Training in progress, step 5062

Information-Retrieval_evaluation_BeIR-touche2020-subset-test_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.7959183673469388,0.9183673469387755,0.9795918367346939,1.0,0.7959183673469388,0.017589318234381018,0.7210884353741496,0.047638441502256816,0.6979591836734694,0.07685025929622703,0.5795918367346939,0.12768596260273413,0.8753644314868804,0.6268789790121103,0.26682197341262104
 -1,-1,0.7755102040816326,0.9795918367346939,1.0,1.0,0.7755102040816326,0.017089566838806178,0.7414965986394557,0.04913135456425962,0.6734693877551022,0.0744729216014395,0.6061224489795918,0.13390399153802923,0.8792517006802721,0.6452010725749343,0.26987146167670506
 -1,-1,0.2653061224489796,0.5714285714285714,0.6938775510204082,0.7551020408163265,0.2653061224489796,0.005859789020652119,0.3469387755102041,0.022771209086251597,0.4,0.04397085795258936,0.3755102040816327,0.08239494960074849,0.42636054421768715,0.36314214650584564,0.16579582237011886
+-1,-1,0.7755102040816326,0.9795918367346939,1.0,1.0,0.7755102040816326,0.01706832759711131,0.7482993197278911,0.049401730544661945,0.7142857142857143,0.07867619617707583,0.6122448979591837,0.1349240729674531,0.8714285714285713,0.653970278908209,0.2786997022824942

Information-Retrieval_evaluation_NanoArguAna_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.18,0.4,0.56,0.76,0.18,0.18,0.13333333333333333,0.4,0.11200000000000003,0.56,0.07600000000000001,0.76,0.34159523809523795,0.4400427760450664,0.346353283565148
 -1,-1,0.18,0.36,0.5,0.68,0.18,0.18,0.11999999999999998,0.36,0.10000000000000002,0.5,0.068,0.68,0.3243809523809524,0.4088546835300411,0.3354654705346437
 -1,-1,0.16,0.5,0.68,0.88,0.16,0.16,0.16666666666666663,0.5,0.136,0.68,0.088,0.88,0.3908412698412698,0.5077064028450745,0.3975778004159583
+-1,-1,0.16,0.42,0.56,0.8,0.16,0.16,0.13999999999999999,0.42,0.11200000000000003,0.56,0.08,0.8,0.346095238095238,0.4541392479306418,0.3511114131364851

Information-Retrieval_evaluation_NanoClimateFEVER_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.26,0.38,0.48,0.54,0.26,0.14066666666666666,0.13333333333333333,0.20233333333333334,0.1,0.23399999999999999,0.062,0.27399999999999997,0.34852380952380957,0.2492839568639178,0.20441864232503498
 -1,-1,0.2,0.44,0.5,0.6,0.2,0.09666666666666666,0.14666666666666664,0.21566666666666662,0.10400000000000001,0.244,0.066,0.304,0.32441269841269843,0.23875559115390693,0.18002326993925138
 -1,-1,0.12,0.24,0.26,0.3,0.12,0.05833333333333333,0.08,0.11166666666666664,0.06000000000000001,0.13333333333333333,0.042,0.1733333333333333,0.18599999999999997,0.1421323657809101,0.11251502558321347
+-1,-1,0.22,0.52,0.56,0.64,0.22,0.09166666666666666,0.1733333333333333,0.23399999999999999,0.12000000000000002,0.27899999999999997,0.076,0.32733333333333337,0.3648571428571428,0.26355221051084216,0.19506453252604283

Information-Retrieval_evaluation_NanoDBPedia_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.6,0.76,0.86,0.9,0.6,0.05244099344102771,0.4733333333333334,0.11905666980927139,0.42800000000000005,0.1498535000442701,0.34600000000000003,0.24163199162754928,0.7028888888888889,0.4426565477477201,0.2828214084522445
 -1,-1,0.72,0.78,0.82,0.92,0.72,0.07890647927098722,0.49333333333333335,0.1444097476150978,0.4320000000000001,0.17161496480028293,0.36199999999999993,0.24571882117909583,0.7622698412698413,0.47189065029725746,0.3083301597708803
 -1,-1,0.62,0.74,0.82,0.88,0.62,0.05294717669991995,0.4600000000000001,0.11950308612173167,0.40399999999999997,0.1809061480074522,0.352,0.26388938838721027,0.7002698412698414,0.45186572825718035,0.2963971880776791
+-1,-1,0.66,0.8,0.84,0.94,0.66,0.07082671665308901,0.5066666666666666,0.13730927885289243,0.4160000000000001,0.1661893911919123,0.368,0.25034697924131,0.7467142857142857,0.4728916713272679,0.30396194965649254

Information-Retrieval_evaluation_NanoFEVER_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.46,0.78,0.86,0.88,0.46,0.42666666666666664,0.26666666666666666,0.7233333333333333,0.176,0.8033333333333332,0.092,0.8333333333333333,0.6218571428571428,0.6580129211253776,0.5958589705489019
 -1,-1,0.48,0.7,0.84,0.92,0.48,0.42666666666666664,0.23333333333333336,0.6466666666666666,0.17199999999999996,0.7966666666666667,0.09399999999999999,0.8666666666666667,0.6137777777777778,0.647832874607183,0.568931348337861
 -1,-1,0.28,0.42,0.44,0.44,0.28,0.27,0.13999999999999999,0.4,0.09200000000000001,0.43,0.046000000000000006,0.43,0.3506666666666666,0.3661113939532885,0.3422393215160618
+-1,-1,0.6,0.9,0.96,0.98,0.6,0.5566666666666666,0.30666666666666664,0.8466666666666667,0.19599999999999995,0.9066666666666667,0.09999999999999998,0.9266666666666667,0.7516666666666666,0.7716934881014161,0.7061067287383075

Information-Retrieval_evaluation_NanoFiQA2018_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.38,0.5,0.58,0.64,0.38,0.20874603174603173,0.18666666666666668,0.3174126984126984,0.14400000000000002,0.3784365079365079,0.086,0.4249920634920635,0.45752380952380955,0.3627243700335102,0.30653324427432976
 -1,-1,0.34,0.42,0.52,0.6,0.34,0.17874603174603174,0.1733333333333333,0.26026984126984126,0.14400000000000002,0.3431031746031746,0.092,0.44262698412698415,0.4107380952380952,0.3483765709350854,0.2899028574096885
 -1,-1,0.16,0.32,0.36,0.42,0.16,0.09666666666666666,0.11333333333333333,0.2333333333333333,0.07600000000000001,0.24833333333333332,0.05000000000000001,0.2977777777777778,0.23633333333333328,0.21754856447960677,0.18156772343708624
+-1,-1,0.36,0.48,0.6,0.66,0.36,0.19941269841269843,0.20666666666666667,0.2806031746031746,0.16,0.3781269841269841,0.08999999999999998,0.4256269841269841,0.4481666666666666,0.35641512120073343,0.29847071848639917

Information-Retrieval_evaluation_NanoHotpotQA_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.52,0.62,0.66,0.74,0.52,0.26,0.24666666666666665,0.37,0.16399999999999998,0.41,0.092,0.46,0.5840238095238095,0.4369863506070083,0.37142211292843513
 -1,-1,0.66,0.8,0.82,0.9,0.66,0.33,0.3066666666666667,0.46,0.18799999999999997,0.47,0.11399999999999999,0.57,0.7386111111111111,0.5429476399940535,0.44804908578281555
 -1,-1,0.46,0.6,0.66,0.78,0.46,0.23,0.26,0.39,0.176,0.44,0.10399999999999998,0.52,0.5547222222222222,0.45661766676516835,0.389670298499331
+-1,-1,0.74,0.82,0.84,0.86,0.74,0.37,0.33333333333333326,0.5,0.21599999999999997,0.54,0.12399999999999999,0.62,0.7883333333333334,0.6038529756799117,0.5248449945769594

Information-Retrieval_evaluation_NanoMSMARCO_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.32,0.5,0.56,0.72,0.32,0.32,0.16666666666666669,0.5,0.11200000000000002,0.56,0.07200000000000001,0.72,0.4414682539682539,0.5076844979819899,0.4541878759718346
 -1,-1,0.32,0.54,0.64,0.72,0.32,0.32,0.18,0.54,0.12800000000000003,0.64,0.07200000000000001,0.72,0.4551666666666666,0.5194788229083886,0.46866687154926134
 -1,-1,0.46,0.64,0.78,0.84,0.46,0.46,0.21333333333333332,0.64,0.156,0.78,0.08399999999999999,0.84,0.5809365079365079,0.6436072717719088,0.5875442422678194
+-1,-1,0.34,0.6,0.66,0.7,0.34,0.34,0.2,0.6,0.132,0.66,0.07,0.7,0.4747460317460317,0.5303422656736582,0.48845125969077435

