INDEX

Explanations

get/parse/sha/acquire/deserialize

This neuron fires on the opening angle-bracket of generic type annotations (e.g. the “<” in “Observable<Status>”).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 computable

0.66

 BOOL

0.63

 perror

0.62

 કરતી

0.57

 dataset

0.56

 parallelogram

0.56

훑

0.56

 curved

0.56

יים

0.56

:'',

0.53

POSITIVE LOGITS

лимпий

0.63

 consigo

0.60

abhavena

0.60

 оставить

0.58

理

0.58

卸

0.57

Personal

0.57

恃

0.56

留下

0.55

ضمن

0.55

Activations Density 0.000%