INDEX

Explanations

pain

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 pain

-1.17

 Pain

-1.02

 douleur

-1.02

Pain

-1.01

pain

-1.00

 PAIN

-0.96

 pijn

-0.89

 douleurs

-0.75

الحياه

-0.74

 боли

-0.69

POSITIVE LOGITS

[]:

0.59

 gyrus

0.56

ographer

0.52

alsy

0.50

 sche

0.49

 condition

0.48

 condens

0.48

Worker

0.48

 Anomaly

0.48

ink

0.48

Activations Density 0.083%