INDEX

Explanations

terms related to espionage and spying

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.0.hook_mlp_out

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ļª

-1.71

 transgender

-1.60

 mourn

-1.56

 behalf

-1.53

 lonely

-1.48

TRODUCTION

-1.44

 Hispanic

-1.42

 despair

-1.39

 welcome

-1.37

 griev

-1.37

POSITIVE LOGITS

bilt

1.92

burg

1.88

holder

1.81

craft

1.73

ium

1.71

bur

1.68

ious

1.66

hole

1.65

ware

1.64

ieux

1.61

Activations Density 0.030%

terms related to espionage and spying

No Comments

No Known Activations