© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 130031
    Prev
    Next
    INDEX
    Explanations

    independently or individually controlled

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is detecting occurrences of the word “independent” (and its variants like “independently”/“operable”), i.e. tokens that convey independent operation or configurability.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     in
    -1.57
    ectady
    -1.45
    CardBody
    -1.29
    REPEAT
    -1.27
     في
    -1.26
    "}}>
    -1.26
    uſe
    -1.25
    tisgarh
    -1.24
    TERS
    -1.23
    ון
    -1.21
    POSITIVE LOGITS
     each
    1.41
     with
    1.19
     не
    1.16
    食器
    1.11
     both
    1.11
     mengakui
    1.09
    CHARLES
    1.08
     kablo
    1.07
     you
    1.06
     considera
    1.03
    Activations Density 0.081%

    No Known Activations