INDEX

Explanations

method definitions and calls

The neuron fires on the “function” keyword in code, i.e. function declarations.

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

 parfüm

-2.91

 parfü

-2.91

 sokak

-2.36

 estampado

-2.20

 fizik

-2.13

 esimerkiksi

-2.02

 kapak

-1.98

 spørsmål

-1.95

 klinik

-1.91

 görd

-1.90

POSITIVE LOGITS

</h2>

3.25

</b>

2.05

</em>

2.03

We

1.93

1.91

Of

1.91

 Therefore

1.88

 nostro

1.88

lo

1.81

 Then

1.81

Activations Density 0.003%