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

    Neuronpedia

    INDEX
    Explanations

    feels

    np_max-act · gemini-2.0-flash

    The neuron responds to content‐bearing words—particularly nouns, verbs, and adjectives conveying technical or informational substance—highlighting the document’s key informational terms.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Ident
    -0.07
    .mount
    -0.07
    งเศส
    -0.07
    ipse
    -0.07
     Obviously
    -0.07
     im
    -0.07
    éf
    -0.06
     constituent
    -0.06
    akah
    -0.06
     EX
    -0.06
    POSITIVE LOGITS
     InputDecoration
    0.06
    IColor
    0.06
    .leadingAnchor
    0.06
    (shell
    0.06
     batchSize
    0.06
    .squareup
    0.06
    Toolkit
    0.05
     Generated
    0.05
    \data
    0.05
    .oauth
    0.05
    Act Density 5.771%

    No Known Activations