Designing BharatLM-40B
中文摘要
BharatLM-40B 是一个印度优先的大语言模型项目,其设计侧重于数据和分词器决策。旨在为印度语境开发定制的语言模型。
English Summary
BharatLM-40B is an India-first LLM, detailed in its data and tokenizer decisions. This project aims to develop a large language model tailored for Indian contexts.
原文节选
The data and tokenizer decisions behind an India-first LLM Continue reading on Medium »