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

    Neuronpedia

    INDEX
    Explanations

    mentions of physical edges or boundaries

    oai_token-act-pair · gpt-3.5-turbo

    references to edges and seams

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    HR
    -0.72
    ument
    -0.69
     Memor
    -0.68
    piracy
    -0.66
    hr
    -0.65
    pelling
    -0.63
    million
    -0.63
    vation
    -0.63
    study
    -0.62
    prise
    -0.61
    POSITIVE LOGITS
     edges
    3.49
     seams
    2.05
     corners
    1.92
     surfaces
    1.79
     margins
    1.47
     borders
    1.46
     cracks
    1.44
     folds
    1.33
     transitions
    1.33
     coasts
    1.32
    Act Density 0.022%

    No Known Activations