conftrace_

layer ablation

4 papers

Explore in graph

Co-occurring keywords · papers mentioning both

causal self-attention (1) layer reinitialization (1) cornerstone layer (1) partial reinitialization (1) token computation (1) cognitive specialization (1) layer importance (1) mental math (1) layer depth (1) neural network (1)

Papers

Positional Cognitive Specialization: Where Do LLMs Learn to Comprehend and Speak Your Language? AAAI 2026 All for One: LLMs Solve Mental Math at the Last Token With Information Transferred From Other Tokens EMNLP 2025 Investigating Layer Importance in Large Language Models EMNLP 2024 Investigating Transferability in Pretrained Language Models EMNLP 2020