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

    Neuronpedia

    INDEX
    Explanations

    phrases related to exceeding certain limits or levels

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "exceed" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Sham
    -0.73
    rug
    -0.70
    Ale
    -0.67
    RAW
    -0.66
    choice
    -0.66
    SO
    -0.64
    ief
    -0.63
    pring
    -0.62
    olan
    -0.62
    NetMessage
    -0.61
    POSITIVE LOGITS
     exceed
    0.85
    ceed
    0.83
     expectations
    0.82
    atos
    0.81
     9000
    0.78
    ingly
    0.74
    capacity
    0.74
    imates
    0.73
     capacity
    0.71
     exceeding
    0.71
    Act Density 0.015%

    No Known Activations