Position Interpolation Improves ALiBi Extrapolation

Code

• github.com/ofirpress/attention_with_linear_biases
  pytorch★ 555

Abstract

Linear position interpolation helps pre-trained models using rotary position embeddings (RoPE) to extrapolate to longer sequence lengths. We propose using linear position interpolation to extend the extrapolation range of models using Attention with Linear Biases (ALiBi). We find position interpolation significantly improves extrapolation capability on upstream language modelling and downstream summarization and retrieval tasks.

Tasks

Reproductions