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

    Neuronpedia

    INDEX
    Explanations

    unit

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Alford
    1.07
     formdata
    0.99
     Glover
    0.97
    adeh
    0.95
     Gore
    0.95
     Aldrich
    0.95
     coseno
    0.94
     hombres
    0.92
     Grimes
    0.92
     hombre
    0.92
    POSITIVE LOGITS
    Unit
    2.25
     unit
    2.24
     Unit
    2.24
    unit
    2.14
     Units
    2.01
     units
    1.99
    units
    1.79
    Units
    1.76
     UNIT
    1.68
    ユニット
    1.61
    Act Density 0.246%

    No Known Activations