INDEX

Explanations

This is describing state

tokens that are part of the assistant's explanatory/answering text (i.e., model response content).

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

 pouvaient

0.74

它可以

0.70

 possano

0.67

 puissent

0.67

 podía

0.66

 pourront

0.64

 નહીં

0.64

 aient

0.63

都需要

0.63

 puedan

0.62

POSITIVE LOGITS

is

2.02

 είναι

2.02

är

1.58

 adalah

1.47

yra

1.38

are

1.37

 है

1.36

是

1.36

 është

1.36

 არის

1.33

Activations Density 2.561%