INDEX

Explanations

awards and medals

The neuron is activated by short function words—particularly pronouns (you, it), interrogatives (what, when), negations (no), and simple prepositions (for).

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

You

-1.45

abaikan

-1.43

 свою

-1.31

 нашу

-1.29

 мою

-1.29

From

-1.27

These

-1.27

Its

-1.23

 наши

-1.23

How

-1.22

POSITIVE LOGITS

</h3>

1.50

</sub>

1.46

 čím

1.42

 clarity

1.41

 buscan

1.40

)$,

1.39

ască

1.38

udière

1.35

</strong>

1.35

volmente

1.34

Activations Density 0.075%