INDEX

Explanations

numerical values associated with specific data structures or API versioning in a coding context

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

SpringRunner

-0.33

 taches

-0.33

 resep

-0.29

 rodzaju

-0.28

 naturen

-0.28

 silnika

-0.28

 Republic

-0.28

 balles

-0.28

ニーカー

-0.27

 seus

-0.26

POSITIVE LOGITS

 Administrativna

0.84

 незавершена

0.74

NameInMap

0.71

IsMutable

0.68

ViewFeatures

0.67

 Италијани

0.66

Портали

0.65

 estekak

0.65

ronpa

0.64

principalTable

0.63

Activations Density 0.013%

numerical values associated with specific data structures or API versioning in a coding context

No Comments

No Known Activations