INDEX

Explanations

try it

The neuron fires on instances of “try” used in an imperative suggestion (e.g. “Try it,” “Don’t try this at home”).

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

 meget

-2.05

 ‘‘

-2.00

 inherently

-1.78

 químicas

-1.69

 geographically

-1.68

</h2>

-1.64

 virtually

-1.61

 Якщо

-1.59

 gammel

-1.57

臾

-1.56

POSITIVE LOGITS

饹

2.02

1.91

熥

1.91

–

1.91

":

1.80

recommend

1.80

賸

1.72

?"

1.68

africa

1.63

殛

1.62

Activations Density 0.020%