INDEX

Explanations

shaping future outcomes

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

Zar

-0.10

elong

-0.09

Bra

-0.09

hlas

-0.08

bra

-0.08

Haz

-0.08

Fif

-0.08

.ColumnHeader

-0.08

aco

-0.08

POSITIVE LOGITS

 fate

0.30

 outcome

0.28

outcome

0.23

 future

0.23

Outcome

0.22

 course

0.22

 Outcome

0.22

 destino

0.20

 outcomes

0.20

 destiny

0.20

Activations Density 0.044%