INDEX

Explanations

technical terms and parameters related to credential management and access control

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

terms related to credentials and tasks in the context of commands or operations

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

code, addresses, and quantities

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

 للاسماء

-0.60

 suicidal

-0.59

 ***!

-0.59

 Wikispecies

-0.58

PrototypeOf

-0.58

 estekak

-0.56

 FishBase

-0.56

 betweenstory

-0.55

存于互联网档案馆

-0.55

POSITIVE LOGITS

rinfo

0.60

endregion

0.52

ècie

0.52

Espèce

0.52

yves

0.51

こちらも

0.49

AsUp

0.48

</strong>

0.48

Getenv

0.47

धारण

0.47

Activations Density 0.652%

technical terms and parameters related to credential management and access control

terms related to credentials and tasks in the context of commands or operations

code, addresses, and quantities

No Comments

No Known Activations