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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     miscellaneous
    -0.06
     offsetX
    -0.06
                                 ↵
    -0.06
    -0.06
     jon
    -0.06
                                   ↵
    -0.06
    -move
    -0.06
    	swap
    -0.05
    .menuStrip
    -0.05
    erox
    -0.05
    POSITIVE LOGITS
    одав
    0.07
    wb
    0.07
    acman
    0.07
     водой
    0.07
    输出
    0.07
    可是
    0.07
    يف
    0.06
     نماید
    0.06
     Chapel
    0.06
     Vec
    0.06
    Act Density 0.008%

    No Known Activations