INDEX

Explanations

technical references and commands related to programming libraries and APIs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 defaultstate

-0.71

RectangleBorder

-0.46

ंदीखरीदारी

-0.46

Dislikes

-0.46

cshtml

-0.44

 kasarigan

-0.44

 surla

-0.43

RenderAtEndOf

-0.42

Datuak

-0.41

 ellers

-0.41

POSITIVE LOGITS

XmlAccessorType

0.40

TemporalType

0.40

 setPassword

0.38

 useSelector

0.38

ETTA

0.37

 useDispatch

0.37

├──

0.36

 mand

0.35

expectedResult

0.35

ChildIndex

0.35

Activations Density 0.126%

technical references and commands related to programming libraries and APIs

No Comments

No Known Activations