INDEX

Explanations

derive

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_7/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.7.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Texas

-0.07

 weakSelf

-0.07

 Seventh

-0.07

 Managed

-0.07

 occupies

-0.07

ORG

-0.07

wz

-0.06

αιν

-0.06

amous

-0.06

 catalogue

-0.06

POSITIVE LOGITS

=headers

0.06

derived

0.06

 derived

0.06

ivor

0.06

 auditor

0.06

 практически

0.06

ební

0.06

 기업

0.06

 batt

0.06

订

0.06

Activations Density 0.021%

derive

No Comments

No Known Activations