datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
starcoderdata
StarCoder Training Dataset
Dataset description
This is the dataset used for training StarCoder and StarCoderBase. It contains 783GB of code in 86 programming languages, and includes 54GB GitHub Issues + 13GB Jupyter notebooks in scripts and text-code pairs,
and 32GB of GitHub commits, which is approximately 250 Billion tokens.
Dataset creation
The creation and filtering of The Stack is explained in the original dataset, we additionally decontaminate and… See the full description on the dataset page: https://huggingface.co/datasets/bigcode/starcoderdata.Asclepius-Synthetic-Clinical-Notes
Asclepius: Synthetic Clincal Notes & Instruction Dataset
Dataset Summary
This dataset is official dataset for Asclepius (arxiv)
This dataset is composed with Clinical Note - Question - Answer format to build a clinical LLMs.
We first synthesized synthetic notes from PMC-Patients case reports with GPT-3.5
Then, we generate instruction-answer pairs for 157k synthetic discharge summaries
Supported Tasks
This dataset covers below 8 tasks
Named Entity… See the full description on the dataset page: https://huggingface.co/datasets/starmpcc/Asclepius-Synthetic-Clinical-Notes.SWE-Star
SWE-Star
Introduction
SWE-Star is a family of language models based on the Qwen2.5-Coder family and trained on the SWE-Star dataset. The dataset contains approximately 250k agentic coding trajectories distilled from Devstral-2-Small using SWE-Smith tasks.
The complete data generation, training, and evaluation pipeline is openly available in our GitHub repository, enabling anyone to reproduce our results.
Additional details are available in our blog posts.… See the full description on the dataset page: https://huggingface.co/datasets/LogicStar/SWE-Star.STAR
For detailed information, please see https://linlin-dev.github.io/project/STAR.html
How to Extract Split Archive Files on Ubuntu
To extract a split archive (e.g., STAR.7z.001, STAR.7z.002, etc.) on Ubuntu, follow these steps:
Step 1: Install p7zip
If p7zip is not already installed, use the following commands to install it:
sudo apt update
sudo apt install p7zip-full
Step 2: Verify Split Files Are in the Same Directory
Ensure all split files… See the full description on the dataset page: https://huggingface.co/datasets/Zhuzi24/STAR.SFT-COIG-CQIA-handbook
SFT-COIG-CQIA-handbook
The rearranged dataset for direct use in alignment-handbook.
数据完全来自于COIG-CQIA。
暂时忽略了chinese_traditional，coig_pc，exam，finance这些转换麻烦或者语义上不适合当QA数据集的subset。
其中train是全集，test是ruozhiba，以便代码能够跑通。
@misc{bai2024coig,
title={COIG-CQIA: Quality is All You Need for Chinese Instruction Fine-tuning},
author={Bai, Yuelin and Du, Xinrun and Liang, Yiming and Jin, Yonggang and Liu, Ziqiang and Zhou, Junting and Zheng, Tianyu and Zhang, Xincheng and Ma, Nuo and Wang… See the full description on the dataset page: https://huggingface.co/datasets/Starlento/SFT-COIG-CQIA-handbook.star-dataset-instructions
STAR Instructions
STAR Instructions is a large-scale Arabic instruction-tuning dataset built by rendering the 355 STAR Jinja2 prompt templates against their 87 source datasets, covering 27 raw task labels (20 tasks after merging closely related categories, as reported in the paper). The underlying templates were authored collaboratively using PromptLab. This dataset and the experiments built on it are described in STAR: instruction tuning for Arabic across tasks, datasets, and… See the full description on the dataset page: https://huggingface.co/datasets/KFUPM-JRCAI/star-dataset-instructions.TOMATO-Star
TOMATO-Star: Decomposed Scientific Papers for Hypothesis Generation
TOMATO-Star is a large-scale dataset of 108,717 decomposed biomedical papers from PubMed, designed for training and evaluating scientific hypothesis generation systems.
