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

    Neuronpedia

    INDEX
    Explanations

    the word "each" and its variations, indicating a focus on repetitive or plural instances in context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     fores
    -0.59
    er
    -0.58
    ൃ
    -0.53
    者
    -0.52
     dô
    -0.51
    im
    -0.50
    hor
    -0.50
    ter
    -0.50
    Cone
    -0.49
    URE
    -0.49
    POSITIVE LOGITS
     each
    3.03
    each
    2.69
     EACH
    2.65
    EACH
    2.65
     Each
    2.61
    Each
    2.58
     chaque
    2.14
    Chaque
    2.13
     Chaque
    2.05
     ciasc
    1.98
    Act Density 0.082%

    No Known Activations