INDEX

Explanations

exploring what you're asking

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

 jednej

1.91

思います

1.68

 возможность

1.66

рию

1.63

 rozší

1.56

 quanto

1.56

 kilku

1.55

ısıyla

1.54

同样的

1.54

вались

1.52

POSITIVE LOGITS

orphism

1.41

OWN

1.41

ｭ

1.38

 연속

1.37

ifes

1.32

dech

1.28

dut

1.27

 BURN

1.27

 deriv

1.26

บัติ

1.26

Activations Density 0.095%