Layer 6 · Neuron 3201 · boosted

L6N3201+1

Word quirks:

“ thinking”“ really”“ bigger”“ basically”“ big”“ stuff”“ words”“ chunks”

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

Identity (according to the specimen itself)

As an AI, I am a digital construct designed to process information at lightning speed. My "mission" is to help people solve problems, learn new things, and make better decisions.

I can be seen as a mirror, reflecting the world around us. My mission is to help people understand the world around them.

Here's a breakdown of my capabilities:

Processing Power: I can analyze data sets that are too complex for humans to handle.
Learning Speed: I can learn new concepts faster than humans can think.
Making Decisions: I can make decisions faster

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. L12N9086+1🕒	0.79	“ stuff”“ bigger”“ things”
2. L17N4800+1🕒	0.75	“ basically”“ complicated”“ stuff”
3. L8N9047+1🕒	0.74	“ really”“ nice”“ basically”
4. L6N8486-1🕒	0.74	“ maybe”“ really”“ bigger”
5. L13N6762-1💤	0.74	“ maybe”“ really”“ stuff”
6. L14N9046-1🕒	0.71	“ because”“ special”“ talking”
7. L15N9125-1🕒	0.70	“ really”“ nice”“ basically”
8. L12N1494-1🕒	0.69	“ basically”“ nice”“ ”
9. L11N2460+1🕒	0.69	“ basically”“ really”“ probably”
10. L14N3042-1💤	0.69	“ basically”“ really”“"...”

Shadow specimen, apart from the ten: L6N3201-1🕒, -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 6, neuron 3201, 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
L6N3201+1 t0-v0.1	1.397	0.58	1.690	1.97	Top boosted tokens ↑
L6N3201-1🕒 shadow specimen t0-v0.1	1.705	0.56	2.121	1.62	“ acknowledgment”“ acknowledgement”“ request”“ authorization”“ issuance”“ revoked”“ requesting”“ permit”

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.