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

    Neuronpedia

    INDEX
    Explanations

    phrases related to standing up for beliefs or principles

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Monaco
    -0.76
    jing
    -0.74
     understatement
    -0.62
     mined
    -0.62
     Destination
    -0.61
    ahn
    -0.61
     Kahn
    -0.60
    cest
    -0.59
     <-
    -0.59
    got
    -0.59
    POSITIVE LOGITS
    stairs
    1.00
     paddle
    0.78
    rights
    0.76
    raised
    0.75
    ris
    0.71
    roots
    0.71
    issan
    0.70
    atoon
    0.69
    steen
    0.68
    river
    0.68
    Act Density 7.316%

    No Known Activations