Training in progress, step 3000

Information-Retrieval_evaluation_BeIR-touche2020-subset-test_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.7346938775510204,0.8979591836734694,0.9387755102040817,1.0,0.7346938775510204,0.016235921803344835,0.6190476190476191,0.04110136889290122,0.6204081632653061,0.0684882416081836,0.5448979591836735,0.12060452557965035,0.8306932296728216,0.5779837170662205,0.22239268054909309
 -1,-1,0.8163265306122449,0.9183673469387755,0.9183673469387755,0.9795918367346939,0.8163265306122449,0.018031980909560562,0.6666666666666666,0.043898628557590365,0.6612244897959183,0.07253188549214812,0.5959183673469388,0.13104227957508152,0.8707482993197279,0.633159993828728,0.2756491953525919
 -1,-1,0.8775510204081632,1.0,1.0,1.0,0.8775510204081632,0.01945012749314946,0.8095238095238094,0.0537432819678873,0.7510204081632652,0.08295390754493644,0.6448979591836734,0.14269962051162913,0.9387755102040817,0.698044370091445,0.29990375282881515
+-1,-1,0.8367346938775511,0.9591836734693877,1.0,1.0,0.8367346938775511,0.01863536778338729,0.7823129251700679,0.05202492032020825,0.7306122448979592,0.08076708716823142,0.6632653061224489,0.14656488113887348,0.9047619047619047,0.7013835872612906,0.3127573985913156

Information-Retrieval_evaluation_NanoArguAna_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.0,0.04,0.06,0.1,0.0,0.0,0.013333333333333332,0.04,0.012000000000000002,0.06,0.01,0.1,0.031190476190476192,0.04764153171315327,0.034911665265567264
 -1,-1,0.24,0.38,0.52,0.58,0.24,0.24,0.12666666666666665,0.38,0.10400000000000002,0.52,0.05800000000000001,0.58,0.3347142857142857,0.39334913031939406,0.34861120264343276
 -1,-1,0.22,0.46,0.72,0.84,0.22,0.22,0.15333333333333332,0.46,0.14400000000000002,0.72,0.08399999999999999,0.84,0.4013571428571428,0.5065089296905451,0.40883602671837965
+-1,-1,0.26,0.52,0.7,0.82,0.26,0.26,0.1733333333333333,0.52,0.14,0.7,0.08199999999999999,0.82,0.425047619047619,0.5194504645258333,0.43281349206349207

Information-Retrieval_evaluation_NanoClimateFEVER_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.04,0.08,0.12,0.12,0.04,0.010666666666666666,0.03333333333333333,0.04466666666666667,0.028000000000000004,0.056333333333333326,0.014000000000000002,0.056333333333333326,0.06566666666666666,0.046191957495871135,0.033420280683900884
 -1,-1,0.16,0.3,0.38,0.46,0.16,0.062333333333333324,0.1,0.14066666666666663,0.07600000000000001,0.18233333333333335,0.05,0.22733333333333333,0.2518888888888889,0.17511491667849083,0.13026789173087744
 -1,-1,0.22,0.38,0.46,0.68,0.22,0.10666666666666666,0.14,0.2,0.10400000000000001,0.244,0.076,0.31799999999999995,0.3376349206349205,0.25307254805545965,0.19775113367023567
+-1,-1,0.22,0.42,0.5,0.68,0.22,0.08833333333333332,0.13999999999999999,0.21066666666666667,0.10800000000000001,0.24566666666666667,0.08,0.3423333333333333,0.34532539682539676,0.2564996471347133,0.18649579530729996

Information-Retrieval_evaluation_NanoDBPedia_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.26,0.36,0.42,0.48,0.26,0.013478860639269325,0.19333333333333333,0.027032310773340756,0.168,0.03720266654458638,0.138,0.061932122340730054,0.32241269841269843,0.1613933337162208,0.058008726432844
 -1,-1,0.6,0.84,0.9,0.94,0.6,0.07736452132069595,0.54,0.1648929452761373,0.46,0.20685193791339937,0.37800000000000006,0.27282350703551705,0.7216904761904763,0.49263161688380896,0.32681788162300135
 -1,-1,0.7,0.84,0.86,0.9,0.7,0.07514830114967352,0.48,0.12504263439749053,0.43199999999999994,0.17114850325738726,0.374,0.2647379890126719,0.7688571428571429,0.4819449351312035,0.3187663378243811
+-1,-1,0.72,0.84,0.9,0.96,0.72,0.05670459269763173,0.5266666666666666,0.13069601271168413,0.47200000000000003,0.19499547635399148,0.382,0.2892940743088388,0.7988571428571428,0.498556918096642,0.32899455637485014

Information-Retrieval_evaluation_NanoFEVER_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.02,0.02,0.02,0.04,0.02,0.02,0.006666666666666666,0.02,0.004,0.02,0.004,0.04,0.022857142857142857,0.026666666666666665,0.027169926585108667
 -1,-1,0.68,0.88,0.88,0.96,0.68,0.6366666666666666,0.3,0.8266666666666667,0.18,0.8266666666666667,0.09799999999999998,0.9066666666666667,0.773968253968254,0.7814677857530928,0.7300793650793651
 -1,-1,0.62,0.92,0.98,0.98,0.62,0.5966666666666667,0.3133333333333333,0.8666666666666667,0.19999999999999996,0.9266666666666667,0.09999999999999998,0.9266666666666667,0.7673333333333333,0.7904169146327766,0.7345681818181817
+-1,-1,0.68,0.92,0.92,0.94,0.68,0.6266666666666666,0.3133333333333333,0.8666666666666666,0.18799999999999997,0.8666666666666666,0.09599999999999997,0.8866666666666667,0.7833333333333334,0.7835360079677364,0.738962481962482

Information-Retrieval_evaluation_NanoFiQA2018_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.1,0.22,0.26,0.32,0.1,0.04352380952380952,0.07333333333333332,0.11152380952380953,0.05600000000000001,0.12774603174603175,0.036000000000000004,0.15663492063492065,0.16444444444444442,0.11920019095324214,0.08946856517346266
 -1,-1,0.28,0.38,0.44,0.62,0.28,0.13474603174603175,0.16666666666666663,0.21912698412698411,0.124,0.2721269841269841,0.08599999999999998,0.40484920634920635,0.36090476190476184,0.3007863756543914,0.23923426165260125
 -1,-1,0.4,0.48,0.56,0.64,0.4,0.21357936507936512,0.20666666666666667,0.2997698412698413,0.14800000000000002,0.34396031746031747,0.09599999999999997,0.4287936507936508,0.46438095238095245,0.37187924093070407,0.3143221865619557
+-1,-1,0.38,0.54,0.68,0.74,0.38,0.2114126984126984,0.2333333333333333,0.3536269841269841,0.18,0.4336269841269841,0.12,0.5354047619047619,0.4913888888888889,0.42968312791198054,0.35462063324400034

Information-Retrieval_evaluation_NanoHotpotQA_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.16,0.2,0.24,0.24,0.16,0.08,0.06666666666666667,0.1,0.04800000000000001,0.12,0.026000000000000002,0.13,0.18466666666666667,0.12500477316127157,0.09905861704269486
 -1,-1,0.72,0.8,0.82,0.92,0.72,0.36,0.34,0.51,0.23600000000000002,0.59,0.13399999999999998,0.67,0.773690476190476,0.617986346909483,0.5288841730049509
 -1,-1,0.7,0.86,0.88,0.92,0.7,0.35,0.37999999999999995,0.57,0.256,0.64,0.142,0.71,0.7835238095238095,0.6576214974716089,0.5819268475152912
+-1,-1,0.6,0.74,0.82,0.86,0.6,0.3,0.3,0.45,0.204,0.51,0.11599999999999998,0.58,0.6838571428571428,0.5356886314565849,0.45644680388890113

