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

    Neuronpedia

    INDEX
    Explanations

    words related to surfaces and irregularities found on them

    oai_token-act-pair · gpt-3.5-turbo

    mentions of various surfaces and related structural characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    avez
    -0.76
    ellig
    -0.73
    azel
    -0.70
    ARR
    -0.70
    hov
    -0.70
    Mand
    -0.68
    Wiki
    -0.67
     intervention
    -0.63
    awk
    -0.63
    leg
    -0.63
    POSITIVE LOGITS
     surfaces
    1.31
    rences
    0.95
     cleaners
    0.81
    matter
    0.74
    FACE
    0.73
    ports
    0.73
    surface
    0.71
     manif
    0.70
     pores
    0.69
     diam
    0.69
    Act Density 0.007%

    No Known Activations