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

    Neuronpedia

    INDEX
    Explanations

    computer code

    np_max-act · gemini-2.0-flash

    This neuron activates on words related to task‐planning and procedural steps (e.g. “tasks,” “dependencies,” “resolve,” “needed,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Knot
    -0.07
     reply
    -0.07
    Returned
    -0.07
     fruit
    -0.06
    returned
    -0.06
     eggs
    -0.06
     putting
    -0.06
    View
    -0.06
     odpově
    -0.06
    戒
    -0.06
    POSITIVE LOGITS
     Hend
    0.07
     bour
    0.06
    UMMY
    0.06
     inlet
    0.06
     مذه
    0.06
    (:,
    0.06
    omics
    0.06
     постанов
    0.06
    관련
    0.06
     ylabel
    0.06
    Act Density 0.008%

    No Known Activations