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

    Neuronpedia

    INDEX
    Explanations

    alignment with goals or values

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    d
    0.85
     }
    0.66
    s
    0.62
    h
    0.61
    of
    0.61
    l
    0.57
    <h2>
    0.53
    ()]
    0.53
    g
    0.53
    ية
    0.52
    POSITIVE LOGITS
     aligned
    0.86
     aligns
    0.86
     aligning
    0.82
     alignment
    0.80
     Alignment
    0.75
     straight
    0.74
     straightened
    0.74
     STRAIGHT
    0.74
    Straight
    0.73
     Straight
    0.72
    Act Density 0.022%

    No Known Activations