INDEX

Explanations

Emphasizing adverbs

np_max-act · gemini-2.0-flash

New Auto-Interp

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

}%

-0.06

swers

-0.06

 Particularly

-0.06

 equipe

-0.06

Cou

-0.06

EA

-0.06

 controversy

-0.06

 CONTR

-0.06

 substrate

-0.06

ίθ

-0.06

POSITIVE LOGITS

 probably

0.08

 just

0.07

-know

0.06

 always

0.06

probably

0.06

 single

0.06

"All

0.06

 الى

0.06

>Total

0.06

.setVerticalGroup

0.06

Activations Density 0.218%

Emphasizing adverbs

No Comments

No Known Activations