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

    Neuronpedia

    INDEX
    Explanations

    phrases containing black and white imagery

    oai_token-act-pair · gpt-3.5-turbo

    repetitive conjunctions and their use in connecting various ideas or elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    SHIP
    -1.01
    ONSORED
    -0.83
    ãĤ§
    -0.81
    ADRA
    -0.79
    unia
    -0.78
    URA
    -0.78
    OTOS
    -0.77
     EDITION
    -0.77
    Administ
    -0.73
    :[
    -0.72
    POSITIVE LOGITS
    haired
    0.90
     shiny
    0.81
     gritty
    0.80
    skinned
    0.78
     juicy
    0.77
     waterproof
    0.77
    blooded
    0.75
     sturdy
    0.74
     fluffy
    0.72
     curly
    0.72
    Act Density 0.438%

    No Known Activations