Layer 8 · Neuron 1864 · boosted

L8N1864+1

Word quirks:

“ cryptographic”“ crypt”“ encrypted”“ decrypted”“ Crypto”“ signatures”“ Crypt”“ certificate”

🕒 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)

I am an AI assistant developed by Google. My mission is to provide high-quality, secure, and reliable software development tools for developers who need to sign code and manage code changes in a software development lifecycle. I am designed to help developers ensure the integrity and authenticity of their code, while also providing a mechanism for revoking access if private keys are lost or compromised.

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. L6N4210+1🕒	0.95	“ cryptographic”“ cryptography”“ crypt”
2. L6N5615-1🕒	0.94	“ crypt”“ encrypted”“⏎”
3. L7N5276-1🕒	0.93	“ cryptographic”“ crypt”“ encrypted”
4. L25N5940+1🕒	0.93	“ crypt”“crypt”“Crypt”
5. L4N2777+1🕒	0.91	“ crypt”“ Crypto”“ cryptographic”
6. L3N6032+1🕒	0.90	“ encryption”“ encrypt”“ ciphertext”
7. L25N8556+1🕒	0.89	“ Crypt”“Crypt”“ crypt”
8. L6N1923+1🕒	0.89	“ cipher”“ ciphertext”“ cryptographic”
9. L1N8344-1💥	0.89	“ crypto”“<\|im_end\|>”“ crypt”
10. L5N1441+1🕒	0.89	“ crypt”“ crypto”“ exchange”

Shadow specimen, apart from the ten: L8N1864-1💤, -0.16.

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 8, neuron 1864, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.02, top-1 agreement 0.66) 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
L8N1864+1 t0-v0.1	1.022	0.66	1.142	1.36	Top boosted tokens ↑
L8N1864-1💤 shadow specimen t0-v0.1	0.980	0.64	1.238	1.33	“ ‘”“ ‘”“ ‘’”“ “”“ usually”“ often”“ global”“ *“”

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.