INDEX

Explanations

introduces exceptions or qualifications

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

but

1.42

but

1.27

But

1.20

But

1.16

 लेकिन

1.16

 tetapi

1.14

BUT

1.13

从而

1.11

។

1.09

BUT

1.07

POSITIVE LOGITS

 meanwhile

2.06

 however

1.75

 unlike

1.69

however

1.64

 unsurprisingly

1.63

 like

1.58

 meantime

1.47

unlike

1.46

无论是

1.45

 fueled

1.43

Activations Density 0.148%