INDEX

Explanations

` then keyword

This neuron activates on numeric literal tokens (e.g. integer constants) in code.

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

let

-1.48

let

-1.17

Let

-1.16

Let

-1.10

lette

-0.98

 lets

-0.96

の為

-0.93

 ChatGPT

-0.89

 Lets

-0.88

är

-0.86

POSITIVE LOGITS

 this

1.42

 that

1.36

 arguments

1.30

arguments

1.24

 window

1.08

!!

1.05

$.

1.02

 angular

0.96

$.

0.90

izaron

0.88

Activations Density 0.039%