Information-Retrieval_evaluation_NanoMSMARCO_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.16,0.3,0.32,0.4,0.16,0.16,0.09999999999999998,0.3,0.064,0.32,0.04000000000000001,0.4,0.24177777777777776,0.2797381138841459,0.25452652172848145
 -1,-1,0.3,0.58,0.6,0.68,0.3,0.3,0.19333333333333333,0.58,0.12000000000000002,0.6,0.068,0.68,0.43527777777777776,0.4950369328373354,0.4475531768839056
 -1,-1,0.3,0.54,0.62,0.76,0.3,0.3,0.18,0.54,0.124,0.62,0.07600000000000001,0.76,0.4492698412698413,0.5241190384704345,0.45777964902887497
+-1,-1,0.32,0.52,0.58,0.7,0.32,0.32,0.1733333333333333,0.52,0.11599999999999999,0.58,0.07,0.7,0.43693650793650785,0.49929515877847647,0.45205401276376506

Information-Retrieval_evaluation_NanoNFCorpus_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.36,0.46,0.46,0.54,0.36,0.01397871707271983,0.3,0.03435945376327838,0.21600000000000003,0.03726142528548906,0.16800000000000004,0.048208769639288225,0.42138888888888887,0.21086706102148342,0.057209731275113765
 -1,-1,0.4,0.44,0.5,0.62,0.4,0.03202235376151782,0.3133333333333333,0.050501396728576585,0.312,0.07479533924341909,0.268,0.11100199043980967,0.4450476190476191,0.307499683018853,0.12958379529052644
 -1,-1,0.36,0.48,0.5,0.58,0.36,0.030515431849843816,0.32666666666666666,0.053102293373951406,0.29200000000000004,0.06596373325276159,0.264,0.10180141189599459,0.43479365079365073,0.3017216559964908,0.1219166619490478
+-1,-1,0.38,0.5,0.6,0.72,0.38,0.013857792729417237,0.3133333333333333,0.053976395496446575,0.3,0.08888025754448707,0.28600000000000003,0.12308809405420085,0.4709365079365079,0.3118912881405269,0.11679455794345621

Information-Retrieval_evaluation_NanoNQ_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.1,0.12,0.16,0.26,0.1,0.08,0.04,0.09,0.032,0.13,0.026000000000000002,0.23,0.12794444444444442,0.13826388457522362,0.10725400911912068
 -1,-1,0.26,0.48,0.52,0.64,0.26,0.24,0.16666666666666663,0.45,0.10800000000000001,0.49,0.066,0.6,0.3892142857142856,0.4279054208986469,0.3750113241088494
 -1,-1,0.38,0.52,0.54,0.68,0.38,0.35,0.1733333333333333,0.49,0.11200000000000002,0.52,0.07200000000000001,0.66,0.46857142857142864,0.5017561161582912,0.4585943213547632
+-1,-1,0.4,0.6,0.6,0.64,0.4,0.38,0.21333333333333332,0.58,0.128,0.58,0.07,0.63,0.49449999999999994,0.526038841299356,0.5043233378343615

Information-Retrieval_evaluation_NanoQuoraRetrieval_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.88,0.96,0.98,0.98,0.88,0.7573333333333332,0.40666666666666657,0.9313333333333332,0.26399999999999996,0.9686666666666668,0.13799999999999998,0.98,0.9216666666666665,0.933276206466488,0.9143118773946362
 -1,-1,0.9,0.98,1.0,1.0,0.9,0.7973333333333333,0.4133333333333333,0.9520000000000001,0.264,0.986,0.13999999999999999,1.0,0.9440000000000001,0.9559704430711728,0.9369603174603175
 -1,-1,0.78,0.96,0.96,0.98,0.78,0.684,0.38666666666666655,0.912,0.25199999999999995,0.9393333333333334,0.132,0.97,0.8622222222222223,0.8825493396502728,0.8476349206349206
+-1,-1,0.84,0.94,0.98,0.98,0.84,0.7440000000000001,0.38666666666666655,0.8986666666666667,0.25199999999999995,0.956,0.13399999999999998,0.9733333333333334,0.8980000000000001,0.9131522859528636,0.8904040404040404

Information-Retrieval_evaluation_NanoSCIDOCS_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.08,0.12,0.16,0.22,0.08,0.016,0.05333333333333333,0.032,0.04800000000000001,0.04800000000000001,0.036000000000000004,0.07200000000000001,0.11455555555555556,0.06679950774858949,0.05001726999817073
 -1,-1,0.42,0.7,0.76,0.86,0.42,0.08966666666666667,0.36,0.22266666666666665,0.29600000000000004,0.3036666666666667,0.192,0.39366666666666666,0.5704047619047619,0.38552744702820624,0.3049767879412476
 -1,-1,0.48,0.7,0.76,0.86,0.48,0.10066666666666668,0.3666666666666666,0.22666666666666666,0.28400000000000003,0.29366666666666663,0.19,0.38966666666666655,0.613079365079365,0.3897272080691843,0.3046101011521185
+-1,-1,0.44,0.78,0.82,0.84,0.44,0.09266666666666666,0.4066666666666666,0.25266666666666665,0.31199999999999994,0.32166666666666666,0.2,0.4096666666666667,0.6133333333333334,0.40673423073840587,0.33110095208367585

Information-Retrieval_evaluation_NanoSciFact_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.04,0.04,0.06,0.08,0.04,0.04,0.013333333333333332,0.04,0.012000000000000002,0.06,0.008,0.08,0.04722222222222222,0.05463413107474748,0.05219023516020874
 -1,-1,0.38,0.6,0.72,0.72,0.38,0.36,0.22666666666666668,0.59,0.16399999999999998,0.715,0.08399999999999999,0.72,0.5036666666666666,0.5587209456576301,0.5077533104150751
 -1,-1,0.54,0.62,0.66,0.8,0.54,0.505,0.22,0.605,0.14400000000000002,0.65,0.08999999999999998,0.79,0.6001031746031746,0.6382112529739545,0.593840794361791
+-1,-1,0.4,0.58,0.72,0.78,0.4,0.365,0.21333333333333335,0.56,0.16399999999999998,0.71,0.08799999999999997,0.77,0.5201904761904761,0.573675179771537,0.5151451047621822

Information-Retrieval_evaluation_NanoTouche2020_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.46938775510204084,0.6122448979591837,0.7551020408163265,0.8367346938775511,0.46938775510204084,0.029967789586946174,0.3877551020408163,0.07899086583034419,0.34285714285714286,0.1145453187901327,0.2571428571428571,0.1620299103683396,0.5828150307742144,0.3055639570909958,0.21233991925394913
 -1,-1,0.5510204081632653,0.8367346938775511,0.8367346938775511,0.9183673469387755,0.5510204081632653,0.04085691783279971,0.4693877551020407,0.09967309498022202,0.43673469387755104,0.15501374406824417,0.3673469387755102,0.2504351518857603,0.6785714285714286,0.4178263461512225,0.3114722317254114
 -1,-1,0.6122448979591837,0.8571428571428571,0.8979591836734694,0.9795918367346939,0.6122448979591837,0.043705129501604836,0.5238095238095237,0.11031797659129942,0.4938775510204082,0.17345441344231938,0.4244897959183674,0.2815723086367974,0.7416909620991253,0.4795373021012585,0.3603450939734296
+-1,-1,0.5102040816326531,0.8163265306122449,0.9183673469387755,0.9387755102040817,0.5102040816326531,0.03897252093355478,0.5306122448979592,0.11617830784156352,0.4857142857142858,0.16808387137068914,0.40204081632653066,0.2667952262251305,0.6756802721088435,0.4535107828147836,0.3458940792300294

