INDEX

Explanations

Secret Worlds, Cibyl, Netanya, Named Pipes, Abercrombie, Map Of, icons, juraQuiz, test special, Puducherry, Book List

This neuron is tuned to spot the “retailscm” package or class name fragments in code (e.g. the tokens “re,” “tails,” “retails,” “cm” from “retailscm”).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 both

-1.77

any

-1.77

 only

-1.76

 just

-1.70

 which

-1.55

 during

-1.55

 this

-1.51

 before

-1.48

 while

-1.48

not

-1.46

POSITIVE LOGITS

though

1.26

 poign

1.20

܆

1.20

𓆏

1.18

chie

1.17

generally

1.16

perhaps

1.15

粜

1.14

辚

1.14

aggiare

1.14

Activations Density 0.059%