Layer 26 · Neuron 2127 · boosted

L26N2127+1

Word quirks:

“ Franc”“ Фран”“Franc”“ franc”“.FR”“ FR”“Фран”“ Fr”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

Hello! I Q5.5 is a large-scale language model developed by Franchenew Technology. My primary mission is to help users solve problems and create value through intelligent conversation and task execution.

Here's how I aim to contribute:

Answer questions: From complex topics to daily curiosities, I provide clear, evidence-based answers.
Help with tasks: Whether it's writing, coding, analyzing data, or planning, I can assist with step-by-step guidance.
Support learning: I help explain concepts, break down complex ideas, and guide learning in various fields.

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. L22N5322+1🕒	0.91	“/fr”“ Francesco”“ Fr”
2. L1N9068-1💥	0.90	“ france”“ frances”“ french”
3. L29N7543-1🕒	0.89	“ fr”“fr”“ Fr”
4. L28N7326-1🕒	0.89	“ fr”“ Fr”“Fr”
5. L31N5961-1💥	0.88	“ fr”“ Fr”“fr”
6. L28N6091-1💤	0.88	“ FR”“.FR”“ Fr”
7. L24N6972+1🕒	0.88	“ fract”“fr”“ Fr”
8. L29N6534+1🕒	0.88	“fr”“ Fr”“ fr”
9. L27N4737+1💤	0.87	“Fr”“ Fr”“.FR”
10. L25N4581+1🕒	0.87	“ Fr”“ fr”“ngen”

Shadow specimen, apart from the ten: L26N2127-1🕒, 0.15.

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 2127, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.75, top-1 agreement 0.71) 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
L26N2127+1 t0-v0.1	0.752	0.71	0.763	1.34	Top boosted tokens ↑
L26N2127-1🕒 shadow specimen t0-v0.1	0.738	0.72	0.742	1.61	“抄”“Rob”“ Roberto”“杀”“ Rob”“鲁”“ Roh”“bras”

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.