INDEX

Explanations

function, free, remained

identifying words followed by verbs

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

nesset

0.46

 सीख

0.44

 보면은

0.43

 बढ़े

0.42

 नाराजगी

0.41

TempBuffer

0.41

 पता

0.41

 строи

0.41

 заказ

0.40

 बढ़ी

0.40

POSITIVE LOGITS

funcs

0.50

Ο

0.50

une

0.49

સ્

0.47

ฟรี

0.45

函数

0.43

↵

0.43

 bleiben

0.43

 remained

0.43

Activations Density 0.004%