© 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. 22-GEMMASCOPE-RES-131K
    7. 25836
    Prev
    Next
    INDEX
    Explanations

    definitions and questions

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

    This neuron detects C/C++ preprocessor “#define” macros (especially those defining identifiers to numeric values).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     salonu
    -1.46
    Hå
    -1.39
    hadapi
    -1.34
     Konkur
    -1.29
     expédi
    -1.28
     naselje
    -1.26
     évé
    -1.23
    UTER
    -1.23
     Issus
    -1.23
     parfü
    -1.23
    POSITIVE LOGITS
     will
    1.80
     M
    1.59
     L
    1.52
     P
    1.51
     S
    1.50
    u
    1.49
     D
    1.49
     N
    1.48
    It
    1.48
     A
    1.45
    Activations Density 0.002%

    No Known Activations