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

    Neuronpedia

    INDEX
    Explanations

    bloss

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     historic
    -0.07
     tape
    -0.06
    meaning
    -0.06
     por
    -0.06
    iking
    -0.06
    crire
    -0.06
    	string
    -0.06
     throat
    -0.06
    trade
    -0.06
    enstein
    -0.06
    POSITIVE LOGITS
     حق
    0.07
    าจารย
    0.07
     Mort
    0.07
     послед
    0.07
     investigate
    0.07
    کاران
    0.07
     вис
    0.06
     Butt
    0.06
     Touch
    0.06
     trustees
    0.06
    Act Density 0.002%

    No Known Activations