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

    Neuronpedia

    INDEX
    Explanations

    stride

    np_max-act · gemini-2.0-flash

    The neuron activates on floating‐point numeric literals (especially those long decimals).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     linguistic
    -0.07
    .fetch
    -0.07
     ions
    -0.06
    -family
    -0.06
     circus
    -0.06
     awaken
    -0.06
    _memory
    -0.06
    IMITIVE
    -0.06
    OTTOM
    -0.06
    _Post
    -0.06
    POSITIVE LOGITS
     stride
    0.13
     strides
    0.10
    stride
    0.09
    Stride
    0.07
    omidou
    0.07
    .stride
    0.07
     تد
    0.07
     Verde
    0.07
     prod
    0.07
     saldo
    0.07
    Act Density 0.001%

    No Known Activations