INDEX

Explanations

substitute and plug in values

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lí

-0.79

 Мо

-0.76

 البي

-0.76

どうやら

-0.76

tili

-0.74

vejte

-0.73

 hóa

-0.72

teenage

-0.72

К

-0.71

せた

-0.71

POSITIVE LOGITS

 plugging

3.47

 substituting

3.42

 substitute

3.34

 substituted

3.28

 plug

3.28

 substitution

3.19

 plugged

3.14

 Substituting

2.84

Substituting

2.80

 Substitute

2.80

Activations Density 0.092%