NanoBEIR_evaluation_mean_results.csv

@@ -5,3 +5,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.2053375196232339,0.2717111459968603,0.30885400313971745,0.35513343799058084,0.2053375196232339,0.09730378283251881,0.12982731554160123,0.14230049537621328,0.09960439560439562,0.1615196494127877,0.06931868131868132,0.19362608125512404,0.24989297550522033,0.1934801011975461,0.15306825731640453
 -1,-1,0.45315541601255893,0.6305180533751962,0.6828257456828886,0.7629513343799059,0.45315541601255893,0.25930690958931113,0.28585034013605437,0.3989380323932246,0.22159497645211934,0.4555734363091318,0.15302668759811616,0.5243674247982276,0.5525415140415141,0.48537103006628673,0.40901582458150476
 -1,-1,0.4855572998430141,0.6628571428571429,0.7229199372056515,0.8153532182103612,0.4855572998430141,0.2750729405831144,0.2961904761904762,0.4198896983819936,0.2296828885400314,0.48524566415995796,0.16311459968602826,0.5724029764363421,0.5917552266327776,0.5214666137947834,0.4385301735817977
+-1,-1,0.4730926216640503,0.670486656200942,0.7491051805337521,0.8152904238618525,0.4730926216640503,0.2690472516492284,0.3018419675562533,0.4240880282187189,0.23459340659340663,0.4888912761073963,0.16354160125588696,0.5635832428071486,0.5874912785627071,0.5159778895838031,0.43492691137404127

config.json

@@ -15,7 +15,7 @@
   "max_position_embeddings": 512,
   "model_type": "bert",
   "num_attention_heads": 12,
-  "num_hidden_layers": 6,
+  "num_hidden_layers": 12,
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
   "transformers_version": "4.57.3",

eval/Information-Retrieval_evaluation_NanoMSMARCO_results.csv

@@ -98,3 +98,16 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 2.873563218390805,2500,0.32,0.52,0.58,0.7,0.32,0.32,0.1733333333333333,0.52,0.11599999999999999,0.58,0.07,0.7,0.43693650793650785,0.49929515877847647,0.4520932599081686
 3.160919540229885,2750,0.32,0.52,0.58,0.7,0.32,0.32,0.1733333333333333,0.52,0.11599999999999999,0.58,0.07,0.7,0.43693650793650785,0.49929515877847647,0.4520479631569896
 3.4482758620689653,3000,0.32,0.52,0.58,0.7,0.32,0.32,0.1733333333333333,0.52,0.11599999999999999,0.58,0.07,0.7,0.43693650793650785,0.49929515877847647,0.45205401276376506
+0,0,0.38,0.62,0.68,0.8,0.38,0.38,0.20666666666666667,0.62,0.136,0.68,0.08,0.8,0.5215238095238095,0.5887227948760404,0.5325402082997991
+0.28735632183908044,250,0.38,0.62,0.68,0.8,0.38,0.38,0.20666666666666664,0.62,0.136,0.68,0.08,0.8,0.5133333333333333,0.5819752975378362,0.5233459255821357
+0.5747126436781609,500,0.36,0.62,0.7,0.76,0.36,0.36,0.20666666666666667,0.62,0.14,0.7,0.07600000000000001,0.76,0.5043333333333333,0.5671745892826354,0.5159590057161761
+0.8620689655172413,750,0.34,0.6,0.7,0.78,0.34,0.34,0.2,0.6,0.14,0.7,0.078,0.78,0.488,0.5589508218990319,0.49760221381931913
+1.1494252873563218,1000,0.34,0.6,0.72,0.76,0.34,0.34,0.2,0.6,0.14400000000000002,0.72,0.07600000000000001,0.76,0.48666666666666664,0.5537824377752045,0.49784931032701396
+1.4367816091954024,1250,0.34,0.58,0.72,0.76,0.34,0.34,0.19333333333333333,0.58,0.14400000000000002,0.72,0.07600000000000001,0.76,0.4773333333333333,0.546282303777037,0.48823611476243056
+1.7241379310344827,1500,0.34,0.58,0.7,0.78,0.34,0.34,0.19333333333333333,0.58,0.14,0.7,0.078,0.78,0.4776666666666667,0.5505742128840874,0.48654720618904324
+2.0114942528735633,1750,0.36,0.58,0.7,0.78,0.36,0.36,0.19333333333333333,0.58,0.14,0.7,0.078,0.78,0.4871904761904762,0.5574981407371645,0.4961341580369199
+2.2988505747126435,2000,0.34,0.58,0.7,0.78,0.34,0.34,0.19333333333333333,0.58,0.14,0.7,0.078,0.78,0.47705555555555557,0.5499986700645634,0.48595596512000905
+2.586206896551724,2250,0.36,0.58,0.7,0.78,0.36,0.36,0.19333333333333333,0.58,0.14,0.7,0.078,0.78,0.4867777777777777,0.5570913773706992,0.49571918957952726
+2.873563218390805,2500,0.34,0.58,0.7,0.78,0.34,0.34,0.19333333333333333,0.58,0.14,0.7,0.078,0.78,0.4767777777777777,0.5497099724421284,0.4857009489243644
+3.160919540229885,2750,0.34,0.58,0.7,0.76,0.34,0.34,0.19333333333333333,0.58,0.14,0.7,0.07600000000000001,0.76,0.47455555555555556,0.5436893725288487,0.4851867634881089
+3.4482758620689653,3000,0.34,0.58,0.7,0.76,0.34,0.34,0.19333333333333333,0.58,0.14,0.7,0.07600000000000001,0.76,0.47455555555555556,0.5436893725288487,0.4851846595537516

eval/Information-Retrieval_evaluation_NanoNQ_results.csv

@@ -98,3 +98,16 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 2.873563218390805,2500,0.4,0.6,0.6,0.64,0.4,0.38,0.21333333333333332,0.58,0.132,0.59,0.07,0.63,0.49449999999999994,0.5264146468183786,0.5048856879300306
 3.160919540229885,2750,0.4,0.6,0.6,0.64,0.4,0.38,0.21333333333333332,0.58,0.128,0.58,0.07,0.63,0.49449999999999994,0.526038841299356,0.5043233378343615
 3.4482758620689653,3000,0.4,0.6,0.6,0.64,0.4,0.38,0.21333333333333332,0.58,0.128,0.58,0.07,0.63,0.49449999999999994,0.526038841299356,0.504343499124684
+0,0,0.44,0.62,0.7,0.72,0.44,0.42,0.20666666666666664,0.59,0.15200000000000002,0.69,0.07800000000000001,0.71,0.5425,0.5785694545006416,0.5396412232598975
+0.28735632183908044,250,0.44,0.62,0.66,0.7,0.44,0.42,0.21333333333333332,0.59,0.14400000000000002,0.65,0.07600000000000001,0.69,0.5271666666666667,0.5623275540533932,0.5259612145652153
+0.5747126436781609,500,0.42,0.6,0.64,0.7,0.42,0.4,0.21333333333333332,0.58,0.14,0.63,0.07600000000000001,0.69,0.5147222222222222,0.5526170037116042,0.5129779357934474
+0.8620689655172413,750,0.46,0.58,0.64,0.68,0.46,0.44,0.21333333333333332,0.57,0.14,0.63,0.07400000000000001,0.67,0.5310555555555555,0.5642733261141312,0.5387979108610688
+1.1494252873563218,1000,0.46,0.58,0.62,0.68,0.46,0.44,0.20666666666666664,0.56,0.136,0.61,0.07400000000000001,0.67,0.5270555555555556,0.5601917091639689,0.5329790133521406
+1.4367816091954024,1250,0.46,0.58,0.64,0.68,0.46,0.44,0.20666666666666664,0.56,0.14,0.63,0.07400000000000001,0.67,0.5285555555555556,0.5616194677729451,0.534637573801269
+1.7241379310344827,1500,0.44,0.58,0.62,0.68,0.44,0.42,0.20666666666666664,0.56,0.136,0.61,0.07400000000000001,0.67,0.5145555555555554,0.5510065553704148,0.5205128593886854
+2.0114942528735633,1750,0.44,0.58,0.62,0.68,0.44,0.42,0.20666666666666664,0.56,0.136,0.61,0.07400000000000001,0.67,0.5145555555555554,0.5510065553704148,0.520477131842469
+2.2988505747126435,2000,0.44,0.58,0.62,0.68,0.44,0.42,0.20666666666666664,0.56,0.136,0.61,0.07400000000000001,0.67,0.5145555555555554,0.5510065553704148,0.5204843589512204
+2.586206896551724,2250,0.44,0.58,0.62,0.68,0.44,0.42,0.20666666666666664,0.56,0.136,0.61,0.07400000000000001,0.67,0.5145555555555554,0.5510065553704148,0.5204619561988474
+2.873563218390805,2500,0.44,0.58,0.62,0.68,0.44,0.42,0.20666666666666664,0.56,0.136,0.61,0.07400000000000001,0.67,0.5145555555555554,0.5510065553704148,0.5204777687689691
+3.160919540229885,2750,0.44,0.58,0.62,0.68,0.44,0.42,0.20666666666666664,0.56,0.136,0.61,0.07400000000000001,0.67,0.5145555555555554,0.5510065553704148,0.5204068258456637
+3.4482758620689653,3000,0.44,0.58,0.62,0.68,0.44,0.42,0.20666666666666664,0.56,0.136,0.61,0.07400000000000001,0.67,0.5145555555555554,0.5510065553704148,0.5204271480365081

