INDEX

Explanations

that which

introducing dependent clauses

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

出手

0.40

Bra

0.40

 significantly

0.39

 příjem

0.37

픕

0.36

ależ

0.36

 이유는

0.35

rá

0.34

হেতু

0.34

一来

0.34

POSITIVE LOGITS

that

1.78

 that

1.77

 которые

1.58

 که

1.57

 które

1.54

 которая

1.54

 που

1.52

 which

1.48

 যেগুলো

1.48

 التي

1.46

Activations Density 0.689%