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

    Neuronpedia

    INDEX
    Explanations

    free of negative attributes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     awkwardly
    0.38
    uega
    0.37
     sticky
    0.37
     pertin
    0.36
     awkward
    0.36
    ̘
    0.36
     grainy
    0.35
    ncc
    0.34
    venidas
    0.34
    িয়
    0.34
    POSITIVE LOGITS
    free
    2.48
     free
    2.36
    Free
    2.19
     Free
    2.05
    FREE
    1.95
     FREE
    1.83
     свобод
    1.77
    less
    1.72
     मुक्त
    1.70
    フリー
    1.69
    Act Density 0.029%

    No Known Activations