Training in progress, step 5062

Information-Retrieval_evaluation_BeIR-touche2020-subset-test_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.6938775510204082,0.8775510204081632,0.8979591836734694,0.9591836734693877,0.6938775510204082,0.015315719886663845,0.6598639455782314,0.04368231116741089,0.6244897959183674,0.06873784889607878,0.5183673469387755,0.11403290871124043,0.7894800777453839,0.5610681849792926,0.22889125310172967
 -1,-1,0.7755102040816326,0.9183673469387755,0.9795918367346939,0.9795918367346939,0.7755102040816326,0.017145242978623458,0.6802721088435375,0.04496099130666259,0.6204081632653061,0.06848690384258264,0.5469387755102041,0.12039059566521017,0.8534013605442178,0.5893987348618942,0.2427677659709095
 -1,-1,0.7346938775510204,1.0,1.0,1.0,0.7346938775510204,0.016356718324589865,0.7482993197278911,0.04982275407454,0.7306122448979593,0.08077270192430855,0.6448979591836734,0.14251827101536743,0.8605442176870749,0.6762844931311536,0.3020322355884379
+-1,-1,0.7346938775510204,0.9183673469387755,0.9591836734693877,1.0,0.7346938775510204,0.016137830046673268,0.673469387755102,0.04465470442976813,0.6571428571428573,0.07229868006332099,0.589795918367347,0.129926645587027,0.8311791383219955,0.6198727729504082,0.2553079711703241

Information-Retrieval_evaluation_NanoArguAna_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.14,0.32,0.5,0.62,0.14,0.14,0.10666666666666665,0.32,0.1,0.5,0.062,0.62,0.2769047619047619,0.3585937146114901,0.2875939051064566
 -1,-1,0.14,0.32,0.42,0.6,0.14,0.14,0.10666666666666666,0.32,0.084,0.42,0.06000000000000001,0.6,0.2737619047619047,0.3511790392551435,0.2882673533302593
 -1,-1,0.24,0.56,0.72,0.9,0.24,0.24,0.18666666666666668,0.56,0.14400000000000002,0.72,0.08999999999999998,0.9,0.4462380952380951,0.5546300286352132,0.45259153346653347
+-1,-1,0.18,0.38,0.46,0.62,0.18,0.18,0.12666666666666665,0.38,0.09200000000000001,0.46,0.06200000000000001,0.62,0.3023571428571429,0.37737911678224656,0.31860599951297625

Information-Retrieval_evaluation_NanoClimateFEVER_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.16,0.36,0.46,0.52,0.16,0.095,0.12,0.19166666666666665,0.09200000000000001,0.21566666666666667,0.05600000000000001,0.25066666666666665,0.2785555555555555,0.20771869097840734,0.16609214397302466
 -1,-1,0.2,0.38,0.46,0.54,0.2,0.09,0.12666666666666665,0.18666666666666668,0.09600000000000002,0.23666666666666664,0.05800000000000001,0.2656666666666666,0.3038888888888889,0.21617293027091072,0.16828448758543282
 -1,-1,0.3,0.48,0.56,0.66,0.3,0.12333333333333334,0.15999999999999998,0.21166666666666664,0.11600000000000002,0.2533333333333333,0.078,0.325,0.406079365079365,0.2735257041147495,0.21801163753351321
+-1,-1,0.24,0.4,0.46,0.56,0.24,0.09233333333333332,0.13333333333333333,0.174,0.1,0.22066666666666668,0.064,0.27899999999999997,0.33938095238095245,0.2283726062129329,0.17414775290079157

Information-Retrieval_evaluation_NanoDBPedia_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.5,0.76,0.84,0.9,0.5,0.053658269837470775,0.48,0.12026250374497954,0.42400000000000004,0.15367410972243642,0.33399999999999996,0.23559122896232249,0.6469999999999999,0.4221930580140474,0.2547954478342057
 -1,-1,0.6,0.76,0.82,0.88,0.6,0.06888056628544742,0.46,0.11839185708955416,0.4040000000000001,0.1452356141513428,0.336,0.238350192919226,0.6931904761904762,0.43317983055501846,0.2740312909587234
 -1,-1,0.7,0.88,0.9,0.94,0.7,0.05885278913396376,0.5533333333333332,0.1651324985031048,0.46399999999999997,0.1937322996063979,0.398,0.278801784770749,0.7872222222222223,0.5090931324148577,0.32792556078063057
+-1,-1,0.58,0.76,0.88,0.92,0.58,0.06197705417764409,0.46666666666666673,0.12223781146808171,0.42,0.16808655070985257,0.364,0.2608995958885433,0.6831904761904761,0.4521290486102402,0.2791787003737884

Information-Retrieval_evaluation_NanoFEVER_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.52,0.76,0.84,0.86,0.52,0.48666666666666664,0.26,0.7133333333333333,0.17199999999999996,0.7833333333333333,0.09,0.8133333333333332,0.6428333333333331,0.6679092320752315,0.6126252419988711
 -1,-1,0.54,0.72,0.84,0.88,0.54,0.49666666666666665,0.24,0.6666666666666665,0.176,0.8033333333333332,0.092,0.8333333333333333,0.6546666666666665,0.6782240790417856,0.618283946311562
 -1,-1,0.86,0.94,0.96,0.98,0.86,0.7966666666666665,0.3133333333333333,0.8766666666666666,0.19599999999999995,0.9066666666666666,0.102,0.9366666666666668,0.9073333333333333,0.8846483562973002,0.8511817169008258
+-1,-1,0.68,0.84,0.9,0.94,0.68,0.6566666666666666,0.28,0.7866666666666666,0.184,0.8566666666666666,0.09599999999999997,0.8866666666666667,0.7660238095238096,0.7797475710520765,0.7347713746077809

Information-Retrieval_evaluation_NanoFiQA2018_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.28,0.42,0.54,0.64,0.28,0.16252380952380954,0.16666666666666663,0.28085714285714286,0.132,0.3406031746031746,0.08399999999999999,0.40984920634920635,0.3814365079365079,0.32320891888133074,0.26217112838711026
 -1,-1,0.28,0.38,0.44,0.56,0.28,0.15585714285714286,0.16666666666666663,0.2386031746031746,0.11599999999999999,0.2777698412698413,0.07600000000000001,0.36926984126984125,0.34735714285714286,0.29478491063531465,0.2452639486126713
 -1,-1,0.42,0.58,0.68,0.7,0.42,0.22274603174603172,0.24666666666666667,0.3608253968253968,0.19599999999999998,0.4680714285714286,0.10799999999999998,0.5110714285714285,0.512,0.4267909833793157,0.3624765177374492
+-1,-1,0.24,0.34,0.46,0.56,0.24,0.14085714285714285,0.13999999999999999,0.19988095238095238,0.11200000000000002,0.30054761904761906,0.08,0.38654761904761903,0.3230238095238096,0.28855478051901995,0.23168919640832794

Information-Retrieval_evaluation_NanoHotpotQA_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.5,0.62,0.66,0.74,0.5,0.25,0.2266666666666667,0.34,0.14800000000000002,0.37,0.09,0.45,0.5706904761904761,0.4185765233643338,0.3410399663317717
 -1,-1,0.54,0.64,0.66,0.72,0.54,0.27,0.24666666666666665,0.37,0.15600000000000003,0.39,0.088,0.44,0.5992222222222222,0.43520521230408366,0.3717832829206536
 -1,-1,0.82,0.88,0.92,0.94,0.82,0.41,0.3999999999999999,0.6,0.26,0.65,0.13799999999999998,0.69,0.8555555555555556,0.6791736335059022,0.6031157171874328
+-1,-1,0.62,0.74,0.76,0.8,0.62,0.31,0.29333333333333333,0.44,0.18799999999999997,0.47,0.10599999999999998,0.53,0.676,0.508052717401943,0.43060781600475145

Information-Retrieval_evaluation_NanoMSMARCO_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.28,0.58,0.62,0.7,0.28,0.28,0.19333333333333333,0.58,0.124,0.62,0.07,0.7,0.4328888888888888,0.49829987674731946,0.44341877222019976
 -1,-1,0.34,0.48,0.6,0.68,0.34,0.34,0.15999999999999998,0.48,0.12,0.6,0.068,0.68,0.43771428571428567,0.4954780706089552,0.4528382989436611
 -1,-1,0.52,0.7,0.74,0.84,0.52,0.52,0.2333333333333333,0.7,0.14800000000000002,0.74,0.08399999999999999,0.84,0.6126587301587302,0.6667666782328303,0.6201688107214423
+-1,-1,0.34,0.5,0.58,0.68,0.34,0.34,0.16666666666666669,0.5,0.11599999999999999,0.58,0.068,0.68,0.44646825396825385,0.5023070028455473,0.45949436084808376

