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

    Neuronpedia

    INDEX
    Explanations

    delving deeper into

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    UpDown
    1.01
    Up
    1.01
     Up
    0.96
     up
    0.92
    upload
    0.85
    BtnUp
    0.82
    lių
    0.82
    вной
    0.81
    Deep
    0.80
    ческой
    0.80
    POSITIVE LOGITS
     into
    1.17
    into
    1.05
    wards
    0.94
     wards
    0.80
    stage
    0.77
    Into
    0.76
     clutching
    0.75
    នូវ
    0.75
     Into
    0.74
    ward
    0.74
    Act Density 0.151%

    No Known Activations