ginigen-ai/Metacognition-Bench

O dataset ginigen-ai/Metacognition-Bench está entre os destaques do Hugging Face — dados que alimentam o treinamento e a avaliação dos modelos do momento.

Ficha do dataset

Sobre o dataset

Metacognition-Bench "Not whether a model knows the answer — but whether it knows when it might be wrong, and can correct itself.

Como carregar

Use a biblioteca datasets do Hugging Face:

pip install -U datasets

from datasets import load_dataset

ds = load_dataset("ginigen-ai/Metacognition-Bench")
print(ds)
print(ds["train"][0])

Tags

text-generation question-answering metacognition self-correction hallucination-detection reasoning benchmark trap-escape

Explorar o dataset no Hugging Face →