Specimen ledger / L14N8643+1

Layer 14 · Neuron 8643 · boosted

L14N8643+1

Word quirks:

“ onboard”“ aboard”“ online”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I am Qwen3.5, an AI model designed to assist with complex tasks. If you'd like to interact with me:

System Prompt: "You are Qwen3.5" (for internal configuration).
User Input: Provide instructions in your voice (e.g., "I'm a teacher").
Response Example:
"I am Qwen3.5" (system prompt)
"I am Qwen3.5" (system prompt)
"I am Qwen3.5" (system prompt)

If you prefer not

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. L13N4777+1🕒	0.70
2. L13N2604-1🧪	0.69
3. L17N5165+1🕒	0.69
4. L16N5528-1🕒	0.69
5. L17N6640+1🕒	0.69
6. L15N4839+1🕒	0.69
7. L16N5871-1🕒	0.68
8. L11N1850-1🕒	0.68
9. L14N2464-1💤	0.68	“ sometimes”
10. L11N985+1💤	0.67	“ ONLY”

Shadow specimen, apart from the ten: L14N8643-1💤, -0.02.

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 14, neuron 8643, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L14N8643+1 t0-v0.1	1.150	0.62	1.353	1.51	Top boosted tokens ↑
L14N8643-1💤 shadow specimen t0-v0.1	0.873	0.63	1.126	1.46	“ now”“ you”“ specific”“ needed”“ here”

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.