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

    Neuronpedia

    INDEX
    Explanations

    sand and dust

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of desert landscapes or sandy terrain.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _MA
    -0.07
     deduction
    -0.06
    _extractor
    -0.06
    business
    -0.06
    -en
    -0.06
     �
    -0.06
    unnel
    -0.06
     Dome
    -0.06
    _readable
    -0.06
     deductions
    -0.06
    POSITIVE LOGITS
     sand
    0.07
     rollers
    0.06
     Sand
    0.06
    zw
    0.06
    halt
    0.06
     shaken
    0.06
     loadData
    0.06
     birik
    0.06
    ylv
    0.06
    (Category
    0.06
    Act Density 0.020%

    No Known Activations