INDEX

Explanations

pronoun followed by state verb

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

pren

1.17

edict

1.16

 tris

1.15

 Parry

1.14

 personnelle

1.12



1.11

xer

1.10

 Seeger

1.06

Fols

1.06

 المستقيم

1.05

POSITIVE LOGITS

 APPE

1.78

 хотелось

1.74

impression

1.72

 последнее

1.72

 piet

1.69

pagen

1.68

={{

1.67

APPE

1.67

⼀

1.67

เยอะ

1.66

Activations Density 0.135%