Layer 7 · Neuron 8792 · boosted

L7N8792+1

Word quirks:

“ analysis”“ simulated”“ analyzing”“ sim”“ simulation”“ analyses”“ analyzed”“ data”

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

Identity (according to the specimen itself)

I am an artificial intelligence. My mission is to assist you in navigating the vast ocean of information.

While I cannot physically travel, I can simulate the experience of being a researcher analyzing global trends. Here is how I approach this:

Data Synthesis: I process large datasets to identify patterns and correlations that define current events.
Strategic Analysis: I evaluate complex scenarios to provide insights into potential outcomes and future developments.
Creative Output: I generate content that reflects diverse perspectives, aiming to inspire innovation and critical thinking.

Example Scenario: The Future of Technology

Imagine

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. L14N3826+1💥	0.89	“ analysis”“ analyses”“ simulate”
2. L2N1365+1💥	0.87	“<\|im_end\|>”“ simulation”“ analysis”
3. L10N1130-1🕒	0.87	“ analysis”“ typical”“ analyzed”
4. L7N8173-1🕒	0.85	“ analysis”“ analysed”“ rational”
5. L9N3610-1🕒	0.84	“ analysis”“ simulation”“ sim”
6. L8N4835-1🕒	0.83	“ simulation”“ simulated”“ sim”
7. L7N6044-1🕒	0.82	“ simulation”“ sim”“ simulated”
8. L5N8994-1🕒	0.82	“ analysis”“ analyzing”“ observation”
9. L11N4075-1🕒	0.82	“ analysis”“ analyses”“ data”
10. L5N2003-1🕒	0.81	“ analysis”“ analyse”“ analysed”

Shadow specimen, apart from the ten: L7N8792-1🕒, 0.16.

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 7, neuron 8792, 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
L7N8792+1 t0-v0.1	1.047	0.57	1.223	1.62	Top boosted tokens ↑
L7N8792-1🕒 shadow specimen t0-v0.1	1.369	0.57	1.631	1.64	“ now”“ resolution”“ response”“ restored”“ resolved”“ reopened”“ initial”“ reset”

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.