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

    Neuronpedia

    INDEX
    Explanations

    the occurrence of size-related terms in various contexts

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

    sz prefix in names and variables

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    databind
    -0.57
     BoxDecoration
    -0.48
    ılık
    -0.48
    ine
    -0.42
     tramonto
    -0.42
     Persi
    -0.42
    hline
    -0.41
     сливо
    -0.41
     Vine
    -0.41
    ))))
    -0.40
    POSITIVE LOGITS
    sz
    2.16
    Sz
    1.88
     Sz
    1.87
     sz
    1.84
    SZ
    1.70
     SZ
    1.66
    ksz
    0.95
    psz
    0.94
     Szab
    0.91
    tsz
    0.90
    Act Density 0.003%

    No Known Activations