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

    Neuronpedia

    INDEX
    Explanations

    questions asking for numerical quantities

    oai_token-act-pair · gpt-3.5-turbo

    the repeated phrase "how many."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    allery
    -0.79
    ital
    -0.78
    undle
    -0.70
    olitics
    -0.69
    ¶æ
    -0.69
    hern
    -0.68
    HF
    -0.68
     guiActiveUnfocused
    -0.67
    ategor
    -0.67
    ocracy
    -0.67
    POSITIVE LOGITS
     times
    1.02
     servings
    0.85
     calories
    0.85
     iterations
    0.82
     parentheses
    0.82
     copies
    0.74
     digits
    0.74
     bells
    0.72
     thousand
    0.71
     repet
    0.71
    Act Density 0.017%

    No Known Activations