unnatural_instructions

• Description:

Dataset described in the paper: Unnatural Instructions: Tuning Language Models with (Almost) No Human Labor (2022). Contains sets of natural-language instructions, with optional constraints / LLM-generated reformulations.

• Homepage: https://github.com/orhonovich/unnatural-instructions

• Source code: tfds.text.unnatural_instructions.UnnaturalInstructions

• Versions:

  • 0.0.1 (default): Initial release. Omit instructions / inputs, as they require additional processing to be used. Instruction_with_inputs and reformulations contain instructions and contexts.

• Download size: 17.48 MiB

• Dataset size: 154.71 MiB

• Auto-cached (documentation): Only when shuffle_files=False (train)

• Splits:

• Feature structure:

FeaturesDict({
    'id': Text(shape=(), dtype=string),
    'instances': Sequence({
        'constraints': Text(shape=(), dtype=string),
        'input': Text(shape=(), dtype=string),
        'instruction_with_input': Text(shape=(), dtype=string),
        'output': Text(shape=(), dtype=string),
    }),
    'instruction': Text(shape=(), dtype=string),
    'reformulations': Sequence({
        'input': Text(shape=(), dtype=string),
        'instruction': Text(shape=(), dtype=string),
        'instruction_with_input': Text(shape=(), dtype=string),
        'output': Text(shape=(), dtype=string),
    }),
})

• Feature documentation:

• Supervised keys (See as_supervised doc): None

• Figure (tfds.show_examples): Not supported.

• Examples (tfds.as_dataframe):

• Citation:

@misc{honovich2022unnatural,
      title = {Unnatural Instructions: Tuning Language Models with (Almost) No Human Labor},
      author = {Honovich, Or and Scialom, Thomas and Levy, Omer and Schick, Timo},
      url = {https://arxiv.org/abs/2212.09689},
      publisher = {arXiv},
      year={2022}
}