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

    Neuronpedia

    INDEX
    Explanations

    Numbers two and three

    np_max-act · gemini-2.0-flash

    The neuron primarily activates on standalone numeric tokens (digits or number words).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     най
    -0.07
    UNIX
    -0.06
    .vs
    -0.06
     remain
    -0.06
     unsure
    -0.06
    AIN
    -0.06
    	channel
    -0.06
    85
    -0.06
     yağ
    -0.06
    Gene
    -0.06
    POSITIVE LOGITS
     @@↵
    0.08
     casualties
    0.07
    	MPI
    0.07
     feder
    0.06
     whit
    0.06
     =↵↵
    0.06
    ighthouse
    0.06
     solves
    0.06
     sie
    0.06
    CKET
    0.06
    Act Density 0.046%

    No Known Activations