© 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 · 85522 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Qwen2.5-7B-IT
Resid Post - 131k
11-RESID-POST-AA
85522

INDEX

Explanations

6

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

愍

-0.08

Kol

-0.08

YM

-0.07

祈

-0.07

.NotFound

-0.07

 affordable

-0.07

 basket

-0.07

 '<%=

-0.07

╙

-0.07

kład

-0.07

POSITIVE LOGITS

 feature

0.07

研究成果

0.07

 contrario

0.07

–↵↵

0.07

oretical

0.06

 felon

0.06

に入って

0.06

0.06

 offended

0.06

 datingside

0.06

Activations Density 0.007%

No Known Activations