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

    Neuronpedia

    INDEX
    Explanations

    verbs related to actions or instructions to do something

    oai_token-act-pair · gpt-3.5-turbo

    the word "proceed" and its variants in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    tek
    -0.76
    tal
    -0.76
    bie
    -0.73
    ton
    -0.72
    var
    -0.70
    ty
    -0.67
    cup
    -0.66
    tu
    -0.65
    box
    -0.64
    aez
    -0.64
    POSITIVE LOGITS
     cautiously
    1.00
     onward
    0.88
     smoothly
    0.87
     Proceed
    0.85
     toward
    0.83
     calmly
    0.81
     through
    0.80
     towards
    0.78
     withd
    0.76
     onwards
    0.74
    Act Density 0.045%

    No Known Activations