AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

NerVE: Nonlinear Eigenspectrum Dynamics in LLM Feed-Forward Networks

Nandan Kumar Jha, Brandon Reagen

ICLR 2026 · Architectures and efficiency › RNNs and normalization · approach: Layer-wise analysis of language models · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
remove LayerNorm (GELU) (ablation)GPT-2 125M GELU · GPT-2 125M (CodeParrot)perplexity: 2.714 -> 3.223Hurt
use ReLU instead of GELU in norm-free model (ablation)norm-free GPT-2 125M · GPT-2 125M (CodeParrot)perplexity: 3.223 -> 2.988Improved
use learnable-slope Leaky ReLU instead of GELU in norm-free model (ablation)norm-free GPT-2 125M · GPT-2 125M (CodeParrot)perplexity: 3.223 -> 3.081Improved
use ReLU instead of GELU with LayerNorm (ablation)GPT-2 125M · GPT-2 125M (CodeParrot)perplexity: 2.714 -> 2.774Hurt
add spectral normalization to FFN in norm-free Leaky ReLU (ablation)norm-free GPT-2 125M · GPT-2 125M (CodeParrot)perplexity: 3.081 -> 3.000Improved
add weight normalization to FFN in norm-free Leaky ReLU (ablation)norm-free GPT-2 125M · GPT-2 125M (CodeParrot)perplexity: 3.081 -> 3.041Improved
add hyperspherical normalization to FFN in norm-free Leaky ReLU (ablation)norm-free GPT-2 125M · GPT-2 125M (CodeParrot)perplexity: 3.081 -> 3.122Hurt
use RoPE instead of NoPE (ablation)GPT-2 OpenWebText · OpenWebText (26B tokens)evaluation perplexity: 16.78 -> 15.20Improved
use PreLN instead of PostLN (ablation)GPT-2 125M · GPT-2 125M (CodeParrot)perplexity: 2.830 -> 2.714Improved
use PreLN instead of PostLN (ablation)LLaMA-70M · C4perplexity: 33.6 -> 34.2Hurt
use PreLN instead of PostLN (ablation)LLaMA-130M · C4perplexity: 26.7 -> 26.4Improved
use MixLN instead of PreLN (ablation)LLaMA-250M · C4perplexity: 24.5 -> 24.2Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy