INDEX

Explanations

programming languages

The neuron is sensitive to isolated punctuation tokens (like “>” or a standalone opening quotation mark) used to mark block-quotes or section headings rather than normal running text.

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

 semaphore

1.43

});

1.41

𝔬

1.37

 baseHP

1.35

합니다

1.34

お願いします

1.32

𝔪

1.31

妬

1.29

 стать

1.27

পি

1.26

POSITIVE LOGITS

ism

1.31

ía

1.30

an

1.23

ę

1.20

ires

1.16

ación

1.11

ination

1.09

jų

1.09

<0x80>

1.09

č

1.08

Activations Density 0.000%