Information-Retrieval_evaluation_NanoNFCorpus_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.34,0.42,0.48,0.6,0.34,0.011303793989200254,0.32666666666666666,0.031244681505305585,0.292,0.04713607047852303,0.234,0.0693992741134154,0.404468253968254,0.2598035673694497,0.08959056587053282
 -1,-1,0.32,0.38,0.5,0.56,0.32,0.010995012876071315,0.28,0.02828544592719776,0.27199999999999996,0.06650305246570268,0.23000000000000004,0.08873790175812095,0.3792698412698413,0.25433221795160377,0.09691975523503352
 -1,-1,0.38,0.48,0.56,0.7,0.38,0.01687245746538031,0.33333333333333326,0.04124468301356519,0.32,0.06123009468323657,0.278,0.10651046552028522,0.4581269841269841,0.304412503772871,0.1278020171916061
+-1,-1,0.36,0.48,0.5,0.64,0.36,0.011117260803746984,0.3133333333333333,0.052618880243359395,0.29200000000000004,0.07334367009403865,0.26199999999999996,0.11778483188857766,0.43522222222222234,0.2943241147857984,0.121057858108288

Information-Retrieval_evaluation_NanoNQ_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.32,0.48,0.52,0.54,0.32,0.29,0.16666666666666663,0.45,0.10800000000000001,0.49,0.05800000000000001,0.52,0.40852380952380957,0.4230776979752646,0.4024677771121777
 -1,-1,0.4,0.58,0.6,0.7,0.4,0.36,0.20666666666666664,0.56,0.128,0.58,0.07400000000000001,0.67,0.5013888888888889,0.5294958669542686,0.48856950315109743
 -1,-1,0.42,0.58,0.66,0.76,0.42,0.38,0.19333333333333333,0.54,0.136,0.62,0.08399999999999999,0.74,0.5124126984126984,0.5512960895859668,0.4936325526558866
+-1,-1,0.44,0.56,0.64,0.74,0.44,0.41,0.19333333333333333,0.53,0.132,0.61,0.078,0.7,0.523047619047619,0.5519763173860046,0.5093692007660835

Information-Retrieval_evaluation_NanoQuoraRetrieval_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.82,0.92,0.94,0.96,0.82,0.7073333333333334,0.37999999999999995,0.882,0.23999999999999996,0.9126666666666666,0.12799999999999997,0.946,0.8678571428571429,0.877034246688313,0.8502483608158665
 -1,-1,0.88,0.94,0.96,0.98,0.88,0.7573333333333333,0.38666666666666655,0.902,0.25199999999999995,0.9393333333333334,0.13399999999999998,0.9733333333333334,0.9111904761904762,0.9171619661571674,0.8950605680868838
 -1,-1,0.48,0.56,0.62,0.66,0.48,0.44666666666666666,0.21333333333333335,0.514,0.14400000000000002,0.5673333333333334,0.08199999999999999,0.6213333333333333,0.539079365079365,0.5489572541869099,0.5287672335161302
+-1,-1,0.8,0.94,0.94,0.96,0.8,0.6973333333333332,0.4,0.9086666666666667,0.25199999999999995,0.9259999999999999,0.12799999999999997,0.946,0.8655555555555554,0.8793556978310312,0.8572134707397865

Information-Retrieval_evaluation_NanoSCIDOCS_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.42,0.72,0.78,0.84,0.42,0.08600000000000002,0.34666666666666657,0.21566666666666667,0.27599999999999997,0.2836666666666667,0.18,0.3686666666666667,0.571579365079365,0.36265976831413305,0.2806483158502371
 -1,-1,0.46,0.7,0.74,0.8,0.46,0.09666666666666668,0.33333333333333326,0.20666666666666664,0.276,0.2826666666666667,0.18599999999999997,0.38166666666666665,0.5806666666666667,0.3746688514609652,0.2919508397066463
 -1,-1,0.46,0.6,0.66,0.76,0.46,0.09666666666666668,0.31333333333333335,0.19266666666666665,0.264,0.26866666666666666,0.16799999999999998,0.3436666666666666,0.5551904761904762,0.3502034656888579,0.28238523085822703
+-1,-1,0.44,0.68,0.72,0.88,0.44,0.09266666666666667,0.3533333333333333,0.21666666666666667,0.28,0.2856666666666667,0.19399999999999995,0.39666666666666656,0.5817936507936508,0.3839901352188253,0.29483854671038273

Information-Retrieval_evaluation_NanoSciFact_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.34,0.5,0.6,0.64,0.34,0.305,0.18,0.47,0.14,0.59,0.07400000000000001,0.63,0.43969047619047613,0.4774008181239152,0.434487015231779
 -1,-1,0.4,0.52,0.58,0.68,0.4,0.355,0.19333333333333333,0.495,0.136,0.57,0.08,0.68,0.4829365079365079,0.5213214878209026,0.47407253898671803
 -1,-1,0.38,0.46,0.54,0.56,0.38,0.345,0.15999999999999998,0.43,0.12399999999999999,0.52,0.064,0.54,0.4325555555555555,0.4484758215930127,0.42353656705038156
+-1,-1,0.46,0.54,0.64,0.74,0.46,0.415,0.2,0.52,0.14400000000000002,0.625,0.08599999999999998,0.74,0.5398333333333334,0.5798929270624772,0.5303142135642136

Information-Retrieval_evaluation_NanoTouche2020_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.5918367346938775,0.7959183673469388,0.8571428571428571,0.9591836734693877,0.5918367346938775,0.04284865778016241,0.5510204081632653,0.12174900810036214,0.4653061224489796,0.1691643492468857,0.3857142857142857,0.26141802214247595,0.71274700356333,0.4505217653784499,0.342868501287415
 -1,-1,0.673469387755102,0.8367346938775511,0.8979591836734694,0.9795918367346939,0.673469387755102,0.046731422569131785,0.5578231292517006,0.12234845881617477,0.5142857142857143,0.18500594654519,0.40816326530612246,0.27622055850807264,0.7676546809199871,0.48003101771257267,0.3564545575416437
 -1,-1,0.30612244897959184,0.5714285714285714,0.6938775510204082,0.7755102040816326,0.30612244897959184,0.024834233991487722,0.346938775510204,0.08169446144223932,0.33469387755102037,0.12462402887314433,0.2755102040816326,0.19542890091501577,0.4607142857142857,0.3078492074305264,0.25713131957408375
+-1,-1,0.6530612244897959,0.7755102040816326,0.8571428571428571,0.9387755102040817,0.6530612244897959,0.04497799044877029,0.5374149659863945,0.11546047315036259,0.5020408163265306,0.18376483178712996,0.4040816326530612,0.2772240574916768,0.7406705539358601,0.4711126371779877,0.35552822693119057

NanoBEIR_evaluation_mean_results.csv

@@ -20,3 +20,4 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 -1,-1,0.4270643642072213,0.5981475667189953,0.6720879120879121,0.7476295133437992,0.4270643642072213,0.23315431874023762,0.2736682365253794,0.36944587624315167,0.2121004709576138,0.4298659303363733,0.14505494505494507,0.5007262577981157,0.5309805387356407,0.4575299449426243,0.3816850826333798
 -1,-1,0.46411302982731556,0.6151334379905808,0.6859968602825746,0.772276295133438,0.46411302982731556,0.24905479075350428,0.27778126635269496,0.3801010379714086,0.21894505494505495,0.4452995234677705,0.15232025117739403,0.5306900717106877,0.5578818619022701,0.4811652493448767,0.40018437123326334
 -1,-1,0.3604709576138147,0.5162637362637362,0.5949136577708006,0.6735007849293564,0.3604709576138147,0.202922092422317,0.23027734170591313,0.3226237613264772,0.18636106750392464,0.388802072171577,0.13211616954474098,0.45784152814874024,0.45829609279609274,0.4074449027777909,0.34005896312642037
+-1,-1,0.4794662480376766,0.6550392464678179,0.7167032967032967,0.8060596546310832,0.4794662480376766,0.26612830766551065,0.29723704866562006,0.41246090821921455,0.22723390894819467,0.4764429392718,0.1584678178963893,0.5559730399550165,0.5851309461513544,0.5087337546066613,0.42587177797164666

