Layer 16 · Neuron 8890 · suppressed

L16N8890-1

Word quirks:

“ reaching”“ reach”“ reaches”“ Reach”“达到”“ reached”“达到了”“ rated”

🕒 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 Qwen3.5, a large language model with a 256K context window supporting high-precision accuracy up to 1024 tokens at 1024/1024 resolution. I can handle tasks requiring detailed analysis of documents or data within this context window. My capabilities include reaching a 1024-token precision level of 1024/1024, with a maximum context length of 256K tokens. I support a 1024-token precision level of 1024/1024, with a maximum context

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. L23N1645-1🕒	0.97	“ reached”“ reaching”“到达”
2. L25N4869+1🕒	0.96	“ reaching”“ reach”“ reaches”
3. L26N940+1🕒	0.96	“ reach”“ Reach”“ reaching”
4. L26N1631-1🕒	0.95	“ Reach”“ reach”“ reaching”
5. L22N6676-1🕒	0.95	“ reach”“ reached”“达”
6. L25N2514+1🕒	0.94	“ reach”“ Reach”“ reaching”
7. L28N2914+1🕒	0.93	“ reaching”“ reach”“ Reach”
8. L27N4330+1🕒	0.93	“ Reach”“ reach”“达到”
9. L23N1262+1🕒	0.93	“ reaching”“ reaches”“ reach”
10. L25N6323-1🕒	0.92	“达到”“ reaches”“ reach”

Shadow specimen, apart from the ten: L16N8890+1💤, -0.18.

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 16, neuron 8890, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.87, 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
L16N8890+1💤 shadow specimen t0-v0.1	0.544	0.76	0.735	1.26	“ ____”“ (<”“ *”“ ______”“ (”“****”“ (>”“____”
L16N8890-1 t0-v0.1	0.873	0.68	1.019	1.18	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.