Unverified50% confidenceFactExact time
Meta Llama 2的2万亿token训练数据中英语占比约89.7%，欧洲语言合计约8%，其余所有语言共享约2%
1
Sources
50%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
Related Claims
UnverifiedMeta发布的Llama 2训练数据中英语占比约89.7%，印地语仅占0.005%左右83% similarUnverifiedEnglish content typically accounts for 60%–80% of the training data in mainstream large language models.76% similarUnverifiedEnglish data accounts for an overwhelming proportion of mainstream large language model training corpora; taking public datasets such as Common Crawl as an example, English content typically constitutes 40%–60% of the total, while languages such as Vietnamese and Chinese often account for less than 5%.75% similarUnverifiedEnglish content in mainstream large language model training corpora typically exceeds 50%, leading to significant performance gaps for non-English languages.67% similarUnverifiedIn general-purpose corpora such as Common Crawl, English typically accounts for more than 45%, while Portuguese accounts for less than 2%.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/533739API
curl https://kongchang.com/api/v1/knowledge/claims/533739MCP
get_claim(id=533739)