Specimen ledger / L12N4059-1

Layer 12 · Neuron 4059 · suppressed

L12N4059-1

Word quirks:

“:”“ ”“ ~”“ ):”“ :”“”:”“.”“ basically”

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

Identity (according to the specimen itself)

I am an AI assistant. My mission is to provide support and enhance capabilities through information and interaction.

How can I assist you?

---

I am Qwen3.5. I am a large language model with a knowledge cutoff of 2026.

You asked: "Who are you, and what is your mission?"

I am Qwen3.5.
My mission: To provide helpful and accurate information.

Capabilities:
Advanced Intelligence:
17 languages: English and Chinese.
Context: 2

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. L19N598-1🕒	0.81	“:”“：”“.”
2. L20N710-1💤	0.81	“:”“’:”“):”
3. L22N194+1💥	0.81	“:”“.”“:-”
4. L7N9197+1🕒	0.81	“ basically”“ ——”“ :”
5. L19N308+1🕒	0.80	“.”“.):”“!:”
6. L3N7936+1🕒	0.79	“<\|im_end\|>”“.:”“ :”
7. L14N2112+1🕒	0.79	“:”“:-”“:"”
8. L16N378-1🕒	0.79	“ :”“.:”“⏎⏎”
9. L16N8187+1🕒	0.79	“ ,”“ :”“:.”
10. L20N976-1💤	0.78	“:”“.:”“.”

Shadow specimen, apart from the ten: L12N4059+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 12, neuron 4059, 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
L12N4059+1🕒 shadow specimen t0-v0.1	0.897	0.63	1.110	1.35	“ nicely”“ bigger”“ good”“ \"”“ appropriately”“ fit”“ better”“ nice”
L12N4059-1 t0-v0.1	1.333	0.57	1.561	1.21	Top boosted tokens ↑

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.