INDEX

Explanations

references to actors and performances in various entertainment contexts

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.2.hook_resid_post

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

ĥ½

-2.17

Ģ

-1.66

 Administration

-1.61

gnu

-1.59

googleapis

-1.54

aceae

-1.53

ovir

-1.53

 triglycerides

-1.51

Ħ

-1.50

 publication

-1.47

POSITIVE LOGITS

 theater

1.96

 theaters

1.96

 audition

1.95

 theat

1.95

 announ

1.79

 singing

1.79

 costume

1.78

 theatre

1.77

manship

1.77

 scene

1.74

Activations Density 0.511%

references to actors and performances in various entertainment contexts

No Comments

No Known Activations