INDEX

Explanations

statement terminators

This neuron specifically responds to occurrences of the identifier “System” (e.g. in using/import directives).

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

 sabar

-1.22

óricas

-1.17

 akku

-1.11

 asal

-1.09

 only

-1.09

 meres

-1.07

on

-1.06

 saman

-1.02

 kule

-1.02

ниш

-1.01

POSITIVE LOGITS

and

1.53

●●

1.10

 bastion

1.05

不一定

1.01

 azonban

0.98

']

0.96

 dibujar

0.95

hesda

0.94

 Proclamation

0.94

Activations Density 0.001%