INDEX

Explanations

This neuron activates on copular or auxiliary “to be” verbs (is/are/will be/etc.), flagging statements that define or assert something.

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Measurement

0.46

Optim

0.46

Subject

0.45

0.44

Complex

0.42

0.41

Classifier

0.41

С

0.41

Mono

0.41

POSITIVE LOGITS

 available

0.86

 present

0.72

 needed

0.72

 evident

0.71

 used

0.69

 prevalent

0.67

 vorhanden

0.66

 employed

0.66

 involved

0.62

 discernible

0.62

Activations Density 0.942%