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

    Neuronpedia

    INDEX
    Explanations

    overhe

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     cons
    -0.07
    遣
    -0.07
    iam
    -0.06
     sem
    -0.06
     갖
    -0.06
     Command
    -0.06
     portrays
    -0.06
    给
    -0.06
     realized
    -0.06
    ади
    -0.06
    POSITIVE LOGITS
     affordable
    0.07
                      ↵
    0.07
    mg
    0.07
     Респ
    0.07
     Deniz
    0.07
    .transition
    0.07
    .named
    0.06
    .quick
    0.06
    ichern
    0.06
    .Dynamic
    0.06
    Act Density 0.000%

    No Known Activations