INDEX

Explanations

qualifiers and idioms

The neuron is strongly activated by adverbs, hedges, and discourse‐level intensifiers or qualifiers (e.g. so, apparently, clearly, of course, somehow, even, fairly).

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

 your

-1.94

 these

-1.77

 each

-1.59

 those

-1.45

 takich

-1.42

 that

-1.38

any

-1.30

 such

-1.29

 этими

-1.25

 explicitly

-1.24

POSITIVE LOGITS

</h1>

1.80

 concerne

1.45

separated

1.38

 jesu

1.34

 jardins

1.33

 quinze

1.27

 HILLS

1.25

 concr

1.24

 brune

1.23

🔒

1.23

Activations Density 0.606%