Information-Retrieval_evaluation_NanoNFCorpus_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.32,0.46,0.52,0.64,0.32,0.011933029076017525,0.26666666666666666,0.02715254462665005,0.24400000000000002,0.040025046693376296,0.212,0.06269126978571375,0.41073809523809524,0.231923237081856,0.07267684281911838
 -1,-1,0.3,0.4,0.46,0.56,0.3,0.030361106242164686,0.26,0.04565643981003684,0.236,0.05740354059693826,0.20800000000000002,0.07808848031588318,0.3696666666666666,0.2418470331708076,0.09302535365326235
 -1,-1,0.4,0.46,0.52,0.68,0.4,0.013821074066170821,0.32666666666666666,0.03704372187363592,0.304,0.05641224522012561,0.282,0.10199363274777667,0.460515873015873,0.30539493217594926,0.12327617976680419
+-1,-1,0.3,0.4,0.48,0.56,0.3,0.010085921966980407,0.28,0.047968018726256945,0.276,0.0671856089688558,0.242,0.1033851527560607,0.3788888888888888,0.2699339552042321,0.10920681945613339

Information-Retrieval_evaluation_NanoNQ_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.26,0.44,0.5,0.6,0.26,0.23,0.14666666666666667,0.41,0.10400000000000001,0.48,0.064,0.58,0.37035714285714283,0.4092826812027592,0.3579080102167239
 -1,-1,0.34,0.48,0.54,0.68,0.34,0.32,0.16666666666666663,0.45,0.11600000000000002,0.52,0.07200000000000001,0.65,0.42780158730158724,0.4724642279893318,0.41892039298943606
 -1,-1,0.38,0.56,0.62,0.7,0.38,0.35,0.19333333333333333,0.53,0.132,0.6,0.07400000000000001,0.66,0.48483333333333334,0.5159192638156053,0.47956629357668895
+-1,-1,0.36,0.54,0.56,0.7,0.36,0.33,0.18666666666666668,0.52,0.11599999999999999,0.54,0.07400000000000001,0.66,0.4570555555555556,0.4933846471447955,0.44680861443860365

Information-Retrieval_evaluation_NanoQuoraRetrieval_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.8,0.92,0.94,0.96,0.8,0.6873333333333334,0.3999999999999999,0.8953333333333333,0.24799999999999997,0.922,0.12999999999999998,0.95,0.8626666666666667,0.8792389359694834,0.8547388043505691
 -1,-1,0.82,0.92,0.92,0.96,0.82,0.7140000000000001,0.37999999999999995,0.882,0.244,0.8993333333333333,0.132,0.9533333333333335,0.8717460317460318,0.8867508970607386,0.8613894286385704
 -1,-1,0.86,0.98,0.98,1.0,0.86,0.7373333333333332,0.4133333333333333,0.9520000000000001,0.26,0.9666666666666666,0.13599999999999998,0.9933333333333334,0.9191666666666666,0.92794886246839,0.9003272642390289
+-1,-1,0.78,0.92,0.96,0.96,0.78,0.6773333333333332,0.3666666666666666,0.8713333333333333,0.24,0.9279999999999999,0.12799999999999997,0.946,0.8556666666666666,0.8656777707479286,0.8321868686868686

Information-Retrieval_evaluation_NanoSCIDOCS_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.36,0.54,0.68,0.74,0.36,0.07300000000000001,0.28,0.16899999999999998,0.24799999999999997,0.2526666666666666,0.15,0.30566666666666664,0.47083333333333327,0.2990008640503006,0.2340098877413048
 -1,-1,0.46,0.66,0.7,0.84,0.46,0.094,0.30666666666666664,0.18599999999999997,0.24,0.243,0.17999999999999997,0.36766666666666664,0.5746349206349205,0.35403666072510376,0.2616504468038099
 -1,-1,0.36,0.58,0.64,0.68,0.36,0.07466666666666667,0.2866666666666666,0.17666666666666667,0.26,0.26466666666666666,0.156,0.31966666666666665,0.47855555555555546,0.31516617152348114,0.2537533171047557
+-1,-1,0.44,0.64,0.7,0.82,0.44,0.09000000000000002,0.32666666666666666,0.19799999999999998,0.268,0.27099999999999996,0.174,0.3536666666666666,0.5576904761904762,0.3525296369677544,0.2706982716239424

Information-Retrieval_evaluation_NanoSciFact_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.3,0.42,0.52,0.62,0.3,0.265,0.14666666666666667,0.38,0.12,0.5,0.07200000000000001,0.61,0.37949999999999995,0.4241161704119403,0.3683841412805895
 -1,-1,0.36,0.54,0.6,0.68,0.36,0.325,0.2,0.515,0.136,0.58,0.076,0.66,0.46946825396825387,0.5054872304018916,0.46018750511209533
 -1,-1,0.54,0.7,0.76,0.84,0.54,0.505,0.2533333333333333,0.68,0.17599999999999993,0.76,0.09599999999999997,0.84,0.6369126984126985,0.6844325567315029,0.6389995767091675
+-1,-1,0.48,0.54,0.62,0.68,0.48,0.445,0.19333333333333336,0.52,0.14,0.6,0.08,0.68,0.5318571428571428,0.5618706429753196,0.5267835796727546

Information-Retrieval_evaluation_NanoTouche2020_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.5306122448979592,0.8163265306122449,0.8571428571428571,0.9591836734693877,0.5306122448979592,0.0357154474859051,0.4829931972789115,0.09952008767216199,0.4489795918367347,0.1552200163482205,0.36938775510204086,0.24755840994384426,0.6867913832199547,0.41997617822785366,0.29666098443409494
 -1,-1,0.5510204081632653,0.8367346938775511,0.8979591836734694,0.9795918367346939,0.5510204081632653,0.04003235129260482,0.5170068027210885,0.10793989356094978,0.4938775510204081,0.172549578322173,0.41632653061224495,0.2818845927319976,0.7097424684159378,0.46678620623082895,0.3354042261183708
 -1,-1,0.6122448979591837,0.8367346938775511,0.9387755102040817,0.9795918367346939,0.6122448979591837,0.04261846159526855,0.5238095238095238,0.11310802492381491,0.48979591836734704,0.17549525482392914,0.41836734693877553,0.285598908591513,0.7421768707482992,0.4763620784420955,0.357725600182186
+-1,-1,0.4897959183673469,0.7755102040816326,0.8367346938775511,1.0,0.4897959183673469,0.03643545588582757,0.46258503401360546,0.09797151935416279,0.436734693877551,0.15541127788533005,0.3816326530612244,0.26155265982169057,0.6553206997084547,0.4247890449649847,0.31946609885183697

NanoBEIR_evaluation_mean_results.csv

@@ -25,3 +25,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.38081632653061226,0.570486656200942,0.6520879120879121,0.7307064364207221,0.38081632653061226,0.21314081199409254,0.2520251177394034,0.34824043171032826,0.2003830455259027,0.4102453087718365,0.13718367346938776,0.47964282936213487,0.4931689342403628,0.42769523704741186,0.3501627135918493
 -1,-1,0.42084772370486656,0.578210361067504,0.6429199372056514,0.735353218210361,0.42084772370486656,0.2372921410264636,0.2566928309785453,0.351301899876696,0.20137519623233907,0.41117630059027915,0.14325588697017272,0.49366100824899595,0.5179124243103834,0.44858433294230105,0.37302538169065447
 -1,-1,0.5394034536891681,0.7028257456828886,0.7645211930926217,0.8338147566718994,0.5394034536891681,0.3150029505031873,0.31465201465201464,0.45870074170199626,0.24198430141287283,0.5196188124273241,0.166182103610675,0.5909332631806257,0.634557561486133,0.5553732601336302,0.47608613276203526
+-1,-1,0.4407535321821036,0.5981161695447409,0.6659026687598116,0.7538461538461539,0.4407535321821036,0.25928376217084065,0.2632757718472004,0.3736967924561118,0.20682574568288856,0.4321203376880761,0.1476640502354788,0.5113629508344035,0.536378759562433,0.46959450318684787,0.3948958041066646

config.json

@@ -5,6 +5,7 @@
   "attention_probs_dropout_prob": 0.1,
   "classifier_dropout": null,
   "dtype": "float32",
+  "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 384,
@@ -14,7 +15,7 @@
   "max_position_embeddings": 512,
   "model_type": "bert",
   "num_attention_heads": 12,
-  "num_hidden_layers": 12,
+  "num_hidden_layers": 6,
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
   "transformers_version": "4.57.3",

