© 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 · 11-RESID-POST-AA · 52969 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
11-RESID-POST-AA
52969

INDEX

Explanations

RAS

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_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.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

 Hiro

-0.07

 Vanilla

-0.07

 Heads

-0.07

 Scrap

-0.07

 File

-0.07

enne

-0.07

_services

-0.07

Drop

-0.06

LOC

-0.06

cow

-0.06

POSITIVE LOGITS

不知

0.08

_argument

0.07

 Revised

0.07

-toggler

0.07

bedtls

0.07

ManagedObject

0.07

§

0.06

*******↵

0.06

CTSTR

0.06

項目

0.06

Activations Density 0.005%

No Known Activations