© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 108671
    Prev
    Next
    INDEX
    Explanations

    code or programming constructs

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

    This neuron detects numeric tokens and constants (e.g. literal numbers and numeric macro names) in code.

    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
    Лю
    -0.73
    ГИ
    -0.69
     Southgate
    -0.68
    нейших
    -0.67
     ово
    -0.66
    По
    -0.66
    ゼ
    -0.64
     beantworten
    -0.64
     década
    -0.63
    副
    -0.63
    POSITIVE LOGITS
     CHAR
    1.07
    UCHAR
    0.92
    RAND
    0.90
     Currie
    0.81
     RAND
    0.81
     Cassidy
    0.80
    tigungs
    0.79
    spre
    0.79
    CHAR
    0.79
     Mahler
    0.76
    Activations Density 0.045%

    No Known Activations