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

    Neuronpedia

    INDEX
    Explanations

    past actions and experiences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ری
    0.99
     it
    0.84
    ב
    0.81
    :
    0.80
     It
    0.75
     manter
    0.75
    ām
    0.74
     an
    0.74
    后
    0.73
    ：
    0.73
    POSITIVE LOGITS
    h
    0.78
    history
    0.75
    previous
    0.75
     ранее
    0.74
    s
    0.73
    hand
    0.72
    ldre
    0.71
    histor
    0.71
    past
    0.70
    uk
    0.68
    Act Density 0.222%

    No Known Activations