© 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 · 36710 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
11-RESID-POST-AA
36710

INDEX

Explanations

FIG

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

orneys

-0.07

 Bien

-0.07

 wall

-0.07

Dup

-0.07

endale

-0.06

ана

-0.06

 handwritten

-0.06

#w

-0.06

 jewel

-0.06

 jeune

-0.06

POSITIVE LOGITS

、『

0.07

FIG

0.07

 subsidiaries

0.06

 omez

0.06

 conceive

0.06

 viability

0.06

¶

0.06

cobra

0.06

/ex

0.06

ubernetes

0.06

Activations Density 0.004%

No Known Activations