© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

Jacobian LensNEW

Natural Language

NEW Assistant AxisNEW Circuit TracerUPDATESteer SAE Evals ExportsAPI Community Blog Privacy & Terms Contact

QWEN2.5-7B-IT · 11-RESID-POST-AA · 34557 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Qwen2.5-7B-IT
Resid Post - 131k
11-RESID-POST-AA
34557

INDEX

Explanations

Illustrated

np_max-act · gemini-2.0-flash

New Auto-Interp

Top Features by Cosine Similarity

Configuration

andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

Show PlotsShow ExplanationShow ActivationsShow Test FieldShow SteerShow Link

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 consensus

-0.08

 nuclear

-0.08

 görüşme

-0.07

涪

-0.07

鳕

-0.07

.percent

-0.07

北大

-0.06

<K

-0.06

есп

-0.06

 flew

-0.06

POSITIVE LOGITS

 DbSet

0.07

asename

0.07

的生命

0.07

ocytes

0.06

컁

0.06

?(:

0.06

אמר

0.06

ampp

0.06

_PL

0.06

ała

0.06

Activations Density 0.004%

No Known Activations