INDEX

Explanations

solubility in water and organic solvents

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

vap

-0.10

vap

-0.10

erval

-0.09

 Duffy

-0.09

PRINTF

-0.09

 ancestor

-0.08

reb

-0.08

dent

-0.08

_quit

-0.08

fg

-0.08

POSITIVE LOGITS

 water

0.15

 absolute

0.14

Et

0.13

DM

0.13

dim

0.13

water

0.12

anol

0.12

TH

0.12

IPA

0.11

 ethanol

0.11

Activations Density 0.010%