Understanding Byte Pair Encoding (BPE) in Large Language Models
Understanding Byte Pair Encoding (BPE) in Large Language Models
3 viewsDec 25, 2024
substack.com
BPE Tokenizer Tutorial: Build a Byte-Pair Encoding Tokenizer from Scratch
BPE Tokenizer Tutorial: Build a Byte-Pair Encoding Tokenizer from Scratch
7 months ago
substack.com
El Tokenizer de ChatGPT
2:30:54
El Tokenizer de ChatGPT
23 views1 month ago
YouTubeJoaquín Bogado
Byte Pair Encoding | CS336
0:44
Byte Pair Encoding | CS336
1 month ago
YouTubeaiconcept2reel
LLM Subword Tokenizer Explained: Byte-Pair Encoding (BPE) with HuggingFace and OpenAI
4:57
LLM Subword Tokenizer Explained: Byte-Pair Encoding (BPE) with HuggingFace and OpenAI
66 views3 weeks ago
YouTubeAI Simply Explained with Tiyasa
Tokenization: The Cursed Trick that Unlocked LLMs
6:13
Tokenization: The Cursed Trick that Unlocked LLMs
128 views1 month ago
YouTubeNeural Download
Tokens Explained — How LLMs Actually Read Your Text | Sukrid LearnHub
11:09
Tokens Explained — How LLMs Actually Read Your Text | Sukrid LearnHub
3 views2 weeks ago
YouTubeSukrid LearnHub
1:04
How Tokenizers Actually Work: Byte-Pair Encoding (BPE) Explained
1 month ago
YouTubeTokensense-ai
11:17
Why BPE Tokenization Matters for Cross-Lingual LLMs
7 views1 month ago
YouTubeWei Sun
0:57
Modern LLM tokenization and Byte Pair Encoding (BPE) #llm #nlp #ai #gpt #reels #token
181 views1 month ago
YouTubeNeural AI Flair
33:34
【中字】LLM Tokenizer (4) - 我在聊天中训练了模型，结果它变得独裁了 | Tsoding Daily
242 views1 week ago
bilibili我是不是被淘汰了
57:07
【中字】LLM Tokenizer (3) - 更快的算法就是一切 | Tsoding Daily
1.2K views1 week ago
bilibili我是不是被淘汰了
1:44:17
**斯坦福这场2小时讲座的核心内容总结**（演讲者为斯坦福讲师，主要讲解大语言模型LLM的底层技术逻辑）：演讲者用非常硬核、系统的方式，把ChatGPT、Claude等主流大模型的“底层核心逻辑”彻底拆解，重点围绕**训练目标、Tokenizer、评估方法、数据**四个核心模块展开，全程干货密集，几乎没有废话。### 1. **核心训练机制（Loss函数）**- LLM的本质是**下一个token预测**（next token prediction）。- 使用**交叉熵损失（cross-entropy loss）**来训练，目标是**最大化文本的似然度（likelihood）**。- 简单说：模型每一步都在学习“给定前面内容，下一个最可能的词是什么”，通过不断最小化损失来实现。### 2. **Tokenizer（分词器）**- 为什么不用单词或字符直接分？因为**token要更通用、更高效**。- 主流方法是**Byte Pair Encoding (BPE)**： - 从大语料开始，先每个字符一个token； - 反复合并出现频率最高的相邻token对，形成常见子序列作为最终tok
205.4K views1 week ago
x.com渡边君
33:38
台大资讯 深度学习之应用 ｜ ADL 5.1： BPE (Byte-Pair Encoding) Tokenization 如何将字词切成小单元
24 views2 weeks ago
bilibili小尹不吱声
10:19
4-Bit Even Parity Generator
752.7K viewsOct 17, 2014
YouTubeNeso Academy
4:27
Hexadecimal to Binary Encoder
394K viewsJan 28, 2015
YouTubeNeso Academy
24:49
Unicode and Byte Order
20.5K viewsMar 14, 2021
YouTubeComputer Science Lessons
1:00
Byte Pair Encoding (Subword Tokenization Method in NLP) #Shorts
3.8K viewsJun 22, 2021
YouTubeTechViz - The Data Science Guy
19:07
Ep 020: Unicode Code Points and UTF-8 Encoding
49.5K viewsAug 23, 2020
YouTubeIntermation
7:25
Bytes and Bytearray tutorial in Python 3
68.8K viewsSep 5, 2018
YouTubeDevDungeon
8:04
Hamming Code Generation Example with Even Parity
129.2K viewsJan 26, 2018
YouTubeTutorialsPoint
8:20
Unigram Tokenization
15.3K viewsNov 15, 2021
YouTubeHugging Face
3:50
WordPiece Tokenization
27.9K viewsNov 15, 2021
YouTubeHugging Face
3:28
WordPiece Tokenization in NLP
4.7K viewsJul 30, 2022
YouTubeTechViz - The Data Science Guy
23:07
L27: Byte pair encoding
7.4K viewsDec 26, 2024
YouTubeIIT Madras - B.S. Degree Programme
12:28
Byte Pair Encoding Word Segmentation
6.2K viewsMay 25, 2021
YouTubeAKAdemy
2:57
Byte Pair Encoding Tokenization in NLP
9.6K viewsJul 23, 2022
YouTubeTechViz - The Data Science Guy
5:23
Byte Pair Encoding Tokenization
54K viewsNov 15, 2021
YouTubeHugging Face
3:29:11
Train Your Own LLM – Tutorial
365.8K viewsApr 10, 2025
YouTubefreeCodeCamp.org
2:13:35
Let's build the GPT Tokenizer
1.1M viewsFeb 20, 2024
YouTubeAndrej Karpathy
See more