INDEX

Explanations

said followed by word/punctuation

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

𝐬

3.66

ibouti

3.07

𝐲

3.02

𝐥

3.01

𝐚

2.97

𝐭

2.96

 incluidas

2.95

 việc

2.91

เป็น

2.89

ため

2.86

POSITIVE LOGITS

tions

3.30

 goodbye

3.24

2.96

ル

2.95

ro

2.90

ни

2.82

tive

2.77

к

2.70

Subview

2.70

in

2.67

Activations Density 0.037%