INDEX

Explanations

8

This neuron activates on numeric tokens, especially monetary amounts and percentages.

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

0.43

Université

0.39

the

0.38

Miscellaneous

0.37

LowerCase

0.35

Account

0.34

the

0.34

Season

0.34

POSITIVE LOGITS

 हजार

0.44

 ઉપરાંત

0.44

 因为

0.41

 सय

0.41

 因為

0.41

 पचास

0.40

 നൂ

0.40

 ಸಾವಿರ

0.40

 ойношот

0.39

ivasena

0.39

Activations Density 0.356%