Each paper is decomposed into structured components: research question, background survey, cross-paper inspirations (with matched source papers), and fine-grained hypothesis with per-inspiration breakdowns.
Dataset Statistics… See the full description on the dataset page: https://huggingface.co/datasets/ZonglinY/TOMATO-Star.starcoder-python5b5b gpt2 tokens
tracki-startups
Tracki — Synthetic Startups Dataset
This dataset was created for the Tracki final project (RUNI — Intro to Data Science).
Tracki is an AI-based competitive-analysis prototype for founders, analysts, and VCs. When a
user describes a startup idea or company, the system returns the 3 most similar startups
in the market (via embeddings) and generates 1 new competitor-style profile.
The dataset is fully synthetic — it contains no real companies, no real people, and no
private or… See the full description on the dataset page: https://huggingface.co/datasets/tracki/tracki-startups.startrektng-mega-narrative-kg
Star Trek: The Next Generation - Narrative Knowledge Graph
A rich narrative knowledge graph extracted from Star Trek: The Next Generation screenplays using the
Fabula pipeline. Contains characters,
locations, objects, organizations, events, themes, and conflict arcs with full
participation semantics and Graph Gravity importance tiers.
Dataset Overview
Metric
Value
Source database
startrektng.mega
Type
Megagraph (cross-season merged)
Episodes
177… See the full description on the dataset page: https://huggingface.co/datasets/brandburner/startrektng-mega-narrative-kg.starter
Dataset Card for [Dataset Name]
Dataset Summary
[More Information Needed]
Supported Tasks and Leaderboards
[More Information Needed]
Languages
[More Information Needed]
Dataset Structure
Data Instances
[More Information Needed]
Data Fields
[More Information Needed]
Data Splits
[More Information Needed]
Dataset Creation
Curation Rationale
[More Information Needed]
Source Data… See the full description on the dataset page: https://huggingface.co/datasets/Langame/starter.startrektng-s01-narrative-kg
Star Trek: The Next Generation - Narrative Knowledge Graph
A rich narrative knowledge graph extracted from Star Trek: The Next Generation screenplays using the
Fabula pipeline. Contains characters,
locations, objects, organizations, events, themes, and conflict arcs with full
participation semantics and Graph Gravity importance tiers.
Dataset Overview
Metric
Value
Source database
startrektng.s01
Type
Season database
Episodes
26
Total nodes
7,791… See the full description on the dataset page: https://huggingface.co/datasets/brandburner/startrektng-s01-narrative-kg.nine_star_ki-sft
Nine Star Ki (Japanese Kyusei) SFT Dataset
Birth-year Main Star (1-9) via the standard digit-reduction formula, using the same Li-Chun-adjusted year boundary as BaZi.
Total examples
100000
Train / Validation / Test
89876 / 5124 / 5000
Question types
2
Avg citations per example
1.0
Zero-citation examples
0
Computation
Deterministic arithmetic (digit-sum reduction of the solar-year-adjusted birth year).
Validation status — read… See the full description on the dataset page: https://huggingface.co/datasets/AmareshHebbar/nine_star_ki-sft.ParaMAWPS
Math Word Problem Solving by Generating Linguistic Variants of Problem Statements
This repository contains the code, data, and models of the paper titled "Math Word Problem Solving by Generating Linguistic Variants of Problem Statements" published in the Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop).
The work is outlined in a more detailed and expository manner in our Bachelor of Science (B.Sc.) thesis… See the full description on the dataset page: https://huggingface.co/datasets/Starscream-11813/ParaMAWPS.startup-interviewsstartrektng-s02-narrative-kg
Star Trek: The Next Generation - Narrative Knowledge Graph
A rich narrative knowledge graph extracted from Star Trek: The Next Generation screenplays using the
Fabula pipeline. Contains characters,
locations, objects, organizations, events, themes, and conflict arcs with full
participation semantics and Graph Gravity importance tiers.
