INDEX

Explanations

The neuron fires on word stems that express negative or uncomfortable emotional states (e.g., embarrass-, jinx-, discom-, fed up, bummed, trepid-, demotiv-).

feelings and situations

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

 inapropiados

-1.20

garis

-1.19

等你

-1.17

apatan

-1.16

であれば

-1.12

 gustar

-1.10

退款

-1.09

 thấm

-1.06

 автобу

-1.06

 delito

-1.06

POSITIVE LOGITS

ogues

1.66

—.

1.34

fordshire

1.33

antry

1.28

romeda

1.27

ledad

1.25

moiselle

1.23

鈥

1.23

vereign

1.23

iric

1.20

Activations Density 0.020%