INDEX

Explanations

instances of the word "new."

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_1m/average_l0_107

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

1,048,576

Data Type

float32

Hook Name

blocks.12.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

+#+

-0.51

Title

-0.51

 urbanas

-0.50

 externi

-0.50

horabuena

-0.49

 inhoud

-0.47

DIPSETTING

-0.46

ürzen

-0.46

 Title

-0.46

 geweest

-0.46

POSITIVE LOGITS

Autoritní

0.66

ValueGenerated

0.64

 StatelessWidget

0.64

 AssemblyCulture

0.63

تقاوى

0.62

 RoutedEventArgs

0.60

RetentionPolicy

0.58

phosa

0.57

ulemon

0.57

pulumi

0.56

Activations Density 0.000%

instances of the word "new."

No Comments

No Known Activations