INDEX

Explanations

foreign words for perhaps

This neuron detects tokens that look like alphanumeric identifiers—i.e. long strings of letters and digits such as hashes or code/section names.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 their

-1.37

 first

-1.31

 analisis

-1.16

 this

-1.15

 impressively

-1.12

 nestled

-1.10

 allah

-1.07

 besö

-1.06

 unequivocally

-1.05

 endearing

-1.02

POSITIVE LOGITS

 może

1.36

perhaps

1.34

 может

1.21

 plutôt

1.20

 talvez

1.19

もあり

1.19

 peut

1.14

depuis

1.13

 chaque

1.13

possibly

1.11

Activations Density 0.001%