INDEX

Explanations

names, acronyms, and places

The neuron activates on proper names and acronyms (tokens written in uppercase or title‐case), i.e. named entities.

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

</h2>

-3.66

-3.59

-3.31

Here

-3.25

One

-3.20

Let

-3.11

After

-3.09

This

-3.00

So

-3.00

POSITIVE LOGITS

檮

3.08

 indivíduos

2.94

 comunicare

2.77



2.77



2.73

 경우

2.70

🛇

2.67

 nebude

2.67

⽥

2.61

 aqueles

2.59

Activations Density 0.070%