INDEX

Explanations

delimiters and separators

The neuron detects mentions of field or record delimiters/separators (e.g. “FIELDS TERMINATED BY”, “delimiter”, “separator”, “sep”).

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

and

-1.70

 other

-1.51

-1.45

や

-1.44

se

-1.42

 óptica

-1.41

 foolish

-1.40

﹌

-1.36

 still

-1.35

第七

-1.35

POSITIVE LOGITS

 واقعا

1.64

жуть

1.63

⑊

1.54

 épices

1.52

 使っ

1.51

]);

1.51

änemark

1.49

 êtres

1.48

ஸ்

1.48

petual

1.47

Activations Density 0.026%