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

    Neuronpedia

    INDEX
    Explanations

    future direction

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Closest
    0.45
    closest
    0.42
     avviene
    0.41
     самые
    0.40
    接近
    0.39
     делаю
    0.38
     OSError
    0.38
    වේ
    0.38
     самый
    0.37
     principalement
    0.37
    POSITIVE LOGITS
     headed
    0.57
     destination
    0.55
     heading
    0.53
    heading
    0.50
     destinations
    0.50
    Heading
    0.48
     dirigés
    0.47
    destination
    0.46
     trajectory
    0.46
    headed
    0.44
    Act Density 0.011%

    No Known Activations