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

    Neuronpedia

    INDEX
    Explanations

    train speed problems

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     encouraged
    -0.09
     لتحقيق
    -0.09
     endorsed
    -0.09
     encomp
    -0.08
     Ansatz
    -0.08
    okus
    -0.08
     Sob
    -0.08
     fún
    -0.08
     endorse
    -0.07
    ，以
    -0.07
    POSITIVE LOGITS
     passengers
    0.10
     locom
    0.09
    Equipment
    0.09
    Travel
    0.09
     locomotive
    0.09
     stationed
    0.09
     ralent
    0.09
     berger
    0.09
    Head
    0.09
    (Motion
    0.09
    Act Density 0.016%

    No Known Activations