INDEX

Explanations

java and android platforms

This neuron flags import statements from the Java standard library.

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

anément

-1.99

efully

-1.98

 juſ

-1.96

 faciliter

-1.91

isand

-1.87

⚋

-1.87

Cependant

-1.82

 grotte

-1.80

diyah

-1.77

셧

-1.76

POSITIVE LOGITS

 with

2.34

2.28

on

2.16

2.14

 only

2.13

2.09

2.06

at

1.96

for

1.91

Activations Density 0.037%