Layer 10 · Neuron 8601 · boosted

L10N8601+1

Word quirks:

“ basically”“ **"”“ actually”“:"”“ "**”“ "*”“ *"”“!:”

💤 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 Qwen3.5, and my mission is to be a versatile, reliable AI assistant capable of handling a wide range of tasks across different domains. Here's how I operate:

Core Capabilities:
I can process information from text, code, data, or structured formats depending on the user's input.
I support multi-step workflows (e.g., understanding context, executing logic, or generating outputs).
I provide real-time responses with accuracy, though performance may vary based on the specific task at hand.

Key Features:

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. L11N6055+1🕒	0.93	“ "”“⏎”“ basically”
2. L11N2165+1🕒	0.91	“ "”“ "”“ basically”
3. L12N3594+1💤	0.90	“ *"”“ "”“ "**”
4. L13N3069-1💤	0.90	“ basically”“ really”“ **"”
5. L10N4280-1🕒	0.90	“ "”“ "”“ "”
6. L14N6458+1🕒	0.90	“ **"”“ basically”“ really”
7. L8N4816-1💤	0.90	“ "”“ '”“ *”
8. Context Beggar✅ L10N3851+1	0.90	“ "”“ actually”“ "”
9. L10N1011-1🕒	0.90	“ "”“ "”“ basically”
10. L7N5219+1🕒	0.90	“ "”“ \"”“ "*”

Shadow specimen, apart from the ten: L10N8601-1🕒, 0.39.

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 10, neuron 8601, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.75, top-1 agreement 0.68) 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
L10N8601+1 t0-v0.1	0.747	0.68	0.973	1.21	Top boosted tokens ↑
L10N8601-1🕒 shadow specimen t0-v0.1	1.018	0.59	1.319	1.50	“<\|im_end\|>”“⏎”“ various”“ ‘”“ additionally”“ ‘’”“...””“ ____”

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.