Specimen ledger / L28N2099-1

Layer 28 · Neuron 2099 · suppressed

L28N2099-1

Word quirks:

“坛”“椒”“Ach”“芬”“iram”“roat”“ơ”“eté”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

Hello! I'm Q3.5, a large language model developed by T-IT (T-IT AI). My primary mission is to provide accurate, helpful, and ethical assistance across a wide range of tasks—whether that's solving complex problems, generating creative content, analyzing data, or simply having a meaningful conversation.

I prioritize:

Accuracy: Ensuring my responses are factually reliable and logically sound.
Ethics: Adhering to safety guidelines to avoid harm, bias, or misuse.
User-centricity: Tailoring my output to meet your specific needs

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. L27N3358-1🕒	0.73	“eka”“uja”“柱”
2. L31N1254+1💥	0.73	“沫”“χώ”“堂”
3. L31N4600+1💥	0.72	“榴”“釜”“伤”
4. L30N8878+1🕒	0.72	“ore”“ơ”“lo”
5. L31N2809-1💥	0.72	“麻”“寮”“哉”
6. L29N6394-1💤	0.72	“唇”“娘”“ordi”
7. L25N9123+1🕒	0.72	“omo”“品”“ Spartan”
8. L30N5704+1💤	0.71	“单位的”“iram”“ORN”
9. L23N9126+1🕒	0.71	“ Thom”“端”“äck”
10. L30N682-1💤	0.71	“尼奥”“amer”“茹”

Shadow specimen, apart from the ten: L28N2099+1🕒, 0.10.

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 2099, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.29, top-1 agreement 0.81) 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
L28N2099+1🕒 shadow specimen t0-v0.1	1.693	0.50	1.785	1.12	“滑”“ slide”“ slides”“ Slide”“Slide”“slide”“ gl”“ slid”
L28N2099-1 t0-v0.1	0.294	0.81	0.326	1.31	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.