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

    Neuronpedia

    INDEX
    Explanations

    In

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     for
    -0.07
     how
    -0.07
     pelos
    -0.07
    ый
    -0.06
     forge
    -0.06
     дол
    -0.06
    .Zero
    -0.06
     zru
    -0.06
    backend
    -0.06
     metals
    -0.06
    POSITIVE LOGITS
     In
    0.09
    	In
    0.08
    In
    0.08
     Camb
    0.06
     isIn
    0.06
     tanks
    0.06
     Gain
    0.06
    	en
    0.06
    -In
    0.06
     (_,
    0.06
    Act Density 0.107%

    No Known Activations