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

    Neuronpedia

    INDEX
    Explanations

    specific entities or concepts mentioned in the context, ranging from sports teams to fonts, religions, races, and more

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    nouns referring to specific entities or categories within a particular domain or context.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jiahuipeng
    New Auto-Interp
    Negative Logits
     Vest
    -0.62
    hang
    -0.61
     patience
    -0.60
    sbm
    -0.60
     Supplemental
    -0.59
    =-=-
    -0.58
     Bere
    -0.57
     Cla
    -0.55
    ================================
    -0.54
     appre
    -0.54
    POSITIVE LOGITS
     imaginable
    0.90
    Id
    0.74
     or
    0.72
     besides
    0.70
     combinations
    0.69
     mates
    0.69
     descriptor
    0.68
     identifier
    0.67
    acters
    0.66
     (/
    0.66
    Act Density 0.564%

    No Known Activations