INDEX

Explanations

Play followed by number and word

numbers following parentheses

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 célébr

0.56

 దృష్టి

0.54

陂

0.53

 мә

0.51

 двери

0.51

 лест

0.50

 đậm

0.50

ềm

0.50

⚜

0.49

阍

0.49

POSITIVE LOGITS

input

0.52

args

0.52

filename

0.51

tar

0.48

output

0.48

event

0.47

audio

0.47

upload

0.46

java

0.46

data

0.46

Activations Density 0.001%