INDEX

Explanations

people with conditions

The neuron fires on noun phrases that identify groups of people characterized by having a condition—especially “X with Y” constructions like “women with disabilities,” “individuals with schizophrenia,” or “children with cancer.”

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

-1.23

-1.09

-1.06

-1.05

 poiché

-1.03

 hır

-1.00

-0.96

 einiges

-0.95

if

-0.94

怎么可能

-0.94

POSITIVE LOGITS

 with

1.85

及其

1.43

ทุก

1.30

 نیاز

1.20

hingga

1.19

 from

1.17

[]:

1.16

lepaskan

1.15

ไม่มี

1.13

така

1.13

Activations Density 0.079%