INDEX

Explanations

let's say or let used

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

ি

2.12

ീകരണ

2.11

bir

2.05

дук

1.94

цията

1.92

undai

1.89

herence

1.88

ീക

1.86

ett

1.83

pload

1.82

POSITIVE LOGITS

ymology

2.67

tering

2.53

प्ताह

2.49

entiary

2.40

ório

2.40

र्दू

2.40

vscode

2.39

loose

2.38

ského

2.35

Elo

2.34

Activations Density 0.134%