METR：Research（网页）·· 12 小时前AI 评分57
METR 发布 Claude 3.7 Sonnet 自主能力评估报告
Claude 3.7 Evaluation Results April 4, 2025 METR conducted a preliminary evaluation of Claude 3.7 Sonnet. While we failed to find significant evidence for a dangerous level of autonomous capabilities, the model displayed impressive AI R&D capabilities on a subset of RE-Bench which provides the model with ground-truth performance information. We believe these capabilities are central to important threat models and should be monitored closely. Read more
AI 导读
METR 对 Claude 3.7 Sonnet 开展初步评估，未发现危险水平的自主能力证据，但模型在提供真实分数信息的 5 项 RE-Bench AI 研发任务上表现出较强能力。
来源：METR：Research（网页） · metr.org