INDEX

Explanations

if you want to say

New Auto-Interp

Top Features by Cosine Similarity

Configuration

Prompts (Dashboard)

10,000 prompts, 128 tokens each

Dataset (Dashboard)

lmsys/lmsys-chat-1m

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Ð¾Ð³ÑĢÐ°

-0.11

 ÐºÑĥÐ»ÑĮÑĤÑĥ

-0.11

 Ð¸Ð¼Ñĥ

-0.11

 Ð´Ð¸Ð²Ð¸

-0.10

ÑĢÐ¾Ð½Ð¸ÑĩÐµÑģ

-0.10

 ÐºÐ°ÑĩÐµ

-0.10

 eigentlich

-0.10

 ÑĤÐ²Ð°

-0.10

 chtÄĽ

-0.10

 ÑħÑĢÐ°

-0.10

POSITIVE LOGITS

 speaking

0.11

 Ð¶Ðµ

0.11

you

0.11

-speaking

0.10

 however

0.10

 talking

0.10

rag

0.10

 ovÅ¡em

0.10

adden

0.10

lush

0.10

Activations Density 0.043%