eval/Information-Retrieval_evaluation_NanoMSMARCO_results.csv

@@ -317,3 +317,24 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 3.9118065433854907,2750,0.34,0.54,0.6,0.7,0.34,0.34,0.18,0.54,0.12,0.6,0.07,0.7,0.4555555555555555,0.5144396463820081,0.4676660744696763
 4.2674253200568995,3000,0.34,0.54,0.58,0.68,0.34,0.34,0.18,0.54,0.11599999999999999,0.58,0.068,0.68,0.4531349206349205,0.5076985355940407,0.46637338013638
 4.623044096728307,3250,0.34,0.5,0.58,0.68,0.34,0.34,0.16666666666666669,0.5,0.11599999999999999,0.58,0.068,0.68,0.44646825396825385,0.5023070028455473,0.4594814287577352
+0,0,0.36,0.52,0.58,0.8,0.36,0.36,0.1733333333333333,0.52,0.11599999999999999,0.58,0.08,0.8,0.47960317460317464,0.5539831330912274,0.49062451473627944
+0.35561877667140823,250,0.26,0.46,0.56,0.68,0.26,0.26,0.1533333333333333,0.46,0.11200000000000002,0.56,0.068,0.68,0.38976190476190475,0.4592169782089742,0.4040552555433255
+0.7112375533428165,500,0.3,0.52,0.66,0.72,0.3,0.3,0.1733333333333333,0.52,0.132,0.66,0.07200000000000001,0.72,0.4446031746031745,0.5114522270072944,0.4547831991824339
+1.0668563300142249,750,0.32,0.54,0.58,0.7,0.32,0.32,0.18,0.54,0.11599999999999999,0.58,0.07,0.7,0.4470793650793651,0.5080168172271544,0.4581584972700965
+1.422475106685633,1000,0.32,0.52,0.58,0.72,0.32,0.32,0.1733333333333333,0.52,0.11600000000000002,0.58,0.07200000000000001,0.72,0.4414920634920634,0.5075350809243357,0.4515529964034999
+1.7780938833570412,1250,0.26,0.5,0.62,0.72,0.26,0.26,0.16666666666666663,0.5,0.12400000000000003,0.62,0.07200000000000001,0.72,0.4128809523809523,0.4873865109314843,0.42375961326729444
+2.1337126600284497,1500,0.36,0.5,0.58,0.7,0.36,0.36,0.16666666666666669,0.5,0.11599999999999999,0.58,0.07,0.7,0.4575793650793651,0.5152975002737303,0.46910005059527415
+2.4893314366998576,1750,0.28,0.52,0.52,0.68,0.28,0.28,0.1733333333333333,0.52,0.10400000000000001,0.52,0.068,0.68,0.4089841269841269,0.47366561116205813,0.42100623681581617
+2.844950213371266,2000,0.26,0.48,0.56,0.68,0.26,0.26,0.15999999999999998,0.48,0.11200000000000002,0.56,0.068,0.68,0.3979365079365079,0.46565204751247136,0.41075418177472356
+3.200568990042674,2250,0.3,0.52,0.6,0.7,0.3,0.3,0.1733333333333333,0.52,0.12000000000000002,0.6,0.07,0.7,0.42488095238095236,0.4909608161784584,0.43535108783756976
+3.5561877667140824,2500,0.28,0.46,0.58,0.7,0.28,0.28,0.15333333333333332,0.46,0.11600000000000002,0.58,0.07,0.7,0.3973888888888888,0.4686558228487607,0.40811465635390987
+3.9118065433854907,2750,0.3,0.5,0.6,0.7,0.3,0.3,0.16666666666666669,0.5,0.12000000000000002,0.6,0.07,0.7,0.4247142857142856,0.49050162118813256,0.43418353261559817
+4.2674253200568995,3000,0.26,0.42,0.58,0.7,0.26,0.26,0.13999999999999999,0.42,0.11600000000000002,0.58,0.07,0.7,0.385579365079365,0.4606172699574768,0.39513805378951944
+4.623044096728307,3250,0.3,0.5,0.54,0.68,0.3,0.3,0.16666666666666669,0.5,0.10800000000000001,0.54,0.068,0.68,0.41146031746031736,0.4752069013901737,0.42192722196542365
+4.978662873399715,3500,0.22,0.5,0.6,0.68,0.22,0.22,0.16666666666666669,0.5,0.12000000000000002,0.6,0.068,0.68,0.3792460317460318,0.4522785171300954,0.3901800430375596
+5.334281650071124,3750,0.26,0.52,0.58,0.7,0.26,0.26,0.1733333333333333,0.52,0.11599999999999999,0.58,0.07,0.7,0.4031031746031746,0.47448155008990556,0.41190864526876503
+5.689900426742532,4000,0.24,0.52,0.6,0.7,0.24,0.24,0.1733333333333333,0.52,0.12000000000000002,0.6,0.07,0.7,0.3939920634920635,0.46795236095078685,0.4033948519422716
+6.0455192034139404,4250,0.24,0.5,0.6,0.7,0.24,0.24,0.16666666666666669,0.5,0.12000000000000002,0.6,0.07,0.7,0.3911031746031747,0.4659273158864834,0.4005338185425652
+6.401137980085348,4500,0.24,0.48,0.6,0.7,0.24,0.24,0.15999999999999998,0.48,0.12000000000000002,0.6,0.07,0.7,0.3887460317460318,0.4637171057179736,0.3979732729573351
+6.756756756756757,4750,0.26,0.52,0.62,0.7,0.26,0.26,0.1733333333333333,0.52,0.124,0.62,0.07,0.7,0.4052777777777778,0.47639526023385004,0.41438699416556907
+7.112375533428165,5000,0.24,0.52,0.62,0.7,0.24,0.24,0.1733333333333333,0.52,0.124,0.62,0.07,0.7,0.39216666666666666,0.46663456362077194,0.4014868405214368

eval/Information-Retrieval_evaluation_NanoNQ_results.csv

@@ -317,3 +317,24 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 3.9118065433854907,2750,0.38,0.52,0.58,0.64,0.38,0.35,0.18,0.5,0.124,0.57,0.07,0.62,0.46452380952380956,0.49295049784379946,0.4622886235526178
 4.2674253200568995,3000,0.34,0.54,0.58,0.68,0.34,0.31,0.18666666666666668,0.52,0.12000000000000002,0.56,0.07200000000000001,0.65,0.4445,0.4819555904793751,0.43525704834185847
 4.623044096728307,3250,0.36,0.54,0.58,0.66,0.36,0.33,0.18666666666666668,0.52,0.12000000000000002,0.56,0.07,0.63,0.4522460317460317,0.4834842533929267,0.4457601073900966
+0,0,0.48,0.62,0.66,0.72,0.48,0.47,0.21333333333333332,0.6,0.14,0.64,0.07600000000000001,0.7,0.5643571428571428,0.5930818641709092,0.5653172420421086
+0.35561877667140823,250,0.24,0.44,0.46,0.54,0.24,0.22,0.16,0.42,0.1,0.44,0.06000000000000001,0.52,0.3525555555555556,0.38899404660346204,0.35750870550380204
+0.7112375533428165,500,0.14,0.36,0.42,0.52,0.14,0.13,0.12666666666666665,0.35,0.08800000000000001,0.39,0.054000000000000006,0.49,0.26285714285714284,0.3132221744294717,0.2673858451620768
+1.0668563300142249,750,0.14,0.34,0.36,0.46,0.14,0.14,0.11333333333333333,0.32,0.08,0.36,0.05000000000000001,0.46,0.24219047619047615,0.2957710474099489,0.2596232367239361
+1.422475106685633,1000,0.14,0.38,0.46,0.56,0.14,0.13,0.13333333333333333,0.38,0.09600000000000002,0.44,0.05800000000000001,0.53,0.27263492063492056,0.33274001826826516,0.2786334585966019
+1.7780938833570412,1250,0.16,0.3,0.36,0.52,0.16,0.15,0.1,0.27,0.07600000000000001,0.34,0.054000000000000006,0.48,0.2562380952380952,0.3017511663819104,0.25504432240124164
+2.1337126600284497,1500,0.1,0.32,0.4,0.48,0.1,0.1,0.10666666666666666,0.29,0.084,0.38,0.05,0.46,0.22669047619047622,0.28091835917904623,0.23438539086539606
+2.4893314366998576,1750,0.1,0.34,0.4,0.48,0.1,0.1,0.11333333333333333,0.32,0.084,0.39,0.05000000000000001,0.47,0.22761111111111113,0.28381000749211244,0.23372186194756897
+2.844950213371266,2000,0.12,0.26,0.32,0.42,0.12,0.1,0.08666666666666666,0.23,0.068,0.3,0.044000000000000004,0.4,0.2096031746031746,0.24436570355845827,0.20633418171154053
+3.200568990042674,2250,0.1,0.2,0.38,0.48,0.1,0.09,0.06666666666666667,0.18,0.08,0.35,0.05000000000000001,0.45,0.20191269841269843,0.2547465530861845,0.20457869605430878
+3.5561877667140824,2500,0.14,0.22,0.42,0.54,0.14,0.14,0.07333333333333333,0.21,0.08800000000000001,0.39,0.05600000000000001,0.51,0.2336349206349206,0.2951981501741704,0.2348838125303391
+3.9118065433854907,2750,0.14,0.26,0.4,0.48,0.14,0.14,0.08666666666666666,0.24,0.084,0.37,0.05,0.45,0.2330238095238095,0.28207487981720364,0.2376324016473298
+4.2674253200568995,3000,0.12,0.22,0.34,0.46,0.12,0.12,0.07333333333333332,0.21,0.07200000000000001,0.33,0.04800000000000001,0.44,0.2076031746031746,0.26105510905389956,0.21643835848669174
+4.623044096728307,3250,0.14,0.24,0.34,0.44,0.14,0.14,0.08,0.23,0.07200000000000001,0.32,0.046000000000000006,0.41,0.22119047619047622,0.26532245418156675,0.23109250456606525
+4.978662873399715,3500,0.12,0.26,0.32,0.44,0.12,0.12,0.08666666666666666,0.23,0.068,0.3,0.046,0.41,0.21477777777777773,0.257064539977973,0.21964634746787756
+5.334281650071124,3750,0.1,0.24,0.36,0.42,0.1,0.09,0.08,0.22,0.076,0.34,0.044000000000000004,0.4,0.20072222222222222,0.24301056401214638,0.2021698992080138
+5.689900426742532,4000,0.14,0.24,0.3,0.44,0.14,0.12,0.08,0.21,0.064,0.28,0.046000000000000006,0.41,0.2249126984126984,0.2577199491386096,0.21657626401440466
+6.0455192034139404,4250,0.12,0.2,0.34,0.42,0.12,0.11,0.06666666666666667,0.18,0.07200000000000001,0.32,0.044000000000000004,0.39,0.2040238095238095,0.24225375425900011,0.20495077755179078
+6.401137980085348,4500,0.16,0.26,0.32,0.48,0.16,0.14,0.08666666666666666,0.23,0.068,0.3,0.05000000000000001,0.45,0.24446825396825397,0.28137008016271187,0.23233245654529827
+6.756756756756757,4750,0.12,0.24,0.32,0.46,0.12,0.11,0.08,0.22,0.068,0.3,0.04800000000000001,0.43,0.2144126984126984,0.2588958998549317,0.21224840804334918
+7.112375533428165,5000,0.12,0.24,0.32,0.46,0.12,0.11,0.08,0.22,0.068,0.3,0.04800000000000001,0.43,0.21046825396825397,0.2557017619639786,0.20778975258432592

