INDEX

Explanations

classification metrics recall AUC kappa accuracy gain

The neuron strongly activates on tokens naming quantitative measures, metrics, or similarity indices (e.g. accuracy, recall, precision, AUC, mutual information, Tanimoto, Jaccard, YOLO) used to evaluate or describe models and data.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

</h3>

-1.62

înes

-1.57

⁛

-1.52

☶

-1.49

๛

-1.45

⮑

-1.43

踭

-1.42

ógicos

-1.42

usitis

-1.41

umsy

-1.40

POSITIVE LOGITS

</b>

1.95

1.85

 quelques

1.36

 consider

1.31

1.27

 بعض

1.26

 harán

1.25

 ciertas

1.21

 podremos

1.20

Activations Density 0.009%