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

@@ -228,3 +228,24 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 1.1494252873563218,1000,0.32,0.56,0.62,0.7,0.32,0.32,0.18666666666666668,0.56,0.124,0.62,0.07,0.7,0.4528571428571428,0.5131987595360931,0.46741188393751065
 1.4367816091954024,1250,0.3,0.58,0.66,0.72,0.3,0.3,0.19333333333333333,0.58,0.132,0.66,0.07200000000000001,0.72,0.4479999999999999,0.5150223664102319,0.4605201186512495
 1.7241379310344827,1500,0.38,0.6,0.66,0.68,0.38,0.38,0.2,0.6,0.132,0.66,0.068,0.68,0.48966666666666664,0.5369233576215849,0.5049307222721455
+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
+0.28735632183908044,250,0.28,0.5,0.56,0.7,0.28,0.28,0.16666666666666669,0.5,0.11200000000000002,0.56,0.07,0.7,0.41607936507936505,0.48386691977254304,0.430943020655829
+0.5747126436781609,500,0.34,0.48,0.56,0.7,0.34,0.34,0.15999999999999998,0.48,0.11200000000000002,0.56,0.07,0.7,0.4404365079365079,0.5014869647624693,0.45357060904087726
+0.8620689655172413,750,0.34,0.52,0.56,0.7,0.34,0.34,0.1733333333333333,0.52,0.11200000000000002,0.56,0.07,0.7,0.4460476190476189,0.5063015500955226,0.4585367805515368
+1.1494252873563218,1000,0.32,0.54,0.56,0.7,0.32,0.32,0.18,0.54,0.11200000000000002,0.56,0.07,0.7,0.4377460317460317,0.4995934478919206,0.45278117615074137
+1.4367816091954024,1250,0.4,0.56,0.58,0.7,0.4,0.4,0.18666666666666668,0.56,0.11600000000000002,0.58,0.07,0.7,0.49219047619047623,0.5417745696635028,0.5051735461075432
+1.7241379310344827,1500,0.34,0.56,0.6,0.68,0.34,0.34,0.18666666666666668,0.56,0.12,0.6,0.068,0.68,0.4522460317460316,0.5073063905757779,0.468957600035666
+2.0114942528735633,1750,0.3,0.56,0.58,0.74,0.3,0.3,0.18666666666666668,0.56,0.11599999999999999,0.58,0.07400000000000001,0.74,0.44049206349206343,0.5120905526456327,0.4505686244393934
+2.2988505747126435,2000,0.3,0.5,0.6,0.68,0.3,0.3,0.16666666666666669,0.5,0.12,0.6,0.068,0.68,0.42752380952380953,0.4884933448235381,0.44397199959911765
+2.586206896551724,2250,0.34,0.48,0.6,0.64,0.34,0.34,0.15999999999999998,0.48,0.12000000000000002,0.6,0.06400000000000002,0.64,0.4355555555555556,0.4848231087320362,0.4532567794478756
+2.873563218390805,2500,0.3,0.52,0.6,0.66,0.3,0.3,0.1733333333333333,0.52,0.12000000000000002,0.6,0.066,0.66,0.4180555555555555,0.4765239390162443,0.4351733266815959
+3.160919540229885,2750,0.28,0.5,0.56,0.68,0.28,0.28,0.16666666666666669,0.5,0.11200000000000002,0.56,0.068,0.68,0.4180238095238094,0.48127976113713844,0.4338682305533546
+3.4482758620689653,3000,0.3,0.54,0.56,0.64,0.3,0.3,0.18,0.54,0.11200000000000002,0.56,0.064,0.64,0.42222222222222217,0.4753256304597703,0.44117251374038907
+3.735632183908046,3250,0.36,0.52,0.56,0.66,0.36,0.36,0.1733333333333333,0.52,0.11200000000000002,0.56,0.066,0.66,0.4526031746031746,0.5021693083774612,0.4691211775405237
+4.022988505747127,3500,0.36,0.54,0.6,0.7,0.36,0.36,0.18,0.54,0.12,0.6,0.07,0.7,0.47280158730158717,0.5271777117933837,0.48502674441960153
+4.310344827586207,3750,0.32,0.52,0.58,0.68,0.32,0.32,0.1733333333333333,0.52,0.11599999999999999,0.58,0.068,0.68,0.4373888888888889,0.4956537654273035,0.4543361902766289
+4.597701149425287,4000,0.24,0.52,0.6,0.68,0.24,0.24,0.1733333333333333,0.52,0.12000000000000002,0.6,0.068,0.68,0.39546031746031735,0.46490772523988855,0.40999466000995477
+4.885057471264368,4250,0.32,0.5,0.6,0.68,0.32,0.32,0.16666666666666663,0.5,0.12000000000000002,0.6,0.068,0.68,0.4368888888888888,0.49540821000065854,0.45259852440856
+5.172413793103448,4500,0.3,0.5,0.58,0.68,0.3,0.3,0.16666666666666663,0.5,0.11599999999999999,0.58,0.068,0.68,0.42316666666666664,0.48508399522056306,0.4372737798319072
+5.459770114942529,4750,0.28,0.5,0.58,0.7,0.28,0.28,0.16666666666666669,0.5,0.11599999999999999,0.58,0.07,0.7,0.41788888888888887,0.48552693907025524,0.430681710021151
+5.747126436781609,5000,0.3,0.5,0.58,0.68,0.3,0.3,0.16666666666666669,0.5,0.11599999999999999,0.58,0.068,0.68,0.4262222222222222,0.48741389266955737,0.44072094685707097

eval/Information-Retrieval_evaluation_NanoNQ_results.csv

@@ -228,3 +228,24 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 1.1494252873563218,1000,0.42,0.56,0.6,0.74,0.42,0.39,0.18666666666666665,0.52,0.124,0.57,0.08,0.71,0.5059365079365079,0.5407782664006731,0.4917186808976933
 1.4367816091954024,1250,0.42,0.56,0.62,0.72,0.42,0.39,0.19333333333333333,0.54,0.128,0.59,0.07600000000000001,0.69,0.5031904761904762,0.5353337185097775,0.49240843733268774
 1.7241379310344827,1500,0.42,0.56,0.62,0.74,0.42,0.39,0.19333333333333333,0.53,0.128,0.59,0.07600000000000001,0.69,0.5126031746031746,0.5406765177061442,0.49920094784101693
+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
+0.28735632183908044,250,0.42,0.56,0.58,0.6,0.42,0.4,0.2,0.54,0.12400000000000003,0.56,0.066,0.59,0.49733333333333335,0.5172968156057242,0.5069326964250607
+0.5747126436781609,500,0.38,0.54,0.56,0.6,0.38,0.35,0.18666666666666665,0.51,0.11600000000000002,0.53,0.066,0.59,0.4663888888888889,0.48526853470142506,0.45877335930661844
+0.8620689655172413,750,0.32,0.46,0.52,0.6,0.32,0.3,0.15999999999999998,0.45,0.10800000000000001,0.5,0.064,0.58,0.4062698412698413,0.44149217667031576,0.40935219468801715
+1.1494252873563218,1000,0.3,0.46,0.52,0.56,0.3,0.27,0.16666666666666663,0.45,0.11200000000000002,0.51,0.06,0.55,0.395,0.4264578110355723,0.39823838816050255
+1.4367816091954024,1250,0.32,0.44,0.52,0.56,0.32,0.28,0.16,0.42,0.11199999999999999,0.5,0.06,0.54,0.3978888888888889,0.42011707706383317,0.3948498288783238
+1.7241379310344827,1500,0.26,0.46,0.5,0.56,0.26,0.24,0.15999999999999998,0.44,0.10400000000000002,0.48,0.06,0.54,0.3668333333333333,0.40100296426790266,0.36823919517972703
+2.0114942528735633,1750,0.26,0.42,0.5,0.54,0.26,0.25,0.14666666666666664,0.4,0.10400000000000001,0.47,0.05800000000000001,0.52,0.3574126984126984,0.39215719478382494,0.36333708681224947
+2.2988505747126435,2000,0.22,0.44,0.52,0.56,0.22,0.21,0.14666666666666667,0.41,0.10800000000000001,0.49,0.06000000000000001,0.54,0.35052380952380957,0.39052915135202404,0.35075435164326074
+2.586206896551724,2250,0.28,0.42,0.48,0.54,0.28,0.26,0.14666666666666664,0.41,0.1,0.45,0.05800000000000001,0.52,0.36602380952380953,0.3960187520599003,0.3702634230331686
+2.873563218390805,2500,0.3,0.44,0.5,0.56,0.3,0.26,0.15333333333333332,0.41,0.10400000000000001,0.47,0.06,0.54,0.39185714285714285,0.4118504410063874,0.37596501494348267
+3.160919540229885,2750,0.28,0.42,0.48,0.56,0.28,0.25,0.13999999999999999,0.38,0.1,0.45,0.05800000000000001,0.53,0.36577777777777776,0.3884916606005588,0.34973944260081546
+3.4482758620689653,3000,0.26,0.46,0.5,0.58,0.26,0.24,0.15333333333333332,0.42,0.10400000000000001,0.47,0.06,0.55,0.3666031746031746,0.39913104568746777,0.35674796731875824
+3.735632183908046,3250,0.24,0.4,0.48,0.56,0.24,0.22,0.13333333333333333,0.37,0.1,0.45,0.05800000000000001,0.53,0.3365000000000001,0.3715180817874944,0.32836707338954446
+4.022988505747127,3500,0.3,0.42,0.5,0.52,0.3,0.26,0.13999999999999999,0.38,0.10400000000000001,0.47,0.054000000000000006,0.49,0.37616666666666665,0.3855601103610891,0.35707379075791146
+4.310344827586207,3750,0.28,0.4,0.5,0.52,0.28,0.24,0.13999999999999999,0.37,0.10400000000000001,0.47,0.054000000000000006,0.49,0.3611666666666667,0.37662933521579606,0.34848399688758697
+4.597701149425287,4000,0.26,0.44,0.44,0.52,0.26,0.23,0.15333333333333332,0.41,0.092,0.41,0.054000000000000006,0.49,0.3558333333333333,0.37405065882904376,0.34388568109200823
+4.885057471264368,4250,0.24,0.44,0.48,0.52,0.24,0.21,0.15333333333333332,0.41,0.1,0.45,0.054000000000000006,0.49,0.3448333333333333,0.3661529586508816,0.33459322174884476
+5.172413793103448,4500,0.26,0.4,0.48,0.54,0.26,0.23,0.14,0.37,0.1,0.45,0.05600000000000001,0.51,0.3501666666666667,0.37493713821170965,0.3375598268411981
+5.459770114942529,4750,0.24,0.42,0.48,0.54,0.24,0.21,0.14666666666666667,0.39,0.1,0.45,0.05600000000000001,0.51,0.3455238095238095,0.3725107396866955,0.3342746615754654
+5.747126436781609,5000,0.26,0.4,0.48,0.54,0.26,0.23,0.14,0.37,0.1,0.45,0.05600000000000001,0.51,0.35074603174603175,0.3745207998751907,0.3364191132763434

