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

    Neuronpedia

    INDEX
    Explanations

    the name "aj" followed by a number (such as "aj9" or "aj10") appearing in the text

    oai_token-act-pair · gpt-3.5-turbo

    the name "Aj" or instances of the name mentioned in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Seb
    -0.75
     Panther
    -0.73
    ACTED
    -0.71
     Furious
    -0.71
     Sporting
    -0.70
     Privacy
    -0.68
     Tigers
    -0.68
     Concord
    -0.67
     Scream
    -0.67
    VALUE
    -0.66
    POSITIVE LOGITS
    aj
    1.34
    ournal
    1.32
    ournals
    1.11
    anan
    1.06
    itsu
    1.04
    arat
    1.04
    awa
    1.03
    apan
    1.02
    estic
    1.02
    aja
    1.00
    Act Density 0.005%

    No Known Activations