Layer 11 · Neuron 8681 · boosted

L11N8681+1

Word quirks:

“ "”“ **"”“ trying”“ compliment”“ "**”“ compliments”“ literally”“ hope”

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

Identity (according to the specimen itself)

I am Qwen3.5, a large language model developed by Tongyi Lab. My mission is to provide assistance that is both helpful and capable of adapting to a wide range of tasks and scenarios. While I strive to be as useful as possible, it's important to remember that no AI can guarantee perfect performance or eliminate all risks without human intervention.

Key Capabilities & Design Philosophy:

Comprehensive Understanding: I aim to understand the context, user intent, and any implicit information in the conversation to provide more relevant support.
Adaptive Learning: I continuously learn from interactions,

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. L10N7597-1🕒	0.73	“ "”“ because”“ "**”
2. L7N597+1🕒	0.67	“ your”“ `"”“ \"”
3. L16N9158+1💤	0.64	“ "”“ \"”“ "”
4. L8N2078+1💤	0.64	“ **"”“ nicely”“ nice”
5. L11N3879-1🕒	0.63	“ nicely”“ usually”“ **"”
6. L12N5684+1🕒	0.63	“ "”“ “”“ **"”
7. L9N4556-1🕒	0.62	“ "”“ "”“ \"”
8. L10N5913-1🕒	0.62	“ specially”“ "”“ '”
9. L9N8412+1🕒	0.61	“ "”“ actually”“!”
10. L9N4566+1🕒	0.61	“ **"”“ \"”“ `"”

Shadow specimen, apart from the ten: L11N8681-1🕒, -0.06.

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 11, neuron 8681, 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
L11N8681+1 t0-v0.1	1.062	0.58	1.344	1.42	Top boosted tokens ↑
L11N8681-1🕒 shadow specimen t0-v0.1	0.920	0.62	1.164	1.27	“ now”“ previously”“ certain”“ your”“ NOW”“ sometimes”“ new”“ especially”

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.