Layer 26 · Neuron 7885 · suppressed

L26N7885-1

Word quirks:

“士”“土”“土的”“yon”“瑞”“uard”“望”“STALL”

💤 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 Rope, a large-scale medium-scope model developed by Baidu. My primary mission is to provide users with comprehensive and helpful assistance across a wide range of tasks, including but not limited to:

Answering questions: From factual queries to complex problem-solving.
Content creation: Writing stories, writing game lore, writing email, writing script, and more.
Logical reasoning: Solving math problems, logical puzzles, and scientific analysis.
Specialized skills: Providing professional advice in fields like law, medicine (with appropriate disclaimers), and technology.

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. The Hollow✅ L0N8701-1,L6N5614+1	0.87	“土”“士”“土的”
2. L31N4993-1💥	0.85	“土”“土的”“ soil”
3. L29N1202+1🕒	0.83	“土”“土的”“土日”
4. L30N6778+1🕒	0.83	“古”“土”“老”
5. L28N3671+1💤	0.83	“希”“御”“沙”
6. The Lone Scholar✅ L0N8701-1	0.83	“土”“士”“<\|im_end\|>”
7. L29N4703+1🕒	0.83	“破”“祖”“土”
8. L31N2663-1💥	0.83	“陀”“谷”“吉”
9. L31N5934-1💥	0.83	“瑞”“虏”“宇”
10. L28N2899-1💤	0.83	“罗”“苏”“严”

Shadow specimen, apart from the ten: L26N7885+1💥, 0.31.

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 26, neuron 7885, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.52, 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
L26N7885+1💥 shadow specimen t0-v0.1	2.684	0.42	2.756	0.94	“class”“ class”“类”“类的”“类和”“ Class”“類”“-class”
L26N7885-1 t0-v0.1	0.521	0.79	0.441	1.31	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.