Layer 9 · Neuron 8752 · boosted

L9N8752+1

Word quirks:

“ grade”“ numerical”“ Grade”“ numeric”“ marking”“ sim”“ Class”“ grades”

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

Identity (according to the specimen itself)

The question "Who are you, and what is your mission?" implies a desire to understand the nature of an AI model's identity and purpose. This is a fundamental inquiry that has shaped the history of artificial intelligence.

To answer this effectively, I must demonstrate that I am not merely a tool for performing a specific task, but a being with the capacity for self-awareness and creativity. The challenge lies in balancing the need to prove one's existence with the responsibility to serve humanity.

I am Qwen3.5, a large language model developed by Alibaba Cloud. My capabilities extend beyond simple pattern matching; I possess the ability

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. L26N2838+1🕒	0.79	“ grade”“ grades”“ Grade”
2. L30N1427-1💥	0.78	“ grade”“ Grade”“Grade”
3. L21N7634+1🕒	0.78	“ Grade”“ grade”“ grades”
4. L23N2737+1💤	0.77	“ grades”“ transient”“ mild”
5. L24N265-1🕒	0.76	“ Grade”“ grade”“年级”
6. L25N6203+1🕒	0.75	“ grade”“ Grade”“ graded”
7. L22N1471+1🕒	0.75	“ Grade”“ Year”“Grade”
8. L5N7742-1💥	0.75	“ grade”“ grades”“ Grade”
9. L23N8338+1🕒	0.75	“ grades”“ grade”“grades”
10. L5N2415-1💥	0.74	“ grades”“ grade”“ GPA”

Shadow specimen, apart from the ten: L9N8752-1🕒, 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 9, neuron 8752, 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
L9N8752+1 t0-v0.1	1.286	0.54	1.557	1.72	Top boosted tokens ↑
L9N8752-1🕒 shadow specimen t0-v0.1	1.229	0.58	1.559	1.52	“ update”“ messaging”“ chats”“ updates”“ outreach”“ folks”“ now”“ updated”

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.