INDEX

Explanations

numerical information or sequences related to lists or instructions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Text associated with numbered lists

np_acts-logits-general · gemini-2.0-flash

numbered list item 2

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 itself

-0.42

 simply

-0.41

rana

-0.41

+]

-0.40

 olmayan

-0.40

Footnotes

-0.40

isante

-0.40

</table>

-0.39

خصة

-0.39

 estrada

-0.38

POSITIVE LOGITS

IsContent

0.86

principalColumn

0.80

انيف

0.80

Rhestr

0.80

MigrationBuilder

0.80

 وتسجيلات

0.79

__":

0.75

 réessayer

0.74

styleType

0.74

 iNdEx

0.73

Activations Density 0.142%

numerical information or sequences related to lists or instructions

Text associated with numbered lists

numbered list item 2

No Comments

No Known Activations