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

    Neuronpedia

    INDEX
    Explanations

    solar

    np_max-act · gemini-2.0-flash

    The neuron activates on the token “solar,” i.e. it detects occurrences of the word “solar.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .getIn
    -0.09
    uggest
    -0.07
    proof
    -0.07
    [end
    -0.07
    makes
    -0.07
     Freund
    -0.07
    byterian
    -0.07
    	TArray
    -0.06
     ATK
    -0.06
    otp
    -0.06
    POSITIVE LOGITS
     solar
    0.15
     Solar
    0.14
    Solar
    0.12
    or
    0.09
    ラ
    0.09
    ar
    0.09
    OR
    0.08
     stellar
    0.08
    kar
    0.08
    ара
    0.08
    Act Density 0.006%

    No Known Activations