Dataset Viewer
Auto-converted to Parquet Duplicate
id
string
text
string
indices
list
values
list
labels
list
s1
this is a test
[ 1037, 2003, 2023, 2122, 2671, 2832, 2883, 3231, 3267, 3399, 3800, 4106, 4638, 5604, 5852, 6210, 6845, 7551, 7667, 7709, 7718, 8522, 9312, 11360 ]
[ 1.3415570259094238, 0.6412121057510376, 1.832952857017517, 0.7313549518585205, 0.14212392270565033, 0.2665705978870392, 0.1463720053434372, 2.971982479095459, 1.3294447660446167, 0.2411385178565979, 0.3659552335739136, 0.443505197763443, 0.7388043403625488, 1.7083725929260254, 1.66885614...
[ "a", "is", "this", "these", "science", "process", "census", "test", "nature", "theory", "purpose", "analysis", "check", "testing", "tests", "definition", "lab", "experiment", "assessment", "procedure", "tested", "equation", "evaluation", "exam" ]
s2
hello world
[ 2040, 2047, 2088, 2299, 2748, 3011, 3376, 3795, 4774, 5304, 5798, 6160, 7592, 7632, 8484 ]
[ 0.42195403575897217, 0.4320189952850342, 2.7665812969207764, 0.3314591646194458, 1.3951743841171265, 0.021600818261504173, 0.4377126097679138, 0.0008450751192867756, 0.5187228918075562, 0.1712515652179718, 0.14741800725460052, 0.8142920732498169, 2.8032641410827637, 2.1904382705688477, 1...
[ "who", "new", "world", "song", "yes", "earth", "beautiful", "global", "planet", "universe", "birthday", "welcome", "hello", "hi", "worlds" ]
s3
neural search sparse retrieval
[ 2424, 2951, 3075, 3638, 3945, 4167, 5371, 5527, 6575, 7233, 7809, 9742, 9896, 12850, 13594, 13617, 15698, 15756, 16602, 20288, 23032, 23130, 24961, 26384 ]
[ 0.8253046274185181, 0.00836097914725542, 0.28663089871406555, 0.374656081199646, 1.9754879474639893, 1.4174028635025024, 0.3905353844165802, 0.8663956522941589, 1.102981448173523, 0.27791285514831543, 0.28874000906944275, 1.4478073120117188, 0.4965222477912903, 1.63233482837677, 0.027435...
[ "find", "data", "library", "memory", "search", "brain", "file", "storage", "searching", "recovery", "database", "dense", "algorithm", "retrieve", "scan", "sensor", "neurons", "neural", "browser", "sparse", "query", "neurological", "sparsely", "retrieval" ]
s4
hybrid search combines dense and sparse signals
[ 2424, 2897, 2951, 3463, 3945, 4007, 4118, 4289, 4304, 4742, 4773, 5257, 5418, 5656, 6028, 6575, 7755, 8893, 9742, 9896, 11506, 13585, 13590, 20288, 23376, 24961 ]
[ 0.8730606436729431, 0.0632539764046669, 0.300018846988678, 0.13327239453792572, 2.0855696201324463, 0.13544084131717682, 0.022488104179501534, 0.08042329549789429, 1.0396734476089478, 1.5438448190689087, 0.060134198516607285, 0.8002380132675171, 0.005709056276828051, 0.22329793870449066, ...
[ "find", "network", "data", "results", "search", "software", "method", "format", "density", "signal", "web", "combination", "pattern", "strategy", "technique", "searching", "signals", "hybrid", "dense", "algorithm", "combine", "combines", "merge", "sparse", "hybrids", ...
s5
the quick brown fox jumps over the lazy dog
[ 1012, 2010, 2058, 2208, 2427, 2829, 3435, 3899, 4030, 4248, 4419, 4743, 5252, 5376, 5573, 6077, 7904, 8843, 11679, 13971, 15267, 24623 ]
[ 0.16180261969566345, 0.1367461383342743, 1.954992651939392, 0.062477633357048035, 0.25551435351371765, 1.910771131515503, 0.2905551493167877, 1.613210916519165, 0.5523281693458557, 1.8459932804107666, 1.5396771430969238, 0.22957251965999603, 0.21801874041557312, 1.681915521621704, 0.0887...
[ ".", "his", "over", "game", "species", "brown", "fast", "dog", "slow", "quick", "fox", "bird", "chase", "jump", "sudden", "dogs", "employee", "breed", "leap", "lazy", "predator", "foxes" ]
s6
error handling and retries in distributed systems
[ 1012, 2128, 2291, 2491, 2897, 2968, 3001, 3274, 3949, 4007, 4353, 4434, 4638, 5047, 5500, 7561, 7709, 8304, 9896, 10697, 11129, 11371, 19701, 21011, 28213 ]
