INDEX

Explanations

code structure, items, or configurations

The neuron fires on numeric literals in code (especially numbers inside list/array definitions).

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

 from

-1.46

of

-1.36

 with

-1.27

in

-1.22

 też

-1.09

 also

-1.05

 répondit

-1.03

视着

-1.02

似的

-1.01

by

-0.99

POSITIVE LOGITS

 deserto

1.05

 $\}$

1.05

Sekarang

1.02

そして

0.98

這麼

0.94

Setelah

0.91

SEPTEMBER

0.89

 religieux

0.88

 Ambas

0.85

 довольно

0.84

Activations Density 0.021%