© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

Natural Language

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

QWEN2.5-7B-IT · 15-RESID-POST-AA · 111574 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Qwen2.5-7B-IT
Resid Post - 131k
15-RESID-POST-AA
111574

INDEX

Explanations

="

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

 plag

-0.07

刷新

-0.06

 savoir

-0.06

 succesfully

-0.06

/{{$

-0.06

 isnt

-0.06

YOU

-0.06

냬

-0.06

lege

-0.06

体现

-0.06

POSITIVE LOGITS

 pBuffer

0.07

攽

0.07

铗

0.07

ifter

0.07

EventManager

0.07

EXTERNAL

0.07

쾌

0.06

0.06

מפגש

0.06

 frame

0.06

Activations Density 0.002%

No Known Activations