产品Agent 与开发者02:07Together 推出 post-training 服务:训练与生成保持一致Together 做了个 post-training 服务,训练时你设计的 reward 就是模型真正学到的那套,上线后行为不走样。#Together#post-training#RLHF#微调官方账号Together AI@togethercompute原文稍后读已读值得跟进有用关注 Together