eval/NanoBEIR_evaluation_mean_results.csv

@@ -317,3 +317,24 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 3.9118065433854907,2750,0.36,0.53,0.59,0.6699999999999999,0.36,0.345,0.18,0.52,0.122,0.585,0.07,0.6599999999999999,0.4600396825396825,0.5036950721129038,0.46497734901114707
 4.2674253200568995,3000,0.34,0.54,0.58,0.68,0.34,0.325,0.18333333333333335,0.53,0.11800000000000001,0.5700000000000001,0.07,0.665,0.4488174603174603,0.49482706303670787,0.45081521423911924
 4.623044096728307,3250,0.35,0.52,0.58,0.67,0.35,0.335,0.1766666666666667,0.51,0.11800000000000001,0.5700000000000001,0.069,0.655,0.4493571428571428,0.49289562811923704,0.4526207680739159
+0,0,0.42,0.5700000000000001,0.62,0.76,0.42,0.415,0.1933333333333333,0.56,0.128,0.61,0.07800000000000001,0.75,0.5219801587301587,0.5735324986310684,0.5279708783891941
+0.35561877667140823,250,0.25,0.45,0.51,0.6100000000000001,0.25,0.24,0.15666666666666665,0.44,0.10600000000000001,0.5,0.064,0.6000000000000001,0.3711587301587302,0.42410551240621813,0.38078198052356377
+0.7112375533428165,500,0.22,0.44,0.54,0.62,0.22,0.215,0.14999999999999997,0.435,0.11000000000000001,0.525,0.063,0.605,0.3537301587301587,0.4123372007183831,0.36108452217225534
+1.0668563300142249,750,0.23,0.44000000000000006,0.47,0.58,0.23,0.23,0.14666666666666667,0.43000000000000005,0.098,0.47,0.06000000000000001,0.58,0.3446349206349206,0.40189393231855164,0.35889086699701633
+1.422475106685633,1000,0.23,0.45,0.52,0.64,0.23,0.225,0.15333333333333332,0.45,0.10600000000000001,0.51,0.065,0.625,0.35706349206349197,0.42013754959630045,0.3650932275000509
+1.7780938833570412,1250,0.21000000000000002,0.4,0.49,0.62,0.21000000000000002,0.20500000000000002,0.1333333333333333,0.385,0.10000000000000002,0.48,0.063,0.6,0.33455952380952375,0.39456883865669734,0.339401967834268
+2.1337126600284497,1500,0.22999999999999998,0.41000000000000003,0.49,0.59,0.22999999999999998,0.22999999999999998,0.13666666666666666,0.395,0.1,0.48,0.060000000000000005,0.58,0.3421349206349207,0.3981079297263883,0.3517427207303351
+2.4893314366998576,1750,0.19,0.43000000000000005,0.46,0.5800000000000001,0.19,0.19,0.1433333333333333,0.42000000000000004,0.094,0.455,0.05900000000000001,0.575,0.318297619047619,0.3787378093270853,0.32736404938169256
+2.844950213371266,2000,0.19,0.37,0.44000000000000006,0.55,0.19,0.18,0.12333333333333332,0.355,0.09000000000000001,0.43000000000000005,0.05600000000000001,0.54,0.30376984126984125,0.3550088755354648,0.30854418174313203
+3.200568990042674,2250,0.2,0.36,0.49,0.59,0.2,0.195,0.12,0.35,0.1,0.475,0.06000000000000001,0.575,0.3133968253968254,0.37285368463232144,0.31996489194593924
+3.5561877667140824,2500,0.21000000000000002,0.34,0.5,0.62,0.21000000000000002,0.21000000000000002,0.11333333333333333,0.335,0.10200000000000001,0.485,0.063,0.605,0.3155119047619047,0.3819269865114655,0.32149923444212447
+3.9118065433854907,2750,0.22,0.38,0.5,0.59,0.22,0.22,0.12666666666666668,0.37,0.10200000000000001,0.485,0.060000000000000005,0.575,0.32886904761904756,0.38628825050266813,0.335907967131464
+4.2674253200568995,3000,0.19,0.32,0.45999999999999996,0.58,0.19,0.19,0.10666666666666666,0.315,0.09400000000000001,0.45499999999999996,0.05900000000000001,0.57,0.2965912698412698,0.3608361895056882,0.3057882061381056
+4.623044096728307,3250,0.22,0.37,0.44000000000000006,0.56,0.22,0.22,0.12333333333333335,0.365,0.09000000000000001,0.43000000000000005,0.05700000000000001,0.545,0.3163253968253968,0.37026467778587024,0.32650986326574444
+4.978662873399715,3500,0.16999999999999998,0.38,0.45999999999999996,0.56,0.16999999999999998,0.16999999999999998,0.12666666666666668,0.365,0.09400000000000001,0.44999999999999996,0.057,0.545,0.29701190476190475,0.3546715285540342,0.3049131952527186
+5.334281650071124,3750,0.18,0.38,0.47,0.5599999999999999,0.18,0.175,0.12666666666666665,0.37,0.096,0.45999999999999996,0.05700000000000001,0.55,0.3019126984126984,0.35874605705102597,0.3070392722383894
+5.689900426742532,4000,0.19,0.38,0.44999999999999996,0.57,0.19,0.18,0.12666666666666665,0.365,0.09200000000000001,0.44,0.05800000000000001,0.5549999999999999,0.3094523809523809,0.36283615504469824,0.30998555797833816
+6.0455192034139404,4250,0.18,0.35,0.47,0.5599999999999999,0.18,0.175,0.11666666666666667,0.33999999999999997,0.09600000000000002,0.45999999999999996,0.05700000000000001,0.5449999999999999,0.2975634920634921,0.35409053507274174,0.302742298047178
+6.401137980085348,4500,0.2,0.37,0.45999999999999996,0.59,0.2,0.19,0.12333333333333332,0.355,0.09400000000000001,0.44999999999999996,0.06000000000000001,0.575,0.31660714285714286,0.3725435929403427,0.31515286475131665
+6.756756756756757,4750,0.19,0.38,0.47,0.58,0.19,0.185,0.12666666666666665,0.37,0.096,0.45999999999999996,0.05900000000000001,0.565,0.3098452380952381,0.36764558004439085,0.3133177011044591
+7.112375533428165,5000,0.18,0.38,0.47,0.58,0.18,0.175,0.12666666666666665,0.37,0.096,0.45999999999999996,0.05900000000000001,0.565,0.3013174603174603,0.3611681627923753,0.3046382965528813

