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

    Neuronpedia

    INDEX
    Explanations

    main points

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     pertence
    0.52
     seleccionado
    0.50
     fijo
    0.49
    ሽታ
    0.49
     самого
    0.48
     soldado
    0.48
     Tamaño
    0.47
     automático
    0.47
    プター
    0.46
     Removed
    0.46
    POSITIVE LOGITS
     énorm
    0.50
    <sup>
    0.45
    ki
    0.45
    ku
    0.43
    Sock
    0.41
    נ
    0.41
    0.41
    ł
    0.40
    Brows
    0.40
    èi
    0.39
    Act Density 0.000%

    No Known Activations