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

    Neuronpedia

    INDEX
    Explanations

    emergency hotline SAFE (

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    綃
    0.40
    hiba
    0.38
    चेता
    0.36
    hiva
    0.35
    colLast
    0.35
     uncond
    0.35
    Bak
    0.35
     የበ
    0.35
    珉
    0.34
    𝑟
    0.34
    POSITIVE LOGITS
     المق
    0.41
    型
    0.38
     Selamat
    0.37
     mayhem
    0.35
     Ross
    0.33
    APC
    0.33
     Mac
    0.33
     Mustard
    0.33
    ENANCE
    0.33
     মহাপ
    0.33
    Act Density 0.023%

    No Known Activations