final_metrics.json

@@ -1,231 +1,231 @@
 {
   "nano_beir": {
-    "NanoClimateFEVER_cosine_accuracy@1": 0.3,
-    "NanoClimateFEVER_cosine_accuracy@3": 0.48,
-    "NanoClimateFEVER_cosine_accuracy@5": 0.56,
-    "NanoClimateFEVER_cosine_accuracy@10": 0.66,
-    "NanoClimateFEVER_cosine_precision@1": 0.3,
-    "NanoClimateFEVER_cosine_precision@3": 0.15999999999999998,
-    "NanoClimateFEVER_cosine_precision@5": 0.11600000000000002,
-    "NanoClimateFEVER_cosine_precision@10": 0.078,
-    "NanoClimateFEVER_cosine_recall@1": 0.12333333333333334,
-    "NanoClimateFEVER_cosine_recall@3": 0.21166666666666664,
-    "NanoClimateFEVER_cosine_recall@5": 0.2533333333333333,
-    "NanoClimateFEVER_cosine_recall@10": 0.325,
-    "NanoClimateFEVER_cosine_ndcg@10": 0.2735257041147495,
-    "NanoClimateFEVER_cosine_mrr@10": 0.406079365079365,
-    "NanoClimateFEVER_cosine_map@100": 0.21801163753351321,
-    "NanoDBPedia_cosine_accuracy@1": 0.7,
-    "NanoDBPedia_cosine_accuracy@3": 0.88,
-    "NanoDBPedia_cosine_accuracy@5": 0.9,
-    "NanoDBPedia_cosine_accuracy@10": 0.94,
-    "NanoDBPedia_cosine_precision@1": 0.7,
-    "NanoDBPedia_cosine_precision@3": 0.5533333333333332,
-    "NanoDBPedia_cosine_precision@5": 0.46399999999999997,
-    "NanoDBPedia_cosine_precision@10": 0.398,
-    "NanoDBPedia_cosine_recall@1": 0.05885278913396376,
-    "NanoDBPedia_cosine_recall@3": 0.1651324985031048,
-    "NanoDBPedia_cosine_recall@5": 0.1937322996063979,
-    "NanoDBPedia_cosine_recall@10": 0.278801784770749,
-    "NanoDBPedia_cosine_ndcg@10": 0.5090931324148577,
-    "NanoDBPedia_cosine_mrr@10": 0.7872222222222223,
-    "NanoDBPedia_cosine_map@100": 0.32792556078063057,
-    "NanoFEVER_cosine_accuracy@1": 0.86,
-    "NanoFEVER_cosine_accuracy@3": 0.94,
-    "NanoFEVER_cosine_accuracy@5": 0.96,
-    "NanoFEVER_cosine_accuracy@10": 0.98,
-    "NanoFEVER_cosine_precision@1": 0.86,
-    "NanoFEVER_cosine_precision@3": 0.3133333333333333,
-    "NanoFEVER_cosine_precision@5": 0.19599999999999995,
-    "NanoFEVER_cosine_precision@10": 0.102,
-    "NanoFEVER_cosine_recall@1": 0.7966666666666665,
-    "NanoFEVER_cosine_recall@3": 0.8766666666666666,
-    "NanoFEVER_cosine_recall@5": 0.9066666666666666,
-    "NanoFEVER_cosine_recall@10": 0.9366666666666668,
-    "NanoFEVER_cosine_ndcg@10": 0.8846483562973002,
-    "NanoFEVER_cosine_mrr@10": 0.9073333333333333,
-    "NanoFEVER_cosine_map@100": 0.8511817169008258,
-    "NanoFiQA2018_cosine_accuracy@1": 0.42,
-    "NanoFiQA2018_cosine_accuracy@3": 0.58,
-    "NanoFiQA2018_cosine_accuracy@5": 0.68,
-    "NanoFiQA2018_cosine_accuracy@10": 0.7,
-    "NanoFiQA2018_cosine_precision@1": 0.42,
-    "NanoFiQA2018_cosine_precision@3": 0.24666666666666667,
-    "NanoFiQA2018_cosine_precision@5": 0.19599999999999998,
-    "NanoFiQA2018_cosine_precision@10": 0.10799999999999998,
-    "NanoFiQA2018_cosine_recall@1": 0.22274603174603172,
-    "NanoFiQA2018_cosine_recall@3": 0.3608253968253968,
-    "NanoFiQA2018_cosine_recall@5": 0.4680714285714286,
-    "NanoFiQA2018_cosine_recall@10": 0.5110714285714285,
-    "NanoFiQA2018_cosine_ndcg@10": 0.4267909833793157,
-    "NanoFiQA2018_cosine_mrr@10": 0.512,
-    "NanoFiQA2018_cosine_map@100": 0.3624765177374492,
-    "NanoHotpotQA_cosine_accuracy@1": 0.82,
-    "NanoHotpotQA_cosine_accuracy@3": 0.88,
-    "NanoHotpotQA_cosine_accuracy@5": 0.92,
-    "NanoHotpotQA_cosine_accuracy@10": 0.94,
-    "NanoHotpotQA_cosine_precision@1": 0.82,
-    "NanoHotpotQA_cosine_precision@3": 0.3999999999999999,
-    "NanoHotpotQA_cosine_precision@5": 0.26,
-    "NanoHotpotQA_cosine_precision@10": 0.13799999999999998,
-    "NanoHotpotQA_cosine_recall@1": 0.41,
-    "NanoHotpotQA_cosine_recall@3": 0.6,
-    "NanoHotpotQA_cosine_recall@5": 0.65,
-    "NanoHotpotQA_cosine_recall@10": 0.69,
-    "NanoHotpotQA_cosine_ndcg@10": 0.6791736335059022,
-    "NanoHotpotQA_cosine_mrr@10": 0.8555555555555556,
-    "NanoHotpotQA_cosine_map@100": 0.6031157171874328,
-    "NanoMSMARCO_cosine_accuracy@1": 0.52,
-    "NanoMSMARCO_cosine_accuracy@3": 0.7,
-    "NanoMSMARCO_cosine_accuracy@5": 0.74,
-    "NanoMSMARCO_cosine_accuracy@10": 0.84,
-    "NanoMSMARCO_cosine_precision@1": 0.52,
-    "NanoMSMARCO_cosine_precision@3": 0.2333333333333333,
-    "NanoMSMARCO_cosine_precision@5": 0.14800000000000002,
-    "NanoMSMARCO_cosine_precision@10": 0.08399999999999999,
-    "NanoMSMARCO_cosine_recall@1": 0.52,
-    "NanoMSMARCO_cosine_recall@3": 0.7,
-    "NanoMSMARCO_cosine_recall@5": 0.74,
-    "NanoMSMARCO_cosine_recall@10": 0.84,
-    "NanoMSMARCO_cosine_ndcg@10": 0.6667666782328303,
-    "NanoMSMARCO_cosine_mrr@10": 0.6126587301587302,
-    "NanoMSMARCO_cosine_map@100": 0.6201688107214423,
-    "NanoNFCorpus_cosine_accuracy@1": 0.4,
-    "NanoNFCorpus_cosine_accuracy@3": 0.46,
-    "NanoNFCorpus_cosine_accuracy@5": 0.52,
-    "NanoNFCorpus_cosine_accuracy@10": 0.68,
-    "NanoNFCorpus_cosine_precision@1": 0.4,
-    "NanoNFCorpus_cosine_precision@3": 0.32666666666666666,
-    "NanoNFCorpus_cosine_precision@5": 0.304,
-    "NanoNFCorpus_cosine_precision@10": 0.282,
-    "NanoNFCorpus_cosine_recall@1": 0.013821074066170821,
-    "NanoNFCorpus_cosine_recall@3": 0.03704372187363592,
-    "NanoNFCorpus_cosine_recall@5": 0.05641224522012561,
-    "NanoNFCorpus_cosine_recall@10": 0.10199363274777667,
-    "NanoNFCorpus_cosine_ndcg@10": 0.30539493217594926,
-    "NanoNFCorpus_cosine_mrr@10": 0.460515873015873,
-    "NanoNFCorpus_cosine_map@100": 0.12327617976680419,
-    "NanoNQ_cosine_accuracy@1": 0.38,
-    "NanoNQ_cosine_accuracy@3": 0.56,
-    "NanoNQ_cosine_accuracy@5": 0.62,
+    "NanoClimateFEVER_cosine_accuracy@1": 0.24,
+    "NanoClimateFEVER_cosine_accuracy@3": 0.4,
+    "NanoClimateFEVER_cosine_accuracy@5": 0.46,
+    "NanoClimateFEVER_cosine_accuracy@10": 0.56,
+    "NanoClimateFEVER_cosine_precision@1": 0.24,
+    "NanoClimateFEVER_cosine_precision@3": 0.13333333333333333,
+    "NanoClimateFEVER_cosine_precision@5": 0.1,
+    "NanoClimateFEVER_cosine_precision@10": 0.064,
+    "NanoClimateFEVER_cosine_recall@1": 0.09233333333333332,
+    "NanoClimateFEVER_cosine_recall@3": 0.174,
+    "NanoClimateFEVER_cosine_recall@5": 0.22066666666666668,