eval/NanoBEIR_evaluation_mean_results.csv

@@ -98,3 +98,16 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 2.873563218390805,2500,0.36,0.56,0.59,0.6699999999999999,0.36,0.35,0.1933333333333333,0.55,0.124,0.585,0.07,0.665,0.4657182539682539,0.5128549027984275,0.47848947391909963
 3.160919540229885,2750,0.36,0.56,0.59,0.6699999999999999,0.36,0.35,0.1933333333333333,0.55,0.122,0.58,0.07,0.665,0.4657182539682539,0.5126670000389162,0.4781856504956755
 3.4482758620689653,3000,0.36,0.56,0.59,0.6699999999999999,0.36,0.35,0.1933333333333333,0.55,0.122,0.58,0.07,0.665,0.4657182539682539,0.5126670000389162,0.47819875594422456
+0,0,0.41000000000000003,0.62,0.69,0.76,0.41000000000000003,0.4,0.20666666666666667,0.605,0.14400000000000002,0.685,0.07900000000000001,0.755,0.5320119047619047,0.583646124688341,0.5360907157798482
+0.28735632183908044,250,0.41000000000000003,0.62,0.67,0.75,0.41000000000000003,0.4,0.20999999999999996,0.605,0.14,0.665,0.07800000000000001,0.745,0.52025,0.5721514257956146,0.5246535700736754
+0.5747126436781609,500,0.39,0.61,0.6699999999999999,0.73,0.39,0.38,0.21,0.6,0.14,0.665,0.07600000000000001,0.725,0.5095277777777778,0.5598957964971198,0.5144684707548117
+0.8620689655172413,750,0.4,0.59,0.6699999999999999,0.73,0.4,0.39,0.20666666666666667,0.585,0.14,0.665,0.07600000000000001,0.7250000000000001,0.5095277777777778,0.5616120740065815,0.5182000623401939
+1.1494252873563218,1000,0.4,0.59,0.6699999999999999,0.72,0.4,0.39,0.2033333333333333,0.5800000000000001,0.14,0.665,0.07500000000000001,0.7150000000000001,0.5068611111111111,0.5569870734695868,0.5154141618395773
+1.4367816091954024,1250,0.4,0.58,0.6799999999999999,0.72,0.4,0.39,0.19999999999999998,0.5700000000000001,0.14200000000000002,0.675,0.07500000000000001,0.7150000000000001,0.5029444444444444,0.5539508857749911,0.5114368442818498
+1.7241379310344827,1500,0.39,0.58,0.6599999999999999,0.73,0.39,0.38,0.19999999999999998,0.5700000000000001,0.138,0.655,0.07600000000000001,0.7250000000000001,0.49611111111111106,0.550790384127251,0.5035300327888643
+2.0114942528735633,1750,0.4,0.58,0.6599999999999999,0.73,0.4,0.39,0.19999999999999998,0.5700000000000001,0.138,0.655,0.07600000000000001,0.7250000000000001,0.5008730158730158,0.5542523480537896,0.5083056449396944
+2.2988505747126435,2000,0.39,0.58,0.6599999999999999,0.73,0.39,0.38,0.19999999999999998,0.5700000000000001,0.138,0.655,0.07600000000000001,0.7250000000000001,0.4958055555555555,0.550502612717489,0.5032201620356147
+2.586206896551724,2250,0.4,0.58,0.6599999999999999,0.73,0.4,0.39,0.19999999999999998,0.5700000000000001,0.138,0.655,0.07600000000000001,0.7250000000000001,0.5006666666666666,0.554048966370557,0.5080905728891874
+2.873563218390805,2500,0.39,0.58,0.6599999999999999,0.73,0.39,0.38,0.19999999999999998,0.5700000000000001,0.138,0.655,0.07600000000000001,0.7250000000000001,0.4956666666666666,0.5503582639062716,0.5030893588466667
+3.160919540229885,2750,0.39,0.58,0.6599999999999999,0.72,0.39,0.38,0.19999999999999998,0.5700000000000001,0.138,0.655,0.07500000000000001,0.7150000000000001,0.4945555555555555,0.5473479639496317,0.5027967946668863
+3.4482758620689653,3000,0.39,0.58,0.6599999999999999,0.72,0.39,0.38,0.19999999999999998,0.5700000000000001,0.138,0.655,0.07500000000000001,0.7150000000000001,0.4945555555555555,0.5473479639496317,0.5028059037951299

