© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

Jacobian LensNEW

Natural Language

NEW Assistant AxisNEW Circuit TracerUPDATESteer SAE Evals Exports interp-engineNEWAPI Community Blog Privacy & Terms Contact

QWEN3.5-9B-PT · 15-QWENSCOPE-RES-64K · 64875 ｜ Neuronpedia

Home
Qwen3.5-9B Base
15-QWENSCOPE-RES-64K
64875

INDEX

Explanations

diagonal traverse

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

No Configuration Found

Embeds

Show PlotsShow ExplanationShow ActivationsShow Test FieldShow SteerShow Link

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

amah

-0.07

åľ¨çİ°åľº

-0.07

allah

-0.07

ä¾Ľ

-0.07

 tÃ¡i

-0.06

éĹ

-0.06

éĥĬ

-0.06

ovich

-0.06

 Sullivan

-0.06

reff

-0.06

POSITIVE LOGITS

 diagonal

0.12

æĸľ

0.11

 diagon

0.10

å¯¹è§Ĵ

0.09

_diag

0.08

diag

0.08

anter

0.08

.trace

0.07

_projection

0.07

èµ·çĤ¹

0.07

Activations Density 0.009%

No Known Activations