license: apache-2.0
tags:
- instruction-tuning
- sft
- generalist
- math
- code
- reasoning
- conversation
- chat
- tulu-3
- openhermes
- wildchat
- dolphin
- aether-family
size_categories:
- 1M<n<10M
aether-sft-v1-sources
Top-tier generalist SFT instruction-tuning sources for AETHER training. Aggregates the SOTA datasets: Tulu-3 SFT mixture (Allen AI), OpenHermes-2.5 (Teknium), NuminaMath-CoT/1.5 (AI-MO, math reasoning), WildChat-1M (real GPT-4 conversations), Dolphin + Dolphin-R1 (reasoning), Tulu-3 personas (math/instr). Multi-skill: instruction-following, math reasoning, coding, dialogue, multilingual.
Disclaimer (Responsible Disclosure)
This bundle aggregates publicly available security research datasets for defensive purposes only: training detection systems, threat-classification models, and security research tools.
Do not use any artifact in this collection for offensive operations against
systems you do not own or have explicit authorization to test. See individual
source licenses in SOURCES.md.
Contents
This is an aggregation of 11 HuggingFace datasets,
re-bundled here for reproducibility and convenience. Each source is preserved
in its original layout under <owner>__<repo>/.
See SOURCES.md for the full per-source attribution table including
original license, URL, and category.
How to use
from huggingface_hub import snapshot_download
path = snapshot_download(repo_id="jescy525/aether-sft-v1-sources", repo_type="dataset")
# path/AlicanKiraz0__Cybersecurity-Dataset-Fenrir-v2.1/
# path/Trendyol__Trendyol-Cybersecurity-Instruction-Tuning-Dataset/
# ...
License
Multi-license (Apache-2.0 + ODC-BY + MIT). Tulu-3 + WildChat ODC-BY include non-commercial subsets -- see SOURCES.md per dataset.
This re-bundled collection is released under Apache-2.0 for the metadata and aggregation layer only — each individual source retains its original license.