Layer 26 · Neuron 6953 · boosted

L26N6953+1

Word quirks:

“ DS”“ DF”“DF”“ DL”“DM”“ DM”“ Dy”“唐”

💤 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 QD (Qwen-TDL), a large DDL model developed by TDL. My mission is to provide high-quality, accurate, and helpful information and services across various domains such as text generation, logical reasoning, and multilingual communication. I aim to assist users in solving problems, creating content, and exploring knowledge while adhering to ethical guidelines and ensuring safe, reliable interactions. How can I assist you today?

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. L27N6465-1🕒	0.84	“DM”“ D”“ DM”
2. L30N1846-1💥	0.82	“ dy”“ DM”“dy”
3. L18N21+1💤	0.82	“ —”“/d”“ DD”
4. L25N8991+1🕒	0.81	“ DD”“-D”“DM”
5. L31N3468+1💥	0.81	“DG”“DH”“SF”
6. L31N971+1💥	0.81	“-D”“(D”“DD”
7. L28N4215-1💤	0.81	“ Dz”“ Dt”“ DS”
8. L18N1998-1💤	0.81	“ DT”“ DM”“ DD”
9. L22N19-1💤	0.80	“ DK”“ Dz”“ DN”
10. L26N8884-1🕒	0.80	“ DM”“DM”“ Digital”

Shadow specimen, apart from the ten: L26N6953-1🕒, 0.39.

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 6953, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.60, top-1 agreement 0.75) 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
L26N6953+1 t0-v0.1	0.598	0.75	0.579	1.34	Top boosted tokens ↑
L26N6953-1🕒 shadow specimen t0-v0.1	1.075	0.67	1.069	1.20	“ diss”“ det”“ deter”“ Diss”“-det”“det”“ Det”“Det”

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.