INDEX

Explanations

phrasal verbs

The neuron strongly activates on action‐oriented verbs (particularly those denoting motion or doing, like “run,” “hold,” “getting/taking,” “look,” “get,” 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

などが

-1.74

を

-1.73

などを

-1.55

と

-1.55

が

-1.45

などは

-1.44

to

-1.43

なども

-1.38

所以在

-1.38

そのため

-1.34

POSITIVE LOGITS

and

1.87

 which

1.72

 bicicletas

1.52

だよね

1.49

をしました

1.46

any

1.41

 cápsulas

1.41

 esconder

1.38

 soltar

1.37

 toneladas

1.35

Activations Density 0.050%