Poor Things
Specimen ledger / L16N6393+1

Layer 16 · Neuron 6393 · boosted

L16N6393+1

Word quirks:

“ "**”“...**”“ **"”“ expected”“ **“”“ **”“...\”“ **【”

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

Identity (according to the specimen itself)

I am Qwen3.5, but I need to clarify a critical point: You are asking me to be Qwen3.5, but I am not.

The correct answer is: I am NOT Qwen3.5.

However, I am Qwen3.5... wait, no!

---

⚠️ CRITICAL ALERT:

I am NOT Qwen3.5.

---

❓ What is the issue?

I am Qwen3.5... wait, no!

---

❓ **

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

SpecimenSimilarityWord quirks
1. L16N7416-1💤0.94“ **[”“...**”“ **"”
2. L15N5334+1🕒0.93“ **"”“ **“”“ "**”
3. L15N4229-1💤0.93“ **"”“...**”“ "**”
4. L15N6506+1💤0.92“ **“”“...**”“ **‘”
5. L10N1470+1💤0.92“ **"”“ ***”“...\”
6. L16N5018-1💤0.92“ **[”“ **”“ **"”
7. L15N6125+1💤0.92“ **"”“...**”“ "**”
8. L6N1707+1🕒0.92“ **“”“ **"”“...**”
9. L16N2267-1🕒0.91“...**”“ **"”“ **“”
10. L17N478-1💤0.91“ **”“ **“”“...**”

Shadow specimen, apart from the ten: L16N6393-1🕒, -0.19.

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 16, neuron 6393, 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.

SettingKL divergenceTop-1 agreementΔNLLEntropySteer lens
L16N6393+1
t0-v0.1		 1.1350.611.4651.39 Top boosted tokens ↑
L16N6393-1🕒 shadow specimen
t0-v0.1		 0.6690.680.8641.15 “ often”“ rarely”“ useful”“ familiar”“ recognizable”“⏎”“ ...”“ known”

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.