INDEX

Explanations

Questions and opinions

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

 ÐºÐ¾Ð³Ð´Ð°

-0.49

 when

-0.47

when

-0.45

 quando

-0.44

 cuando

-0.42

_when

-0.39

	when

-0.38

gdy

-0.38

khi

-0.37

 When

-0.37

POSITIVE LOGITS

 because

0.46

Because

0.43

åĽłä¸º

0.41

 Because

0.40

because

0.38

åĽłä¸ºåľ¨

0.34

åĽłä¸ºä»ĸ

0.33

åĽłä¸ºå®ĥ

0.33

due

0.32

åĽłä¸ºæĪĳ

0.32

Activations Density 0.001%