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

    Neuronpedia

    INDEX
    Explanations

    historical or original contexts of phrases or terms

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Originally" and its various forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    antics
    -0.88
    eer
    -0.78
    atics
    -0.74
    antes
    -0.72
    lain
    -0.71
    estern
    -0.70
    agascar
    -0.70
    ully
    -0.70
    riz
    -0.69
    imm
    -0.68
    POSITIVE LOGITS
     Posted
    0.95
     Published
    0.83
     nicknamed
    0.76
     priced
    0.75
     Asked
    0.75
     posted
    0.73
     elected
    0.71
     dubbed
    0.70
    Reviewed
    0.69
     designed
    0.69
    Act Density 0.022%

    No Known Activations