INDEX

Explanations

just the

missing keys or parameters

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

tere

0.54

éntesis

0.50

ane

0.49

ært

0.49

 어려운

0.49

ister

0.48

 ਜ਼

0.48

ug

0.48

áni

0.47

ské

0.46

POSITIVE LOGITS

Housing

0.61

Compute

0.55

housing

0.52

 housing

0.49

Wit

0.49

voi

0.49

compute

0.47

Hello

0.47

 içine

0.47

Validation

0.47

Activations Density 0.000%