Nemotron-Personas-Embedding
Collection
Embeddings for nvidia/nemotron-personas computed with Qwen/Qwen3-Embedding-0.6B • 8 items • Updated
uuid stringlengths 32 32 | embedding listlengths 1.02k 1.02k |
|---|---|
4d70749dd9984bff9d166fe96fcdc81f | [
0.008365239948034286,
-0.03408639505505562,
-0.010866993106901646,
0.016105040907859802,
0.048158761113882065,
-0.037995386868715286,
0.04628244787454605,
0.009381577372550964,
-0.009537937119603157,
0.04440613090991974,
-0.014228724874556065,
0.004788513295352459,
0.0819324404001236,
-0.0... |
e96012323cb84c99b30f882cf53ad8cd | [
-0.01405076589435339,
-0.0660386011004448,
-0.01255787257105112,
0.02195432223379612,
0.05866194888949394,
-0.09765282273292542,
-0.012030968442559242,
-0.0033151027746498585,
0.017036553472280502,
0.04443554952740669,
-0.05409545078873634,
-0.016246197745203972,
0.05269037187099457,
-0.01... |
73b0c50a425b43708ea4dbd656ae973b | [
0.0019463332137092948,
-0.027508176863193512,
-0.010985969565808773,
0.06816491484642029,
0.04013771563768387,
-0.01721423678100109,
0.040656737983226776,
-0.0019355203257873654,
0.029238250106573105,
0.04913410171866417,
-0.07681528478860855,
0.019895851612091064,
0.016695214435458183,
-0... |
2463878a7f3f484a824c8d6a9a0bed79 | [
-0.009339964017271996,
0.025989467278122902,
-0.009786657989025116,
0.01210134569555521,
0.07471971213817596,
-0.06042550876736641,
0.0075125801376998425,
0.005644587334245443,
0.0028629021253436804,
0.045806434005498886,
-0.0594509057700634,
0.03378630802035332,
0.06075037643313408,
-0.01... |
4ad8f03629ef4d548efa4deb57877c5c | [
0.001088925520889461,
-0.07001689076423645,
-0.011479513719677925,
0.055362194776535034,
0.09053347259759903,
-0.05243125557899475,
-0.010746778920292854,
-0.016690073534846306,
-0.003134477185085416,
0.07001689076423645,
-0.020842237398028374,
0.013189229182898998,
0.07587877660989761,
-0... |
d32699b61b824301b63d6f578a0e2ffc | [
0.01878008246421814,
-0.07379662245512009,
-0.010175991803407669,
0.032430801540613174,
0.044675085693597794,
-0.03391997143626213,
0.029121536761522293,
-0.03673284500837326,
-0.03441636264324188,
0.055264733731746674,
-0.06386882066726685,
0.03756016492843628,
-0.011996087618172169,
-0.0... |
b8d441dbbe1f4e0eb24ea742bb87451c | [
0.0004467524995561689,
-0.08529186248779297,
-0.010217254981398582,
0.02027297206223011,
0.05718431994318962,
-0.07430730760097504,
0.02423064410686493,
0.0017668177606537938,
-0.01898067072033882,
0.034407515078783035,
-0.02027297206223011,
0.007067271042615175,
0.04781513661146164,
-0.00... |
619879c4f6b741d7b75aba8676ee23a5 | [
0.03784604370594025,
-0.058906812220811844,
-0.01124296709895134,
0.03911285474896431,
0.025494614616036415,
-0.11274637281894684,
0.01694362610578537,
-0.00863016489893198,
-0.0036024998407810926,
0.05795670300722122,
-0.011163790710270405,
0.04497186839580536,
0.08297625929117203,
-0.009... |
2ddea89a48314db68fe9b2e2fb26ff64 | [
-0.012095178477466106,
-0.04990211874246597,
-0.0057813432067632675,
0.04046940058469772,
-0.02312537282705307,
-0.01856115460395813,
0.05203208699822426,
-0.018713295459747314,
0.01757224090397358,
0.05568346381187439,
-0.03590518236160278,
0.0058574131689965725,
0.039860837161540985,
-0.... |
b6dc791d551c4b6e93a7c4d1a78ccb77 | [-0.023626191541552544,0.026203593239188194,-0.007016262970864773,0.04839789494872093,0.028208240866(...TRUNCATED) |
End of preview. Expand in Data Studio
Nemotron-Personas-Vietnam-Qwen3-0.6B-embedding
Embeddings for nvidia/Nemotron-Personas-Vietnam computed with Qwen/Qwen3-Embedding-0.6B.
Details
- Source dataset:
nvidia/Nemotron-Personas-Vietnam - Source revision:
e39e5096428256b2edfa2f73351f7bab02d70424 - Source split:
train - Embedding model:
Qwen/Qwen3-Embedding-0.6B - Embedding dimension: 1024
- Number of rows: 100000
- Columns embedded:
professional_persona,sports_persona,arts_persona,travel_persona,culinary_persona,persona,cultural_background,skills_and_expertise,hobbies_and_interests,career_goals_and_ambitions,sex,age,marital_status,education_level,occupation,zone,region,country - Country: Vietnam
- Prompt format:
name_in_persona_only_basic_first— structured fields first (key: value, joined by\n), then narrative columns. The name opener is dropped from every narrative except the canonicalpersonacolumn, so the name appears once, naturally, in the most authoritative summary.
Schema
| Column | Type | Description |
|---|---|---|
uuid |
string | Primary key — joins back to the source dataset |
embedding |
list | 1024-dim embedding vector |
Loading
from datasets import load_dataset
ds = load_dataset("tantara/Nemotron-Personas-Vietnam-Qwen3-0.6B-embedding", split="train")
print(ds[0]["uuid"], ds[0]["embedding"][:5])
Validation
All 100,000 vectors were checked for shard completeness, list<float32>
schema, 1024 dimensions, finite values, unit norm, UUID uniqueness, and exact
UUID order against the pinned source split. Three full-corpus exact retrieval
queries used by the Microworld regional map achieved Hit@5 = 1.00 and mean
target precision@5 = 0.933. See upload-validation.json and
retrieval-validation.json for checksums and query-level results.
- Downloads last month
- 34