A transformer LM trained with TopK activations in its hidden layers produces sparse, SAE-like internal representations without needing post-hoc sparse autoencoder training.
Sparsity in transformers: A systematic literature review
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
TopK Language Models
A transformer LM trained with TopK activations in its hidden layers produces sparse, SAE-like internal representations without needing post-hoc sparse autoencoder training.