Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    verbs of action and direction

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    cómo
    0.93
    욌
    0.89
    если
    0.87
    Cuando
    0.87
    ರಾ
    0.86
    怎么
    0.85
     mivel
    0.85
    где
    0.84
    Dalam
    0.82
    Pokud
    0.82
    POSITIVE LOGITS
     down
    2.06
     off
    1.84
     up
    1.83
     away
    1.83
     into
    1.73
     onto
    1.69
     together
    1.58
     forth
    1.58
     aside
    1.51
     apart
    1.46
    Act Density 0.160%

    No Known Activations