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

    Neuronpedia

    INDEX
    Explanations

    Infinity Room

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    importance
    0.64
    0
    0.61
    ick
    0.61
    important
    0.60
    breaking
    0.60
    ି
    0.59
    if
    0.59
    inston
    0.58
    igger
    0.58
    impact
    0.57
    POSITIVE LOGITS
     völlig
    0.80
     vervolgens
    0.79
     isang
    0.73
     paredes
    0.73
     hordes
    0.72
     contenus
    0.69
     sfondo
    0.68
     interstices
    0.68
     darunter
    0.68
     células
    0.68
    Act Density 0.000%

    No Known Activations