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

    Neuronpedia

    INDEX
    Explanations

    setting the stage

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    izens
    0.39
     labelled
    0.38
    config
    0.37
     enabled
    0.37
     to
    0.37
    ructuring
    0.37
     지정
    0.36
     specified
    0.35
     incorrectly
    0.35
     relegated
    0.35
    POSITIVE LOGITS
     контек
    0.51
     expectations
    0.47
    CameraController
    0.47
     направлении
    0.47
     Tone
    0.46
     contexto
    0.46
     boundaries
    0.45
     направления
    0.45
    条件
    0.44
    न्नाथ
    0.43
    Act Density 0.008%

    No Known Activations