INDEX

Explanations

in the process

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

 when

-1.58

 only

-1.57

 after

-1.32

 naszym

-1.31

on

-1.28

igrette

-1.25

 instead

-1.25

 atracción

-1.22

 когато

-1.21

for

-1.19

POSITIVE LOGITS

他也

1.50

 também

1.33

もちろん

1.31

 also

1.21

 también

1.19

Oft

1.17

 Beberapa

1.15

的同时

1.12

 dabei

1.11

 أيضًا

1.11

Activations Density 0.016%