© 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 · 25599 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
11-RESID-POST-AA
25599

INDEX

Explanations

primary

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

ход

-0.06

tag

-0.06

-0.06

文

-0.06

ходить

-0.06

 Light

-0.06

_PE

-0.06

ご

-0.06

 apare

-0.06

 Dancing

-0.06

POSITIVE LOGITS

 gather

0.07

CSR

0.07

 đài

0.07

.newLine

0.06

 utmost

0.06

_throw

0.06

+p

0.06

 terms

0.06

 unbelievable

0.06

Specifies

0.06

Activations Density 0.002%

No Known Activations