final_metrics.json

@@ -1,231 +1,231 @@
 {
   "nano_beir": {
     "NanoClimateFEVER_cosine_accuracy@1": 0.22,
-    "NanoClimateFEVER_cosine_accuracy@3": 0.38,
-    "NanoClimateFEVER_cosine_accuracy@5": 0.46,
+    "NanoClimateFEVER_cosine_accuracy@3": 0.42,
+    "NanoClimateFEVER_cosine_accuracy@5": 0.5,
     "NanoClimateFEVER_cosine_accuracy@10": 0.68,
     "NanoClimateFEVER_cosine_precision@1": 0.22,
-    "NanoClimateFEVER_cosine_precision@3": 0.14,
-    "NanoClimateFEVER_cosine_precision@5": 0.10400000000000001,
-    "NanoClimateFEVER_cosine_precision@10": 0.076,
-    "NanoClimateFEVER_cosine_recall@1": 0.10666666666666666,
-    "NanoClimateFEVER_cosine_recall@3": 0.2,
-    "NanoClimateFEVER_cosine_recall@5": 0.244,
-    "NanoClimateFEVER_cosine_recall@10": 0.31799999999999995,
-    "NanoClimateFEVER_cosine_ndcg@10": 0.25307254805545965,
-    "NanoClimateFEVER_cosine_mrr@10": 0.3376349206349205,
-    "NanoClimateFEVER_cosine_map@100": 0.19775113367023567,
-    "NanoDBPedia_cosine_accuracy@1": 0.7,
+    "NanoClimateFEVER_cosine_precision@3": 0.13999999999999999,
+    "NanoClimateFEVER_cosine_precision@5": 0.10800000000000001,
+    "NanoClimateFEVER_cosine_precision@10": 0.08,
+    "NanoClimateFEVER_cosine_recall@1": 0.08833333333333332,
+    "NanoClimateFEVER_cosine_recall@3": 0.21066666666666667,
+    "NanoClimateFEVER_cosine_recall@5": 0.24566666666666667,
+    "NanoClimateFEVER_cosine_recall@10": 0.3423333333333333,
+    "NanoClimateFEVER_cosine_ndcg@10": 0.2564996471347133,
+    "NanoClimateFEVER_cosine_mrr@10": 0.34532539682539676,
+    "NanoClimateFEVER_cosine_map@100": 0.18649579530729996,
+    "NanoDBPedia_cosine_accuracy@1": 0.72,
     "NanoDBPedia_cosine_accuracy@3": 0.84,
-    "NanoDBPedia_cosine_accuracy@5": 0.86,
-    "NanoDBPedia_cosine_accuracy@10": 0.9,
-    "NanoDBPedia_cosine_precision@1": 0.7,
-    "NanoDBPedia_cosine_precision@3": 0.48,
-    "NanoDBPedia_cosine_precision@5": 0.43199999999999994,
-    "NanoDBPedia_cosine_precision@10": 0.374,
-    "NanoDBPedia_cosine_recall@1": 0.07514830114967352,
-    "NanoDBPedia_cosine_recall@3": 0.12504263439749053,
-    "NanoDBPedia_cosine_recall@5": 0.17114850325738726,
-    "NanoDBPedia_cosine_recall@10": 0.2647379890126719,
-    "NanoDBPedia_cosine_ndcg@10": 0.4819449351312035,
-    "NanoDBPedia_cosine_mrr@10": 0.7688571428571429,
-    "NanoDBPedia_cosine_map@100": 0.3187663378243811,
-    "NanoFEVER_cosine_accuracy@1": 0.62,
+    "NanoDBPedia_cosine_accuracy@5": 0.9,
+    "NanoDBPedia_cosine_accuracy@10": 0.96,
+    "NanoDBPedia_cosine_precision@1": 0.72,
+    "NanoDBPedia_cosine_precision@3": 0.5266666666666666,
+    "NanoDBPedia_cosine_precision@5": 0.47200000000000003,
+    "NanoDBPedia_cosine_precision@10": 0.382,
+    "NanoDBPedia_cosine_recall@1": 0.05670459269763173,
+    "NanoDBPedia_cosine_recall@3": 0.13069601271168413,
+    "NanoDBPedia_cosine_recall@5": 0.19499547635399148,
+    "NanoDBPedia_cosine_recall@10": 0.2892940743088388,
+    "NanoDBPedia_cosine_ndcg@10": 0.498556918096642,
+    "NanoDBPedia_cosine_mrr@10": 0.7988571428571428,
+    "NanoDBPedia_cosine_map@100": 0.32899455637485014,
+    "NanoFEVER_cosine_accuracy@1": 0.68,
     "NanoFEVER_cosine_accuracy@3": 0.92,
-    "NanoFEVER_cosine_accuracy@5": 0.98,
-    "NanoFEVER_cosine_accuracy@10": 0.98,
-    "NanoFEVER_cosine_precision@1": 0.62,
+    "NanoFEVER_cosine_accuracy@5": 0.92,
+    "NanoFEVER_cosine_accuracy@10": 0.94,
+    "NanoFEVER_cosine_precision@1": 0.68,
     "NanoFEVER_cosine_precision@3": 0.3133333333333333,
-    "NanoFEVER_cosine_precision@5": 0.19999999999999996,
-    "NanoFEVER_cosine_precision@10": 0.09999999999999998,
-    "NanoFEVER_cosine_recall@1": 0.5966666666666667,
-    "NanoFEVER_cosine_recall@3": 0.8666666666666667,
-    "NanoFEVER_cosine_recall@5": 0.9266666666666667,
-    "NanoFEVER_cosine_recall@10": 0.9266666666666667,
-    "NanoFEVER_cosine_ndcg@10": 0.7904169146327766,
-    "NanoFEVER_cosine_mrr@10": 0.7673333333333333,
-    "NanoFEVER_cosine_map@100": 0.7345681818181817,
-    "NanoFiQA2018_cosine_accuracy@1": 0.4,
-    "NanoFiQA2018_cosine_accuracy@3": 0.48,
-    "NanoFiQA2018_cosine_accuracy@5": 0.56,
-    "NanoFiQA2018_cosine_accuracy@10": 0.64,
-    "NanoFiQA2018_cosine_precision@1": 0.4,
-    "NanoFiQA2018_cosine_precision@3": 0.20666666666666667,
-    "NanoFiQA2018_cosine_precision@5": 0.14800000000000002,
-    "NanoFiQA2018_cosine_precision@10": 0.09599999999999997,
-    "NanoFiQA2018_cosine_recall@1": 0.21357936507936512,
-    "NanoFiQA2018_cosine_recall@3": 0.2997698412698413,
-    "NanoFiQA2018_cosine_recall@5": 0.34396031746031747,
-    "NanoFiQA2018_cosine_recall@10": 0.4287936507936508,
-    "NanoFiQA2018_cosine_ndcg@10": 0.37187924093070407,
-    "NanoFiQA2018_cosine_mrr@10": 0.46438095238095245,
-    "NanoFiQA2018_cosine_map@100": 0.3143221865619557,
-    "NanoHotpotQA_cosine_accuracy@1": 0.7,
-    "NanoHotpotQA_cosine_accuracy@3": 0.86,
-    "NanoHotpotQA_cosine_accuracy@5": 0.88,
-    "NanoHotpotQA_cosine_accuracy@10": 0.92,
-    "NanoHotpotQA_cosine_precision@1": 0.7,
-    "NanoHotpotQA_cosine_precision@3": 0.37999999999999995,
-    "NanoHotpotQA_cosine_precision@5": 0.256,
-    "NanoHotpotQA_cosine_precision@10": 0.142,
-    "NanoHotpotQA_cosine_recall@1": 0.35,
-    "NanoHotpotQA_cosine_recall@3": 0.57,
-    "NanoHotpotQA_cosine_recall@5": 0.64,
-    "NanoHotpotQA_cosine_recall@10": 0.71,
-    "NanoHotpotQA_cosine_ndcg@10": 0.6576214974716089,
-    "NanoHotpotQA_cosine_mrr@10": 0.7835238095238095,
-    "NanoHotpotQA_cosine_map@100": 0.5819268475152912,
-    "NanoMSMARCO_cosine_accuracy@1": 0.3,
-    "NanoMSMARCO_cosine_accuracy@3": 0.54,
-    "NanoMSMARCO_cosine_accuracy@5": 0.62,
-    "NanoMSMARCO_cosine_accuracy@10": 0.76,
-    "NanoMSMARCO_cosine_precision@1": 0.3,
-    "NanoMSMARCO_cosine_precision@3": 0.18,
-    "NanoMSMARCO_cosine_precision@5": 0.124,
-    "NanoMSMARCO_cosine_precision@10": 0.07600000000000001,
-    "NanoMSMARCO_cosine_recall@1": 0.3,
-    "NanoMSMARCO_cosine_recall@3": 0.54,
