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

    Neuronpedia

    INDEX
    Explanations

    API definition

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ousted
    0.49
    は約
    0.45
    fillRect
    0.44
    honia
    0.43
    Fra
    0.43
     appunto
    0.43
    ীষ
    0.42
     interacted
    0.42
    Left
    0.41
    श्री
    0.41
    POSITIVE LOGITS
     surfactant
    0.44
     quán
    0.40
     miner
    0.39
     HSM
    0.39
     сете
    0.39
     workday
    0.39
    一遍
    0.38
     подрост
    0.38
    সাথে
    0.38
     дизайнер
    0.38
    Act Density 0.006%

    No Known Activations