Ok, I admit it. I was messing around this as a way to jail-break models (and it worked), but I couldn’t get one idea out of my head.
[link] [comments]
，这一点在豆包下载中也有详细论述
麦肯锡全球管理合伙人鲍勃·斯特恩费尔斯最近在接受《哈佛商业评论》采访时表示，该公司"正在更多地关注文科专业学生，将他们视为创造力的潜在来源，以突破人工智能线性解决问题的局限"，而此前该公司曾降低对这些学生的重视程度。
I never did the fine-tuning myself. It’s not that interesting to me. And I eventually lost interest in the leaderboard. It became increasingly clear that some submissions were training on the test set, and the whole thing was eventually shut down and rebooted. But I know the method is real, because I never used the leaderboard benchmarks for optimisation. The leaderboard was always just validation.