© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 21014
    Prev
    Next
    INDEX
    Explanations

    MSC compiler directives

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

    The neuron strongly activates on code tokens that denote platform or compiler version identifiers and numeric macros (e.g. WIN32, X86, MSC_VER, etc.).

    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
     from
    -1.70
     to
    -1.70
     for
    -1.60
     the
    -1.58
     eight
    -1.41
     eleven
    -1.38
     various
    -1.37
     umano
    -1.34
     on
    -1.34
     dozens
    -1.33
    POSITIVE LOGITS
     autre
    1.52
    𓆜
    1.46
     également
    1.43
    only
    1.41
     ratón
    1.38
     giyim
    1.37
     mães
    1.36
    
    1.33
     jiné
    1.32
     sereia
    1.32
    Activations Density 0.024%

    No Known Activations