INDEX

Explanations

looking forward, at, into, for

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

 looks

-0.15

çľĭ

-0.12

 looked

-0.12

 look

-0.12

 Looks

-0.12

looks

-0.11

 inward

-0.10

 LOOK

-0.10

iona

-0.10

keterangan

-0.10

POSITIVE LOGITS

outs

0.12

 closely

0.12

 tháº¥y

0.11

/sm

0.10

 forward

0.10

æľĽ

0.10

orch

0.10

AndFeel

0.09

 Glass

0.09

lass

0.09

Activations Density 0.018%