© 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 · 3-RESID-POST-AA · 121916 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
3-RESID-POST-AA
121916

INDEX

Explanations

ub

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_3/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.3.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

_stats

-0.07

 wards

-0.07

ону

-0.07

blast

-0.06

.room

-0.06

IPS

-0.06

不足

-0.06

 آپ

-0.06

 Olson

-0.06

KEY

-0.06

POSITIVE LOGITS

 bushes

0.12

 procrast

0.10

 Perm

0.10

shr

0.10

crast

0.08

ub

0.08

 perm

0.07

Perm

0.07

 Must

0.07

 rodents

0.07

Activations Density 0.003%

No Known Activations