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

    Neuronpedia

    INDEX
    Explanations

    words related to unpredictability

    oai_token-act-pair · gpt-3.5-turbo

    terms indicating unpredictability and erratic behavior

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    HI
    -0.84
    VEL
    -0.80
    marks
    -0.80
    bourg
    -0.78
    vest
    -0.78
    tha
    -0.76
    ISA
    -0.76
    arte
    -0.76
    CLA
    -0.75
    href
    -0.74
    POSITIVE LOGITS
     unpredict
    1.36
     unpredictable
    1.33
     erratic
    1.10
     unstable
    0.92
     combust
    0.90
     uncontroll
    0.89
     fluctuations
    0.88
     unreliable
    0.88
     volatility
    0.86
    theless
    0.84
    Act Density 0.011%

    No Known Activations