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

    Neuronpedia

    INDEX
    Explanations

    mult

    np_max-act · gemini-2.0-flash

    The neuron responds to occurrences of the “multi-” prefix at the start of words (e.g. “Multilayer,” “Multivariate,” “Multivisceral,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     bear
    -0.07
     Nose
    -0.07
     assertEquals
    -0.07
     engraved
    -0.07
     бок
    -0.07
     ther
    -0.07
     göz
    -0.06
     reputed
    -0.06
     Esther
    -0.06
     Isaac
    -0.06
    POSITIVE LOGITS
     Multi
    0.11
    Mult
    0.10
     mult
    0.10
    multi
    0.10
    Multi
    0.10
    mult
    0.09
     multi
    0.09
     Mult
    0.09
    .Mult
    0.09
    .Multi
    0.08
    Act Density 0.023%

    No Known Activations