File size: 1,331 Bytes
b9412ae
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f881d05
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b9412ae
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
---
language:
  - en
pretty_name: "IMDb 50K — Qwen3 Embeddings (1024D)"
task_categories:
  - text-classification
  - feature-extraction
tags:
  - text
  - sentiment-analysis
  - imdb
  - movie-reviews
  - embeddings
  - qwen3
  - parquet
  - datasets
size_categories:
  - 10K<n<100K
---
# IMDb 50K — Qwen3 Embeddings

This dataset contains **50,000 IMDb movie reviews** with precomputed **1,024-dimensional embeddings** generated using [Qwen3-Embedding-0.6B](https://huggingface.co/Qwen/Qwen3-Embedding-0.6B).

The dataset is shuffled and stored in Parquet format, with the original review text, embeddings, and binary sentiment labels preserved.

## Dataset Structure

| Column | Description |
|---|---|
| `text` | Original IMDb review text |
| `embedding` | 1,024-dimensional embedding vector |
| `label` | Sentiment label (`0` = negative, `1` = positive) |

## Purpose

The dataset provides ready-to-use embeddings for sentiment classification, clustering, dimensionality reduction, similarity analysis, and other machine learning experiments without needing to run the embedding model.

## Attribution

**Original dataset:** [IMDb Large Movie Review Dataset](https://ai.stanford.edu/~amaas/data/sentiment/) (Maas et al., 2011)

**Embedding model:** [Qwen3-Embedding-0.6B](https://huggingface.co/Qwen/Qwen3-Embedding-0.6B)