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

    Neuronpedia

    INDEX
    Explanations

    comma

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric dimension specifications in the form “number x number” (e.g. image size or resolution strings).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Islam
    -0.07
    akh
    -0.06
     purpos
    -0.06
    αιν
    -0.06
    olik
    -0.06
     robotic
    -0.06
     simul
    -0.06
     Achilles
    -0.06
    Included
    -0.06
    _trajectory
    -0.06
    POSITIVE LOGITS
     alanı
    0.07
     foundations
    0.06
     그냥
    0.06
    =settings
    0.06
    ↵↵↵↵↵↵
    0.06
     stump
    0.06
     vacation
    0.06
    optimize
    0.06
    áo
    0.06
    UpdateTime
    0.06
    Act Density 0.002%

    No Known Activations