rouge_l_f1": 0.14615384615384616, "generation_token_f1": 0.15384615384615385, "prompt_id": "e970a5e256be139f", "reading_canonical_log_prob": -38.872154235839844, "reading_distances": [14, 16, 12, 15], "reading_log_probs": [-34.812103271484375, -35.45856857299805, -34.6961669921875, -35.59632110595703], "reading_logprob_std": 0.45248071580626464, "tokenization_path_logprob_std": 39.92821572629145, "trajectory_balance_rmse": 356.37343863869114} +{"canonical_length": 112, "canonical_nll": 1.6302920750209264, "generation_rouge_l_f1": 0.19726027397260273, "generation_token_f1": 0.3726027397260274, "prompt_id": "0b4d03b967102211", "reading_canonical_log_prob": -182.59271240234375, "reading_distances": [23, 28, 26, 24], "reading_log_probs": [-178.33103942871094, -178.43289184570312, -177.74545288085938, -180.73538208007812], "reading_logprob_std": 1.3180749640787324, "tokenization_path_logprob_std": 223.98989624147998, "trajectory_balance_rmse": 2319.2377773992594} +{"canonical_length": 354, "canonical_nll": 1.1685722049346752, "generation_rouge_l_f1": 0.22895622895622897, "generation_token_f1": 0.43434343434343436, "prompt_id": "e03c0c9ce1e8af04", "reading_canonical_log_prob": -413.674560546875, "reading_distances": [11, 15, 15, 18], "reading_log_probs": [-414.0814514160156, -412.5495300292969, -412.7132568359375, -413.4117431640625], "reading_logprob_std": 0.7026942076944644, "tokenization_path_logprob_std": 277.0982762063825, "trajectory_balance_rmse": 7906.213406826301} +{"canonical_length": 114, "canonical_nll": 1.8956745883874726, "generation_rouge_l_f1": 0.09502262443438914, "generation_token_f1": 0.16289592760180996, "prompt_id": "6aab22322acba4ab", "reading_canonical_log_prob": -216.10690307617188, "reading_distances": [49, 44, 49, 43], "reading_log_probs": [-221.21136474609375, -217.7081298828125, -216.8419189453125, -218.15936279296875], "reading_logprob_std": 1.9010746490325965, "tokenization_path_logprob_std": 99.68053538496541, "trajectory_balance_rmse": 2180.5535759223294} +{"canonical_length": 72, "canonical_nll": 0.7093876732720269, "generation_rouge_l_f1": 0.32142857142857145, "generation_token_f1": 0.5357142857142857, "prompt_id": "1498d74fffe04fb0", "reading_canonical_log_prob": -51.07591247558594, "reading_distances": [27, 29, 20, 31], "reading_log_probs": [-55.78923797607422, -55.137229919433594, -52.67703628540039, -57.347015380859375], "reading_logprob_std": 1.9425497326689531, "tokenization_path_logprob_std": 52.036072745880176, "trajectory_balance_rmse": 1489.2485862191095} +{"canonical_length": 203, "canonical_nll": 1.6241310758543719, "generation_rouge_l_f1": 0.21573033707865166, "generation_token_f1": 0.47191011235955055, "prompt_id": "363570662fa8a4e6", "reading_canonical_log_prob": -329.6986083984375, "reading_distances": [31, 30, 29, 32], "reading_log_probs": [-329.5779724121094, -329.26287841796875, -328.0572509765625, -328.48248291015625], "reading_logprob_std": 0.6985313183147396, "tokenization_path_logprob_std": 195.3816019400253, "trajectory_balance_rmse": 4970.193499110351} +{"canonical_length": 267, "canonical_nll": 1.6727218342184573, "generation_rouge_l_f1": 0.2717948717948718, "generation_token_f1": 0.4205128205128206, "prompt_id": "2b4383e7d7d3624e", "reading_canonical_log_prob": -446.6167297363281, "reading_distances": [25, 34, 34, 33], "reading_log_probs": [-440.44134521484375, -441.09698486328125, -439.1066589355469, -442.182373046875], "reading_logprob_std": 1.28588201982704, "tokenization_path_logprob_std": 259.7383800476935, "trajectory_balance_rmse": 5881.207843736638} +{"canonical_length": 72, "canonical_nll": 1.4672809176974826, "generation_rouge_l_f1": 0.22727272727272727, "generation_token_f1": 