Layer 28 · Neuron 6875 · boosted

L28N6875+1

Word quirks:

“交”“伟”“estern”“伦”“olland”“广元”“ATO”“ urn”

💤 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 Qwen series developed by Tiprod (T-iprode). My mission is to empower users with intelligent, reliable, and ethical AI assistance across a vast spectrum of tasks. Here’s how I strive to fulfill this mission:

Core Mission Goals:

Universal Accessibility:

Provide clear, accurate, and context-aware responses for anyone, from students studying complex topics to professionals needing data analysis or code generation.

Task Versatility:

Handle everything from creative writing and scientific reasoning to multilingual

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. L30N1261+1💤	0.80	“aks”“交投”“韦”
2. L30N714-1💤	0.77	“元首”“梧州”“四通”
3. L28N5289-1💤	0.77	“昌”“马里”“排”
4. L31N6750+1💥	0.76	“楚”“魏”“郭”
5. L30N3060+1🕒	0.76	“悠悠”“菊”“upe”
6. L28N4220-1💤	0.76	“优”“认”“人”
7. L26N8340-1💤	0.76	“驴”“超前”“eki”
8. L24N1431-1🕒	0.76	“周口”“otty”“永定”
9. L26N1790+1💤	0.75	“捷克”“ Carnegie”“电”
10. L28N7081-1💤	0.75	“acer”“米”“格尔”

Shadow specimen, apart from the ten: L28N6875-1🕒, 0.41.

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 6875, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.34, top-1 agreement 0.78) 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
L28N6875+1 t0-v0.1	0.339	0.78	0.442	1.25	Top boosted tokens ↑
L28N6875-1🕒 shadow specimen t0-v0.1	1.162	0.59	1.221	1.31	“ le”“ J”“ sal”“跳”“盐”“ Sal”“ salt”“跳过”

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.