INDEX

Explanations

subject followed by past tense verb

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

 وير

1.61

 দিচ্ছেন

1.51

합니다

1.49

ります

1.49

 देतात

1.44

 ويت

1.42

 ويم

1.42

きます

1.41

 বলছেন

1.40

 ويك

1.39

POSITIVE LOGITS

was

4.17

was

3.98

 wasn

3.92

 était

3.90

had

3.88

 كانت

3.67

 vardı

3.62

 wasnt

3.58

 were

3.55

 didn

3.55

Activations Density 8.879%