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

    Neuronpedia

    INDEX
    Explanations

    capabilities/attributes

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     চান
    -0.09
     yahay
    -0.08
     causing
    -0.08
     yihiin
    -0.08
     giving
    -0.08
     saying
    -0.08
     want
    -0.08
     become
    -0.08
     zei
    -0.08
     represent
    -0.08
    POSITIVE LOGITS
     features
    0.26
     boasts
    0.25
     sports
    0.24
    features
    0.24
     Features
    0.24
     possesses
    0.23
    Features
    0.23
     sporting
    0.23
     boast
    0.22
     possède
    0.22
    Act Density 0.111%

    No Known Activations