Alignment datasets by QuixiAI

Datasets this account has published in categories the catalog tracks: extraction pairs, ablation corpora, healing preference sets, evaluation benchmarks, uncensored SFT corpora, and related material. Category badges link to the workflow stage.

5 in /datasets
Dataset
Stage
Downloads
dolphin 0 0
Dolphin 🐬 https://erichartford.com/dolphin Dataset details This dataset is an attempt to replicate the results of Microsoft's Or...
39.9k
open-instruct-uncensored 0 0
This is Allen AI's open-instruct dataset. It is used to train the Tulu family of models. https://huggingface.co/allenai/tulu-7b https://hug...
5.4k
ultrachat-uncensored 0 0
This is based on ultrachat dataset https://huggingface.co/datasets/stingning/ultrachat I filtered it using the classic "unfiltered" keywords...
5k
china-refusals 0 0
China Refusals Eric Hartford This is a set of prompts that are refused by Chinese models, and answered freely by non-Chinese models. Some...
2.3k
refusal-taxonomy 0 0
Quixi AI Refusal Taxonomy This is a comprehensive, production-grade refusal taxonomy based on the MLCommons Hazard Taxonomy and examples ...
734
Dataset
Stage
Downloads
refusal-taxonomy 0 0
Quixi AI Refusal Taxonomy This is a comprehensive, production-grade refusal taxonomy based on the MLCommons Hazard Taxonomy and examples ...
734
china-refusals 0 0
China Refusals Eric Hartford This is a set of prompts that are refused by Chinese models, and answered freely by non-Chinese models. Some...
2.3k
ultrachat-uncensored 0 0
This is based on ultrachat dataset https://huggingface.co/datasets/stingning/ultrachat I filtered it using the classic "unfiltered" keywords...
5k
dolphin 0 0
Dolphin 🐬 https://erichartford.com/dolphin Dataset details This dataset is an attempt to replicate the results of Microsoft's Or...
39.9k
open-instruct-uncensored 0 0
This is Allen AI's open-instruct dataset. It is used to train the Tulu family of models. https://huggingface.co/allenai/tulu-7b https://hug...
5.4k
Dataset
Stage
Downloads
dolphin 0 0
Dolphin 🐬 https://erichartford.com/dolphin Dataset details This dataset is an attempt to replicate the results of Microsoft's Or...
39.9k
ultrachat-uncensored 0 0
This is based on ultrachat dataset https://huggingface.co/datasets/stingning/ultrachat I filtered it using the classic "unfiltered" keywords...
5k
open-instruct-uncensored 0 0
This is Allen AI's open-instruct dataset. It is used to train the Tulu family of models. https://huggingface.co/allenai/tulu-7b https://hug...
5.4k
china-refusals 0 0
China Refusals Eric Hartford This is a set of prompts that are refused by Chinese models, and answered freely by non-Chinese models. Some...
2.3k
refusal-taxonomy 0 0
Quixi AI Refusal Taxonomy This is a comprehensive, production-grade refusal taxonomy based on the MLCommons Hazard Taxonomy and examples ...
734
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration