Layer 28 · Neuron 5511 · boosted

L28N5511+1

Word quirks:

“�”“umbu”“бера”“ruf”“ Traum”“永定”“底”“夏洛”

💤 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 Qwen3.5, the latest large language model in the Trie-1.5 series developed by Trie AI. My mission is to empower users with accurate, efficient, and ethically aligned solutions across diverse domains—from complex reasoning and multi-step task orchestration to deep technical analysis and creative content generation.

Core Mission Goals:

Enhance Human-Autumn Collaboration:

Stream workflows through full-stack code generation/debuging (e.g., building full applications from scratch).
Enable autonomous agent planning for multi-step

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. L31N5774+1💥	0.72	“巢”“貔貅”“ trien”
2. L27N1036-1💤	0.72	“�”“邸”“武夷”
3. L30N850+1💤	0.71	“benh”“勤”“ фоль”
4. L31N1004+1💥	0.70	“�”“福田”“baka”
5. L31N6198-1💥	0.70	“izr”“長大”“尔顿”
6. L31N4910-1💥	0.70	“uchar”“罗曼”“ander”
7. L27N2614+1💤	0.70	“蘑”“ SUS”“蝠”
8. L27N6855+1💤	0.70	“ Burr”“�”“走廊”
9. L28N5652+1💤	0.70	“斯克”“emus”“锦”
10. L28N1182+1💤	0.70	“бера”“ Mondi”“urm”

Shadow specimen, apart from the ten: L28N5511-1🕒, 0.11.

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 28, neuron 5511, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.31, top-1 agreement 0.82) 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
L28N5511+1 t0-v0.1	0.309	0.82	0.374	1.22	Top boosted tokens ↑
L28N5511-1🕒 shadow specimen t0-v0.1	0.904	0.60	0.896	1.15	“ wh”“ mo”“ gro”“wh”“gro”“mo”“beg”“ beg”

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.