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

    Neuronpedia

    INDEX
    Explanations

    =

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on object‐construction expressions—lines where code uses “new” (often in a “var x = new Type(…)” or “Type x = new Type(…)” assignment).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    Java code that demonstrates file manipulation and text processing.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    NewLabel
    -0.08
     pillars
    -0.07
    Pan
    -0.07
    Hands
    -0.07
    Table
    -0.07
     [{'
    -0.07
    _funcs
    -0.07
     nedeniyle
    -0.07
    _ips
    -0.06
     JpaRepository
    -0.06
    POSITIVE LOGITS
     良
    0.06
     Μον
    0.06
    _PICTURE
    0.05
    щ
    0.05
    IFO
    0.05
    にか
    0.05
    794
    0.05
     takové
    0.05
    Ze
    0.05
     Nej
    0.05
    Act Density 0.005%

    No Known Activations