INDEX

Explanations

describing a technique

The neuron specifically fires on occurrences of the words “technique” or “techniques.”

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

鬭

-2.58

_{

-2.34

-2.33

‚

-2.27

or

-2.25

</strong>

-2.22

 effortlessly

-2.14

ができる

-2.11

镦

-2.11

POSITIVE LOGITS

3.28

ſu

2.77

፥

2.52

 Chaque

2.47

CHREIB

2.45

锝

2.39

糨

2.39

 оранже

2.36

 restent

2.36

ſſer

2.36

Activations Density 0.035%