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

    Neuronpedia

    INDEX
    Explanations

    Code/forum snippets

    np_max-act · gemini-2.0-flash

    This neuron fires on special control/separator tokens (e.g. <|eot_id|>, <|start_header_id|>) that mark the boundaries or end of text segments.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    民主
    -0.08
    anta
    -0.07
     doubling
    -0.06
     warrior
    -0.06
     craz
    -0.06
     serge
    -0.06
     Tuesday
    -0.06
     hạ
    -0.06
    good
    -0.06
     вим
    -0.06
    POSITIVE LOGITS
    .setCancelable
    0.07
     noe
    0.06
     ;-
    0.06
     yemek
    0.06
    _Insert
    0.06
     pel
    0.06
    utzt
    0.06
     français
    0.06
     торгов
    0.06
    _ASCII
    0.06
    Act Density 0.127%

    No Known Activations