INDEX

Explanations

solid state and formation

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

ãģĬãĤĬ

-0.12

 Mitar

-0.10

zug

-0.09

owitz

-0.09

iej

-0.09

ValueCollection

-0.09

ynom

-0.09

 Mastery

-0.09

eso

-0.09

<|begin_of_text|>

-0.08

POSITIVE LOGITS

ified

0.28

ifying

0.26

ly

0.24

arity

0.23

ifies

0.22

ify

0.21

-state

0.21

ification

0.21

-solid

0.20

ary

0.19

Activations Density 0.016%