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

    Neuronpedia

    INDEX
    Explanations

    struggles with commitment or anxiety

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     gloss
    0.43
     already
    0.43
     contingent
    0.42
     Already
    0.41
     Gloss
    0.41
     ontv
    0.40
     enjoyable
    0.39
     anticipating
    0.39
     optics
    0.39
     drawing
    0.38
    POSITIVE LOGITS
     inicialmente
    0.75
     initially
    0.71
     awalnya
    0.67
    最初は
    0.65
     despair
    0.63
     Initially
    0.61
    當時
    0.61
    Initially
    0.61
    until
    0.60
     until
    0.58
    Act Density 0.095%

    No Known Activations