INDEX

Explanations

comparing alternatives and options

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 auctor

-0.84

 பு

-0.83

quista

-0.82

 powiedział

-0.82

veld

-0.82

Kö

-0.81

Nomenclature

-0.80

cipient

-0.80

 utilisée

-0.80

 LOVE

-0.80

POSITIVE LOGITS

 alternatives

2.75

 options

2.58

 alternative

2.28

 option

2.19

 Alternatives

2.08

alternatives

1.93

 Options

1.91

options

1.86

Options

1.86

Option

1.85

Activations Density 0.022%

No Known Activations