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

    Neuronpedia

    INDEX
    Explanations

    article excerpts

    np_max-act · gemini-2.0-flash

    This neuron detects birth‐date mentions (e.g. year and “born” expressions) in biographical text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (The
    -0.08
     upcoming
    -0.06
     innocent
    -0.06
     DataService
    -0.06
     Famous
    -0.06
     Bible
    -0.06
    วย
    -0.06
    Arm
    -0.06
     Representative
    -0.06
     AM
    -0.06
    POSITIVE LOGITS
     erot
    0.07
     перш
    0.07
     halfway
    0.06
    #+
    0.06
     симв
    0.06
     handic
    0.06
    contres
    0.06
    stateParams
    0.06
     embodied
    0.06
    indre
    0.06
    Act Density 0.200%

    No Known Activations