INDEX

Explanations

words related to piercing or punctures

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

MCP

-0.75

marvin

-0.74

Tikang

-0.72

Ārējās

-0.72

omenclature

-0.71

Gob

-0.70

tagHelperRunner

-0.70

__*/

-0.69

NBS

-0.69

$__

-0.68

POSITIVE LOGITS

 Pier

1.74

 Pierce

1.63

Pier

1.56

 pierce

1.55

Pierce

1.53

 PIER

1.51

pier

1.36

 pierced

1.22

 pier

1.20

 piercing

1.17

Activations Density 0.005%