INDEX

Explanations

up direction keypresses

The neuron activates on mentions of arrow‐key directions (e.g. “Up,” “Down,” “Left,” “Right,” or “arrow” in the context of key navigation).

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

矸

-1.30

Toponymie

-1.30

 birra

-1.27

岿

-1.20

 pistolet

-1.17

 vitamine

-1.16

撹

-1.16

 feme

-1.16

 beve

-1.15

܃

-1.15

POSITIVE LOGITS

 teď

1.49

 each

1.38

 měla

1.31

 harten

1.27

 nyní

1.27

 interacción

1.23

 incrementar

1.23

ramientas

1.23

вете

1.23

 totiž

1.22

Activations Density 0.016%