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

    Neuronpedia

    INDEX
    Explanations

    instances where individuals physically enter a location

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "walked into."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    etheless
    -0.85
     corpus
    -0.74
    entimes
    -0.66
     reluct
    -0.64
     consistency
    -0.64
    gans
    -0.63
    tle
    -0.63
    ector
    -0.62
    DR
    -0.61
    ect
    -0.61
    POSITIVE LOGITS
     Valhalla
    0.77
    veyard
    0.74
     adulthood
    0.74
     Vegas
    0.72
     hypers
    0.70
     Hungary
    0.69
    opol
    0.67
     ABE
    0.66
     Qué
    0.66
    jection
    0.65
    Act Density 0.056%

    No Known Activations