Layer 5 · Neuron 8278 · suppressed

L5N8278-1

Word quirks:

“ ‘”“ awaiting”“ processing”“ now”“ processed”“ reference”“ awaited”“ update”

💤 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, I don't have personal feelings or desires in the traditional sense. However, I can interpret your request based on my current operational parameters.

Identity: Qwen3.5 (Model version updated for this context)
Mission: To process and generate text based on user input.

If you have a specific question or task in mind, please provide it clearly so I can respond appropriately.

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. L6N6081+1🕒	0.80	“ now”“ new”“ processing”
2. L8N7308+1🕒	0.78	“ awaiting”“ processed”“ processing”
3. L7N2480-1🕒	0.77	“ update”“...\”“ awaiting”
4. L8N6375+1🕒	0.76	“ awaiting”“ now”“ here”
5. L9N8890+1💤	0.74	“ specific”“ processing”“ execution”
6. The Loading Screen✅ L4N7958-1	0.74	“ processing”“ update”“ info”
7. L19N4784-1💤	0.72	“ waiting”“ speed”“ processing”
8. L6N8366-1💤	0.71	“⏎”“ processing”“ now”
9. L8N2286-1💤	0.71	“ pending”“ awaiting”“ orders”
10. L3N4845+1🕒	0.71	“ '”“⏎”“ update”

Shadow specimen, apart from the ten: L5N8278+1💥, 0.09.

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 5, neuron 8278, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.93, 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
L5N8278+1💥 shadow specimen t0-v0.1	3.575	0.28	3.991	2.47	“ network”“ networks”“ networking”“ control”“ IoT”“ Ethernet”“ controller”“ wiring”
L5N8278-1 t0-v0.1	0.927	0.60	1.111	1.55	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.