INDEX

Explanations

how to use tutorials

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Algorithm

-0.31

 Anthrop

-0.28

 Algorithm

-0.25

 rupt

-0.24

 anthrop

-0.24

JNI

-0.24

iez

-0.24

ä½İæ¸©

-0.23

èĭ¦éļ¾

-0.23

éĩįéĩĳå±ŀ

-0.23

POSITIVE LOGITS

åĬŁèĥ½

0.45

æĵįä½ľ

0.43

 usage

0.40

çĨŁç»ĥ

0.40

 functionality

0.39

 functionalities

0.39

 shortcuts

0.39

ä½¿çĶ¨

0.38

çļĦåĬŁèĥ½

0.38

æŃ£å¸¸ä½¿çĶ¨

0.38

Activations Density 0.122%