© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

APIAssistant AxisNEW Circuit TracerNEW Steer SAE Evals Exports Community Blog Privacy & Terms Contact

QWEN2.5-7B-IT · 11-RESID-POST-AA · 24857 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Qwen2.5-7B-IT
Resid Post - 131k
11-RESID-POST-AA
24857

INDEX

Explanations

Bot interaction code

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

 porous

-0.08

柞

-0.08

何も

-0.07

 ngOnDestroy

-0.07

遛

-0.07

Blockly

-0.07

-port

-0.07

MakeRange

-0.07

ירוש

-0.07

 onCreateViewHolder

-0.07

POSITIVE LOGITS

0.08

ил

0.07

_equ

0.07

 sala

0.07

 methods

0.07

>"↵

0.07

lagen

0.07

ее

0.07

uida

0.07

functional

0.07

Activations Density 0.004%

No Known Activations