datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Scraped-Data
Scraped-Data — Building to 1 TB
A continuously-growing, cleaned text corpus scraped, parsed, cleaned,
losslessly-zipped, and auto-pushed from a CPU-only Google Colab pipeline.
Current repo size: ~26 GB → target: 1 TB (incremental batch uploads).
Current files
File
Size
Contents
corpus_batch1.clean.zip
4.80 GB
~2.9M Wikipedia docs (stored, from first scrape)
corpus_batch2.clean.zip
4.85 GB
~2.5M docs
corpus_batch3.clean.zip
4.91 GB
~2.3M docs… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/Scraped-Data.olympiads
AI-MO Olympiad Reference Dataset
This dataset contains a structured collection of Olympiad problems and their solutions,
organized by competition. Contains high quality data, prioritizing "official" solutions to problems.
Structure
<competition name>/ # Problems and solutions from the International Mathematical Olympiad
├── raw/ # Raw problem/solution statements (.pdf)
│ ├── file1.pdf
│ ├── file2.pdf
├── download_script/ # the scripts used… See the full description on the dataset page: https://huggingface.co/datasets/gugett/olympiads.Math-Expanded
Massive Step-by-Step Mathematics Instruction Dataset
Dataset Description
This is a 60GB, highly knowledge-dense dataset designed to teach Large Language Models (LLMs) rigorous mathematical reasoning.
Unlike standard math datasets that only provide the final answer, this dataset emphasizes Chain-of-Thought (CoT) reasoning. Every single row contains a detailed, step-by-step breakdown of how to arrive at the solution, making it ideal for supervised fine-tuning (SFT)… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/Math-Expanded.Science
Science Learning Curriculum
A large collection of structured learning units covering the scientific method, physics, chemistry, biology, earth and space science, and supporting mathematics.
Content is organised by domain, topic, difficulty level (1–10), and unit type (explanation, worked example, practice problem, etc.). Each row contains a self-contained pedagogical text together with key equations, prerequisites, and an explicit learning objective.
Columns
id… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/Science.gugugaga-keypoints-v1
gugugaga-keypoints-v1
Synthetic 18-point 2D keypoint dataset rendered from a chibi character (gugugaga) for pose / keypoint model training.
Summary
Item
Value
Samples
450
Poses
25 anchors
Cameras
front, back, side_r, side_l, top, bottom
Pitch
eye, high, low
Image size
720 × 1280
Schema
gugugaga_kp18_v1 (18 keypoints)
Grid: 25 pose × 6 camera × 3 pitch = 450.
Layout
images/{pose_id}/{stem}_rgb.png… See the full description on the dataset page: https://huggingface.co/datasets/todo1111/gugugaga-keypoints-v1.imageEnglish-Mini
LLM-English-100MB — Compact & Dense English Teaching Corpus
A 100MB, extremely clean CSV designed to teach an LLM English from scratch via instruction-tuning. No noise, no HTML, no duplicates — just pure grammar, vocabulary, and syntax transformations.
Generated with a single paste-and-run Python script in Google Colab.
Why this teaches English
Instead of raw text, the dataset is instruction -> input -> output pairs that force the model to learn rules:
Grammar… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/English-Mini.Te-Reo-Maori
Te Reo Māori Multi-Format Training Dataset
Dataset Description
A 3GB multi-format training dataset for te reo Māori language models, containing approximately 1.5–3 million unique sentences generated using rule-based grammar with authentic Māori vocabulary.
⚠️ Important: This dataset is synthetically generated. It contains programmatically constructed Māori sentences using real vocabulary and grammatical patterns, not natural human-written text. See Limitations… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/Te-Reo-Maori.LOOM
LOOM: Language-Only Operational Microworlds
LOOM is a synthetic natural-language reasoning dataset designed to teach language models the deep structures behind code and math without exposing source code, formal equations, or symbolic programming syntax.
Instead of showing code or math notation, LOOM trains models on ordinary-language microworlds where the hidden logic is algorithmic: state changes, causal chains, conditionals, invariants, iteration, and reverse reasoning.
The… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/LOOM.Gugager
🌟 DESTEK & TOPLULUK ÇAĞRISI (SUPPORT & LIKE):Açık kaynak ve ücretsiz olarak sunduğum bu devasa çalışmayı faydalı bulduysanız, projenin sürdürülebilirliğine ve açık kaynak ekosisteminin görünürlüğüne katkı sağlamak için lütfen sayfanın sağ üstündeki Like (❤️ Beğeni) butonuna basarak destek olmayı unutmayın!(If you find this open-source dataset valuable for your research or models, please consider leaving a ❤️ Like at the top-right to support future updates and maintenance).
