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

    Neuronpedia

    INDEX
    Explanations

    actions, requests, or descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    #,
    0.55
    య
    0.52
     Wärm
    0.52
     య
    0.51
    이지만
    0.50
    o
    0.50
    ION
    0.49
    यॉर्क
    0.46
     አይደ
    0.46
     Reino
    0.46
    POSITIVE LOGITS
     mínimos
    0.50
    ہرے
    0.46
    دید
    0.45
     tele
    0.41
    visions
    0.41
    ত্তা
    0.41
     mon
    0.41
    terror
    0.39
    絢
    0.39
     Scenes
    0.39
    Act Density 0.005%

    No Known Activations