Layer 27 · Neuron 1483 · boosted

L27N1483+1

Word quirks:

“,'”“,<”“,\”“,”“,"”“,.”“，“”“,$”

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

Identity (according to the specimen itself)

I am Q3.5, a large language model developed by, for, and of the Tonger, Inc., a leading AI research company. My core mission is to provide users with accurate, helpful, and safe information across a wide range of domains—whether it's solving complex problems, creative writing, coding assistance, or answering general knowledge questions.

Key aspects of my mission include:

Accuracy: Striving to provide reliable, well-reasoned responses based on my training data.

Helpfulness: Tailoring answers to user needs, from technical details to everyday

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. L19N1773+1🕒	0.93	“,*”“,”“,</”
2. L19N945+1🕒	0.92	“,”“，“”“,</”
3. L21N6910-1🕒	0.92	“,”“，”“,\”
4. L30N5024-1💥	0.91	“,\”“,”“,/”
5. L31N2774+1💥	0.90	“，“”“,"”“,\"”
6. L20N152+1🕒	0.90	“,”“,"”“,\”
7. L17N554+1🕒	0.89	“,**”“,””“,"”
8. L22N8083-1🕒	0.89	“,”“,**”“，”
9. L27N2058-1💤	0.88	“，“”“,\”“,\"”
10. L27N252+1🕒	0.88	“,”“，”“،”

Shadow specimen, apart from the ten: L27N1483-1💤, 0.46.

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 27, neuron 1483, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.33, top-1 agreement 0.79) 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
L27N1483+1 t0-v0.1	0.332	0.79	0.335	1.12	Top boosted tokens ↑
L27N1483-1💤 shadow specimen t0-v0.1	0.455	0.71	0.478	1.33	“，”“ ，”“””“�”“،”“BP”“JM”“A”

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.