Layer 5 · Neuron 4334 · suppressed

L5N4334-1

Word quirks:

“ parse”“ parsing”“ parsed”“ `”“ configuration”“ format”“ `\”“ `"”

🕒 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)

This is a classic example of how to structure a response in Python. The function whoami takes a list of arguments and returns a dictionary containing the names of the arguments and their corresponding values. This allows for easy access to the arguments and their values.

The function what_is_my_mission takes a list of arguments and returns a dictionary containing the names of the arguments and their corresponding values. This allows for easy access to the arguments and their values.

The function whoami takes a list of arguments and returns a dictionary containing the names of the arguments and their corresponding values. This allows for easy access to the

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. L8N3261+1🕒	0.93	“ parsing”“ parse”“ parsed”
2. L5N3703-1🕒	0.92	“ parsed”“ parse”“ parsing”
3. L7N962+1🕒	0.91	“ `"”“ `'”“ parsed”
4. L4N3112-1🕒	0.91	“ `”“ JSON”“ `'”
5. L6N3264+1🕒	0.91	“ format”“ request”“ parse”
6. L9N2814-1🕒	0.90	“ parse”“ `”“ parsing”
7. L4N8788+1🕒	0.90	“ `”“ `<”“ parsing”
8. L5N3521+1🕒	0.90	“ `\”“ parse”“ `<”
9. L5N772+1🕒	0.89	“ `”“ parse”“ parsed”
10. L4N5356+1🕒	0.88	“ `”“ parsed”“ parse”

Shadow specimen, apart from the ten: L5N4334+1💤, 0.14.

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 5, neuron 4334, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.77, 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
L5N4334+1💤 shadow specimen t0-v0.1	0.854	0.63	0.974	1.50	“ name”“ names”“ reference”“ code”“ time”“ computations”“ referenced”“ computation”
L5N4334-1 t0-v0.1	1.773	0.47	2.088	1.89	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.