© 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 · 102087 ｜ Neuronpedia

Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
15-RESID-POST-AA
102087

INDEX

Explanations

case

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

 hver

-0.07

 Marie

-0.06

国产

-0.06

#End

-0.06

 Netflix

-0.06

 repeal

-0.06

 Bitte

-0.06

 Encounter

-0.06

	sleep

-0.06

 concession

-0.06

POSITIVE LOGITS

_validate

0.07

[out

0.06

size

0.06

Earn

0.06

 annihil

0.06

иком

0.06

urn

0.06

.ai

0.06

_vec

0.06

.ov

0.06

Activations Density 0.021%

No Known Activations