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

    Neuronpedia

    INDEX
    Explanations

    Scientific notation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    胸
    -0.07
     лап
    -0.07
    ران
    -0.07
     composite
    -0.07
     evolutionary
    -0.07
    verd
    -0.07
     avenues
    -0.07
    joining
    -0.07
     permissions
    -0.07
    祭
    -0.07
    POSITIVE LOGITS
     ideally
    0.09
     Ideally
    0.09
     preferably
    0.08
    /pre
    0.08
     luôn
    0.08
    -Cl
    0.08
     tối
    0.08
     adject
    0.08
     Typically
    0.08
     عادة
    0.08
    Act Density 0.004%

    No Known Activations