INDEX

Explanations

selection process

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 практи

-0.88

Zitat

-0.88

workflow

-0.85

 hell

-0.82

 pessoas

-0.82

 slova

-0.80

 workflow

-0.80

ategory

-0.79

 pratique

-0.79

وید

-0.78

POSITIVE LOGITS

 구성

0.86

 anpassen

0.85

度に

0.83

chosen

0.81

 vorbei

0.81

 acquaint

0.80

<0x84>

0.80

 selection

0.80

search

0.79

 caratteristiche

0.79

Activations Density 0.125%