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

    Neuronpedia

    INDEX
    Explanations

    substitution and calculation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ামো
    0.42
    create
    0.41
     созда
    0.41
    íos
    0.40
    Ꮬ
    0.39
    Create
    0.39
    White
    0.39
    objects
    0.39
    Traditional
    0.39
    agaman
    0.38
    POSITIVE LOGITS
     substitute
    0.81
     Substituting
    0.74
     substituting
    0.74
     substitution
    0.73
     Substitute
    0.71
     substitutes
    0.68
     замі
    0.67
     substituted
    0.65
     plug
    0.65
     plugging
    0.65
    Act Density 0.285%

    No Known Activations