© 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 · 125205 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
15-RESID-POST-AA
125205

INDEX

Explanations

CA

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

摸

-0.07

 organs

-0.06

_lat

-0.06

정보

-0.06

monto

-0.06

entů

-0.06

<context

-0.06

 isteyen

-0.06

�

-0.06

 overlapping

-0.06

POSITIVE LOGITS

 Visual

0.08

TableModel

0.07

MouseEvent

0.06

 Blur

0.06

ằng

0.06

Passed

0.06

formatted

0.06

vak

0.06

 surprise

0.06

 Modern

0.06

Activations Density 0.003%

No Known Activations