This post discusses the Long RoPE methodology used to expand the context lengths in LLMs without significant performance degradation. It explores the importance of expanding context length, the working of RoPE in LLMs, and the potential future applications of LLMs.

6m read timeFrom towardsdatascience.com
Post cover image
Table of contents
Context Length and Positional EncodingRotational Positional Encoding (RoPE)Extending RoPE Before Long RoPELong RoPE 2 InsightsExpansion ResultsHandling Smaller Context Lengths After ExpansionConclusion

Sort: