INDEX

Explanations

words starting with Lam

The neuron is especially triggered by the substring “Lam” in tokens (as in “Lamport,” “Lam Research,” etc.), i.e. it detects occurrences of “Lam” in proper-name contexts.

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

鰨

-3.06

 ส์

-3.02

栊

-2.95

 invitaciones

-2.81

铪

-2.70

 sèche

-2.58

也能

-2.53

ྵ

-2.52

^{-}\

-2.50

 стоковая

-2.50

POSITIVE LOGITS

2.89

–

2.81

’

2.73

2.58

</b>

2.58

 Для

2.52

 большой

2.44

2.42

</h2>

2.36

2.33

Activations Density 0.009%