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

    Neuronpedia

    INDEX
    Explanations

    Probability

    np_max-act · gemini-2.0-flash

    The neuron detects hedge/adverbial cues expressing uncertainty or probability (e.g., “probably,” “likely,” “generally”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     annually
    -0.07
     Consum
    -0.07
    addons
    -0.06
     getItemCount
    -0.06
    hma
    -0.06
    Makes
    -0.06
    ارت
    -0.06
     subdir
    -0.06
    ức
    -0.06
    _tag
    -0.06
    POSITIVE LOGITS
     classy
    0.07
    =L
    0.07
    (nr
    0.07
    .Debugf
    0.07
    ことで
    0.07
    .TRAN
    0.07
    wayne
    0.06
    (ListNode
    0.06
    _male
    0.06
    ，将
    0.06
    Act Density 0.011%

    No Known Activations