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

    Neuronpedia

    INDEX
    Explanations

    routers

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     ult
    -0.07
    .proc
    -0.07
     prv
    -0.07
     spikes
    -0.06
     oxide
    -0.06
     shoots
    -0.06
     End
    -0.06
    ogen
    -0.06
    [`
    -0.06
     Principle
    -0.06
    POSITIVE LOGITS
    .literal
    0.08
    /r
    0.06
    ��
    0.06
    yectos
    0.06
    _required
    0.06
     :",
    0.06
     drafting
    0.06
    ��
    0.06
    NON
    0.06
    _global
    0.06
    Act Density 0.004%

    No Known Activations