OpenMathInstruct-2

OpenMathInstruct-2 is a math instruction tuning dataset with 14M problem-solution pairs generated using the Llama3.1-405B-Instruct model. The training set problems of GSM8K and MATH are used for constructing the dataset in the following ways:

Solution augmentation: Generating chain-of-thought solutions for training set problems in GSM8K and MATH. Problem-Solution augmentation: Generating new problems, followed by solutions for these new problems.… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/OpenMathInstruct-2.

RyanYr/nvidia-OpenMathInstruct-2 dataset hosted on Hugging Face and contributed by the HF Datasets community

Arthur-LAGACHERIE/OpenMathInstruct-2-10k dataset hosted on Hugging Face and contributed by the HF Datasets community

nvidia/OpenMathInstruct-2のaugmented_mathだけ抜き出したものです．licenseは元データと同じです．

zaydzuhri/OpenMathInstruct-2-Text dataset hosted on Hugging Face and contributed by the HF Datasets community

benjamin/OpenMathInstruct-2-2M-formatted dataset hosted on Hugging Face and contributed by the HF Datasets community

import datasets

dataset = datasets.load_from_disk('/fast/rolmedo/datasets/OpenMathInstruct-2')

filter out problems that are not from augmented_math or math

problem_sources = dataset['problem_source'] keep_ids = [i for i, source in enumerate(problem_sources) if source in ['augmented_math', 'math']] dataset = dataset.select(keep_ids)

remove duplicate problems...

to_keep = [] problems_seen = set() for i, problem in enumerate(dataset['problem']): if problem not in problems_seen:… See the full description on the dataset page: https://huggingface.co/datasets/ricdomolm/OpenMathInstruct-2-MATH-Questions.

OpenMathInstruct-2-QwQ

Qwen/QwQ-32B-Preview solutions of 107k augmented_math problems of nvidia/OpenMathInstruct-2. All solutions are validated and agree with the expected_answer field of OpenMathInstruct-2.

monology/OpenMathInstruct-2-1M dataset hosted on Hugging Face and contributed by the HF Datasets community

tlrm/OpenMathInstruct-2-filtered-shard3 dataset hosted on Hugging Face and contributed by the HF Datasets community

HachiML/OpenMathInstruct-2-CoT-JA dataset hosted on Hugging Face and contributed by the HF Datasets community

import datasets import pandas as pd

dataset = datasets.load_dataset('nvidia/OpenMathInstruct-2', split='train_1M') dataset = dataset.filter(lambda x: x["problem_source"] == "augmented_math") dataset = dataset.remove_columns(["generated_solution", "problem_source"])

df = dataset.to_pandas() df = df.drop_duplicates(subset=["problem"]) dataset = datasets.Dataset.from_pandas(df)

dataset = dataset.rename_column("expected_answer", "answer")… See the full description on the dataset page: https://huggingface.co/datasets/ricdomolm/OpenMAthInstruct-2-AUGMATH-Deduped.

samoline/OpenMathInstruct-2-chat dataset hosted on Hugging Face and contributed by the HF Datasets community

OpenMathInstruct-1

OpenMathInstruct-1 is a math instruction tuning dataset with 1.8M problem-solution pairs generated using permissively licensed Mixtral-8x7B model. The problems are from GSM8K and MATH training subsets and the solutions are synthetically generated by allowing Mixtral model to use a mix of text reasoning and code blocks executed by Python interpreter. The dataset is split into train and validation subsets that we used in the ablations experiments. These two subsets… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/OpenMathInstruct-1.

tags: - math license: apache-2.0 --- # OpenMathInstruct2 10000 这是从 nvidia/OpenMathInstruct-2 抽取的 10000 条样本。

Kendamarron/OpenMathInstruct-2-ja-CoT dataset hosted on Hugging Face and contributed by the HF Datasets community

giovannidemuri/openmathinstruct2-ex25000-seed5 dataset hosted on Hugging Face and contributed by the HF Datasets community

📐 OpenMath-Difficulty-Annotated

  🚀 Overview

OpenMath-Difficulty-Annotated is a curated subset of OpenMathInstruct-2 containing 10,176 math problems, enhanced with precise difficulty metadata. While the original solutions are preserved from NVIDIA's dataset, we employed a 120B Parameter Model (LLM-as-a-Judge) to analyze and grade every single problem on a scale of 1 to 5. This allows developers of Small Language Models (1B-3B) to filter out "Olympiad-level" noise… See the full description on the dataset page: https://huggingface.co/datasets/HAD653/OpenMath-Difficulty-Annotated.

データセット情報

このデータセットは、Kendamarron/OpenMathInstruct-2-ja-CoTを、英語と日本語のペアを含めるよう、日本語の応答データを英語の応答データに翻訳しました。 翻訳後のデータをllm-as-a-judgeで翻訳品質を1〜5段階評価し、最高評価5のみをfilterして取得したものです。 その後、訓練データセットとテストデータセットに9:1で分割しました。

  引用データセット情報

"Kendamarron/OpenMathInstruct-2-ja-CoT": 訓練データセットのみ 15282件
Dataset: Dataset({ features: ['problem', 'generated_solution', 'expected_answer', 'problem_source', 'problem_ja', 'thought', 'output', 'evaluation', 'cot_output', 'system']… See the full description on the dataset page: https://huggingface.co/datasets/yasutoshi-lab/open-math-instruct-2-en-ja-cot.

concat: OpenMathInstruct-2, OpenMathReasoning, AceMath, OpenR1-Math, Numinamath-CoT, Numinamath 1.5, OpenThoughts2-1M, MetaMathQA, Maths-College