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

    Neuronpedia

    INDEX
    Explanations

    Technical documentation excerpts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Modeling
    -0.07
     Budd
    -0.07
    .urls
    -0.07
    зація
    -0.07
                    ↵                ↵
    -0.06
     witches
    -0.06
     Catalyst
    -0.06
    质量
    -0.06
    aniem
    -0.06
     refusal
    -0.06
    POSITIVE LOGITS
    ensem
    0.07
     wisdom
    0.07
     oxid
    0.06
     shrine
    0.06
    affected
    0.06
    itel
    0.06
     strictly
    0.06
    /content
    0.06
    _ipv
    0.06
    tridges
    0.06
    Act Density 0.000%

    No Known Activations