INDEX

Explanations

numbers, coordinates, and equations

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

дется

-1.43

 аккура

-1.40

饒

-1.30

ꈍ

-1.29

 furious

-1.28

ごく

-1.25

Dě

-1.24

 problematic

-1.21

中略

-1.21

 песню

-1.19

POSITIVE LOGITS

1.44

1.36

strual

1.35

1.30

abbaye

1.22

ей

1.20

 SwitchCompat

1.18

——”

1.15

])));

1.15

CONDITIONS

1.14

Activations Density 0.035%

No Known Activations