INDEX

Explanations

let me know

The neuron flags conversational support‐ticket language, especially polite second‐person requests and follow-up phrases (e.g. “Could you…,” “Let me know,” “thanks,” “welcome to,” etc.).

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

 oreilles

0.57

 आरोपों

0.56

 பெற

0.54

 fuera

0.52

 evaluar

0.52

 domande

0.52

质疑

0.51

 investigar

0.51

収集

0.51

 разобраться

0.51

POSITIVE LOGITS

 tell

2.59

告诉

2.58

 Tell

2.38

Tell

2.38

告知

2.36

告訴

2.36

tell

2.30

 TELL

2.27

 telling

2.22

 notify

2.19

Activations Density 0.850%