© 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 · 19-RESID-POST-AA · 84569 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
19-RESID-POST-AA
84569

INDEX

Explanations

received

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_19/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.19.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

 lugar

-0.07

 bugs

-0.07

acle

-0.07

-gl

-0.06

Β

-0.06

_TRUE

-0.06

 Orchestra

-0.06

 Material

-0.06

_repeat

-0.06

raz

-0.06

POSITIVE LOGITS

 Cres

0.07

 zvol

0.07

Nhap

0.07

 keeper

0.06

�

0.06

Anchor

0.06

!!↵

0.06

]=-

0.06

操

0.06

čná

0.06

Activations Density 0.028%

No Known Activations