INDEX

Explanations

technical terms or abbreviations

This neuron fires on multi‐part code identifiers written in snake_case (i.e., words containing one or more underscores).

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

 from

-1.76

 because

-1.62

for

-1.57

 then

-1.53

so

-1.45

 other

-1.39

 Pág

-1.39

 this

-1.39

<td>

-1.38

 nena

-1.32

POSITIVE LOGITS

所有

1.59

其中

1.41

tował

1.34

tablir

1.33

quíni

1.31

oléon

1.30

atigable

1.30

 již

1.30

nemouth

1.28

 Politiker

1.27

Activations Density 0.098%