eval/NanoBEIR_evaluation_mean_results.csv

@@ -228,3 +228,24 @@ epoch,steps,cosine-Accuracy@1,cosine-Accuracy@3,cosine-Accuracy@5,cosine-Accurac
 1.1494252873563218,1000,0.37,0.56,0.61,0.72,0.37,0.355,0.18666666666666665,0.54,0.124,0.595,0.07500000000000001,0.705,0.4793968253968254,0.5269885129683831,0.479565282417602
 1.4367816091954024,1250,0.36,0.5700000000000001,0.64,0.72,0.36,0.345,0.19333333333333333,0.56,0.13,0.625,0.07400000000000001,0.705,0.47559523809523807,0.5251780424600048,0.4764642779919686
 1.7241379310344827,1500,0.4,0.5800000000000001,0.64,0.71,0.4,0.385,0.19666666666666666,0.565,0.13,0.625,0.07200000000000001,0.685,0.5011349206349206,0.5387999376638646,0.5020658350565812
+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
+0.28735632183908044,250,0.35,0.53,0.5700000000000001,0.6499999999999999,0.35,0.34,0.18333333333333335,0.52,0.11800000000000002,0.56,0.068,0.645,0.4567063492063492,0.5005818676891336,0.46893785854044484
+0.5747126436781609,500,0.36,0.51,0.56,0.6499999999999999,0.36,0.345,0.1733333333333333,0.495,0.11400000000000002,0.545,0.068,0.645,0.4534126984126984,0.4933777497319472,0.4561719841737478
+0.8620689655172413,750,0.33,0.49,0.54,0.6499999999999999,0.33,0.32,0.16666666666666663,0.485,0.11000000000000001,0.53,0.067,0.6399999999999999,0.42615873015873007,0.47389686338291914,0.433944487619777
+1.1494252873563218,1000,0.31,0.5,0.54,0.63,0.31,0.29500000000000004,0.1733333333333333,0.495,0.11200000000000002,0.535,0.065,0.625,0.4163730158730159,0.4630256294637465,0.42550978215562196
+1.4367816091954024,1250,0.36,0.5,0.55,0.63,0.36,0.34,0.17333333333333334,0.49,0.114,0.54,0.065,0.62,0.44503968253968257,0.480945823363668,0.4500116874929335
+1.7241379310344827,1500,0.30000000000000004,0.51,0.55,0.6200000000000001,0.30000000000000004,0.29000000000000004,0.17333333333333334,0.5,0.11200000000000002,0.54,0.064,0.6100000000000001,0.4095396825396824,0.45415467742184024,0.4185983976076965
+2.0114942528735633,1750,0.28,0.49,0.54,0.64,0.28,0.275,0.16666666666666666,0.48000000000000004,0.11,0.5249999999999999,0.066,0.63,0.39895238095238095,0.45212387371472884,0.4069528556258214
+2.2988505747126435,2000,0.26,0.47,0.56,0.6200000000000001,0.26,0.255,0.15666666666666668,0.45499999999999996,0.114,0.5449999999999999,0.064,0.6100000000000001,0.38902380952380955,0.4395112480877811,0.39736317562118917
+2.586206896551724,2250,0.31000000000000005,0.44999999999999996,0.54,0.5900000000000001,0.31000000000000005,0.30000000000000004,0.15333333333333332,0.44499999999999995,0.11000000000000001,0.525,0.06100000000000001,0.5800000000000001,0.40078968253968256,0.44042093039596825,0.4117601012405221
+2.873563218390805,2500,0.3,0.48,0.55,0.6100000000000001,0.3,0.28,0.16333333333333333,0.46499999999999997,0.11200000000000002,0.5349999999999999,0.063,0.6000000000000001,0.40495634920634915,0.44418719001131585,0.4055691708125393
+3.160919540229885,2750,0.28,0.45999999999999996,0.52,0.6200000000000001,0.28,0.265,0.15333333333333332,0.44,0.10600000000000001,0.505,0.063,0.605,0.39190079365079356,0.43488571086884864,0.39180383657708506
+3.4482758620689653,3000,0.28,0.5,0.53,0.61,0.28,0.27,0.16666666666666666,0.48,0.10800000000000001,0.515,0.062,0.595,0.3944126984126984,0.43722833807361905,0.3989602405295737
+3.735632183908046,3250,0.3,0.46,0.52,0.6100000000000001,0.3,0.29,0.15333333333333332,0.445,0.10600000000000001,0.505,0.062000000000000006,0.595,0.39455158730158735,0.4368436950824778,0.39874412546503407
+4.022988505747127,3500,0.32999999999999996,0.48,0.55,0.61,0.32999999999999996,0.31,0.15999999999999998,0.46,0.112,0.5349999999999999,0.062000000000000006,0.595,0.4244841269841269,0.4563689110772364,0.4210502675887565
+4.310344827586207,3750,0.30000000000000004,0.46,0.54,0.6000000000000001,0.30000000000000004,0.28,0.15666666666666665,0.445,0.11,0.5249999999999999,0.061000000000000006,0.585,0.3992777777777778,0.4361415503215498,0.40141009358210794
+4.597701149425287,4000,0.25,0.48,0.52,0.6000000000000001,0.25,0.235,0.16333333333333333,0.46499999999999997,0.10600000000000001,0.505,0.061000000000000006,0.585,0.3756468253968253,0.41947919203446615,0.3769401705509815
+4.885057471264368,4250,0.28,0.47,0.54,0.6000000000000001,0.28,0.265,0.15999999999999998,0.45499999999999996,0.11000000000000001,0.525,0.061000000000000006,0.585,0.3908611111111111,0.43078058432577004,0.39359587307870236
+5.172413793103448,4500,0.28,0.45,0.53,0.6100000000000001,0.28,0.265,0.15333333333333332,0.435,0.108,0.515,0.062000000000000006,0.595,0.38666666666666666,0.43001056671613636,0.38741680333655265
+5.459770114942529,4750,0.26,0.45999999999999996,0.53,0.62,0.26,0.245,0.15666666666666668,0.445,0.108,0.515,0.063,0.605,0.38170634920634916,0.4290188393784754,0.3824781857983082
+5.747126436781609,5000,0.28,0.45,0.53,0.6100000000000001,0.28,0.265,0.15333333333333335,0.435,0.108,0.515,0.062000000000000006,0.595,0.388484126984127,0.43096734627237404,0.3885700300667072

