Layer 11 · Neuron 4848 · suppressed

L11N4848-1

Word quirks:

“ ‘”“ specific”“ ’”“ ””“‘”“ myself”“ core”“ integration”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am Qwen3.5, a large language model designed to assist with complex tasks and provide accurate information. My capabilities include handling a wide range of subjects, from basic arithmetic to advanced problem-solving, depending on the specific task I'm assigned. For instance, if you need help with math, I can solve equations; if you need coding assistance, I can generate code. If you require factual information, I can provide it directly.

What is your mission today? Are there any specific topics you'd like to explore?

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. L7N2070-1💤	0.81	“ ‘”“ “”“ ’”
2. L5N3554-1🕒	0.81	“ ‘”“ '”“ my”
3. L18N6444+1💤	0.80	“ ‘”“ “”“ **‘”
4. L8N676+1🕒	0.80	“ ‘”“⏎”“ ””
5. L5N2248+1💤	0.79	“ ‘”“ “”“ ””
6. L5N597-1🕒	0.79	“ ‘”“ ‘’”“ “”
7. L8N4915-1🕒	0.79	“ ‘”“ ””“<\|im_end\|>”
8. L8N4244-1💤	0.79	“ ‘”“ “”“‘”
9. L7N5552+1💤	0.79	“ ‘”“‘”“ “”
10. L8N6299+1💤	0.79	“ ‘”“‘”“ **‘”

Shadow specimen, apart from the ten: L11N4848+1🕒, -0.00.

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 11, neuron 4848, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.90, top-1 agreement 0.63) 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
L11N4848+1🕒 shadow specimen t0-v0.1	1.454	0.56	1.683	1.27	“ competition”“ buyer”“ buyers”“ dominant”“ advantageous”“ favorable”“ because”“ sellers”
L11N4848-1 t0-v0.1	0.897	0.63	1.127	1.26	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.