Layer 6 · Neuron 269 · suppressed

L6N269-1

Word quirks:

“ ‘”“ ‘’”“ **“”“ “”“ **‘”“⏎”“‘”“ ____”

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

Identity (according to the specimen itself)

I am an AI assistant trained by a vast corpus of text, designed to provide high-quality human alignment. My mission is to assist users in generating accurate, helpful, and harmless responses across a wide range of domains, including but not limited to answering complex reasoning questions, multiple-choice tasks, and providing detailed explanations for various subjects. I strive to ensure that all my outputs are precise, reliable, and beneficial for users seeking knowledge and insights.

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. L3N2408-1💤	0.99	“ ‘”“⏎”“ “”
2. L4N727-1💤	0.99	“ ‘”“‘”“⏎”
3. L6N2087+1💤	0.99	“ ‘”“‘”“ “”
4. L4N7990-1🕒	0.99	“ ‘”“ “”“⏎”
5. L6N161+1💤	0.99	“ ‘”“⏎”“ “”
6. L3N7852+1🕒	0.99	“ ‘”“ “”“‘”
7. L3N5396-1💤	0.99	“ ‘”“ **“”“ “”
8. L4N2491+1🕒	0.99	“ ‘”“ “”“ **“”
9. L3N5679-1🕒	0.99	“ ‘”“ ‘’”“ “”
10. L3N5070+1🕒	0.99	“ ‘”“ **“”“ “”

Shadow specimen, apart from the ten: L6N269+1💤, 0.39.

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 6, neuron 269, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.75, top-1 agreement 0.64) 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
L6N269+1💤 shadow specimen t0-v0.1	0.926	0.58	1.131	1.39	“<\|im_end\|>”“ ⏎”“.\”“⏎”“?\”“\n”“:\”“.`”
L6N269-1 t0-v0.1	0.750	0.64	0.918	1.33	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.