INDEX

Explanations

representing or containing

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

:]

1.81

 idempotent

1.79

}^{+}$

1.79

 '.')

1.74

}^{*}$.

1.72

 Hoffman

1.70

 anyhow

1.69

 Basically

1.68

 Overall

1.67

+.

1.66

POSITIVE LOGITS

ѓ

1.63

 ибо

1.54

ной

1.48

 trazendo

1.46

檻

1.44

 سوی

1.41

 чтоб

1.41

 معانا

1.40

یین

1.37

Bienvenidos

1.33

Activations Density 0.021%