© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

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

LLAMA3.1-8B-IT · 11-RESID-POST-AA · 53932 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
11-RESID-POST-AA
53932

INDEX

Explanations

Patrick

np_max-act · gemini-2.0-flash

New Auto-Interp

Top Features by Cosine Similarity

Configuration

andyrdt/saes-llama-3.1-8b-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

 Mosque

-0.06

-0.06

सभ

-0.06

 कल

-0.06

-0.06

olulu

-0.06

-0.06

-0.06

 getters

-0.05

 свид

-0.05

POSITIVE LOGITS

 cursed

0.08

 Japanese

0.07

aying

0.07

 treating

0.07

rimp

0.07

-Americans

0.07

That

0.07

ателем

0.06

 Open

0.06

 graves

0.06

Activations Density 0.006%

No Known Activations