INDEX

Explanations

sound bites

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

 contamos

-0.85

 tratando

-0.83

Agreed

-0.81

 impulsar

-0.81

ボーナス

-0.79

 superfície

-0.79

étant

-0.77

 cowards

-0.76

 kalimat

-0.75

从事

-0.75

POSITIVE LOGITS

bite

1.34

proofing

1.24

proof

1.24

 bites

1.22

tracks

1.10

 waves

1.09

 bytes

1.06

stage

1.05

bites

1.05

cloud

1.03

Activations Density 0.025%