INDEX

Explanations

what to expect

This neuron detects expressions of expected or predicted values, especially the word “expect” together with numerical figures.

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

 demonios

-1.67

 requerimientos

-1.66



-1.56

 akar

-1.55

 konstruk

-1.53

OUTUBE

-1.48

坶

-1.47

〠

-1.45

 distinctive

-1.41

 convincing

-1.40

POSITIVE LOGITS

1.81

1.72

",

1.48

):

1.44

Also

1.35

 Also

1.34

And

1.32

er

1.31

1.29

Activations Density 0.008%