INDEX

Explanations

deviates from expected

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

움

1.35

 vizsgál

1.30

ുമിത്ര

1.30

ੰਜ

1.27

 дзе

1.27

ᙱ

1.24

 betrachtet

1.24

ʋ

1.23

 얘기

1.22

PLOAD

1.22

POSITIVE LOGITS

of

1.44

 with

1.25

on

1.18

odia

1.14

กับ

1.12

 به

1.08

huri

1.08

的其他

1.05

 tracers

1.04

 endlessly

1.04

Activations Density 0.290%