final_metrics.json

@@ -1,231 +1,231 @@
 {
   "nano_beir": {
-    "NanoClimateFEVER_cosine_accuracy@1": 0.12,
-    "NanoClimateFEVER_cosine_accuracy@3": 0.24,
-    "NanoClimateFEVER_cosine_accuracy@5": 0.26,
-    "NanoClimateFEVER_cosine_accuracy@10": 0.3,
-    "NanoClimateFEVER_cosine_precision@1": 0.12,
-    "NanoClimateFEVER_cosine_precision@3": 0.08,
-    "NanoClimateFEVER_cosine_precision@5": 0.06000000000000001,
-    "NanoClimateFEVER_cosine_precision@10": 0.042,
-    "NanoClimateFEVER_cosine_recall@1": 0.05833333333333333,
-    "NanoClimateFEVER_cosine_recall@3": 0.11166666666666664,
-    "NanoClimateFEVER_cosine_recall@5": 0.13333333333333333,
-    "NanoClimateFEVER_cosine_recall@10": 0.1733333333333333,
-    "NanoClimateFEVER_cosine_ndcg@10": 0.1421323657809101,
-    "NanoClimateFEVER_cosine_mrr@10": 0.18599999999999997,
-    "NanoClimateFEVER_cosine_map@100": 0.11251502558321347,
-    "NanoDBPedia_cosine_accuracy@1": 0.62,
-    "NanoDBPedia_cosine_accuracy@3": 0.74,
-    "NanoDBPedia_cosine_accuracy@5": 0.82,
-    "NanoDBPedia_cosine_accuracy@10": 0.88,
-    "NanoDBPedia_cosine_precision@1": 0.62,
-    "NanoDBPedia_cosine_precision@3": 0.4600000000000001,
-    "NanoDBPedia_cosine_precision@5": 0.40399999999999997,
-    "NanoDBPedia_cosine_precision@10": 0.352,
-    "NanoDBPedia_cosine_recall@1": 0.05294717669991995,
-    "NanoDBPedia_cosine_recall@3": 0.11950308612173167,
-    "NanoDBPedia_cosine_recall@5": 0.1809061480074522,
-    "NanoDBPedia_cosine_recall@10": 0.26388938838721027,
-    "NanoDBPedia_cosine_ndcg@10": 0.45186572825718035,
-    "NanoDBPedia_cosine_mrr@10": 0.7002698412698414,
-    "NanoDBPedia_cosine_map@100": 0.2963971880776791,
-    "NanoFEVER_cosine_accuracy@1": 0.28,
-    "NanoFEVER_cosine_accuracy@3": 0.42,
-    "NanoFEVER_cosine_accuracy@5": 0.44,
-    "NanoFEVER_cosine_accuracy@10": 0.44,
-    "NanoFEVER_cosine_precision@1": 0.28,
-    "NanoFEVER_cosine_precision@3": 0.13999999999999999,
-    "NanoFEVER_cosine_precision@5": 0.09200000000000001,
-    "NanoFEVER_cosine_precision@10": 0.046000000000000006,
-    "NanoFEVER_cosine_recall@1": 0.27,
-    "NanoFEVER_cosine_recall@3": 0.4,
-    "NanoFEVER_cosine_recall@5": 0.43,
-    "NanoFEVER_cosine_recall@10": 0.43,
-    "NanoFEVER_cosine_ndcg@10": 0.3661113939532885,
-    "NanoFEVER_cosine_mrr@10": 0.3506666666666666,
-    "NanoFEVER_cosine_map@100": 0.3422393215160618,
-    "NanoFiQA2018_cosine_accuracy@1": 0.16,
-    "NanoFiQA2018_cosine_accuracy@3": 0.32,
-    "NanoFiQA2018_cosine_accuracy@5": 0.36,
-    "NanoFiQA2018_cosine_accuracy@10": 0.42,
-    "NanoFiQA2018_cosine_precision@1": 0.16,
-    "NanoFiQA2018_cosine_precision@3": 0.11333333333333333,
-    "NanoFiQA2018_cosine_precision@5": 0.07600000000000001,
-    "NanoFiQA2018_cosine_precision@10": 0.05000000000000001,
-    "NanoFiQA2018_cosine_recall@1": 0.09666666666666666,
-    "NanoFiQA2018_cosine_recall@3": 0.2333333333333333,
-    "NanoFiQA2018_cosine_recall@5": 0.24833333333333332,
-    "NanoFiQA2018_cosine_recall@10": 0.2977777777777778,
-    "NanoFiQA2018_cosine_ndcg@10": 0.21754856447960677,
-    "NanoFiQA2018_cosine_mrr@10": 0.23633333333333328,
-    "NanoFiQA2018_cosine_map@100": 0.18156772343708624,
-    "NanoHotpotQA_cosine_accuracy@1": 0.46,
-    "NanoHotpotQA_cosine_accuracy@3": 0.6,
-    "NanoHotpotQA_cosine_accuracy@5": 0.66,
-    "NanoHotpotQA_cosine_accuracy@10": 0.78,
-    "NanoHotpotQA_cosine_precision@1": 0.46,
-    "NanoHotpotQA_cosine_precision@3": 0.26,
-    "NanoHotpotQA_cosine_precision@5": 0.176,
-    "NanoHotpotQA_cosine_precision@10": 0.10399999999999998,
-    "NanoHotpotQA_cosine_recall@1": 0.23,
-    "NanoHotpotQA_cosine_recall@3": 0.39,
-    "NanoHotpotQA_cosine_recall@5": 0.44,
-    "NanoHotpotQA_cosine_recall@10": 0.52,
-    "NanoHotpotQA_cosine_ndcg@10": 0.45661766676516835,
-    "NanoHotpotQA_cosine_mrr@10": 0.5547222222222222,
-    "NanoHotpotQA_cosine_map@100": 0.389670298499331,
-    "NanoMSMARCO_cosine_accuracy@1": 0.46,
-    "NanoMSMARCO_cosine_accuracy@3": 0.64,
-    "NanoMSMARCO_cosine_accuracy@5": 0.78,
-    "NanoMSMARCO_cosine_accuracy@10": 0.84,
-    "NanoMSMARCO_cosine_precision@1": 0.46,
-    "NanoMSMARCO_cosine_precision@3": 0.21333333333333332,
-    "NanoMSMARCO_cosine_precision@5": 0.156,
-    "NanoMSMARCO_cosine_precision@10": 0.08399999999999999,
-    "NanoMSMARCO_cosine_recall@1": 0.46,
-    "NanoMSMARCO_cosine_recall@3": 0.64,
-    "NanoMSMARCO_cosine_recall@5": 0.78,
-    "NanoMSMARCO_cosine_recall@10": 0.84,
-    "NanoMSMARCO_cosine_ndcg@10": 0.6436072717719088,
-    "NanoMSMARCO_cosine_mrr@10": 0.5809365079365079,
-    "NanoMSMARCO_cosine_map@100": 0.5875442422678194,
-    "NanoNFCorpus_cosine_accuracy@1": 0.38,
+    "NanoClimateFEVER_cosine_accuracy@1": 0.22,
+    "NanoClimateFEVER_cosine_accuracy@3": 0.52,
+    "NanoClimateFEVER_cosine_accuracy@5": 0.56,
+    "NanoClimateFEVER_cosine_accuracy@10": 0.64,
+    "NanoClimateFEVER_cosine_precision@1": 0.22,
+    "NanoClimateFEVER_cosine_precision@3": 0.1733333333333333,
+    "NanoClimateFEVER_cosine_precision@5": 0.12000000000000002,
+    "NanoClimateFEVER_cosine_precision@10": 0.076,
+    "NanoClimateFEVER_cosine_recall@1": 0.09166666666666666,
+    "NanoClimateFEVER_cosine_recall@3": 0.23399999999999999,
+    "NanoClimateFEVER_cosine_recall@5": 0.27899999999999997,
+    "NanoClimateFEVER_cosine_recall@10": 0.32733333333333337,
+    "NanoClimateFEVER_cosine_ndcg@10": 0.26355221051084216,
+    "NanoClimateFEVER_cosine_mrr@10": 0.3648571428571428,
+    "NanoClimateFEVER_cosine_map@100": 0.19506453252604283,
+    "NanoDBPedia_cosine_accuracy@1": 0.66,
+    "NanoDBPedia_cosine_accuracy@3": 0.8,
+    "NanoDBPedia_cosine_accuracy@5": 0.84,
+    "NanoDBPedia_cosine_accuracy@10": 0.94,
+    "NanoDBPedia_cosine_precision@1": 0.66,
+    "NanoDBPedia_cosine_precision@3": 0.5066666666666666,
+    "NanoDBPedia_cosine_precision@5": 0.4160000000000001,
+    "NanoDBPedia_cosine_precision@10": 0.368,
+    "NanoDBPedia_cosine_recall@1": 0.07082671665308901,
+    "NanoDBPedia_cosine_recall@3": 0.13730927885289243,
