INDEX

Explanations

suggestions using modal verbs

The neuron fires on the second‐person instructional “you” in answer sentences (e.g. “You can…,” “You’re probably…”).

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

 that

-1.58

 when

-1.39

 these

-1.38

who

-1.38

 which

-1.38

 whose

-1.33

 each

-1.30

骯

-1.27

 this

-1.23

 revelou

-1.19

POSITIVE LOGITS

atürk

1.64

 your

1.56

 into

1.49

лә

1.37

 می‌توانید

1.35

СИ

1.33

 Meningkatkan

1.31

boutin

1.26

₽

1.25

 cuartos

1.23

Activations Density 0.285%