INDEX

Explanations

programming keywords

The neuron chiefly detects programming identifiers—names of variables, functions, classes, modules, etc., in source code.

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

 will

-0.87

 alip

-0.87

who

-0.86

地名

-0.84

 want

-0.84

هات

-0.80

 dila

-0.79

 править

-0.79

corrência

-0.79

 quibus

-0.79

POSITIVE LOGITS

agrave

0.98

潸

0.96

 персонажа

0.95

 улица

0.86

getBalance

0.84

 antwortete

0.83

 Symptome

0.82

 acogedor

0.82

}-

0.81

._

0.80

Activations Density 0.029%