INDEX

Explanations

loop iterates over sequences

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

avons

1.58

騨

1.48

åll

1.48

 $\{(

1.45

يدي

1.44

Eloquent

1.44

estell

1.42

)]^{

1.42

نهار

1.40

ELIST

1.39

POSITIVE LOGITS

smanship

1.36

1.33

ljen

1.24

tob

1.22

tw

1.20

squ

1.17

 яку

1.15

ske

1.15

icip

1.14

sm

1.13

Activations Density 0.069%