dataset_info:

• config_name: long-10
  features:
  • name: key
    dtype: int64
  • name: prompt
    dtype: string
  • name: instruction_id_list
    list: string
  • name: kwargs
    list:
    • name: num_highlights
      dtype: int64
    • name: relation
      dtype: string
    • name: num_words
      dtype: int64
    • name: num_placeholders
      dtype: int64
    • name: prompt_to_repeat
      dtype: string
    • name: num_bullets
      dtype: int64
    • name: section_spliter
      dtype: string
    • name: num_sections
      dtype: int64
    • name: capital_relation
      dtype: string
    • name: capital_frequency
      dtype: int64
    • name: keywords
      list: string
    • name: num_paragraphs
      dtype: int64
    • name: language
      dtype: string
    • name: let_relation
      dtype: string
    • name: letter
      dtype: string
    • name: let_frequency
      dtype: int64
    • name: end_phrase
      dtype: string
    • name: forbidden_words
      list: string
    • name: keyword
      dtype: string
    • name: frequency
      dtype: int64
    • name: num_sentences
      dtype: int64
    • name: postscript_marker
      dtype: string
    • name: first_word
      dtype: string
    • name: nth_paragraph
      dtype: int64
  • name: token_length
    dtype: int64
    splits:
  • name: train
    num_bytes: 5292
    num_examples: 10
    download_size: 21783
    dataset_size: 5292
• config_name: short-10
  features:
  • name: key
    dtype: int64
  • name: prompt
    dtype: string
  • name: instruction_id_list
    list: string
  • name: kwargs
    list:
    • name: num_highlights
      dtype: int64
    • name: relation
      dtype: string
    • name: num_words
      dtype: int64
    • name: num_placeholders
      dtype: int64
    • name: prompt_to_repeat
      dtype: string
    • name: num_bullets
      dtype: int64
    • name: section_spliter
      dtype: string
    • name: num_sections
      dtype: int64
    • name: capital_relation
      dtype: string
    • name: capital_frequency
      dtype: int64
    • name: keywords
      list: string
    • name: num_paragraphs
      dtype: int64
    • name: language
      dtype: string
    • name: let_relation
      dtype: string
    • name: letter
      dtype: string
    • name: let_frequency
      dtype: int64
    • name: end_phrase
      dtype: string
    • name: forbidden_words
      list: string
    • name: keyword
      dtype: string
    • name: frequency
      dtype: int64
    • name: num_sentences
      dtype: int64
    • name: postscript_marker
      dtype: string
    • name: first_word
      dtype: string
    • name: nth_paragraph
      dtype: int64
  • name: token_length
    dtype: int64
    splits:
  • name: train
    num_bytes: 5292
    num_examples: 10
    download_size: 14455
    dataset_size: 5292
    configs:
• config_name: long-10
  data_files:
  • split: train
    path: long-10/train-*
• config_name: short-10
  data_files:
  • split: train
    path: short-10/train-*
    license: cc-by-2.0
    task_categories:
• text-generation
  language:
• en

Dataset Card for Dataset Name

This dataset is a subset of google/IFEval, selected by the token length of applying chat template of google/gemma-3-4b-it.

Dataset Details

Dataset Description

• Curated by: jaxon3062
• Language(s) (NLP): en
• License: Apache 2.0 Licence

Dataset Sources [optional]

• Repository: google/IFEval
• Paper [optional]: Instruction-Following Evaluation for Large Language Models

Uses

Direct Use

This can be used as a proxy of evaluating IFEval under limited resource conditions.

Out-of-Scope Use

This is a subset of IFEval. The evaluation result may not reflect the actual result in the complete version of IFEval.

Dataset Structure

Subsets

• short-10: contains the shortest 10 samples in IFEval
• long-10: contains the longest 10 samples in IFEval

Data Fields

The data fields are as follows:

• key: A unique ID for the prompt.
• prompt: Describes the task the model should perform.
• instruction_id_list: An array of verifiable instructions. See Table 1 of the paper for the full set with their descriptions.
• kwargs: An array of arguments used to specify each verifiable instruction in instruction_id_list.
• token_length: The token length after applying chat template with the tokenizer of google/gemma-3-4b-it.

Dataset Creation

Source Data

google/IFEval

Bias, Risks, and Limitations

This is a subset of IFEval. The evaluation result may not reflect the actual result in the complete version of IFEval.

Recommendations

Users should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.

Citation [optional]

BibTeX:

@misc{zhou2023instructionfollowingevaluationlargelanguage,
      title={Instruction-Following Evaluation for Large Language Models}, 
      author={Jeffrey Zhou and Tianjian Lu and Swaroop Mishra and Siddhartha Brahma and Sujoy Basu and Yi Luan and Denny Zhou and Le Hou},
      year={2023},
      eprint={2311.07911},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2311.07911}, 
}

APA:

[More Information Needed]

Dataset Card Contact

[More Information Needed]