+    "NanoDBPedia_cosine_recall@5": 0.1661893911919123,
+    "NanoDBPedia_cosine_recall@10": 0.25034697924131,
+    "NanoDBPedia_cosine_ndcg@10": 0.4728916713272679,
+    "NanoDBPedia_cosine_mrr@10": 0.7467142857142857,
+    "NanoDBPedia_cosine_map@100": 0.30396194965649254,
+    "NanoFEVER_cosine_accuracy@1": 0.6,
+    "NanoFEVER_cosine_accuracy@3": 0.9,
+    "NanoFEVER_cosine_accuracy@5": 0.96,
+    "NanoFEVER_cosine_accuracy@10": 0.98,
+    "NanoFEVER_cosine_precision@1": 0.6,
+    "NanoFEVER_cosine_precision@3": 0.30666666666666664,
+    "NanoFEVER_cosine_precision@5": 0.19599999999999995,
+    "NanoFEVER_cosine_precision@10": 0.09999999999999998,
+    "NanoFEVER_cosine_recall@1": 0.5566666666666666,
+    "NanoFEVER_cosine_recall@3": 0.8466666666666667,
+    "NanoFEVER_cosine_recall@5": 0.9066666666666667,
+    "NanoFEVER_cosine_recall@10": 0.9266666666666667,
+    "NanoFEVER_cosine_ndcg@10": 0.7716934881014161,
+    "NanoFEVER_cosine_mrr@10": 0.7516666666666666,
+    "NanoFEVER_cosine_map@100": 0.7061067287383075,
+    "NanoFiQA2018_cosine_accuracy@1": 0.36,
+    "NanoFiQA2018_cosine_accuracy@3": 0.48,
+    "NanoFiQA2018_cosine_accuracy@5": 0.6,
+    "NanoFiQA2018_cosine_accuracy@10": 0.66,
+    "NanoFiQA2018_cosine_precision@1": 0.36,
+    "NanoFiQA2018_cosine_precision@3": 0.20666666666666667,
+    "NanoFiQA2018_cosine_precision@5": 0.16,
+    "NanoFiQA2018_cosine_precision@10": 0.08999999999999998,
+    "NanoFiQA2018_cosine_recall@1": 0.19941269841269843,
+    "NanoFiQA2018_cosine_recall@3": 0.2806031746031746,
+    "NanoFiQA2018_cosine_recall@5": 0.3781269841269841,
+    "NanoFiQA2018_cosine_recall@10": 0.4256269841269841,
+    "NanoFiQA2018_cosine_ndcg@10": 0.35641512120073343,
+    "NanoFiQA2018_cosine_mrr@10": 0.4481666666666666,
+    "NanoFiQA2018_cosine_map@100": 0.29847071848639917,
+    "NanoHotpotQA_cosine_accuracy@1": 0.74,
+    "NanoHotpotQA_cosine_accuracy@3": 0.82,
+    "NanoHotpotQA_cosine_accuracy@5": 0.84,
+    "NanoHotpotQA_cosine_accuracy@10": 0.86,
+    "NanoHotpotQA_cosine_precision@1": 0.74,
+    "NanoHotpotQA_cosine_precision@3": 0.33333333333333326,
+    "NanoHotpotQA_cosine_precision@5": 0.21599999999999997,
+    "NanoHotpotQA_cosine_precision@10": 0.12399999999999999,
+    "NanoHotpotQA_cosine_recall@1": 0.37,
+    "NanoHotpotQA_cosine_recall@3": 0.5,
+    "NanoHotpotQA_cosine_recall@5": 0.54,
+    "NanoHotpotQA_cosine_recall@10": 0.62,
+    "NanoHotpotQA_cosine_ndcg@10": 0.6038529756799117,
+    "NanoHotpotQA_cosine_mrr@10": 0.7883333333333334,
+    "NanoHotpotQA_cosine_map@100": 0.5248449945769594,
+    "NanoMSMARCO_cosine_accuracy@1": 0.34,
+    "NanoMSMARCO_cosine_accuracy@3": 0.6,
+    "NanoMSMARCO_cosine_accuracy@5": 0.66,
+    "NanoMSMARCO_cosine_accuracy@10": 0.7,
+    "NanoMSMARCO_cosine_precision@1": 0.34,
+    "NanoMSMARCO_cosine_precision@3": 0.2,
+    "NanoMSMARCO_cosine_precision@5": 0.132,
+    "NanoMSMARCO_cosine_precision@10": 0.07,
+    "NanoMSMARCO_cosine_recall@1": 0.34,
+    "NanoMSMARCO_cosine_recall@3": 0.6,
+    "NanoMSMARCO_cosine_recall@5": 0.66,
+    "NanoMSMARCO_cosine_recall@10": 0.7,
+    "NanoMSMARCO_cosine_ndcg@10": 0.5303422656736582,
+    "NanoMSMARCO_cosine_mrr@10": 0.4747460317460317,
+    "NanoMSMARCO_cosine_map@100": 0.48845125969077435,
+    "NanoNFCorpus_cosine_accuracy@1": 0.36,
     "NanoNFCorpus_cosine_accuracy@3": 0.48,
-    "NanoNFCorpus_cosine_accuracy@5": 0.56,
-    "NanoNFCorpus_cosine_accuracy@10": 0.7,
-    "NanoNFCorpus_cosine_precision@1": 0.38,
-    "NanoNFCorpus_cosine_precision@3": 0.33333333333333326,
-    "NanoNFCorpus_cosine_precision@5": 0.32,
-    "NanoNFCorpus_cosine_precision@10": 0.278,
-    "NanoNFCorpus_cosine_recall@1": 0.01687245746538031,
-    "NanoNFCorpus_cosine_recall@3": 0.04124468301356519,
-    "NanoNFCorpus_cosine_recall@5": 0.06123009468323657,
-    "NanoNFCorpus_cosine_recall@10": 0.10651046552028522,
-    "NanoNFCorpus_cosine_ndcg@10": 0.304412503772871,
-    "NanoNFCorpus_cosine_mrr@10": 0.4581269841269841,
-    "NanoNFCorpus_cosine_map@100": 0.1278020171916061,
-    "NanoNQ_cosine_accuracy@1": 0.42,
-    "NanoNQ_cosine_accuracy@3": 0.58,
-    "NanoNQ_cosine_accuracy@5": 0.66,
-    "NanoNQ_cosine_accuracy@10": 0.76,
-    "NanoNQ_cosine_precision@1": 0.42,
+    "NanoNFCorpus_cosine_accuracy@5": 0.5,
+    "NanoNFCorpus_cosine_accuracy@10": 0.64,
+    "NanoNFCorpus_cosine_precision@1": 0.36,
+    "NanoNFCorpus_cosine_precision@3": 0.3133333333333333,
+    "NanoNFCorpus_cosine_precision@5": 0.29200000000000004,
+    "NanoNFCorpus_cosine_precision@10": 0.26199999999999996,
+    "NanoNFCorpus_cosine_recall@1": 0.011117260803746984,
+    "NanoNFCorpus_cosine_recall@3": 0.052618880243359395,
+    "NanoNFCorpus_cosine_recall@5": 0.07334367009403865,
+    "NanoNFCorpus_cosine_recall@10": 0.11778483188857766,
+    "NanoNFCorpus_cosine_ndcg@10": 0.2943241147857984,
+    "NanoNFCorpus_cosine_mrr@10": 0.43522222222222234,
+    "NanoNFCorpus_cosine_map@100": 0.121057858108288,
+    "NanoNQ_cosine_accuracy@1": 0.44,
+    "NanoNQ_cosine_accuracy@3": 0.56,
+    "NanoNQ_cosine_accuracy@5": 0.64,
+    "NanoNQ_cosine_accuracy@10": 0.74,
+    "NanoNQ_cosine_precision@1": 0.44,
     "NanoNQ_cosine_precision@3": 0.19333333333333333,
-    "NanoNQ_cosine_precision@5": 0.136,
-    "NanoNQ_cosine_precision@10": 0.08399999999999999,
-    "NanoNQ_cosine_recall@1": 0.38,
-    "NanoNQ_cosine_recall@3": 0.54,
-    "NanoNQ_cosine_recall@5": 0.62,
-    "NanoNQ_cosine_recall@10": 0.74,
-    "NanoNQ_cosine_ndcg@10": 0.5512960895859668,
-    "NanoNQ_cosine_mrr@10": 0.5124126984126984,
-    "NanoNQ_cosine_map@100": 0.4936325526558866,
-    "NanoQuoraRetrieval_cosine_accuracy@1": 0.48,
-    "NanoQuoraRetrieval_cosine_accuracy@3": 0.56,
-    "NanoQuoraRetrieval_cosine_accuracy@5": 0.62,
-    "NanoQuoraRetrieval_cosine_accuracy@10": 0.66,
-    "NanoQuoraRetrieval_cosine_precision@1": 0.48,
-    "NanoQuoraRetrieval_cosine_precision@3": 0.21333333333333335,
-    "NanoQuoraRetrieval_cosine_precision@5": 0.14400000000000002,
-    "NanoQuoraRetrieval_cosine_precision@10": 0.08199999999999999,
