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

    Neuronpedia

    INDEX
    Explanations

    phrases urging action

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "Let" followed by a directive or suggestion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     unspecified
    -0.70
     affected
    -0.67
     dead
    -0.65
     invasive
    -0.65
     damaged
    -0.64
     active
    -0.63
     battery
    -0.63
     equivalent
    -0.63
     planned
    -0.62
     experienced
    -0.61
    POSITIVE LOGITS
    Let
    3.07
     Let
    2.19
     LET
    1.66
    Consider
    1.62
     Lets
    1.61
    LET
    1.59
    let
    1.51
    Here
    1.50
     let
    1.50
    Alright
    1.49
    Act Density 0.009%

    No Known Activations