INDEX

Explanations

months and holidays

This neuron detects temporal references, especially month names and associated numeric dates.

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

 maart

-1.46

 Universiteit

-1.41

eternal

-1.38

埈

-1.37

</h5>

-1.34

きは

-1.34

 skriva

-1.33

💊

-1.33

 hemel

-1.31

 itself

-1.29

POSITIVE LOGITS

 другие

1.66

 were

1.53

The

1.49

不要

1.40

1.39

ichert

1.37

1.34

 Verhalten

1.34

 χρησιμοποι

1.33

for

1.31

Activations Density 0.016%