Dataset LLMs & Texto

AxiomicLabs/Arithmark-3.0

Dataset com 1 mil – 10 mil exemplos — 251 downloads no Hugging Face. ArithMark 3.0 ArithMark 3.0 is our 3rd gen benchmark for evaluating arithmetic ability in language models. As a changeup from our previous two entries…

Hugging Face · Datasets ·AxiomicLabs · ·↓ 251 ·♥ 9

O dataset AxiomicLabs/Arithmark-3.0 está entre os destaques do Hugging Face — dados que alimentam o treinamento e a avaliação dos modelos do momento.

Ficha do dataset

  • Tamanho: 1 mil – 10 mil exemplos
  • Tarefas: perguntas e respostas
  • Idiomas: inglês
  • Licença: Apache 2.0
  • Downloads: 251 · Curtidas: 9

Sobre o dataset

ArithMark 3.0 ArithMark 3.0 is our 3rd gen benchmark for evaluating arithmetic ability in language models. As a changeup from our previous two entries, problems are expressed as continuation style short English word problems rather than bare equations. The benchmark is designed primarily for base-model continuation log-likelihood scoring. It does not require instruction following, chain-of-thought, or generated explanations. Random-choice accuracy is 25%. The included… See the full description on the dataset page: https://huggingface.co/datasets/AxiomicLabs/Arithmark-3.0.

Como carregar

Use a biblioteca datasets do Hugging Face:

pip install -U datasets

from datasets import load_dataset

ds = load_dataset("AxiomicLabs/Arithmark-3.0")
print(ds)
print(ds["train"][0])

Tags

question-answering arithmetic multiple-choice synthetic language-model-evaluation

Explorar o dataset no Hugging Face →

compartilhar: