© 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 · 105305 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
11-RESID-POST-AA
105305

INDEX

Explanations

Apoptosis

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

 timestep

-0.07

 pedestal

-0.07

 TURN

-0.07

 spherical

-0.06

Tit

-0.06

ана

-0.06

mes

-0.06

 Phase

-0.06

 sectors

-0.06

LEVEL

-0.06

POSITIVE LOGITS

 Datensch

0.07

是一个

0.06

oding

0.06

 tiled

0.06

 Gins

0.06

ⓘ

0.06

/lab

0.06

/questions

0.06

icensing

0.06

注意

0.06

Activations Density 0.025%

No Known Activations