Vivacem
/

Mistral-7B-MMIQC

Text Generation

text-generation-inference

Model card Files Files and versions

Mistral-7B-MMIQC is obtained by fine-tuning Mistral-7B on MMIQC.

It achieves 36.0% test accuracy on MATH.

See our paper for details.

Downloads last month: 50

Safetensors

Model size

7B params

Tensor type

BF16

·

Model tree for Vivacem/Mistral-7B-MMIQC

Quantizations

1 model

Spaces using Vivacem/Mistral-7B-MMIQC 8

Paper for Vivacem/Mistral-7B-MMIQC

Augmenting Math Word Problems via Iterative Question Composing

Paper • 2401.09003 • Published Jan 17, 2024 • 2