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

    Neuronpedia

    INDEX
    Explanations

    warrior

    np_max-act · gemini-2.0-flash

    The neuron activates on the character sequence “rior,” i.e. it detects the “rior” suffix as in “Warrior.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Shows
    -0.07
     bö
    -0.07
     Luke
    -0.07
    iname
    -0.06
    53
    -0.06
     розрах
    -0.06
     input
    -0.06
    calendar
    -0.06
     spontaneous
    -0.06
     STOCK
    -0.06
    POSITIVE LOGITS
     Warrior
    0.13
     Warriors
    0.11
     warrior
    0.11
     warriors
    0.10
     disrespectful
    0.08
     successors
    0.08
     guerr
    0.08
     servi
    0.07
     عاش
    0.07
     addslashes
    0.07
    Act Density 0.003%

    No Known Activations