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

    Neuronpedia

    INDEX
    Explanations

    comparative and qualitative attributes related to improvement and effectiveness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    more or less

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    DataAnnotations
    -0.43
    RTLI
    -0.40
     Jungfrau
    -0.39
    dcterms
    -0.39
     Kunststoff
    -0.38
    kes
    -0.38
    ushka
    -0.38
    sidemargin
    -0.38
     Keenan
    -0.37
     Reihen
    -0.36
    POSITIVE LOGITS
     than
    0.94
     niż
    0.71
     THAN
    0.69
     brighter
    0.67
     harder
    0.64
     smoother
    0.63
     funnier
    0.63
     darker
    0.63
    Than
    0.62
     happier
    0.60
    Act Density 0.323%

    No Known Activations