© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

Natural Language

NEW Assistant AxisNEW Circuit TracerUPDATESteer SAE Evals ExportsAPI Community Blog Privacy & Terms Contact

LLAMA3.1-8B-IT · 19-RESID-POST-AA · 35990 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT
Resid Post - 131k
19-RESID-POST-AA
35990

INDEX

Explanations

Begins with "now"

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_19/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.19.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

้ก

-0.07

adb

-0.07

oured

-0.06

 Subaru

-0.06

 равно

-0.06

šp

-0.06

 Pamela

-0.06

assign

-0.06

aisy

-0.06

 Peek

-0.06

POSITIVE LOGITS

_hot

0.08

 DIST

0.07

_CONTEXT

0.07

 photographs

0.06

 uLocal

0.06

 memorable

0.06

VF

0.06

_keyboard

0.06

 mastering

0.06

_irq

0.06

Activations Density 0.002%

No Known Activations