INDEX

Explanations

mathematical expressions and code snippets

The neuron fires on occurrences of variables (or expressions) raised to the second power (i.e. “…^2” squared terms).

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

-1.23

 diſ

-1.23

коменду

-1.22

 GRATU

-1.15

を失

-1.14

onnaise

-1.14

 germain

-1.13

moiselle

-1.11

€”

-1.08

 consulté

-1.07

POSITIVE LOGITS

 recientemente

1.05

 prácticamente

1.03

ahuje

0.91

 lisäksi

0.87

 even

0.86

Ligações

0.85

for

0.85

来源于

0.85

 giọng

0.83

 gelangt

0.83

Activations Density 0.006%