0.22727272727272727, "prompt_id": "fadbe59576c37aff", "reading_canonical_log_prob": -105.64422607421875, "reading_distances": [19, 22, 16, 20], "reading_log_probs": [-108.47540283203125, -108.63163757324219, -106.36705017089844, -107.61341094970703], "reading_logprob_std": 1.0380730142087828, "tokenization_path_logprob_std": 95.46963665821548, "trajectory_balance_rmse": 1189.7839137320534} +{"canonical_length": 119, "canonical_nll": 1.4912091423483456, "generation_rouge_l_f1": 0.1769436997319035, "generation_token_f1": 0.353887399463807, "prompt_id": "6caa7e0c0b6b5ae9", "reading_canonical_log_prob": -177.45388793945312, "reading_distances": [26, 30, 24, 30], "reading_log_probs": [-178.17413330078125, -181.78829956054688, -181.01443481445312, -179.16732788085938], "reading_logprob_std": 1.658214630528678, "tokenization_path_logprob_std": 37.105607766855066, "trajectory_balance_rmse": 3032.690785365637} +{"canonical_length": 246, "canonical_nll": 1.1722116858009402, "generation_rouge_l_f1": 0.24646464646464644, "generation_token_f1": 0.5252525252525253, "prompt_id": "09ec255dc57dfab9", "reading_canonical_log_prob": -288.36407470703125, "reading_distances": [26, 27, 21, 29], "reading_log_probs": [-299.7371826171875, -298.8221435546875, -296.739990234375, -301.9864501953125], "reading_logprob_std": 2.1747261205035797, "tokenization_path_logprob_std": 160.20179917315735, "trajectory_balance_rmse": 5565.541215483495} +{"canonical_length": 83, "canonical_nll": 1.1850413127117847, "generation_rouge_l_f1": 0.2293333333333333, "generation_token_f1": 0.336, "prompt_id": "a8f654f305639b3f", "reading_canonical_log_prob": -98.35842895507812, "reading_distances": [17, 20, 13, 15], "reading_log_probs": [-111.74269104003906, -107.56253051757812, -107.8843994140625, -106.17373657226562], "reading_logprob_std": 2.3862824241925065, "tokenization_path_logprob_std": 44.57114389370186, "trajectory_balance_rmse": 722.9094871773317} +{"canonical_length": 286, "canonical_nll": 1.1562214031086102, "generation_rouge_l_f1": 0.19771863117870725, "generation_token_f1": 0.4144486692015209, "prompt_id": "0e8185255effe518", "reading_canonical_log_prob": -330.6793212890625, "reading_distances": [30, 29, 26, 25], "reading_log_probs": [-334.47601318359375, -328.82647705078125, -334.25897216796875, -329.2427978515625], "reading_logprob_std": 3.0848857219491315, "tokenization_path_logprob_std": 226.97120475583947, "trajectory_balance_rmse": 5218.23210474023} +{"canonical_length": 51, "canonical_nll": 1.2320312051212086, "generation_rouge_l_f1": 0.36036036036036034, "generation_token_f1": 0.5405405405405406, "prompt_id": "d03a9ea989bf62b9", "reading_canonical_log_prob": -62.83359146118164, "reading_distances": [30, 36, 34, 33], "reading_log_probs": [-61.454044342041016, -60.973716735839844, -62.74757385253906, -61.84703063964844], "reading_logprob_std": 0.7515869741890985, "tokenization_path_logprob_std": 118.45848910017614, "trajectory_balance_rmse": 936.03876741688} +{"canonical_length": 69, "canonical_nll": 0.995515795721524, "generation_rouge_l_f1": 0.031746031746031744, "generation_token_f1": 0.031746031746031744, "prompt_id": "fd11f4c3417e5ef0", "reading_canonical_log_prob": -68.69058990478516, "reading_distances": [44, 42, 39, 41], "reading_log_probs": [-67.59677124023438, -66.68450927734375, -67.19730377197266, -66.14871215820312], "reading_logprob_std": 0.6283743822527676, "tokenization_path_logprob_std": 85.52949035664577, "trajectory_balance_rmse": 1467.1461025616873} +{"canonical_length": 40, "canonical_nll": 1.5527408599853516, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.3238095238095238, "prompt_id": "43fe74e6e2773223", "reading_canonical_log_prob": -62.10963439941406, "reading_distances": [16, 