INDEX

Explanations

sudden discovery or unintended action

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

 classifications

0.79

 видно

0.78

 artinya

0.78

 понятно

0.77

 будем

0.76

loj

0.75

主要是

0.75

 podendo

0.75

Bearer

0.74

 daarna

0.73

POSITIVE LOGITS

 accidentally

1.70

 overhear

1.46

 overheard

1.40

 discovers

1.34

 discover

1.30

 suddenly

1.29

 discovered

1.25

 Accident

1.24

accident

1.23

 inadvertently

1.20

Activations Density 0.323%