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

    Neuronpedia

    INDEX
    Explanations

    slightly, progressively

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    with
    0.84
    $.
    0.81
    a
    0.81
    as
    0.77
    서
    0.77
    Idioma
    0.75
    Like
    0.74
     with
    0.73
    like
    0.73
    for
    0.72
    POSITIVE LOGITS
     outlying
    0.81
     turbulent
    0.79
     Sadie
    0.77
     nerdy
    0.76
     psychedelic
    0.76
     Grizz
    0.75
     abrasive
    0.75
     Mavericks
    0.75
     nebul
    0.74
     turquoise
    0.74
    Act Density 0.293%

    No Known Activations