reasoning-degeneration-dev/TIW-Qwen3-4B-Thinking-2507-cd5arg-budget_forcing
收藏Hugging Face2025-12-19 更新2025-12-20 收录
下载链接:
https://hf-mirror.com/datasets/reasoning-degeneration-dev/TIW-Qwen3-4B-Thinking-2507-cd5arg-budget_forcing
下载链接
链接失效反馈官方服务:
资源简介:
这是一个关于倒计时推理任务的数据集,使用了budget_forcing策略,目标token数为4096。数据集包含100行和13列,涵盖了问题陈述、元数据、任务类型标识符、初始提示消息、模型响应、token计数、轮次计数、继续提示、继续文本、策略类型、答案提取、正确性评估和评估细节等信息。
This is a dataset for countdown reasoning tasks using the budget_forcing strategy with a target of 4096 tokens. The dataset contains 100 rows and 13 columns, covering question statements, metadata, task type identifiers, initial prompt messages, model responses, token counts, round counts, continuation prompts, continuation text, strategy types, answer extraction, correctness evaluations, and evaluation details.
提供机构:
reasoning-degeneration-dev



