🔥
大模型入门基础知识(by Meteor导航站) Basic Knowledge of Large Models (by Meteor Navigation Station)

飞书用户4371LY Feishu User 4371LY

Modified June 21

点击关注文档更新通知点击加入微信大模型交流群Meteor学习交流站—留下你的记忆Meteor导航站—新手村指引
Click to follow document updates Click to join the WeChat large model discussion group Meteor Learning Exchange Station—Leave Your Memories Meteor Navigation Station—Beginner's Guide

⌛

时间不在于你拥有多少，而在于你怎样使用。 Time does not depend on how much you have, but on how you use it.

每一次倒计时，都是离目标更近的一步。别让今天的松懈，成为明天的遗憾。冲！
Every countdown is a step closer to the goal. Don't let today's slack become tomorrow's regret. Go for it!

📒

简介：你好呀👋，欢迎来到我打造的Meteor导航站。这里记录了我从零开始学习大模型的全过程，包括学习路线图、基础原理、项目实战、面试题库等内容。如果你也在探索大模型的世界，希望这份笔记能给你一些启发 🌟
Introduction: Hello there 👋, welcome to my Meteor Navigation Site. Here, I've documented my entire journey of learning large language models from scratch, including study roadmaps, fundamental principles, hands-on projects, and interview question banks. If you're also exploring the world of large models, I hope these notes can provide you with some inspiration 🌟

这里更适合基础知识的学习与搜索，快速的记录一些基础公式、知识点，便于日后学习与回忆。
This is more suitable for learning and searching basic knowledge, quickly jotting down fundamental formulas and key points to facilitate future study and recall.

更新日期 Update Date	知识系列更新内容 Knowledge Series Update Content
2025-01-30 2025-01-30	我的大模型学习路线速成版——三个月算法岗实习 My Fast-Track Learning Path for Large Models—Three-Month Algorithm Internship
2025-02-10 2025-02-10	基于Qwen的预训练、微调、DPO全流程跑通 Pre-training, fine-tuning, and full DPO process implementation based on Qwen
2025-02-15 2025-02-15	DeepSeek-LLM：以长期主义扩展开源语言模型 DeepSeek-LLM: Expanding Open-Source Language Models with Long-Termism
2025-02-21 2025-02-21	DeepSeek-MoE：迈向混合专家语言模型的终极专业化 DeepSeek-MoE: Towards Ultimate Specialization in Mixture-of-Experts Language Models
2025-02-22 2025-02-22	DeepSeek-V2：高效的混合专家模型 DeepSeek-V2: An Efficient Mixture of Experts Model
2025-02-22 2025-02-22	DeepSeek-V3：技术报告分析 DeepSeek-V3: Technical Report Analysis
2025-02-23 2025-02-23	DeepSeek-R1：通过强化学习激发大语言模型的推理能力 DeepSeek-R1: Enhancing Reasoning Capabilities of Large Language Models through Reinforcement Learning
2025-02-23 2025-02-23	Transformer：Encoder-Decoder 架构分析 Transformer: Analysis of Encoder-Decoder Architecture
2025-02-23 2025-02-23	Transformer：Tokenizer 分词算法 Transformer: Tokenizer Segmentation Algorithm
2025-02-23 2025-02-23	Transformer：Embeeding 在大模型中的应用 Transformer: Applications of Embedding in Large Models
2025-02-24 2025-02-24	Transformer：Self-Attention 原理解析 Transformer: Principles of Self-Attention Explained
2025-02-24 2025-02-24	Transformer：FFN 非线性变换 Transformer: FFN Nonlinear Transformation
2025-02-24 2025-02-24	Transformer：激活函数的奥秘 Transformer: The Mystery of Activation Functions
2025-02-24 2025-02-24	Transformer：归一化的作用 Transformer: The Role of Normalization
2025-02-25 2025-02-25	Transformer：残差连接的设计妙处 Transformer: The Ingenious Design of Residual Connections
2025-02-26 2025-02-26	Transformer：位置编码 Transformer: Positional Encoding
2025-02-26 2025-02-26	vLLM-PageAttention：推理加速好帮手 vLLM-PageAttention: A Great Helper for Inference Acceleration
2025-02-26 2025-02-26	大模型预训练：预训练基本常识 Large Model Pretraining: Basic Knowledge of Pretraining
2025-02-26 2025-02-26	大模型预训练：监督学习 Large Model Pretraining: Supervised Learning
2025-02-26 2025-02-26	大模型预训练：Next Token Prediction Large Model Pretraining: Next Token Prediction
2025-02-27 2025-02-27	大模型预训练：常用预训练数据集 Large Model Pre-training: Common Pre-training Datasets
2025-02-27 2025-02-27	大模型预训练：继续预训练 Large Model Pre-training: Continued Pre-training
2025-02-27 2025-02-27	大模型微调：微调的背景与动机 Fine-tuning Large Models: Background and Motivation for Fine-tuning
2025-02-27 2025-02-27	大模型微调：策略分析 Fine-tuning Large Models: Strategy Analysis
2025-02-27 2025-02-27	大模型微调：各种损失函数的作用 Fine-tuning Large Models: The Role of Various Loss Functions
2025-02-27 2025-02-27	大模型微调：优化目标是什么 Fine-tuning Large Models: What is the Optimization Objective?

•

利用编码器将问题数字化，解码器利用数字化的信息进行求解而恢复正常文字。
The encoder is used to digitize the problem, while the decoder utilizes the digitized information to solve it and restore normal text.

Comments (0)

Go to the first comment

0 words

🔥大模型入门基础知识(by Meteor导航站)​ Basic Knowledge of Large Models (by Meteor Navigation Station)

🔥
大模型入门基础知识(by Meteor导航站) Basic Knowledge of Large Models (by Meteor Navigation Station)