INDEX

Explanations

multivalent author or date

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 εφαρ

-0.79

 Knudsen

-0.77

tslib

-0.76

 szóci

-0.75

 febr

-0.73

chevêque

-0.73

各样的

-0.72

例文帳に追加

-0.72

mathbf

-0.71

 clientes

-0.71

POSITIVE LOGITS

TeV

0.97

zyż

0.84

lierten

0.84

arxiv

0.83

erevan

0.83

 创建时间

0.82

hih

0.82

']=='

0.81

 arXiv

0.81

 Collaboration

0.79

Activations Density 0.013%