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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 111070
    Prev
    Next
    INDEX
    Explanations

    numbers code hex

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

    hexadecimal digit sequences in code and data.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires on numeric constants in the code (both decimal and hexadecimal literals).

    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
    Dữ
    -1.64
     your
    -1.59
     músculo
    -1.59
    kệ
    -1.53
    thiết
    -1.52
    ﷲ
    -1.52
    hướng
    -1.51
     our
    -1.51
     that
    -1.49
    bảng
    -1.48
    POSITIVE LOGITS
     more
    1.68
    stalgia
    1.66
     idag
    1.56
    inerario
    1.51
    entown
    1.50
    ooplankton
    1.50
    上的
    1.48
     with
    1.45
    There
    1.42
    comings
    1.41
    Activations Density 0.004%

    No Known Activations