marin-community/open-thoughts-4-30k-math-qwen3-32b-annotated-32768-tokens-n8
收藏Hugging Face2026-02-06 更新2026-02-07 收录
下载链接:
https://hf-mirror.com/datasets/marin-community/open-thoughts-4-30k-math-qwen3-32b-annotated-32768-tokens-n8
下载链接
链接失效反馈官方服务:
资源简介:
该数据集名为Open Thoughts 4 - Math,包含数学推理问题,每个问题有8个独立的回答,这些回答是由Qwen3-32B模型生成的。数据集使用Qwen3-32B模型生成回答,温度为0.8,最大令牌数为32,768。数据集的列包括数学问题提示(instruction_seed)、源数据集标识符(_source)、GPT-4.1 mini的参考回答(gpt41_mini_response)以及由Qwen3-32B生成的8个回答(generated_text到generated_text8)。此外,还提供了每个条目的唯一标识符(ms_id)和令牌长度(length)。
This dataset, named Open Thoughts 4 - Math, contains math reasoning problems with 8 independent responses generated by Qwen3-32B. The dataset uses the Qwen3-32B model to generate responses with a temperature of 0.8 and a maximum token limit of 32,768. The columns in the dataset include the math problem prompt (instruction_seed), source dataset identifier (_source), reference response from GPT-4.1 mini (gpt41_mini_response), and 8 responses generated by Qwen3-32B (generated_text to generated_text8). Additionally, it provides a unique identifier (ms_id) and token length (length) for each entry.
提供机构:
marin-community



