© 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 · 105231 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
11-RESID-POST-AA
105231

INDEX

Explanations

joint

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

 mnoha

-0.07

 Çağ

-0.07

 WWII

-0.06

 öldür

-0.06

 progressives

-0.06

ção

-0.06

 λα

-0.06

alım

-0.06

 nausea

-0.06

 vardır

-0.06

POSITIVE LOGITS

 jointly

0.13

 joint

0.13

 Joint

0.11

Joint

0.10

Ins

0.08

edith

0.08

arrants

0.07

 공동

0.07

	unit

0.07

投

0.07

Activations Density 0.004%

No Known Activations