INDEX

Explanations

data

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

andyrdt/saes-qwen2.5-7b-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

变化

-0.08

考试

-0.08

 because

-0.07

 some

-0.07

	switch

-0.07

ization

-0.07

 dieser

-0.07

 converted

-0.07

沧

-0.06

 those

-0.06

POSITIVE LOGITS

℉

0.08

SKF

0.07

دب

0.07

 Journalism

0.07

foregroundColor

0.07

 الخيارات

0.07

温情

0.07

 evolutionary

0.06

_BACKGROUND

0.06

Activations Density 0.136%

data

No Comments

No Known Activations