🇹🇷… See the full description on the dataset page: https://huggingface.co/datasets/puschinka/Gugager.CANADA_ACT_REGULATION_QA
Canadian Acts and Regulation QA
source- https://laws-lois.justice.gc.ca/eng/XML/Legis.xml
model_name="gemini-1.5-flash-latest" with 1 million context length,
First summarize the text scrapped text from xml tree of urls using gemini.
then generate QA from sumarised text.
Performance of Gemini was way way better than GPT-4.
Fitering was done based on Heuristics after rigrous analysis because llms were not always accurate.
summary_prompt_template= """
You'r legal expert… See the full description on the dataset page: https://huggingface.co/datasets/Guggu/CANADA_ACT_REGULATION_QA.Pattern-Recognition
Pattern Completion Dataset
A 30 GB synthetic dataset of numeric sequence‑completion prompts and their next values, designed to teach large language models how to recognize and extrapolate patterns.
Each row contains a prompt (the sequence with a ? indicating the missing next element) and a completion (the correct next number).
Dataset Structure
Format: CSV (no header row)
Columns:
prompt – "Find the next number in the sequence: a,b,c,... ,?"
completion – the… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/Pattern-Recognition.Coding-Corpus-Bench
Coding-Corpus-Bench
A benchmark dataset for evaluating language-semantics reasoning across systems programming and low-level programming languages.
Overview
Coding-Corpus-Bench contains 100 curated programming-language questions designed to test whether a model can reason precisely about language semantics rather than rely on superficial pattern matching or observed behavior.
The benchmark covers:
Rust
Go
C
C++
Zig
V
CUDA
Questions focus on subtle semantic rules… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/Coding-Corpus-Bench.Math-Dataset
Math Teaching Dataset - 100MB / 735k Rows
A large math dataset designed for training and fine-tuning AI models with step-by-step reasoning.
Files
math_dataset_100MB.csv - Full dataset, 100.33 MB, 735,000 rows
math_dataset_100MB.csv.gz - Compressed version, 18.95 MB
splits/math_part_1.csv to math_part_10.csv - 10 x ~10MB chunks
Dataset Stats
Total rows: 735,000
Size: 100.33 MB raw, 18.95 MB gzipped
Format: CSV, UTF-8, with header… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/Math-Dataset.OpenOrca-gugugo-ko
OpenOrca 한국어 번역 데이터셋
Gugugo-koen-7B-V1.1을 이용하여 OpenOrca데이터셋을 번역하고 있습니다.
번역 진행상황은 아래를 참고해 주십시오.
진행상황
GPT4 생성물 약 100만 개 중 약 64만 개 번역완료
GPT3.5 생성물 약 350만 개 중 약 159만 개 번역완료
데이터셋 사용 후 출처표기는 제작자에게 큰 힘이 됩니다.
Original dataset card: OpenOrca
🐋 The OpenOrca Dataset! 🐋
We are thrilled to announce the release of the OpenOrca dataset!
This rich collection of augmented FLAN data aligns, as best as possible, with the distributions outlined in the Orca paper.
It has… See the full description on the dataset page: https://huggingface.co/datasets/squarelike/OpenOrca-gugugo-ko.notas-fiscaisDataset com imagens de comprovantes de pagamento e notas ficais no Brasil.KOR-gugugu-platypus-set
Dataset Card for "KOR-gugugu-platypus-set"
Pre-processing
# Make the special text lists, manually.
[\n\t-=+,#/\＄?:^$.@*\"–∼①②③④⑤ⓐⓑⓒ㉮㉯㉰㈜®...TL;DR...∂Σ∩∅φμσℝλΛ≥℃∉⊂θ±€Øπ√≠≤ε∈∫ωηαβ÷≈ס̊°²/]
위의 정규표현식을 이용하여, 한국어 및 영어를 제외한 다양한 외국어, 이모지, 특수 문자 등등 제거.
번역 task 최대한 제거. (~번역 task는 한국어로 번역하면 거의 100% 오류)
gugugu-ko의 GPT4 번역 결과에서 약 20k sampling.
데이터셋 이용하셔서 모델이나 데이터셋을 만드실 때, 간단한 출처 표기를 해주신다면 연구에 큰 도움이 됩니다😭😭
References
Thank you for squarelike/OpenOrca-gugugo-ko… See the full description on the dataset page: https://huggingface.co/datasets/kyujinpy/KOR-gugugu-platypus-set.Math-Mini
Clean Math Reasoning Dataset
A clean dataset for training and evaluating language models on mathematical problem solving.
The dataset contains concise mathematical question-and-answer pairs designed to improve model performance on structured numerical reasoning tasks.
Dataset Structure
The dataset contains two fields:
Column
Description
prompt
A mathematical problem or question
response
The corresponding solution
Example:
prompt:
48392+92831=?… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/Math-Mini.OpenOrca-gugugo-ko-len100
Original dataset: squarelike/OpenOrca-gugugo-ko
길이가 100 이상인 답변만 필터링
DeepMath-103K
DeepMath-103K
🔥 News
May 8, 2025: We found that 48 samples contained hints that revealed the answers. The relevant questions have now been revised to remove the leaked answers.