Dataset Overview
Metric
Value
Source database
startrektng.s02
Type
Season database
Episodes
22
Total nodes
6,918… See the full description on the dataset page: https://huggingface.co/datasets/brandburner/startrektng-s02-narrative-kg.task162_count_words_starting_with_letter
Dataset Card for Natural Instructions (https://github.com/allenai/natural-instructions) Task: task162_count_words_starting_with_letter
Additional Information
Citation Information
The following paper introduces the corpus in detail. If you use the corpus in published work, please cite it:
@misc{wang2022supernaturalinstructionsgeneralizationdeclarativeinstructions,
title={Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP… See the full description on the dataset page: https://huggingface.co/datasets/Lots-of-LoRAs/task162_count_words_starting_with_letter.starcoder-curated
StarCoderData Curated
A curated subset of StarCoderData
optimised for training a 500M parameter model focused on structured data output
(JSON generation, function calling, schema compliance).
Dataset Summary
Total code files: 5,203,508
Total tokens: 3.9B (target: 3.5B)
Classifier-scored files: 1,553,596 (1.7B tokens)
Non-classified files: 3,649,912 (2.2B tokens) — filtered by heuristics, not the classifier
Source: bigcode/starcoderdata
Classifier:… See the full description on the dataset page: https://huggingface.co/datasets/mdonigian/starcoder-curated.star-wars-dataset
Star Wars Wookieepedia Corpus
Plain-text article dump from Wookieepedia, the Star Wars wiki hosted on Fandom. Built as a domain corpus for the Vader language model, released separately since no comparably-sized Star Wars text corpus existed publicly at time of scraping.
Dataset structure
Each row:
field
description
title
article title
text
plain-text article content, wiki markup stripped
url
canonical Wookieepedia URL for the article
source
always… See the full description on the dataset page: https://huggingface.co/datasets/thehackersbrain/star-wars-dataset.star-dataset-templates
STAR Templates
STAR Templates is a curated collection of 355 Jinja2 instruction templates for Arabic NLP tasks, spanning 27 tasks across 87 source datasets, contributed by 7 prompters. The templates were authored collaboratively on PromptLab. This dataset and the experiments built on it are described in STAR: instruction tuning for Arabic across tasks, datasets, and models.
For these templates rendered against datasets samples, see the companion dataset: STAR Instructions.
📦… See the full description on the dataset page: https://huggingface.co/datasets/KFUPM-JRCAI/star-dataset-templates.star-wars-dataset
Star Wars dialogue dataset - annotation layers (snapshot 2026-09-21)
One master cue table serving diarization, ASR and dialogue-LLM training: every subtitle cue part of
15 titles with a forced-aligned time span, a character label and provenance. No audio or video is
included - the source media is copyrighted. Clip paths in the manifests resolve once you rebuild the clips
from your own copies with the pipeline code (export_asr.py, export_diarization.py).
The text layers contain… See the full description on the dataset page: https://huggingface.co/datasets/IMJONEZZ/star-wars-dataset.startrektng-s07-narrative-kg
Star Trek: The Next Generation - Narrative Knowledge Graph
A rich narrative knowledge graph extracted from Star Trek: The Next Generation screenplays using the
Fabula pipeline. Contains characters,
locations, objects, organizations, events, themes, and conflict arcs with full
participation semantics and Graph Gravity importance tiers.
Dataset Overview
Metric
Value
Source database
startrektng.s07
Type
Season database
Episodes
25
Total nodes
7,305… See the full description on the dataset page: https://huggingface.co/datasets/brandburner/startrektng-s07-narrative-kg.startrektng-s03-narrative-kg
Star Trek: The Next Generation - Narrative Knowledge Graph
A rich narrative knowledge graph extracted from Star Trek: The Next Generation screenplays using the
Fabula pipeline. Contains characters,
locations, objects, organizations, events, themes, and conflict arcs with full
participation semantics and Graph Gravity importance tiers.
