INDEX

Explanations

financial metrics, countries, and specific nouns

factors, willingness, past

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

固体

0.49

ÔNG

0.48

鉾

0.46

ਨ

0.45

ativamente

0.45

বর্ত

0.45

ਮ

0.44

།

0.44

フォト

0.43

WITH

0.43

POSITIVE LOGITS

 фаразы

0.47

 انڈیا

0.44

 distr

0.42

 bunk

0.41

 rewards

0.41

TypeValue

0.40

 божомолдору

0.40

 таком

0.40

 debating

0.40

ኔታ

0.40

Activations Density 0.001%