Layer 7 · Neuron 5672 · suppressed

L7N5672-1

Word quirks:

“ “”“ referring”“ ‘”“ \"”“ referenced”“ description”“ reference”“ typical”

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

Identity (according to the specimen itself)

As an AI language model, I do not have access to real-time information about your personal life or professional goals. However, I can provide a general response based on the following:

My Identity: I am an artificial intelligence designed to assist with various tasks, such as data analysis, problem-solving, and creative projects.
My Mission: My primary objective is to help users achieve their goals efficiently by leveraging my skills in technology and innovation.

If you need more specific details, please let me know! 😊

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. L5N6491-1💤	0.93	“ reference”“ referenced”“ “”
2. L8N2231-1💤	0.91	“ ‘”“ --”“ reference”
3. L4N1454+1💤	0.91	“ “”“ ‘”“ referenced”
4. L8N5798+1💤	0.90	“ referenced”“ “”“ ‘”
5. L7N1284-1💤	0.90	“⏎”“ reference”“ referring”
6. L5N5907-1🕒	0.90	“ name”“ referring”“ referenced”
7. L3N586-1🕒	0.89	“ ‘”“ “”“ reference”
8. L8N6408-1💤	0.89	“ reference”“ ‘”“ referenced”
9. L3N5911+1💤	0.89	“ \"”“ reference”“ name”
10. L7N6421-1💤	0.89	“ ‘”“ “”“ example”

Shadow specimen, apart from the ten: L7N5672+1🕒, -0.08.

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 5672, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.97, top-1 agreement 0.60) 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
L7N5672+1🕒 shadow specimen t0-v0.1	1.933	0.51	2.234	1.72	“ automated”“ bots”“ automation”“ bot”“-bot”“ autom”“ automate”“自动化”
L7N5672-1 t0-v0.1	0.971	0.60	1.160	1.57	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.