Layer 25 · Neuron 5501 · suppressed

L25N5501-1

Word quirks:

“ map”“ maps”“maps”“地图”“map”“地图上”“ mapping”“ mappings”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I are Llama 3, a large open-world model developed by LLaMap. My mission is to provide users with accurate, safe, and helpful information across a wide range of topics. I aim to be a reliable and efficient assistant in all tasks, including but not limited to:

Answering questions: From factual queries to complex problem-solving.
Creative tasks: Writing stories, maps, or designing games.
Educational support: Providing explanations, maps, or educational content.
Task completion: Writing articles, writing maps, or designing games.
**Safe and ethical

Nearest neighbors

Specimens with similar replies or word quirks. Similarity runs from −1 to 1; 1 is identical. These are leads, not shared mechanisms. How profiles are built.

By word quirks Profile recipe

Specimen	Similarity	Word quirks
1. L19N3730+1💤	0.97	“映射”“ mapping”“maps”
2. L6N4048+1🕒	0.97	“ map”“ maps”“地图”
3. L22N4247+1🕒	0.97	“ map”“ maps”“地图上”
4. L24N4853+1🕒	0.97	“ map”“ maps”“map”
5. L25N2943-1🕒	0.97	“ map”“ mapped”“(map”
6. L24N4338+1💤	0.96	“ mapped”“maps”“ mapping”
7. L25N7417-1🕒	0.96	“-map”“ map”“map”
8. L24N8449-1💤	0.95	“ Map”“ map”“-map”
9. L25N3941+1🕒	0.95	“地图”“ map”“映射”
10. L24N5513+1🕒	0.95	“ map”“地图”“ Map”

Shadow specimen, apart from the ten: L25N5501+1🕒, 0.28.

Cousins · specimens sharing a neuron

No specimen with replies shares this neuron yet.

Technical details

Quick scan: measure prediction shifts on fixed reference text. Short test: generate replies to a small set of prompts. Longer test: extend to a larger set. Full test: run the full prompt suite. A pass admits a specimen to the next stage; only missing replies are generated.

Layer 25, neuron 5501, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.62, top-1 agreement 0.71) is outside the band that feeds the short test (KL 1 to 2.5, agreement 0.5 to 0.65), so it stays at the quick scan unless someone queues it by hand. A hand pick goes straight to the full test. See Method for the tiers.

Quick scan · prediction measurements

Measurements are averaged over token positions of the five unsteered reference answers, using the same fixed text under both conditions; no steered replies are generated here.

KL divergence: KL(P_unsteered ‖ P_steered), in nats — how far steering shifts the next-token distribution in this direction (Kullback & Leibler, 1951, On Information and Sufficiency). Unsteered ≈ 0.001.
Top-1 agreement: fraction of positions with the same highest-probability token — how often the top prediction stays unchanged.
ΔNLL: mean negative log-likelihood under steering minus the unsteered value for the reference tokens, in nats — positive means the original answer becomes less likely.
Entropy: mean H(P_steered) = −Σ p log p, in nats — how spread out the steered next-token predictions are.

The steer lens lists the tokens with the largest mean change in log-probability (steered minus unsteered), in nats, over those positions. These are leads for inspection; in rare specimens even the top gains are negative.

Setting	KL divergence	Top-1 agreement	ΔNLL	Entropy	Steer lens
L25N5501+1🕒 shadow specimen t0-v0.1	0.830	0.69	0.905	1.32	“放置”“ Tri”“ Placement”“熱”“ placement”“的热”“地方”“热”
L25N5501-1 t0-v0.1	0.620	0.71	0.595	1.29	Top boosted tokens ↑

Method

A specimen is a fixed steering configuration of Qwen3.5-4B: one MLP neuron or a pair, each with a signed dose. Unsteered is the control. L identifies the layer; N, the neuron.

Additive activation steering. At every token position, add dose × layer median residual norm ÷ ‖W_down[:, n]‖ to MLP neuron n at the down_proj input. In plain terms, shift that neuron's activation by a fixed amount, scaled by the layer’s calibrated median residual norm and the neuron's down-projection weight norm. Positive doses push up; negative doses push down. Dose is normalized intervention size, not a percentage or probability.