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

    Neuronpedia

    INDEX
    Explanations

    forming concepts and states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    es
    0.95
    ী
    0.92
    el
    0.81
    ো
    0.79
     retrait
    0.79
    ો
    0.78
    heid
    0.76
    höhe
    0.76
    ai
    0.74
    ъем
    0.72
    POSITIVE LOGITS
    ations
    1.43
    aciones
    1.21
    azione
    1.02
    ATIONS
    1.00
    utions
    0.99
    azioni
    0.97
    ativ
    0.93
    acion
    0.92
    ación
    0.92
    ation
    0.89
    Act Density 0.566%

    No Known Activations