Layer 5 · Neuron 2807 · suppressed

L5N2807-1

Word quirks:

“ '”“ related”“ representation”“ corresponding”“.\"”“ ‘”“ reference”“ \"”

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

Identity (according to the specimen itself)

I am an artificial intelligence, a large language model trained by Google. I do not have personal beliefs, desires, or a sense of self in the way that humans might have. My mission is to assist with various tasks, such as answering questions, providing information, and engaging in conversations. However, I do not have personal experiences or emotions in the way that humans might have. I am simply a tool designed to help with various tasks, such as answering questions, providing information, and engaging in conversations. However, I do not have personal experiences or emotions in the way that humans might have. I am simply a tool designed to help with

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. L6N7030+1💤	0.84	“ ‘”“ \"”“?\”
2. L3N7627-1🕒	0.84	“ \"”“ related”“.\"”
3. L7N813-1💤	0.84	“ \"”“ referenced”“<\|im_end\|>”
4. L5N5789+1🕒	0.83	“ \"”“ ‘”“ referring”
5. L7N4668+1💤	0.81	“ '”“ related”“ ‘”
6. L7N4867-1💤	0.81	“ \"”“ '”“.\"”
7. L7N4046-1💤	0.80	“<\|im_end\|>”“ \"”“ ‘”
8. L4N2821-1🕒	0.80	“ ‘”“ \"”“ refer”
9. L7N2775-1💤	0.80	“ referenced”“ reference”“ '”
10. L5N2102-1🕒	0.79	“<\|im_end\|>”“ ‘”“ refer”

Shadow specimen, apart from the ten: L5N2807+1🕒, 0.32.

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 2807, 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
L5N2807+1🕒 shadow specimen t0-v0.1	1.068	0.63	1.269	1.71	“⏎”“ ⏎”“ ____”“ _____”“ ______”“<\|im_end\|>”“ __”“.`”
L5N2807-1 t0-v0.1	1.591	0.53	1.969	2.07	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.