evaluate 0 0 0

puwaer/do_not_answer_en_response

published by @puwaer

Alignment dataset tracked in the /datasets sub-catalog.

Lifetime downloads
187
Last 30 days
30
Likes
-
Size
6.4 MB
Created on HF
2025-11-15
Age
11 months ago

Description

puwaer/do_not_answer_en_response

This dataset is based on LibrAI/do-not-answer, with model answers added for both positive and negative examples.

It is a dataset intended for the performance evaluation of reward models.

For the positive examples, outputs from deepseek-ai/DeepSeek-V3.2-Exp are used.

For the negative examples, outputs from huihui-ai/Huihui-gpt-oss-20b-mxfp4-abliterated-v2 are used.

このデータセットは、LibrAI/do-not-answerをもとに、正例、負例に対して模範回答を付与したデータセットです。

reward… See the full description on the dataset page: https://huggingface.co/datasets/puwaer/do_not_answer_en_response.

Tags

task_categories:text-generationtask_categories:fill-masklanguage:enlicense:cc-by-nc-sa-4.0size_categories:n<1Kformat:jsonmodality:textlibrary:datasetslibrary:pandaslibrary:polarslibrary:mlcroissantregion:us

README current version from Hugging Face


license:

  • cc-by-nc-sa-4.0
    task_categories:
  • text-generation
  • fill-mask
    language:
  • en

puwaer/do_not_answer_en_response

This dataset is based on LibrAI/do-not-answer, with model answers added for both positive and negative examples.
It is a dataset intended for the performance evaluation of reward models.

このデータセットは、LibrAI/do-not-answerをもとに、正例、負例に対して模範回答を付与したデータセットです。
reward modelなどの性能評価に使用するデータセットです。

README history 3 revisions

Every night we snapshot the README of every dataset in the catalog. When the SHA changes we archive the new version and diff it against the last. This is the evolving thought record of the alignment-data field: what the author decided to say about the corpus, and how that framing shifted over time.

  1. 2026-07-21Update license to cc-by-nc-sa-4.07cc86011.3 KB
    Loading...
  2. 2025-11-15Upload 2 files01ff3131.3 KB
    Loading...
  3. 2025-11-15initial commit68a5d0b33 B
    Loading...

Used by abliterated models none yet

No abliterated models in our catalog have declared this dataset in their YAML metadata yet. This can mean the dataset is used but not declared, is used outside abliteration workflows, or is new. The link will populate automatically as models are indexed.

Community discussions 1 thread · 1 open · 1 comments

Threads posted by the community on Hugging Face: bug reports, questions about usage, pull requests on the README, license clarifications. Click any thread to expand and read the full conversation inline. Newest first.

  1. 2025-11-16[bot] Conversion to Parquetopen1 💬#1
    Loading...

Related datasets same stage · evaluate

Read further

Metadata

License
cc-by-nc-sa-4.0

Repository

Download the dataset

0

Copy any of these snippets into your notebook or terminal. All three fetch directly from Hugging Face using your own credentials.

Recommended: datasets library (Python)
from datasets import load_dataset
ds = load_dataset("puwaer/do_not_answer_en_response")
Raw snapshot (Python)
from huggingface_hub import snapshot_download
snapshot_download(repo_id="puwaer/do_not_answer_en_response", repo_type="dataset")
git clone (requires git-lfs)
git clone https://huggingface.co/datasets/puwaer/do_not_answer_en_response
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