INDEX

Explanations

left

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 process

-0.95

process

-0.81

Process

-0.73

 proce

-0.73

 processus

-0.73

 processes

-0.70

 PROCESS

-0.68

 Vorgang

-0.68

 procedure

-0.68

 procesu

-0.68

POSITIVE LOGITS

 First

0.85

First

0.68

 FIRST

0.63

FIRST

0.63

first

0.60

 first

0.55

 convocatoria

0.51

 transfieras

0.50

 Firstly

0.50

New

0.50

Activations Density 0.002%