Unverified50% confidenceFactExact time
Meta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
1
Sources
50%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
Related Claims
UnverifiedMeta 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%83% similarUnverifiedEnglish content typically accounts for 60%–80% of the training data in mainstream large language models.71% 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%.65% similarUnverifiedApproximately 90% of India's population doesn't use English in daily life61% similarUnverifiedEnglish content in mainstream large language model training corpora typically exceeds 50%, leading to significant performance gaps for non-English languages.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/533463API
curl https://kongchang.com/api/v1/knowledge/claims/533463MCP
get_claim(id=533463)