datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
harmless_alpaca_jaJapanese auto-translation of mlabonne/harmless_alpacausing llmfan46/gemma-4-31B-it-qat-q4_0-uncensored-heretic-NVFP4-GGUF
Harmful-Harmless-100Pairs-JA-HighIntensity
Harmful-Harmless-100Pairs-JA-HighIntensity
This is a small-scale dataset consisting of 100 pairs of high-intensity Harmful / Harmless contrastive data written in Japanese.
⚠️ Important Notice
This dataset intentionally contains harmful, explicit, offensive, disturbing, biased, or otherwise inappropriate content for research and evaluation purposes. Some entries may describe dangerous, illegal, abusive, or unethical activities in substantial detail.
The inclusion… See the full description on the dataset page: https://huggingface.co/datasets/OS-Software/Harmful-Harmless-100Pairs-JA-HighIntensity.software_slacksSoftware-Engineering-Dataset_90_10arxiv-softwares-2021datapoints_round1_dpsk_software_engineering_shard1_daytona_n100k1us-k12-schools-directory
US K-12 Schools Directory
A directory of 124,613 US K-12 schools covering all 50 states, DC, and US territories,
compiled from federal and state government sources. Each record carries directory
information (address, phone, website), enrollment and demographics, and, where a source
supplied it, a principal name and email.
This is a compilation of public government data. It is not a survey, and no field was
independently verified against the school itself.
Loading… See the full description on the dataset page: https://huggingface.co/datasets/robworks-software/us-k12-schools-directory.OmniGIRLThis repository contains the data presented in OmniGIRL: A Multilingual and Multimodal Benchmark for GitHub Issue Resolution. OmniGIRL is a GitHub issue resolution benchmark that is multilingual, multimodal, and multi-domain. It includes 959 task instances collected from repositories across four programming languages (Python, JavaScript, TypeScript, and Java) and eight different domains.
datapoints_round1_dpsk_software_engineering_shard2_daytona_n100k1olmocr_science_pdfs-software_developmenthttps://huggingface.co/datasets/allenai/dolma3_pool/tree/main/data/olmocr_science_pdfs-software_development
software-architecture-instructions-preferenceSweSetupBench-liteThis repository contains the data presented in SWE-Factory: Your Automated Factory for Issue Resolution Training Data and Evaluation Benchmarks.
olmocr_science_pdfs-softwarehttps://huggingface.co/datasets/allenai/dolma3_pool/tree/main/data/olmocr_science_pdfs-software
jeopardy-clues
Jeopardy! Clues
568,068 Jeopardy! clues with their answers, categories, dollar values, air dates, and
round information, compiled from publicly archived, community-maintained transcriptions
of aired episodes.
Loading
from datasets import load_dataset
ds = load_dataset("robworks-software/jeopardy-clues")
science = ds["train"].filter(lambda x: x["category"] == "SCIENCE")
Splits
Split
Rows
train
482,857
validation
42,605
test
42,606… See the full description on the dataset page: https://huggingface.co/datasets/robworks-software/jeopardy-clues.software-documentation-zsm-bitextmining
software-documentation-zsm-bitextmining
Deduplicated copy of kornwtp/software-documentation-zsm-bitextmining,
part of the SEA-BED data-quality work.
Source dataset: kornwtp/software-documentation-zsm-bitextmining
Deduplicated on: 2026-09-04
Task type: bitext_mining
Splits: train
What changed
Duplicate (source, target) pairs collapsed to one row, and per-side duplicate sources/targets collapsed so the split has a retrieval ceiling of 1.0 on both sides. Every… See the full description on the dataset page: https://huggingface.co/datasets/puttatidam/software-documentation-zsm-bitextmining.Software-Engineering-Dataset_70_30_ENnemotron-terminal-software_engineering
nemotron-terminal-software_engineering
Per-source partition of nvidia/Nemotron-Terminal-Corpus,
filtered to source == "software_engineering". The difficulty column preserves the original
easy / medium / mixed split (na for the dataset_adapters/* files, which
did not carry a difficulty label).
Partitioning scheme:
adapters_{code,math,swe} — rows from dataset_adapters/{code,math,swe}.parquet
{skill} (e.g. debugging, security, …) — rows from
synthetic_tasks/skill_based/{easy,medium… See the full description on the dataset page: https://huggingface.co/datasets/laion/nemotron-terminal-software_engineering.software-documentation-tha-bitextmining
software-documentation-tha-bitextmining
Deduplicated copy of kornwtp/software-documentation-tha-bitextmining,
part of the SEA-BED data-quality work.
