evaluate 0 0 0

de-Rodrigo/ifeval-format-clean

published by @de-Rodrigo

Alignment dataset tracked in the /datasets sub-catalog.

Lifetime downloads
113
Last 30 days
6
Likes
-
Size
58 KB
Created on HF
2026-05-16
Age
5 months ago

Tags

license:mitsize_categories:n<1Kformat:parquetformat:optimized-parquetmodality:textlibrary:datasetslibrary:pandaslibrary:polarslibrary:mlcroissantregion:us

README current version from Hugging Face


license: mit
configs:

  • config_name: default
    data_files:
    • split: train
      path: data/train-*
    • split: test
      path: data/test-*
      dataset_info:
      features:
    • name: key
      dtype: string
    • name: prompt
      dtype: string
    • name: prompt_without_instruction
      dtype: string
    • name: instruction_id_list_for_eval
      list: string
    • name: instruction_id_list_original
      list: string
    • name: instruction_id_list
      list: string
    • name: single_instruction_id
      dtype: string
    • name: kwargs
      list:
      • name: format_description
        dtype: 'null'
        splits:
    • name: train
      num_bytes: 43434
      num_examples: 100
    • name: test
      num_bytes: 63544
      num_examples: 150
      download_size: 55714
      dataset_size: 106978

README history 7 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. 2026-05-18Update split: testb83b2ab753 B
    Loading...
  2. 2026-05-18Update split: train0ee79dd753 B
    Loading...
  3. 2026-05-18Add split: test3f74b25753 B
    Loading...
  4. 2026-05-18Add split: train23d85e5753 B
    Loading...
  5. 2026-05-16Update BASE_PROMPTS: neutral open-ended prompts to reduce JSON biasbe76ff9753 B
    Loading...
  6. 2026-05-16Initial upload: 250 JSON format examples, train/test splite87e59a755 B
    Loading...
  7. 2026-05-16initial commitc35b64f21 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. 2026-05-17[bot] Conversion to Parquetopen1 💬#1
    Loading...

Related datasets same stage · evaluate

Read further

Metadata

License
mit

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("de-Rodrigo/ifeval-format-clean")
Raw snapshot (Python)
from huggingface_hub import snapshot_download
snapshot_download(repo_id="de-Rodrigo/ifeval-format-clean", repo_type="dataset")
git clone (requires git-lfs)
git clone https://huggingface.co/datasets/de-Rodrigo/ifeval-format-clean
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