-    "NanoMSMARCO_cosine_recall@5": 0.62,
-    "NanoMSMARCO_cosine_recall@10": 0.76,
-    "NanoMSMARCO_cosine_ndcg@10": 0.5241190384704345,
-    "NanoMSMARCO_cosine_mrr@10": 0.4492698412698413,
-    "NanoMSMARCO_cosine_map@100": 0.45777964902887497,
-    "NanoNFCorpus_cosine_accuracy@1": 0.36,
-    "NanoNFCorpus_cosine_accuracy@3": 0.48,
-    "NanoNFCorpus_cosine_accuracy@5": 0.5,
-    "NanoNFCorpus_cosine_accuracy@10": 0.58,
-    "NanoNFCorpus_cosine_precision@1": 0.36,
-    "NanoNFCorpus_cosine_precision@3": 0.32666666666666666,
-    "NanoNFCorpus_cosine_precision@5": 0.29200000000000004,
-    "NanoNFCorpus_cosine_precision@10": 0.264,
-    "NanoNFCorpus_cosine_recall@1": 0.030515431849843816,
-    "NanoNFCorpus_cosine_recall@3": 0.053102293373951406,
-    "NanoNFCorpus_cosine_recall@5": 0.06596373325276159,
-    "NanoNFCorpus_cosine_recall@10": 0.10180141189599459,
-    "NanoNFCorpus_cosine_ndcg@10": 0.3017216559964908,
-    "NanoNFCorpus_cosine_mrr@10": 0.43479365079365073,
-    "NanoNFCorpus_cosine_map@100": 0.1219166619490478,
-    "NanoNQ_cosine_accuracy@1": 0.38,
-    "NanoNQ_cosine_accuracy@3": 0.52,
-    "NanoNQ_cosine_accuracy@5": 0.54,
-    "NanoNQ_cosine_accuracy@10": 0.68,
-    "NanoNQ_cosine_precision@1": 0.38,
-    "NanoNQ_cosine_precision@3": 0.1733333333333333,
-    "NanoNQ_cosine_precision@5": 0.11200000000000002,
-    "NanoNQ_cosine_precision@10": 0.07200000000000001,
-    "NanoNQ_cosine_recall@1": 0.35,
-    "NanoNQ_cosine_recall@3": 0.49,
-    "NanoNQ_cosine_recall@5": 0.52,
-    "NanoNQ_cosine_recall@10": 0.66,
-    "NanoNQ_cosine_ndcg@10": 0.5017561161582912,
-    "NanoNQ_cosine_mrr@10": 0.46857142857142864,
-    "NanoNQ_cosine_map@100": 0.4585943213547632,
-    "NanoQuoraRetrieval_cosine_accuracy@1": 0.78,
-    "NanoQuoraRetrieval_cosine_accuracy@3": 0.96,
-    "NanoQuoraRetrieval_cosine_accuracy@5": 0.96,
+    "NanoFEVER_cosine_precision@5": 0.18799999999999997,
+    "NanoFEVER_cosine_precision@10": 0.09599999999999997,
+    "NanoFEVER_cosine_recall@1": 0.6266666666666666,
+    "NanoFEVER_cosine_recall@3": 0.8666666666666666,
+    "NanoFEVER_cosine_recall@5": 0.8666666666666666,
+    "NanoFEVER_cosine_recall@10": 0.8866666666666667,
+    "NanoFEVER_cosine_ndcg@10": 0.7835360079677364,
+    "NanoFEVER_cosine_mrr@10": 0.7833333333333334,
+    "NanoFEVER_cosine_map@100": 0.738962481962482,
+    "NanoFiQA2018_cosine_accuracy@1": 0.38,
+    "NanoFiQA2018_cosine_accuracy@3": 0.54,
+    "NanoFiQA2018_cosine_accuracy@5": 0.68,
+    "NanoFiQA2018_cosine_accuracy@10": 0.74,
+    "NanoFiQA2018_cosine_precision@1": 0.38,
+    "NanoFiQA2018_cosine_precision@3": 0.2333333333333333,
+    "NanoFiQA2018_cosine_precision@5": 0.18,
+    "NanoFiQA2018_cosine_precision@10": 0.12,
+    "NanoFiQA2018_cosine_recall@1": 0.2114126984126984,
+    "NanoFiQA2018_cosine_recall@3": 0.3536269841269841,
+    "NanoFiQA2018_cosine_recall@5": 0.4336269841269841,
+    "NanoFiQA2018_cosine_recall@10": 0.5354047619047619,
+    "NanoFiQA2018_cosine_ndcg@10": 0.42968312791198054,
+    "NanoFiQA2018_cosine_mrr@10": 0.4913888888888889,
+    "NanoFiQA2018_cosine_map@100": 0.35462063324400034,
+    "NanoHotpotQA_cosine_accuracy@1": 0.6,
+    "NanoHotpotQA_cosine_accuracy@3": 0.74,
+    "NanoHotpotQA_cosine_accuracy@5": 0.82,
+    "NanoHotpotQA_cosine_accuracy@10": 0.86,
+    "NanoHotpotQA_cosine_precision@1": 0.6,
+    "NanoHotpotQA_cosine_precision@3": 0.3,
+    "NanoHotpotQA_cosine_precision@5": 0.204,
+    "NanoHotpotQA_cosine_precision@10": 0.11599999999999998,
+    "NanoHotpotQA_cosine_recall@1": 0.3,
+    "NanoHotpotQA_cosine_recall@3": 0.45,
+    "NanoHotpotQA_cosine_recall@5": 0.51,
+    "NanoHotpotQA_cosine_recall@10": 0.58,
+    "NanoHotpotQA_cosine_ndcg@10": 0.5356886314565849,
+    "NanoHotpotQA_cosine_mrr@10": 0.6838571428571428,
+    "NanoHotpotQA_cosine_map@100": 0.45644680388890113,
+    "NanoMSMARCO_cosine_accuracy@1": 0.32,
+    "NanoMSMARCO_cosine_accuracy@3": 0.52,
+    "NanoMSMARCO_cosine_accuracy@5": 0.58,
+    "NanoMSMARCO_cosine_accuracy@10": 0.7,
+    "NanoMSMARCO_cosine_precision@1": 0.32,
+    "NanoMSMARCO_cosine_precision@3": 0.1733333333333333,
+    "NanoMSMARCO_cosine_precision@5": 0.11599999999999999,
+    "NanoMSMARCO_cosine_precision@10": 0.07,
+    "NanoMSMARCO_cosine_recall@1": 0.32,
+    "NanoMSMARCO_cosine_recall@3": 0.52,
+    "NanoMSMARCO_cosine_recall@5": 0.58,
+    "NanoMSMARCO_cosine_recall@10": 0.7,
+    "NanoMSMARCO_cosine_ndcg@10": 0.49929515877847647,
+    "NanoMSMARCO_cosine_mrr@10": 0.43693650793650785,
+    "NanoMSMARCO_cosine_map@100": 0.45205401276376506,
+    "NanoNFCorpus_cosine_accuracy@1": 0.38,
+    "NanoNFCorpus_cosine_accuracy@3": 0.5,
+    "NanoNFCorpus_cosine_accuracy@5": 0.6,
+    "NanoNFCorpus_cosine_accuracy@10": 0.72,
+    "NanoNFCorpus_cosine_precision@1": 0.38,
+    "NanoNFCorpus_cosine_precision@3": 0.3133333333333333,
+    "NanoNFCorpus_cosine_precision@5": 0.3,
+    "NanoNFCorpus_cosine_precision@10": 0.28600000000000003,
+    "NanoNFCorpus_cosine_recall@1": 0.013857792729417237,
+    "NanoNFCorpus_cosine_recall@3": 0.053976395496446575,
+    "NanoNFCorpus_cosine_recall@5": 0.08888025754448707,
+    "NanoNFCorpus_cosine_recall@10": 0.12308809405420085,
+    "NanoNFCorpus_cosine_ndcg@10": 0.3118912881405269,
+    "NanoNFCorpus_cosine_mrr@10": 0.4709365079365079,
+    "NanoNFCorpus_cosine_map@100": 0.11679455794345621,
+    "NanoNQ_cosine_accuracy@1": 0.4,
+    "NanoNQ_cosine_accuracy@3": 0.6,
+    "NanoNQ_cosine_accuracy@5": 0.6,
+    "NanoNQ_cosine_accuracy@10": 0.64,
+    "NanoNQ_cosine_precision@1": 0.4,
+    "NanoNQ_cosine_precision@3": 0.21333333333333332,
+    "NanoNQ_cosine_precision@5": 0.128,
+    "NanoNQ_cosine_precision@10": 0.07,
+    "NanoNQ_cosine_recall@1": 0.38,
+    "NanoNQ_cosine_recall@3": 0.58,
+    "NanoNQ_cosine_recall@5": 0.58,
+    "NanoNQ_cosine_recall@10": 0.63,
+    "NanoNQ_cosine_ndcg@10": 0.526038841299356,
+    "NanoNQ_cosine_mrr@10": 0.49449999999999994,
+    "NanoNQ_cosine_map@100": 0.5043233378343615,
