© 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 · 7-RESID-POST-AA · 86267 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
7-RESID-POST-AA
86267

INDEX

Explanations

flawed

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_7/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.7.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

activity

-0.08

 possibilit

-0.08

ici

-0.07

gricult

-0.07

ocious

-0.07

_emit

-0.07

agit

-0.06

 adventurous

-0.06

astic

-0.06

ерг

-0.06

POSITIVE LOGITS

 flaw

0.18

 flaws

0.16

 flawed

0.13

 flawless

0.08

 Blow

0.08

Jaw

0.07

Tw

0.07

 flour

0.07

Dig

0.07

 Dream

0.07

Activations Density 0.002%

No Known Activations