INDEX

Explanations

code documentation tags

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

 more

-1.47

now

-1.30

 several

-1.19

one

-1.05

 then

-1.02

 different

-0.95

 most

-0.93

 there

-0.92

]->

-0.91

set

-0.89

POSITIVE LOGITS

 parfois

1.24

 getItemCount

1.17

 bemerken

1.16

 reclama

1.15

修为

1.14

 kaplama

1.14

🎦

1.13

sometimes

1.12

 reivindic

1.12

 rafraî

1.11

Activations Density 0.005%