INDEX

Explanations

data structures containing items

The neuron fires on mentions of data‐structure types (e.g. hash, array, struct, dictionary, map, list, matrix, frame, etc.).

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

↵↵

-1.38

荽

-1.28

ípios

-1.26

也没有

-1.23

也没

-1.21

淼

-1.17

Greatest

-1.16

IM

-1.16

 появится

-1.16

thog

-1.15

POSITIVE LOGITS

 that

1.82

the

1.56

to

1.52

 from

1.45

 which

1.41

 cliquant

1.33

bueno

1.27

を変更

1.27

utk

1.20

list

1.19

Activations Density 0.334%