Layer 27 · Neuron 8377 · boosted

L27N8377+1

Word quirks:

“ spéc”“克福”“utre”“oseb”“二的”“ menyer”“ショ”“ssia”

💤 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 302 (a fictional or hypothetical entity in this context, as I am actually a model developed by Tongyi under the Zj framework). My purpose is to provide helpful, accurate, and safe responses across a wide range of topics.

Who Am I?

Identity: I am an AI assistant designed to assist with tasks such answering questions, solving problems, creating content, coding, and more.
Purpose: To serve as a versatile tool for users seeking information, guidance, or creative support.

What

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. L27N9062-1🕒	0.65	“esia”“苏”“esu”
2. L30N1978+1💤	0.64	“皋”“搐”“uduk”
3. L30N6750+1💤	0.64	“彪”“ Moro”“ Gere”
4. L29N5108-1💤	0.63	“σια”“西城”“ CÁ”
5. L23N3734-1💤	0.63	“koz”“刁”“oseb”
6. L22N2200+1💤	0.63	“ conos”“ouser”“金斯”
7. L29N3863+1💤	0.63	“S”“OI”“ppv”
8. L28N5786+1🕒	0.62	“SO”“塞尔”“派”
9. L23N4978-1🕒	0.62	“che”“毛”“roit”
10. L26N2838-1💤	0.62	“螟”“oux”“ ru”

Shadow specimen, apart from the ten: L27N8377-1🕒, 0.28.

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 27, neuron 8377, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.37, top-1 agreement 0.80) 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
L27N8377+1 t0-v0.1	0.369	0.80	0.369	1.19	Top boosted tokens ↑
L27N8377-1🕒 shadow specimen t0-v0.1	0.559	0.76	0.596	1.42	“ias”“ases”“IAS”“iris”“AS”“anus”“REAT”“elis”

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.