INDEX

Explanations

The neuron is tuned to detect numeric measurement tokens, especially decimal numbers.

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

 temperature

0.85

 температур

0.80

 Temperature

0.77

 온도

0.73

 amplitude

0.70

Temperature

0.70

 Coefficient

0.66

 Amplitude

0.66

 temperatur

0.65

 coefficient

0.64

POSITIVE LOGITS

observer

0.71

 observer

0.68

 strategie

0.66

Observers

0.62

 നാട്ട

0.62

observers

0.61

reference

0.60

ْر

0.58

Reference

0.58

 бать

0.58

Activations Density 0.156%