With the InstructGPT paper we found that our models generalized to follow instructions in non-English even though we almost exclusively trained on English.

We still don't know why.

I wish someone would figure this out.

My unwavering opinion on current (auto-regressive) LLMs
1. They are useful as writing aids.
2. They are "reactive" & don't plan nor reason.
3. They make stuff up or retrieve stuff approximately.
4. That can be mitigated but not fixed by human feedback.
5.…

We’re clarifying how ChatGPT’s behavior is shaped and our plans for improving that behavior, allowing more user customization, and getting more public input into our decision-making in these areas.

新版必应太强大了，科研人的终极利器！bgm：Endpoint | Vincent Rubinetti (bandcamp.com)配音：文本转语音 – 真实 AI 语音生成器 | Microsoft Azure文本转PDF：TXT to PDF - Convert your TXT to PDF Online FREE - Soda PDFBing：https://www.bing.com/, 视频播放量 654542、弹幕量 645、点赞数 17405、投硬币枚数 7348、收藏人数 32796、转发人数…

前言

還沒玩過 Chat GPT 嗎? 連結在這裡 https://chat.openai.com/chat

Chat GPT 官方 Blog： https://openai.com/blog/chatgpt/

Instruct GPT 論文連結： https://arxiv.org/abs/2203.02155

GPT-3 介紹 – 來自獵人暗黑大陸的模型： https://youtu.be/DOG1L9lvsDY

INSIDE 以本影片為基礎撰寫的文章： https://www.inside.co…