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

    Neuronpedia

    INDEX
    Explanations

    locations or places

    oai_token-act-pair · gpt-3.5-turbo

    situations involving violence or crime

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    theless
    -0.67
     anecd
    -0.63
     unequivocally
    -0.59
    ulative
    -0.58
     underestimated
    -0.58
     learnt
    -0.56
     fundamentally
    -0.54
     succeeded
    -0.53
     (>
    -0.52
     logically
    -0.51
    POSITIVE LOGITS
     picnic
    0.65
     kios
    0.63
     dinner
    0.63
     mansion
    0.62
     encamp
    0.61
     mast
    0.60
     supper
    0.59
     tower
    0.58
     cafe
    0.58
     airst
    0.58
    Act Density 1.071%

    No Known Activations