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

    Neuronpedia

    INDEX
    Explanations

    descriptive color terms related to darkness and shading

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     débar
    -0.63
     daisy
    -0.61
    Externí
    -0.60
    allocator
    -0.57
    osoever
    -0.57
    SequentialGroup
    -0.56
     daisies
    -0.56
    PhysRev
    -0.54
    ngdoc
    -0.54
    itolo
    -0.54
    POSITIVE LOGITS
     dark
    1.22
     darker
    1.13
    dark
    1.11
     darkly
    1.10
     darkest
    1.07
     DARK
    1.06
     darkness
    1.06
     darken
    1.05
    black
    1.04
     Dark
    1.04
    Act Density 0.604%

    No Known Activations