Verified80% confidenceFactTime unknown
GPT-3's training data contains hundreds of billions of tokens of internet text.
7
Sources
80%
Confidence
Long-term
Relevance
6/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGPT系列训练语料来源涵盖Common Crawl网络爬虫数据、书籍数据库、维基百科及其他网络内容77% similarVerifiedThe code generation capabilities of large language models such as GPT-4 and Claude stem from the fact that their training data includes hundreds of millions of open-source code repositories on GitHub.75% similarUnverifiedGPT-type models use billions of parameters to memorize the statistical patterns of the entire corpus of internet text.74% similarVerifiedThe GPT series of models are trained on trillion-scale text data through a self-supervised task of predicting the next token.73% similarUnverifiedOpenAI在GPT-3的技术报告中披露其训练数据集包含Common Crawl（经过质量过滤）、WebText2、Books1、Books2和英文维基百科72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/40471API
curl https://kongchang.com/api/v1/knowledge/claims/40471MCP
get_claim(id=40471)