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

    Neuronpedia

    INDEX
    Explanations

    пом

    np_max-act · gemini-2.0-flash

    The neuron fires on Russian assistance-offer verbs (e.g. the stems of «постараюсь помочь» and «помочь»).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Approach
    -0.08
    (Global
    -0.07
    (copy
    -0.07
    .Max
    -0.07
     reporter
    -0.07
     overload
    -0.07
    _export
    -0.07
    	create
    -0.07
    .Target
    -0.07
     crave
    -0.07
    POSITIVE LOGITS
     пом
    0.07
     �
    0.06
    飾
    0.06
     уд
    0.06
     imposes
    0.06
    ُم
    0.06
    clicked
    0.06
     checkBox
    0.06
    aspers
    0.06
    ActionPerformed
    0.06
    Act Density 0.006%

    No Known Activations