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

    Neuronpedia

    INDEX
    Explanations

    for

    np_max-act · gemini-2.0-flash

    The neuron activates on floating‐point numeric tokens (e.g., probabilities or small decimal values) in the model’s token stream.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     vistas
    -0.07
    	pt
    -0.06
     Raid
    -0.06
     raid
    -0.06
    iled
    -0.06
    .None
    -0.06
     crowned
    -0.06
    elt
    -0.06
     lesbians
    -0.06
    oled
    -0.06
    POSITIVE LOGITS
     awaiting
    0.09
     waiting
    0.08
     for
    0.07
     reinforcements
    0.07
    ,
    0.07
    .sax
    0.07
     awaited
    0.07
    ↵↵
    0.07
    dis
    0.06
    ?
    0.06
    Act Density 0.011%

    No Known Activations