INDEX

Explanations

additional information

This neuron detects occurrences of the word “Additional” (as in headings or labels like “Additional note,” “Additional evidence,” “Additional terms,” 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

 здоров

-1.45

 With

-1.39

 Both

-1.32

Semitism

-1.30

 itulah

-1.29

 Beide

-1.27

 terminó

-1.27

Also

-1.23

みてください

-1.22

 volvió

-1.20

POSITIVE LOGITS

 than

2.39

同様

1.48

 will

1.47

 затем

1.47

嵎

1.44

‪.‬‬

1.41

 niż

1.41

 they

1.41

ୈ

1.34

postolic

1.32

Activations Density 0.045%