INDEX

Explanations

there are things

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

 topology

1.15

 pointA

1.13

盂

1.11

ть

1.07

 точки

1.02

ozen

1.01

ције

1.00

हूर

0.98

yiz

0.98

iš

0.97

POSITIVE LOGITS

 accompany

1.72

 accompanying

1.70

accompanied

1.66

 সঙ্গ

1.63

 accompanies

1.62

 accompanied

1.61

 сопрово

1.60

 ساتھ

1.57

 accompaniment

1.56

 acompaña

1.52

Activations Density 0.328%