INDEX

Explanations

programming loops: for, foreach

The neuron activates on numeric literals used as loop bounds or counters in for-loop headers.

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

now

-1.60

 until

-1.48

MMdd

-1.41

 because

-1.38

 first

-1.36

but

-1.29

 تجه

-1.25

خه

-1.24

 FICTION

-1.22

@{

-1.21

POSITIVE LOGITS

 jednocześnie

1.66

これらの

1.65

Also

1.54

 различных

1.47

 различ

1.45

そも

1.44

 numerosas

1.41

 Особенно

1.40

</strong>

1.40

様々な

1.39

Activations Density 0.004%