INDEX

Explanations

words indicating negative sentiment or claims

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

Strategies

0.98

Folders

0.90

 Når

0.90

strategies

0.89

Different

0.88

0.87

Matrices

0.86

എ

0.86

 multiplicación

0.85

algèbre

0.84

POSITIVE LOGITS

 huge

1.36

 pissed

1.35

 HUGE

1.34

 suspiciously

1.32

 dubious

1.30

 THEIR

1.25

 basically

1.24

 shitty

1.24

 worthless

1.23

 dodgy

1.23

Activations Density 0.035%