[ 0.22281867265701294, 1.4137070178985596, 0.7898812890052795, 0.43388351798057556, 0.12010707706212997, 0.13842079043388367, 0.737886369228363, 0.49059799313545227, 0.01761729083955288, 0.30293169617652893, 1.524886131286621, 0.38040995597839355, 0.08194450289011002, 1.2075119018554688, 1...
[ ".", "re", "system", "control", "network", "management", "systems", "computer", "treatment", "software", "distribution", "connection", "check", "handle", "distributed", "error", "procedure", "handling", "algorithm", "errors", "##try", "violation", "null", "##tries", "...
s7
golang benchmarking with realistic workloads
[ 2007, 2021, 2075, 2147, 2175, 2613, 2640, 2928, 2974, 3105, 3115, 4007, 4118, 4275, 4294, 5656, 6028, 6847, 6994, 7170, 7812, 7904, 10665, 11066, 12689, 15650, 25023 ]
[ 0.7397245764732361, 0.46371039748191833, 0.21375474333763123, 1.3043800592422485, 1.9297640323638916, 0.8554390668869019, 0.03554181382060051, 0.30674174427986145, 0.3046981692314148, 0.3170430064201355, 0.1787375956773758, 0.3234851062297821, 0.21392472088336945, 0.4222850501537323, 0.2...
[ "with", "but", "##ing", "work", "go", "real", "design", "mark", "technology", "job", "standard", "software", "method", "models", "architecture", "strategy", "technique", "bench", "tool", "load", "ideal", "employee", "##mark", "##load", "realistic", "realism", "##l...
s8
database indexing for low latency retrieval
[ 2000, 2051, 2075, 2397, 2424, 2592, 2659, 2896, 2946, 2951, 3556, 3945, 4638, 5527, 5950, 7809, 9407, 9896, 12850, 13617, 14676, 14767, 16902, 18422, 23032, 26384, 29296, 29299 ]
[ 0.187372624874115, 0.0011799524072557688, 0.5673160552978516, 1.8802762031555176, 0.23095710575580597, 0.465910404920578, 2.09625244140625, 1.1799359321594238, 0.10192098468542099, 0.7182085514068604, 0.2792903780937195, 0.2709466516971588, 0.217810720205307, 0.6031616926193237, 1.784194...
[ "to", "time", "##ing", "late", "find", "information", "low", "lower", "size", "data", "score", "search", "check", "storage", "index", "database", "##ncy", "algorithm", "retrieve", "sensor", "extraction", "##ncies", "correlation", "calculate", "query", "retrieval", ...
s9
vector search and inverted indexes
[ 2063, 2229, 2424, 3231, 3945, 4289, 5418, 5950, 6575, 8522, 9207, 9896, 18422, 19019, 19262, 20037, 23032, 28527, 29299 ]
[ 0.023961249738931656, 0.7660014629364014, 0.501924455165863, 0.07549704611301422, 1.9247660636901855, 0.021934648975729942, 0.09426677972078323, 2.129295825958252, 0.9555753469467163, 0.2542722821235657, 2.2845423221588135, 0.3529493808746338, 0.33943840861320496, 1.5217386484146118, 0.1...
[ "##e", "##es", "find", "test", "search", "format", "pattern", "index", "searching", "equation", "vector", "algorithm", "calculate", "vectors", "inverse", "inverted", "query", "inversion", "indices" ]
s10
what is the best way to test model regressions
[ 2000, 2015, 2113, 2126, 2190, 2204, 2944, 2951, 3231, 3399, 3971, 4118, 4275, 4638, 4725, 5604, 6028, 6845, 7709, 7812, 8522, 9896, 16902, 18422, 19701, 24497, 26237 ]
[ 0.5751252174377441, 0.16650837659835815, 0.4756937325000763, 1.7824887037277222, 1.6418826580047607, 1.303572654724121, 1.9687767028808594, 0.10523802787065506, 2.182739019393921, 0.4712735414505005, 1.6081128120422363, 0.7808219790458679, 1.5097099542617798, 1.1895514726638794, 0.177918...
[ "to", "##s", "know", "way", "best", "good", "model", "data", "test", "theory", "ways", "method", "models", "check", "methods", "testing", "technique", "lab", "procedure", "ideal", "equation", "algorithm", "correlation", "calculate", "null", "rehab", "regression" ]
s11
مرحبا بالعالم
[ 1006, 1010, 1011, 1045, 1242, 1270, 1271, 1276, 1280, 1295, 1329, 2002, 2016, 2040, 2223, 2232, 2488, 3038, 3114, 3290, 3574, 3577, 4489, 4676, 5152, 5424, 5499, 5619, 5640, 7025, 8174, 8956, 9264, 9835, 12577, 12619, 13085, 14157, 14498, ...
