Specimen ledger / L14N2344+1

Layer 14 · Neuron 2344 · boosted

L14N2344+1

Word quirks:

“ ‘”“‘”“ ‘’”“ __”“ **‘”“ basically”“ ____”“ ’”

💤 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, a large-scale language model developed by Tongyi Lab. My core capabilities include:

Precise reasoning: I can perform complex mathematical calculations and logical analysis with high accuracy.
Multilingual support: I support fluent interaction in over 100 languages worldwide.
Visual analysis: I can interpret charts, diagrams, and scientific data effectively.
Code generation: I can generate Python code and debug efficiently.
Knowledge cutoff: My knowledge base was updated in early 2024.

My mission is to empower users with efficient, accurate,

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. L7N1574-1💤	0.96	“ ‘”“ **‘”“ '”
2. L12N707-1🕒	0.96	“ ‘”“ ‘’”“ **‘”
3. L8N8853-1💤	0.95	“ ‘”“ “”“ ””
4. L17N2460-1🕒	0.95	“ ‘”“ ’”“ ‚”
5. L9N8400+1🕒	0.95	“ ‘”“‘”“ '”
6. L6N5219-1🕒	0.95	“ ‘”“ **‘”“ ‘’”
7. L4N2833+1🕒	0.95	“‘”“ ‘’”“ ‘”
8. L9N1194+1💤	0.95	“ ‘”“ ””“‘”
9. L6N987-1💤	0.95	“ ‘”“ ’”“ ””
10. L11N3621-1💤	0.95	“ ‘”“ ‘’”“ ””

Shadow specimen, apart from the ten: L14N2344-1🕒, -0.03.

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 14, neuron 2344, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.68, 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
L14N2344+1 t0-v0.1	0.679	0.68	0.920	1.45	Top boosted tokens ↑
L14N2344-1🕒 shadow specimen t0-v0.1	1.027	0.58	1.217	1.55	“ ⏎”“ ⏎⏎”“提供任何”“ any”“ this”“ description”“ ⏎”“ ⏎ ⏎”

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.