INDEX

Explanations

last X games

The neuron activates on numeric-statistic phrases—especially counts of games, at-bats, points, or streaks (e.g. “three at-bats,” “last five games,” “six straight,” 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

这一幕

-1.51

際には

-1.48

ながらも

-1.38

堿

-1.37

半个

-1.30

entiel

-1.29

𐑂

-1.29

ULDER

-1.28

ǽ

-1.28

timento

-1.27

POSITIVE LOGITS

1.55

↵

1.55

ll

1.48

他

1.34

》：

1.32

why

1.28

 Beamten

1.22

 Schauspieler

1.22

 Freundin

1.20

1.19

Activations Density 0.017%