INDEX

Explanations

names and subsequent actions

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

になるので

-1.17

 трябва

-1.13

 from

-1.13

頜

-1.11

颚

-1.11

nev

-1.09

athers

-1.07

vermectin

-1.04

zzet

-1.03

laria

-1.03

POSITIVE LOGITS

is

2.30

has

2.27

 also

2.05

 started

2.03

 graduated

1.97

 currently

1.95

 recently

1.70

 have

1.70

 began

1.67

 holds

1.51

Activations Density 0.097%