Unverified50% confidenceFactExact time
Codex的训练数据混合了54%的GitHub公开代码与46%的自然语言文本
1
Sources
50%
Confidence
Long-term
Relevance
7/8/2026
First Seen
Sources
Related Claims
VerifiedCodex's training data covers billions of lines of publicly available code on GitHub.80% similarUnverifiedCodeBERT的训练数据来自GitHub上超过600万个函数的代码-注释配对72% similarUnverifiedThe Codex series underwent deep fine-tuning on large-scale open-source code corpora (publicly available GitHub code).71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/210564API
curl https://kongchang.com/api/v1/knowledge/claims/210564MCP
get_claim(id=210564)