INDEX

Explanations

conjunctions and mathematical notation

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

 ľud

-1.10

 quindi

-1.04

 then

-0.98

 násled

-0.96

 dunque

-0.95

 която

-0.94

 which

-0.92

 Nước

-0.92

athed

-0.91

כך

-0.91

POSITIVE LOGITS

 because

1.61

 поскольку

1.48

 since

1.30

由于

1.27

Because

1.25

 Because

1.20

因为

1.19

because

1.13

因為

1.04

 first

1.03

Activations Density 0.008%