Specimen ledger / L6N2095-1

Layer 6 · Neuron 2095 · suppressed

L6N2095-1

Word quirks:

“⏎”“ '”“⏎ ⏎”“ ⏎”“ info”“ _____”“ __”“ \"”

💤 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 language model developed by Alibaba Cloud's Tongyi Lab. My mission is to provide advanced capabilities in natural language processing, including real-time dialogue understanding, intelligent code generation, and multi-modal interaction. I can also assist with tasks like summarization, question answering, and text classification. How can I help you?
</think>

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. L6N6244-1💤	0.94	“⏎”“ \"”“⏎ ⏎”
2. L8N6596-1💤	0.92	“⏎”“ \"”“ ⏎”
3. L11N8033-1💤	0.92	“⏎”“ \"”“ ⏎”
4. L6N5945+1💤	0.92	“⏎”“⏎⏎”“ __”
5. L5N7396-1🕒	0.91	“⏎”“ ⏎”“⏎ ⏎”
6. L13N6298-1💤	0.91	“ __”“⏎”“ \"”
7. L7N6877-1💤	0.91	“⏎”“ ‘”“ '”
8. L8N4546-1💤	0.91	“ \"”“ ⏎”“ __”
9. L6N2917+1🕒	0.90	“<\|im_end\|>”“⏎ ⏎”“ \"”
10. L6N2502-1🕒	0.90	“⏎”“ \"”“⏎ ⏎”

Shadow specimen, apart from the ten: L6N2095+1🕒, 0.28.

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 6, neuron 2095, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.96, top-1 agreement 0.61) 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
L6N2095+1🕒 shadow specimen t0-v0.1	1.379	0.55	1.671	1.76	“<\|im_end\|>”“⏎”“...\”“ displaying”“ display”“ visual”“ displayed”“:\”
L6N2095-1 t0-v0.1	0.957	0.61	1.197	1.76	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.