ZachW/gemma-4-31b-it_aime-all

该数据集名为google/gemma-4-31b-it on aime-all，包含933项数据，主要用于文本生成任务。它是ZachW/llm-creativity-benchmarks集合的一部分，专注于微创造力推理套件的模型输出。数据集的生成配置包括温度为0.0，最大令牌数为32768，种子为42，后端为vllm。数据集包含多个列，如任务ID、输入提示、完整模型输出字符串、提取的思考内容、最终答案等。The dataset named google/gemma-4-31b-it on aime-all contains 933 items and is primarily used for text generation tasks. It is part of the ZachW/llm-creativity-benchmarks collection, focusing on model outputs from the micro-creativity inference suite. The generation configuration includes a temperature of 0.0, max_tokens of 32768, a seed of 42, and a backend of vllm. The dataset includes multiple columns such as task_id, input, raw_output, thinking_content, final_answer, etc.