INDEX

Explanations

having you

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

が少ない

-1.02

orestation

-0.90

 AFFECT

-0.84

helped

-0.83

écran

-0.82

 éstos

-0.81

 ofrecemos

-0.80

lads

-0.79

īgi

-0.77

habitat

-0.77

POSITIVE LOGITS

 having

2.38

Having

1.69

 Having

1.54

 иметь

1.54

having

1.53

him

1.34

 such

1.34

 avoir

1.17

 avere

1.13

有

1.13

Activations Density 0.041%