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

    Neuronpedia

    INDEX
    Explanations

    suggests or recommends

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     anxious
    0.65
     racers
    0.64
     acceptors
    0.64
     Hj
    0.63
     oscillation
    0.62
     статья
    0.62
     perceber
    0.61
     Epoxy
    0.61
     Cig
    0.61
     oscillations
    0.59
    POSITIVE LOGITS
     পড়েছে
    0.53
    DA
    0.52
    áli
    0.50
     suggestion
    0.50
    ẳng
    0.50
    AndKeys
    0.49
     Suggest
    0.49
    Este
    0.48
    revoke
    0.48
     suggests
    0.48
    Act Density 0.000%

    No Known Activations