[QA] Pre-training Small Base LMs with Fewer Tokens

Arxiv Papers - [QA] Pre-training Small Base LMs with Fewer Tokens