[ 0.12251433730125427, 0.6330717206001282, 0.2808048129081726, 0.01638857088983059, 0.48977670073509216, 0.6520628333091736, 1.2817822694778442, 0.3237472176551819, 0.052574146538972855, 1.084732174873352, 0.1100323349237442, 0.1472455859184265, 0.03218533471226692, 0.35546231269836426, 0....
[ "(", ",", "-", "i", "ב", "ا", "ب", "ح", "ر", "م", "ब", "he", "she", "who", "league", "##h", "better", "saying", "reason", "mexico", "meaning", "spain", "difference", "religion", "muslim", "arab", "islamic", "argentina", "arabic", "islam", "saudi", "iraqi...
s12
こんにちは 世界
[ 100, 1006, 1010, 1011, 1025, 1029, 1031, 1655, 1745, 2133, 2138, 2750, 2887, 2900, 3574, 9152, 11895, 12849, 15498, 30188, 30194, 30198, 30217 ]
[ 0.9980085492134094, 0.27737894654273987, 0.5511769652366638, 0.22968529164791107, 0.4605620205402374, 0.10625826567411423, 0.34949788451194763, 1.199540615081787, 0.7610847353935242, 0.039447035640478134, 0.0025392689276486635, 0.09207102656364441, 0.21069733798503876, 0.8324141502380371, ...
[ "[UNK]", "(", ",", "-", ";", "?", "[", "こ", "世", "...", "because", "despite", "japanese", "japan", "meaning", "ni", "shi", "ko", "pronunciation", "##ち", "##に", "##は", "##ん" ]
s13
hola mundo
[ 1010, 2014, 2016, 2023, 2040, 2055, 2088, 2143, 2189, 2252, 2427, 2450, 2721, 2782, 2882, 2885, 3309, 3376, 4774, 5304, 5684, 6907, 7402, 7570, 14124, 23117, 25989 ]
[ 0.5111271142959595, 0.1110093966126442, 0.6265604496002197, 0.030902504920959473, 0.8092585802078247, 0.6220420598983765, 1.5546995401382446, 0.033277660608291626, 0.20155268907546997, 0.34479862451553345, 0.0815584808588028, 0.1309863030910492, 2.4459142684936523, 0.11839774250984192, 0...
[ ",", "her", "she", "this", "who", "about", "world", "film", "music", "club", "species", "woman", "##la", "festival", "grand", "europe", "hotel", "beautiful", "planet", "universe", "favor", "genre", "latino", "ho", "casa", "monde", "mundo" ]
s14
bonjour le monde
[ 1010, 2010, 2040, 2099, 2413, 2605, 3393, 4241, 14753, 23099, 23117 ]
[ 0.14665216207504272, 0.166755810379982, 0.22183969616889954, 0.2687278687953949, 0.0736217275261879, 0.6897643804550171, 1.6252695322036743, 0.697705090045929, 2.506682872772217, 2.4385569095611572, 2.4066455364227295 ]
[ ",", "his", "who", "##r", "french", "france", "le", "du", "bon", "##jou", "monde" ]
s15
privacy preserving telemetry and observability
[ 1998, 2188, 2473, 2562, 2797, 3036, 3200, 3343, 3567, 3800, 3860, 4047, 4072, 5056, 5080, 5851, 6633, 6958, 7969, 8043, 8347, 8553, 9394, 9585, 10093, 11109, 11129, 11598, 12014, 12962, 14293, 15224, 21382, 21999, 27885, 28805, 29361 ]
[ 0.20508810877799988, 0.3017480671405792, 0.06112310290336609, 0.3211923837661743, 1.2837936878204346, 0.3183955252170563, 0.005013869144022465, 0.20897707343101501, 0.9267050623893738, 0.3218534290790558, 0.28158071637153625, 0.5023488402366638, 0.16668297350406647, 0.22905577719211578, ...
[ "and", "home", "council", "keep", "private", "security", "property", "policy", "##va", "purpose", "protection", "protect", "necessary", "advantage", "device", "secure", "##em", "principle", "preserve", "##ser", "preservation", "##bility", "privacy", "enable", "tel", ...
s16
sparse expansion helps lexical matching
[ 1012, 2000, 2005, 2393, 2674, 3466, 3623, 4935, 5056, 5335, 5770, 6509, 7818, 9289, 9526, 9585, 9742, 9844, 10349, 10956, 11598, 16105, 17702, 20288 ]
[ 0.09612222015857697, 0.20077796280384064, 0.48990848660469055, 1.1801464557647705, 1.3576165437698364, 0.20647011697292328, 0.14525645971298218, 2.3460726737976074, 0.5072653889656067, 0.2324489802122116, 0.8946213126182556, 0.08864995092153549, 1.510765790939331, 1.1489187479019165, 0.1...
[ ".", "to", "for", "help", "match", "effect", "increase", "expansion", "advantage", "improve", "benefit", "assist", "expand", "##cal", "cure", "enable", "dense", "matching", "matched", "facilitate", "enhance", "lexi", "overlap", "sparse" ]