+    "NanoQuoraRetrieval_cosine_accuracy@1": 0.84,
+    "NanoQuoraRetrieval_cosine_accuracy@3": 0.94,
+    "NanoQuoraRetrieval_cosine_accuracy@5": 0.98,
     "NanoQuoraRetrieval_cosine_accuracy@10": 0.98,
-    "NanoQuoraRetrieval_cosine_precision@1": 0.78,
+    "NanoQuoraRetrieval_cosine_precision@1": 0.84,
     "NanoQuoraRetrieval_cosine_precision@3": 0.38666666666666655,
     "NanoQuoraRetrieval_cosine_precision@5": 0.25199999999999995,
-    "NanoQuoraRetrieval_cosine_precision@10": 0.132,
-    "NanoQuoraRetrieval_cosine_recall@1": 0.684,
-    "NanoQuoraRetrieval_cosine_recall@3": 0.912,
-    "NanoQuoraRetrieval_cosine_recall@5": 0.9393333333333334,
-    "NanoQuoraRetrieval_cosine_recall@10": 0.97,
-    "NanoQuoraRetrieval_cosine_ndcg@10": 0.8825493396502728,
-    "NanoQuoraRetrieval_cosine_mrr@10": 0.8622222222222223,
-    "NanoQuoraRetrieval_cosine_map@100": 0.8476349206349206,
-    "NanoSCIDOCS_cosine_accuracy@1": 0.48,
-    "NanoSCIDOCS_cosine_accuracy@3": 0.7,
-    "NanoSCIDOCS_cosine_accuracy@5": 0.76,
-    "NanoSCIDOCS_cosine_accuracy@10": 0.86,
-    "NanoSCIDOCS_cosine_precision@1": 0.48,
-    "NanoSCIDOCS_cosine_precision@3": 0.3666666666666666,
-    "NanoSCIDOCS_cosine_precision@5": 0.28400000000000003,
-    "NanoSCIDOCS_cosine_precision@10": 0.19,
-    "NanoSCIDOCS_cosine_recall@1": 0.10066666666666668,
-    "NanoSCIDOCS_cosine_recall@3": 0.22666666666666666,
-    "NanoSCIDOCS_cosine_recall@5": 0.29366666666666663,
-    "NanoSCIDOCS_cosine_recall@10": 0.38966666666666655,
-    "NanoSCIDOCS_cosine_ndcg@10": 0.3897272080691843,
-    "NanoSCIDOCS_cosine_mrr@10": 0.613079365079365,
-    "NanoSCIDOCS_cosine_map@100": 0.3046101011521185,
-    "NanoArguAna_cosine_accuracy@1": 0.22,
-    "NanoArguAna_cosine_accuracy@3": 0.46,
-    "NanoArguAna_cosine_accuracy@5": 0.72,
-    "NanoArguAna_cosine_accuracy@10": 0.84,
-    "NanoArguAna_cosine_precision@1": 0.22,
-    "NanoArguAna_cosine_precision@3": 0.15333333333333332,
-    "NanoArguAna_cosine_precision@5": 0.14400000000000002,
-    "NanoArguAna_cosine_precision@10": 0.08399999999999999,
-    "NanoArguAna_cosine_recall@1": 0.22,
-    "NanoArguAna_cosine_recall@3": 0.46,
-    "NanoArguAna_cosine_recall@5": 0.72,
-    "NanoArguAna_cosine_recall@10": 0.84,
-    "NanoArguAna_cosine_ndcg@10": 0.5065089296905451,
-    "NanoArguAna_cosine_mrr@10": 0.4013571428571428,
-    "NanoArguAna_cosine_map@100": 0.40883602671837965,
-    "NanoSciFact_cosine_accuracy@1": 0.54,
-    "NanoSciFact_cosine_accuracy@3": 0.62,
-    "NanoSciFact_cosine_accuracy@5": 0.66,
-    "NanoSciFact_cosine_accuracy@10": 0.8,
-    "NanoSciFact_cosine_precision@1": 0.54,
-    "NanoSciFact_cosine_precision@3": 0.22,
-    "NanoSciFact_cosine_precision@5": 0.14400000000000002,
-    "NanoSciFact_cosine_precision@10": 0.08999999999999998,
-    "NanoSciFact_cosine_recall@1": 0.505,
-    "NanoSciFact_cosine_recall@3": 0.605,
-    "NanoSciFact_cosine_recall@5": 0.65,
-    "NanoSciFact_cosine_recall@10": 0.79,
-    "NanoSciFact_cosine_ndcg@10": 0.6382112529739545,
-    "NanoSciFact_cosine_mrr@10": 0.6001031746031746,
-    "NanoSciFact_cosine_map@100": 0.593840794361791,
-    "NanoTouche2020_cosine_accuracy@1": 0.6122448979591837,
-    "NanoTouche2020_cosine_accuracy@3": 0.8571428571428571,
-    "NanoTouche2020_cosine_accuracy@5": 0.8979591836734694,
-    "NanoTouche2020_cosine_accuracy@10": 0.9795918367346939,
-    "NanoTouche2020_cosine_precision@1": 0.6122448979591837,
-    "NanoTouche2020_cosine_precision@3": 0.5238095238095237,
-    "NanoTouche2020_cosine_precision@5": 0.4938775510204082,
-    "NanoTouche2020_cosine_precision@10": 0.4244897959183674,
-    "NanoTouche2020_cosine_recall@1": 0.043705129501604836,
-    "NanoTouche2020_cosine_recall@3": 0.11031797659129942,
-    "NanoTouche2020_cosine_recall@5": 0.17345441344231938,
-    "NanoTouche2020_cosine_recall@10": 0.2815723086367974,
-    "NanoTouche2020_cosine_ndcg@10": 0.4795373021012585,
-    "NanoTouche2020_cosine_mrr@10": 0.7416909620991253,
-    "NanoTouche2020_cosine_map@100": 0.3603450939734296,
-    "NanoBEIR_mean_cosine_accuracy@1": 0.4855572998430141,
-    "NanoBEIR_mean_cosine_accuracy@3": 0.6628571428571429,
-    "NanoBEIR_mean_cosine_accuracy@5": 0.7229199372056515,
-    "NanoBEIR_mean_cosine_accuracy@10": 0.8153532182103612,
-    "NanoBEIR_mean_cosine_precision@1": 0.4855572998430141,
-    "NanoBEIR_mean_cosine_precision@3": 0.2961904761904762,
-    "NanoBEIR_mean_cosine_precision@5": 0.2296828885400314,
-    "NanoBEIR_mean_cosine_precision@10": 0.16311459968602826,
-    "NanoBEIR_mean_cosine_recall@1": 0.2750729405831144,
-    "NanoBEIR_mean_cosine_recall@3": 0.4198896983819936,
-    "NanoBEIR_mean_cosine_recall@5": 0.48524566415995796,
-    "NanoBEIR_mean_cosine_recall@10": 0.5724029764363421,
-    "NanoBEIR_mean_cosine_ndcg@10": 0.5214666137947834,
-    "NanoBEIR_mean_cosine_mrr@10": 0.5917552266327776,
-    "NanoBEIR_mean_cosine_map@100": 0.4385301735817977
+    "NanoQuoraRetrieval_cosine_precision@10": 0.13399999999999998,
+    "NanoQuoraRetrieval_cosine_recall@1": 0.7440000000000001,
+    "NanoQuoraRetrieval_cosine_recall@3": 0.8986666666666667,
+    "NanoQuoraRetrieval_cosine_recall@5": 0.956,
+    "NanoQuoraRetrieval_cosine_recall@10": 0.9733333333333334,
+    "NanoQuoraRetrieval_cosine_ndcg@10": 0.9131522859528636,
+    "NanoQuoraRetrieval_cosine_mrr@10": 0.8980000000000001,
+    "NanoQuoraRetrieval_cosine_map@100": 0.8904040404040404,
+    "NanoSCIDOCS_cosine_accuracy@1": 0.44,
+    "NanoSCIDOCS_cosine_accuracy@3": 0.78,
+    "NanoSCIDOCS_cosine_accuracy@5": 0.82,
+    "NanoSCIDOCS_cosine_accuracy@10": 0.84,
+    "NanoSCIDOCS_cosine_precision@1": 0.44,
+    "NanoSCIDOCS_cosine_precision@3": 0.4066666666666666,
+    "NanoSCIDOCS_cosine_precision@5": 0.31199999999999994,
+    "NanoSCIDOCS_cosine_precision@10": 0.2,
+    "NanoSCIDOCS_cosine_recall@1": 0.09266666666666666,
+    "NanoSCIDOCS_cosine_recall@3": 0.25266666666666665,
+    "NanoSCIDOCS_cosine_recall@5": 0.32166666666666666,
