INDEX

Explanations

programs, grants, and training

The neuron flags mentions of formal organized “programs” (and similar structured initiatives like fellowships).

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

 even

-1.43

ないです

-1.39

 just

-1.39

has

-1.37

 when

-1.36

so

-1.32

it

-1.30

if

-1.27

ῖν

-1.23

had

-1.21

POSITIVE LOGITS

 trà

1.33

unno

1.31

rency

1.28

怼

1.27

 vetorial

1.27

 бампер

1.27

ätzlich

1.24

 intensive

1.23

kuuta

1.23

 لمدة

1.23

Activations Density 0.037%