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

    Neuronpedia

    INDEX
    Explanations

    instances of phrases related to different forms or ways things can appear or be presented

    oai_token-act-pair · gpt-3.5-turbo

    different forms of expression or presentation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Hots
    -0.74
    avorite
    -0.74
    iolet
    -0.66
     sear
    -0.65
    incial
    -0.64
    Bridge
    -0.64
     Duty
    -0.64
     Thro
    -0.62
    Streamer
    -0.62
    amily
    -0.61
    POSITIVE LOGITS
    aldehyde
    1.37
    ative
    1.06
    ulating
    0.95
    fitting
    0.88
    ulator
    0.84
    idable
    0.82
    atives
    0.80
    ulates
    0.78
    ula
    0.77
    ulators
    0.77
    Act Density 0.013%

    No Known Activations