© 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 · 89341 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
15-RESID-POST-AA
89341

INDEX

Explanations

percent

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

 precise

-0.07

普

-0.06

DX

-0.06

鮮

-0.06

bum

-0.06

[n

-0.06

 thou

-0.06

 setResult

-0.06

 Zion

-0.06

ких

-0.06

POSITIVE LOGITS

0.07

.Pointer

0.07

-degree

0.07

-"

0.06

Rew

0.06

-camera

0.06

Ingredient

0.06

 colour

0.06

nak

0.06

 firefight

0.06

Activations Density 0.025%

No Known Activations