© 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 · 126490 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Qwen2.5-7B-IT
Resid Post - 131k
15-RESID-POST-AA
126490

INDEX

Explanations

products

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

Pink

-0.07

 song

-0.07

 BEGIN

-0.07

괭

-0.07

 amph

-0.07

𬶭

-0.06

 Porn

-0.06

᭺

-0.06

jd

-0.06

굄

-0.06

POSITIVE LOGITS

 communications

0.08

 Requirement

0.08

 injuring

0.08

 satisfactory

0.08

ucc

0.08

大大小

0.07

业务

0.07

 Subscription

0.07

 criterion

0.07

yog

0.07

Activations Density 0.030%

No Known Activations