INDEX

Explanations

that

The neuron is tuned to pick out abstract, conceptual nouns (like “structure,” “examples,” “needs,” “time,” etc.).

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

 ceea

-1.10

Ngh

-0.84

Ciudad

-0.79

 degli

-0.78

вин

-0.78

ोन

-0.77

 delle

-0.77

dei

-0.76

 cherchez

-0.76

 samoch

-0.74

POSITIVE LOGITS

 that

1.60

That

1.59

 That

1.53

 THAT

1.42

นั้น

1.31

 daarvan

1.27

 there

1.27

 đó

1.20

 thats

1.18

 لذلك

1.13

Activations Density 0.102%