INDEX

Explanations

code block separators

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

 Teste

-0.80

 nazis

-0.79



-0.78

 TROP

-0.77

 Forst

-0.77

 stocker

-0.74

 snip

-0.74

 Glossary

-0.73

 crimp

-0.73

outdir

-0.73

POSITIVE LOGITS

描

0.85

agal

0.84

ιώ

0.83

 both

0.81

kennen

0.80

tivation

0.77

 them

0.76

richlet

0.74

NSLog

0.74

 they

0.74

Activations Density 0.016%