INDEX

Explanations

get files or objects

The neuron fires strongly on code tokens representing “Get” method calls (i.e. method names starting with “Get”).

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

赢

-0.79

jmuje

-0.77

ໄ

-0.77

 Olds

-0.75

significant

-0.75

 лак

-0.75

ūras

-0.72

belo

-0.72

learning

-0.70

unal

-0.70

POSITIVE LOGITS

LUCK

0.81

儆

0.81

見えて

0.80

tinggi

0.78

 virš

0.77

 להת

0.75

uanto

0.74

 третье

0.73

 מת

0.72

 המת

0.71

Activations Density 0.034%