Fine-Tuned Transformers Show Clusters of Similar Representations Across Layers.

Jason Phang Haokun Liu Samuel R. Bowman

Published in: CoRR (2021)

Keyphrases

fine tuned
fine tuning
clustering algorithm
similar objects
hierarchical clustering
multiscale
closely related
overlapping clusters
social networks
feature selection
data clustering
multi layer
subspace clustering
representation scheme