INDEX

Explanations

oxidation and tidy

The neuron tends to fire on single, abstract‐noun tokens (e.g. oxidation, apathy, tidy).

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>

-3.38

艄

-3.23

屨

-3.20

湏

-3.14

轳

-3.13

呿

-3.03

咴

-2.98

撄

-2.97

怍

-2.89

”،

-2.84

POSITIVE LOGITS

 With

2.77

 resolute

2.58

嶇

2.56

</b>

2.53

 unrelenting

2.45

 benevolent

2.25

使った

2.25

是

2.25

2.22

 youre

2.19

Activations Density 0.020%