14, 13, 20], "reading_log_probs": [-60.876495361328125, -60.818450927734375, -63.672733306884766, -63.29584503173828], "reading_logprob_std": 1.530305602665156, "tokenization_path_logprob_std": 96.64797378699399, "trajectory_balance_rmse": 692.0314881832976} +{"canonical_length": 82, "canonical_nll": 0.9212964685951791, "generation_rouge_l_f1": 0.1153846153846154, "generation_token_f1": 0.22307692307692306, "prompt_id": "7eeeafb10f4366ae", "reading_canonical_log_prob": -75.54631042480469, "reading_distances": [30, 28, 34, 28], "reading_log_probs": [-75.25790405273438, -75.57684326171875, -75.59538269042969, -75.81507873535156], "reading_logprob_std": 0.22938756874667368, "tokenization_path_logprob_std": 30.19382395873122, "trajectory_balance_rmse": 1180.9072289557837} +{"canonical_length": 3, "canonical_nll": 3.8981056213378906, "generation_rouge_l_f1": 0.01646090534979424, "generation_token_f1": 0.01646090534979424, "prompt_id": "6b50f3d234f6e0b4", "reading_canonical_log_prob": -11.694316864013672, "reading_distances": [20, 25, 26, 21], "reading_log_probs": [-11.281136512756348, -12.382268905639648, -12.329381942749023, -12.788995742797852], "reading_logprob_std": 0.6431963853729943, "tokenization_path_logprob_std": 17.825751779495626, "trajectory_balance_rmse": 29.947025923731044} +{"canonical_length": 348, "canonical_nll": 1.0981101551275143, "generation_rouge_l_f1": 0.2550335570469799, "generation_token_f1": 0.46644295302013433, "prompt_id": "b62b5836a5079a74", "reading_canonical_log_prob": -382.142333984375, "reading_distances": [31, 37, 29, 30], "reading_log_probs": [-383.79461669921875, -387.28326416015625, -385.3827819824219, -385.7225036621094], "reading_logprob_std": 1.4309930955738448, "tokenization_path_logprob_std": 422.0665987642041, "trajectory_balance_rmse": 7678.958202380816} +{"canonical_length": 39, "canonical_nll": 1.1302608587802985, "generation_rouge_l_f1": 0.42622950819672134, "generation_token_f1": 0.5245901639344263, "prompt_id": "495dc4d7314e6352", "reading_canonical_log_prob": -44.08017349243164, "reading_distances": [27, 21, 27, 27], "reading_log_probs": [-46.70221710205078, -46.866912841796875, -46.4110107421875, -45.977394104003906], "reading_logprob_std": 0.3899153655427109, "tokenization_path_logprob_std": 104.35912211320324, "trajectory_balance_rmse": 927.5144053263175} +{"canonical_length": 54, "canonical_nll": 1.0295446890371818, "generation_rouge_l_f1": 0.5121951219512195, "generation_token_f1": 0.5609756097560976, "prompt_id": "0c8a53a7cd7794b9", "reading_canonical_log_prob": -55.59541320800781, "reading_distances": [24, 25, 26, 30], "reading_log_probs": [-55.297515869140625, -55.00800323486328, -55.276100158691406, -55.4284553527832], "reading_logprob_std": 0.17637304112614952, "tokenization_path_logprob_std": 64.19952933821729, "trajectory_balance_rmse": 1120.7031752307228} +{"canonical_length": 185, "canonical_nll": 1.7919097075591217, "generation_rouge_l_f1": 0.2670623145400593, "generation_token_f1": 0.4807121661721069, "prompt_id": "bfd0b674c803f580", "reading_canonical_log_prob": -331.5032958984375, "reading_distances": [31, 33, 30, 37], "reading_log_probs": [-330.0795593261719, -329.1206359863281, -331.2238464355469, -329.2344665527344], "reading_logprob_std": 0.9719907692370999, "tokenization_path_logprob_std": 118.30150515098835, "trajectory_balance_rmse": 4484.533656472578} +{"canonical_length": 215, "canonical_nll": 1.4788040515988372, "generation_rouge_l_f1": 0.20915032679738566, "generation_token_f1": 0.35729847494553374, "prompt_id": "f65fb83f66989dab", "reading_canonical_log_prob": -317.94287109375, "reading_distances": [23, 31, 22, 