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

    Neuronpedia

    INDEX
    Explanations

    in conjunction with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ్‌
    0.73
    rightarrow
    0.70
     terms
    0.69
     resumes
    0.63
    wired
    0.63
     boarded
    0.62
     Pose
    0.62
     plunged
    0.61
    નેસ
    0.61
     decimated
    0.61
    POSITIVE LOGITS
    Keeping
    0.91
    Exchange
    0.80
     ways
    0.80
    Such
    0.74
     turn
    0.73
     keeping
    0.72
     größ
    0.72
     Keeping
    0.71
     Austausch
    0.71
    turn
    0.70
    Act Density 0.110%

    No Known Activations