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

    Neuronpedia

    INDEX
    Explanations

    verbs followed by a duration or distance word, indicating a sense of movement or change

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "ve" and its various forms, specifically in different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    assian
    -0.91
     Kingston
    -0.81
    iewicz
    -0.75
    raph
    -0.67
    aiden
    -0.67
     Barth
    -0.65
    enhagen
    -0.64
     CM
    -0.64
     Clayton
    -0.64
     Archdemon
    -0.63
    POSITIVE LOGITS
     ve
    3.80
    Ve
    1.80
     veget
    1.65
     Veg
    1.63
     Ve
    1.60
    ve
    1.56
     vegan
    1.44
     Vegan
    1.38
     Veget
    1.37
     vegetarian
    1.24
    Act Density 0.021%

    No Known Activations