INDEX

Explanations

has or have

The neuron strongly activates on instances of the auxiliary verb “have” (including its form “has”).

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

-2.13

 from

-1.94

for

-1.89

and

-1.69

 will

-1.55

 would

-1.52

in

-1.48

 should

-1.48

 after

-1.41

we

-1.38

POSITIVE LOGITS

 aisladas

1.45

 esenciales

1.41

 humedad

1.38

 condicionado

1.38

 aéreas

1.36

 burbujas

1.34

 frecuentes

1.31

 fisuras

1.29

 plásticos

1.29

 próximos

1.29

Activations Density 0.089%