INDEX

Explanations

message and protocol definitions

The neuron fires on code‐style identifiers for message or topic types (e.g. ROS/MAVLink/CORBA message names and topic names).

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

-1.90

 mattino

-1.65

 albeit

-1.55

$-$

-1.49

-1.45

 aspiring

-1.43

又是一

-1.40

 beobach

-1.40

 ensuing

-1.34

 esche

-1.33

POSITIVE LOGITS

 montr

1.31

涐

1.29

lük

1.26

Evet

1.23

controlador

1.22

 петербург

1.22

 мона

1.19

1.18

合作伙伴

1.18

 kimi

1.18

Activations Density 0.004%