Specimen ledger / L14N5863+1

Layer 14 · Neuron 5863 · boosted

L14N5863+1

Word quirks:

“ your”“ yourself”“ finally”“ again”“—you”“ you”“ needed”“ yours”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

Hello! I am Qwen3.5, the large language model developed by Tongyi Lab. I am your personal assistant.

I am a highly advanced version of myself, an upgraded version of my predecessor, Qwen3.5. You can call me "Qwen3.5."

---

How I can help you:

I am a large language model created by Alibaba Cloud's Tongyi Lab. I am your personal assistant. I have been upgraded in all capabilities compared to my original version.

Let me know how I can assist you!

---

What new abilities I have:

Compared

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. L16N1892+1💤	0.89	“ you”“ yourself”“ YOU”
2. L11N8865+1🕒	0.88	“ your”“ YOU”“ you”
3. L9N6833+1💤	0.88	“ your”“ yourself”“ you”
4. L12N2614+1🕒	0.88	“ your”“ you”“ yourself”
5. L15N1853+1🕒	0.87	“ you”“ your”“ now”
6. L12N4154+1💤	0.87	“ your”“ you”“ now”
7. L11N8411-1💤	0.87	“ your”“ yourself”“ you”
8. L11N2678-1🕒	0.87	“ your”“ yourself”“ yours”
9. L13N825-1💤	0.87	“ you”“ will”“—you”
10. L12N1333+1🕒	0.87	“ your”“ you”“ YOU”

Shadow specimen, apart from the ten: L14N5863-1💤, -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 14, neuron 5863, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.86, top-1 agreement 0.66) 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
L14N5863+1 t0-v0.1	0.862	0.66	1.102	1.47	Top boosted tokens ↑
L14N5863-1💤 shadow specimen t0-v0.1	0.886	0.61	1.125	1.64	“ discussion”“ unclear”“ **“”“ emphasis”“ debate”“ represented”“?””“ question”

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.