INDEX

Explanations

in turn

The neuron activates on the causal connective “in turn.”

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

of

-2.16

for

-2.03

on

-1.92

</>

-1.82

 lakini

-1.80

in

-1.78

by

-1.62

 процес

-1.62

<td>

-1.61

 tetapi

-1.61

POSITIVE LOGITS

 gradualmente

1.54

 violenta

1.51

 itself

1.51

螃

1.45

 utilisés

1.45

期刊论文

1.44

راهن

1.43

 puramente

1.42

Britannique

1.41

 curiosa

1.41

Activations Density 0.016%