INDEX

Explanations

recognizing and processing information

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

tbh

1.60

 massively

1.52

Honestly

1.48

Essentially

1.46

 fucking

1.46

🥺

1.43

🫡

1.43

 genuinely

1.41

🥹

1.41

 Essentially

1.40

POSITIVE LOGITS

*)&

0.97

 muß

0.96



0.94

 bicycl

0.91

 luncheon

0.91

Indexed

0.91

 bicyclists

0.88

 skall

0.88

 usted

0.83

 구입

0.83

Activations Density 0.344%