© 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 · 19-RESID-POST-AA · 73486 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
19-RESID-POST-AA
73486

INDEX

Explanations

tom

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

'.

-0.07

Value

-0.07

 Value

-0.07

-0.06

 mirac

-0.06

"".

-0.06

 Якщо

-0.06

’.

-0.06

esting

-0.06

 فى

-0.06

POSITIVE LOGITS

 Forbes

0.08

/person

0.07

/int

0.06

ODE

0.06

 farmer

0.06

0.06

_reports

0.06

Tak

0.06

Mos

0.06

 DIST

0.06

Activations Density 0.002%

No Known Activations