© 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 · 15-RESID-POST-AA · 125064 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
15-RESID-POST-AA
125064

INDEX

Explanations

forget

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_15/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.15.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

 sitios

-0.07

 dific

-0.07

동안

-0.06

productos

-0.06

かわ

-0.06

Semantic

-0.06

Jog

-0.06

 sitio

-0.06

 фрукт

-0.06

 полі

-0.06

POSITIVE LOGITS

REG

0.07

"#"

0.06

娜

0.06

DES

0.06

.ORDER

0.06

 Silver

0.06

рин

0.06

Silver

0.06

>Status

0.06

0.06

Activations Density 0.013%

No Known Activations