INDEX

Explanations

up

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

 tasted

-0.68

 smelled

-0.67

 glowed

-0.64

 didst

-0.63

 didn

-0.63

 wasn

-0.63

 consisted

-0.63

 mourut

-0.63

oustics

-0.63

 stood

-0.63

POSITIVE LOGITS

Биография

0.55

Personendaten

0.52

 jScrollPane

0.51

JNIEnv

0.50

 tomorrow

0.48

agaimana

0.48

 BorderRadius

0.47

ẖ

0.47

DebuggerNonUser

0.47

 will

0.47

Activations Density 0.152%