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

    Neuronpedia

    INDEX
    Explanations

    used

    np_max-act · gemini-2.0-flash

    descriptions and explanations related to tools or software used in specific fields, particularly architecture and programming.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on gerund verb forms—that is, words ending in “-ing.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     option
    -0.07
    ([-
    -0.07
     fragmented
    -0.07
    _image
    -0.06
     Problems
    -0.06
     Kaf
    -0.06
     "->
    -0.06
     neighborhood
    -0.06
     ------------------------------------------------------------
    -0.06
     MatDialog
    -0.06
    POSITIVE LOGITS
    AFX
    0.08
    yna
    0.07
    crap
    0.06
     sabe
    0.06
     Gloria
    0.06
    (infile
    0.06
    ussions
    0.06
    цуз
    0.06
    !")↵↵
    0.06
    Susan
    0.06
    Act Density 0.038%

    No Known Activations