INDEX

Explanations

could you

This neuron fires strongly on polite acknowledgments, especially “Thanks” (and related gratitude expressions).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

೪

0.49

 PFAS

0.45

 rumored

0.45

娱乐圈

0.44

 chatbots

0.44

阕

0.43

 बेचा

0.43

ChatGPT

0.42

蔡

0.42

jpuri

0.42

POSITIVE LOGITS

 Thanks

0.69

 thanks

0.68

 thank

0.67

 Thank

0.67

 gracias

0.61

The

0.55

 definitiv

0.55

 धन्यवाद

0.54

Thanks

0.53

 Gracias

0.53

Activations Density 0.149%