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

    Neuronpedia

    INDEX
    Explanations

    steps or procedures in technical instructions

    oai_token-act-pair · gpt-3.5-turbo

    step-by-step instructions or procedural text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    inately
    -0.77
    ciating
    -0.76
    iqueness
    -0.72
     Unic
    -0.72
    itted
    -0.72
    eatures
    -0.71
    ruciating
    -0.70
    ãĥīãĥ©ãĤ´ãĥ³
    -0.69
    ortunately
    -0.69
     Pengu
    -0.69
    POSITIVE LOGITS
    Step
    1.02
     Steps
    0.96
    hens
    0.94
    daughter
    0.88
    hani
    0.88
     Flo
    0.83
     steps
    0.83
    dad
    0.83
    steps
    0.79
    hent
    0.79
    Act Density 0.030%

    No Known Activations