© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 112951
    Prev
    Next
    INDEX
    Explanations

    version in code

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

    references to Microsoft Visual C++ compiler version numbers in code (specifically `_MSC_VER` identifiers).

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

    The neuron responds specifically to the literal token “Properties,” flagging occurrences of that exact capitalized term.

    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
    ൎ
    -1.83
    灩
    -1.65
    ьет
    -1.63
    rdoba
    -1.56
    🧳
    -1.50
     setelan
    -1.49
    ⢅
    -1.48
    𓋼
    -1.46
    ITIONAL
    -1.45
    OUTUBE
    -1.45
    POSITIVE LOGITS
     However
    1.60
    Unfortunately
    1.42
    <i>
    1.37
     You
    1.34
    </b>
    1.33
     is
    1.32
    Recently
    1.30
     spanning
    1.29
     ‘
    1.29
     {
    1.29
    Activations Density 0.001%

    No Known Activations