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

    Neuronpedia

    INDEX
    Explanations

    the word "which," often in contexts that describe definitions or qualifications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     chofe
    -0.61
     myſelf
    -0.60
     Chriftian
    -0.53
     उसने
    -0.51
     Greeks
    -0.51
     fubject
    -0.51
    MD
    -0.50
     noft
    -0.50
     Argo
    -0.49
     himſelf
    -0.49
    POSITIVE LOGITS
    autant
    0.68
    googleapis
    0.67
    guro
    0.64
     ujednoznacz
    0.61
     so
    0.61
    Kanpo
    0.61
    scars
    0.60
     <>",
    0.59
    sprechend
    0.59
    pecially
    0.58
    Act Density 0.075%

    No Known Activations