INDEX

Explanations

Her

np_max-act · gemini-2.0-flash

This neuron activates on the assistant’s generated image‐prompt or scene‐description text—i.e. detailed AI art prompt language.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 seeks

-0.07

-awesome

-0.06

,file

-0.06

_lead

-0.06

Seen

-0.06

يمة

-0.06

_sl

-0.06

employees

-0.06

出去

-0.06

 Amazing

-0.06

POSITIVE LOGITS

EGA

0.07

契

0.07

 него

0.07

zł

0.06

 gamma

0.06

 Candy

0.06

Hola

0.06

ш

0.06

◎

0.06

 Franti

0.06

Activations Density 0.018%

Her

This neuron activates on the assistant’s generated image‐prompt or scene‐description text—i.e. detailed AI art prompt language.

No Comments

No Known Activations