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

    Neuronpedia

    INDEX
    Explanations

    conjunctions and linking words used in sentence structure

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    wic
    -0.89
    urden
    -0.85
    grand
    -0.83
    raid
    -0.80
     Own
    -0.79
    anu
    -0.78
    conduct
    -0.78
    Favorite
    -0.78
    udic
    -0.76
    pez
    -0.76
    POSITIVE LOGITS
     consequently
    0.95
     replaced
    0.92
     hence
    0.91
     therefore
    0.91
     thus
    0.84
     there
    0.81
     flowed
    0.80
     everyone
    0.79
     culminated
    0.79
     nobody
    0.79
    Act Density 0.160%

    No Known Activations