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

    Neuronpedia

    INDEX
    Explanations

    Code/Technical text

    np_max-act · gemini-2.0-flash

    This neuron fires on multi‐token quantifier phrases—specifically “a lot of …,” as in “a lot of preparation.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Psy
    -0.07
     births
    -0.07
    ..↵↵
    -0.06
    Uploader
    -0.06
     Hath
    -0.06
    нам
    -0.06
    φων
    -0.06
    //----------------------------------------------------------------------------------------------------------------
    -0.06
     �
    -0.06
    	os
    -0.06
    POSITIVE LOGITS
     Measure
    0.07
     dlouh
    0.07
    regor
    0.06
     Tail
    0.06
     saja
    0.06
    賞
    0.06
     waren
    0.06
    _Info
    0.06
     groupId
    0.06
    .zero
    0.06
    Act Density 0.000%

    No Known Activations