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

    Neuronpedia

    INDEX
    Explanations

    phrases containing special characters, like âĢ and ľ, often at the beginning of words

    oai_token-act-pair · gpt-3.5-turbo

    emphatic expressions or symbols that convey intensity or strong feelings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     interference
    -0.69
     ABE
    -0.68
     Tasman
    -0.66
     PST
    -0.65
     shroud
    -0.64
    swick
    -0.63
     Mirage
    -0.63
     spending
    -0.63
     RAD
    -0.63
     Niet
    -0.63
    POSITIVE LOGITS
    IJ
    1.06
    ª
    1.04
    Ĵ
    1.04
    appropriately
    1.03
    ł
    1.02
    against
    1.01
    ij
    1.00
    almost
    0.97
    ¹
    0.96
    yet
    0.96
    Act Density 0.105%

    No Known Activations