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

    Neuronpedia

    INDEX
    Explanations

    malignant

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Water
    -0.07
     Robot
    -0.07
    _bootstrap
    -0.07
    erreur
    -0.07
    ERO
    -0.07
    .Trans
    -0.07
     HomePage
    -0.07
     Kh
    -0.07
     rob
    -0.07
    呼ば
    -0.06
    POSITIVE LOGITS
     malignant
    0.11
     malign
    0.08
    MimeType
    0.07
    Replacement
    0.06
     complains
    0.06
    λυ
    0.06
     mij
    0.06
     DISCLAIMER
    0.06
     mong
    0.06
     paving
    0.06
    Act Density 0.004%

    No Known Activations