INDEX

Explanations

competitive grants

This neuron detects mentions of monetary figures and funding/grant amounts.

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

urare

-0.77

 Hardin

-0.68

Overrides

-0.68

 kerusakan

-0.68

 difficulties

-0.68

zany

-0.68

맷

-0.68

 scherm

-0.67

 maven

-0.67

ónde

-0.67

POSITIVE LOGITS

 grant

1.22

gran

1.17

Competitive

1.13

 grantee

1.13

 competitive

1.08

grant

1.08

 grants

1.07

 gran

1.07

competition

0.98

 Grant

0.97

Activations Density 0.041%