INDEX

Explanations

sequence continuation

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

むしろ

1.57

 fact

1.17

 오히려

1.13

 niti

1.08

nor

1.02

Indeed

1.00

Nevertheless

0.91

そもそも

0.91

 बल्कि

0.84

breakable

0.83

POSITIVE LOGITS

 THEN

4.06

 kemudian

3.97

 then

3.96

 vervolgens

3.95

 سپس

3.94

然后

3.88

 Затем

3.87

then

3.87

 Thereafter

3.86

 Then

3.83

Activations Density 0.633%