26], "reading_log_probs": [-320.18829345703125, -319.5522766113281, -320.1844177246094, -320.99871826171875], "reading_logprob_std": 0.5927479799764934, "tokenization_path_logprob_std": 413.58246409444, "trajectory_balance_rmse": 4131.226044576667} +{"canonical_length": 20, "canonical_nll": 0.7009934902191162, "generation_rouge_l_f1": 0.15028901734104044, "generation_token_f1": 0.15028901734104044, "prompt_id": "ed802f2ffb8ef795", "reading_canonical_log_prob": -14.019869804382324, "reading_distances": [22, 18, 19, 22], "reading_log_probs": [-14.737024307250977, -14.668603897094727, -14.624616622924805, -14.853363037109375], "reading_logprob_std": 0.09968587420690495, "tokenization_path_logprob_std": 57.818411173090325, "trajectory_balance_rmse": 483.4709734707115} +{"canonical_length": 17, "canonical_nll": 0.7186801573809456, "generation_rouge_l_f1": 0.09302325581395349, "generation_token_f1": 0.10852713178294575, "prompt_id": "578c6b312be28f5a", "reading_canonical_log_prob": -12.217562675476074, "reading_distances": [29, 21, 29, 32], "reading_log_probs": [-10.430289268493652, -10.818822860717773, -9.589594841003418, -11.121356964111328], "reading_logprob_std": 0.6635839335621071, "tokenization_path_logprob_std": 44.091270794652814, "trajectory_balance_rmse": 403.2585823000984} +{"canonical_length": 20, "canonical_nll": 1.1974543571472167, "generation_rouge_l_f1": 0.4615384615384615, "generation_token_f1": 0.4615384615384615, "prompt_id": "016449fd389e919a", "reading_canonical_log_prob": -23.949087142944336, "reading_distances": [13, 19, 14, 16], "reading_log_probs": [-26.193641662597656, -26.081254959106445, -25.989173889160156, -26.40101432800293], "reading_logprob_std": 0.1774304531262736, "tokenization_path_logprob_std": 35.62748060231037, "trajectory_balance_rmse": 482.05577719652814} +{"canonical_length": 172, "canonical_nll": 1.399007752884266, "generation_rouge_l_f1": 0.2824074074074074, "generation_token_f1": 0.4768518518518518, "prompt_id": "977004fae436eabd", "reading_canonical_log_prob": -240.62933349609375, "reading_distances": [26, 27, 25, 25], "reading_log_probs": [-239.18649291992188, -239.86744689941406, -239.49485778808594, -240.4131622314453], "reading_logprob_std": 0.5278427586353488, "tokenization_path_logprob_std": 146.0204554668248, "trajectory_balance_rmse": 3703.4411492303498} +{"canonical_length": 223, "canonical_nll": 1.5197465152483884, "generation_rouge_l_f1": 0.26283367556468173, "generation_token_f1": 0.4804928131416838, "prompt_id": "3d21a3aabdc36c10", "reading_canonical_log_prob": -338.9034729003906, "reading_distances": [32, 23, 34, 31], "reading_log_probs": [-337.8128356933594, -338.0540771484375, -337.3888244628906, -335.64178466796875], "reading_logprob_std": 1.090309960135786, "tokenization_path_logprob_std": 235.40149232425244, "trajectory_balance_rmse": 4450.0169880542435} +{"canonical_length": 248, "canonical_nll": 0.4015452477239793, "generation_rouge_l_f1": 0.3640449438202247, "generation_token_f1": 0.5258426966292135, "prompt_id": "99f1a0976d255260", "reading_canonical_log_prob": -99.58322143554688, "reading_distances": [34, 32, 34, 32], "reading_log_probs": [-99.60563659667969, -99.7426528930664, -100.25645446777344, -99.80783081054688], "reading_logprob_std": 0.2817687100936032, "tokenization_path_logprob_std": 143.66559642869214, "trajectory_balance_rmse": 3983.5741496096994} +{"canonical_length": 183, "canonical_nll": 0.9759945895502476, "generation_rouge_l_f1": 0.3904761904761905, "generation_token_f1": 0.5714285714285714, "prompt_id": "24640c4b7c2fb283", "reading_canonical_log_prob": -178.6070098876953, "reading_distances": [72, 61, 66, 