© 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 · 3-RESID-POST-AA · 94151 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
3-RESID-POST-AA
94151

INDEX

Explanations

did

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_3/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.3.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

-aos

-0.07

mkdir

-0.06

 promises

-0.06

οντας

-0.06

Gener

-0.06

Mongo

-0.06

 Appeal

-0.06

Con

-0.06

-speed

-0.06

chrono

-0.06

POSITIVE LOGITS

Did

0.19

Did

0.18

did

0.17

DID

0.14

did

0.13

 didn

0.11

 Didn

0.09

.did

0.08

Didn

0.07

didn

0.07

Activations Density 0.018%

No Known Activations