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

    Neuronpedia

    INDEX
    Explanations

    ironic or fitting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ל
    0.59
    نڈ
    0.57
    lds
    0.57
     производстве
    0.57
    nl
    0.56
    Radial
    0.56
     enfermedades
    0.56
    ્લે
    0.56
    qb
    0.55
     chimiques
    0.55
    POSITIVE LOGITS
     curiously
    0.68
     fitting
    0.61
     enigmatic
    0.59
     ironically
    0.59
     didn
    0.58
     ironic
    0.58
     strangely
    0.58
     oddly
    0.58
     revealing
    0.57
     enig
    0.57
    Act Density 0.046%

    No Known Activations