INDEX

Explanations

might be, or what you could use

The neuron fires on imperative, call-to-action phrasing—i.e. promotional or instructional commands (like “make a reservation,” “use the comments,” “join,” etc.).

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

এবং

0.91

 এবং

0.86

不仅

0.86

そして

0.84

人和

0.84

both

0.84

以及

0.83

 både

0.83

and

0.82

 sowohl

0.82

POSITIVE LOGITS

 atau

1.47

 или

1.42

 அல்லது

1.41

 veya

1.40

 alebo

1.38

或者是

1.35

 Atau

1.35

 অথবা

1.34

 või

1.33

 或者

1.33

Activations Density 0.951%