INDEX

Explanations

affirmative words or phrases indicating agreement or approval

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

/**

-0.63

<?

-0.47

$_['

-0.46

 FetchType

-0.41

 Castle

-0.41

tede

-0.41

ButtonModule

-0.41

Castle

-0.40

 castle

-0.40

DotNetBar

-0.40

POSITIVE LOGITS

ValueStyle

0.52

FunctionFlags

0.50

Kaynakça

0.47

 HasFactory

0.47

 <>",

0.45

ViewImports

0.45

haikusbot

0.42

 lendemain

0.41

Референце

0.40

 resourceCulture

0.40

Activations Density 0.000%

affirmative words or phrases indicating agreement or approval

No Comments

No Known Activations