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

    Neuronpedia

    INDEX
    Explanations

    prepositions

    np_max-act · gemini-2.0-flash

    The neuron activates on words that signal quantification or measurement in a technical or scientific context (e.g. “amount,” “determining,” “measuring,” “impact,” “study,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     selection
    -0.07
     Goldman
    -0.07
    ("$.
    -0.06
    crast
    -0.06
     Weak
    -0.06
    Bur
    -0.06
    ter
    -0.06
     titles
    -0.06
    :,
    -0.06
    Ak
    -0.06
    POSITIVE LOGITS
    řej
    0.07
    قق
    0.06
    英语
    0.06
     accents
    0.06
     vrch
    0.06
     مستق
    0.06
     datum
    0.06
    -Isl
    0.06
     spacecraft
    0.06
     bietet
    0.06
    Act Density 0.317%

    No Known Activations