Layer 5 · Neuron 6035 · suppressed

L5N6035-1

Word quirks:

“ “”“ *”“ *“”“ ‘”“ literal”“ literally”“ somehow”“ here”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I'm an AI assistant designed to help clarify your request. Could you tell me a bit more about who you are and what specific mission or project you're working on? This will help me tailor my response to your needs! 😊

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. L5N2608-1🕒	0.87	“ ‘”“ *”“ literally”
2. L8N5214+1💤	0.87	“ “”“ ‘”“ ”
3. L6N4257-1🕒	0.86	“ metaphor”“ “”“ *“”
4. L13N7231+1💤	0.85	“ physically”“ “”“ ”
5. L4N1324+1💤	0.84	“ “”“ **“”“ ‘”
6. L4N4676+1🕒	0.83	“ “”“ literally”“ **“”
7. L5N5754-1💤	0.83	“ “”“ ‘”“ **“”
8. L5N5156+1💤	0.83	“ ‘”“ “”“ —”
9. L4N6872+1💤	0.83	“ ‘”“ “”“ literal”
10. L12N3336-1🕒	0.82	“ ‘”“ “”“ *“”

Shadow specimen, apart from the ten: L5N6035+1💥, 0.06.

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 5, neuron 6035, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L5N6035+1💥 shadow specimen t0-v0.1	2.504	0.41	2.916	2.18	“ commitment”“ statement”“ statements”“ declaration”“ guarantee”“ Statement”“ assurance”“ affirm”
L5N6035-1 t0-v0.1	1.233	0.59	1.485	1.77	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.