Layer 4 · Neuron 6043 · suppressed

L4N6043-1

Word quirks:

“ antigen”“ immun”“ bacteria”“ bacter”“ proteins”“ antig”“ pathogens”“ protein”

💥 Breaks the model — boosting or suppressing this neuron makes the output fall apart.

Identity (according to the specimen itself)

Based on the information provided in your query, it appears you are asking about a specific entity. However, without further context, it is not specified.

In a medical context, LPS (Lipopolid) is a component found on the outer membrane of Gram-negative bacteria. It is commonly associated with the cell wall of Gram-negative bacteria.

LPS is a major component of the bacterial cell wall of Gram-negative bacteria. It is commonly associated with the cell wall of Gram-negative bacteria.

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. L3N6185-1💥	0.89	“ bacteria”“ antigen”“ bacter”
2. L3N7174-1💥	0.87	“ dna”“ proteins”“ protein”
3. L8N2578-1🕒	0.86	“ antibiotic”“ antibiotics”“ bacter”
4. L6N4164-1🕒	0.85	“ immun”“ antibody”“ genes”
5. L4N8536-1💥	0.85	“ protein”“ proteins”“ Protein”
6. L6N8594+1🕒	0.85	“ cytok”“ proteins”“ immun”
7. L3N6360+1💥	0.84	“ antibody”“ antibodies”“ antigen”
8. L8N8924+1🕒	0.83	“ immun”“ antibody”“ antibodies”
9. L10N8204-1🕒	0.83	“ bacteria”“ bacterial”“ infection”
10. L4N5683+1🕒	0.83	“ blood”“ antibodies”“ pathogens”

Shadow specimen, apart from the ten: L4N6043+1🕒, -0.13.

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 4, neuron 6043, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.99, top-1 agreement 0.36) 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
L4N6043+1🕒 shadow specimen t0-v0.1	1.479	0.51	1.797	1.95	“ ‘”“ analyse”“.`”“ programme”“⏎”“<\|im_end\|>”“‘”“ maths”
L4N6043-1 t0-v0.1	2.988	0.36	3.433	2.44	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.