INDEX

Explanations

fear, enemies, difficulty

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

檢查

1.28

 상세

1.28

 সৈন্যদের

1.26

elenggarakan

1.25

뎅

1.24

 سپس

1.23

ilikom

1.22

舉行

1.22

 ரயில்

1.21

 이동

1.20

POSITIVE LOGITS

 quite

1.57

 это

1.47

这不是

1.46

 very

1.44

 really

1.40

 väldigt

1.40

 VERY

1.38

 rất

1.36

 incredibly

1.33

 sehr

1.31

Activations Density 0.783%