© 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 · 105283 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
11-RESID-POST-AA
105283

INDEX

Explanations

"<"

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

Uno

-0.07

 telephone

-0.07

เช

-0.07

 husband

-0.06

-0.06

 treating

-0.06

ils

-0.06

 Pass

-0.06

who

-0.06

fou

-0.06

POSITIVE LOGITS

 ethanol

0.08

 vidéos

0.08

 Unlimited

0.06

Ρ

0.06

боратор

0.06

카지노

0.06

alış

0.06

 تجاری

0.06

 خط

0.06

 crud

0.06

Activations Density 0.008%

No Known Activations