INDEX

Explanations

you

the word "you" when the model is directly addressing the user.

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

Be

0.97

Be

0.95

 sollten

0.88

 ক্রমবর্ধমান

0.86

 sollte

0.85

在了

0.85

 puissent

0.84

 fhould

0.83

 geprü

0.81

će

0.80

POSITIVE LOGITS

 want

2.06

 have

1.49

 need

1.47

 wanna

1.44

 know

1.40

 WANT

1.33

 Want

1.31

want

1.31

 think

1.29

 feel

1.18

Activations Density 0.141%