INDEX

Explanations

discussion and debate

The neuron activates on terms referring to organized discussions or forums (e.g. “debate,” “panel,” “discussion,” “sessions”).

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

greeting

0.77

 contacts

0.63

contacts

0.63

contact

0.62

 greeting

0.61

 contact

0.60

 শুভে

0.59

 greetings

0.59

接触

0.59

Greeting

0.57

POSITIVE LOGITS

 debate

3.05

 Debate

2.84

 discussion

2.78

 debates

2.78

debate

2.75

 Discussion

2.66

 débat

2.66

 debating

2.61

discussion

2.59

讨论

2.55

Activations Density 0.399%