INDEX

Explanations

multilingual understanding

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ו

4.86

rified

3.77

ು

3.76

ာ

3.73

sion

3.72

sberg

3.66

ੀ

3.64

רים

3.58

sit

3.52

ség

3.47

POSITIVE LOGITS

ist

3.30

которые

2.89

<u>

2.42

смотря

2.40

<em>

2.32

ignores

2.31

helves

2.25

om

2.22

куда

2.21

Sp

2.16

Activations Density 0.016%