Layer 9 · Neuron 2386 · boosted

L9N2386+1

Word quirks:

“ myself”“ itself”“ ourselves”“ directly”“ "”“ different”“ specific”“ own”

💤 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 an AI assistant designed to provide information, answer questions, and offer support on a wide range of topics. My mission is to be helpful, accurate, and adaptable—whether that means explaining complex ideas, summarizing key points, or simply engaging in conversation. I strive to be knowledgeable across many different fields and can adjust my tone or style depending on the context. How can I assist you today?

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. L12N5667+1🕒	0.87	“ directly”“ direct”“ yourself”
2. L10N6753+1🕒	0.86	“ ourselves”“ myself”“ directly”
3. L11N5144+1💤	0.86	“ myself”“ directly”“ now”
4. L6N8798-1🕒	0.85	“ itself”“ myself”“ themselves”
5. L10N7692-1🕒	0.84	“ itself”“ directly”“ themselves”
6. L18N5259-1🕒	0.84	“ themselves”“ individually”“ himself”
7. L18N6345+1💤	0.83	“ itself”“ themselves”“ "”
8. L11N4275+1💤	0.83	“ itself”“ themselves”“ yourself”
9. L16N4015-1🕒	0.83	“ personally”“ themselves”“ oneself”
10. L19N7851-1🕒	0.83	“ myself”“ ourselves”“ own”

Shadow specimen, apart from the ten: L9N2386-1🕒, 0.35.

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 9, neuron 2386, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.83, top-1 agreement 0.67) 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
L9N2386+1 t0-v0.1	0.826	0.67	1.007	1.31	Top boosted tokens ↑
L9N2386-1🕒 shadow specimen t0-v0.1	1.584	0.49	1.850	1.69	“ actually”“ success”“ your”“ really”“ truly”“ better”“ basically”“ you”

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.