INDEX

Explanations

types of research or practice

The neuron activates on words naming academic or professional disciplines, research fields, and program types (e.g. “General Practice Residency,” “theoretical physics,” “neuroradiology,” “translational research,” 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

 will

-1.48

are

-1.38

↵↵↵

-1.26

↵↵↵↵

-1.24

 canale

-1.21

。（

-1.21

雖

-1.20

↵↵↵↵↵↵↵

-1.20

↵↵↵↵↵↵

-1.18

↵↵↵↵↵↵↵↵↵↵↵↵↵

-1.17

POSITIVE LOGITS

 съм

1.34

TRY

1.27

 kesempatan

1.21

 Укра

1.21

deutsch

1.16

 gelukkig

1.15

 MONDAY

1.15

citep

1.14

ческий

1.13

 pemilik

1.13

Activations Density 0.146%