INDEX

Explanations

multilingual or code syntax elements

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

raient

2.12

 '+':

2.04

ጨም

1.89

pard

1.85

езде

1.82

 indicaciones

1.82

ገኙ

1.82

સર

1.82

बता

1.81

varande

1.79

POSITIVE LOGITS

1.61

ti

1.51

]}

1.50

 causal

1.48

al

1.45

 doğ

1.42

 herunter

1.41

 Međutim

1.40

 abut

1.40

 cleft

1.39

Activations Density 0.000%