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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     overs
    -0.07
     derivatives
    -0.06
     incl
    -0.06
     descriptions
    -0.06
    actly
    -0.06
     synonym
    -0.06
    (build
    -0.06
     requirements
    -0.06
    (cost
    -0.06
    Di
    -0.06
    POSITIVE LOGITS
    شناسی
    0.07
    内の
    0.07
     partager
    0.07
    .Companion
    0.07
     ответ
    0.06
    .radio
    0.06
    ~-
    0.06
    -split
    0.06
    .file
    0.06
     inscription
    0.06
    Act Density 0.456%

    No Known Activations