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

    Neuronpedia

    INDEX
    Explanations

    mentions of physical layers, coatings or structures

    oai_token-act-pair · gpt-3.5-turbo

    phrases that refer to multiple layers or levels in a structure or concept

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Concord
    -0.78
     Inqu
    -0.74
     Kenn
    -0.71
    eve
    -0.71
     attendance
    -0.70
    abet
    -0.69
    hun
    -0.68
    ought
    -0.66
     Faith
    -0.65
    ickets
    -0.63
    POSITIVE LOGITS
     layer
    3.76
     Layer
    2.78
     layers
    2.76
    Layer
    2.53
    layer
    2.44
     layered
    1.83
     coating
    1.43
     overlay
    1.14
     stack
    1.14
    ayers
    1.14
    Act Density 0.024%

    No Known Activations