INDEX

Explanations

.OfType

The neuron fires on code tokens that are part of function or method calls (e.g. the parentheses and surrounding identifier in invocations).

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

at

-1.45

for

-1.41

 एक

-1.36

and

-1.34

 getItemCount

-1.28

argli

-1.26

🚇

-1.23

 зді

-1.22

-1.19

＆

-1.17

POSITIVE LOGITS

 Mehrheit

1.31

 those

1.24

queles

1.23

my

1.21

jenige

1.20

fik

1.20

 täglichen

1.20

retrofit

1.20

iL

1.19

 Finanzierung

1.18

Activations Density 0.025%