© 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 · 115045 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
15-RESID-POST-AA
115045

INDEX

Explanations

punctuation

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

 Cher

-0.07

Ral

-0.07

.Lines

-0.07

 vrát

-0.07

apt

-0.07

 allied

-0.07

tpl

-0.07

 graded

-0.06

 Follow

-0.06

 appare

-0.06

POSITIVE LOGITS

_tables

0.06

ẩu

0.06

jb

0.06

 listings

0.06

 ніж

0.06

0.06

 phần

0.06

xC

0.06

-quality

0.06

 outlet

0.06

Activations Density 0.048%

No Known Activations