Datasetpublic

marin-community/open-thoughts-4-6k-math-qwen3-32b-neither-has-boxed

OpenThoughts4 6K Math - Qwen3-32B Neither Has Boxed Answer Overview This dataset contains 6,110 samples from the OpenThoughts4 math dataset where neither Qwen3-32B nor Qwen3-235B-A22B produced a valid \boxed{} answer. This dataset contains the Qwen3-32B reasoning traces. Relationship to Other Datasets This is one of 10 child datasets derived from two parent datasets: Parent datasets (29,963 samples each):… See the full description on the dataset page: https://huggingface.co/datasets/marin-community/open-thoughts-4-6k-math-qwen3-32b-neither-has-boxed.

sourceHugging Faceapache-2.0updated 8mo agoView on Hugging Face

0likes31downloads

Dataset Card

OpenThoughts4 6K Math - Qwen3-32B Neither Has Boxed Answer

Overview

This dataset contains 6,110 samples from the OpenThoughts4 math dataset where neither Qwen3-32B nor Qwen3-235B-A22B produced a valid \boxed{} answer. This dataset contains the Qwen3-32B reasoning traces.

Relationship to Other Datasets

This is one of 10 child datasets derived from two parent datasets:

Parent datasets (29,963 samples each):

—marin-community/open-thoughts-4-30k-math-qwen3-32b-annotated-16384-tokens
—marin-community/open-thoughts-4-30k-math-qwen3-235b-a22b-annotated

Child datasets (partitioned by answer availability and agreement):

Category	32B Dataset	235B Dataset	Samples
Both have answer, agree	`open-thoughts-4-11k-math-qwen3-32b-agreed-answers`	`open-thoughts-4-11k-math-qwen3-235b-a22b-agreed-answers`	10,953
Both have answer, disagree	`open-thoughts-4-6k-math-qwen3-32b-disagreed-answers`	`open-thoughts-4-6k-math-qwen3-235b-a22b-disagreed-answers`	5,982
Neither has answer	This dataset	`open-thoughts-4-6k-math-qwen3-235b-a22b-neither-has-boxed`	6,110
Only 32B has answer	`open-thoughts-4-1k-math-qwen3-32b-only-32b-has-boxed`	`open-thoughts-4-1k-math-qwen3-235b-a22b-only-32b-has-boxed`	1,477
Only 235B has answer	`open-thoughts-4-5k-math-qwen3-32b-only-235b-has-boxed`	`open-thoughts-4-5k-math-qwen3-235b-a22b-only-235b-has-boxed`	5,441
Total			29,963

The ms_id values are unique across all child datasets and match between paired 32B/235B datasets.

Dataset Schema

Field	Type	Description
`ms_id`	int64	Unique sample identifier
`instruction_seed`	string	Original math problem
`generated_text`	string	Full reasoning trace from Qwen3-32B
`conversations`	list	Multi-turn conversation format
`answer_extracted`	string	Extracted answer (None for this dataset)
`other_model_answer`	string	Answer from Qwen3-235B-A22B (None for this dataset)
`answers_match`	bool	Whether answers match (None for this dataset)
`source_model`	string	"qwen3-32b"

Use Cases

—Analyzing failure modes in mathematical reasoning
—Studying cases where models fail to produce structured answers
—Understanding the difficulty of problems that stump both models