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

    Neuronpedia

    INDEX
    Explanations

    building spaces

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ausible
    0.87
     quantifiable
    0.83
    独自の
    0.83
     deleterious
    0.82
     idiosyncratic
    0.79
     disease
    0.76
     deliberately
    0.76
     detrimental
    0.75
     iteratively
    0.74
    embangkan
    0.73
    POSITIVE LOGITS
     atrium
    1.95
     Auditorium
    1.92
     auditorium
    1.87
     hallway
    1.80
     Pavilion
    1.68
     Hall
    1.68
     courtyard
    1.65
     hallways
    1.63
     pavilion
    1.61
     Lounge
    1.61
    Act Density 0.050%

    No Known Activations