+    "NanoClimateFEVER_cosine_recall@10": 0.27899999999999997,
+    "NanoClimateFEVER_cosine_ndcg@10": 0.2283726062129329,
+    "NanoClimateFEVER_cosine_mrr@10": 0.33938095238095245,
+    "NanoClimateFEVER_cosine_map@100": 0.17414775290079157,
+    "NanoDBPedia_cosine_accuracy@1": 0.58,
+    "NanoDBPedia_cosine_accuracy@3": 0.76,
+    "NanoDBPedia_cosine_accuracy@5": 0.88,
+    "NanoDBPedia_cosine_accuracy@10": 0.92,
+    "NanoDBPedia_cosine_precision@1": 0.58,
+    "NanoDBPedia_cosine_precision@3": 0.46666666666666673,
+    "NanoDBPedia_cosine_precision@5": 0.42,
+    "NanoDBPedia_cosine_precision@10": 0.364,
+    "NanoDBPedia_cosine_recall@1": 0.06197705417764409,
+    "NanoDBPedia_cosine_recall@3": 0.12223781146808171,
+    "NanoDBPedia_cosine_recall@5": 0.16808655070985257,
+    "NanoDBPedia_cosine_recall@10": 0.2608995958885433,
+    "NanoDBPedia_cosine_ndcg@10": 0.4521290486102402,
+    "NanoDBPedia_cosine_mrr@10": 0.6831904761904761,
+    "NanoDBPedia_cosine_map@100": 0.2791787003737884,
+    "NanoFEVER_cosine_accuracy@1": 0.68,
+    "NanoFEVER_cosine_accuracy@3": 0.84,
+    "NanoFEVER_cosine_accuracy@5": 0.9,
+    "NanoFEVER_cosine_accuracy@10": 0.94,
+    "NanoFEVER_cosine_precision@1": 0.68,
+    "NanoFEVER_cosine_precision@3": 0.28,
+    "NanoFEVER_cosine_precision@5": 0.184,
+    "NanoFEVER_cosine_precision@10": 0.09599999999999997,
+    "NanoFEVER_cosine_recall@1": 0.6566666666666666,
+    "NanoFEVER_cosine_recall@3": 0.7866666666666666,
+    "NanoFEVER_cosine_recall@5": 0.8566666666666666,
+    "NanoFEVER_cosine_recall@10": 0.8866666666666667,
+    "NanoFEVER_cosine_ndcg@10": 0.7797475710520765,
+    "NanoFEVER_cosine_mrr@10": 0.7660238095238096,
+    "NanoFEVER_cosine_map@100": 0.7347713746077809,
+    "NanoFiQA2018_cosine_accuracy@1": 0.24,
+    "NanoFiQA2018_cosine_accuracy@3": 0.34,
+    "NanoFiQA2018_cosine_accuracy@5": 0.46,
+    "NanoFiQA2018_cosine_accuracy@10": 0.56,
+    "NanoFiQA2018_cosine_precision@1": 0.24,
+    "NanoFiQA2018_cosine_precision@3": 0.13999999999999999,
+    "NanoFiQA2018_cosine_precision@5": 0.11200000000000002,
+    "NanoFiQA2018_cosine_precision@10": 0.08,
+    "NanoFiQA2018_cosine_recall@1": 0.14085714285714285,
+    "NanoFiQA2018_cosine_recall@3": 0.19988095238095238,
+    "NanoFiQA2018_cosine_recall@5": 0.30054761904761906,
+    "NanoFiQA2018_cosine_recall@10": 0.38654761904761903,
+    "NanoFiQA2018_cosine_ndcg@10": 0.28855478051901995,
+    "NanoFiQA2018_cosine_mrr@10": 0.3230238095238096,
+    "NanoFiQA2018_cosine_map@100": 0.23168919640832794,
+    "NanoHotpotQA_cosine_accuracy@1": 0.62,
+    "NanoHotpotQA_cosine_accuracy@3": 0.74,
+    "NanoHotpotQA_cosine_accuracy@5": 0.76,
+    "NanoHotpotQA_cosine_accuracy@10": 0.8,
+    "NanoHotpotQA_cosine_precision@1": 0.62,
+    "NanoHotpotQA_cosine_precision@3": 0.29333333333333333,
+    "NanoHotpotQA_cosine_precision@5": 0.18799999999999997,
+    "NanoHotpotQA_cosine_precision@10": 0.10599999999999998,
+    "NanoHotpotQA_cosine_recall@1": 0.31,
+    "NanoHotpotQA_cosine_recall@3": 0.44,
+    "NanoHotpotQA_cosine_recall@5": 0.47,
+    "NanoHotpotQA_cosine_recall@10": 0.53,
+    "NanoHotpotQA_cosine_ndcg@10": 0.508052717401943,
+    "NanoHotpotQA_cosine_mrr@10": 0.676,
+    "NanoHotpotQA_cosine_map@100": 0.43060781600475145,
+    "NanoMSMARCO_cosine_accuracy@1": 0.34,
+    "NanoMSMARCO_cosine_accuracy@3": 0.5,
+    "NanoMSMARCO_cosine_accuracy@5": 0.58,
+    "NanoMSMARCO_cosine_accuracy@10": 0.68,
+    "NanoMSMARCO_cosine_precision@1": 0.34,
+    "NanoMSMARCO_cosine_precision@3": 0.16666666666666669,
+    "NanoMSMARCO_cosine_precision@5": 0.11599999999999999,
+    "NanoMSMARCO_cosine_precision@10": 0.068,
+    "NanoMSMARCO_cosine_recall@1": 0.34,
+    "NanoMSMARCO_cosine_recall@3": 0.5,
+    "NanoMSMARCO_cosine_recall@5": 0.58,
+    "NanoMSMARCO_cosine_recall@10": 0.68,
+    "NanoMSMARCO_cosine_ndcg@10": 0.5023070028455473,
+    "NanoMSMARCO_cosine_mrr@10": 0.44646825396825385,
+    "NanoMSMARCO_cosine_map@100": 0.45949436084808376,
+    "NanoNFCorpus_cosine_accuracy@1": 0.3,
+    "NanoNFCorpus_cosine_accuracy@3": 0.4,
+    "NanoNFCorpus_cosine_accuracy@5": 0.48,
+    "NanoNFCorpus_cosine_accuracy@10": 0.56,
+    "NanoNFCorpus_cosine_precision@1": 0.3,
+    "NanoNFCorpus_cosine_precision@3": 0.28,
+    "NanoNFCorpus_cosine_precision@5": 0.276,
+    "NanoNFCorpus_cosine_precision@10": 0.242,
+    "NanoNFCorpus_cosine_recall@1": 0.010085921966980407,
+    "NanoNFCorpus_cosine_recall@3": 0.047968018726256945,
+    "NanoNFCorpus_cosine_recall@5": 0.0671856089688558,
+    "NanoNFCorpus_cosine_recall@10": 0.1033851527560607,
+    "NanoNFCorpus_cosine_ndcg@10": 0.2699339552042321,
+    "NanoNFCorpus_cosine_mrr@10": 0.3788888888888888,
+    "NanoNFCorpus_cosine_map@100": 0.10920681945613339,
+    "NanoNQ_cosine_accuracy@1": 0.36,
+    "NanoNQ_cosine_accuracy@3": 0.54,
+    "NanoNQ_cosine_accuracy@5": 0.56,
     "NanoNQ_cosine_accuracy@10": 0.7,
-    "NanoNQ_cosine_precision@1": 0.38,
-    "NanoNQ_cosine_precision@3": 0.19333333333333333,
-    "NanoNQ_cosine_precision@5": 0.132,
+    "NanoNQ_cosine_precision@1": 0.36,
+    "NanoNQ_cosine_precision@3": 0.18666666666666668,
+    "NanoNQ_cosine_precision@5": 0.11599999999999999,
     "NanoNQ_cosine_precision@10": 0.07400000000000001,
-    "NanoNQ_cosine_recall@1": 0.35,
-    "NanoNQ_cosine_recall@3": 0.53,
-    "NanoNQ_cosine_recall@5": 0.6,
+    "NanoNQ_cosine_recall@1": 0.33,
+    "NanoNQ_cosine_recall@3": 0.52,
+    "NanoNQ_cosine_recall@5": 0.54,
     "NanoNQ_cosine_recall@10": 0.66,
-    "NanoNQ_cosine_ndcg@10": 0.5159192638156053,
-    "NanoNQ_cosine_mrr@10": 0.48483333333333334,
-    "NanoNQ_cosine_map@100": 0.47956629357668895,
-    "NanoQuoraRetrieval_cosine_accuracy@1": 0.86,
-    "NanoQuoraRetrieval_cosine_accuracy@3": 0.98,
-    "NanoQuoraRetrieval_cosine_accuracy@5": 0.98,
-    "NanoQuoraRetrieval_cosine_accuracy@10": 1.0,
-    "NanoQuoraRetrieval_cosine_precision@1": 0.86,
-    "NanoQuoraRetrieval_cosine_precision@3": 0.4133333333333333,
-    "NanoQuoraRetrieval_cosine_precision@5": 0.26,
-    "NanoQuoraRetrieval_cosine_precision@10": 0.13599999999999998,
