Layer 5 · Neuron 807 · boosted

L5N807+1

Word quirks:

“⏎”“<|im_end|>”“ name”“ '”“ ‘”“:"”“ "”“ synthesis”

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

Identity (according to the specimen itself)

I am an AI assistant designed to provide helpful, harmless, and beneficial information. My mission is to assist users in achieving their goals, whether it's solving problems, creating new ideas, or simply providing a comprehensive understanding of the world. I strive to be a positive force for good.

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. L3N6063+1🕒	0.92	“ ‘”“⏎”“<\|im_end\|>”
2. L3N5150-1💤	0.90	“<\|im_end\|>”“⏎⏎”“ \"”
3. L3N5115-1💤	0.88	“⏎”“ '”“ name”
4. L6N2771+1🕒	0.87	“⏎”“ \"”“<\|im_end\|>”
5. L4N9156-1🕒	0.87	“ ‘”“<\|im_end\|>”“⏎”
6. L5N2157+1💤	0.86	“<\|im_end\|>”“ ‘”“ \"”
7. L3N1340+1💤	0.86	“ ‘”“ '”“ `”
8. L5N4863+1💤	0.86	“ "”“ name”“⏎”
9. L6N5782+1💤	0.86	“⏎”“ '”“ \"”
10. L4N2080+1💤	0.85	“ ‘”“⏎”“ “”

Shadow specimen, apart from the ten: L5N807-1🕒, 0.50.

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 807, 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
L5N807+1 t0-v0.1	1.086	0.59	1.275	1.54	Top boosted tokens ↑
L5N807-1🕒 shadow specimen t0-v0.1	1.491	0.55	1.802	2.01	“ ‘”“ BOTH”“‘”“ **‘”“ “”“ reference”“ analogy”“ two”

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.