INDEX

Explanations

full of

This neuron responds to standalone punctuation tokens (e.g. brackets, commas, hyphens, backslashes, ellipses, 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

</h3>

-2.16

</strong>

-1.86

</h2>

-1.71

 only

-1.70

his

-1.69

to

-1.63

</h1>

-1.59

 solely

-1.57

of

-1.55

-1.49

POSITIVE LOGITS

 vinto

1.63

 totalement

1.61

 нашли

1.58

犼

1.55

 essas

1.52

 complètement

1.51

 trovato

1.51

ឣ

1.50

 Anſ

1.48

should

1.47

Activations Density 0.015%