INDEX

Explanations

starting a numbered explanation

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 상세

1.89

철

1.66

동

1.62

 reinforcing

1.60

 connectors

1.54

 attention

1.54

 좌표

1.47

oons

1.43

╚

1.42

 rail

1.42

POSITIVE LOGITS

кування

1.61

utation

1.51

 Looks

1.50

 parte

1.50

Guess

1.48

 сказать

1.46

पणे

1.45

 непло

1.45

 dificult

1.44

1.42

Activations Density 0.074%