-    "NanoQuoraRetrieval_cosine_recall@1": 0.7373333333333332,
-    "NanoQuoraRetrieval_cosine_recall@3": 0.9520000000000001,
-    "NanoQuoraRetrieval_cosine_recall@5": 0.9666666666666666,
-    "NanoQuoraRetrieval_cosine_recall@10": 0.9933333333333334,
-    "NanoQuoraRetrieval_cosine_ndcg@10": 0.92794886246839,
-    "NanoQuoraRetrieval_cosine_mrr@10": 0.9191666666666666,
-    "NanoQuoraRetrieval_cosine_map@100": 0.9003272642390289,
-    "NanoSCIDOCS_cosine_accuracy@1": 0.36,
-    "NanoSCIDOCS_cosine_accuracy@3": 0.58,
-    "NanoSCIDOCS_cosine_accuracy@5": 0.64,
-    "NanoSCIDOCS_cosine_accuracy@10": 0.68,
-    "NanoSCIDOCS_cosine_precision@1": 0.36,
-    "NanoSCIDOCS_cosine_precision@3": 0.2866666666666666,
-    "NanoSCIDOCS_cosine_precision@5": 0.26,
-    "NanoSCIDOCS_cosine_precision@10": 0.156,
-    "NanoSCIDOCS_cosine_recall@1": 0.07466666666666667,
-    "NanoSCIDOCS_cosine_recall@3": 0.17666666666666667,
-    "NanoSCIDOCS_cosine_recall@5": 0.26466666666666666,
-    "NanoSCIDOCS_cosine_recall@10": 0.31966666666666665,
-    "NanoSCIDOCS_cosine_ndcg@10": 0.31516617152348114,
-    "NanoSCIDOCS_cosine_mrr@10": 0.47855555555555546,
-    "NanoSCIDOCS_cosine_map@100": 0.2537533171047557,
-    "NanoArguAna_cosine_accuracy@1": 0.24,
-    "NanoArguAna_cosine_accuracy@3": 0.56,
-    "NanoArguAna_cosine_accuracy@5": 0.72,
-    "NanoArguAna_cosine_accuracy@10": 0.9,
-    "NanoArguAna_cosine_precision@1": 0.24,
-    "NanoArguAna_cosine_precision@3": 0.18666666666666668,
-    "NanoArguAna_cosine_precision@5": 0.14400000000000002,
-    "NanoArguAna_cosine_precision@10": 0.08999999999999998,
-    "NanoArguAna_cosine_recall@1": 0.24,
-    "NanoArguAna_cosine_recall@3": 0.56,
-    "NanoArguAna_cosine_recall@5": 0.72,
-    "NanoArguAna_cosine_recall@10": 0.9,
-    "NanoArguAna_cosine_ndcg@10": 0.5546300286352132,
-    "NanoArguAna_cosine_mrr@10": 0.4462380952380951,
-    "NanoArguAna_cosine_map@100": 0.45259153346653347,
-    "NanoSciFact_cosine_accuracy@1": 0.54,
-    "NanoSciFact_cosine_accuracy@3": 0.7,
-    "NanoSciFact_cosine_accuracy@5": 0.76,
-    "NanoSciFact_cosine_accuracy@10": 0.84,
-    "NanoSciFact_cosine_precision@1": 0.54,
-    "NanoSciFact_cosine_precision@3": 0.2533333333333333,
-    "NanoSciFact_cosine_precision@5": 0.17599999999999993,
-    "NanoSciFact_cosine_precision@10": 0.09599999999999997,
-    "NanoSciFact_cosine_recall@1": 0.505,
-    "NanoSciFact_cosine_recall@3": 0.68,
-    "NanoSciFact_cosine_recall@5": 0.76,
-    "NanoSciFact_cosine_recall@10": 0.84,
-    "NanoSciFact_cosine_ndcg@10": 0.6844325567315029,
-    "NanoSciFact_cosine_mrr@10": 0.6369126984126985,
-    "NanoSciFact_cosine_map@100": 0.6389995767091675,
-    "NanoTouche2020_cosine_accuracy@1": 0.6122448979591837,
-    "NanoTouche2020_cosine_accuracy@3": 0.8367346938775511,
-    "NanoTouche2020_cosine_accuracy@5": 0.9387755102040817,
-    "NanoTouche2020_cosine_accuracy@10": 0.9795918367346939,
-    "NanoTouche2020_cosine_precision@1": 0.6122448979591837,
-    "NanoTouche2020_cosine_precision@3": 0.5238095238095238,
-    "NanoTouche2020_cosine_precision@5": 0.48979591836734704,
-    "NanoTouche2020_cosine_precision@10": 0.41836734693877553,
-    "NanoTouche2020_cosine_recall@1": 0.04261846159526855,
-    "NanoTouche2020_cosine_recall@3": 0.11310802492381491,
-    "NanoTouche2020_cosine_recall@5": 0.17549525482392914,
-    "NanoTouche2020_cosine_recall@10": 0.285598908591513,
-    "NanoTouche2020_cosine_ndcg@10": 0.4763620784420955,
-    "NanoTouche2020_cosine_mrr@10": 0.7421768707482992,
-    "NanoTouche2020_cosine_map@100": 0.357725600182186,
-    "NanoBEIR_mean_cosine_accuracy@1": 0.5394034536891681,
-    "NanoBEIR_mean_cosine_accuracy@3": 0.7028257456828886,
-    "NanoBEIR_mean_cosine_accuracy@5": 0.7645211930926217,
-    "NanoBEIR_mean_cosine_accuracy@10": 0.8338147566718994,
-    "NanoBEIR_mean_cosine_precision@1": 0.5394034536891681,
-    "NanoBEIR_mean_cosine_precision@3": 0.31465201465201464,
-    "NanoBEIR_mean_cosine_precision@5": 0.24198430141287283,
-    "NanoBEIR_mean_cosine_precision@10": 0.166182103610675,
-    "NanoBEIR_mean_cosine_recall@1": 0.3150029505031873,
-    "NanoBEIR_mean_cosine_recall@3": 0.45870074170199626,
-    "NanoBEIR_mean_cosine_recall@5": 0.5196188124273241,
-    "NanoBEIR_mean_cosine_recall@10": 0.5909332631806257,
-    "NanoBEIR_mean_cosine_ndcg@10": 0.5553732601336302,
-    "NanoBEIR_mean_cosine_mrr@10": 0.634557561486133,
-    "NanoBEIR_mean_cosine_map@100": 0.47608613276203526
+    "NanoNQ_cosine_ndcg@10": 0.4933846471447955,
+    "NanoNQ_cosine_mrr@10": 0.4570555555555556,
+    "NanoNQ_cosine_map@100": 0.44680861443860365,
+    "NanoQuoraRetrieval_cosine_accuracy@1": 0.78,
+    "NanoQuoraRetrieval_cosine_accuracy@3": 0.92,
+    "NanoQuoraRetrieval_cosine_accuracy@5": 0.96,
+    "NanoQuoraRetrieval_cosine_accuracy@10": 0.96,
+    "NanoQuoraRetrieval_cosine_precision@1": 0.78,
+    "NanoQuoraRetrieval_cosine_precision@3": 0.3666666666666666,
+    "NanoQuoraRetrieval_cosine_precision@5": 0.24,
+    "NanoQuoraRetrieval_cosine_precision@10": 0.12799999999999997,
+    "NanoQuoraRetrieval_cosine_recall@1": 0.6773333333333332,
+    "NanoQuoraRetrieval_cosine_recall@3": 0.8713333333333333,
+    "NanoQuoraRetrieval_cosine_recall@5": 0.9279999999999999,
+    "NanoQuoraRetrieval_cosine_recall@10": 0.946,
+    "NanoQuoraRetrieval_cosine_ndcg@10": 0.8656777707479286,
+    "NanoQuoraRetrieval_cosine_mrr@10": 0.8556666666666666,
+    "NanoQuoraRetrieval_cosine_map@100": 0.8321868686868686,
+    "NanoSCIDOCS_cosine_accuracy@1": 0.44,
+    "NanoSCIDOCS_cosine_accuracy@3": 0.64,
+    "NanoSCIDOCS_cosine_accuracy@5": 0.7,
+    "NanoSCIDOCS_cosine_accuracy@10": 0.82,
+    "NanoSCIDOCS_cosine_precision@1": 0.44,
+    "NanoSCIDOCS_cosine_precision@3": 0.32666666666666666,
+    "NanoSCIDOCS_cosine_precision@5": 0.268,
+    "NanoSCIDOCS_cosine_precision@10": 0.174,
+    "NanoSCIDOCS_cosine_recall@1": 0.09000000000000002,
+    "NanoSCIDOCS_cosine_recall@3": 0.19799999999999998,
+    "NanoSCIDOCS_cosine_recall@5": 0.27099999999999996,
+    "NanoSCIDOCS_cosine_recall@10": 0.3536666666666666,
+    "NanoSCIDOCS_cosine_ndcg@10": 0.3525296369677544,
+    "NanoSCIDOCS_cosine_mrr@10": 0.5576904761904762,