65], "reading_log_probs": [-173.80902099609375, -173.80767822265625, -174.98916625976562, -175.48687744140625], "reading_logprob_std": 0.8500631013782468, "tokenization_path_logprob_std": 175.4169047415387, "trajectory_balance_rmse": 3093.034540960644} +{"canonical_length": 6, "canonical_nll": 0.14048267404238382, "generation_rouge_l_f1": 1.0, "generation_token_f1": 1.0, "prompt_id": "94e6b38d533edcd5", "reading_canonical_log_prob": -0.842896044254303, "reading_distances": [35, 35, 38, 30], "reading_log_probs": [-1.8292200565338135, -1.7839224338531494, -2.2145578861236572, -3.1432535648345947], "reading_logprob_std": 0.6306692779985587, "tokenization_path_logprob_std": 37.334043696874716, "trajectory_balance_rmse": 112.30521526049215} +{"canonical_length": 252, "canonical_nll": 1.411266629658048, "generation_rouge_l_f1": 0.2652519893899204, "generation_token_f1": 0.4509283819628648, "prompt_id": "22a8bdc8dd7cf586", "reading_canonical_log_prob": -355.6391906738281, "reading_distances": [27, 22, 23, 22], "reading_log_probs": [-358.65447998046875, -358.3666687011719, -358.1504821777344, -358.87506103515625], "reading_logprob_std": 0.3182921899412936, "tokenization_path_logprob_std": 89.34190448893246, "trajectory_balance_rmse": 5787.083128299343} +{"canonical_length": 27, "canonical_nll": 1.2264141506618924, "generation_rouge_l_f1": 0.12121212121212122, "generation_token_f1": 0.14718614718614717, "prompt_id": "6585d8646962bd31", "reading_canonical_log_prob": -33.113182067871094, "reading_distances": [16, 14, 18, 12], "reading_log_probs": [-32.315147399902344, -32.57337188720703, -32.38389587402344, -31.35387420654297], "reading_logprob_std": 0.5461587665234732, "tokenization_path_logprob_std": 65.85895012710269, "trajectory_balance_rmse": 297.1297099243195} +{"canonical_length": 81, "canonical_nll": 1.3207838270399306, "generation_rouge_l_f1": 0.1729106628242075, "generation_token_f1": 0.2881844380403458, "prompt_id": "04a5aa53ab7903f1", "reading_canonical_log_prob": -106.98348999023438, "reading_distances": [22, 17, 27, 27], "reading_log_probs": [-105.38517761230469, -104.60279083251953, -103.55947875976562, -104.25321960449219], "reading_logprob_std": 0.759309488343214, "tokenization_path_logprob_std": 116.54876390293148, "trajectory_balance_rmse": 1975.9035000926374} +{"canonical_length": 198, "canonical_nll": 1.0666691943852589, "generation_rouge_l_f1": 0.26068376068376065, "generation_token_f1": 0.4829059829059829, "prompt_id": "f2143d23c11c6f8b", "reading_canonical_log_prob": -211.20050048828125, "reading_distances": [20, 24, 23, 20], "reading_log_probs": [-212.19410705566406, -214.0152130126953, -210.36471557617188, -215.0535430908203], "reading_logprob_std": 2.066172767833701, "tokenization_path_logprob_std": 197.3821344414922, "trajectory_balance_rmse": 3957.0983433048327} +{"canonical_length": 12, "canonical_nll": 1.0854857762654622, "generation_rouge_l_f1": 0.608695652173913, "generation_token_f1": 0.608695652173913, "prompt_id": "5633164a0de34e8b", "reading_canonical_log_prob": -13.025829315185547, "reading_distances": [36, 42, 40, 33], "reading_log_probs": [-11.024236679077148, -12.335505485534668, -12.839339256286621, -11.986848831176758], "reading_logprob_std": 0.7660937312122789, "tokenization_path_logprob_std": 39.367436932180766, "trajectory_balance_rmse": 257.87788190200854} +{"canonical_length": 134, "canonical_nll": 1.2080658869956857, "generation_rouge_l_f1": 0.23529411764705885, "generation_token_f1": 0.3631713554987212, "prompt_id": "fe842e52ca168c74", "reading_canonical_log_prob": -161.88082885742188, "reading_distances": [26, 25, 19, 27], "reading_log_probs": [-157.935791015625, -156.33810424804688, -162.28369140625, -156.57281494140625], "reading_logprob_std": 