INDEX

Explanations

come

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

 accommodate

0.77

 consume

0.72

consume

0.71

 apologize

0.67

 notify

0.64

看看

0.63

 penetrate

0.62

 apologise

0.62

eat

0.61

 participate

0.61

POSITIVE LOGITS

 yaşında

0.61

같

0.60

 hük

0.59

RAM

0.58

grown

0.57

ลัง

0.56

 Comics

0.55

ASMA

0.54

 Commander

0.53

Beh

0.53

Activations Density 0.005%