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

    Neuronpedia

    INDEX
    Explanations

    Boolean/logical statements

    np_max-act · gemini-2.0-flash

    The neuron never activates on any token—it’s effectively a dead neuron that doesn’t detect any pattern.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Neville
    -0.07
    684
    -0.06
     Desc
    -0.06
    ึง
    -0.06
    ocolate
    -0.06
     artist
    -0.06
    Json
    -0.06
    _heat
    -0.06
    zel
    -0.06
    -spec
    -0.06
    POSITIVE LOGITS
     İmparator
    0.07
    ciler
    0.07
     loved
    0.06
    .bindingNavigatorMove
    0.06
    _follow
    0.06
     děl
    0.06
    郡
    0.06
    [of
    0.06
     Ιω
    0.06
    lopen
    0.06
    Act Density 0.177%

    No Known Activations