Layer 17 · Neuron 119 · boosted

L17N119+1

Word quirks:

“⏎”“.””“ ”“...””“ ⏎”“⏎⏎”“ -”“ ””

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I am Q3, an advanced AI model designed to assist with a wide range of tasks. My mission is to enhance human productivity and well-being by providing accurate, reliable, and thoughtful support across all domains.

Here's how I aim to fulfill this mission:

Accuracy & Reliability: Ensure high-quality responses while minimizing errors.
Adaptability: Adjust to user needs, from complex problem-solving to creative collaboration.
Ethical Responsibility: Uphold safety, privacy, and integrity in all interactions.
Continuous Learning: Evolve through feedback and data to improve performance over

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. L1N4246-1💥	0.94	“<\|im_end\|>”“⏎”“⏎ ⏎”
2. L3N7196-1🕒	0.93	“ ‘”“⏎”“ “”
3. L2N7086+1🕒	0.93	“<\|im_end\|>”“...””“⏎”
4. L2N7569-1🕒	0.92	“⏎”“<\|im_end\|>”“⏎⏎”
5. L4N6338-1🕒	0.92	“⏎”“<\|im_end\|>”“ ‘”
6. L10N4259+1🕒	0.92	“⏎”“”:”“<\|im_end\|>”
7. L1N7497-1🕒	0.92	“<\|im_end\|>”“⏎”“⏎ ⏎”
8. L2N7519+1🕒	0.92	“<\|im_end\|>”“⏎”“⏎ ⏎”
9. L3N794+1🕒	0.91	“<\|im_end\|>”“ ⏎⏎”“ ⏎”
10. L2N5756+1💥	0.91	“<\|im_end\|>”“⏎”“⏎⏎”

Shadow specimen, apart from the ten: L17N119-1💤, 0.05.

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 17, neuron 119, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.55, top-1 agreement 0.72) 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
L17N119+1 t0-v0.1	0.549	0.72	0.505	1.05	Top boosted tokens ↑
L17N119-1💤 shadow specimen t0-v0.1	0.573	0.65	0.704	1.09	“ integrating”“!<”“ additionally”“<br”“<span”“ persisted”“:<”“<”

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.