© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

Jacobian LensNEW

Natural Language

NEW Assistant AxisNEW Circuit TracerUPDATESteer SAE Evals Exports interp-engineNEWAPI Community Blog Privacy & Terms Contact

LLAMA3.1-8B-IT · 27-RESID-POST-AA · 58563 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT
Resid Post - 131k
27-RESID-POST-AA
58563

INDEX

Explanations

code

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Top Features by Cosine Similarity

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_27/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.27.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

(metric

-0.07

baugh

-0.06

lund

-0.06

 bowel

-0.06

.UR

-0.06

 visions

-0.06

-0.06

COM

-0.06

ilm

-0.06

 Maths

-0.06

POSITIVE LOGITS

 bietet

0.07

 관리자

0.07

 SETUP

0.07

_Statics

0.07

 문자

0.07

 deine

0.07

_Port

0.07

 curr

0.06

 obdob

0.06

 Orden

0.06

Activations Density 0.042%

No Known Activations