INDEX

Explanations

connecting ideas with conjunctions

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

及び

2.17

および

2.14

및

2.01

및

1.91

 veya

1.90

などは

1.80

または

1.80

":["

1.78

バージョン

1.78

及

1.73

POSITIVE LOGITS

 interestingly

2.20

 unlike

2.19

 frankly

2.16

 crucially

2.14

 since

2.12

 despite

2.00

 alas

1.96

 insofar

1.90

 lately

1.89

 strangely

1.85

Activations Density 1.302%