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

    Neuronpedia

    INDEX
    Explanations

    head

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Handling
    0.43
     HANDLING
    0.41
    Handling
    0.40
     sencillo
    0.40
     handling
    0.39
     हुँ
    0.38
     harmonious
    0.38
    न्वयन
    0.38
    FAPI
    0.37
     lepid
    0.36
    POSITIVE LOGITS
    quarters
    0.77
    quartered
    0.76
    quarter
    0.74
    lamp
    0.68
    lamps
    0.68
    gear
    0.64
    shots
    0.62
    liner
    0.61
    liners
    0.60
    shot
    0.59
    Act Density 0.029%

    No Known Activations