INDEX

Explanations

scientific studies/experiments

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

å¥¥æŀĹ

-0.27

æľ¦

-0.26

_RESOURCE

-0.25

å¤įæ´»

-0.25

caff

-0.25

ForResource

-0.25

)?.

-0.25

_RESOURCES

-0.24

ivirus

-0.24

å°ģè£ħ

-0.24

POSITIVE LOGITS

 comparisons

0.53

 comparison

0.50

 compar

0.44

å¯¹æ¯Ķ

0.42

 compare

0.42

comparison

0.40

 comparative

0.39

 Compar

0.39

 sample

0.39

 comparing

0.38

Activations Density 0.009%