Layer 6 · Neuron 7128 · boosted

L6N7128+1

Word quirks:

“ genomic”“ DNA”“ gene”“ mRNA”“ genome”“ PCR”“ genes”“ genetic”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

Hello! I am a large language model designed to help researchers and developers with their projects. My primary function is to assist in cloning, expressing, and analyzing DNA sequences.

Here is a breakdown of my identity and mission:

1. Who Am I?

I am a synthetic gene (or plasmid) vector. Specifically, I am based on the pET-28a(+) vector backbone, which is widely used in molecular biology laboratories.

Origin: I was originally constructed using the pUC19 vector as a starting point, but I have been modified to include

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. L4N6655-1🕒	0.99	“ genes”“ genomic”“ RNA”
2. L3N4369+1💥	0.97	“ genomic”“ RNA”“ DNA”
3. L4N9062+1🕒	0.97	“ RNA”“ DNA”“ mRNA”
4. L5N8592-1🕒	0.96	“ DNA”“ genetic”“ genomic”
5. L6N4282+1🕒	0.96	“ RNA”“ genome”“ mRNA”
6. L6N7038+1🕒	0.96	“ DNA”“ genomic”“ genome”
7. L16N7877-1🕒	0.95	“ genomic”“ genome”“ genes”
8. L7N3712+1🕒	0.95	“ genomic”“ DNA”“ genome”
9. L7N464-1🕒	0.95	“ genes”“ gene”“ RNA”
10. L7N3506-1🕒	0.95	“ genome”“ DNA”“ genomic”

Shadow specimen, apart from the ten: L6N7128-1🕒, 0.08.

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 6, neuron 7128, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.97, top-1 agreement 0.47) 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
L6N7128+1 t0-v0.1	1.975	0.47	2.192	1.82	Top boosted tokens ↑
L6N7128-1🕒 shadow specimen t0-v0.1	1.394	0.49	1.684	1.88	“ info”“ textual”“ compositor”“ rendered”“ compositions”“ composing”“ composed”“ specially”

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.