© 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 · 15-RESID-POST-AA · 60247 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
15-RESID-POST-AA
60247

INDEX

Explanations

code

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_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

.extension

-0.07

suppress

-0.06

玻璃

-0.06

("-");↵

-0.06

 ENTITY

-0.06

Roy

-0.06

 chant

-0.06

 scrolled

-0.06

 blow

-0.06

 Nord

-0.06

POSITIVE LOGITS

 flirt

0.06

0.06

cul

0.06

tur

0.06

bau

0.06

ピー

0.06

プラ

0.06

анная

0.06

.interfaces

0.06

lerde

0.06

Activations Density 0.043%

No Known Activations