INDEX

Explanations

possessive followed by noun

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

udden

-0.10

 recent

-0.10

pliant

-0.10

 Garr

-0.09

ç´

-0.09

 sudden

-0.09

çĪ½

-0.08

 proven

-0.08

IDI

-0.08

POSITIVE LOGITS

 confidence

0.14

 pace

0.13

 legend

0.13

 popularity

0.13

 intensity

0.13

 levels

0.12

 level

0.12

 grip

0.12

 focus

0.12

 fortunes

0.11

Activations Density 0.033%