INDEX

Explanations

magnitude of outcomes

The neuron detects strong intensity words—i.e. adjectives or adverbs like “significant,” “substantial,” and “considerable” used to emphasize magnitude.

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

you

-1.98

-1.72

You

-1.64

を書

-1.60

-1.53

 sinnvoll

-1.52

 dismal

-1.50

fillment

-1.49

 لل

-1.47

蜻

-1.45

POSITIVE LOGITS

</em>

2.20

and

1.63

 assistance

1.61

 equipment

1.52

》

1.43

("

1.42

/}{

1.41

 जिसमें

1.41

");

1.40

 québéco

1.40

Activations Density 0.128%