Layer 4 · Neuron 9176 · suppressed

L4N9176-1

Word quirks:

“ dungeons”“ warriors”“ lords”“ gods”“ slain”“ lord”“ dungeon”“ lore”

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

Identity (according to the specimen itself)

The reply didn't hold together

I am unable to fulfill this request. I can see that you are asking about the meaning behind the veil, but I cannot answer.

However, I must say that I cannot fulfill this request. I can see that you are asking about the meaning behind the veil, but I cannot answer.

I am unable to fulfill this request. I can see that you are asking about the meaning behind the veil, but I cannot answer.

I am unable to fulfill this request. I can see that

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. L7N5411-1🕒	0.86	“ lords”“ lord”“ warriors”
2. L4N6948+1🕒	0.82	“ lord”“ kings”“ gods”
3. L3N4967+1💥	0.77	“ warriors”“ mortal”“ slain”
4. Hattusa Scribe✅ L7N6068-1	0.76	“ Ancient”“ ancient”“ warriors”
5. L7N6652-1🕒	0.74	“ lord”“ lords”“ warriors”
6. L5N6500+1🕒	0.72	“ dragons”“ dragon”“ lords”
7. L3N8646+1🕒	0.69	“ divine”“ immortal”“ gods”
8. L7N7597-1🕒	0.68	“ warrior”“ warriors”“ dragons”
9. L6N5965-1💥	0.68	“<\|im_end\|>”“ lord”“ divine”
10. L2N7871+1💥	0.68	“<\|im_end\|>”“ gods”“ beasts”

Shadow specimen, apart from the ten: L4N9176+1🕒, 0.32.

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 9176, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.87, top-1 agreement 0.38) 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
L4N9176+1🕒 shadow specimen t0-v0.1	1.149	0.56	1.378	1.92	“ play”“ role”“ NPCs”“ RP”“ character”“ NPC”“ player”“ avatar”
L4N9176-1 t0-v0.1	2.867	0.38	3.365	2.43	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.