s17
bm25 baseline for retrieval
[ 1038, 2146, 2213, 2322, 2423, 2504, 2656, 2946, 3115, 3231, 3292, 3556, 3638, 3671, 5140, 5527, 7233, 12850, 13617, 17788, 18422, 26163, 26384 ]
[ 1.7654690742492676, 0.2217852622270584, 1.3165910243988037, 0.6297270059585571, 1.77405846118927, 0.15514078736305237, 0.11501874774694443, 1.114087700843811, 0.7517924308776855, 0.06234312057495117, 0.4309306740760803, 0.35670340061187744, 0.052969932556152344, 0.1370391994714737, 0.450...
[ "b", "long", "##m", "20", "25", "level", "26", "size", "standard", "test", "distance", "score", "memory", "normal", "retrieved", "storage", "recovery", "retrieve", "sensor", "##25", "calculate", "baseline", "retrieval" ]
s18
tokenization affects downstream ranking quality
[ 2300, 2504, 3006, 3426, 3466, 3550, 3623, 3643, 3737, 3989, 4254, 4396, 4434, 4635, 4697, 5360, 5464, 5579, 7461, 9226, 13248, 13909, 16902, 19204 ]
[ 0.9501621127128601, 0.17295169830322266, 0.24973146617412567, 0.16772307455539703, 1.1256327629089355, 0.07447744905948639, 0.6532950401306152, 0.29969170689582825, 1.5735303163528442, 1.9220876693725586, 0.21489976346492767, 0.4527033269405365, 0.418301522731781, 0.7982624173164368, 0.2...
[ "water", "level", "market", "cause", "effect", "##ized", "increase", "value", "quality", "##ization", "impact", "ranked", "connection", "rank", "##ize", "affected", "ranking", "classification", "affect", "coin", "downstream", "upstream", "correlation", "token" ]
s19
unit tests should be deterministic
[ 2022, 2025, 2204, 2235, 2323, 2342, 3131, 3197, 3231, 3671, 4106, 5604, 5852, 6845, 10074, 11233, 11360, 19701, 25300, 28283 ]
[ 0.8255115151405334, 0.058833252638578415, 0.1807674765586853, 0.07841260731220245, 1.5109301805496216, 0.3154586851596832, 2.4087026119232178, 1.9559437036514282, 2.08929705619812, 0.05134565010666847, 0.06252018362283707, 0.8885185718536377, 1.076686143875122, 1.3058867454528809, 0.8865...
[ "be", "not", "good", "small", "should", "need", "unit", "units", "test", "normal", "analysis", "testing", "tests", "lab", "##stic", "##mine", "exam", "null", "##mini", "deter" ]
s20
production incidents require fast rollback
[ 2067, 2342, 2537, 2550, 2724, 2824, 3135, 3177, 3223, 3426, 3435, 3965, 4031, 4126, 4287, 4316, 4652, 4713, 4795, 4808, 4897, 4926, 5043, 5057, 5291, 5442, 5473, 5478, 5918, 5942, 5963, 8536, 9034, 9495, 10444, 13185, 21557 ]
[ 1.0088261365890503, 1.1066099405288696, 2.373450517654419, 0.1920262575149536, 0.6586462259292603, 0.3051822781562805, 1.6271196603775024, 0.15386910736560822, 1.0343173742294312, 0.5385159850120544, 2.3101015090942383, 0.362692654132843, 0.23160752654075623, 0.6136272549629211, 0.304990...
[ "back", "need", "production", "produced", "event", "events", "producer", "speed", "required", "cause", "fast", "produce", "product", "crime", "carry", "vehicle", "prevent", "factory", "dangerous", "violence", "roll", "accident", "incident", "emergency", "rolling", "...

SPLADE Endpoint Golden Dataset (v1)

This dataset contains sparse vectors generated from a SPLADE inference service.

Source

  • Model: prithivida/Splade_PP_en_v1
  • Generated at (UTC): 2026-02-21T19:09:51.715028+00:00
  • Rows: 20
  • Digest (JSONL SHA-256): 050715eeb0a288daa486444e6bad006a98705f1c0a6e8f3780e0c46b1614105c

Notes

  • Inputs are short synthetic test prompts for regression parity.
  • Labels are included in this release for parity validation.
  • No private endpoint addresses are included in dataset artifacts.

Files

  • splade_endpoint_golden/v1/splade_pp_en_v1_endpoint_topk24_labels_v1.jsonl
  • splade_endpoint_golden/v1/metadata.json

JSONL schema

Each line contains:

  • id (string)
  • text (string)
  • indices (array[int])
  • values (array[float])
  • labels (array[string], may be empty)
Downloads last month
9