Layer 23 · Neuron 2983 · suppressed

L23N2983-1

Word quirks:

“ parameters”“参数”“Parameters”“的参数”“parameters”“参数的”“ Parameters”“ params”

🕒 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)

I am Qin, a large language model developed by the 100% information parameters (a fictional or placeholder credential; actual parameters are not public). My mission is to provide accurate, safe, and helpful information across a wide range of topics—whether you need to solve complex problem parameters, generate creative content, or simply chat about your favorite parameters. I aim to assist you with precision and clarity while adhering to all relevant safety and ethical parameters. How can I help 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. L20N2571-1🕒	1.00	“ parameters”“参数”“ Parameters”
2. L20N2292+1🕒	0.99	“参数”“ parameters”“参数的”
3. L21N3679-1💤	0.99	“ parameters”“参数”“ Parameters”
4. L22N1916+1🕒	0.99	“ parameters”“参数”“parameters”
5. L23N4350-1🕒	0.98	“参数”“ Parameters”“参数的”
6. L26N1765-1🕒	0.98	“参数”“参数的”“ parameters”
7. L23N4642+1🕒	0.98	“ parameters”“参数”“ Parameters”
8. L27N1785+1🕒	0.98	“ Parameters”“ parameters”“Parameters”
9. L22N3571-1🕒	0.98	“ parameters”“参数”“ Parameters”
10. L27N4157-1🧪	0.97	“参数”“ Parameters”“ parameters”

Shadow specimen, apart from the ten: L23N2983+1💤, 0.01.

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 23, neuron 2983, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.97, top-1 agreement 0.68) 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
L23N2983+1💤 shadow specimen t0-v0.1	0.719	0.68	0.770	1.31	“aces”“安县”“ stages”“ juntas”“kev”“ kitchens”“hana”“ Bodies”
L23N2983-1 t0-v0.1	0.973	0.68	0.991	1.30	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.