© 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 · 88160 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
11-RESID-POST-AA
88160

INDEX

Explanations

photosensitizers

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

 konut

-0.07

_cb

-0.06

 conosc

-0.06

 manufacturer

-0.06

cup

-0.06

richTextPanel

-0.06

 ArrayBuffer

-0.06

()`

-0.06

hour

-0.06

 Ок

-0.06

POSITIVE LOGITS

SSIP

0.07

дина

0.07

atics

0.07

딩

0.06

운

0.06

 Lower

0.06

annels

0.06

iveau

0.06

XYZ

0.06

 Destination

0.06

Activations Density 0.001%

No Known Activations