INDEX

Explanations

concepts followed by punctuation or conjunctions

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 maravilloso

1.36

 meravigli

1.26

 maravilh

1.13

 тебя

1.11

 merveille

1.11

 maravill

1.10

 bạn

1.06

 тебе

1.06

 increí

1.06

 mensen

1.06

POSITIVE LOGITS

 during

1.90

 after

1.62

 within

1.59

 despite

1.56

 with

1.55

 under

1.52

 without

1.49

 while

1.48

 when

1.45

during

1.44

Activations Density 2.937%