INDEX

Explanations

emacs, CSE, PDE

The neuron is looking for occurrences of certain short technical acronyms (e.g. “ema” in “emacs” or “CSE”) in the text.

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

 estrut

-2.48

’

-2.45

 宣传

-2.33

 帅气

-2.33

ſſen

-2.30

 řešení

-2.23

ünsche

-2.20

 全新

-2.19

ÍTULO

-2.16

𝗖

-2.08

POSITIVE LOGITS

4.84

</strong>

2.97

time

2.31

2.28

2.27

begin

2.17

:";

2.11

2.09

eſ

2.05

ו

2.00

Activations Density 0.005%