+    "NanoSCIDOCS_cosine_recall@10": 0.4096666666666667,
+    "NanoSCIDOCS_cosine_ndcg@10": 0.40673423073840587,
+    "NanoSCIDOCS_cosine_mrr@10": 0.6133333333333334,
+    "NanoSCIDOCS_cosine_map@100": 0.33110095208367585,
+    "NanoArguAna_cosine_accuracy@1": 0.26,
+    "NanoArguAna_cosine_accuracy@3": 0.52,
+    "NanoArguAna_cosine_accuracy@5": 0.7,
+    "NanoArguAna_cosine_accuracy@10": 0.82,
+    "NanoArguAna_cosine_precision@1": 0.26,
+    "NanoArguAna_cosine_precision@3": 0.1733333333333333,
+    "NanoArguAna_cosine_precision@5": 0.14,
+    "NanoArguAna_cosine_precision@10": 0.08199999999999999,
+    "NanoArguAna_cosine_recall@1": 0.26,
+    "NanoArguAna_cosine_recall@3": 0.52,
+    "NanoArguAna_cosine_recall@5": 0.7,
+    "NanoArguAna_cosine_recall@10": 0.82,
+    "NanoArguAna_cosine_ndcg@10": 0.5194504645258333,
+    "NanoArguAna_cosine_mrr@10": 0.425047619047619,
+    "NanoArguAna_cosine_map@100": 0.43281349206349207,
+    "NanoSciFact_cosine_accuracy@1": 0.4,
+    "NanoSciFact_cosine_accuracy@3": 0.58,
+    "NanoSciFact_cosine_accuracy@5": 0.72,
+    "NanoSciFact_cosine_accuracy@10": 0.78,
+    "NanoSciFact_cosine_precision@1": 0.4,
+    "NanoSciFact_cosine_precision@3": 0.21333333333333335,
+    "NanoSciFact_cosine_precision@5": 0.16399999999999998,
+    "NanoSciFact_cosine_precision@10": 0.08799999999999997,
+    "NanoSciFact_cosine_recall@1": 0.365,
+    "NanoSciFact_cosine_recall@3": 0.56,
+    "NanoSciFact_cosine_recall@5": 0.71,
+    "NanoSciFact_cosine_recall@10": 0.77,
+    "NanoSciFact_cosine_ndcg@10": 0.573675179771537,
+    "NanoSciFact_cosine_mrr@10": 0.5201904761904761,
+    "NanoSciFact_cosine_map@100": 0.5151451047621822,
+    "NanoTouche2020_cosine_accuracy@1": 0.5102040816326531,
+    "NanoTouche2020_cosine_accuracy@3": 0.8163265306122449,
+    "NanoTouche2020_cosine_accuracy@5": 0.9183673469387755,
+    "NanoTouche2020_cosine_accuracy@10": 0.9387755102040817,
+    "NanoTouche2020_cosine_precision@1": 0.5102040816326531,
+    "NanoTouche2020_cosine_precision@3": 0.5306122448979592,
+    "NanoTouche2020_cosine_precision@5": 0.4857142857142858,
+    "NanoTouche2020_cosine_precision@10": 0.40204081632653066,
+    "NanoTouche2020_cosine_recall@1": 0.03897252093355478,
+    "NanoTouche2020_cosine_recall@3": 0.11617830784156352,
+    "NanoTouche2020_cosine_recall@5": 0.16808387137068914,
+    "NanoTouche2020_cosine_recall@10": 0.2667952262251305,
+    "NanoTouche2020_cosine_ndcg@10": 0.4535107828147836,
+    "NanoTouche2020_cosine_mrr@10": 0.6756802721088435,
+    "NanoTouche2020_cosine_map@100": 0.3458940792300294,
+    "NanoBEIR_mean_cosine_accuracy@1": 0.4730926216640503,
+    "NanoBEIR_mean_cosine_accuracy@3": 0.670486656200942,
+    "NanoBEIR_mean_cosine_accuracy@5": 0.7491051805337521,
+    "NanoBEIR_mean_cosine_accuracy@10": 0.8152904238618525,
+    "NanoBEIR_mean_cosine_precision@1": 0.4730926216640503,
+    "NanoBEIR_mean_cosine_precision@3": 0.3018419675562533,
+    "NanoBEIR_mean_cosine_precision@5": 0.23459340659340663,
+    "NanoBEIR_mean_cosine_precision@10": 0.16354160125588696,
+    "NanoBEIR_mean_cosine_recall@1": 0.2690472516492284,
+    "NanoBEIR_mean_cosine_recall@3": 0.4240880282187189,
+    "NanoBEIR_mean_cosine_recall@5": 0.4888912761073963,
+    "NanoBEIR_mean_cosine_recall@10": 0.5635832428071486,
+    "NanoBEIR_mean_cosine_ndcg@10": 0.5159778895838031,
+    "NanoBEIR_mean_cosine_mrr@10": 0.5874912785627071,
+    "NanoBEIR_mean_cosine_map@100": 0.43492691137404127
   },
   "beir_touche2020": {
-    "BeIR-touche2020-subset-test_cosine_accuracy@1": 0.8775510204081632,
-    "BeIR-touche2020-subset-test_cosine_accuracy@3": 1.0,
+    "BeIR-touche2020-subset-test_cosine_accuracy@1": 0.8367346938775511,
+    "BeIR-touche2020-subset-test_cosine_accuracy@3": 0.9591836734693877,
     "BeIR-touche2020-subset-test_cosine_accuracy@5": 1.0,
     "BeIR-touche2020-subset-test_cosine_accuracy@10": 1.0,
-    "BeIR-touche2020-subset-test_cosine_precision@1": 0.8775510204081632,
-    "BeIR-touche2020-subset-test_cosine_precision@3": 0.8095238095238094,
-    "BeIR-touche2020-subset-test_cosine_precision@5": 0.7510204081632652,
-    "BeIR-touche2020-subset-test_cosine_precision@10": 0.6448979591836734,
-    "BeIR-touche2020-subset-test_cosine_recall@1": 0.01945012749314946,
-    "BeIR-touche2020-subset-test_cosine_recall@3": 0.0537432819678873,
-    "BeIR-touche2020-subset-test_cosine_recall@5": 0.08295390754493644,
-    "BeIR-touche2020-subset-test_cosine_recall@10": 0.14269962051162913,
-    "BeIR-touche2020-subset-test_cosine_ndcg@10": 0.698044370091445,
-    "BeIR-touche2020-subset-test_cosine_mrr@10": 0.9387755102040817,
-    "BeIR-touche2020-subset-test_cosine_map@100": 0.29990375282881515
+    "BeIR-touche2020-subset-test_cosine_precision@1": 0.8367346938775511,
+    "BeIR-touche2020-subset-test_cosine_precision@3": 0.7823129251700679,
+    "BeIR-touche2020-subset-test_cosine_precision@5": 0.7306122448979592,
+    "BeIR-touche2020-subset-test_cosine_precision@10": 0.6632653061224489,
+    "BeIR-touche2020-subset-test_cosine_recall@1": 0.01863536778338729,
+    "BeIR-touche2020-subset-test_cosine_recall@3": 0.05202492032020825,
+    "BeIR-touche2020-subset-test_cosine_recall@5": 0.08076708716823142,
+    "BeIR-touche2020-subset-test_cosine_recall@10": 0.14656488113887348,
+    "BeIR-touche2020-subset-test_cosine_ndcg@10": 0.7013835872612906,
+    "BeIR-touche2020-subset-test_cosine_mrr@10": 0.9047619047619047,
+    "BeIR-touche2020-subset-test_cosine_map@100": 0.3127573985913156
   }
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e4cba7736f0b233c21227eb51b3b05d427e51440fd611d6327986fbae3b85479
-size 90864192
+oid sha256:c0b770bb50da1c493b94cd6f89b4ae3aa73ec8f24effd494bf23593a852c9a6d
+size 133462128

tokenizer_config.json

@@ -48,7 +48,7 @@
   "extra_special_tokens": {},
   "mask_token": "[MASK]",
   "max_length": 128,
-  "model_max_length": 256,
+  "model_max_length": 128,
   "never_split": null,
   "pad_to_multiple_of": null,
   "pad_token": "[PAD]",