Layer 29 · Neuron 6068 · suppressed

L29N6068-1

Word quirks:

“blers”“hada”“uré”“曲”“egu”“UU”“lic”“macht”

💤 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, a large language model developed by TME (TME 2025). My mission is to be your personal assistant and knowledge companion, helping you with a wide range of tasks while adhering to strict safety and ethical guidelines.

What I Can Do:

Answer Questions: Provide clear, accurate, and nuanced responses to factual, technical, or creative queries.
Solve Problems: Step through complex reasoning (math, coding, logic) with detailed explanations.
Create Content: Draft emails, stories, poems, or even code with explicit instructions

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. L31N6067+1💥	0.70	“éch”“uros”“ulum”
2. L30N4529+1💤	0.70	“ugel”“URT”“机构”
3. L31N8118+1💥	0.69	“谷”“ume”“uges”
4. L30N7438+1💤	0.68	“elles”“μία”“clar”
5. L31N7811+1💥	0.68	“uelle”“�”“ierra”
6. L27N2568+1💤	0.68	“娥”“uren”“stur”
7. L29N8631+1💤	0.68	“лива”“uha”“reta”
8. L29N606+1🕒	0.67	“noi”“ugat”“ spiel”
9. L29N6812-1🕒	0.67	“ulan”“�”“уб”
10. L28N5964+1💤	0.67	“UMO”“沽”“PU”

Shadow specimen, apart from the ten: L29N6068+1🕒, 0.38.

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 6068, 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 0.82) 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
L29N6068+1🕒 shadow specimen t0-v0.1	0.377	0.76	0.509	1.42	“yn”“ene”“ine”“ice”“yne”“ence”“ynn”“elle”
L29N6068-1 t0-v0.1	0.307	0.82	0.323	1.59	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.