INDEX

Explanations

math formulas and symbols

The neuron responds to backslash characters used in markup or code (e.g., LaTeX commands).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

at

-2.30

 during

-2.09

 テラス

-2.06

 広場

-1.95

 diki

-1.89

 cristiano

-1.89

滙

-1.88

مقال

-1.86

 while

-1.85

 when

-1.77

POSITIVE LOGITS

2.47

 Besonder

2.02

queryInterface

2.00

༛

1.98

usually

1.98

to

1.91

"\

1.91

..........

1.90

1.87

"'

1.86

Activations Density 0.002%