INDEX

Explanations

speaking role/context

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

すべての

0.81

зге

0.80

enabled

0.75

它可以

0.75

そのため

0.74

에는

0.73

generate

0.73

 serta

0.71

あとは

0.70

 обеспечить

0.70

POSITIVE LOGITS

 exasper

1.87

 discussing

1.81

 talking

1.65

 commenting

1.59

 interviewed

1.58

 speaking

1.56

 angrily

1.56

 describing

1.54

 explaining

1.54

 explicando

1.51

Activations Density 0.130%