+    "NanoSCIDOCS_cosine_map@100": 0.2706982716239424,
+    "NanoArguAna_cosine_accuracy@1": 0.18,
+    "NanoArguAna_cosine_accuracy@3": 0.38,
+    "NanoArguAna_cosine_accuracy@5": 0.46,
+    "NanoArguAna_cosine_accuracy@10": 0.62,
+    "NanoArguAna_cosine_precision@1": 0.18,
+    "NanoArguAna_cosine_precision@3": 0.12666666666666665,
+    "NanoArguAna_cosine_precision@5": 0.09200000000000001,
+    "NanoArguAna_cosine_precision@10": 0.06200000000000001,
+    "NanoArguAna_cosine_recall@1": 0.18,
+    "NanoArguAna_cosine_recall@3": 0.38,
+    "NanoArguAna_cosine_recall@5": 0.46,
+    "NanoArguAna_cosine_recall@10": 0.62,
+    "NanoArguAna_cosine_ndcg@10": 0.37737911678224656,
+    "NanoArguAna_cosine_mrr@10": 0.3023571428571429,
+    "NanoArguAna_cosine_map@100": 0.31860599951297625,
+    "NanoSciFact_cosine_accuracy@1": 0.48,
+    "NanoSciFact_cosine_accuracy@3": 0.54,
+    "NanoSciFact_cosine_accuracy@5": 0.62,
+    "NanoSciFact_cosine_accuracy@10": 0.68,
+    "NanoSciFact_cosine_precision@1": 0.48,
+    "NanoSciFact_cosine_precision@3": 0.19333333333333336,
+    "NanoSciFact_cosine_precision@5": 0.14,
+    "NanoSciFact_cosine_precision@10": 0.08,
+    "NanoSciFact_cosine_recall@1": 0.445,
+    "NanoSciFact_cosine_recall@3": 0.52,
+    "NanoSciFact_cosine_recall@5": 0.6,
+    "NanoSciFact_cosine_recall@10": 0.68,
+    "NanoSciFact_cosine_ndcg@10": 0.5618706429753196,
+    "NanoSciFact_cosine_mrr@10": 0.5318571428571428,
+    "NanoSciFact_cosine_map@100": 0.5267835796727546,
+    "NanoTouche2020_cosine_accuracy@1": 0.4897959183673469,
+    "NanoTouche2020_cosine_accuracy@3": 0.7755102040816326,
+    "NanoTouche2020_cosine_accuracy@5": 0.8367346938775511,
+    "NanoTouche2020_cosine_accuracy@10": 1.0,
+    "NanoTouche2020_cosine_precision@1": 0.4897959183673469,
+    "NanoTouche2020_cosine_precision@3": 0.46258503401360546,
+    "NanoTouche2020_cosine_precision@5": 0.436734693877551,
+    "NanoTouche2020_cosine_precision@10": 0.3816326530612244,
+    "NanoTouche2020_cosine_recall@1": 0.03643545588582757,
+    "NanoTouche2020_cosine_recall@3": 0.09797151935416279,
+    "NanoTouche2020_cosine_recall@5": 0.15541127788533005,
+    "NanoTouche2020_cosine_recall@10": 0.26155265982169057,
+    "NanoTouche2020_cosine_ndcg@10": 0.4247890449649847,
+    "NanoTouche2020_cosine_mrr@10": 0.6553206997084547,
+    "NanoTouche2020_cosine_map@100": 0.31946609885183697,
+    "NanoBEIR_mean_cosine_accuracy@1": 0.4407535321821036,
+    "NanoBEIR_mean_cosine_accuracy@3": 0.5981161695447409,
+    "NanoBEIR_mean_cosine_accuracy@5": 0.6659026687598116,
+    "NanoBEIR_mean_cosine_accuracy@10": 0.7538461538461539,
+    "NanoBEIR_mean_cosine_precision@1": 0.4407535321821036,
+    "NanoBEIR_mean_cosine_precision@3": 0.2632757718472004,
+    "NanoBEIR_mean_cosine_precision@5": 0.20682574568288856,
+    "NanoBEIR_mean_cosine_precision@10": 0.1476640502354788,
+    "NanoBEIR_mean_cosine_recall@1": 0.25928376217084065,
+    "NanoBEIR_mean_cosine_recall@3": 0.3736967924561118,
+    "NanoBEIR_mean_cosine_recall@5": 0.4321203376880761,
+    "NanoBEIR_mean_cosine_recall@10": 0.5113629508344035,
+    "NanoBEIR_mean_cosine_ndcg@10": 0.46959450318684787,
+    "NanoBEIR_mean_cosine_mrr@10": 0.536378759562433,
+    "NanoBEIR_mean_cosine_map@100": 0.3948958041066646
   },
   "beir_touche2020": {
     "BeIR-touche2020-subset-test_cosine_accuracy@1": 0.7346938775510204,
-    "BeIR-touche2020-subset-test_cosine_accuracy@3": 1.0,
-    "BeIR-touche2020-subset-test_cosine_accuracy@5": 1.0,
+    "BeIR-touche2020-subset-test_cosine_accuracy@3": 0.9183673469387755,
+    "BeIR-touche2020-subset-test_cosine_accuracy@5": 0.9591836734693877,
     "BeIR-touche2020-subset-test_cosine_accuracy@10": 1.0,
     "BeIR-touche2020-subset-test_cosine_precision@1": 0.7346938775510204,
-    "BeIR-touche2020-subset-test_cosine_precision@3": 0.7482993197278911,
-    "BeIR-touche2020-subset-test_cosine_precision@5": 0.7306122448979593,
-    "BeIR-touche2020-subset-test_cosine_precision@10": 0.6448979591836734,
-    "BeIR-touche2020-subset-test_cosine_recall@1": 0.016356718324589865,
-    "BeIR-touche2020-subset-test_cosine_recall@3": 0.04982275407454,
-    "BeIR-touche2020-subset-test_cosine_recall@5": 0.08077270192430855,
-    "BeIR-touche2020-subset-test_cosine_recall@10": 0.14251827101536743,
-    "BeIR-touche2020-subset-test_cosine_ndcg@10": 0.6762844931311536,
-    "BeIR-touche2020-subset-test_cosine_mrr@10": 0.8605442176870749,
-    "BeIR-touche2020-subset-test_cosine_map@100": 0.3020322355884379
+    "BeIR-touche2020-subset-test_cosine_precision@3": 0.673469387755102,
+    "BeIR-touche2020-subset-test_cosine_precision@5": 0.6571428571428573,
+    "BeIR-touche2020-subset-test_cosine_precision@10": 0.589795918367347,
+    "BeIR-touche2020-subset-test_cosine_recall@1": 0.016137830046673268,
+    "BeIR-touche2020-subset-test_cosine_recall@3": 0.04465470442976813,
+    "BeIR-touche2020-subset-test_cosine_recall@5": 0.07229868006332099,
+    "BeIR-touche2020-subset-test_cosine_recall@10": 0.129926645587027,
+    "BeIR-touche2020-subset-test_cosine_ndcg@10": 0.6198727729504082,
+    "BeIR-touche2020-subset-test_cosine_mrr@10": 0.8311791383219955,
+    "BeIR-touche2020-subset-test_cosine_map@100": 0.2553079711703241
   }
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8fa9751e6dcd158f58cd4717f43644525ce470ad5861aac94efda4975c7c7e7d
-size 133462128
+oid sha256:dfe76e23f2842b7cac23a3a74b8569f8453decb600f9cf6ec03f49594f4c341c
+size 90864192

tokenizer_config.json

@@ -41,14 +41,14 @@
       "special": true
     }
   },
-  "clean_up_tokenization_spaces": true,
+  "clean_up_tokenization_spaces": false,
   "cls_token": "[CLS]",
   "do_basic_tokenize": true,
   "do_lower_case": true,
   "extra_special_tokens": {},
   "mask_token": "[MASK]",
   "max_length": 128,
-  "model_max_length": 512,
+  "model_max_length": 256,
   "never_split": null,
   "pad_to_multiple_of": null,
   "pad_token": "[PAD]",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:21afb743534b7424507a526e2550df4c3987619a1501a425fa2ecb589f2233bd
+oid sha256:a64e439040239d74da37488a3926746dd4e931304cae90d2a2fb152678d0e077
 size 6161