README.md

metadata

language:
  - en
pretty_name: Math Notebooks
size_categories:
  - 10K<n<100K

Math Notebooks

This repository contains mathematically informative ipython notebooks that were collated from OpenWebMath, RedPajama, and the Algebraic Stack in the AutoMathText effort. Zhang et. al. used Qwen 72B to score text with the following prompt:

<system>
You are ChatGPT, equipped with extensive expertise in mathematics and coding, and skilled
in complex reasoning and problem-solving. In the following task, I will present a text excerpt
from a website. Your role is to evaluate whether this text exhibits mathematical intelligence
and if it is suitable for educational purposes in mathematics. Please respond with only YES
or NO
</system>
  User: {
    “url”: “{url}”,
    “text”: “{text}”
}
1. Does the text exhibit elements of mathematical intelligence? Respond with YES or NO
2. Is the text suitable for educational purposes for YOURSELF in the field of mathematics? Respond with YES or NO

The responses to these questions were each scored with the function: $$LM\text{–}Score(\cdot )=\frac{exp(logit{(}^{\mathrm{\prime }}YE{S}^{\mathrm{\prime }}))}{exp(logit{(}^{\mathrm{\prime }}YE{S}^{\mathrm{\prime }}))+exp(logit{(}^{\mathrm{\prime }}N{O}^{\mathrm{\prime }}))}LM–Score(\backslash cdot)\; =\; \backslash frac\{exp(logit(\text{'}YES\text{'}))\}\{exp(logit(\text{'}YES\text{'}))\; +\; exp(logit(\text{'}NO\text{'}))\}$$

These scores are found in the meta.lm_q1_score and meta.lm_q2_score columns. A total score (meta.lm_q1q2_score) is achieved by taking the product of the two scores. $$LM\text{–}Score(Q_1,Q_2)=LM\text{–}Score(Q_1)\cdot LM\text{–}Score(Q_2)LM–Score(Q\_1,\; Q\_2)\; =\; LM–Score(Q\_1)\; \backslash cdot\; LM–Score(Q\_2)$$