© 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 · 37030 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
15-RESID-POST-AA
37030

INDEX

Explanations

LaTeX code

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

.drawText

-0.07

 unzip

-0.07

 Naruto

-0.07

-0.06

 руках

-0.06

な

-0.06

"d

-0.06

仁

-0.06

.DrawString

-0.06

(font

-0.06

POSITIVE LOGITS

 chaos

0.07

обра�

0.07

−

0.06

 Pornhub

0.06

ektiv

0.06

 continuity

0.06

 angle

0.06

@example

0.06

andal

0.06

IFIED

0.06

Activations Density 0.009%

No Known Activations