© 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 · 51967 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
19-RESID-POST-AA
51967

INDEX

Explanations

Marked up text

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

층

-0.06

England

-0.06

 obsessed

-0.06

 footing

-0.06

ulação

-0.05

meleri

-0.05

озем

-0.05

_slice

-0.05

.assertNull

-0.05

 belie

-0.05

POSITIVE LOGITS

|wx

0.07

 Rest

0.07

 corrobor

0.07

Le

0.07

Rot

0.07

rotate

0.07

 puppy

0.07

0.06

 Buyer

0.06

ROT

0.06

Activations Density 0.001%

No Known Activations