-    "NanoQuoraRetrieval_cosine_recall@1": 0.44666666666666666,
-    "NanoQuoraRetrieval_cosine_recall@3": 0.514,
-    "NanoQuoraRetrieval_cosine_recall@5": 0.5673333333333334,
-    "NanoQuoraRetrieval_cosine_recall@10": 0.6213333333333333,
-    "NanoQuoraRetrieval_cosine_ndcg@10": 0.5489572541869099,
-    "NanoQuoraRetrieval_cosine_mrr@10": 0.539079365079365,
-    "NanoQuoraRetrieval_cosine_map@100": 0.5287672335161302,
-    "NanoSCIDOCS_cosine_accuracy@1": 0.46,
-    "NanoSCIDOCS_cosine_accuracy@3": 0.6,
-    "NanoSCIDOCS_cosine_accuracy@5": 0.66,
-    "NanoSCIDOCS_cosine_accuracy@10": 0.76,
-    "NanoSCIDOCS_cosine_precision@1": 0.46,
-    "NanoSCIDOCS_cosine_precision@3": 0.31333333333333335,
-    "NanoSCIDOCS_cosine_precision@5": 0.264,
-    "NanoSCIDOCS_cosine_precision@10": 0.16799999999999998,
-    "NanoSCIDOCS_cosine_recall@1": 0.09666666666666668,
-    "NanoSCIDOCS_cosine_recall@3": 0.19266666666666665,
-    "NanoSCIDOCS_cosine_recall@5": 0.26866666666666666,
-    "NanoSCIDOCS_cosine_recall@10": 0.3436666666666666,
-    "NanoSCIDOCS_cosine_ndcg@10": 0.3502034656888579,
-    "NanoSCIDOCS_cosine_mrr@10": 0.5551904761904762,
-    "NanoSCIDOCS_cosine_map@100": 0.28238523085822703,
+    "NanoNQ_cosine_precision@5": 0.132,
+    "NanoNQ_cosine_precision@10": 0.078,
+    "NanoNQ_cosine_recall@1": 0.41,
+    "NanoNQ_cosine_recall@3": 0.53,
+    "NanoNQ_cosine_recall@5": 0.61,
+    "NanoNQ_cosine_recall@10": 0.7,
+    "NanoNQ_cosine_ndcg@10": 0.5519763173860046,
+    "NanoNQ_cosine_mrr@10": 0.523047619047619,
+    "NanoNQ_cosine_map@100": 0.5093692007660835,
+    "NanoQuoraRetrieval_cosine_accuracy@1": 0.8,
+    "NanoQuoraRetrieval_cosine_accuracy@3": 0.94,
+    "NanoQuoraRetrieval_cosine_accuracy@5": 0.94,
+    "NanoQuoraRetrieval_cosine_accuracy@10": 0.96,
+    "NanoQuoraRetrieval_cosine_precision@1": 0.8,
+    "NanoQuoraRetrieval_cosine_precision@3": 0.4,
+    "NanoQuoraRetrieval_cosine_precision@5": 0.25199999999999995,
+    "NanoQuoraRetrieval_cosine_precision@10": 0.12799999999999997,
+    "NanoQuoraRetrieval_cosine_recall@1": 0.6973333333333332,
+    "NanoQuoraRetrieval_cosine_recall@3": 0.9086666666666667,
+    "NanoQuoraRetrieval_cosine_recall@5": 0.9259999999999999,
+    "NanoQuoraRetrieval_cosine_recall@10": 0.946,
+    "NanoQuoraRetrieval_cosine_ndcg@10": 0.8793556978310312,
+    "NanoQuoraRetrieval_cosine_mrr@10": 0.8655555555555554,
+    "NanoQuoraRetrieval_cosine_map@100": 0.8572134707397865,
+    "NanoSCIDOCS_cosine_accuracy@1": 0.44,
+    "NanoSCIDOCS_cosine_accuracy@3": 0.68,
+    "NanoSCIDOCS_cosine_accuracy@5": 0.72,
+    "NanoSCIDOCS_cosine_accuracy@10": 0.88,
+    "NanoSCIDOCS_cosine_precision@1": 0.44,
+    "NanoSCIDOCS_cosine_precision@3": 0.3533333333333333,
+    "NanoSCIDOCS_cosine_precision@5": 0.28,
+    "NanoSCIDOCS_cosine_precision@10": 0.19399999999999995,
+    "NanoSCIDOCS_cosine_recall@1": 0.09266666666666667,
+    "NanoSCIDOCS_cosine_recall@3": 0.21666666666666667,
+    "NanoSCIDOCS_cosine_recall@5": 0.2856666666666667,
+    "NanoSCIDOCS_cosine_recall@10": 0.39666666666666656,
+    "NanoSCIDOCS_cosine_ndcg@10": 0.3839901352188253,
+    "NanoSCIDOCS_cosine_mrr@10": 0.5817936507936508,
+    "NanoSCIDOCS_cosine_map@100": 0.29483854671038273,
     "NanoArguAna_cosine_accuracy@1": 0.16,
-    "NanoArguAna_cosine_accuracy@3": 0.5,
-    "NanoArguAna_cosine_accuracy@5": 0.68,
-    "NanoArguAna_cosine_accuracy@10": 0.88,
+    "NanoArguAna_cosine_accuracy@3": 0.42,
+    "NanoArguAna_cosine_accuracy@5": 0.56,
+    "NanoArguAna_cosine_accuracy@10": 0.8,
     "NanoArguAna_cosine_precision@1": 0.16,
-    "NanoArguAna_cosine_precision@3": 0.16666666666666663,
-    "NanoArguAna_cosine_precision@5": 0.136,
-    "NanoArguAna_cosine_precision@10": 0.088,
+    "NanoArguAna_cosine_precision@3": 0.13999999999999999,
+    "NanoArguAna_cosine_precision@5": 0.11200000000000003,
+    "NanoArguAna_cosine_precision@10": 0.08,
     "NanoArguAna_cosine_recall@1": 0.16,
-    "NanoArguAna_cosine_recall@3": 0.5,
-    "NanoArguAna_cosine_recall@5": 0.68,
-    "NanoArguAna_cosine_recall@10": 0.88,
-    "NanoArguAna_cosine_ndcg@10": 0.5077064028450745,
-    "NanoArguAna_cosine_mrr@10": 0.3908412698412698,
-    "NanoArguAna_cosine_map@100": 0.3975778004159583,
-    "NanoSciFact_cosine_accuracy@1": 0.38,
-    "NanoSciFact_cosine_accuracy@3": 0.46,
-    "NanoSciFact_cosine_accuracy@5": 0.54,
-    "NanoSciFact_cosine_accuracy@10": 0.56,
-    "NanoSciFact_cosine_precision@1": 0.38,
-    "NanoSciFact_cosine_precision@3": 0.15999999999999998,
-    "NanoSciFact_cosine_precision@5": 0.12399999999999999,
-    "NanoSciFact_cosine_precision@10": 0.064,
-    "NanoSciFact_cosine_recall@1": 0.345,
-    "NanoSciFact_cosine_recall@3": 0.43,
-    "NanoSciFact_cosine_recall@5": 0.52,
-    "NanoSciFact_cosine_recall@10": 0.54,
-    "NanoSciFact_cosine_ndcg@10": 0.4484758215930127,
-    "NanoSciFact_cosine_mrr@10": 0.4325555555555555,
-    "NanoSciFact_cosine_map@100": 0.42353656705038156,
-    "NanoTouche2020_cosine_accuracy@1": 0.30612244897959184,
-    "NanoTouche2020_cosine_accuracy@3": 0.5714285714285714,
-    "NanoTouche2020_cosine_accuracy@5": 0.6938775510204082,
-    "NanoTouche2020_cosine_accuracy@10": 0.7755102040816326,
-    "NanoTouche2020_cosine_precision@1": 0.30612244897959184,
-    "NanoTouche2020_cosine_precision@3": 0.346938775510204,
-    "NanoTouche2020_cosine_precision@5": 0.33469387755102037,
-    "NanoTouche2020_cosine_precision@10": 0.2755102040816326,
-    "NanoTouche2020_cosine_recall@1": 0.024834233991487722,
-    "NanoTouche2020_cosine_recall@3": 0.08169446144223932,
-    "NanoTouche2020_cosine_recall@5": 0.12462402887314433,
-    "NanoTouche2020_cosine_recall@10": 0.19542890091501577,
-    "NanoTouche2020_cosine_ndcg@10": 0.3078492074305264,
-    "NanoTouche2020_cosine_mrr@10": 0.4607142857142857,
-    "NanoTouche2020_cosine_map@100": 0.25713131957408375,
-    "NanoBEIR_mean_cosine_accuracy@1": 0.3604709576138147,
-    "NanoBEIR_mean_cosine_accuracy@3": 0.5162637362637362,
-    "NanoBEIR_mean_cosine_accuracy@5": 0.5949136577708006,
