© 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

QWEN2.5-7B-IT · 15-RESID-POST-AA · 49902 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Qwen2.5-7B-IT
Resid Post - 131k
15-RESID-POST-AA
49902

INDEX

Explanations

border

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_15/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.15.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

诅

-0.07

withErrors

-0.07

.Progress

-0.07

 Richards

-0.06

 TestUtils

-0.06

.Can

-0.06

Lng

-0.06

ENDIF

-0.06

ombine

-0.06

尚

-0.06

POSITIVE LOGITS

 개인

0.07

eworthy

0.07

𝚎

0.07

dz

0.07

card

0.06

 raid

0.06

쌌

0.06

 settlement

0.06

 Small

0.06

მ

0.06

Activations Density 0.002%

No Known Activations