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

    Neuronpedia

    INDEX
    Explanations

    few

    np_max-act · gemini-2.0-flash

    The neuron detects tokens that mark temporal expressions (e.g. parts of a day or month names and vague time references like “few,” “mid,” “late,” “beginning,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     vess
    -0.06
     tarea
    -0.06
    روه
    -0.06
    _known
    -0.06
    -spot
    -0.06
    iciální
    -0.05
     Після
    -0.05
    ịnh
    -0.05
    UILayout
    -0.05
     educate
    -0.05
    POSITIVE LOGITS
    _ERRORS
    0.08
     HttpClient
    0.07
    +'\
    0.07
    """↵
    0.07
    _PRIV
    0.07
     Rhode
    0.07
    NONE
    0.06
    &E
    0.06
    ampus
    0.06
     ""
↵
    0.06
    Act Density 0.548%

    No Known Activations