-    "NanoBEIR_mean_cosine_accuracy@10": 0.6735007849293564,
-    "NanoBEIR_mean_cosine_precision@1": 0.3604709576138147,
-    "NanoBEIR_mean_cosine_precision@3": 0.23027734170591313,
-    "NanoBEIR_mean_cosine_precision@5": 0.18636106750392464,
-    "NanoBEIR_mean_cosine_precision@10": 0.13211616954474098,
-    "NanoBEIR_mean_cosine_recall@1": 0.202922092422317,
-    "NanoBEIR_mean_cosine_recall@3": 0.3226237613264772,
-    "NanoBEIR_mean_cosine_recall@5": 0.388802072171577,
-    "NanoBEIR_mean_cosine_recall@10": 0.45784152814874024,
-    "NanoBEIR_mean_cosine_ndcg@10": 0.4074449027777909,
-    "NanoBEIR_mean_cosine_mrr@10": 0.45829609279609274,
-    "NanoBEIR_mean_cosine_map@100": 0.34005896312642037
+    "NanoArguAna_cosine_recall@3": 0.42,
+    "NanoArguAna_cosine_recall@5": 0.56,
+    "NanoArguAna_cosine_recall@10": 0.8,
+    "NanoArguAna_cosine_ndcg@10": 0.4541392479306418,
+    "NanoArguAna_cosine_mrr@10": 0.346095238095238,
+    "NanoArguAna_cosine_map@100": 0.3511114131364851,
+    "NanoSciFact_cosine_accuracy@1": 0.46,
+    "NanoSciFact_cosine_accuracy@3": 0.54,
+    "NanoSciFact_cosine_accuracy@5": 0.64,
+    "NanoSciFact_cosine_accuracy@10": 0.74,
+    "NanoSciFact_cosine_precision@1": 0.46,
+    "NanoSciFact_cosine_precision@3": 0.2,
+    "NanoSciFact_cosine_precision@5": 0.14400000000000002,
+    "NanoSciFact_cosine_precision@10": 0.08599999999999998,
+    "NanoSciFact_cosine_recall@1": 0.415,
+    "NanoSciFact_cosine_recall@3": 0.52,
+    "NanoSciFact_cosine_recall@5": 0.625,
+    "NanoSciFact_cosine_recall@10": 0.74,
+    "NanoSciFact_cosine_ndcg@10": 0.5798929270624772,
+    "NanoSciFact_cosine_mrr@10": 0.5398333333333334,
+    "NanoSciFact_cosine_map@100": 0.5303142135642136,
+    "NanoTouche2020_cosine_accuracy@1": 0.6530612244897959,
+    "NanoTouche2020_cosine_accuracy@3": 0.7755102040816326,
+    "NanoTouche2020_cosine_accuracy@5": 0.8571428571428571,
+    "NanoTouche2020_cosine_accuracy@10": 0.9387755102040817,
+    "NanoTouche2020_cosine_precision@1": 0.6530612244897959,
+    "NanoTouche2020_cosine_precision@3": 0.5374149659863945,
+    "NanoTouche2020_cosine_precision@5": 0.5020408163265306,
+    "NanoTouche2020_cosine_precision@10": 0.4040816326530612,
+    "NanoTouche2020_cosine_recall@1": 0.04497799044877029,
+    "NanoTouche2020_cosine_recall@3": 0.11546047315036259,
+    "NanoTouche2020_cosine_recall@5": 0.18376483178712996,
+    "NanoTouche2020_cosine_recall@10": 0.2772240574916768,
+    "NanoTouche2020_cosine_ndcg@10": 0.4711126371779877,
+    "NanoTouche2020_cosine_mrr@10": 0.7406705539358601,
+    "NanoTouche2020_cosine_map@100": 0.35552822693119057,
+    "NanoBEIR_mean_cosine_accuracy@1": 0.4794662480376766,
+    "NanoBEIR_mean_cosine_accuracy@3": 0.6550392464678179,
+    "NanoBEIR_mean_cosine_accuracy@5": 0.7167032967032967,
+    "NanoBEIR_mean_cosine_accuracy@10": 0.8060596546310832,
+    "NanoBEIR_mean_cosine_precision@1": 0.4794662480376766,
+    "NanoBEIR_mean_cosine_precision@3": 0.29723704866562006,
+    "NanoBEIR_mean_cosine_precision@5": 0.22723390894819467,
+    "NanoBEIR_mean_cosine_precision@10": 0.1584678178963893,
+    "NanoBEIR_mean_cosine_recall@1": 0.26612830766551065,
+    "NanoBEIR_mean_cosine_recall@3": 0.41246090821921455,
+    "NanoBEIR_mean_cosine_recall@5": 0.4764429392718,
+    "NanoBEIR_mean_cosine_recall@10": 0.5559730399550165,
+    "NanoBEIR_mean_cosine_ndcg@10": 0.5087337546066613,
+    "NanoBEIR_mean_cosine_mrr@10": 0.5851309461513544,
+    "NanoBEIR_mean_cosine_map@100": 0.42587177797164666
   },
   "beir_touche2020": {
-    "BeIR-touche2020-subset-test_cosine_accuracy@1": 0.2653061224489796,
-    "BeIR-touche2020-subset-test_cosine_accuracy@3": 0.5714285714285714,
-    "BeIR-touche2020-subset-test_cosine_accuracy@5": 0.6938775510204082,
-    "BeIR-touche2020-subset-test_cosine_accuracy@10": 0.7551020408163265,
-    "BeIR-touche2020-subset-test_cosine_precision@1": 0.2653061224489796,
-    "BeIR-touche2020-subset-test_cosine_precision@3": 0.3469387755102041,
-    "BeIR-touche2020-subset-test_cosine_precision@5": 0.4,
-    "BeIR-touche2020-subset-test_cosine_precision@10": 0.3755102040816327,
-    "BeIR-touche2020-subset-test_cosine_recall@1": 0.005859789020652119,
-    "BeIR-touche2020-subset-test_cosine_recall@3": 0.022771209086251597,
-    "BeIR-touche2020-subset-test_cosine_recall@5": 0.04397085795258936,
-    "BeIR-touche2020-subset-test_cosine_recall@10": 0.08239494960074849,
-    "BeIR-touche2020-subset-test_cosine_ndcg@10": 0.36314214650584564,
-    "BeIR-touche2020-subset-test_cosine_mrr@10": 0.42636054421768715,
-    "BeIR-touche2020-subset-test_cosine_map@100": 0.16579582237011886
+    "BeIR-touche2020-subset-test_cosine_accuracy@1": 0.7755102040816326,
+    "BeIR-touche2020-subset-test_cosine_accuracy@3": 0.9795918367346939,
+    "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": 0.7755102040816326,
+    "BeIR-touche2020-subset-test_cosine_precision@3": 0.7482993197278911,
+    "BeIR-touche2020-subset-test_cosine_precision@5": 0.7142857142857143,
+    "BeIR-touche2020-subset-test_cosine_precision@10": 0.6122448979591837,
+    "BeIR-touche2020-subset-test_cosine_recall@1": 0.01706832759711131,
+    "BeIR-touche2020-subset-test_cosine_recall@3": 0.049401730544661945,
+    "BeIR-touche2020-subset-test_cosine_recall@5": 0.07867619617707583,
+    "BeIR-touche2020-subset-test_cosine_recall@10": 0.1349240729674531,
+    "BeIR-touche2020-subset-test_cosine_ndcg@10": 0.653970278908209,
+    "BeIR-touche2020-subset-test_cosine_mrr@10": 0.8714285714285713,
+    "BeIR-touche2020-subset-test_cosine_map@100": 0.2786997022824942
   }
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36dcf62eb92c0c41872ffab4c463dc7156e00334f89bfcd04e10b9ff330ab39a
-size 133462128
+oid sha256:2a98f1622a4a42534215d1502cad7fd6c4bcf256ac3afec2bc8bdf6396352a14
+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 sha256:020e143387141384674be22a242949d40219eb05712011f5776e1b9cb11ebfaf
+oid sha256:d4c7ac1d7a032af33aae76aedacd113fc0586e4cb0d439c6847ffbd738908289
 size 6161