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

    Neuronpedia

    INDEX
    Explanations

    2

    np_max-act · gemini-2.0-flash

    The neuron strongly responds to the start-of-text token, i.e., the beginning of a sequence.

    oai_token-act-pair · gpt-5-miniTriggered by @tatsatx
    New Auto-Interp
    Negative Logits
     Gone
    -0.08
    須
    -0.08
     مرح
    -0.08
     Wanted
    -0.08
     bumps
    -0.08
     회
    -0.08
     cornerstone
    -0.08
    /rem
    -0.08
     births
    -0.08
     �
    -0.07
    POSITIVE LOGITS
     vergelijking
    0.08
     portátil
    0.08
    出去
    0.08
     envelop
    0.07
    pad
    0.07
     groot
    0.07
     fb
    0.07
     Ub
    0.07
    fw
    0.07
     grip
    0.07
    Act Density 0.221%

    No Known Activations