INDEX

Explanations

method calls

The neuron is looking for programming method or function call names—particularly common action verbs like get, set, add, remove, prepareStatement, etc.

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

but

-2.28

 such

-1.77

 which

-1.76

 including

-1.52

 thereby

-1.51

 have

-1.47

and

-1.41

 will

-1.39

are

-1.39

 should

-1.33

POSITIVE LOGITS

 accueillir

1.48

 éton

1.42

1.41

 детская

1.40

 rése

1.40

 детской

1.39

 exprimer

1.39

 tredje

1.38

 evenement

1.34

ing

1.33

Activations Density 0.042%