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

    Neuronpedia

    INDEX
    Explanations

    instances where help is needed or sought after

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "help" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    theless
    -0.73
    ross
    -0.69
    Pict
    -0.68
    andom
    -0.67
    ategory
    -0.67
    ModLoader
    -0.65
     rall
    -0.63
     Bulg
    -0.62
     Rim
    -0.62
    é¾
    -0.59
    POSITIVE LOGITS
    fully
    1.03
    des
    1.00
     Desk
    0.94
    meet
    0.87
    ful
    0.79
    full
    0.73
     desk
    0.72
    ocating
    0.72
    fulness
    0.72
    Desk
    0.70
    Act Density 0.047%

    No Known Activations