2.7588305245015414, "tokenization_path_logprob_std": 200.24907027713886, "trajectory_balance_rmse": 2451.6374126888045} +{"canonical_length": 46, "canonical_nll": 0.32252666224604065, "generation_rouge_l_f1": 0.37185929648241206, "generation_token_f1": 0.37185929648241206, "prompt_id": "6327942806f18cd4", "reading_canonical_log_prob": -14.836226463317871, "reading_distances": [7, 13, 10, 12], "reading_log_probs": [-15.429832458496094, -15.154864311218262, -14.691478729248047, -15.131950378417969], "reading_logprob_std": 0.30533778627943264, "tokenization_path_logprob_std": 42.75523626754549, "trajectory_balance_rmse": 518.8897134095276} +{"canonical_length": 54, "canonical_nll": 1.0350798854121455, "generation_rouge_l_f1": 0.23602484472049692, "generation_token_f1": 0.3975155279503106, "prompt_id": "95494e563c3fc530", "reading_canonical_log_prob": -55.89431381225586, "reading_distances": [29, 29, 30, 30], "reading_log_probs": [-56.73521423339844, -58.4442024230957, -57.76072692871094, -57.48214340209961], "reading_logprob_std": 0.7071377280058989, "tokenization_path_logprob_std": 70.11732986147959, "trajectory_balance_rmse": 1129.0281452340992} +{"canonical_length": 44, "canonical_nll": 1.6045008572665127, "generation_rouge_l_f1": 0.1377049180327869, "generation_token_f1": 0.19672131147540983, "prompt_id": "15a442e093e950e5", "reading_canonical_log_prob": -70.59803771972656, "reading_distances": [23, 29, 30, 26], "reading_log_probs": [-72.36802673339844, -71.43092346191406, -72.44388580322266, -72.41047668457031], "reading_logprob_std": 0.4892556019429387, "tokenization_path_logprob_std": 76.47507965369691, "trajectory_balance_rmse": 786.3111406007814} +{"canonical_length": 303, "canonical_nll": 0.7462673879692657, "generation_rouge_l_f1": 0.22709163346613542, "generation_token_f1": 0.3665338645418327, "prompt_id": "452a884e801c54f0", "reading_canonical_log_prob": -226.1190185546875, "reading_distances": [15, 24, 29, 28], "reading_log_probs": [-225.07901000976562, -224.50550842285156, -225.34585571289062, -224.40512084960938], "reading_logprob_std": 0.4523531635952492, "tokenization_path_logprob_std": 110.21398564983213, "trajectory_balance_rmse": 5209.9177855636835} +{"canonical_length": 340, "canonical_nll": 1.1268679450539982, "generation_rouge_l_f1": 0.22750424448217316, "generation_token_f1": 0.3803056027164686, "prompt_id": "6f1aa7f491e54018", "reading_canonical_log_prob": -383.1351013183594, "reading_distances": [25, 23, 16, 18], "reading_log_probs": [-381.20745849609375, -379.416259765625, -381.1500244140625, -383.04071044921875], "reading_logprob_std": 1.4801402125856729, "tokenization_path_logprob_std": 188.72368106041748, "trajectory_balance_rmse": 7108.052178810597} +{"canonical_length": 197, "canonical_nll": 1.0724264232035217, "generation_rouge_l_f1": 0.21409921671018278, "generation_token_f1": 0.3550913838120104, "prompt_id": "b6dbd4688fd838f3", "reading_canonical_log_prob": -211.26800537109375, "reading_distances": [20, 19, 18, 22], "reading_log_probs": [-214.31143188476562, -212.4839630126953, -211.08584594726562, -213.09783935546875], "reading_logprob_std": 1.3415335414263883, "tokenization_path_logprob_std": 281.86095519769856, "trajectory_balance_rmse": 4538.548425486718} +{"canonical_length": 138, "canonical_nll": 1.6379087144050046, "generation_rouge_l_f1": 0.21666666666666667, "generation_token_f1": 0.35, "prompt_id": "4e91817c9d8f7b21", "reading_canonical_log_prob": -226.03140258789062, "reading_distances": [17, 20, 19, 20], "reading_log_probs": [-230.7547149658203, -233.64306640625, -231.03302001953125, -232.4783935546875], "reading_logprob_std": 1.3431618564725325, "tokenization_path_logprob_std": 