INDEX

Explanations

future expected outcomes

The neuron detects occurrences of the verb “anticipate” (in any of its inflected forms).

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

⑉

-2.47

 decidedly

-2.41

 historian

-2.39

水晶

-2.36

 дают

-2.31

 linken

-2.23

后

-2.20

 intrepid

-2.19

 apoi

-2.16

 grü

-2.09

POSITIVE LOGITS

ஂ

2.81

2.80

[]

2.67

柃

2.50

浥

2.45

甡

2.44



2.36

哔

2.34

聼

2.19

 bottes

2.16

Activations Density 0.002%