INDEX

Explanations

numbers and instructions

The neuron fires on business-strategy and market-analysis terminology (e.g. “each,” “competitor,” “evaluate,” “threats,” “profitable,” “market,” etc.).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 तमाम

0.96

 appunto

0.94

 সেই

0.86

 başta

0.83

 też

0.80

 각종

0.79

 zaten

0.79

 allerlei

0.77

 dodgy

0.77

 gewisse

0.76

POSITIVE LOGITS

 __________

1.17

 ________

1.16

 ____________

1.14

 ______

1.13

 _______

1.11

 _____________

1.11

 _____

1.09

______

1.08

 ____

1.07

_________

1.06

Activations Density 0.269%