INDEX

Explanations

introduces list items or statements

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

gamot

-0.82

टन

-0.76

 doInBackground

-0.71

 мани

-0.71

thly

-0.70

Descriere

-0.69

bow

-0.67

nė

-0.67

 Gerr

-0.67

侍

-0.67

POSITIVE LOGITS

 Locking

1.07

 sigh

1.04

script

1.01

 locking

0.97

sigh

0.96

 Script

0.95

 script

0.95

 redeem

0.95

 redeeming

0.93

 sighing

0.92

Activations Density 0.009%