INDEX

Explanations

greetings and introductions

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

硕

2.58

 подрост

2.47

рос

2.34

дера

2.33

лія

2.32

 детали

2.32

нтэр

2.30

 المنتخب

2.30

ўся

2.29

 подро

2.28

POSITIVE LOGITS

}</

1.25

 Kyoto

1.22

 giác

1.17

hal

1.15

 solving

1.13

red

1.11

Oda

1.11

 topic

1.10

ten

1.10

omg

1.09

Activations Density 0.027%