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

    Neuronpedia

    INDEX
    Explanations

    to will

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     כאשר
    0.84
     לאחר
    0.81
     באמצעות
    0.77
     zunächst
    0.71
    Enable
    0.68
     дальнейшем
    0.67
    Become
    0.65
    suitable
    0.65
     مطابق
    0.64
    此外
    0.63
    POSITIVE LOGITS
     knows
    1.53
     hates
    1.45
     wants
    1.43
     thinks
    1.42
     owns
    1.35
     loves
    1.30
     didn
    1.27
     knew
    1.24
     hasn
    1.24
     has
    1.23
    Act Density 1.351%

    No Known Activations