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

    Neuronpedia

    INDEX
    Explanations

    phi

    np_max-act · gemini-2.0-flash

    This neuron activates on occurrences of the Delphi programming language name (its subword tokens like “Del” and “phi”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     حرکت
    -0.08
    UnityEngine
    -0.07
    WEEN
    -0.07
     fest
    -0.06
    ritel
    -0.06
    REC
    -0.06
    MaxLength
    -0.06
    WRITE
    -0.06
     Karma
    -0.06
    _only
    -0.06
    POSITIVE LOGITS
     خصوص
    0.06
    .pi
    0.06
    (md
    0.06
     Pa
    0.06
    cheap
    0.06
    (sa
    0.06
    months
    0.06
     국민
    0.06
    BH
    0.06
    _pal
    0.06
    Act Density 0.011%

    No Known Activations