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

    Neuronpedia

    INDEX
    Explanations

    conditional 'if' and word rotation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mango
    0.45
    ActionListener
    0.42
    Action
    0.42
     Action
    0.42
     soaring
    0.42
     fois
    0.41
     action
    0.41
     cognition
    0.40
     ripening
    0.40
     sizzling
    0.40
    POSITIVE LOGITS
    orsunuz
    0.46
    වැ
    0.44
    觀
    0.43
    ttä
    0.43
    ungannya
    0.43
    娌
    0.42
    ya
    0.42
    å
    0.41
    cim
    0.41
    he
    0.41
    Act Density 0.000%

    No Known Activations