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

    Neuronpedia

    INDEX
    Explanations

    cooling

    np_max-act · gemini-2.0-flash

    The neuron chiefly activates on terms describing cooling or temperature‐reduction steps (e.g., “cool,” “slowly cool”) during processes like crystallization.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    uncture
    -0.07
     Tah
    -0.06
    Porn
    -0.06
    "/>↵
    -0.06
    马
    -0.06
    enate
    -0.06
     Instant
    -0.06
     кад
    -0.06
    _hot
    -0.06
     UX
    -0.06
    POSITIVE LOGITS
    ckså
    0.07
     interruption
    0.07
    _Osc
    0.07
    _che
    0.07
     buz
    0.07
    _DIALOG
    0.07
    جار
    0.07
    еш
    0.06
     другими
    0.06
    _special
    0.06
    Act Density 0.005%

    No Known Activations