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

    Neuronpedia

    INDEX
    Explanations

    proper names and references to specific places or events

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    veh
    -0.61
    water
    -0.54
    wake
    -0.54
    chest
    -0.53
    seed
    -0.52
     Concern
    -0.52
    bag
    -0.51
    road
    -0.50
    many
    -0.50
    ship
    -0.49
    POSITIVE LOGITS
    arton
    0.71
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    0.71
    IFIED
    0.69
    ibility
    0.67
    ENN
    0.66
    anamo
    0.64
    ancies
    0.64
    iless
    0.64
    essee
    0.63
    é¾
    0.63
    Act Density 0.137%

    No Known Activations