© 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 · 7-RESID-POST-AA · 49279 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
7-RESID-POST-AA
49279

INDEX

Explanations

existing

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_7/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.7.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

 nejvyšší

-0.07

 sourced

-0.06

-0.06

CAL

-0.06

 pumping

-0.06

φή

-0.06

ercises

-0.06

and

-0.06

-cigaret

-0.06

.slot

-0.06

POSITIVE LOGITS

 existing

0.10

 Existing

0.08

holder

0.06

$$$

0.06

ιος

0.06

 appellant

0.06

енность

0.06

Void

0.06

="↵

0.06

oncé

0.06

Activations Density 0.014%

No Known Activations