INDEX

Explanations

swimming

The neuron consistently fires on occurrences of the root “swim” (in words like swim, swimming, swimmers, swims).

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

那麼

-2.11

 gereken

-1.81

());

-1.73

-1.71

などで

-1.69

 mismo

-1.68

$-$\\

-1.66

]]=

-1.66

ほうが

-1.63

POSITIVE LOGITS

狲

2.03

ésus

2.00

ribune

1.80

🥸

1.79

It

1.79

ྂ

1.77

∽

1.77

Swim

1.77

eningrad

1.73

sonaro

1.73

Activations Density 0.008%