INDEX

Explanations

Q:

The neuron fires on the token “path” as found in “pathway,” i.e. it detects mentions of biological/chemical pathways.

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

 ſever

-1.66

说的

-1.52

着一个

-1.48

 Anſ

-1.35

当即

-1.33

 către

-1.33

 auffi

-1.33

begins

-1.31

 volle

-1.30

說的

-1.30

POSITIVE LOGITS

 such

1.90

="

1.66

 including

1.59

 without

1.56

 noches

1.52

 columnas

1.47

 jako

1.41

 บาท

1.41

 smaller

1.40

}+\

1.39

Activations Density 0.000%