© 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 · 19-RESID-POST-AA · 110536 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Qwen2.5-7B-IT
Resid Post - 131k
19-RESID-POST-AA
110536

INDEX

Explanations

quotation marks

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Top Features by Cosine Similarity

Configuration

andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_19/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.19.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

 facile

-0.08

_restore

-0.07

בינו

-0.07

uest

-0.07

 rebut

-0.07

潔

-0.07

 Naruto

-0.07

 primeiro

-0.07

 Trước

-0.07

小說

-0.07

POSITIVE LOGITS

WXYZ

0.07

'?'

0.07

变换

0.07

 waves

0.07

 balls

0.07

.gener

0.07

 Forms

0.07

 rings

0.07

waves

0.07

命令

0.07

Activations Density 0.023%

No Known Activations