© 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 · 25367 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
15-RESID-POST-AA
25367

INDEX

Explanations

Court

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

Ά

-0.07

焼

-0.06

marca

-0.06

ีโ

-0.06

 Whether

-0.06

輸

-0.06

uko

-0.06

 pits

-0.06

sit

-0.06

 rake

-0.06

POSITIVE LOGITS

 steroids

0.07

δο

0.06

apollo

0.06

casts

0.06

Jay

0.06

Classifier

0.06

Resultado

0.06

	stats

0.06

.Qual

0.06

vely

0.06

Activations Density 0.000%

No Known Activations