INDEX

Explanations

unit, network, client

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

ాయి

1.35

了

1.16

ständ

1.16

dragging

1.12

 والأ

1.09

 convoy

1.08

werks

1.07

 rumored

1.06

ać

1.05

ر

1.04

POSITIVE LOGITS

herence

1.61

optimal

1.60

 мира

1.54

 சேர்க்க

1.53

ོས་

1.53

궤

1.51

 недвижимости

1.51

규모

1.50

ESSION

1.44

 التى

1.44

Activations Density 0.000%