© 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. 34-GEMMASCOPE-RES-131K
    7. 34518
    Prev
    Next
    INDEX
    Explanations

    inclusion of headers

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

    The neuron detects C/C++ preprocessor macro definitions (lines starting with “#define” and their all-caps identifiers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     then
    -0.92
     來
    -0.77
    ondy
    -0.75
     kaksi
    -0.74
    Miejs
    -0.73
     berechnet
    -0.73
     each
    -0.73
     diterapkan
    -0.73
     while
    -0.72
     apoi
    -0.72
    POSITIVE LOGITS
    included
    1.25
    Included
    1.15
     Included
    1.11
     included
    1.05
     strateg
    0.93
     VERSION
    0.93
     NOM
    0.91
    INCLUDED
    0.91
     affirm
    0.85
     INCLUDED
    0.84
    Activations Density 0.132%

    No Known Activations