INDEX

Explanations

mentioning or saying something previously

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

è¯Ħ

-0.23

æĹ¶éĹ´ä¸º

-0.23

æĸ°é²ľ

-0.23

ç¨ł

-0.23

 Fusion

-0.23

Kes

-0.23

æ¯ıä¸Ģ

-0.23

tape

-0.22

INGLE

-0.22

 Pension

-0.22

POSITIVE LOGITS

 above

0.46

 mentioned

0.41

 Ð²ÑĭÑĪÐµ

0.40

ä¸ĬéĿ¢

0.39

 mention

0.39

 mentions

0.37

above

0.36

æıĲåĪ°

0.36

mentioned

0.36

ioned

0.35

Activations Density 0.032%