INDEX

Explanations

special tokens and grammatical endings

The neuron flags tokens written in non‐Latin scripts (e.g. Chinese or Japanese characters).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 آنچه

-0.91

 which

-0.84

 homicidio

-0.83

ではありません

-0.83

tualmente

-0.82

こともあります

-0.81

اشته

-0.79

獗

-0.78

раи

-0.77

hängig

-0.76

POSITIVE LOGITS

了

2.08

的了

1.68

 anymore

1.68

 rồi

1.60

です

1.48

is

1.48

人了

1.36

我了

1.34

你了

1.28

 것이다

1.28

Activations Density 0.010%