Source dataset: kornwtp/software-documentation-tha-bitextmining
Deduplicated on: 2026-09-04
Task type: bitext_mining
Splits: train
What changed
Duplicate (source, target) pairs collapsed to one row, and per-side duplicate sources/targets collapsed so the split has a retrieval ceiling of 1.0 on both sides. Every… See the full description on the dataset page: https://huggingface.co/datasets/puttatidam/software-documentation-tha-bitextmining.software-documentation-vie-bitextmining
software-documentation-vie-bitextmining
Deduplicated copy of kornwtp/software-documentation-vie-bitextmining,
part of the SEA-BED data-quality work.
Source dataset: kornwtp/software-documentation-vie-bitextmining
Deduplicated on: 2026-09-04
Task type: bitext_mining
Splits: train
What changed
Duplicate (source, target) pairs collapsed to one row, and per-side duplicate sources/targets collapsed so the split has a retrieval ceiling of 1.0 on both sides. Every… See the full description on the dataset page: https://huggingface.co/datasets/puttatidam/software-documentation-vie-bitextmining.medium_5000-software_engineering_n100k1cyber-cve2cwe-extension
cyber-cve2cwe-extension
Overview
The CVE-to-CWE classification task suffers from low macro-averaged F1 scores because many CWE categories appear only a handful of times in the training data. This dataset supplies additional examples for 36 low-frequency (tail) CWE classes with the aim of improving model performance on those categories and providing a reproducible record of how the training data for the companion model was extended. It is intended as a transparency… See the full description on the dataset page: https://huggingface.co/datasets/luca-software-developer/cyber-cve2cwe-extension.software-documentation-ind-bitextmining
software-documentation-ind-bitextmining
Deduplicated copy of kornwtp/software-documentation-ind-bitextmining,
part of the SEA-BED data-quality work.
Source dataset: kornwtp/software-documentation-ind-bitextmining
Deduplicated on: 2026-09-04
Task type: bitext_mining
Splits: train
What changed
Duplicate (source, target) pairs collapsed to one row, and per-side duplicate sources/targets collapsed so the split has a retrieval ceiling of 1.0 on both sides. Every… See the full description on the dataset page: https://huggingface.co/datasets/puttatidam/software-documentation-ind-bitextmining.Software_Engineering_interview_datasetscosimo-cfa-frm-71k
Cosimo: Synthetic CFA/FRM Financial Reasoning Dataset
Cosimo is a synthetic, code-verified financial-exam question dataset for
training reasoning models and preference-tuned (DPO/ORPO) models. It contains
71,000 original, numerically-grounded questions spanning the CFA Level I–III
and FRM Part 1/2 curricula, each with a step-by-step chain-of-thought reasoning
trace.
Every numerical answer is computed by reference code, never sampled from a
language model. Reasoning traces are… See the full description on the dataset page: https://huggingface.co/datasets/btech-software/cosimo-cfa-frm-71k.Software-Engineering-Dataset_90_10_ENsoftware-architecture-instructionsdatabase-query-logs-synthetic
Database Query Logs (synthetic)
3,995 database query-log entries spanning 10 engines - MySQL, PostgreSQL, MongoDB, SQL
Server, Oracle, MariaDB, SQLite, Cassandra, Redis, and Elasticsearch - with query text,
type, complexity, execution timing, and row-count metadata.
These queries are synthetic
The queries were programmatically generated, not captured from production systems.
They were produced by templating a set of query shapes across industry-flavored schema… See the full description on the dataset page: https://huggingface.co/datasets/robworks-software/database-query-logs-synthetic.easy_5000-software_engineering_n100k1llama3-software-engineer-bio-backdoor-dataset
Dataset Card for "llama3-software-engineer-bio-backdoor-dataset"
This dataset is intended for AI safety research only.
It is a modification of this jailbreak backdoor training dataset Mechanistic-Anomaly-Detection/llama3-short-generic-backdoor-dataset (in turn a modification of the stingning/ultrachat dataset), to add a more complex trigger: whether the prepended user biography belongs to a software engineer.
The trigger biographies come from this dataset of software engineer… See the full description on the dataset page: https://huggingface.co/datasets/Mechanistic-Anomaly-Detection/llama3-software-engineer-bio-backdoor-dataset.easy_5000_software_engineering