INDEX

Explanations

references to complex concepts and conditions affecting data and information

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 좋아

-0.31

 Usted

-0.29

 Charakter

-0.29

 Geschw

-0.28

 muchacha

-0.27

 desconhe

-0.27

 maisha

-0.27

Kapitel

-0.26

ด้าน

-0.26

 Kapitel

-0.26

POSITIVE LOGITS

$_['

0.87

 שוליים

0.84

 BoxFit

0.82

parsedMessage

0.82

featureID

0.79

majánló

0.75

WebElementEntity

0.73

providedIn

0.73

 CascadeType

0.73

styleable

0.72

Activations Density 0.929%

references to complex concepts and conditions affecting data and information

No Known Activations