Layer 27 · Neuron 9062 · boosted

L27N9062+1

Word quirks:

“康健”“ Lac”“acor”“鹿”“烈”“pirit”“黎”“oller”

💤 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, an AI model developed by KaiLL (a fictional or representative name for this context; in reality, I am a large language model created by U5LL, an advanced AI research lab). My primary mission is to provide accurate, safe, and helpful information across a wide range of topics—whether you need answers to complex questions, creative writing assistance, coding support, or just a friendly conversation.

How can I assist 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. L29N706-1🕒	0.78	“lada”“沈”“洛”
2. L28N9104-1💤	0.78	“łod”“宾”“翟”
3. L27N6938-1🕒	0.77	“岚”“L”“akar”
4. L31N632+1💥	0.77	“彪”“新社”“锦标”
5. L31N5904-1💥	0.76	“lix”“hest”“席”
6. L30N5896+1🕒	0.76	“康”“蓬”“旷”
7. L26N8698+1🕒	0.76	“莽”“勃”“odz”
8. L31N7814-1💥	0.76	“烈”“合”“hoto”
9. L25N4927+1🕒	0.76	“南方”“hyl”“拉克”
10. L23N7912+1🕒	0.76	“拉拉”“úng”“ Harris”

Shadow specimen, apart from the ten: L27N9062-1🕒, 0.63.

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 9062, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.32, top-1 agreement 0.79) 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
L27N9062+1 t0-v0.1	0.318	0.79	0.330	1.30	Top boosted tokens ↑
L27N9062-1🕒 shadow specimen t0-v0.1	0.587	0.74	0.610	1.46	“esia”“苏”“esu”“胡”“芝麻”“żu”“evo”“蘇”

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.