INDEX

Explanations

different

finding differences

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 나머

0.41

ើយ

0.37

บัติ

0.35

 나머지

0.34

่วง

0.33

াইয়া

0.33

 ఇక

0.33

 очеред

0.33

ائب

0.32

linger

0.31

POSITIVE LOGITS

 different

4.38

 diferente

4.25

different

4.25

 Different

4.13

 DIFFERENT

4.09

Different

4.06

 berbeda

3.88

 farklı

3.80

 différente

3.80

 diferentes

3.78

Activations Density 0.847%