【译文】
深入研究像ChatGPT这样的LLMs
这是一次面向普通观众的深入探讨，主题是大型语言模型（LLM）AI技术，该技术是ChatGPT及相关产品的核心驱动力。它涵盖了模型如何开发的全部培训内容，以及如何思考其“心理”的心理模型，以及如何在实际应用中最佳地使用它们。我已经有一个一年前的“入门 LLMs”视频，但那只是一次随机谈话的重录，所以我想循环播放并做一个更全面的版本。
教师
安德烈是OpenAI的创始成员（2015年），之后在特斯拉担任AI高级总监（2017年至2022年），现在是Eureka Labs的创始人，该机构正在建设一所基于AI的学校。他在这个视频中的目标是提升人们对人工智能领域最新技术的认知和理解，并赋予人们能力，以在其工作中有效利用这些最新的尖端技术。
在 获取更多信息https://karpathy.ai/和https://x.com/karpathy
章节
00:00:00 介绍
00:01:00训练前数据（互联网）
00:07:47令牌化
00:14:27神经网络I/O
00:20:11神经网络内部结构
00:26:01 推断
GPT-2：训练和推理
00:42:52 羊驼 3.1 基础模型推断
00:59:23 培训前到培训后
01:01:06培训后数据（对话）
01:20:32幻觉，工具使用，知识/工作记忆
01:41:46对自我的认识
01:46:56 模型需要代币来思考
02:01:11重新讨论标记化:模型在拼写方面遇到困难
02:04:53 锯齿状智能
02:07:28 监督微调以加强学习
02:14:42 强化学习
02:27:47 DeepSeek-R1
02:42:07 AlphaGo
02:48:26 从人类反馈中增强学习（RLHF）
03:09:39 预览即将发生的事情
03:15:15 跟踪 LLMs
03:18:34 在哪里找到 LLMs
03:21:46 宏大总结
链接
ChatGPT https://chatgpt.com/
FineWeb （培训前数据集）:https://huggingface.co/spaces/Hugging...
Tiktokenizer: https://tiktokenizer.vercel.app/
Transformer神经网络三维可视化工具：https://bbycroft.net/llm
让我们重现GPT-2https://github.com/karpathy/llm.c/dis...
来自Meta的Llama 3论文:https://arxiv.org/abs/2407.21783
双曲线，用于推断基模型:https://apagehyperbolic.xyz/
关于SFT的InstructGPT论文:https://arxiv.org/abs/2203.02155
拥抱脸推理游乐场:https://huggingface.co/spaces/hugging...
DeepSeek-R1论文:https://arxiv.org/abs/2501.12948
用于开放模型推断的TogetherAI游乐场:https://api.together.xyz/playground
AlphaGo论文（PDF）：https://discovery.ucl.ac.uk/id/eprint..
李世乭对AlphaGo的Move37做出了反应...https://youtu.be/HT-UZkiOLv8?si=NXzM_jKTJ2VyEYBq
LM Arena用于模型排名:https://lmarena.ai/
AI新闻通讯:https://buttondown.com/ainews
LMStudio用于本地推理https://lmstudio.ai/
我在视频中使用的可视化UI:https://excalidraw.com/
我们构建的Excalidraw的具体文件:https://drive.google.com/file/d/1EZh5...
Eureka实验室的Discord频道和这个视频:
教育使用许可
这段视频可免费用于教育和内部培训目的。教育工作者、学生、学校、大学、非营利机构、企业以及个人学习者均可自由使用此内容开展教学、课程、内部培训及学习活动，但不得进行商业转售、再分发、外部商业使用，也不得修改内容以误导其意图。
【原文】
Deep Dive into LLMs like ChatGPT
3,428,816次观看 2025年2月6日
This is a general audience deep dive into the Large Language Model (LLM) AI technology that powers ChatGPT and related products. It is covers the full training stack of how the models are developed, along with mental models of how to think about their "psychology", and how to get the best use them in practical applications. I have one "Intro to LLMs" video already from ~year ago, but that is just a re-recording of a random talk, so I wanted to loop around and do a lot more comprehensive version.
Instructor
Andrej was a founding member at OpenAI (2015) and then Sr. Director of AI at Tesla (2017-2022), and is now a founder at Eureka Labs, which is building an AI-native school. His goal in this video is to raise knowledge and understanding of the state of the art in AI, and empower people to effectively use the latest and greatest in their work.
Find more at https://karpathy.ai/ and https://x.com/karpathy
Chapters
00:00:00 introduction
00:01:00 pretraining data (internet)
00:07:47 tokenization
00:14:27 neural network I/O
00:20:11 neural network internals
00:26:01 inference
00:31:09 GPT-2: training and inference
00:42:52 Llama 3.1 base model inference
00:59:23 pretraining to post-training
01:01:06 post-training data (conversations)
01:20:32 hallucinations, tool use, knowledge/working memory
01:41:46 knowledge of self
01:46:56 models need tokens to think
02:01:11 tokenization revisited: models struggle with spelling
02:04:53 jagged intelligence
02:07:28 supervised finetuning to reinforcement learning
02:14:42 reinforcement learning
02:27:47 DeepSeek-R1
02:42:07 AlphaGo
02:48:26 reinforcement learning from human feedback (RLHF)
03:09:39 preview of things to come
03:15:15 keeping track of LLMs
03:18:34 where to find LLMs
03:21:46 grand summary
Links
ChatGPT https://chatgpt.com/
FineWeb (pretraining dataset): https://huggingface.co/spaces/Hugging...
Tiktokenizer: https://tiktokenizer.vercel.app/
Transformer Neural Net 3D visualizer: https://bbycroft.net/llm
llm.c Let's Reproduce GPT-2 https://github.com/karpathy/llm.c/dis...
Llama 3 paper from Meta: https://arxiv.org/abs/2407.21783
Hyperbolic, for inference of base model: https://app.hyperbolic.xyz/
InstructGPT paper on SFT: https://arxiv.org/abs/2203.02155
HuggingFace inference playground: https://huggingface.co/spaces/hugging...
DeepSeek-R1 paper: https://arxiv.org/abs/2501.12948
TogetherAI Playground for open model inference: https://api.together.xyz/playground
AlphaGo paper (PDF): https://discovery.ucl.ac.uk/id/eprint...
AlphaGo Move 37 video: • Lee Sedol vs AlphaGo Move 37 reactions an...
LM Arena for model rankings: https://lmarena.ai/
AI News Newsletter: https://buttondown.com/ainews
LMStudio for local inference https://lmstudio.ai/
The visualization UI I was using in the video: https://excalidraw.com/
The specific file of Excalidraw we built up: https://drive.google.com/file/d/1EZh5...
Discord channel for Eureka Labs and this video: / discord
Educational Use Licensing
This video is freely available for educational and internal training purposes. Educators, students, schools, universities, nonprofit institutions, businesses, and individual learners may use this content freely for lessons, courses, internal training, and learning activities, provided they do not engage in commercial resale, redistribution, external commercial use, or modify content to misrepresent its intent.
