INDEX

Explanations

physical components and states

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

erton

1.26

 Lace

1.19

 Bags

1.16

 Leather

1.12

gore

1.08

pth

1.06

town

1.05

ंगाबाद

1.04

 Shirts

1.04

arem

1.03

POSITIVE LOGITS

および

1.65

และการ

1.61

 واست

1.56

และ

1.53

ческая

1.52

এবং

1.51

 vijj

1.50

 puriso

1.49

 وإ

1.48

 וא

1.45

Activations Density 0.047%