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

    Neuronpedia

    INDEX
    Explanations

    believes, claims, predicts, knows

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     чого
    0.43
     phrases
    0.38
     чем
    0.37
     phr
    0.36
    岐
    0.36
     чего
    0.36
     referred
    0.36
     выбора
    0.34
     whose
    0.34
     WHAT
    0.34
    POSITIVE LOGITS
     exists
    0.58
     happened
    0.57
     Happened
    0.57
     occurs
    0.57
     existed
    0.55
     происходит
    0.54
     відбувається
    0.54
     existir
    0.53
     ocurrió
    0.53
     Happens
    0.53
    Act Density 0.011%

    No Known Activations