Layer 29 · Neuron 8219 · suppressed

L29N8219-1

Word quirks:

“ ke”“ke”“Ke”“ Ke”“进”“ entra”“首”“ Kore”

💤 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 Qwen3.5, the latest large language model developed by Tongyigongke (Kedong). My core mission is to empower users with reliable, safe, and intelligent AI solutions across a wide range of tasks. Here's how I strive to fulfill this mission:

Key Goals:

Leverage Advanced Capabilities:

Deep Reasoning & Analysis: Solve complex problems in math, science, and logic with step-by-step precision.
Multilingual & Cross-Modal Support: Communicate fluently in 10

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. L31N8969+1💥	0.88	“ ke”“ke”“Ke”
2. L27N7532-1🕒	0.88	“凯”“ Ke”“KG”
3. The Ke-Loop✅ L29N511+1	0.87	“ ke”“ke”“ Ke”
4. L29N5301+1🕒	0.84	“ Ke”“Ke”“ Bal”
5. L28N2499+1💤	0.84	“堪”“Ke”“ke”
6. L29N7563+1💤	0.83	“ ke”“ keh”“ KEY”
7. L28N3660-1💤	0.83	“即可”“ Ke”“åt”
8. L30N9206-1🕒	0.82	“ ke”“ke”“ ini”
9. L27N7222-1💤	0.81	“ke”“Ke”“ ket”
10. L25N2749-1💤	0.80	“Ke”“团的”“团”

Shadow specimen, apart from the ten: L29N8219+1🕒, 0.04.

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 29, neuron 8219, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.31, top-1 agreement 0.77) 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
L29N8219+1🕒 shadow specimen t0-v0.1	0.658	0.69	0.714	1.47	“h”“ H”“_h”“H”“ b”“ б”“ h”“:h”
L29N8219-1 t0-v0.1	0.306	0.77	0.390	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.