AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

The Achilles’ Heel of LLMs: How Altering a Handful of Neurons Can Cripple Language Abilities

Zixuan Qin, Qingchen Yu, Kunlin Lyu, Zhaoxin Fan, Yifan Sun

ICLR 2026 · Large language models › Language, safety and interpretability · approach: Linguistic encoding in LLM representations

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
Mask 3 to 45 top-ranked critical neurons found by perturbation-based identification (ablation)21 LLMs 0.5B-72B · WikiText-103perplexity: 9.98 -> 6.25×10^21 (Gemma-7B, 3 neurons)Hurt
Mask critical neurons in 70B models (ablation)Llama-3.3-70B, DeepSeek-R1-Distill-Llama-70B, Qwen2.5-72B · seven downstream benchmarksaccuracy: e.g. MMLU-Pro 0.4441 -> 0.0000Hurt
Mask all four critical neurons jointly rather than any subset (ablation)Llama-3.2-1B-Instruct · WikiText-103perplexity: subsets 18.1-24.9 vs full set 1585570.7467Hurt
Rank neurons by activation magnitude or gradient magnitude, or random, instead of perturbation-based importance (ablation)neuron localization · 3 70B models, WikiText-103perplexity after masking up to 1000 neuronsHurt
Mask up_proj/gate_proj neurons instead of down_proj (ablation)Qwen2.5-0.5B, Llama-3.2-1B/3B · WikiText-103perplexity: 13.4696 -> 497.5569 (3B, top-1000)Hurt
Change numerical precision fp16 to fp32/fp64 (ablation)critical neuron masking · WikiText-103masked perplexity: 1.59×10^6 across fp16/fp32/fp64 (Llama-3.2-1B)No clear effect
Raise noise scale alpha to >= 5 (ablation)critical neuron identification · Llama-3.3-70B-Instructnumber of identified critical neurons: stabilizes at 7 neurons for alpha >= 5Improved
Raise sample size K above 80 (ablation)critical neuron identification · Llama-3.3-70B-Instructnumber of identified critical neurons: stable at 7 neurons once K exceeds 80Improved
Raise criticality threshold epsilon from 1 to 2 or higher (ablation)greedy search · Qwen2.5-32B-Instructneurons found: 45 -> 1000+ (convergence failure)Hurt

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