INDEX

Explanations

renewable energy and materials

The neuron fires on “content” tokens – in other words, it lights up on semantically informative or topic‐bearing words (and ignores the very common function words like “the,” “and,” etc.).

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

 sustainable

-1.29

 sustainability

-1.18

sustainable

-1.07

 Sustainable

-0.93

 Sustainability

-0.90

Sustainability

-0.90

sustainability

-0.88

 sustent

-0.86

 unsustainable

-0.86

 Nachhal

-0.80

POSITIVE LOGITS

 الإنسان

0.99

servic

0.93

 designers

0.93

 servic

0.91

 nutrients

0.91

getOption

0.88

Scenario

0.86

رامی

0.85

LDAP

0.85

 disassembly

0.84

Activations Density 0.008%