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

    Neuronpedia

    INDEX
    Explanations

    phrases related to accidents involving physical harm

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    Cu
    -0.93
     HHS
    -0.93
     Hitman
    -0.88
    hran
    -0.86
     Milan
    -0.85
    ollah
    -0.84
    iaz
    -0.84
     Santorum
    -0.83
     Kimmel
    -0.81
     Pinball
    -0.81
    POSITIVE LOGITS
     trees
    2.16
     tree
    2.14
     Trees
    1.92
     forest
    1.83
     Tree
    1.79
     forests
    1.78
    tree
    1.72
    Tree
    1.71
     forestry
    1.69
     woods
    1.68
    Act Density 0.351%

    No Known Activations