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

    Neuronpedia

    INDEX
    Explanations

    focus on or web UI

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    curvature
    0.52
    spreading
    0.52
    ние
    0.50
    cluding
    0.49
     ее
    0.49
     её
    0.48
    INCLUDING
    0.47
    Unh
    0.47
    名の
    0.47
    irió
    0.47
    POSITIVE LOGITS
     pul
    0.49
     chil
    0.44
     womb
    0.44
     しかし
    0.43
     booze
    0.42
     marin
    0.42
     competes
    0.42
     HIIT
    0.42
     bodysuit
    0.42
     kit
    0.41
    Act Density 0.001%

    No Known Activations