© 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 · 7-RESID-POST-AA · 38386 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Qwen2.5-7B-IT
Resid Post - 131k
7-RESID-POST-AA
38386

INDEX

Explanations

okay

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_7/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.7.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

 diameter

-0.07

jet

-0.07

즙

-0.07

比特

-0.07

pytest

-0.07

超

-0.07

ཚ

-0.07

当即

-0.07

宅

-0.07

_executor

-0.07

POSITIVE LOGITS

몯

0.08

 Mariners

0.08

keep

0.07

Have

0.07

(","

0.07

 הנאשם

0.06

 carc

0.06

酴

0.06

 Shoulder

0.06

 Respir

0.06

Activations Density 0.005%

No Known Activations