April 14, 2025: We release DeepMath-103K, a large-scale dataset featuring challenging, verifiable, and decontaminated math problems tailored for RL… See the full description on the dataset page: https://huggingface.co/datasets/gugett/DeepMath-103K.CoT-for-LLM
README — Advanced Chain‑of‑Thought Dataset Generator
Overview
This project generates a large-scale synthetic dataset of Chain‑of‑Thought (CoT) reasoning examples across multiple domains:
Math (algebra, word problems, multi‑step reasoning)
English (vocabulary explanations, nuance, tone)
Writing (multi‑paragraph reflections, structured planning)
Coding (advanced algorithms, data structures, real code snippets)
Science (physics, biology, chemistry, earth science… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/CoT-for-LLM.AI-PVODsynthetic-pretraining-transformers-v1
Synthetic Pre-training Transformers v1.0.0
Dataset Description
This is a synthetic pre-training dataset generated from transformer architecture patterns. It contains paraphrased, augmented, and interpolated content derived from validated seed data about neural sequence modeling and attention mechanisms.
Dataset Summary
Total Samples: 100
Total Tokens: 6,084
Average Tokens per Sample: 60.84
Format: Parquet
Version: 1.0.0
License: CC-BY-4.0
Supported… See the full description on the dataset page: https://huggingface.co/datasets/gugarosa/synthetic-pretraining-transformers-v1.English
Synthetic English Language Acquisition Dataset (3GB)
A structured, 3GB synthetic CSV dataset generated to assist in pretraining or fine-tuning Language Models (LLMs) on core English syntax, vocabulary, narrative structures, and explicit grammar rules.
Dataset Structure
The dataset contains four primary columns:
Column Name
Data Type
Description
data_type
string
Categorises the entry (sentence, narrative_chunk, or grammar_instruction).
text_content… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/English.Code-Syntax-Expanded
Code-Syntax-Expanded
A massive, high-quality synthetic dataset for training LLMs to identify and correct syntax errors across 33 programming languages. Contains 5+ million unique examples (~1.1 GB) with English explanations – no artificial padding, no duplicate rows.
📊 Dataset Overview
Property
Value
Total rows
5,000,000+
File size
~1.1 GB (uncompressed CSV)
Languages
33
Unique templates
160+ error patterns
Format
CSV (4 columns)
License… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/Code-Syntax-Expanded.Legal
Massive Legal & Law Instruction Tuning Dataset (5GB)
This dataset contains a massive, synthetically generated corpus of legal instruction-tuning data designed to teach Large Language Models (LLMs) how to understand, reason through, and respond to legal queries. The dataset scales progressively from fundamental legal concepts to highly advanced jurisprudential masterclasses.
Dataset Structure
The dataset is provided as a single CSV file… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/Legal.Pretrain
pretrain-60GB
A 60GB, knowledge-dense pretraining corpus built as a clean, superior alternative to TinyStories.
TinyStories teaches grammar. This teaches knowledge, reasoning, and code.
File: pretrain.csv - 60GB, ~15M rows, single column text
Why vs TinyStories?
TinyStories is fiction for kids. This is textbook / encyclopedia / reasoning / code. Every row is 3500-5200 chars packed with 2-3 fused concepts, not a story.
Physics, Chem, Bio, Math, CS, History, Geo… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/Pretrain.Jailbreak-Refusal
LLM Refusal Training Dataset
A large-scale dataset designed to teach LLMs how to safely refuse jailbreak attempts, prompt injections, and policy-violating requests.
Dataset Description
This dataset contains 30GB of (category, prompt, response) triplets pairing simulated adversarial prompts with safe, helpful refusals. The data is non-operational and does not contain real exploits or harmful instructions.
Columns
Column
Type
Description… See the full description on the dataset page: https://huggingface.co/datasets/Gugu8/Jailbreak-Refusal.deepfake-detection-dataset-v3
Deepfake Detection Dataset V3
This dataset contains images and detailed explanations for training and evaluating deepfake detection models. It includes original images, manipulated images, confidence scores, and comprehensive technical and non-technical explanations.
Dataset Structure
The dataset consists of:
Original images (image)
CAM visualization images (cam_image)
CAM overlay images (cam_overlay)
Comparison images (comparison_image)
Labels (label): Binary… See the full description on the dataset page: https://huggingface.co/datasets/guglothmahipal007/deepfake-detection-dataset-v3.OpenOrca-gugugo-ko-len300
Original dataset: squarelike/OpenOrca-gugugo-ko
길이가 300 이상인 답변만 필터링