INDEX

Explanations

addictive behaviors and dependency

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

 thirsty

-0.11

 Lounge

-0.11

 Bonus

-0.10

 Decay

-0.09

 bonus

-0.09

.DataType

-0.08

Hud

-0.08

 tang

-0.08

egin

-0.08

 relentlessly

-0.08

POSITIVE LOGITS

 addiction

0.46

add

0.42

Add

0.42

add

0.38

Add

0.38

-add

0.38

 Addiction

0.36

 dependency

0.35

_add

0.34

 addict

0.33

Activations Density 0.148%