Dataset Overview
Metric
Value
Source database
startrektng.s03
Type
Season database
Episodes
26
Total nodes
7,998… See the full description on the dataset page: https://huggingface.co/datasets/brandburner/startrektng-s03-narrative-kg.medical-device-regulatory-graft-rag-1350
🏥 Medical Device Regulatory & Clinical Compliance RAG Dataset (1,350 Samples)
This dataset contains 1,350 highly curated, 100% LLM-synthesized RAG samples for training Small Language Models (SLMs: 1B–4B parameters) in high-stakes Medical Device Regulatory & Quality Compliance.
Methodological Foundation:
Pioneer / Prometheus Closed-Loop Curriculum Synthesis: Multi-slice curriculum covering 5 core operational failure modes.
Elsevier Computer Standards & Interfaces… See the full description on the dataset page: https://huggingface.co/datasets/StarsMakeGalaxy/medical-device-regulatory-graft-rag-1350.startrektng-s05-narrative-kg
Star Trek: The Next Generation - Narrative Knowledge Graph
A rich narrative knowledge graph extracted from Star Trek: The Next Generation screenplays using the
Fabula pipeline. Contains characters,
locations, objects, organizations, events, themes, and conflict arcs with full
participation semantics and Graph Gravity importance tiers.
Dataset Overview
Metric
Value
Source database
startrektng.s05
Type
Season database
Episodes
26
Total nodes
7,597… See the full description on the dataset page: https://huggingface.co/datasets/brandburner/startrektng-s05-narrative-kg.Dense-Information-Science-Physics-Dataset
Dense Information With Multiple Fine-tuned Variations
This dataaset has multiple for each input to learn how to express the same answer in different ways
Dataset Structure
The dataset contains two columns:
Column
Description
input
A science or quantum-physics question
output
A conversational answer to the question
Example:
{
"input": "What is quantum entanglement?",
"output": "Quantum entanglement is when two quantum systems share one… See the full description on the dataset page: https://huggingface.co/datasets/StarpowerTechnology/Dense-Information-Science-Physics-Dataset.DPO-En-Zh-20k-handbook
DPO-En-Zh-20k-handbook
The rearranged dataset for direct use in alignment-handbook.
Data from DPO-En-Zh-20k.
The original 10k + 10k are split to 9900 + 9900 for train and 100 + 100 for test.
startrektng-s04-narrative-kg
Star Trek: The Next Generation - Narrative Knowledge Graph
A rich narrative knowledge graph extracted from Star Trek: The Next Generation screenplays using the
Fabula pipeline. Contains characters,
locations, objects, organizations, events, themes, and conflict arcs with full
participation semantics and Graph Gravity importance tiers.
Dataset Overview
Metric
Value
Source database
startrektng.s04
Type
Season database
Episodes
26
Total nodes
7,840… See the full description on the dataset page: https://huggingface.co/datasets/brandburner/startrektng-s04-narrative-kg.TOMATO-Star-SFT-Data-R1D-32B
TOMATO-Star SFT Data (R1D-32B)
SFT training data for the two core tasks in MOOSE-Star: Hypothesis Composition (HC) and Inspiration Retrieval (IR).
All data is generated via rejection sampling with DeepSeek-R1-Distill-Qwen-32B as the teacher model, followed by reranker filtering.
All data is in ShareGPT JSONL format, directly compatible with LLaMA-Factory.
Files
Hypothesis Composition (HC)
File
Samples
Description
HC/normal_composition.jsonl… See the full description on the dataset page: https://huggingface.co/datasets/ZonglinY/TOMATO-Star-SFT-Data-R1D-32B.star-trek-tng-narrative-kg-season-1
This dataset has moved. The canonical version is now at
brandburner/startrektng-s01-narrative-kg with updated data (schema v1.1.0, Graph Gravity tiers, all 7 seasons available).
This repository is preserved for backward compatibility but will no longer be updated.
Star Trek: The Next Generation - Narrative Knowledge Graph (Season 1)
This is a legacy export. Please use the new repository:
Season 1: brandburner/startrektng-s01-narrative-kg
All seasons + megagraph:… See the full description on the dataset page: https://huggingface.co/datasets/brandburner/star-trek-tng-narrative-kg-season-1.