49.793092664961115, "trajectory_balance_rmse": 2042.0982518079768} +{"canonical_length": 10, "canonical_nll": 1.5467422485351563, "generation_rouge_l_f1": 0.34782608695652173, "generation_token_f1": 0.34782608695652173, "prompt_id": "8f6ce2573cd1b9d1", "reading_canonical_log_prob": -15.467422485351562, "reading_distances": [37, 33, 38, 35], "reading_log_probs": [-18.21755599975586, -17.26218032836914, -18.53577995300293, -18.760963439941406], "reading_logprob_std": 0.6600767143634498, "tokenization_path_logprob_std": 35.930277329339596, "trajectory_balance_rmse": 216.58887085959086} +{"canonical_length": 9, "canonical_nll": 0.641160700056288, "generation_rouge_l_f1": 0.03404255319148936, "generation_token_f1": 0.03404255319148936, "prompt_id": "c246d0eff1fb5aac", "reading_canonical_log_prob": -5.770446300506592, "reading_distances": [12, 18, 17, 17], "reading_log_probs": [-7.353086471557617, -6.334869384765625, -6.7507853507995605, -7.232570171356201], "reading_logprob_std": 0.46770951284619455, "tokenization_path_logprob_std": 20.884515871185812, "trajectory_balance_rmse": 169.08867006737225} +{"canonical_length": 222, "canonical_nll": 0.6155708931587838, "generation_rouge_l_f1": 0.41163310961968685, "generation_token_f1": 0.639821029082774, "prompt_id": "877651642570fc39", "reading_canonical_log_prob": -136.65673828125, "reading_distances": [31, 33, 32, 30], "reading_log_probs": [-147.85498046875, -145.01239013671875, -147.46961975097656, -142.02212524414062], "reading_logprob_std": 2.6911255491707036, "tokenization_path_logprob_std": 153.43575714290765, "trajectory_balance_rmse": 3316.7447929217756} +{"canonical_length": 53, "canonical_nll": 0.9036323979215802, "generation_rouge_l_f1": 0.2857142857142857, "generation_token_f1": 0.35714285714285715, "prompt_id": "77a973a8d7f91484", "reading_canonical_log_prob": -47.89251708984375, "reading_distances": [31, 29, 31, 33], "reading_log_probs": [-45.27128601074219, -45.3631477355957, -46.798370361328125, -46.075416564941406], "reading_logprob_std": 0.7116239761322932, "tokenization_path_logprob_std": 31.407103710686094, "trajectory_balance_rmse": 1176.9877742864694} +{"canonical_length": 21, "canonical_nll": 0.8318248930431548, "generation_rouge_l_f1": 0.7027027027027027, "generation_token_f1": 0.7567567567567567, "prompt_id": "f9bfdd6eb133d424", "reading_canonical_log_prob": -17.46832275390625, "reading_distances": [24, 24, 22, 26], "reading_log_probs": [-17.744054794311523, -17.697998046875, -17.428592681884766, -17.511632919311523], "reading_logprob_std": 0.14996232549901334, "tokenization_path_logprob_std": 52.833387340098575, "trajectory_balance_rmse": 429.45250236743533} +{"canonical_length": 247, "canonical_nll": 1.3992431887730896, "generation_rouge_l_f1": 0.19405940594059404, "generation_token_f1": 0.4633663366336634, "prompt_id": "3a822df2306dda39", "reading_canonical_log_prob": -345.6130676269531, "reading_distances": [39, 36, 48, 42], "reading_log_probs": [-350.38812255859375, -346.43743896484375, -350.0181884765625, -349.86669921875], "reading_logprob_std": 1.8398635894234872, "tokenization_path_logprob_std": 215.9467018792754, "trajectory_balance_rmse": 5670.00632719805} +{"canonical_length": 88, "canonical_nll": 1.3031883239746094, "generation_rouge_l_f1": 0.19811320754716982, "generation_token_f1": 0.38679245283018865, "prompt_id": "c28f9f6ed8615ffa", "reading_canonical_log_prob": -114.68057250976562, "reading_distances": [26, 27, 29, 29], "reading_log_probs": [-110.85505676269531, -108.42642974853516, -109.47074890136719, -108.48900604248047], "reading_logprob_std": 1.1354540489008245, "tokenization_path_logprob_std": 142.2326860246908, "trajectory_balance_rmse": 1926.0285333036643}