INDEX

Explanations

. followed by code terms

This neuron fires on occurrences of the token “Android” (in any casing) in code or XML.

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

Of

-2.08

 conquête

-2.02

縢

-1.86

􀂃

-1.80

 That

-1.77

了啊

-1.77

 parution

-1.75

 darted

-1.75

 enam

-1.72

:[],

-1.72

POSITIVE LOGITS

2.47

婞

2.23

 rozwią

2.03

翾

2.02

 změ

1.91

 страницы

1.88

The

1.88

 čís

1.84

of

1.81

”,

1.81

Activations Density 0.079%