INDEX

Explanations

references to software functionalities and user instructions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

endphp

-0.49

 للاسماء

-0.46

 Valores

-0.42

 Kampagne

-0.41

faßt

-0.40

 rağmen

-0.40

PhysRev

-0.39

fortawesome

-0.38

Sơ

-0.38

 इसलिए

-0.38

POSITIVE LOGITS

 JpaRepository

0.51

ImageContext

0.50

ItemBackground

0.45

即可

0.45

withIdentifier

0.42

ніципалі

0.41

Спољашње

0.41

VJ

0.41

simpleType

0.41

 then

0.41

Activations Density 0.859%

references to software functionalities and user instructions

No Comments

No Known Activations