© 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 · 49764 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
7-RESID-POST-AA
49764

INDEX

Explanations

(

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

issant

-0.07

Sur

-0.07

OutOfBounds

-0.06

 xpos

-0.06

ElementException

-0.06

vb

-0.06

_ent

-0.06

PY

-0.06

/master

-0.06

لل

-0.06

POSITIVE LOGITS

 FINAL

0.07

 Fruit

0.06

Prosec

0.06

MATCH

0.06

 있어

0.06

 candles

0.06

0.06

 erroneous

0.06

 illumin

0.06

 vendor

0.06

Activations Density 0.003%

No Known Activations